From 47e2a8b16b0ab46b0ad78911fcfa5ed27da51c6f Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 3 Jun 2026 16:31:54 +0200
Subject: [PATCH 01/81] Add BERT configs

---
 src/sequifier/config/probabilities.py | 81 +++++++++++++++++++++++++++
 src/sequifier/config/train_config.py  | 66 +++++++++++++++++++++-
 src/sequifier/make.py                 |  1 +
 tests/unit/test_train.py              |  1 +
 4 files changed, 148 insertions(+), 1 deletion(-)
 create mode 100644 src/sequifier/config/probabilities.py

diff --git a/src/sequifier/config/probabilities.py b/src/sequifier/config/probabilities.py
new file mode 100644
index 00000000..2591cc97
--- /dev/null
+++ b/src/sequifier/config/probabilities.py
@@ -0,0 +1,81 @@
+import math
+import random
+from abc import ABC, abstractmethod
+from typing import Annotated, Literal, Union
+
+from pydantic import BaseModel, Field
+
+epsilon = 1e-20
+
+
+class ProbabilityDistributionBaseClass(ABC):
+    """
+    Abstract base class for all probability distributions.
+    """
+
+    @abstractmethod
+    def sample(self) -> Union[int, float]:
+        pass
+
+
+class GeometricDistribution(BaseModel, ProbabilityDistributionBaseClass):
+    type: Literal["GeometricDistribution"] = "GeometricDistribution"
+    p: float = Field(..., gt=0.0, le=1.0)
+
+    def sample(self) -> int:
+        if self.p == 1.0:
+            return 1
+        return math.ceil(math.log(random.random() + epsilon) / math.log(1.0 - self.p))
+
+
+class NormalDistributionDiscretizedFloor(BaseModel, ProbabilityDistributionBaseClass):
+    type: Literal["NormalDistributionDiscretizedFloor"] = (
+        "NormalDistributionDiscretizedFloor"
+    )
+    mean: float
+    standard_deviation: float = Field(..., gt=0.0)
+
+    def sample(self) -> int:
+        val = random.gauss(self.mean, self.standard_deviation)
+        return max(round(val), 0) + 1
+
+
+class LogNormalDistributionDistcretizedFloor(
+    BaseModel, ProbabilityDistributionBaseClass
+):
+    type: Literal["LogNormalDistributionDistcretizedFloor"] = (
+        "LogNormalDistributionDistcretizedFloor"
+    )
+    mean: float
+    standard_deviation: float = Field(..., gt=0.0)
+
+    def sample(self) -> int:
+        val = random.lognormvariate(self.mean, self.standard_deviation)
+        return round(val) + 1
+
+
+class PoissonDistributionFloor(BaseModel, ProbabilityDistributionBaseClass):
+    type: Literal["PoissonDistributionFloor"] = "PoissonDistributionFloor"
+    rate: float = Field(..., gt=0.0)
+
+    def sample(self) -> int:
+        L = math.exp(-self.rate)
+        k = 0
+        p = 1.0
+
+        while p > L:
+            k += 1
+            p *= random.random()
+
+        return k
+
+
+ProbabilityDistribution = Annotated[
+    Union[
+        GeometricDistribution,
+        NormalDistributionDiscretizedFloor,
+        LogNormalDistributionDistcretizedFloor,
+        PoissonDistributionFloor,
+    ],
+    Field(discriminator="type"),
+]
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index b37b1f76..eb095ab3 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -1,5 +1,6 @@
 import copy
 import json
+import math
 import os
 import warnings
 from typing import Any, Optional, Union
@@ -10,9 +11,10 @@
 import yaml
 from beartype import beartype
 from loguru import logger
-from pydantic import BaseModel, ConfigDict, Field, field_validator
+from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validator
 
 import sequifier
+from sequifier.config.probabilities import ProbabilityDistribution
 from sequifier.helpers import normalize_path, try_catch_excess_keys
 
 AnyType = str | int | float
@@ -310,6 +312,27 @@ def validate_data_parallelism(cls, v):
         return v
 
 
+class ReplacementDistribution(BaseModel):
+    masked: float = Field(..., gt=0.0, le=1.0)
+    random: float = Field(..., gt=0.0, le=1.0)
+    identical: float = Field(..., gt=0.0, le=1.0)
+
+    @model_validator(mode="after")
+    def validate_sum(self):
+        total = self.masked + self.random + self.identical
+        if not math.isclose(total, 1.0, abs_tol=1e-5):
+            raise ValueError(
+                f"Replacement distribution probabilities must sum to 1.0, got {total}"
+            )
+        return self
+
+
+class BERTSpecModel(BaseModel):
+    masking_probability: float = Field(..., gt=0.0, le=1.0)
+    replacement_distribution: ReplacementDistribution
+    span_masking: ProbabilityDistribution
+
+
 class ModelSpecModel(BaseModel):
     """Pydantic model for model specifications.
 
@@ -324,6 +347,7 @@ class ModelSpecModel(BaseModel):
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
+    training_objective: str = "causal"
     initial_embedding_dim: int
     feature_embedding_dims: Optional[dict[str, int]] = None
     joint_embedding_dim: Optional[int] = None
@@ -344,6 +368,28 @@ class ModelSpecModel(BaseModel):
     rope_theta: float = 10000.0
 
     prediction_length: int
+    bert_spec: Optional[BERTSpecModel] = None
+
+    @field_validator("training_objective")
+    @classmethod
+    def validate_training_objective(cls, v):
+        if v not in ["causal", "bert"]:
+            raise ValueError(f"Only 'causal' and 'bert' are allowed, found {v}")
+        return v
+
+    @field_validator("bert_spec")
+    @classmethod
+    def validate_bert_spec(cls, v, info):
+        training_objective = info.data.get("training_objective")
+        if v and not training_objective == "bert":
+            raise ValueError(
+                "The BERT hyperparameters should only be configured if the training objective is 'bert'"
+            )
+        if not v and training_objective == "bert":
+            raise ValueError(
+                "If the training_objective is 'bert', the BERT hyperparameters must be set"
+            )
+        return v
 
     @field_validator("dim_model")
     @classmethod
@@ -463,6 +509,7 @@ class TrainModel(BaseModel):
         n_classes: The number of classes for each categorical column.
         inference_batch_size: The batch size to be used for inference after model export.
         seed: The random seed for numpy and PyTorch.
+        model_type: Causal or BERT model type
         export_generative_model: If True, exports the generative model.
         export_embedding_model: If True, exports the embedding model.
         export_onnx: If True, exports the model in ONNX format.
@@ -666,4 +713,21 @@ def validate_model_spec(cls, v, info):
                     f"initial_embedding_dim ({embedding_size}) must be a multiple of the number of categorical variables ({n_categorical}: {categorical_columns})."
                 )
 
+        export_generative_model = info.data.get("export_generative_model")
+        export_embedding_model = info.data.get("export_embedding_model")
+        if v.training_objective == "bert":
+            if export_generative_model:
+                raise ValueError(
+                    "'training_objective: bert' is incompatible with generative model export"
+                )
+            if not export_embedding_model:
+                raise ValueError(
+                    "'training_objective: bert' requires embedding model export"
+                )
+        if v.training_objective == "causal":
+            if not export_generative_model and not export_embedding_model:
+                warnings.warn(
+                    "At least one of 'export_generative_model' and 'export_embedding_model' should be true"
+                )
+
         return v
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index fdc29a7a..8943d861 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -36,6 +36,7 @@
 export_onnx: true
 
 model_spec:
+  training_objective: causal
   initial_embedding_dim: 128
   feature_embedding_dims: # the size of the embedding of individual variables, must sum to dim_model
     EXAMPLE_INPUT_COLUMN_NAME: # can be left out if either all input variables are real or all are categorical
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 2f8397b7..cf140984 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -14,6 +14,7 @@ def model_config(tmp_path):
     (tmp_path / "logs").mkdir(exist_ok=True)
 
     model_spec = ModelSpecModel(
+        training_objective="causal",
         initial_embedding_dim=16,
         dim_model=16,
         n_head=4,

From 8fd1493b8d0359f90ffafc6b1efd38a963c17d4e Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 3 Jun 2026 18:31:04 +0200
Subject: [PATCH 02/81] Introduce mask token

---
 src/sequifier/helpers.py                      |  14 +-
 src/sequifier/preprocess.py                   |   8 +-
 tests/configs/train-test-categorical.yaml     |   2 +-
 .../configs/train-test-resume-mid-epoch.yaml  |   2 +-
 tests/integration/test_inference.py           |  13 +-
 tests/integration/test_preprocessing.py       |   4 +-
 .../source_configs/id_maps/itemId.json        |   2 +-
 ...orical-1-best-embedding-3-0-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-1-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-2-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-3-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-4-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-5-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-6-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-7-embeddings.csv |   2 +-
 ...inf-size-best-embedding-3-0-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-1-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-2-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-3-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-4-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-5-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-6-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-7-embeddings.csv |   6 +-
 ...-1-best-3-autoregression-1-predictions.csv |   2 +-
 ...-1-best-3-autoregression-2-predictions.csv |  40 +-
 ...-1-best-3-autoregression-3-predictions.csv |   4 +-
 ...-1-best-3-autoregression-4-predictions.csv |  38 +-
 ...-1-best-3-autoregression-5-predictions.csv |  78 ++--
 ...-1-best-3-autoregression-6-predictions.csv |  40 +-
 ...-1-best-3-autoregression-7-predictions.csv |   4 +-
 ...del-categorical-1-best-3-2-predictions.csv |   2 +-
 ...del-categorical-1-best-3-3-predictions.csv |   2 +-
 ...del-categorical-1-best-3-4-predictions.csv |   2 +-
 ...del-categorical-1-best-3-5-predictions.csv |   4 +-
 ...del-categorical-1-best-3-6-predictions.csv |   2 +-
 ...del-categorical-1-best-3-7-predictions.csv |   2 +-
 ...orical-1-inf-size-best-3-1-predictions.csv |   4 +-
 ...orical-1-inf-size-best-3-2-predictions.csv |   2 +-
 ...orical-1-inf-size-best-3-3-predictions.csv |   4 +-
 ...orical-1-inf-size-best-3-5-predictions.csv |   6 +-
 ...orical-1-inf-size-best-3-6-predictions.csv |   4 +-
 ...orical-1-inf-size-best-3-7-predictions.csv |   2 +-
 ...del-categorical-3-best-3-3-predictions.csv |   2 +-
 ...orical-3-inf-size-best-3-0-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-1-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-2-predictions.csv |   4 +-
 ...orical-3-inf-size-best-3-3-predictions.csv |   8 +-
 ...orical-3-inf-size-best-3-4-predictions.csv |   4 +-
 ...orical-3-inf-size-best-3-5-predictions.csv |  10 +-
 ...orical-3-inf-size-best-3-6-predictions.csv |   4 +-
 ...orical-3-inf-size-best-3-7-predictions.csv |   4 +-
 ...el-categorical-50-best-3-0-predictions.csv |   2 +-
 ...el-categorical-50-best-3-1-predictions.csv |   2 +-
 ...el-categorical-50-best-3-2-predictions.csv |   2 +-
 ...el-categorical-50-best-3-3-predictions.csv |   4 +-
 ...el-categorical-50-best-3-5-predictions.csv |   2 +-
 ...el-categorical-50-best-3-6-predictions.csv |   2 +-
 ...rical-distributed-best-3-0-predictions.csv |   2 +-
 ...rical-distributed-best-3-1-predictions.csv |   2 +-
 ...rical-distributed-best-3-2-predictions.csv |   2 +-
 ...rical-distributed-best-3-3-predictions.csv |   4 +-
 ...rical-distributed-best-3-4-predictions.csv |   2 +-
 ...rical-distributed-best-3-5-predictions.csv |   4 +-
 ...rical-distributed-best-3-7-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-0-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-2-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-3-predictions.csv |   4 +-
 ...-categorical-lazy-best-3-4-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-5-predictions.csv |   4 +-
 ...-categorical-lazy-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-7-predictions.csv |   2 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...custom-eval-run-0-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-7-predictions.csv |  58 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-7-predictions.csv |  58 +--
 ...l-categorical-1-best-3-0-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-1-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-2-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-3-probabilities.csv |   6 +-
 ...l-categorical-1-best-3-4-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-5-probabilities.csv |   6 +-
 ...l-categorical-1-best-3-6-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-7-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-0-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-1-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-2-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-3-probabilities.csv |   6 +-
 ...l-categorical-3-best-3-4-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-5-probabilities.csv |   6 +-
 ...l-categorical-3-best-3-6-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-7-probabilities.csv |   4 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  14 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  14 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  14 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  14 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  14 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  14 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   8 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   8 +-
 ...l-categorical-5-best-3-0-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-1-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-2-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-3-probabilities.csv |   6 +-
 ...l-categorical-5-best-3-4-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-5-probabilities.csv |   6 +-
 ...l-categorical-5-best-3-6-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-7-probabilities.csv |   4 +-
 ...-categorical-50-best-3-0-probabilities.csv |   4 +-
 ...-categorical-50-best-3-1-probabilities.csv |   4 +-
 ...-categorical-50-best-3-2-probabilities.csv |   4 +-
 ...-categorical-50-best-3-3-probabilities.csv |   6 +-
 ...-categorical-50-best-3-4-probabilities.csv |   4 +-
 ...-categorical-50-best-3-5-probabilities.csv |   6 +-
 ...-categorical-50-best-3-6-probabilities.csv |   4 +-
 ...-categorical-50-best-3-7-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   6 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   6 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   6 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   6 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   4 +-
 tests/unit/test_helpers.py                    |  18 +-
 tests/unit/test_preprocess.py                 |   6 +-
 193 files changed, 1889 insertions(+), 1878 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index f2097548..84e39b0c 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -83,11 +83,13 @@ def construct_index_maps(
     the original string or integer identifiers. It only performs this
     operation if `map_to_id` is True and `id_maps` is provided.
 
-    A special mapping for index 0 is added:
+    A special mapping for these indices are added:
     - If original IDs are strings, 0 maps to "unknown".
     - If original IDs are strings, 1 maps to "other".
-    - If original IDs are integers, 0 maps to (minimum original ID) - 2.
-    - If original IDs are integers, 1 maps to (minimum original ID) - 1.
+    - If original IDs are strings, 2 maps to "unknown".
+    - If original IDs are integers, 0 maps to (minimum original ID) - 3.
+    - If original IDs are integers, 1 maps to (minimum original ID) - 2.
+    - If original IDs are integers, 2 maps to (minimum original ID) - 1.
 
     Args:
         id_maps: A nested dictionary mapping column names to their
@@ -120,12 +122,14 @@ def construct_index_maps(
             if isinstance(val, str):
                 map_[0] = "unknown"
                 map_[1] = "other"
+                map_[2] = "mask"
             else:
                 if not isinstance(val, int):
                     raise TypeError(f"Expected integer ID in map, got {type(val)}")
                 min_id = int(min(map_.values()))
-                map_[0] = min_id - 2  # type: ignore
-                map_[1] = min_id - 1
+                map_[0] = min_id - 3  # type: ignore
+                map_[1] = min_id - 2  # type: ignore
+                map_[2] = min_id - 1  # type: ignore
             index_map[target_column] = map_
     return index_map
 
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index f25dc721..dacb92f6 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -973,9 +973,9 @@ def load_precomputed_id_maps(
                     if not len(m) > 0:
                         raise ValueError(f"map in {file} does not contain any values")
                     min_val = min(m.values())
-                    if min_val != 2:
+                    if min_val != 3:
                         raise ValueError(
-                            f"minimum value in map {file} is {min_val}, must be 2."
+                            f"minimum value in map {file} is {min_val}, must be 3."
                         )
                     custom_maps[col_name] = m
     if required_maps:
@@ -1461,7 +1461,7 @@ def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]
     ids = sorted(
         [int(x) if not isinstance(x, str) else x for x in np.unique(data[column])]
     )  # type: ignore
-    id_map = {id_: i + 2 for i, id_ in enumerate(ids)}
+    id_map = {id_: i + 3 for i, id_ in enumerate(ids)}
     return dict(id_map)
 
 
@@ -1529,7 +1529,7 @@ def combine_maps(
         A new, combined, and re-indexed ID map.
     """
     combined_keys = sorted(list(set(list(map1.keys())).union(list(set(map2.keys())))))
-    id_map = {id_: i + 2 for i, id_ in enumerate(combined_keys)}
+    id_map = {id_: i + 3 for i, id_ in enumerate(combined_keys)}
     return id_map
 
 
diff --git a/tests/configs/train-test-categorical.yaml b/tests/configs/train-test-categorical.yaml
index 2da9eace..468aab70 100644
--- a/tests/configs/train-test-categorical.yaml
+++ b/tests/configs/train-test-categorical.yaml
@@ -47,7 +47,7 @@ training_spec:
   loss_weights:
     itemId: 1.0
   class_weights:
-    itemId: [1.0, 1.0, 1.5, 0.5, 0.6, 1.2, 0.66666667, 1.5, 1.2, 0.75, 0.85714286, 0.75, 0.85714286, 0.75, 0.6, 2.,1.5, 1.5, 1.0, 1.0, 0.66666667, 0.75, 0.5, 0.75, 2.0, 1.2, 3.0, 1.0, 0.75, 1.0, 0.85714286, 1.0]
+    itemId: [1.0, 1.0, 1.0, 1.5, 0.5, 0.6, 1.2, 0.66666667, 1.5, 1.2, 0.75, 0.85714286, 0.75, 0.85714286, 0.75, 0.6, 2.,1.5, 1.5, 1.0, 1.0, 0.66666667, 0.75, 0.5, 0.75, 2.0, 1.2, 3.0, 1.0, 0.75, 1.0, 0.85714286, 1.0]
   optimizer:
     name: QHAdam
   scheduler:
diff --git a/tests/configs/train-test-resume-mid-epoch.yaml b/tests/configs/train-test-resume-mid-epoch.yaml
index 17ec8f9e..b0449d91 100644
--- a/tests/configs/train-test-resume-mid-epoch.yaml
+++ b/tests/configs/train-test-resume-mid-epoch.yaml
@@ -48,7 +48,7 @@ training_spec:
     itemId: 1.0
     supCat1: 1.0
   class_weights:
-    itemId: [1.0, 1.0, 1.5, 0.5, 0.6, 1.2, 0.66666667, 1.5, 1.2, 0.75, 0.85714286, 0.75, 0.85714286, 0.75, 0.6, 2.,1.5, 1.5, 1.0, 1.0, 0.66666667, 0.75, 0.5, 0.75, 2.0, 1.2, 3.0, 1.0, 0.75, 1.0, 0.85714286, 1.0]
+    itemId: [1.0, 1.0, 1.0, 1.5, 0.5, 0.6, 1.2, 0.66666667, 1.5, 1.2, 0.75, 0.85714286, 0.75, 0.85714286, 0.75, 0.6, 2.,1.5, 1.5, 1.0, 1.0, 0.66666667, 0.75, 0.5, 0.75, 2.0, 1.2, 3.0, 1.0, 0.75, 1.0, 0.85714286, 1.0]
   optimizer:
     name: QHAdam
   scheduler:
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 10f885c7..540c8e1f 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -26,7 +26,7 @@ def test_predictions_real(predictions):
 
 
 def test_predictions_cat(predictions):
-    valid_values = [str(x) for x in np.arange(100, 130)] + ["unknown"]
+    valid_values = [str(x) for x in np.arange(100, 130)] + ["unknown", "other", "mask"]
     for model_name, model_predictions in predictions.items():
         if "categorical" in model_name or "multitarget" in model_name:
             assert np.all(
@@ -42,6 +42,7 @@ def test_predictions_cat(predictions):
                 admssible_vals = [str(x) for x in np.arange(0, 10)] + [
                     "unknown",
                     "other",
+                    "mask",
                 ]
                 assert np.all(
                     [
@@ -81,9 +82,9 @@ def test_predictions_cat(predictions):
 def test_probabilities(probabilities):
     for model_name, model_probabilities in probabilities.items():
         if "itemId" in model_name:
-            assert model_probabilities.shape[1] == 32
+            assert model_probabilities.shape[1] == 33
         elif "supCat1" in model_name:
-            assert model_probabilities.shape[1] == 12
+            assert model_probabilities.shape[1] == 13
 
         np.testing.assert_almost_equal(
             model_probabilities.sum_horizontal(),
@@ -101,7 +102,11 @@ def test_multi_pred(predictions):
         assert preds.shape[0] > 0, f"{model_name} has no predictions"
         assert preds.shape[1] == 5, f"{model_name} should have 5 columns"
 
-        admssible_vals = [str(x) for x in np.arange(0, 10)] + ["unknown", "other"]
+        admssible_vals = [str(x) for x in np.arange(0, 10)] + [
+            "unknown",
+            "other",
+            "mask",
+        ]
 
         assert np.all(
             [v in admssible_vals for v in preds["supCat1"]]
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 209ad49c..79e13b1f 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -45,7 +45,7 @@ def test_metadata_config(metadata_configs):
 
         if "itemId" in metadata_config["n_classes"]:
             assert len(metadata_config["id_maps"]["itemId"]) == 30
-            assert metadata_config["n_classes"]["itemId"] == 32
+            assert metadata_config["n_classes"]["itemId"] == 33
 
             id_map_keys = np.array(
                 sorted(list(metadata_config["id_maps"]["itemId"].keys()))
@@ -59,7 +59,7 @@ def test_metadata_config(metadata_configs):
             )
             # assert False, id_map_values
             assert np.all(
-                id_map_values == np.arange(2, len(id_map_values) + 2)
+                id_map_values == np.arange(3, len(id_map_values) + 3)
             ), id_map_values
 
         if "itemValue" in metadata_config["selected_columns_statistics"]:
diff --git a/tests/resources/source_configs/id_maps/itemId.json b/tests/resources/source_configs/id_maps/itemId.json
index 20ff05d6..77d5f018 100644
--- a/tests/resources/source_configs/id_maps/itemId.json
+++ b/tests/resources/source_configs/id_maps/itemId.json
@@ -1 +1 @@
-{"100": 2, "101": 3, "102": 4, "103": 5, "104": 6, "105": 7, "106": 8, "107": 9, "108": 10, "109": 11, "110": 12, "111": 13, "112": 14, "113": 15, "114": 16, "115": 17, "116": 18, "117": 19, "118": 20, "119": 21, "120": 22, "121": 23, "122": 24, "123": 25, "124": 26, "125": 27, "126": 28, "127": 29, "128": 30, "129": 31}
+{"100": 3, "101": 4, "102": 5, "103": 6, "104": 7, "105": 8, "106": 9, "107": 10, "108": 11, "109": 12, "110": 13, "111": 14, "112": 15, "113": 16, "114": 17, "115": 18, "116": 19, "117": 20, "118": 21, "119": 22, "120": 23, "121": 24, "122": 25, "123": 26, "124": 27, "125": 28, "126": 29, "127": 30, "128": 31, "129": 32}
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
index 76bd77d7..d63e16f0 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,7,0.8649381,0.24562168,-0.47390407,0.26047117,-0.24761012,1.3031981,0.95492613,-0.36946228,-0.91278917,-0.7628124,1.0646838,-0.84439635,0.16634521,-0.007354275,-1.7519352,-0.033109475,-0.45459607,0.115075,1.1531727,0.9546873,1.2859629,-1.0329498,1.2402246,2.6194525,-0.55354816,-0.63395625,-2.1362736,0.8346781,-1.0040193,-1.0341624,-0.8379355,0.057529,1.3040087,0.12587658,0.8301326,-0.5306769,-1.623395,0.36828652,1.856288,0.320922,-1.8282198,-1.3440526,-0.9354176,0.24922973,1.2998884,-1.236273,-0.055613205,-0.8205791,-0.4832966,1.8687013,-1.4167951,-1.1055189,1.6474618,0.9178678,-0.94887453,1.1648152,0.03853327,-0.36139467,-0.66810036,-1.3988649,-0.1772363,-0.077192,-0.42204562,1.8334289,1.367824,-0.21193206,-0.16146825,-1.3633878,1.9045291,0.64752847,-0.5658707,-1.9270079,-0.7966031,-0.4969335,-1.1129528,-0.3014826,-1.2423565,0.99004376,-0.5590482,0.210285,1.45484,0.78570837,0.4753621,-2.9085505,-0.41873848,-0.30167663,-0.785309,-0.31538683,-0.9972849,0.013730815,-0.4415555,0.761963,-1.4535736,1.3455216,0.027025485,1.8463998,1.8265476,0.24507502,0.7862814,-0.902054,0.41721246,0.29755303,0.40205243,2.0839143,0.08692597,-0.7652936,-1.7749745,0.7529008,-2.0421607,-0.78219783,-0.539601,-0.8647213,0.6763834,0.54173374,1.1983194,0.9964866,0.35037822,1.9535118,-0.9696219,1.1370679,1.1510953,0.2664108,-0.06128295,-0.7830806,-0.050288465,1.0513953,0.33808768,-0.23930782,-0.6178605,0.55309016,-1.3475106,-1.2853458,-0.20373966,-2.6442385,1.2245853,0.4053069,-0.67187744,0.47685108,1.4946405,-0.057367213,0.5233804,-0.41413814,-0.38341537,1.4250722,1.7166282,-0.8751589,-0.07256896,-1.6518211,-0.09111433,0.9957127,0.033603884,0.6290189,-0.51680666,-0.6019538,-1.1488315,0.6561611,-0.9112413,0.5100694,-0.010201941,-0.41575626,-0.99573976,0.51400137,-0.13324496,0.6003735,0.90858644,-1.756543,-0.56579715,-1.2166046,-0.4743553,0.6017386,0.7139283,0.6312486,-0.22448407,0.010416127,1.5952412,-0.31673738,-0.08042968,-1.1402948,-0.39998963,-0.92721313,-1.572918,-0.34202433,1.4257507,-0.44051456,1.3233937,0.4287483,-1.2166924,-0.44838187,0.6182652,-0.20860778,0.8979676,-0.8103249,1.3890587,-1.2660643,0.34599945,0.12778881,1.0039934,0.5634353,0.49836227,-0.8949199
+0,0,7,0.41149214,-0.7849847,-0.28549865,0.4135578,-1.5136797,0.28504756,-0.2592171,0.3586638,0.0096122995,0.68446684,0.27280936,1.1332684,-0.9290369,1.6734341,0.96415323,1.5616682,0.6196785,-0.47138083,-1.7291825,0.6455711,-0.30171967,0.43178305,0.0013747738,-0.076876126,-0.26705164,0.09099751,-1.0307256,0.8569853,0.077619076,0.80527884,-0.3927059,-0.48262763,0.77014446,-1.2891862,-0.6874516,-0.18790592,1.2064557,-0.4788146,1.7160684,-0.37874106,-1.8563964,0.94303226,0.9052338,-0.8621848,-0.46681213,-2.1950614,-0.69286275,0.12579149,0.017325861,-1.8529755,-1.4810988,-1.7104719,-0.908732,1.3428756,1.6590891,1.4576144,-2.0680587,2.074148,0.1254711,-1.3453596,-0.150765,-0.31348786,-0.03713879,0.7493652,-0.09301235,-0.70159227,0.36488333,-1.3442415,-0.74384296,-1.0073787,0.5603107,1.5509155,-1.1920458,0.18350546,-0.26176476,-0.46722165,-0.17439422,0.7131231,0.48848873,-0.995511,0.1133322,-1.4447079,-0.022810481,-0.91964275,0.9241803,-0.87918293,0.80785894,-0.8586497,0.55474204,-0.09638581,-0.28760654,0.24903235,0.24263471,-1.3331411,2.1252306,-1.6045873,0.092501745,1.4959875,-0.7847978,-0.07687911,-1.3790759,-1.2054987,-0.4129643,1.175199,1.6125137,-0.15670183,-0.0052074674,-2.0629342,0.41843984,1.343122,1.3168796,0.4892252,0.8831178,-0.56325597,-0.40071782,-0.2586007,0.8234577,0.3491471,-2.5417411,1.2765507,1.3083528,0.13020971,0.56978697,-1.4101067,-0.31448993,-0.09979597,0.82165474,0.16443604,1.0280895,-0.22518575,-0.7565461,0.22468717,-1.430905,0.5723812,1.408025,-0.66557765,-1.2709581,-0.47722962,-1.0509661,2.1456828,0.35895014,-0.3888584,-2.3072958,0.28834087,0.9454994,1.2007508,0.13654324,-1.0512546,1.1556263,0.28248435,1.1180621,-2.4063637,-0.049587123,-0.4283151,0.34162042,1.1140897,0.74316895,-0.07375317,1.7339555,0.06241791,-0.14042361,-2.4541376,0.76050717,1.1560489,-0.81901497,-1.2859453,-0.049847722,1.0361446,-1.1722547,1.2732593,0.5962597,-0.30993164,-0.031384904,0.20975727,-0.23721394,0.710139,-1.6749394,-0.05069577,1.013451,-0.03550218,0.873963,-1.3086079,-0.043819394,-0.0068593933,0.054935377,2.020619,0.013108201,-1.2887985,-0.85200655,0.054864895,-0.44489822,-1.6583155,-0.63917583,1.483576,-0.23919883,1.4165703,-0.97365326,-1.2111417,0.5758731,0.2508152
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
index db25681f..3e39464b 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,7,0.8438418,0.1794473,-1.0610754,0.49876413,-0.18509679,1.2684177,0.7748202,0.42714408,-0.32632595,0.30871361,1.0232624,-0.531747,0.6962955,0.31235534,-0.97006154,0.4374414,-0.6124084,0.8816563,0.69690573,1.8744665,0.8528751,-1.2626123,1.3722603,2.5256085,-0.5358826,-1.2145865,-1.0444467,1.2561449,-0.30111277,-1.5428085,-0.7596274,0.93373644,1.1946712,0.8260702,0.21434239,0.16956076,-1.6884155,0.5569528,1.8079216,0.4667857,-2.5842013,-1.2288795,-0.69257486,-0.3466202,1.6240005,-0.17097999,-0.021782674,-0.7709626,0.14133824,1.3214102,-1.152685,-1.4690636,1.647548,0.30679667,-1.4932691,1.687439,0.48392302,-1.2188569,-0.082943715,-1.1727151,-0.24080916,-0.48152798,-0.54473346,1.1660991,1.0062953,-0.45552832,-0.66909534,-1.3224851,1.9710338,0.89391387,-0.83821976,-1.2250444,-0.9232601,-0.89614505,-0.1337835,-0.1138566,-0.741236,1.4156529,-0.5768202,-0.19710043,1.0930521,0.9971356,0.41112706,-2.1113818,-0.41081738,-1.1549268,-0.37958524,0.0063747503,-1.3508683,0.07076833,-0.3163995,0.5410095,-0.72964126,0.90011233,-0.29304907,1.7490932,1.6180183,-0.04848024,1.211592,-0.6006584,0.61566734,0.31913698,0.67258716,2.4460082,0.35291326,-1.0352191,-1.1832807,0.51288897,-2.5279958,-1.15038,-0.41004008,-0.3320385,0.3593691,-0.031581197,1.248297,0.87027967,0.4268465,1.9122381,-1.6881752,1.4497418,0.5821907,0.3530083,0.07751833,-0.87054706,-0.10214504,0.92046994,1.1150131,-0.62742823,-0.60133314,0.70472723,-1.2772338,-0.63218576,-0.8797032,-2.7727962,1.1135479,1.0888294,-1.2965524,0.4657736,1.8480299,-0.12805218,0.7858551,0.11978366,-0.9293707,1.4585854,1.1101878,-0.82679045,-0.4033856,-1.705105,-0.119805895,0.92903,-0.5818992,0.87437165,0.041271035,0.010319752,-1.7818887,0.45985395,-0.9583002,1.2992034,-0.17113139,-0.28721648,-1.5636964,-0.015204845,0.24306999,0.60780245,0.8181639,-1.1685832,0.26470146,-0.7132485,-0.5549229,0.40392053,0.9555501,0.25418147,-0.77363104,0.24352336,1.8109292,0.20031126,0.13125075,-1.3413204,0.03766993,-1.3070565,-1.0881963,-0.6965358,1.3141092,-0.7294182,0.76772827,0.016266245,-1.7116737,-1.1352007,0.61998254,-0.58296555,0.80039984,-0.14008771,0.9584638,-0.6666944,0.15143473,-0.16964145,1.2095227,0.10272378,0.3348041,-0.2182458
+1,0,7,0.42257375,-0.17903732,-0.20059693,0.29945794,-1.3389122,0.03707557,-0.299859,0.40720436,0.15333666,0.64613265,-0.47270918,0.8683621,-1.0405228,1.8252618,1.1951929,0.98608077,0.6486402,-0.27345127,-1.749177,0.3916777,-0.06327997,0.5837565,-0.029801903,-0.4157889,-0.8207295,0.56711143,-0.9630651,0.7541865,0.040734746,0.6285889,-0.0074708797,-0.26125398,0.5846216,-1.6676134,-0.6075296,0.39818704,1.1361198,-0.14915793,1.784923,-0.2994867,-1.1560175,1.0586625,0.8207005,-1.4820725,-0.59732884,-2.1600444,-1.0302081,-0.14026557,-0.16139339,-1.6454352,-2.011675,-1.9869673,-0.7221869,1.3151264,1.5895944,1.3244375,-1.8021666,1.6869603,0.08559678,-1.2588723,-0.07605097,0.048324257,-0.35193312,0.9454642,0.104382835,-0.9171362,0.35889798,-0.6452473,-0.6342578,-1.0011637,0.2077998,1.4437941,-0.9567251,0.14908887,-0.24294905,-0.687203,0.05336324,1.0594568,-0.07940463,-1.1028723,-0.08218197,-1.3812248,0.22032668,-0.9392466,1.3229952,-1.090876,1.0943706,-0.5451269,0.12716077,0.06487062,0.0940708,0.37863347,-0.3223036,-1.1197766,1.7777491,-1.5668588,0.25907642,1.4157779,-0.7196445,-0.38355565,-1.5430949,-1.3738754,-0.7312654,1.3590227,1.652726,-0.31939238,-0.023165464,-2.2086487,-0.1569172,1.3741649,0.9475051,0.48269206,0.6875285,-0.534931,-0.24520622,0.24692726,1.1467029,-0.025945317,-2.077082,1.548747,1.3998401,-0.09173701,0.60546964,-1.7211899,-0.32520935,-0.6385531,0.58607954,0.63349676,1.2241516,-0.115974545,-1.2130051,0.3455885,-1.230523,0.3379228,1.3327022,-0.7980125,-1.4078081,-0.5755934,-1.0106425,1.8490473,0.6007134,-0.49850124,-2.4873343,0.3709469,1.0519376,1.3123273,-0.014704171,-1.0918032,0.4931872,0.14534308,1.087156,-2.4626598,0.4272686,-0.5852703,0.11401402,1.3979368,0.96870464,0.11613599,2.0814247,0.08643781,-0.25604638,-2.6740973,0.7021289,1.4160581,-0.9518831,-0.9351058,0.3818723,0.64308923,-1.4009758,0.878512,0.3460209,-0.6776031,0.16450398,-0.31286997,-0.18456456,0.3714644,-1.7571229,0.3403873,1.3923695,-0.17252812,0.6139845,-1.0345956,-0.1693556,-0.044252478,-0.30749786,1.657718,0.78084534,-1.4596986,-1.1058352,0.1094069,-0.053861905,-1.9468046,-0.2616004,1.6512417,-0.49316752,0.82700384,-0.6377279,-0.76203316,0.432102,0.4528831
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
index 5284467d..891aca89 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,7,0.52204627,-0.038092855,-0.82732695,0.52709055,0.06611072,1.0820338,0.6271256,-0.7258145,-1.0272309,0.21715637,1.1717758,-0.38762644,0.54010385,-0.33218145,-1.6269217,0.23357184,-0.8224511,0.7772591,1.2175651,1.905813,1.5552422,-0.47752312,1.6528003,2.6848164,-0.7099252,-0.78975856,-0.9663611,0.39896625,-0.5068774,-0.76180744,-0.22029436,0.49096388,1.3501009,0.4508182,0.6573443,-0.2667651,-1.7489359,0.6311765,2.0538185,-0.38108113,-2.1136858,-1.5755099,-1.4845366,-0.056731053,1.4559208,-0.87732655,0.11649373,-0.31365696,-0.968069,1.8460219,-1.2417718,-0.44821975,1.495183,0.39649376,-1.0719526,1.417287,1.2832841,-0.7492332,-0.40597665,-1.1388721,0.5925744,-0.65348125,-0.08980144,1.4410766,1.9072348,-0.4337984,0.2681212,-0.8927501,1.7772413,0.36656567,-0.4058805,-1.738713,-1.3288599,-0.72347444,-0.014828365,-0.124502756,-0.84998757,0.30157983,0.49887928,-0.1615823,0.41461873,0.18497966,0.69440186,-2.075491,-0.8384113,-1.0419928,-0.9347767,-0.67896277,-0.69964796,-0.4201842,-0.70845145,0.9878649,-1.3933247,1.8833025,0.6312973,1.695937,1.71965,0.10518575,0.3091634,-1.6875365,0.79602194,-0.057294976,0.88798237,2.0516448,0.08196111,-0.44834742,-2.015611,0.43115285,-1.5257589,-1.0001528,-0.7601532,-0.86637354,0.5525021,0.10951839,1.535792,1.4130329,0.1467389,1.818903,-1.1924548,0.6846707,0.49561536,0.82903117,-0.45645127,-0.9932831,-0.903748,0.8548094,1.3870052,-0.24820147,-0.2833696,0.53783894,-0.29969555,-0.42618462,-0.42249247,-2.6975384,0.9544928,0.89196867,-0.7268504,-0.3831172,0.40872112,-0.5766559,0.8267256,0.44900906,-0.20187785,1.2164347,1.406283,-1.4030226,0.24202895,-1.3471564,0.38567412,-0.14396489,-0.02556357,0.8743769,-0.22310354,-0.13754714,-1.4840792,0.9863184,-0.67454904,0.81428856,0.57713085,-0.9858999,-0.8832908,0.30878907,-0.0063148057,0.31424287,1.0139408,-1.8677388,-0.78900766,-1.0641052,-0.9655488,0.4306992,1.3903565,0.5520786,-0.39177105,-0.4996559,1.8079237,-0.34591374,-0.5728688,-1.5090452,-0.07140037,0.038317777,-1.3744234,-0.78208166,0.9468176,-1.0600139,1.9171584,-0.02225734,-1.8697608,-0.7089237,1.139994,-0.06938327,0.06805763,-0.57662845,0.54992193,-0.04600769,-0.17940713,-0.19065502,1.2973483,0.87212723,-0.09027165,0.08437479
+2,0,7,0.34137836,-0.54560816,-0.34839255,0.5934387,-1.4561145,-0.3552921,1.236413,0.2230994,0.52624667,0.35157225,1.0009108,1.453514,-0.99844396,0.52108085,0.99309534,0.9884845,-0.49107695,-0.44872406,-2.3321922,1.2099328,0.02559154,-0.15125763,-0.25792414,1.148591,0.34748212,-0.7123179,-0.9918085,0.72194755,-0.57125664,-0.5386358,-1.4469051,-1.0329736,0.2649161,-1.8303887,-0.22947203,-0.22510229,0.4087416,-0.6991921,0.27918908,0.2699201,-2.0308623,0.73651797,1.1014013,0.28801918,-0.043545906,-1.3224466,-0.728169,1.3165932,-0.38451204,-1.2290868,-0.3865898,-0.82043016,-0.15159693,0.95450556,0.720509,0.58943474,-0.96818477,2.4339142,-0.045729626,-2.2731185,-0.69900125,-0.72985935,-1.0451107,0.16168666,1.275655,0.0921365,-1.3980165,-1.8175794,-1.2656915,-1.2789646,0.65161574,2.2280807,-1.0717951,0.7397777,-0.35529038,0.13295835,-1.1914927,0.17240041,1.2711916,-1.2801683,-0.7685431,-0.809741,-0.24176663,-0.520219,1.2797531,-1.700323,0.7044165,-0.28107253,-0.3459173,-0.61043745,0.52076197,0.051599972,-0.45293874,-0.06818946,0.8897772,0.097269796,0.41846913,1.6724628,-0.749013,0.34057495,-0.7477783,-0.06767678,-1.2013714,0.9092441,1.5504543,0.3554491,1.1866004,-1.1053133,0.1902895,1.1757743,1.115244,0.25665316,0.61796343,-0.47255665,-1.0932264,-1.2121068,0.3602891,1.3848262,-2.0418153,-0.25206277,0.64594483,0.06378719,0.63946444,-1.0031189,0.08267042,0.7659137,0.8319407,-0.17579815,0.7439447,-1.1327,-0.6398448,0.55752283,-1.4105185,0.6037283,0.882441,0.2301522,0.016749382,-0.097922415,-1.7987541,-0.056111883,0.7459678,-0.4149461,-1.682275,0.67691326,-0.4379247,0.76354676,1.4371464,-0.36294943,2.097733,0.2694331,1.5530957,-1.4040655,-0.9152303,-0.5327291,0.41197965,1.040547,-1.1637155,-0.59128654,1.9143434,0.75515777,-0.5771012,-1.1852001,-0.2920808,0.5541353,-1.050848,-0.79349184,0.107663386,0.108831875,0.8791816,2.1749172,-0.9475491,-0.28542468,-0.5886931,0.441553,-0.010180896,0.54911983,-0.48679575,0.46939653,0.29680255,-0.33886543,1.8130943,-1.2549541,-1.0724515,0.56288785,-0.12585218,2.9909089,-1.5593657,-0.92594236,-0.9341071,-0.14739718,-1.5810376,-0.5024728,-1.707612,2.2963934,1.5836017,2.0804906,-0.6040425,0.7326207,0.4728579,-0.083124615
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
index b176d111..d8417f89 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,7,0.5811147,-0.032513455,-0.47977856,0.13629222,-0.26297802,1.0946103,0.70547986,0.2546013,-0.5609061,-1.0476835,0.25876504,-0.47654045,0.6124765,-0.62587225,-2.3766463,-0.87987435,-1.1483063,-0.08349529,0.058731697,2.3143327,1.4411461,-0.7700864,1.6809956,2.562706,-1.1312455,-1.919516,-2.5979652,0.97807324,-0.58193874,-0.4056123,0.100295246,0.10704158,0.4826064,-0.3876764,1.1518319,0.3071064,-1.0756433,1.348694,1.2538676,0.8821212,-3.0179098,-1.1350209,-0.6724623,-0.84317905,2.1001053,-0.8851265,-0.45266476,-0.15717253,0.020399736,0.5467559,-1.436141,-1.0484979,1.1574883,0.36959058,-1.3153619,1.9301816,0.9880461,-0.8081846,0.796931,-0.7006053,-0.043729685,0.29161018,-0.17788444,2.3002772,1.6774901,-0.9066579,-0.12004822,-0.7499011,0.9292784,0.55876046,-1.5865698,-1.2220012,-1.0178283,-0.6064322,-0.17501302,-0.52628666,0.8076905,1.1616044,-0.3377207,-0.054776378,-0.07740326,1.052842,-0.1103256,-1.4683763,-0.9702077,-0.23154487,0.37362558,-1.6633947,-1.0713861,-0.2937544,-0.40128285,1.0818906,-1.1223569,1.1332817,-0.7912049,0.80717605,1.2827631,-0.46637723,0.97363174,-0.8316934,1.044962,0.06701209,0.36239144,1.3466328,0.17860873,-0.8557746,-1.2067758,-0.083446786,-2.647533,-1.4590422,0.3115786,-1.1877053,0.34800985,0.0067028324,1.0329384,0.6848961,0.7815907,0.70181257,-0.31994092,1.1682042,1.5868865,-0.1021363,0.20299423,-0.18094988,0.058076646,0.35854048,1.38783,0.17383336,-0.66786546,0.53052557,-0.8849052,-0.264605,-0.46346378,-2.4525568,0.87664086,1.8091627,-0.887404,-0.5623806,1.4447385,-0.614229,1.0334789,0.23856685,-0.57881784,0.8717396,1.1144764,-0.84959036,-0.7336873,-0.6120007,0.49778894,0.20692903,-0.1316462,0.57095027,0.07802366,-0.045190286,-1.3658072,0.44326764,-0.95905554,0.6394518,-0.5813267,-1.3264008,-0.718985,0.1745243,0.75313646,-0.32120827,1.118506,-0.969537,-0.62831134,-1.4774725,-0.9432864,1.0576622,1.6772305,-0.6164494,-0.5101229,-0.8483868,2.2736018,0.5569865,0.12856804,-1.1824113,0.72462106,-1.1923324,-0.3833196,-0.8572979,1.0559568,-1.1587551,0.47620925,0.71898955,-1.4884728,-1.4193738,1.1060739,0.34149924,-0.081969894,0.7028209,0.4915076,-0.47154474,1.1152692,0.46461847,0.74368006,-0.28396526,0.040208068,0.38362354
-4,0,7,-0.40144852,0.4484332,-1.9290444,0.67508656,-0.35645357,0.58294165,1.0673926,0.32582113,-1.0179955,-0.5669628,1.6901405,-0.597517,0.57684857,0.6212512,-1.3812616,-0.0031751008,-1.4166527,0.31889653,0.8719765,0.92417514,1.434082,-0.9677904,1.0067217,2.4912894,-0.5054417,-0.091208264,-1.6230117,0.6646005,0.14070033,-0.4588424,-2.0376508,0.20500316,0.6994927,0.4712299,-0.017884562,-0.041826397,-1.9103551,-0.8554847,0.9786731,0.5541552,-2.017602,-0.60309106,-0.12291661,-0.078420684,1.9937975,-1.4203902,-0.6049179,-0.4561434,-1.3776398,0.7715638,-1.5148757,-1.4664235,1.4948455,0.59697,-1.5645903,1.4915609,-0.16028762,-0.21107581,-0.6521816,-1.9952201,-0.12593468,-0.0323758,-0.73676616,1.6785882,0.386853,0.30982378,-0.23634642,-0.7451515,1.9966604,0.29396093,-0.4427039,-1.1439474,-1.4322784,-0.86600333,-0.47581324,-0.32586938,-0.4229079,1.7753755,-0.29973793,-0.80312866,0.83222604,0.6190816,0.8152262,-2.184811,-0.36325434,-0.44522017,-0.8133582,0.14973828,-0.54611415,-0.26411292,-1.6611116,0.8977264,-1.8963674,0.6349306,-0.06896561,0.85603106,1.2678437,0.65789634,1.2733372,-1.8506087,0.80270076,0.7760959,0.14549729,1.809624,-0.35460156,-1.1367061,-1.11971,0.44256684,-1.9340838,-1.3446312,-0.52430373,-0.6890866,0.25105867,-0.16935606,1.5000091,1.6533798,0.40652713,1.940414,-0.7145079,0.7842815,0.7231326,0.6299073,0.20868257,-0.022947172,0.0590606,0.015217416,-0.14524586,0.5871516,-0.20919175,0.8956564,-1.2955273,-0.28790218,0.23210756,-2.3776908,1.9674141,0.8105194,-0.7433918,-0.15029533,1.3930359,0.55439985,1.0334952,-0.16596296,-0.87932885,1.2013986,1.6930834,-1.1102449,-0.4110713,-0.88066155,0.48112375,1.262018,0.5744189,-0.23829386,-0.17633875,0.25621903,-1.2790095,0.26927826,-0.4251639,1.4009678,-0.68966246,-0.57853395,-1.9202563,0.73328453,-0.1512125,0.99056023,0.7747929,-1.3844789,-0.92885804,-1.2573935,-0.69143796,0.6731172,0.8143503,0.11032579,-0.865953,0.15180536,0.2987957,-0.6535276,0.73512334,-1.3143648,0.45279935,-0.48555213,-0.52440286,-1.1472527,1.0770756,0.40460247,0.71095395,0.8498973,-1.0886842,-0.76816595,2.1436665,-1.6361245,1.3135405,-0.39040008,0.43102646,-1.0228053,0.79288054,-0.43767598,-0.025315946,0.5831925,-0.122284114,-0.5308977
+3,0,7,0.13949223,-0.8288385,-0.011691158,0.6978597,-1.7423092,-0.41049773,0.4738875,0.43795598,-0.13211526,0.83174855,0.18503393,0.8696095,-0.98063534,1.320197,1.2542927,0.804681,0.4159498,-0.5308098,-2.063883,0.7713375,-0.040647052,0.50756997,-0.47521272,-0.050390225,-0.41869515,0.15082894,-0.91570437,1.1039559,-0.119457014,0.630291,-0.78385514,-0.26605347,0.30441046,-1.7381796,-0.3145166,-0.09518902,0.98777574,-0.5672071,1.5451479,-0.5920584,-1.8862137,0.73404765,0.76614827,-1.2776103,-0.5546785,-1.6537995,-1.1388252,0.5387951,-0.04569923,-1.8089854,-1.3872888,-2.010136,-0.32747167,1.414386,1.5741073,1.1150281,-1.6234998,2.3318863,0.09676073,-1.2265372,-0.32222414,0.23237921,-0.34041348,0.7922307,0.6937952,-0.47161567,0.18344484,-1.5619589,-1.089482,-1.4129832,0.28963518,1.6352048,-1.1390606,0.18842995,-0.38225302,-0.65850925,-0.33246985,0.6151895,0.6373732,-1.5749953,-0.327212,-1.2576468,0.2955655,-0.5931339,1.2506938,-1.0565152,0.96414435,-0.8558795,0.34775218,-0.5049455,0.26373684,-0.20436549,-0.080223046,-0.94886553,1.4435066,-0.92269295,0.47257257,1.7526731,-0.9398457,-0.03447294,-1.2984382,-1.0528736,-1.0664232,0.9039165,1.928867,-0.4283627,0.48781595,-1.5944821,0.19360305,1.8330988,1.3015158,-0.05350088,0.53769165,-0.29242146,-0.8438974,-0.069205664,0.82746214,0.84352636,-1.9790401,0.94773203,1.0971557,0.038646087,0.88122386,-1.4750324,-0.11020492,-0.09284775,0.32929987,-0.14679804,1.1929734,-0.6901023,-0.6029133,0.21264303,-1.3289111,0.8272518,1.2592579,-0.38540855,-1.034611,-0.0327922,-1.550678,1.1788012,0.53695285,-0.36779112,-1.9978998,0.6491437,0.62098545,1.3861,0.43899333,-1.0762482,1.1092504,0.23858577,1.405184,-2.151006,-0.51262474,-0.6709533,0.47459316,1.2390981,0.41524518,0.045998752,2.0420544,0.5617183,-0.0899196,-2.7423296,-0.106221266,1.0869207,-0.7299839,-1.148185,0.08723943,0.6347005,-0.4051242,1.4006757,0.099802785,-0.47955215,-0.47176704,0.27028126,-0.8404873,0.59273887,-1.5649644,0.14975634,0.9384124,-0.36595413,1.3495317,-1.2204468,-0.33850634,0.27997792,-0.07943602,2.6751914,-0.09228743,-1.481454,-0.97814053,0.20486094,-0.731338,-1.5517998,-0.83328533,2.0616615,-0.18377517,1.3525189,-0.7911948,-1.0217361,0.5704211,0.053546384
+4,0,7,1.1079531,0.56694883,-0.51460046,-0.44087544,-0.43843356,0.25955358,0.13465126,-0.19686213,-0.0069119474,1.2750784,-0.36285907,0.29798138,0.2409469,1.3927422,0.2650472,0.2537195,-0.099152684,1.1273077,-1.4502947,-0.10494784,-1.1681588,1.3679986,0.028042292,-0.14520136,-0.9369158,1.5287392,-0.46746886,1.6174921,0.15947264,1.0661743,0.06785149,0.16602738,0.6872166,-0.90826625,-1.1785315,0.49854565,-0.70833915,-0.57204914,2.0476813,-1.371946,-1.3191279,0.64045453,1.1497461,-1.2700158,-0.91564953,-3.0730312,-0.0084550865,1.1021385,-0.20811193,-0.9027882,-1.3729584,-2.1781166,-0.82236534,0.14685307,0.39046445,-0.108870916,-0.63499683,1.680812,1.4627557,0.48326835,-0.4139974,0.048960343,0.16166344,1.8954678,-0.5226412,0.20060515,-0.04435112,-0.754301,-0.71545064,-1.4038323,0.069040895,1.1398026,-1.154028,-0.43103975,0.54638636,-0.74899,-0.14519033,0.9119781,0.7352537,-0.1921951,-0.4378232,-1.6286504,1.354772,-1.0531448,1.8048595,-1.1451042,0.80384815,0.22715564,-0.24490066,0.5841813,-1.1601851,0.73505855,-0.39013195,-0.18186606,1.2028245,-1.022498,1.2703432,2.3782692,-0.9626023,0.44522825,-1.4234884,-1.6156629,-0.5665078,0.95502675,1.4593245,0.47318575,-0.8122251,-1.0151421,-0.82546866,0.6082922,1.1691772,-0.5170128,0.067346595,-0.6556871,-0.5575009,-0.0006836925,2.0677729,-0.63626707,-1.0719508,0.08163928,1.2308236,0.3763118,0.33964872,-1.3674697,0.8556785,-0.69404805,0.6134989,1.0646461,0.9832764,-0.34057733,-0.12112485,0.23851065,-0.026012735,0.8402205,1.8853765,0.90379107,-1.7303153,-0.17408463,-0.7569175,1.1544394,0.10548308,-0.58828104,-2.2133024,0.76151407,0.9882251,-0.33799013,-0.36300513,-0.32815427,-0.16083471,-0.7817147,-0.15756014,-2.7104619,0.73015183,-1.8245586,0.12698413,2.2169824,0.95323753,0.43615663,1.2318543,0.45343706,-0.06706062,-1.7847573,1.080589,2.519145,-0.86268,-0.17361444,1.168513,-0.91562283,-1.569328,1.259442,0.037228383,0.30334857,-0.23501681,-0.49459893,0.3947516,-0.39016256,-0.7743366,0.74605846,0.21725878,1.8476028,0.54618216,-1.5716995,0.9664981,0.12692592,0.3307412,1.9256802,1.494507,-1.164124,-0.9523818,-0.61173975,0.45403442,-1.0929524,-0.046263948,1.1769654,-0.009714122,-0.049784996,-0.15999958,-0.46489516,0.8143899,-0.30355412
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
index 45d59f99..5144f594 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,7,0.4836214,0.15431057,-0.8040745,0.29091778,-0.055900697,1.2060295,0.3452758,-0.8614426,-0.9362209,0.34898758,1.1721082,-0.4112888,0.3945304,0.2336385,-1.4738886,0.73230135,-0.5247049,0.6479498,1.5893649,1.6185886,1.3388404,-0.48648638,1.4070816,2.6896753,-0.5802976,-0.7131316,-0.4670565,0.63593763,-0.47043854,-0.74653304,-0.44683897,0.52990055,1.5743597,0.62231994,0.36112452,-0.41805673,-1.6092434,0.117180616,2.251997,-0.4125149,-1.6764935,-1.4553155,-1.3477093,0.588175,1.2021706,-0.8264717,0.13591708,-0.5834153,-0.6440799,1.8806185,-1.564533,-0.5116288,1.3243208,0.3408409,-0.9232263,1.1233313,1.1755381,-0.87264514,-0.381421,-1.2161983,0.029331272,-0.6922146,-0.08794284,1.609424,1.6640897,-0.5114714,-0.031018924,-0.7680615,1.8714914,0.6504967,-0.3274704,-1.8043759,-1.1517804,-0.81753194,-0.16838285,-0.15660566,-1.4273182,0.27939737,0.25196457,-0.15733546,0.599071,0.2938215,0.43453452,-2.450483,-0.6304489,-1.01529,-0.91700906,-0.041178755,-0.8448184,-0.22622812,-0.7264896,0.9366097,-1.4195501,2.0062637,0.6819727,1.9392235,1.7529082,0.1502222,0.32664576,-1.3147435,0.55803806,0.20468958,0.9131472,2.2310262,-0.1215383,-0.5964191,-2.0484083,0.43182093,-1.5355808,-1.2749287,-0.9757672,-0.9061006,0.8016716,0.12147521,1.4375226,1.4464892,0.098983355,1.8112254,-1.96104,0.54458094,0.40790206,0.6800015,-0.55492187,-1.3341256,-0.56690305,1.023294,1.0548493,-0.70494395,-0.26114026,0.37238428,-0.5541062,-0.5526852,-0.5225235,-2.4509475,1.124607,0.48349783,-0.8573671,0.11351146,0.65686464,-0.6089635,0.48230767,0.38274416,-0.42615348,1.0773033,1.5329567,-1.329324,0.10873803,-1.7785066,0.26635838,0.39991972,-0.10376236,0.7763999,-0.4305313,-0.23850553,-1.2875695,0.73188037,-0.58677316,0.9893427,0.1941753,-0.56284344,-1.1522082,0.4671607,0.093474194,1.139886,1.052885,-1.7801507,-0.3854622,-0.67583334,-0.81050724,0.4004889,0.7293054,0.6006033,-0.7552659,-0.325302,1.7445861,-0.48890847,-0.58957654,-1.4742821,-0.15116952,-0.10655272,-1.2754749,-0.5283034,1.1386613,-0.8321361,1.7763195,-0.21586367,-1.6490741,-0.57824636,0.8866336,-0.26279575,0.35519692,-1.1262823,1.2348262,-0.02918719,-0.45923397,-0.7515425,1.6313766,1.0427995,-0.014255836,0.1959896
+5,0,7,0.30457106,-1.0499812,-0.2708557,0.6033656,-2.1559174,-0.014660421,0.40023395,0.35113475,0.0902287,0.7888887,-0.34383535,0.93346864,-1.0638552,1.6174538,1.5091573,0.9228686,0.086696744,-0.50717336,-2.0618286,0.39436528,-0.021834815,0.6010914,-0.47710696,0.17810024,-0.91974175,-0.12123473,-1.365223,0.79932564,0.19411209,1.1538398,-0.63365114,-0.39145467,0.2746975,-1.130279,-0.6916063,-0.12119007,1.0064207,-0.330205,1.6648016,-0.45908698,-1.8615019,1.1402794,0.75518245,-1.5022097,-0.9118628,-2.2375493,-0.902504,0.6392769,-0.21022926,-1.8474433,-1.5057714,-1.960203,-0.3685194,1.2326003,1.5652041,1.0757546,-1.4903946,2.2756448,-0.32394972,-1.1672122,-0.059360582,-0.04470684,-0.8703109,1.0302415,-0.0895821,-0.403575,-0.37002468,-0.8056304,-1.1433159,-1.2743354,0.85718685,1.1835376,-0.6666402,-0.14671929,-0.04834712,-0.98218584,-0.13421746,0.78738457,0.38280028,-1.2351733,-0.48965338,-1.6121022,0.42098215,-0.85432166,0.9951741,-0.12262745,0.9710562,-0.5171932,0.16173087,0.17513205,-0.07830988,-0.34751597,-0.012498501,-0.66930187,1.9089013,-1.4641144,0.5140048,1.7203206,-0.9557444,0.13077603,-1.1026406,-0.9473788,-0.31690073,0.97551376,1.9448526,-0.476878,0.7957327,-1.8465495,0.67249143,1.7720664,1.2585933,0.45854238,0.5997015,-0.6044482,-0.76524645,-0.30329454,0.5679286,0.5396677,-1.7696911,1.5391197,1.026936,-0.31885958,0.50461066,-1.7645643,-0.50854623,-0.4742418,0.77532613,0.09777384,0.54422975,-0.1796674,-0.8478994,0.0031664053,-1.1256933,0.77754414,1.3871409,-0.26337987,-0.8610575,-0.51069754,-1.2064705,1.7107098,0.7312288,-0.28927425,-2.1887672,0.9429601,0.9586188,1.8643379,0.19833568,-1.2076133,1.2674617,0.3626933,1.0161105,-2.1542332,0.07333201,-0.60430324,0.2051994,0.90605885,0.19795726,0.60270673,2.1178296,0.3766683,-0.5104417,-2.3974814,0.124425165,1.0825484,-0.41810575,-1.0645667,0.04239827,0.48415333,-1.4814224,1.1013521,-0.101820946,-0.6659215,-0.4296166,0.30863544,0.20719278,0.43899494,-1.6712543,0.00065493764,0.7002628,-0.41546488,1.3948044,-1.0820502,-0.11341275,-0.15781793,-0.2200905,1.9745494,-0.07587735,-1.6396667,-0.65765786,-0.13210705,-0.30101702,-1.8123761,-0.7385888,1.5914155,-0.37091118,0.9381928,-1.189511,-0.54202545,0.30190322,-0.34896195
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
index 9a30653c..9f0978c3 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,7,0.6047183,0.61474156,-0.85807425,0.1360568,-0.47004148,1.3154823,0.85255265,-0.2758797,-0.94348884,-0.46887586,1.2464497,-0.17505822,0.22009045,-0.5281095,-1.6458,-0.2863077,-0.7895708,-0.032991536,0.6482207,1.1479981,1.6235281,-0.8318479,1.2820829,2.5700314,-0.74509114,-0.956326,-1.8435808,0.50691956,-0.42929974,-0.82176375,-0.688462,-0.16810319,1.0948266,0.04102623,1.1152531,-0.29527077,-1.4508649,0.15333177,1.8691535,0.65497345,-2.3797927,-1.5487301,-0.8639569,-0.12066294,2.1302707,-1.1985978,-0.4929019,-0.9896101,-0.38555038,1.4556348,-1.5805736,-1.253294,1.9447695,0.7381123,-0.87177855,1.6145153,0.29667908,-0.54530835,-0.6368157,-1.3141624,-0.17360605,0.09582736,-0.50669295,1.6515435,1.0648893,-0.32638404,-0.30910656,-1.0866282,1.7825578,0.1281249,-1.2656724,-1.9909875,-0.67408943,-0.47530064,-0.6126794,-0.78917897,-0.7749392,0.95389134,-0.13919069,0.37404194,1.517078,0.837784,0.7876206,-2.7127264,-0.9797122,0.22331813,-0.5448298,-0.37922773,-0.85313463,0.29169962,-0.72657126,0.6292672,-1.6658629,1.0823836,-0.3303371,1.5071089,1.7243761,0.09441604,0.78559613,-0.8699731,0.83902586,0.43403652,0.27791783,1.5322093,0.39030224,-1.0829405,-1.1652853,0.4393425,-2.4266343,-1.1654254,-0.8020692,-1.2343247,-0.02799856,0.082993135,1.4820228,0.81963,0.81295073,1.5777001,-0.74838066,1.0510573,1.142717,0.67160773,-0.22292933,-0.5321122,0.073675826,0.8971168,0.5200968,0.2786284,-0.6427973,0.5132947,-1.2613649,-1.1026112,-0.2773738,-2.553699,1.4465657,1.2948054,-0.9990521,-0.10682497,1.5314655,0.27273822,1.0103928,0.21165001,-0.5424684,1.0398483,1.2868812,-0.9073566,-0.4446297,-0.86511356,0.28190532,1.1098244,-0.050390765,0.50629246,-0.4595085,-0.39918143,-1.5425367,0.8516377,-0.5174903,0.82666886,-0.24608938,-0.55887544,-1.367161,0.31737787,0.38218778,0.6614556,0.62725943,-1.8203161,-0.7528492,-2.0896697,-0.8917504,0.34349948,1.1796234,0.4355885,0.14362016,-0.1739353,1.5831403,-0.103192456,-0.020828791,-1.1002008,0.052845396,-0.7443734,-1.1041548,-0.86692,1.290353,-0.6568599,0.8881055,0.80793345,-1.1829001,-0.648108,0.781555,-0.75571537,0.790254,-0.30762485,0.82678217,-1.2168562,0.5863828,0.29487747,0.7162153,0.18510085,0.6786084,-0.55795467
-7,0,7,0.6299754,0.050378714,-1.0953621,0.40204868,-0.20064756,1.2504606,0.64857304,-0.5956721,-1.1048201,0.08307071,1.2121466,-0.74361223,0.40176916,-0.19288543,-1.5207201,0.18478414,-0.75672543,0.6516608,1.2947527,1.4226474,1.5715213,-0.7016897,1.537614,2.4972186,-0.66829467,-0.7722992,-1.0611916,0.84529483,-0.6867521,-0.8948152,-0.83675677,0.5516693,1.4557326,0.597971,0.52972084,-0.3124734,-1.4932187,0.23769754,2.1977735,-0.23013769,-2.2219017,-1.3230104,-1.2486353,-0.076617986,1.3280454,-1.2714481,0.05504124,-0.5108662,-0.7244996,1.7406578,-1.2521456,-1.0216031,1.3594753,0.49166912,-1.0463053,1.4819757,0.8894106,-0.78244185,-0.44918662,-1.2543783,0.06708226,-0.60776013,-0.45414528,1.8482741,1.4049805,-0.51085305,0.007999584,-0.81303686,1.9179218,0.6898083,-0.68716633,-1.6860543,-1.1650369,-1.0256021,-0.27775717,-0.43408686,-1.197075,0.7731409,0.45193902,0.13899827,0.5074849,0.6706757,0.6787472,-2.57881,-0.58425534,-0.8426904,-0.9324049,-0.3311055,-0.9196865,-0.534395,-0.88145584,0.8438232,-1.2858,1.7702929,0.14992192,2.1282425,1.7167776,0.39477825,0.50257206,-1.4573762,0.5586969,0.08780366,0.50719386,1.8912692,-0.06565876,-0.49146792,-1.8407304,0.48491526,-1.819101,-0.97141093,-0.6114753,-0.940766,0.5585653,0.06410881,1.4343958,1.449382,0.06305862,1.8221257,-1.2712216,0.8074854,0.7395423,0.63117987,-0.38569003,-0.9561888,-0.34065723,0.9483605,0.6962596,-0.16962893,-0.6836114,0.4338564,-1.3348165,-0.5754306,-0.37627304,-2.8389518,1.3089486,0.59520245,-0.78578126,0.22991493,1.1825001,-0.4132581,0.46656233,0.17132021,-0.51955664,1.1751152,1.520865,-1.29063,-0.16482846,-1.6479331,0.20740972,0.5557091,0.031369217,0.3343893,-0.26854822,-0.23717512,-1.1977962,0.7109176,-0.8197224,0.85080427,0.07489285,-0.57192963,-1.1184924,0.9720554,0.31332842,0.8337498,1.2144716,-1.7497374,-0.4055033,-0.97871673,-0.52501994,0.294604,1.0116265,0.7438575,-0.32010132,-0.2937276,1.6995173,-0.4956802,-0.16089632,-1.3082268,0.05458243,-0.33528122,-1.3450384,-0.57403374,1.2334722,-0.7604033,1.5351142,-0.18204713,-1.5558845,-0.4553292,1.076027,-0.46304265,0.69380623,-0.6540957,1.0922314,-0.59123945,-0.09509767,-0.0446557,1.2391702,1.0227808,-0.0018399156,-0.16837811
+6,0,7,0.7273449,0.009315487,-0.5659161,0.0539636,-1.1747422,0.11731785,-0.24878183,-0.34799668,-0.39417046,1.5134698,0.180432,0.45988694,-0.48523727,1.8745505,1.0106969,0.83485544,0.46109983,0.18401664,-1.8427206,0.5820249,-0.9933856,0.57772356,-0.27879953,-0.88057303,-0.9495125,0.27947068,-0.98555374,0.65584594,-0.7251935,0.63879484,0.27707863,-0.23204361,0.8994904,-1.6550843,-0.41345245,-0.13837196,0.25872716,0.28248805,1.929113,-0.9793597,-1.2198358,1.0012894,0.7334509,-1.2398807,-0.88356876,-2.096589,-0.55407935,0.06886668,-0.776487,-0.90201944,-2.001808,-2.458255,-1.0559742,1.7568047,1.9512438,1.0935278,-1.6113944,1.89258,1.1544857,-0.70098376,-0.571621,0.7649165,-0.5899035,1.3507531,-0.7255975,0.17241018,0.50311905,-0.6263059,-0.36380523,-0.38990656,-0.27967066,1.416065,-1.308391,-0.352508,0.26046076,-0.18396786,0.50395614,0.2636389,0.86980325,-0.877985,0.18548113,-1.2215562,0.33346045,-0.9510769,1.7036265,-1.5579641,0.6795726,-0.18158613,0.011276826,-0.1970394,-0.26797757,0.3612404,-0.5695245,-1.4026842,1.758898,-1.2169626,0.73240304,1.9216541,-0.5449494,-0.14847618,-1.4680748,-1.6136966,-0.1330753,2.083131,2.207608,-0.20060927,-1.1179813,-1.3244749,-0.3105317,1.1410857,1.0294605,0.12662466,0.8479376,-0.43145418,0.2132454,-0.025244722,1.6760103,-0.2850787,-1.5536851,0.5495985,1.1164563,0.2283902,1.3229661,-1.758217,-0.4295333,-0.5159716,0.14204884,0.31548834,1.3842486,-0.54636705,-0.62329435,0.13254096,-0.53141403,0.2421158,1.3180143,0.27180845,-0.8231482,0.23262133,-1.0528578,1.495885,0.19461124,-1.0369684,-2.0902848,-0.4222444,1.0143939,0.5824076,-0.20892456,-0.4718485,0.06580811,-0.58581954,0.8422832,-2.566362,1.0881084,-0.64552516,0.3792994,2.2656064,1.8036302,0.2566192,1.7502713,0.65841585,0.25055784,-2.5194345,0.84616184,1.1300479,-0.60688585,-0.82479155,0.24982473,0.5697576,-0.6672031,1.3321275,0.23418252,0.32546026,0.52113616,0.41371918,-0.2717294,0.47879305,-1.3640656,-0.10326316,1.3061558,0.70007694,0.5445003,-1.2608218,0.51818657,-0.36812475,0.9051562,1.1091714,0.82054716,-1.0394655,-1.0836576,-0.22977057,0.17767808,-1.7726516,0.3174601,1.4640669,-0.17567706,1.0871913,0.17054728,-1.0595478,0.054115366,0.44907287
+7,0,7,0.19220954,-0.49153456,-0.04641067,0.18849212,-1.5853618,0.12341867,-0.11363969,0.13727675,-0.06847534,0.76124924,-0.28678593,0.8900653,-1.0943688,1.5146927,1.2307183,1.5759242,0.80081123,-1.0552121,-1.5437708,0.8671235,-0.33835396,0.6237686,0.060264308,-0.4205532,-0.569102,0.31872287,-1.1516503,0.56125844,0.025306754,0.6017618,-0.29687238,-0.3951651,0.7847005,-1.504742,-0.2827757,-0.21072185,1.2361257,-0.2929404,1.6530923,-0.54373467,-1.6885573,0.8691418,0.6955893,-1.2405534,-0.5988692,-1.744112,-1.3133477,0.17552854,0.058676194,-1.7747884,-1.4241207,-1.7917409,-0.62875485,1.7578604,1.5654361,1.3629228,-1.6896278,1.6548164,-0.14070357,-1.9311712,-0.15432832,-0.25670385,-0.4815931,0.8380533,0.28672034,-0.9617282,0.51135164,-1.1442822,-0.4330058,-0.7627599,0.6438905,1.7754638,-0.89877915,0.2844743,-0.36010924,-0.62290066,0.05098487,0.43144348,0.37151685,-1.350735,-0.3010244,-1.2662656,-0.017898949,-0.89459026,1.0812829,-1.0964411,0.66727376,-0.4757698,0.45214677,-0.045206342,0.10654671,-0.018967189,-0.5759011,-0.76332796,1.7140834,-1.397187,0.37946275,1.4883934,-0.9071629,-0.39879188,-1.280823,-1.1647131,-1.0890014,1.5997056,1.4125903,-0.36653465,0.25216198,-2.067858,0.7692397,1.5507644,1.0313568,0.47055215,1.0971524,-0.27837852,-0.19586825,-0.053837664,0.94147635,0.6782325,-1.8499793,1.3899486,1.4842832,-0.20198905,0.9118752,-1.3156699,-0.32807803,-0.46155733,0.59480095,-0.001722204,0.80457634,-0.21580535,-1.2434112,0.56428885,-1.0542293,0.7155026,1.2230334,-0.9761847,-1.017852,-0.1672411,-1.4611741,2.0070598,0.71912134,-0.6476812,-2.0907934,0.44325376,1.0635188,1.3988785,0.15533331,-0.9266868,1.1053526,0.25385013,1.2198808,-2.3114061,-0.13443123,-0.2354869,0.25447476,0.91357034,0.8403027,-0.059212927,1.9575791,0.24892993,-0.31545094,-2.6089191,0.80578184,1.2054881,-0.7192824,-1.2654837,-0.09570098,0.94349724,-0.8900375,1.2061056,-0.01412794,-0.66192263,-0.2441078,0.041381236,-0.2976183,0.9883735,-1.9825552,0.07066407,1.6886215,-0.68948716,0.5837726,-1.1841415,-0.10145782,-0.008195412,-0.1493982,1.5905204,0.13115445,-1.1599777,-1.202373,0.2164183,-0.3614637,-2.1209643,-0.1951186,1.9769924,-0.68308234,1.2002243,-0.90931207,-0.6301795,0.1339641,0.01462584
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
index 9b428c4a..d965bd46 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,7,0.26348084,0.1233009,-0.9510862,0.46851432,-0.06629313,0.8160627,0.5431626,-0.9199505,-1.360739,-0.2538705,1.1684657,-0.14489299,0.10989469,-0.2309123,-1.7760984,0.3490349,-1.0261083,0.1856724,1.4605979,1.6237235,1.7129359,-0.5611574,1.6723084,2.8874109,-0.98691314,-0.65485513,-1.0679046,0.44296682,-0.5673327,-0.2541984,-0.4929436,0.35431877,1.1681324,0.25615126,0.8050912,-0.35653836,-1.7087184,0.043991115,1.7704822,-0.08917144,-1.9798214,-1.8724368,-1.4108714,0.26632398,1.5392265,-1.3474573,-0.10791889,-0.567187,-0.96988684,1.7185776,-1.6027321,-0.33695924,1.4299693,0.46550983,-0.89209276,1.3031094,1.5477902,-0.6060929,-0.40710664,-1.2166933,0.41607055,-0.3897111,-0.08740451,1.7933946,1.7434844,0.009654752,-0.067560725,-0.5996982,1.7986253,0.3483771,-0.16625577,-1.9332577,-1.0659667,-0.59093297,-0.25158393,-0.40012267,-1.1106377,0.071712896,0.39155295,-0.039082322,0.38537106,0.21039137,0.69452846,-2.3482246,-0.79463667,-0.4639145,-0.52193904,-0.5554572,-0.5127063,-0.5268354,-0.9455389,1.341622,-2.0886319,1.58943,1.0482309,1.3858846,1.7408938,0.02892668,0.5694529,-1.5788714,0.9006992,0.118018724,0.7638617,2.0551603,-0.32436588,-0.6434378,-1.8908632,0.18551913,-1.5123209,-1.4987677,-0.7352275,-0.9231083,0.4445965,0.17023665,1.6077462,1.7081463,0.22440022,1.6947187,-1.0983723,0.48353592,0.5596647,0.8932061,-0.29618305,-0.6437828,-0.31312457,0.5083116,1.2900065,-0.33875406,-0.32123098,0.313723,-0.4940811,-0.8215438,-0.10663929,-2.3547385,1.1263422,0.68281025,-0.9162914,-0.43100452,0.49630877,-0.394641,0.5690176,0.45832327,-0.61618537,1.2726052,1.3995229,-1.3063124,0.4918943,-1.4364437,0.24002609,0.3745131,0.049085103,0.45316574,-0.453979,-0.018170724,-1.1888833,1.0507466,-0.4225001,0.8939625,0.40870383,-0.8803912,-0.66381943,0.4014866,-0.33330816,1.1037079,1.3088257,-1.8285968,-1.0561757,-0.80511665,-1.1772425,0.8244654,1.09275,0.4883135,-0.6731263,-0.55346674,1.5592078,-0.95583314,-0.41993904,-1.1939453,0.3556415,0.42150888,-1.353197,-0.4794684,1.0955988,-0.4739728,1.6267781,0.053663608,-1.6591601,-0.5589501,1.5400091,-0.4875693,-0.023662472,-0.7958585,0.68492943,-0.0061785085,0.0073458375,-0.25019705,0.9735828,0.9391158,0.18115312,-0.13106072
+8,0,7,1.8829988,0.5603056,-0.6805401,0.873459,-0.75400966,0.6241306,-0.022826796,-0.65359944,0.45811382,1.1624973,-0.36711055,1.1964695,-0.9203239,1.9030054,0.92825997,1.2234915,0.3935566,-0.10803351,-1.7747352,1.0153017,-0.22650294,0.07317609,0.7041025,0.083895385,-0.5787242,-1.1430625,-1.0159122,0.3586008,-0.81441945,0.12919727,-0.23688294,-0.8359273,1.0575047,-0.58928937,-0.19002041,-0.26165125,0.6974495,-0.35162425,1.4806678,-0.91830134,-1.5502017,1.193903,1.0963876,-1.9676751,0.075405344,-2.0842352,-0.2591765,1.1972393,-0.45678777,-1.2595507,0.0008469863,-1.209172,-0.04429603,0.6055306,1.2668537,-0.2412063,0.19914478,1.9480219,1.0576845,-1.475916,0.17119397,-0.6035175,-1.7098999,-0.37294888,0.419368,0.268539,-0.70104766,-1.7786356,0.5031316,-1.5002457,0.9026701,0.80984163,-1.2359964,0.22069247,-0.42576674,-0.39630565,-1.4970282,2.616495,0.6003689,-1.1450349,-2.02723,-1.7124385,0.30198652,-0.6123571,1.4489877,-1.4337479,0.3514351,-0.39667892,-0.68725485,0.8674336,-0.02994205,1.0586603,-0.8096146,-1.0183178,0.82786477,0.15899718,1.4175017,1.6423237,-0.7518763,0.46193466,-1.7427615,-0.89581615,0.25311762,1.9719808,0.84027904,0.5098649,0.69561577,-2.01775,0.16369472,0.09517597,0.4123239,0.43099934,-0.5587974,-2.099149,-1.5859907,-0.90635055,-0.023114711,0.15737575,-1.1646839,-0.39253628,1.4398996,1.0370032,-0.44213772,-0.64347506,0.49100915,-0.8488655,0.52995205,-0.12177242,0.4132956,-0.117767714,-0.041891925,1.1423306,-0.9095124,-0.7704257,0.6956402,1.2622151,-1.6017178,1.253087,-0.88926196,0.024840193,-0.6669459,-0.84649396,-1.5089233,0.5503707,1.0900313,0.52742964,1.4928386,1.0869912,0.35689026,-0.5618133,-0.1295253,-2.3059578,0.9978885,-0.25662556,0.12965643,1.9835927,-1.0982833,-0.09509576,1.6303843,-0.04004063,0.85300845,-2.1637661,0.115972474,0.8912982,-1.8220121,-0.67457515,0.4750627,-0.7254808,-1.0370337,1.8635073,-0.7830165,-0.7419083,-0.59229213,0.36369795,0.009352995,-0.652557,-0.9893492,0.13321461,0.05142843,-0.173441,1.0776938,-1.6504616,-0.21089926,0.84927285,-0.2011813,1.2290883,-0.25737432,0.17806257,-0.74700785,-0.22532028,-0.73679703,-1.1095785,-1.5106932,1.1296397,0.6030286,0.907326,-0.8425243,0.72134006,1.3058224,0.4379919
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
index 7701bce1..ea9dd586 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,7,0.76275426,0.18214901,-0.57188916,0.059701614,-0.28557384,1.2528464,0.79763407,-0.63960886,-0.96146375,-0.0512549,0.8373615,-0.3810237,0.24112543,-0.20450255,-1.7736192,0.36160433,-0.7304003,0.41004166,1.2904602,1.6976061,1.2359073,-0.87983704,1.8318158,2.488492,-0.795438,-1.1328822,-1.5530977,1.0552738,-0.77631116,-0.88773954,-0.4693062,0.16485235,1.0138454,0.17359263,0.7859901,-0.23536894,-1.4138858,0.38051438,2.14049,0.16130738,-1.9554428,-1.4978547,-1.225305,-0.007534434,1.8237702,-1.0094837,0.08617119,-0.4849619,-0.25551113,1.6335402,-1.7301743,-0.73824334,1.371595,0.8042116,-0.9446179,1.2468611,0.8701763,-0.7954289,-0.5617317,-1.1564339,-0.08375406,-0.4792437,-0.45387352,1.6733506,1.8638924,-0.5079436,-0.1641803,-1.058204,1.876411,0.6470958,-0.5844382,-2.1348605,-1.0387065,-0.6489218,-0.38448277,-0.53584003,-1.0877233,0.89465255,-0.070738666,0.4274419,0.9511695,0.727988,0.68438584,-2.3704088,-0.6923841,-0.68775594,-0.59505486,-0.51203656,-1.1223544,-0.11988432,-0.28750613,0.5628186,-1.4707938,1.4209483,0.17336716,1.5996127,1.5890402,0.4437642,0.9046561,-1.1071699,0.63461745,0.06332244,0.69563353,2.2999492,0.38046426,-0.77210474,-1.8445017,0.47982848,-2.1513295,-1.4039284,-0.7259288,-0.89535034,0.7229532,0.38970834,1.1962795,1.1890666,0.22372073,1.7938012,-1.1250141,0.9569801,0.8552448,0.5728794,-0.105201356,-1.0274678,0.068610765,1.1812729,0.73605824,-0.69981444,-0.71209633,0.503375,-1.2361196,-1.1348834,-0.5413534,-2.8409722,1.1658453,0.535366,-0.8362626,0.4911523,0.9966621,-0.16169278,0.5466709,0.065419264,-0.50554013,1.3522164,1.6981294,-1.1887987,-0.18314654,-1.4414601,0.17046571,0.5640569,-0.04139424,0.71503925,-0.32278606,-0.39535967,-1.3595438,0.8323629,-0.99404913,0.8279306,0.15347865,-0.36941954,-0.90245074,0.32202348,0.21524166,0.4956704,1.0135361,-1.5879138,-0.6533115,-0.97604185,-0.7554835,0.48038816,0.99044883,0.53412485,-0.16621038,-0.14424634,2.0332005,-0.21379094,-0.06764341,-1.2983478,-0.312222,-0.9194852,-1.2916971,-0.36260626,1.1965542,-0.7213257,1.3095983,-0.08372764,-1.4751688,-0.62356883,0.7289247,-0.069254085,0.41077098,-0.49018452,1.1164718,-0.7885649,0.029319612,-0.09494034,1.3525505,0.58701926,0.36867285,-0.38973525
+9,0,7,0.1209258,-0.7836567,-0.3029543,0.2785893,-1.0025321,0.037851073,0.048571438,0.33848727,-0.3773915,0.7844221,0.24281782,1.148343,-0.13137017,1.279799,0.97056156,1.0246092,0.80062383,-0.3923224,-1.8497859,1.1537666,-0.38878572,0.5604603,-0.58034015,-0.02841155,-0.48242685,0.8719435,-0.9715767,0.86573464,-0.0033297834,1.0148878,-0.93299323,-0.20949924,0.95323825,-1.5853611,-0.5756929,0.1248061,0.5185648,-0.11291537,1.8827963,-0.75403863,-1.9923172,0.3773292,0.76919514,-1.0892999,-0.5601977,-2.0243154,-0.80850774,0.68828875,0.8130654,-1.7597806,-1.9047588,-2.6262157,-1.1613489,1.2342267,1.8630753,0.9635703,-1.9320338,1.985816,-0.055163626,-0.7479937,-0.36713076,0.07263066,-0.35597026,0.9809213,0.099624224,-0.22642994,0.59534913,-1.3628237,-1.0601852,-1.4001173,0.26098126,1.0985968,-1.100743,0.23194666,-0.3807315,-1.1726305,-0.24944744,0.57557213,0.2954017,-1.1652223,0.08526788,-1.5450522,0.21617737,-0.8069991,0.8448825,-0.76840484,0.887578,-0.55467606,0.28331083,0.06474368,-0.70855623,-0.25272217,0.19233815,-1.6127638,1.925872,-1.4058347,0.109641865,2.0278912,-1.168832,-0.40527755,-1.4968951,-0.9538667,-0.92621404,1.25343,1.4133275,0.07042062,-0.17381254,-1.4111978,0.36360025,1.3456023,1.6989193,0.42588353,0.9820415,0.12705263,-0.5005522,0.054737598,0.76003623,0.84157014,-1.7930596,1.0273554,0.44291124,0.082684554,0.61543924,-1.5382718,-0.26630726,-0.56344163,0.47257867,0.62037295,1.3090107,-0.49892786,-0.3358481,-0.06688593,-0.927906,0.33645353,1.4333607,-0.5287842,-1.4776187,0.023554774,-1.1187627,1.6363778,0.19606885,-0.41932362,-2.4530096,0.52753305,1.0463974,1.1490204,0.4760272,-1.2210376,0.7896005,-0.003464025,0.8981069,-2.112449,0.16817431,-0.64763904,0.200444,1.2383525,0.66687185,-0.045646884,1.3314408,0.10086164,0.30638647,-2.6051254,0.5737982,1.2559868,-0.6507682,-1.2285036,0.194903,0.7337301,-0.36549726,1.3863009,0.23165633,-0.05751545,-0.071033075,0.66757375,-1.0468653,0.9908146,-1.5330824,-0.010820814,0.87953395,0.21151483,0.9051806,-1.0364068,-0.45686463,0.33169436,0.61652184,2.6955714,0.38302016,-1.3782058,-0.8940241,0.33452165,-0.84705555,-1.7417383,-0.18060239,1.1929514,-0.091192424,1.401481,-1.0531416,-1.0770215,0.70313454,-0.3485591
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
index 428d289b..04665ca6 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,5,1.7890625,0.41992188,-1.796875,0.6953125,-0.14941406,-0.0019836426,-1.9765625,-0.31445312,0.76953125,-0.75,-1.109375,-1.25,-1.484375,0.765625,-0.36914062,0.71875,-0.046875,0.29296875,0.32617188,0.73046875,0.6640625,0.39453125,1.203125,0.86328125,-0.85546875,0.51953125,-1.0703125,-0.63671875,-0.84765625,0.35546875,-0.7109375,0.8046875,-1.4609375,0.42578125,0.9921875,0.6171875,0.515625,-1.9453125,-1.34375,0.65625,0.24902344,0.48632812,0.80859375,-0.28125,0.37109375,-0.234375,-0.921875,-1.328125,1.84375,-0.41992188,2.078125,0.80078125,-0.6796875,-0.20703125,1.03125,1.9140625,0.23242188,-0.5390625,-0.7734375,-0.734375,1.3984375,-0.038085938,-0.20898438,0.56640625,-1.1484375,1.0390625,-0.31835938,-0.2578125,-1.0234375,-0.921875,1.65625,-0.24707031,-0.021606445,0.46484375,-0.15722656,-1.3671875,-0.60546875,-0.000015974045,-0.3046875,0.546875,-0.36132812,1.953125,-0.41992188,0.110839844,0.5703125,0.09863281,-0.5703125,-0.60546875,-2.453125,1.0859375,-1.515625,-0.94921875,-2.109375,0.796875,1.734375,-0.84765625,1.0625,-0.61328125,-1.25,0.51171875,-1.4453125,1.046875,1.5546875,-1.359375,0.26367188,0.10205078,-0.053222656,-0.65625,0.19628906,2.578125,1.265625,0.671875,-0.02368164,-1.03125,0.17480469,1.0,-1.21875,0.37109375,-0.81640625,-1.359375,-2.1875,0.94140625,0.12158203,0.765625,0.8046875,-0.17773438,0.625,1.0390625,0.28320312,-0.051513672,-1.1875,-0.4921875,0.32421875,0.37890625,-0.50390625,-0.91015625,2.0,-0.13964844,0.33398438,-0.7890625,-0.33984375,-0.81640625,0.828125,0.31835938,-1.2109375,-1.7734375,-1.578125,-0.72265625,0.47460938,-0.91796875,0.3515625,-0.51953125,-1.2109375,-0.57421875,-0.7421875,-1.3203125,-0.2734375,1.1953125,0.16796875,-0.3203125,-0.67578125,-0.43164062,-1.015625,-1.2421875,0.51171875,-0.08154297,-0.36914062,-0.546875,0.12988281,-1.21875,0.2265625,-0.033203125,1.2421875,0.82421875,2.578125,1.109375,1.2421875,-0.21972656,-0.18652344,2.546875,1.6796875,0.33203125,-1.59375,-1.2890625,0.72265625,1.609375,-0.12402344,-1.6484375,1.0390625,-0.23242188,0.4140625,0.69140625,-0.064941406,0.890625,0.17089844,1.21875,2.359375,-0.31054688,-0.51171875,1.390625
-0,0,6,0.16113281,1.0703125,1.09375,0.796875,1.359375,0.56640625,-0.75,-1.1328125,1.875,-2.265625,1.1484375,-0.50390625,-2.078125,0.052001953,-1.0703125,-0.0625,0.73828125,-0.96484375,0.3828125,0.25195312,1.75,0.41796875,0.5625,0.34570312,-0.19140625,-0.33789062,-1.421875,0.45117188,0.056884766,1.71875,0.3203125,0.6953125,0.77734375,-2.09375,-0.58203125,0.96484375,-0.9765625,1.484375,-0.5859375,0.5703125,0.78125,-0.11621094,0.25585938,-0.828125,0.8046875,0.65625,0.22363281,0.50390625,-0.74609375,-1.1640625,0.36914062,0.47265625,-0.6875,-0.10253906,0.46679688,2.1875,-1.4140625,0.25195312,0.51171875,0.5234375,-0.3984375,-0.041748047,-0.18457031,-0.78515625,-0.080566406,1.203125,1.3671875,-0.984375,1.8046875,-0.46875,-1.0234375,0.23046875,0.375,-1.0703125,0.7578125,-0.26367188,1.4140625,-0.37695312,-1.296875,1.7890625,1.2109375,1.5078125,0.765625,-0.35546875,1.09375,-0.32226562,0.11230469,-2.578125,-0.60546875,-0.65234375,-0.30859375,1.3671875,0.61328125,-1.4140625,-1.5625,-1.09375,-0.2578125,-2.90625,0.1484375,-0.6796875,0.45117188,-0.49609375,-0.69921875,-0.46875,1.375,0.21289062,1.4765625,-0.62890625,-2.328125,0.18945312,1.046875,1.359375,-0.39257812,-0.25195312,1.953125,2.109375,-0.05859375,1.1953125,0.31835938,-0.49609375,-0.359375,0.037841797,-0.3046875,0.43359375,-0.921875,-1.0078125,-0.42578125,-0.032714844,1.0546875,0.8359375,-0.8359375,0.107910156,1.015625,0.7890625,0.62890625,-0.6953125,0.33398438,-1.078125,-0.51953125,-1.265625,1.0625,-0.43164062,0.24414062,0.87890625,-0.81640625,0.890625,-0.78125,0.51953125,-1.0625,1.25,-0.26171875,-0.18066406,0.26757812,0.04296875,-1.3046875,-0.6796875,-1.1015625,1.28125,1.0,0.41796875,-0.86328125,-1.09375,-0.25976562,-1.2109375,-1.4296875,0.67578125,-0.011413574,-0.515625,-3.5,-0.56640625,1.1484375,0.47851562,1.0234375,1.015625,0.796875,-1.671875,-0.6484375,0.39453125,1.0078125,-0.5703125,1.1328125,0.40429688,-2.15625,0.74609375,0.51171875,1.5390625,0.796875,-0.546875,-0.18359375,-1.609375,-0.60546875,0.17285156,-0.30078125,0.90234375,-1.125,0.24316406,-0.35546875,-1.34375,-0.16503906,-0.7421875
-0,0,7,-0.71875,-2.15625,-0.546875,1.4375,1.1015625,0.45507812,0.056640625,-1.125,1.0078125,0.859375,2.03125,-1.2265625,-1.71875,0.123046875,1.3984375,-0.9765625,0.7578125,-1.03125,0.40234375,-0.37695312,1.453125,0.045410156,-0.33789062,0.43554688,1.4609375,0.7265625,-0.3984375,0.28710938,-0.70703125,-0.27734375,0.07373047,0.14355469,1.5546875,0.28515625,0.4765625,0.45898438,1.890625,0.65625,-0.6640625,-1.2578125,-1.4609375,0.703125,-2.046875,0.49804688,0.10888672,0.203125,1.7890625,-1.75,-2.34375,-1.203125,0.14746094,-0.0119018555,-1.1953125,1.7890625,0.1875,0.8359375,-0.22070312,-1.1328125,-0.22265625,1.6015625,-0.19726562,-0.14648438,1.109375,-0.13378906,1.5234375,-0.62890625,1.6953125,-1.4140625,0.12158203,-1.3203125,0.8828125,0.17089844,1.28125,0.75,-0.55078125,0.828125,-0.65625,0.859375,-1.4375,1.1953125,-0.55078125,1.078125,-1.4609375,-1.375,-0.19335938,-1.21875,-1.890625,0.11230469,0.77734375,1.390625,-2.015625,0.4609375,0.45507812,-0.22558594,0.10595703,0.83984375,0.81640625,0.23046875,-0.052490234,0.44140625,-0.104003906,-1.1015625,1.1015625,0.6171875,0.87109375,-1.1484375,0.66015625,-0.140625,-0.092285156,-0.92578125,1.625,0.032470703,-0.9296875,0.41210938,0.58203125,-1.3359375,-0.47070312,-0.28710938,0.031982422,0.12207031,-0.8125,0.026977539,0.17578125,1.9453125,0.63671875,-0.6953125,0.25390625,-0.29101562,0.78515625,1.4296875,0.39648438,0.88671875,-0.16308594,0.94140625,-2.109375,0.8984375,1.9453125,0.37109375,-1.015625,0.5546875,1.5,-0.33007812,-0.4765625,-0.20703125,-1.875,-1.5546875,-0.71875,1.375,-0.18261719,-1.4453125,1.546875,-2.40625,-0.9921875,0.44726562,-0.22460938,-0.59375,-0.953125,0.12011719,0.21289062,0.0035095215,-1.1171875,-0.15625,-0.640625,0.23339844,-0.20507812,1.890625,0.4140625,-0.75390625,-1.21875,0.041259766,0.33984375,-0.1796875,-0.20214844,0.07714844,0.04272461,0.91015625,1.625,0.69921875,1.3046875,1.0859375,-0.018676758,0.8515625,-0.38867188,-1.5234375,0.11328125,0.0859375,-2.109375,-1.7734375,0.16992188,-1.15625,1.9921875,-0.21289062,0.35546875,-1.296875,0.4375,0.072753906,-0.51953125,-0.3046875,-1.03125,-0.99609375
+0,0,5,0.99609375,0.23242188,-0.5625,-0.53515625,1.359375,0.578125,-0.7109375,0.4609375,1.2265625,-0.52734375,0.734375,2.5625,0.026000977,0.6171875,-0.11767578,-0.09326172,-0.80859375,0.23925781,2.96875,1.59375,0.6640625,-1.09375,0.49414062,0.060791016,-0.17675781,1.203125,-0.1171875,-0.61328125,-1.765625,-1.46875,-0.15820312,0.8203125,-0.796875,1.34375,-0.6953125,-0.18359375,0.28125,0.734375,0.33203125,0.16992188,0.74609375,-0.34765625,0.31835938,-0.27734375,-0.08642578,0.15820312,0.080566406,-1.7890625,-0.34375,-0.67578125,0.7578125,-1.1875,0.41992188,0.13671875,-1.59375,0.30859375,0.3046875,-0.44921875,1.1484375,-1.109375,-0.12695312,-0.15039062,1.46875,-0.7734375,-0.6171875,-1.25,0.20898438,-1.78125,0.53515625,0.71875,-0.52734375,1.3515625,-0.73828125,-0.76953125,0.81640625,2.46875,-0.80078125,-0.06591797,-1.9296875,-0.19042969,-0.033935547,-0.19824219,0.45117188,-0.26367188,0.83203125,-0.73046875,-0.35351562,-1.9375,0.12988281,1.2421875,0.59765625,0.42578125,-1.46875,1.65625,-0.24609375,-0.08496094,0.008178711,1.2890625,-0.5703125,0.091796875,0.50390625,0.28125,0.6953125,-1.0546875,-1.0703125,-0.13476562,0.060546875,0.8515625,-0.09277344,-0.11279297,1.2734375,0.6875,0.28320312,-0.09716797,1.875,-0.8984375,-0.90234375,-0.94140625,0.27539062,-0.15917969,-0.421875,0.37695312,0.38671875,0.33398438,-1.40625,2.15625,-0.26367188,-1.1953125,1.7734375,1.921875,0.8671875,-2.078125,-1.140625,0.47851562,-1.9140625,0.171875,-1.421875,-0.24023438,0.35351562,-0.984375,1.3515625,-1.40625,-1.671875,2.078125,-0.47070312,-0.41992188,-0.55859375,1.5703125,0.21777344,0.47265625,0.83203125,-1.375,0.6484375,-0.8671875,-1.578125,0.76953125,1.1640625,0.18359375,-0.05810547,-1.171875,0.20117188,1.4375,0.484375,2.296875,1.046875,0.13085938,-0.9921875,-2.859375,1.09375,0.14160156,0.76171875,-0.63671875,-0.19140625,0.546875,-0.5546875,-0.49609375,-0.92578125,-0.52734375,-0.4375,0.76953125,-0.4765625,-0.51171875,-0.3515625,0.7734375,-1.4765625,1.2109375,-1.0,-1.921875,1.875,-1.0546875,-0.859375,-1.03125,-0.6875,-0.38867188,0.021728516,1.46875,-1.6328125,0.27734375,0.15917969,1.0859375
+0,0,6,1.1015625,0.88671875,1.1875,-0.51171875,0.19824219,-0.25,0.55078125,-1.09375,0.76953125,-0.87890625,0.91796875,-0.039794922,1.1015625,-0.43359375,0.59765625,-0.4453125,-0.24804688,-0.140625,2.328125,-0.010864258,-0.47070312,-1.296875,0.25,-0.12890625,-0.6484375,-0.734375,0.859375,-0.5859375,-0.9921875,1.265625,0.45898438,-0.6875,0.78515625,-1.3671875,0.70703125,-0.0072021484,0.11621094,-0.5703125,-0.625,1.109375,-1.421875,0.032470703,-0.5625,-0.08251953,-0.22167969,-1.296875,0.5703125,-1.109375,1.234375,-0.3671875,-0.78515625,0.58203125,-0.71875,1.34375,0.16210938,-0.043945312,0.75390625,-0.73046875,-1.421875,-0.71484375,1.1953125,0.7109375,-0.84375,1.140625,-0.21679688,0.051757812,0.35742188,-2.1875,-0.6640625,1.40625,-1.0546875,0.39257812,0.4453125,0.025634766,0.17382812,1.4609375,1.0390625,1.2109375,-0.828125,-0.265625,-1.3984375,-0.48242188,0.734375,-0.65625,0.11425781,-1.0546875,-0.6640625,-2.46875,1.546875,-1.03125,0.49804688,0.8203125,-0.73828125,0.5859375,-1.21875,-0.65625,-1.9765625,-0.046142578,-1.421875,0.43164062,-0.59765625,1.1015625,1.6015625,0.029663086,1.140625,2.15625,-0.515625,0.72265625,0.40625,-0.64453125,0.95703125,1.484375,0.76953125,0.33203125,0.83203125,0.46484375,-0.140625,-0.30273438,-0.40039062,1.0234375,0.29492188,-0.74609375,1.4140625,-0.640625,-1.5546875,1.3359375,-1.65625,-1.203125,-0.64453125,0.29882812,1.5625,-0.59375,-0.30273438,-0.984375,0.38671875,0.01586914,0.40625,-0.10205078,0.48828125,-1.7421875,2.421875,0.796875,-1.21875,2.296875,0.040771484,-1.84375,2.09375,-0.58203125,-0.61328125,0.98828125,-1.1640625,-0.28125,1.5390625,-1.3125,-0.8984375,0.71875,1.5703125,-1.3046875,0.58203125,0.43359375,-0.35546875,-0.28320312,2.484375,2.078125,0.060058594,-0.87890625,-1.2578125,0.69140625,0.5,-0.24121094,-0.24414062,-0.22265625,-1.7109375,-1.40625,-0.5859375,0.234375,1.3203125,-0.9921875,0.6015625,0.020507812,-1.8828125,-1.2734375,0.546875,-0.35742188,-1.890625,-0.375,0.26953125,-0.79296875,-0.57421875,-0.1875,-1.53125,1.3046875,0.24316406,-0.58984375,-0.90234375,1.3046875,1.2421875,-0.609375,1.5546875,1.1484375
+0,0,7,1.390625,1.453125,-0.890625,-1.4140625,-0.546875,-2.125,0.17285156,1.1171875,-0.43359375,-1.171875,0.104003906,-1.3359375,-2.234375,-0.20605469,-0.25390625,1.15625,-0.24511719,0.6171875,-0.15625,0.0019226074,0.6640625,-0.234375,-0.23242188,1.046875,-1.7109375,-0.34765625,-2.0,0.5859375,1.34375,0.51953125,0.5,1.875,-2.34375,-0.18847656,1.2578125,-0.25390625,1.1640625,2.046875,0.111328125,0.8125,-1.46875,-0.38476562,0.4453125,2.171875,0.5390625,1.6171875,0.43164062,0.63671875,-0.13183594,0.34375,0.36914062,-0.10253906,-0.8984375,-1.234375,-0.30078125,-0.296875,-0.34375,-0.01574707,-0.14160156,-0.546875,-0.24609375,0.88671875,-0.4375,-0.35742188,1.390625,0.0025939941,0.46484375,-0.1328125,-0.26757812,0.4140625,-0.515625,-0.24414062,0.13769531,-0.7734375,0.53125,0.07421875,2.234375,-0.078125,-0.265625,-0.19824219,0.88671875,0.7734375,1.6796875,0.48632812,0.67578125,0.59375,-0.77734375,0.34765625,-1.984375,0.25195312,0.484375,1.8046875,2.078125,-0.26367188,-0.05493164,-1.46875,-0.0390625,0.26953125,1.2265625,1.5234375,1.484375,-0.14941406,0.8515625,-0.28515625,-0.76171875,-0.91796875,-1.015625,-1.28125,-0.6796875,-0.032470703,-0.15234375,-0.25585938,2.6875,-0.6875,-0.076171875,-0.703125,-0.34179688,-1.5703125,0.7265625,0.48046875,1.90625,-2.046875,-0.85546875,-2.578125,-0.41601562,0.08105469,-0.80078125,1.375,0.04638672,-1.734375,0.97265625,-0.48242188,1.0,-0.45117188,-2.046875,0.2578125,0.56640625,-0.20996094,0.29101562,-2.34375,0.32421875,0.9609375,1.296875,1.109375,-0.6171875,-0.44921875,-1.921875,1.6796875,0.46484375,0.10546875,-0.26367188,-1.8828125,-0.58203125,1.5390625,-0.049316406,-0.071777344,-0.49023438,0.84375,-0.7265625,2.09375,-2.0625,0.94921875,-0.59375,0.0859375,-0.51953125,-0.25195312,-0.21875,-0.032714844,0.04711914,-0.5546875,-0.25976562,0.5078125,-1.1953125,0.9140625,1.265625,-0.15917969,1.1953125,-1.1796875,0.052246094,-0.42578125,1.125,0.5234375,-0.20507812,-1.375,-0.45703125,-0.12792969,-0.4140625,-0.37304688,-0.31445312,-0.953125,-0.890625,-0.2890625,-0.74609375,-0.7890625,1.4140625,0.53515625,1.1796875,-0.3984375,0.0047912598,0.24609375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
index a6d6cd23..855bd8e8 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,5,-0.8203125,0.5703125,1.25,1.2265625,0.125,-0.82421875,0.037841797,0.77734375,1.9296875,-2.65625,-0.11425781,-0.12451172,-0.6015625,1.171875,-0.34765625,-0.7890625,-0.4921875,-1.2109375,-0.7890625,-0.22949219,-0.5625,-1.5859375,0.27929688,-2.234375,0.123535156,-2.421875,1.7890625,-1.671875,0.07714844,-0.0859375,-1.65625,1.5859375,0.21679688,-0.921875,0.1015625,-1.046875,0.546875,0.26171875,0.39453125,-0.38085938,0.27148438,1.40625,-1.109375,-1.4375,-0.04711914,-0.7421875,-1.03125,0.45898438,-1.5703125,-2.34375,-0.98046875,1.234375,-0.48828125,-0.6640625,0.9140625,2.0,-0.921875,-0.30859375,1.734375,0.5234375,-0.875,-1.96875,0.27734375,-1.296875,-0.42382812,0.14257812,0.9140625,-1.828125,-0.76953125,-0.27929688,-1.265625,1.171875,-0.234375,0.65625,1.5,-0.72265625,0.83984375,-0.25390625,-0.71484375,1.2265625,1.75,0.70703125,1.109375,-0.5234375,0.21386719,0.671875,0.021362305,0.43945312,-0.8828125,0.28320312,0.80078125,0.421875,2.546875,-0.62890625,-0.48632812,0.24804688,0.9453125,0.15234375,-0.9765625,2.21875,-0.33789062,0.67578125,0.46484375,-1.015625,0.6484375,-0.123535156,-0.140625,0.46484375,-1.9375,0.26757812,0.029296875,-0.14746094,0.5546875,0.66796875,0.35351562,-0.56640625,0.1640625,0.6171875,0.92578125,-0.96484375,0.14257812,1.625,0.28320312,2.515625,-0.6484375,-1.7734375,-0.5703125,0.19628906,0.40820312,1.0625,0.3828125,-0.78125,0.11425781,1.0,1.1875,0.41210938,-0.796875,-0.890625,-0.76171875,0.45117188,2.140625,-0.29296875,-0.546875,0.63671875,0.0037384033,-1.0859375,-0.609375,0.26953125,-0.33398438,1.1875,1.1953125,0.31835938,0.004760742,0.65625,0.44140625,-1.703125,-0.8359375,0.828125,0.24902344,-1.1796875,0.17773438,0.31054688,1.625,0.22070312,-0.17773438,0.484375,-1.109375,-0.546875,-1.0078125,-1.1875,0.13378906,1.3125,-0.65625,-0.14355469,0.24707031,1.5390625,1.3515625,1.4296875,-0.6328125,0.28515625,-0.14160156,0.19042969,-0.11669922,-0.84375,1.1953125,1.15625,-1.3125,-1.296875,0.67578125,-1.8359375,-1.578125,-0.85546875,2.0,1.125,0.41796875,0.33398438,-0.5234375,-0.067871094,-0.41601562,-0.22363281
-1,0,6,-1.3125,-0.34960938,-0.38671875,1.65625,1.3984375,-0.6953125,-1.078125,0.061035156,-0.19335938,0.0146484375,0.53125,-0.6015625,-0.6171875,-1.3046875,0.80078125,0.25976562,-0.36132812,-1.046875,0.5546875,-0.44921875,0.67578125,-1.1015625,-0.15625,0.68359375,-0.25585938,-1.2734375,-0.61328125,0.65234375,-1.578125,-0.6484375,0.30859375,0.53515625,0.7578125,-0.6171875,-0.06982422,0.13769531,0.75,0.21386719,-0.47460938,-0.80078125,-0.47265625,0.62109375,-1.59375,-0.52734375,-0.4609375,0.014709473,0.5390625,-0.83984375,-0.875,-1.140625,0.88671875,1.421875,-2.125,2.21875,0.75390625,1.34375,0.86328125,0.30664062,0.69921875,0.3203125,-0.703125,0.578125,0.9140625,-1.328125,0.66015625,-0.30273438,0.671875,-0.2890625,-0.609375,-0.7421875,1.0078125,-0.7734375,-0.16699219,1.0078125,-0.484375,-0.296875,0.17089844,0.46289062,-2.078125,0.77734375,-0.59765625,0.36132812,-0.13085938,-2.328125,-0.28320312,-0.625,-0.578125,-0.76953125,-0.18945312,0.453125,-1.3515625,0.6171875,1.0859375,-0.6328125,-0.8515625,0.30664062,-0.29296875,-0.59765625,-1.1484375,0.79296875,1.6953125,0.10839844,1.390625,1.4296875,1.6640625,-1.015625,2.140625,0.20898438,0.34179688,-0.21972656,1.4296875,0.49023438,-0.84375,-0.18554688,1.0703125,-0.057861328,0.26367188,1.796875,0.546875,0.5390625,-0.54296875,1.5390625,-0.57421875,3.25,2.015625,-1.6015625,0.2421875,-0.3359375,0.49609375,1.1796875,-0.9140625,1.296875,-1.15625,-0.36132812,-1.828125,1.1484375,1.625,-0.33789062,-0.4765625,-1.7109375,0.8046875,-0.49414062,0.3984375,0.34765625,-0.7734375,0.24121094,-0.578125,1.625,-0.546875,-0.2265625,0.66796875,-1.8515625,-1.8828125,2.125,-0.106933594,-1.9140625,-1.9140625,-0.35546875,1.0390625,0.40234375,-0.19628906,-1.2109375,1.1640625,0.23535156,0.6171875,0.5703125,0.48242188,-1.984375,-0.06347656,1.515625,-0.07861328,0.30273438,1.1484375,0.022705078,-0.049072266,-0.6640625,1.0703125,1.34375,2.203125,0.122558594,0.9609375,-0.70703125,-0.29101562,-1.9453125,0.26171875,-0.3125,-1.6015625,-0.58203125,0.515625,-1.90625,-0.43359375,-0.23730469,1.734375,1.2734375,-0.70703125,-0.7265625,0.055908203,-0.4765625,-1.1796875,-0.734375
-1,0,7,0.76953125,0.265625,-0.25195312,1.2890625,0.87890625,-0.98046875,-0.09375,-1.21875,0.73828125,-0.092285156,0.1484375,-0.8984375,-1.125,-0.64453125,0.6875,0.65625,0.8046875,-0.26367188,1.0234375,-0.39648438,0.26171875,-0.625,-2.171875,1.34375,1.7265625,0.39648438,-0.94140625,1.2109375,0.6796875,-0.62890625,-0.75,0.30273438,0.58203125,-1.078125,-1.875,0.15722656,0.96484375,-1.828125,-0.8125,-1.6328125,-0.78515625,0.6875,-1.0703125,1.1328125,-0.61328125,-0.41796875,1.09375,-0.375,-1.0546875,-0.27734375,2.53125,0.7890625,0.484375,2.03125,-0.8046875,0.17773438,-1.1171875,-0.59765625,-0.24902344,0.013977051,1.265625,1.515625,-0.83203125,-0.515625,0.07421875,-0.81640625,0.48046875,-0.60546875,1.484375,0.19726562,-0.55859375,0.0008087158,-1.546875,0.86328125,-1.4609375,0.0006942749,1.1171875,0.51171875,-1.953125,0.004272461,1.2109375,1.3828125,-0.80078125,-1.4375,0.69140625,1.40625,-1.6640625,0.02709961,0.041259766,0.33398438,0.55078125,0.61328125,0.06542969,-0.12792969,-1.4375,0.19433594,-0.19238281,0.16308594,-1.6640625,-0.38867188,0.12988281,0.5625,-1.390625,1.984375,-0.6796875,1.40625,0.35742188,-0.5078125,-0.45703125,0.9765625,1.640625,0.953125,-1.015625,0.546875,0.33984375,0.8203125,-1.09375,1.0625,-0.7265625,-0.45117188,-1.2421875,2.359375,-1.71875,0.3671875,0.72265625,0.21777344,1.3359375,-0.59375,0.43554688,1.046875,-0.119140625,1.3046875,-0.34179688,0.07714844,-0.25,0.007019043,2.703125,-0.1953125,-0.23925781,-0.77734375,0.06591797,0.12988281,-0.6171875,-0.100097656,0.24707031,-1.484375,-0.30859375,-0.9296875,0.20800781,-1.7734375,-0.62109375,-2.15625,0.68359375,1.78125,-0.50390625,-0.49023438,-1.5625,0.3046875,-0.25976562,0.91015625,-0.671875,0.111328125,-0.578125,-0.5859375,0.90625,0.9296875,0.91015625,-0.32421875,-0.6953125,-0.7109375,-1.1484375,0.012512207,2.296875,-1.5234375,-0.033691406,0.75,0.546875,-0.671875,2.359375,1.7734375,2.34375,-0.31835938,-0.91796875,-0.15234375,0.671875,-1.2734375,0.3984375,-1.3671875,-0.73046875,-0.5390625,0.296875,-0.13671875,0.29492188,1.3828125,-1.8359375,-0.29296875,-1.1171875,-0.44726562,0.44921875,0.1640625
+1,0,5,1.703125,1.15625,-0.33398438,0.65234375,-1.03125,0.07519531,1.015625,1.7890625,0.23632812,-1.8828125,0.053466797,0.5625,0.203125,0.107421875,-0.71875,0.7890625,-0.37304688,0.578125,0.546875,-0.578125,0.45507812,-0.86328125,-0.21875,-0.5390625,0.24707031,-0.18554688,-1.09375,-0.265625,-2.171875,-0.69921875,0.24316406,-0.71484375,0.796875,0.765625,0.66796875,-1.671875,0.546875,-0.765625,-0.55078125,-0.3203125,-0.421875,-1.0234375,-0.89453125,0.75,-2.4375,0.4140625,0.8203125,0.671875,1.203125,0.39257812,-0.95703125,-0.34179688,-0.17578125,1.5078125,-1.0859375,0.66796875,1.7265625,0.42578125,0.071777344,-0.49609375,-0.083984375,0.91796875,-0.41992188,0.15039062,0.359375,-1.0390625,-0.62890625,-0.78125,0.15136719,-0.35351562,0.69140625,-2.09375,0.035888672,-0.76953125,1.8515625,2.703125,1.734375,1.03125,-0.6484375,-0.9375,-0.90234375,-0.984375,-0.74609375,-2.125,1.75,-1.015625,1.8046875,0.09765625,-0.96875,1.7109375,1.6953125,-1.4921875,0.53125,0.921875,-1.2109375,0.42773438,0.48632812,-0.22363281,0.23925781,1.09375,-0.63671875,1.1875,0.90234375,-0.3828125,2.484375,-0.78125,0.47460938,1.1171875,0.9765625,1.4453125,0.58984375,0.515625,0.27929688,0.04711914,1.9296875,-0.033203125,-1.5078125,-0.2578125,-0.5078125,-0.6484375,-0.38867188,-0.9765625,0.8515625,-0.20117188,-1.1328125,0.6484375,0.66015625,-0.44921875,-0.6328125,-0.61328125,0.09326172,-0.18847656,-0.25,-0.37109375,-1.9609375,-1.015625,0.1328125,0.48046875,-0.359375,-1.9296875,2.171875,-0.71484375,-0.640625,-1.7265625,-0.011352539,-0.47070312,-0.890625,-0.46484375,1.2890625,4.03125,-0.69140625,0.15820312,-1.8046875,0.69921875,-0.75390625,-0.61328125,-0.30078125,0.515625,-0.31640625,-0.55859375,0.15625,0.087402344,-0.40820312,0.24316406,-0.41992188,0.9140625,0.14355469,1.03125,1.6875,-1.6796875,-0.21191406,0.107910156,-0.39453125,-0.3671875,0.546875,0.78515625,1.0078125,1.0078125,-0.90234375,-0.54296875,-0.66015625,1.265625,-0.029785156,-0.515625,-0.28320312,-1.203125,1.109375,0.7109375,-0.7890625,-0.30273438,-0.53125,-1.609375,0.30273438,-0.38476562,-1.5,1.859375,-0.35546875,-0.04272461,-0.30273438,-0.111328125
+1,0,6,-0.032958984,1.6171875,-1.140625,-1.5625,0.23144531,-1.4921875,0.41796875,1.671875,1.5625,-0.859375,1.0078125,1.0390625,-1.4609375,0.54296875,-0.5078125,0.70703125,0.020874023,0.046875,-0.18945312,-0.17480469,-0.01928711,-0.37890625,1.4609375,0.07910156,-0.7578125,0.007659912,-1.953125,0.796875,0.59765625,-0.7109375,0.63671875,0.7265625,-1.9140625,0.05029297,1.96875,-1.171875,1.296875,1.734375,-0.33984375,1.6328125,-0.828125,0.625,0.29492188,0.15527344,-0.70703125,2.390625,0.11230469,1.375,1.1015625,0.24609375,-0.59375,1.046875,-0.099609375,-0.40429688,-1.3828125,0.0078125,0.08984375,0.21777344,-0.953125,1.2265625,0.23632812,1.1953125,-0.82421875,0.67578125,0.35351562,-1.421875,0.27539062,0.29101562,-0.85546875,0.7734375,-0.2890625,0.609375,1.6875,-0.57421875,0.20117188,2.078125,0.7578125,0.625,-1.15625,-0.21972656,0.026000977,-0.16992188,0.50390625,-0.78125,0.890625,1.1328125,-0.9921875,-0.734375,-1.3984375,-0.49414062,0.07324219,1.421875,1.234375,0.0016174316,0.43945312,-1.9296875,-0.24121094,0.052490234,0.6796875,-0.07373047,0.9375,0.76953125,0.84375,-0.1484375,-0.58203125,-0.46679688,-0.7890625,0.0003528595,0.008300781,0.49023438,-1.078125,0.29101562,1.625,-1.6796875,1.34375,0.02709961,-1.484375,-0.90625,0.52734375,-0.41015625,1.328125,-2.6875,0.53515625,-1.5546875,-0.072753906,0.921875,-0.33203125,-1.328125,0.73828125,-0.15527344,-1.015625,-1.3125,-0.35546875,1.3828125,-1.9140625,0.94140625,1.671875,0.19921875,1.0,-2.046875,1.3828125,0.37304688,0.38671875,1.0546875,-1.0546875,-1.4765625,-0.94140625,-0.734375,-0.68359375,-0.69921875,0.41015625,-1.734375,-0.25390625,0.71875,-1.421875,2.234375,0.1640625,-0.123046875,0.00065612793,0.515625,-0.30664062,0.8671875,0.3984375,0.78125,0.51171875,-0.49804688,-0.33398438,0.66796875,-0.75390625,-1.71875,-0.9375,-0.15136719,-1.671875,-0.64453125,0.9609375,-0.55859375,1.0625,0.75,-0.57421875,-0.734375,0.3359375,-1.3046875,-0.515625,-0.51953125,-0.032226562,0.21875,-1.140625,-1.21875,-1.0625,-1.1015625,0.31640625,0.23535156,-1.375,0.32421875,-0.66015625,3.34375,1.5625,0.119140625,-0.84765625,-0.39453125
+1,0,7,1.5703125,0.671875,-0.91015625,0.052734375,-1.5078125,0.123046875,0.4375,1.0703125,-1.25,-1.6796875,0.25976562,-0.296875,-0.80078125,0.23828125,-1.171875,-0.7734375,-1.1640625,0.115234375,-0.97265625,-0.46875,0.49023438,-0.42578125,-0.58984375,-0.4296875,-0.02746582,-1.4453125,-1.921875,-0.65234375,-0.81640625,0.055664062,-0.91796875,-1.125,-0.36914062,0.94921875,0.049072266,-0.515625,0.34960938,-1.7578125,0.59765625,0.50390625,1.2109375,0.0015792847,-0.39648438,1.6015625,0.67578125,-0.06225586,-0.82421875,0.3046875,-0.14453125,-0.09082031,0.26171875,-0.33203125,-1.0625,0.40039062,-1.2890625,-0.58203125,0.1640625,1.1875,-1.0078125,-1.3828125,-0.4453125,0.98046875,0.6953125,0.953125,1.6484375,-0.39648438,0.16210938,-0.953125,-1.34375,0.10498047,0.54296875,-0.5078125,1.328125,-0.66796875,0.78515625,0.9140625,2.75,1.3125,0.51171875,0.076660156,-0.98046875,-0.1640625,0.24902344,-1.859375,0.43359375,1.171875,0.47070312,-0.828125,0.099121094,1.1953125,0.114746094,1.1953125,0.58984375,0.5703125,-1.234375,-1.328125,-0.609375,-0.0012664795,1.359375,1.4609375,0.953125,1.3515625,-0.29296875,-1.625,1.2890625,-1.4921875,-0.43945312,1.1015625,0.4921875,-0.40429688,-0.3046875,1.2421875,1.6015625,0.515625,0.72265625,0.46484375,-0.30078125,-0.016967773,-0.13769531,1.609375,0.36914062,-2.265625,-0.81640625,-0.58984375,-0.33203125,-1.3046875,0.21777344,-0.875,-0.044921875,-0.18847656,0.90234375,-0.9296875,-0.018676758,0.58203125,-1.171875,2.265625,-0.47460938,-1.3359375,1.4609375,-2.28125,0.7421875,-0.47265625,-0.49609375,-0.625,1.078125,0.359375,0.28320312,0.77734375,2.71875,3.46875,0.13476562,-0.58203125,-1.4609375,0.8984375,-0.5078125,0.51953125,0.44140625,0.671875,-1.3671875,0.828125,0.8359375,1.53125,0.4140625,-0.40234375,0.080566406,-0.4375,-0.17382812,1.6484375,0.24414062,-1.5703125,-0.53515625,-1.15625,0.09277344,-0.6640625,0.8671875,1.15625,-0.30078125,1.1328125,1.1015625,-0.859375,0.096191406,0.20605469,0.6875,-1.078125,0.20996094,-0.064941406,0.31445312,1.40625,0.5546875,-1.4296875,0.16601562,-2.234375,0.13574219,-2.125,-0.78125,1.84375,-0.037841797,-1.140625,-1.0625,-0.1796875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
index 34da42cc..a4049a56 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,5,-1.140625,-1.0546875,-0.25390625,-0.48242188,0.87109375,0.18457031,0.69921875,-0.37695312,0.9453125,-0.88671875,-0.58984375,-1.2734375,-1.0625,0.084472656,-0.52734375,-0.80078125,0.515625,1.59375,1.4296875,0.65625,0.86328125,0.50390625,-0.32226562,0.9140625,0.28515625,0.55859375,0.49804688,1.28125,0.95703125,-0.40429688,-1.078125,0.100097656,0.41210938,-0.44921875,-0.47265625,-0.4296875,0.8984375,-1.6015625,-0.34375,-0.27148438,0.18945312,-0.048828125,-1.0,0.45898438,0.73828125,-1.2109375,-0.84375,0.98828125,-1.21875,-1.484375,0.453125,0.6640625,-1.40625,0.5234375,0.515625,0.4921875,-0.57421875,0.5546875,0.10888672,0.71875,-1.59375,-0.50390625,0.70703125,0.765625,-0.14355469,0.19628906,1.46875,0.48242188,0.6328125,-1.2109375,0.2109375,-1.0859375,-0.26367188,1.3125,-2.09375,0.390625,0.18261719,1.09375,-1.03125,-1.34375,-0.13378906,2.15625,-0.86328125,-0.55859375,0.94140625,0.06933594,-0.76953125,-1.8515625,0.16992188,0.53515625,-0.74609375,-0.58984375,1.140625,-0.359375,-0.19238281,0.78125,0.90625,-1.0390625,-0.9140625,1.203125,-1.328125,0.33007812,-0.66015625,0.8828125,-0.609375,1.6328125,-0.56640625,0.33984375,-2.359375,-0.70703125,2.84375,0.26367188,1.8046875,-0.92578125,0.8828125,-0.6640625,-1.0078125,-0.5859375,0.49804688,0.51171875,1.1015625,0.23339844,0.7109375,1.25,0.6015625,0.73046875,-0.37304688,-0.70703125,0.84375,1.09375,-1.2578125,-2.171875,0.74609375,1.390625,-0.6953125,-0.52734375,1.59375,-1.1171875,-1.140625,0.24609375,-0.14257812,0.44921875,-0.83984375,-0.8515625,-1.734375,-2.21875,-0.06933594,0.9609375,-1.25,-0.6640625,-0.41992188,-0.984375,0.64453125,0.20800781,-0.43164062,-0.265625,0.80859375,0.69921875,-1.109375,-0.7734375,-0.17382812,0.1015625,-1.8515625,-0.6796875,1.546875,1.3203125,-0.55078125,0.79296875,-1.8203125,2.109375,-1.203125,2.328125,1.7890625,1.2578125,0.515625,0.3125,0.83203125,0.07861328,0.057128906,3.109375,1.09375,2.015625,-1.0390625,-2.03125,-0.26171875,-0.8984375,-0.72265625,-0.38867188,0.20605469,-0.6640625,0.36132812,-1.546875,0.4296875,0.671875,0.43164062,0.29882812,0.5703125,-1.078125,-0.008300781,0.37695312
-2,0,6,-1.984375,0.23242188,-0.41210938,0.8359375,0.40429688,0.859375,-0.28320312,0.55859375,0.28515625,-0.57421875,0.43554688,-1.1484375,-1.0546875,0.66796875,-0.625,-0.41601562,-0.29296875,0.734375,1.21875,0.92578125,-0.20507812,-0.28125,1.5,0.44335938,1.5546875,-0.66796875,0.59375,-0.010253906,-0.03930664,0.2578125,-1.5546875,1.4375,-0.76953125,-1.09375,1.171875,-0.828125,0.15136719,-0.32617188,-1.1484375,0.1484375,-0.45507812,-0.59375,-0.23144531,-0.6640625,0.61328125,0.010803223,0.15429688,-0.20898438,-0.6484375,-0.60546875,0.47265625,0.18652344,-1.484375,-0.23242188,0.1171875,2.578125,-0.8671875,-1.3359375,1.9921875,1.6640625,0.578125,-0.27539062,0.62109375,0.068847656,0.71484375,0.29101562,0.19335938,-0.75,-1.4609375,-2.28125,-2.3125,0.34179688,0.6171875,-1.28125,0.62109375,-0.75390625,0.34960938,-0.43359375,-1.0078125,0.84765625,0.0065612793,1.796875,-0.25976562,-1.65625,0.42382812,-0.024169922,-2.109375,-0.83203125,0.45703125,1.5078125,0.88671875,0.8828125,1.140625,-0.72265625,-0.82421875,-0.22460938,0.84765625,-1.3125,-1.390625,1.796875,-0.22363281,-0.52734375,-0.27148438,0.27148438,-0.5703125,0.59765625,1.828125,-0.5,-1.21875,-0.46679688,-0.83984375,0.8359375,-0.72265625,-1.4140625,0.40234375,0.25195312,-1.8046875,1.2734375,-0.18261719,-0.7890625,0.32421875,-0.123535156,-1.0703125,1.5,0.47070312,-0.82421875,-0.25390625,0.5546875,0.6328125,0.86328125,-0.06591797,-0.13378906,0.18359375,0.4609375,0.1640625,-1.953125,0.23535156,-0.5859375,-1.125,-0.016967773,1.625,-0.07373047,0.29296875,1.0859375,-0.32617188,-1.3671875,-1.8046875,-0.060791016,-0.43164062,0.25976562,0.734375,-0.19433594,0.05810547,0.546875,-1.5,0.6328125,0.69921875,-0.41992188,0.6796875,-1.3515625,-1.3359375,-0.96484375,-1.21875,-0.41015625,0.15136719,-1.3515625,-0.7734375,-0.03930664,-0.98828125,-0.50390625,0.14453125,0.87890625,0.37304688,0.34765625,0.90625,1.6484375,2.359375,-0.15234375,1.0,2.109375,1.453125,2.59375,-0.859375,-2.15625,2.296875,1.828125,0.20019531,-0.51171875,1.4375,-1.453125,-0.33398438,-0.82421875,-0.16699219,2.390625,0.7421875,0.36523438,0.8671875,-0.66796875,1.171875,-1.046875
-2,0,7,1.1640625,-0.16113281,-0.31054688,0.09472656,0.9140625,-0.62109375,0.076171875,-0.022338867,-0.3828125,-0.31640625,-1.5,-3.078125,-1.546875,-0.5625,-0.06689453,0.515625,0.6484375,0.796875,0.76953125,-0.5234375,0.12988281,-0.2734375,-0.33007812,0.62890625,0.6171875,0.86328125,-1.4296875,0.37890625,-0.119140625,0.4375,-1.15625,-0.88671875,-0.6875,-0.91796875,0.23242188,-0.90234375,-0.5703125,-1.5859375,-1.0703125,-0.055908203,0.38085938,-0.578125,-0.203125,0.071777344,0.03112793,0.34765625,0.46679688,0.27148438,0.016357422,-0.8046875,0.92578125,0.68359375,-0.828125,0.30859375,1.4140625,1.3984375,-0.8125,-1.1015625,-0.40039062,0.3671875,0.66015625,0.28710938,-0.23632812,-0.13769531,0.40039062,1.21875,2.546875,0.42382812,-0.0546875,-0.75,0.640625,-1.0390625,-0.7109375,-0.14160156,1.3125,-0.14550781,-0.12988281,0.17285156,-1.828125,-0.6484375,0.31054688,3.0625,-0.7109375,-1.3359375,-0.24804688,0.58984375,-1.9765625,1.0703125,-0.10449219,0.95703125,-0.65234375,-1.1640625,1.6875,0.42578125,-0.006958008,0.5625,-0.022583008,-0.64453125,-1.6484375,-0.34179688,1.609375,0.07470703,0.97265625,0.95703125,0.859375,-0.51171875,0.1796875,-2.1875,-0.390625,-0.8046875,1.4296875,-0.08300781,0.079589844,-1.3125,0.6328125,0.68359375,0.31835938,0.8515625,0.13769531,0.45117188,0.1328125,0.75,-0.84765625,1.7265625,1.3984375,-0.23242188,0.171875,0.14648438,-0.38085938,0.42382812,-0.578125,-0.5390625,-1.1640625,-0.69140625,0.100097656,-2.15625,1.015625,-1.859375,-0.7265625,0.5859375,0.625,-1.171875,0.69921875,0.27734375,-1.421875,-2.0,-1.078125,0.30078125,-0.57421875,-1.28125,0.49414062,-1.8828125,-0.21386719,1.234375,-0.26757812,-1.3828125,0.91015625,0.38476562,1.2265625,-1.234375,-1.6171875,-0.47070312,0.12988281,-0.6875,-1.2421875,-1.1015625,-1.234375,-0.27734375,-0.32226562,1.1328125,-0.41992188,1.34375,1.0703125,1.28125,0.22167969,1.09375,0.91796875,-0.9296875,1.203125,2.3125,3.5,-0.12011719,-2.15625,-0.515625,0.7890625,-0.16699219,0.83203125,-0.17480469,1.4375,0.24902344,0.34960938,-0.04736328,0.24316406,2.359375,-0.42382812,1.0234375,2.125,-0.19238281,0.74609375,0.076171875
+2,0,5,1.7734375,0.96875,-0.4609375,-0.78125,1.203125,-0.75,-0.22753906,1.953125,-0.38671875,-1.609375,0.796875,0.68359375,-1.3359375,-0.5234375,-0.109375,2.40625,-0.0047912598,-1.6796875,-0.3046875,0.029785156,-1.03125,-0.9765625,-0.04248047,1.375,0.21289062,1.390625,-2.5,-0.6953125,-1.6796875,-0.48828125,-0.22753906,1.8203125,-0.025268555,1.0703125,0.97265625,-1.3515625,1.40625,0.859375,0.80859375,1.3515625,1.1796875,0.34179688,1.1640625,-0.32226562,-0.65234375,0.056396484,-0.5546875,0.24707031,-0.20703125,0.06982422,0.10498047,-1.9375,-0.5859375,-0.42773438,-0.95703125,0.44921875,1.625,-0.022583008,0.6796875,-0.51953125,-0.21679688,1.4296875,-0.9375,0.83203125,-0.13964844,-0.95703125,0.58984375,1.0546875,0.77734375,-0.73828125,-1.0390625,-0.16503906,0.5078125,-0.53125,1.875,1.7578125,1.2890625,0.7109375,-0.36914062,-0.28125,0.55859375,0.44726562,0.16210938,-1.296875,1.0859375,-1.53125,1.1015625,1.015625,0.24414062,1.4765625,0.40234375,-0.69140625,-0.84375,1.046875,-0.42773438,-0.13574219,-1.6484375,-0.06640625,-0.83984375,-0.076171875,0.012207031,-0.3671875,0.6640625,0.22363281,0.083984375,-0.60546875,-1.546875,-0.053466797,-0.30664062,0.022338867,-0.203125,-1.546875,1.8515625,-1.1640625,1.4609375,0.75,-2.125,-0.013977051,-0.484375,-1.578125,0.57421875,-1.0234375,-0.52734375,-0.875,-0.4140625,2.46875,-0.95703125,0.13671875,0.1875,-1.3984375,-0.6953125,-1.59375,0.29882812,0.17285156,-1.3203125,-0.26367188,1.3984375,0.890625,0.27539062,-1.234375,1.5625,-1.734375,-0.90234375,0.7734375,-0.65234375,-0.072265625,-0.64453125,1.125,0.7734375,0.53125,-0.84375,-2.234375,0.98046875,-0.59765625,-0.42773438,0.26171875,0.30859375,-0.89453125,0.62109375,1.1328125,-0.76171875,1.515625,1.03125,1.5546875,-0.092285156,1.4765625,0.53125,0.90625,0.2890625,0.5,-0.69140625,-0.19824219,-0.041503906,0.890625,-0.38671875,0.39453125,0.44726562,-0.46289062,0.083984375,-1.09375,0.83984375,-0.265625,-1.015625,-1.9375,-0.45703125,0.36132812,-1.890625,-0.81640625,0.31054688,-2.046875,0.39648438,1.0234375,-1.4765625,-0.38671875,0.071777344,2.359375,-0.5703125,0.671875,-0.88671875,0.515625
+2,0,6,0.92578125,0.53125,-0.47851562,-0.29296875,0.23242188,-0.30859375,0.5390625,0.94921875,-0.23144531,-0.5078125,-0.17089844,-1.015625,0.13867188,0.93359375,-1.0546875,0.89453125,-0.515625,-2.171875,0.359375,-1.046875,0.8359375,-0.67578125,-0.52734375,-0.27539062,-1.203125,-1.0859375,-1.5546875,-0.37890625,-3.03125,0.26757812,-1.09375,-0.5390625,0.010498047,0.46875,1.2734375,1.3984375,1.078125,-0.65625,-1.84375,2.125,-0.41210938,0.68359375,0.0035552979,0.734375,-0.08642578,0.31835938,0.74609375,-0.76953125,0.74609375,0.013366699,-1.71875,-1.03125,-0.86328125,0.875,-2.921875,0.82421875,0.81640625,-1.09375,-0.03466797,-2.0625,1.15625,-0.29882812,-0.4609375,0.13964844,0.014038086,-0.9296875,-0.61328125,-0.53515625,0.38867188,1.3671875,0.5390625,-1.96875,-1.1796875,-0.87109375,0.020263672,2.078125,1.484375,1.3125,-0.84375,-0.8125,1.0234375,0.41601562,0.09814453,-1.484375,0.21191406,-0.13671875,0.5546875,0.625,-0.98828125,0.6796875,1.1484375,-0.19433594,0.515625,0.3828125,-1.859375,-0.52734375,-0.122558594,0.20019531,1.5234375,-0.19921875,-1.3984375,0.104003906,1.6796875,0.98046875,0.15136719,-0.9921875,2.109375,-0.31054688,0.4296875,0.71484375,0.8046875,0.81640625,1.125,-0.66796875,1.15625,0.020874023,0.2578125,-1.09375,-0.640625,-1.5,0.77734375,-1.0625,0.8671875,-1.25,-0.71484375,1.375,-0.95703125,-2.3125,-0.73046875,0.40820312,1.2109375,-0.31640625,0.47265625,-0.20996094,-1.5546875,-0.5546875,-0.24121094,-0.18457031,0.49414062,-1.6328125,3.125,-0.421875,-0.22460938,0.10107422,-0.106933594,1.796875,-1.671875,0.86328125,-0.91796875,2.140625,0.053710938,-0.34765625,1.078125,-0.9765625,-0.73828125,-0.14257812,0.10839844,0.62890625,0.96484375,1.09375,-0.36523438,1.4453125,1.4296875,1.1484375,-0.48046875,0.83203125,-1.015625,-0.734375,1.78125,-0.048583984,0.99609375,-0.09814453,-0.5703125,-0.609375,1.1796875,0.6015625,-1.0,-0.375,1.0625,-0.93359375,0.00065612793,-0.8828125,0.109375,-0.28125,-0.13085938,0.23925781,1.078125,0.2578125,0.2578125,-1.6875,-0.20703125,-0.40429688,0.8515625,-0.35351562,1.4921875,0.71484375,0.53125,-0.33398438,1.515625,-0.057128906
+2,0,7,1.2734375,0.94140625,-0.05908203,0.140625,-0.060302734,-0.80078125,0.24023438,1.0390625,0.39648438,-0.9609375,0.75390625,1.3359375,-0.6328125,1.140625,-1.40625,1.1015625,0.48632812,-1.375,-0.375,-0.15234375,1.2265625,-0.71875,-0.68359375,0.59765625,0.03466797,-0.06298828,-0.3125,-0.041748047,-1.671875,-0.47460938,-1.1953125,0.118652344,1.53125,-0.3359375,2.125,1.171875,1.3984375,0.625,-2.0,2.359375,-0.875,0.69140625,-0.39257812,0.5703125,0.26367188,0.15429688,-0.17089844,-0.4296875,0.047607422,0.76171875,-1.0703125,0.15625,-0.30078125,-0.0859375,-1.28125,0.9609375,1.015625,0.41015625,0.03540039,-1.0859375,1.71875,-0.24121094,1.3203125,-0.1796875,-2.28125,-0.06689453,-0.6015625,0.16796875,0.28320312,0.93359375,-0.7421875,-1.2578125,-0.6953125,0.3203125,-0.21484375,2.046875,1.7265625,0.40429688,-1.3125,-1.609375,1.234375,-0.43359375,0.65625,-2.203125,-0.89453125,-0.73828125,0.44921875,1.2890625,-1.4609375,0.29492188,1.359375,-0.18261719,-0.6640625,0.515625,-2.03125,0.8203125,0.16308594,1.5078125,0.7109375,1.4140625,-1.765625,-0.05859375,1.4140625,0.8515625,-0.86328125,-0.014404297,1.1875,-1.09375,-1.4609375,0.13476562,0.953125,-0.053955078,2.1875,-0.875,0.27734375,-0.9140625,1.0234375,-0.7890625,-0.31835938,-2.078125,1.8984375,0.23339844,-0.35742188,-1.046875,0.484375,0.56640625,-1.0703125,-2.078125,-0.4765625,-0.11376953,0.49609375,-1.5390625,-0.24707031,0.044433594,-1.90625,1.4921875,0.3046875,-0.48046875,-0.25976562,-0.875,1.765625,-1.6953125,-1.3359375,-0.15332031,-0.25976562,1.25,-1.0703125,1.34375,-1.8125,-0.1875,0.60546875,-0.01953125,0.84375,-1.5,-0.9296875,-0.19042969,0.61328125,-0.44140625,0.6953125,0.8359375,-0.25976562,0.32226562,-0.39257812,1.375,-0.921875,1.5703125,-1.296875,-0.67578125,1.0390625,0.5390625,-1.1484375,-0.6484375,-0.62109375,-0.2890625,0.96484375,0.421875,0.94140625,-0.0024261475,0.265625,-0.96484375,-0.41601562,-0.45117188,-0.18652344,-1.3671875,0.54296875,-1.2890625,0.14648438,0.22949219,0.8515625,-1.4609375,1.1484375,0.6953125,-0.37304688,-0.87890625,-0.41796875,0.12402344,1.265625,1.7421875,1.4921875,-0.45117188
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
index 916e7e56..f64c514e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,5,-0.5390625,-1.1171875,-0.26757812,-0.4453125,0.3984375,0.10986328,0.29492188,0.28515625,0.60546875,-0.703125,-1.421875,-1.4296875,-1.328125,-0.004211426,-0.671875,-0.90234375,0.49804688,0.609375,1.875,0.55859375,0.8203125,0.36328125,-0.13867188,0.78515625,0.32617188,1.125,-0.7421875,1.015625,0.83203125,0.5859375,-1.7265625,-0.06738281,-0.24414062,-0.49023438,0.484375,-0.7890625,-0.25,-1.4296875,-0.06298828,0.08886719,0.34570312,0.3515625,-0.578125,0.90625,0.83984375,-0.7890625,-0.80859375,1.5859375,-0.55859375,-1.796875,0.24804688,1.0,-1.3984375,0.6484375,0.8203125,0.8359375,-1.421875,0.11279297,-0.0005912781,0.16210938,-1.171875,-0.0061035156,1.4296875,0.93359375,-0.22363281,0.84375,0.9453125,0.51953125,0.15917969,-0.24414062,0.6171875,-1.40625,-0.30859375,1.0546875,-2.046875,0.45703125,0.90625,1.2578125,-1.6484375,-1.03125,-0.5,2.5625,-0.5,0.265625,1.046875,-0.24609375,-1.140625,-1.1875,0.003967285,0.7421875,-0.50390625,-0.34960938,0.89453125,-0.9609375,-0.35546875,-0.0390625,1.0625,-1.53125,-1.0234375,0.9453125,-0.7421875,0.9296875,-0.59375,0.94921875,-0.66015625,1.453125,-0.9375,0.087890625,-2.015625,-0.12597656,2.234375,0.27929688,2.125,-0.90234375,0.71484375,-0.03930664,-0.58203125,-0.5703125,0.115234375,0.8671875,0.78125,0.578125,0.34570312,1.3515625,0.6484375,1.125,-0.27929688,-0.11767578,1.15625,0.16015625,-1.0703125,-1.8984375,0.29101562,0.64453125,-0.43945312,-1.15625,1.5625,-0.828125,-1.125,0.14550781,-0.8125,0.41601562,-0.27148438,-0.27734375,-2.3125,-2.09375,0.047607422,0.6171875,-1.703125,-0.78125,-0.45703125,-1.1328125,0.31445312,-0.038085938,-0.77734375,-0.91796875,0.30078125,1.0625,-0.42382812,-0.51171875,0.3046875,0.100097656,-1.484375,-1.3046875,1.5,0.8671875,-0.13867188,0.2734375,-1.515625,2.125,-0.87890625,2.015625,1.8828125,1.1875,0.79296875,0.265625,0.6640625,-0.66796875,0.33984375,2.796875,2.140625,1.78125,-1.9140625,-2.171875,-0.37695312,-1.234375,-0.30273438,-1.328125,0.8671875,-0.29492188,0.032958984,-1.0390625,-0.33007812,0.9765625,0.35742188,0.421875,1.0078125,-1.0546875,0.87109375,0.09082031
-3,0,6,1.1171875,-0.46679688,-0.24707031,0.2578125,1.4296875,-2.078125,-1.5078125,-0.8046875,1.203125,0.011291504,-0.3203125,-0.58984375,-1.1640625,-1.4375,0.83203125,1.5,-0.2890625,0.3671875,0.1015625,-0.49414062,1.4453125,-0.5703125,0.515625,0.40820312,-0.28710938,-0.44726562,-1.7109375,0.23730469,0.33398438,0.036376953,-1.3671875,0.86328125,0.48242188,0.72265625,-0.3828125,0.34765625,0.81640625,-0.34179688,-0.29296875,-1.5625,-0.66796875,-0.40039062,0.65234375,-1.7109375,1.2265625,-1.40625,0.9140625,0.07373047,-0.5234375,-1.453125,2.015625,1.7265625,-0.67578125,-0.39453125,-0.52734375,-0.40039062,1.0234375,0.95703125,-0.73828125,-0.9453125,-0.00091171265,-0.52734375,-0.328125,-1.609375,0.029174805,1.4375,-0.09814453,-0.5859375,-0.64453125,-1.265625,1.125,0.8125,-1.5,0.515625,-1.3203125,-1.9296875,-0.265625,-1.84375,0.609375,1.234375,1.8125,1.8359375,0.45703125,-1.140625,0.06640625,0.050048828,-0.70703125,-0.29101562,-0.6328125,1.34375,-1.609375,1.1484375,0.421875,0.32421875,-0.8125,-0.73046875,1.3984375,-1.359375,0.3515625,0.16113281,1.1171875,0.3046875,-0.34960938,0.32421875,-0.87109375,1.75,1.4609375,0.32421875,0.796875,-0.47460938,1.0859375,0.4296875,-1.2421875,-1.1875,-0.10058594,-0.13183594,-0.8125,0.2421875,1.3125,0.59375,-0.68359375,1.1484375,1.46875,0.94140625,0.072753906,-0.62890625,-0.7578125,-1.5703125,0.66796875,1.4453125,0.27539062,-0.421875,-0.296875,0.640625,-0.81640625,-1.0,1.1875,0.007080078,-0.05102539,-0.609375,0.29492188,0.28710938,1.765625,0.7109375,-1.8984375,-0.018066406,-2.78125,1.234375,1.4921875,0.033203125,0.74609375,-1.078125,0.57421875,0.625,-0.22265625,-2.625,-1.2265625,1.3203125,0.75,0.51953125,-0.44921875,-0.4765625,0.4765625,-0.69921875,0.016723633,0.36914062,-0.22851562,0.29882812,0.25195312,0.30859375,1.1640625,0.41015625,0.55859375,0.9765625,-1.546875,-1.1484375,-0.2890625,1.921875,1.4453125,1.5390625,0.18945312,-0.83984375,-1.34375,-1.9296875,1.9765625,0.09814453,0.14648438,-0.3515625,-0.42773438,0.016723633,-0.45898438,-1.8984375,1.0859375,2.015625,-1.0859375,-0.17089844,1.1484375,-0.41601562,0.3515625,-0.5703125
-3,0,7,0.69921875,-0.6015625,-0.67578125,0.35546875,0.42578125,0.049072266,-0.7421875,0.34960938,0.671875,-1.171875,0.328125,-0.5234375,-0.6015625,0.87890625,0.87890625,-0.84375,0.8359375,-0.048583984,1.6171875,-0.75390625,1.5234375,-0.31054688,-0.16992188,1.5625,3.296875,0.95703125,-0.095703125,1.890625,1.2109375,2.15625,-0.65625,1.9609375,0.734375,-0.43164062,0.15625,0.47265625,-0.71484375,0.007080078,-0.65234375,0.890625,-1.09375,-0.2578125,-0.63671875,0.65625,2.03125,-0.62890625,0.6484375,0.078125,-1.828125,-0.671875,-0.98046875,0.24804688,-0.087890625,0.4375,0.12890625,-0.091308594,-2.078125,-0.16699219,-0.041748047,1.078125,-0.92578125,0.70703125,0.20019531,-0.016479492,0.06542969,0.48632812,1.28125,-1.0703125,-0.89453125,-0.49414062,1.03125,-0.7578125,-0.7109375,0.48242188,-0.9765625,-1.0546875,2.1875,-1.875,-1.7265625,1.578125,-0.078125,1.0546875,-0.7109375,-0.8828125,0.25195312,0.265625,-0.09277344,-0.94921875,-0.021728516,0.29296875,0.16796875,1.5703125,1.1796875,-1.3203125,-1.125,-0.8203125,1.7421875,-0.76171875,-1.5234375,0.40820312,-1.0390625,-0.59375,-1.546875,-0.89453125,-1.109375,0.58203125,0.11816406,1.5859375,0.64453125,-0.50390625,-0.095703125,-0.50390625,-1.1640625,0.91015625,1.609375,-1.40625,-1.3515625,-0.30273438,0.44140625,0.6796875,-0.20507812,0.3671875,0.69140625,-0.45117188,0.032958984,0.31640625,0.44140625,0.59765625,0.18945312,1.359375,0.6171875,-0.45898438,0.86328125,1.390625,-0.6484375,-1.5703125,0.671875,-1.0078125,-0.75390625,0.09375,1.78125,-0.53125,0.9921875,0.84375,-0.44140625,-0.9375,-0.890625,-1.5859375,-1.3125,-0.5390625,1.046875,-2.40625,1.7421875,-0.15234375,-0.21484375,-0.29101562,-1.15625,0.47070312,0.30859375,-1.3515625,-0.36328125,-0.33984375,-1.3515625,-0.609375,0.46484375,0.12890625,-1.5078125,1.4921875,-1.0234375,0.87109375,-0.10449219,1.8515625,0.91796875,0.23242188,-1.28125,0.22753906,0.984375,0.8046875,1.1875,1.109375,-0.14355469,1.6796875,-0.48046875,-2.75,1.6015625,-0.06347656,-1.4609375,-1.515625,-0.5703125,-0.34765625,0.3515625,-0.66015625,0.09423828,-0.72265625,0.796875,-1.078125,0.6796875,-0.41601562,0.31640625,-0.87109375
-4,0,5,0.421875,0.3125,-1.03125,0.5859375,-0.43359375,-1.0390625,-1.9453125,-0.5546875,1.125,-0.69921875,0.22167969,-0.18261719,-0.66015625,0.032958984,-0.6796875,0.52734375,0.25585938,0.47851562,0.65625,0.65625,1.6484375,-0.46484375,0.41992188,0.33203125,-0.9296875,-0.27539062,-1.3203125,0.16894531,-0.14160156,-0.5,-0.37109375,0.11328125,0.28320312,-0.38867188,-1.0,-0.07080078,-0.45703125,-0.75,-1.09375,-1.515625,-0.47851562,-0.18261719,-0.043701172,-1.0390625,0.53515625,-0.640625,0.5,-1.890625,0.46289062,-1.375,2.984375,1.359375,-1.2421875,1.1640625,-0.37304688,2.0,0.578125,-0.06640625,-0.52734375,0.13476562,1.640625,-0.095703125,0.55859375,-0.26367188,-0.296875,0.515625,-0.34179688,-0.75390625,1.7109375,-1.5390625,0.77734375,-0.265625,0.18457031,0.75,-2.484375,-0.33007812,-0.953125,-0.16308594,-0.54296875,0.47851562,1.703125,1.5390625,-0.89453125,-1.4921875,1.28125,1.2734375,-1.390625,-0.3359375,-1.6875,0.59375,-1.796875,0.51953125,-2.21875,0.49609375,0.03112793,-0.4140625,1.828125,-0.515625,-1.1328125,0.44726562,-0.48828125,1.2265625,0.31054688,0.4375,-0.34375,1.7578125,1.109375,-0.80078125,0.21875,1.1015625,1.8671875,0.9375,-1.34375,-1.5,0.65234375,0.59765625,-2.34375,0.765625,-0.021484375,-1.34375,-1.703125,1.3671875,-1.4140625,0.31835938,0.625,-0.05517578,0.82421875,0.34375,-0.22851562,1.34375,-1.1484375,0.33984375,0.063964844,0.7578125,-0.11035156,-0.37890625,2.671875,-0.040039062,0.3203125,-0.31835938,-0.7578125,-0.58203125,0.27734375,-0.10058594,-0.42382812,-1.671875,-1.3125,0.071777344,0.62890625,-1.2265625,0.6953125,-1.3046875,-1.0390625,1.21875,-1.5625,-0.53515625,-0.68359375,0.98828125,0.7421875,-0.36328125,-1.265625,-0.11669922,-0.52734375,-1.5390625,0.8671875,1.125,0.40625,-0.94140625,-1.1015625,-1.1875,0.41601562,0.34375,1.140625,0.115234375,1.3046875,-0.46289062,1.59375,0.59765625,1.6640625,2.40625,1.953125,0.05102539,-0.28125,-0.359375,0.52734375,0.94921875,-0.82421875,-1.3359375,-1.1484375,-0.07324219,1.1796875,0.80859375,0.28125,1.359375,-0.036376953,0.75,0.18554688,0.734375,0.15527344,0.74609375
-4,0,6,0.7421875,0.29101562,-1.859375,-0.060058594,1.2578125,0.11621094,-1.2578125,-0.5078125,-0.12792969,-1.046875,0.67578125,-2.953125,-1.8203125,0.45507812,0.6015625,0.83203125,1.1796875,0.63671875,0.12890625,0.38476562,1.109375,1.296875,1.3515625,0.6875,-1.1171875,-1.0078125,-1.125,0.45898438,-1.328125,1.0546875,-0.005706787,0.4296875,-0.30664062,-0.96484375,0.67578125,1.4765625,1.859375,-1.1328125,-0.8671875,-1.078125,-0.14648438,0.640625,0.8046875,0.30078125,0.26367188,0.050048828,-0.48828125,-1.3203125,0.96484375,-1.0859375,1.6015625,-0.30273438,-0.22851562,0.013977051,0.84375,1.4609375,0.19824219,-0.8359375,0.2578125,0.47851562,-1.0390625,-0.71484375,1.21875,-0.546875,0.91796875,0.4296875,2.0,-0.03466797,0.28320312,-1.6640625,0.18164062,1.5234375,1.1015625,0.18261719,0.82421875,-0.8671875,-0.78515625,-0.016967773,-0.37890625,1.421875,-0.29101562,-0.15429688,-0.515625,-0.75390625,1.015625,-0.9140625,0.46679688,-1.40625,-1.1328125,1.46875,-0.6484375,0.0016403198,-0.6796875,0.125,1.421875,0.47851562,0.6875,-1.8828125,-0.43945312,0.08935547,-0.4921875,-0.94921875,1.6953125,-0.8828125,1.5703125,0.17871094,1.1015625,-0.71484375,-1.59375,0.28710938,0.8359375,1.21875,0.60546875,0.18457031,0.14746094,1.328125,-0.87890625,1.53125,-0.76171875,0.36914062,-0.68359375,-1.7265625,0.27148438,0.30664062,0.10839844,0.5078125,0.58984375,0.15136719,0.14160156,-0.27148438,-0.8359375,-0.19921875,-0.671875,1.7109375,-0.921875,-0.4140625,0.765625,-0.28320312,0.109375,-1.1484375,-0.24316406,-0.57421875,1.4296875,0.671875,-1.40625,-0.2578125,-1.3984375,1.0390625,1.3203125,0.3984375,1.1640625,-0.87109375,-0.33203125,-0.66015625,-0.3359375,-1.3359375,-0.24707031,0.8828125,0.17871094,-0.34179688,-0.4921875,-2.015625,-0.97265625,-1.6953125,-1.578125,1.1953125,-0.6484375,-2.0625,-1.6953125,-0.25390625,0.55078125,-0.0071411133,0.20898438,1.234375,1.90625,0.24511719,-0.0045166016,-0.62109375,-0.22753906,1.75,1.1796875,0.4453125,-3.09375,-1.7734375,1.4375,2.21875,-0.17871094,-0.30078125,0.00680542,-0.8984375,-0.203125,-0.43359375,0.61328125,1.9921875,0.1640625,-0.02331543,1.25,-0.5859375,-1.75,0.12792969
-4,0,7,0.890625,-0.67578125,0.047851562,-0.083984375,0.5078125,0.9765625,-1.2734375,0.2890625,2.171875,-0.69921875,0.17285156,-0.49414062,-1.8203125,-1.375,-0.8515625,0.46484375,-0.029052734,-0.20605469,1.359375,-0.19335938,1.359375,-0.96484375,1.140625,0.5546875,1.7421875,-1.0703125,-0.98046875,-0.61328125,0.9609375,1.6875,-1.265625,-0.080078125,1.0390625,-1.3125,-0.076171875,0.017333984,-1.2421875,0.57421875,-0.15039062,0.35351562,-0.04321289,-1.0625,0.6171875,0.27734375,0.07763672,-1.3359375,0.61328125,2.71875,-0.953125,-0.087890625,1.1875,0.10839844,-1.1796875,-0.859375,-0.3359375,1.25,-0.4375,0.4609375,0.16992188,-0.734375,-1.171875,0.921875,0.47265625,-0.734375,0.703125,0.16699219,0.9921875,-0.7421875,0.38476562,-1.1171875,0.734375,0.9609375,0.087402344,0.98828125,0.14941406,-1.8984375,0.61328125,-1.0,-1.375,2.015625,0.01159668,1.1875,1.078125,-0.90625,-1.0390625,0.76953125,-0.796875,-1.3984375,-0.3046875,-0.106933594,0.6953125,0.91796875,0.9453125,-0.6328125,0.33984375,0.12792969,1.375,-1.7890625,-0.3828125,0.107421875,-1.0390625,-0.30859375,-0.15722656,0.64453125,-1.1484375,0.2890625,-0.1875,-0.8125,0.048339844,-0.34375,-0.5703125,1.6640625,0.7265625,-1.3046875,0.85546875,0.21777344,-0.3046875,-0.35546875,-0.11816406,-0.11425781,0.4921875,-1.0234375,-0.8984375,-0.7265625,-0.96484375,-1.1015625,-0.1796875,0.53125,-0.73046875,0.41210938,1.453125,-1.53125,0.58984375,0.68359375,1.71875,-0.28125,0.953125,-2.671875,-0.21484375,0.037841797,3.140625,0.001411438,-0.65234375,1.4453125,-0.28515625,0.859375,-2.40625,-2.3125,0.19042969,0.47070312,-1.1484375,-0.6796875,-1.2109375,0.44335938,-0.93359375,-1.3125,-0.109375,0.66015625,1.8515625,-1.8671875,-0.6171875,-0.83203125,0.11816406,-1.84375,-3.15625,0.296875,-0.23925781,0.53125,-0.91796875,1.0390625,1.140625,1.390625,0.18066406,1.703125,-0.6796875,-0.33789062,0.33398438,-0.13867188,-0.734375,0.41992188,1.5,1.3671875,-0.31640625,-0.08496094,1.078125,1.296875,-0.22265625,-0.31445312,0.6015625,-0.5546875,1.2890625,1.1015625,0.8671875,0.69921875,0.6328125,0.018920898,0.48046875,0.80078125,-0.70703125,-0.19140625
+3,0,5,1.5078125,0.8671875,-0.5859375,0.064941406,0.94140625,-0.8125,0.20703125,2.078125,-0.27734375,-0.875,0.91015625,1.6640625,-0.78515625,-0.12988281,-0.36914062,1.921875,0.6875,-1.421875,0.34179688,0.89453125,-0.060791016,-1.1328125,-0.36328125,0.92578125,0.8359375,1.8515625,-2.640625,-0.26953125,-1.1328125,-1.2890625,-0.10253906,1.203125,0.67578125,1.40625,1.828125,-0.8359375,0.7734375,0.37304688,-0.0008163452,0.87109375,0.890625,0.95703125,0.5546875,-0.6015625,-0.625,0.5078125,-1.5234375,-0.29882812,0.119628906,-0.55078125,0.087402344,-1.5625,-0.5703125,-1.40625,-0.5625,0.49609375,1.4296875,-0.033691406,0.051757812,-0.6015625,0.546875,0.640625,0.390625,0.5546875,-1.1875,-1.25,0.85546875,-0.20507812,0.15917969,-0.9453125,-0.6171875,-0.38867188,0.22949219,-0.55859375,2.25,2.390625,1.5859375,0.4375,-0.74609375,-0.39648438,0.53515625,0.515625,0.24902344,-2.125,1.21875,-0.66015625,0.21777344,0.65234375,0.099121094,2.078125,0.86328125,-0.875,-0.31835938,0.8359375,-1.1015625,-0.07080078,-1.3203125,0.71875,-0.1640625,1.2109375,0.51953125,0.103027344,-0.38867188,0.62109375,-0.5859375,-0.78125,-0.91015625,0.3046875,-1.3671875,-0.013793945,0.011352539,-1.0703125,2.109375,-0.97265625,1.375,0.60546875,-1.328125,-0.40429688,-0.44726562,-0.71875,0.6875,-0.40429688,-0.55859375,-0.7265625,-0.78125,2.546875,-1.0,-0.041748047,0.91015625,-1.3125,-0.38671875,-2.96875,0.04345703,0.296875,-1.5390625,-0.099121094,1.28125,0.19238281,0.12890625,-0.82421875,1.140625,-0.94921875,-1.1953125,0.546875,-0.55078125,0.34179688,-1.0,1.1875,0.75,-0.08935547,0.05078125,-2.125,0.18359375,-1.375,-0.60546875,0.84375,0.75,-0.609375,0.44335938,0.98046875,-0.20800781,2.03125,0.1640625,1.2265625,-0.0625,1.15625,0.4375,0.36132812,0.27539062,1.015625,-0.59375,-0.37695312,-1.078125,0.79296875,-0.59375,0.53515625,0.3984375,-0.5234375,0.32617188,-0.27148438,-0.13183594,-0.55859375,-0.57421875,-2.328125,-0.32617188,0.27539062,-1.3671875,-1.234375,-0.0390625,-1.8203125,0.65625,0.24609375,-1.65625,-0.90625,0.064941406,2.84375,-1.171875,0.5859375,-1.140625,-0.11230469
+3,0,6,0.0043640137,1.171875,-0.86328125,-0.65234375,-0.55078125,0.42773438,0.005432129,1.484375,0.734375,-0.37695312,2.28125,0.69921875,0.921875,0.040283203,0.37304688,1.125,0.16015625,-1.109375,0.5078125,-0.5234375,0.7421875,-0.23046875,0.7109375,0.90625,0.19238281,0.6796875,0.3515625,-0.44921875,-1.84375,-2.296875,-0.6953125,2.15625,0.02709961,2.4375,1.78125,-0.52734375,0.6484375,0.9375,0.91796875,0.765625,0.828125,-0.62109375,0.6015625,-1.2734375,-0.017578125,0.45507812,0.16210938,-0.0625,0.44726562,-1.7734375,-1.5859375,1.2109375,-0.6640625,0.328125,-1.4453125,-0.68359375,0.18847656,-1.015625,0.65234375,0.5859375,-0.09472656,1.109375,0.028198242,1.546875,1.421875,-0.71875,-0.640625,0.15039062,-0.17578125,0.359375,-1.0078125,0.14257812,0.20214844,-0.11816406,-0.33203125,2.109375,-0.42382812,0.31054688,-0.21289062,0.024169922,-1.6015625,-0.099121094,0.91015625,-1.3046875,1.8515625,0.030395508,-0.578125,-0.89453125,0.67578125,-0.49609375,0.099121094,-0.22460938,-0.5234375,0.515625,-0.26757812,-1.0546875,0.53125,-0.3359375,-1.2890625,-0.66015625,-0.47851562,0.6328125,2.671875,-1.015625,-0.54296875,-1.6875,-0.5078125,0.8203125,-0.32226562,0.8671875,0.23144531,0.26171875,1.375,-0.33984375,2.390625,-0.044189453,-0.67578125,-1.3203125,-0.51171875,-1.015625,-0.49023438,-0.98828125,0.81640625,-0.79296875,-1.7109375,1.421875,-1.28125,-1.484375,0.98828125,0.83984375,-0.22363281,-0.70703125,0.5234375,-0.10986328,-1.9609375,1.21875,0.76171875,0.5625,-0.47070312,-1.609375,2.234375,-1.046875,-0.96875,0.63671875,-0.5,-0.39257812,-0.32617188,-1.2734375,-0.265625,0.016723633,0.34765625,0.028930664,1.609375,-2.4375,-0.703125,0.4296875,0.21582031,-1.3984375,1.0,-0.41015625,0.70703125,-0.44921875,0.65234375,-0.37109375,1.015625,2.53125,-1.0390625,0.1796875,1.25,-0.22167969,-0.8984375,-0.8046875,-2.03125,-1.875,-0.026367188,0.609375,0.09277344,-0.08203125,-1.4921875,-1.5078125,0.41992188,-0.8671875,-0.19726562,-0.54296875,-0.21875,-0.66015625,-0.6484375,-0.42773438,-0.3828125,0.671875,1.671875,-0.94921875,-0.36328125,-0.68359375,-0.96484375,1.6015625,0.95703125,1.859375,0.7109375,1.515625
+3,0,7,1.09375,1.171875,-0.28320312,0.38476562,-1.5703125,-1.2421875,0.7890625,0.83203125,-0.3515625,-0.40039062,1.984375,0.8515625,0.39257812,0.55078125,0.7421875,0.9296875,-0.90234375,-0.44335938,0.81640625,-0.69140625,1.0625,-0.34570312,2.046875,1.96875,0.110839844,0.91796875,-0.15820312,-1.2421875,-0.82421875,-0.45703125,-0.06542969,0.84375,0.056152344,1.8125,0.57421875,-0.8125,-0.04638672,1.203125,0.80859375,-1.234375,-0.087402344,-1.8046875,1.28125,0.040039062,0.38085938,-0.3671875,0.33398438,-0.58203125,-2.578125,-1.2734375,0.36914062,0.55078125,-1.265625,0.90234375,1.1640625,0.016357422,-0.1796875,-0.18164062,-0.50390625,0.5078125,0.47070312,0.45898438,-2.203125,-1.40625,1.15625,0.59765625,-0.578125,-0.30859375,1.5546875,-0.064941406,-1.890625,0.8828125,-1.0234375,0.32226562,1.1484375,0.49414062,1.5390625,1.0234375,-0.6015625,0.38671875,-1.4453125,0.47265625,1.671875,-0.34570312,1.7578125,-0.9375,-0.16894531,1.1796875,-0.32226562,0.86328125,-0.012878418,-1.046875,-0.90625,0.43945312,-0.59765625,-0.73828125,1.5,0.33203125,-0.26171875,0.037841797,0.0011062622,-1.421875,1.9140625,-0.33789062,-1.4609375,-2.0625,1.140625,-1.328125,0.83984375,0.546875,0.55078125,-1.03125,1.3359375,-0.43554688,0.78515625,-0.32617188,-0.22753906,-0.34375,-0.10498047,-1.6796875,0.7890625,0.43359375,-0.03491211,-2.375,-0.18359375,1.859375,-1.0078125,0.18066406,1.2578125,-1.1328125,1.03125,1.2109375,1.546875,-1.8125,-2.453125,-0.5078125,-1.15625,0.5546875,0.03881836,-0.9921875,0.47070312,0.21875,-0.31054688,0.96875,-0.62109375,0.3125,-1.5859375,0.13574219,0.071777344,0.42773438,-0.06933594,-1.0234375,-0.60546875,-0.07373047,-0.9765625,-1.671875,-0.8671875,-0.73046875,-0.27929688,1.6484375,-0.3515625,-1.5625,-0.26757812,-0.296875,1.9296875,0.110839844,-2.140625,-1.21875,1.5390625,-0.6875,0.16699219,0.87890625,-1.171875,-0.609375,0.27539062,-0.17675781,1.6171875,0.07519531,-0.34570312,-0.17578125,0.8671875,1.3515625,-0.33007812,-0.66015625,-0.1328125,-0.5625,-0.4453125,-0.32226562,1.1953125,1.1796875,0.56640625,0.012329102,-1.3359375,-0.734375,0.34960938,-0.86328125,-0.10205078,0.33007812,1.8828125,0.09667969
+4,0,5,2.625,0.2265625,-0.76171875,1.46875,0.4453125,-0.65234375,-0.095214844,1.1875,-0.06591797,-1.1875,1.6171875,0.85546875,-0.3828125,-0.008972168,-0.828125,0.9140625,0.1875,-0.16210938,0.22851562,0.36328125,0.65234375,-0.25976562,-0.32617188,0.171875,0.54296875,1.0234375,-0.88671875,0.69140625,-2.078125,-0.24414062,-0.671875,1.0078125,-0.05029297,1.625,0.89453125,0.43554688,0.8671875,-0.39257812,0.48632812,1.1484375,1.3203125,0.40234375,0.25195312,0.2421875,0.010009766,-0.99609375,0.74609375,-0.66015625,-0.73046875,-1.9296875,-0.072753906,-0.8671875,-1.8125,-1.09375,-1.9453125,0.046142578,1.5859375,-0.20703125,0.07763672,-0.42773438,0.053710938,0.119140625,-0.79296875,0.9921875,-0.14453125,-2.046875,0.51171875,-0.21777344,0.47460938,-0.3671875,0.008972168,-2.671875,-0.48632812,-0.45898438,1.3125,1.703125,0.984375,0.0074157715,-0.26367188,-1.6015625,-0.9453125,0.484375,-0.33398438,-2.078125,1.9296875,0.43164062,0.27539062,-0.9140625,0.5859375,0.1875,0.84765625,0.5390625,0.5625,0.71875,-0.05444336,-1.59375,-0.78515625,-0.12011719,-0.57421875,0.22167969,-0.25390625,1.1484375,0.1953125,-0.32421875,0.20605469,-2.15625,-0.71484375,1.046875,-0.35351562,1.0234375,0.16894531,1.0,0.025390625,0.74609375,1.78125,1.421875,-0.42773438,-1.90625,-0.30664062,-0.49414062,-0.70703125,-0.90234375,-0.4296875,0.18261719,-1.5703125,1.5390625,-0.796875,-0.703125,0.22753906,-0.36523438,1.3984375,-0.9609375,0.018798828,0.44921875,-1.6328125,0.9765625,0.66015625,-0.2578125,0.58203125,-2.078125,1.625,-1.125,-1.2578125,1.1171875,0.8515625,0.73046875,-2.15625,1.4140625,1.484375,2.03125,-0.34179688,-0.35742188,0.27148438,-0.13964844,0.59375,1.6015625,-1.546875,-0.08251953,0.8046875,-0.43945312,-0.50390625,1.7421875,0.68359375,1.2265625,0.061279297,0.828125,-0.01977539,0.26953125,0.12597656,-0.94921875,-0.3671875,-1.1484375,-0.1015625,-1.75,-0.68359375,-0.3203125,0.23144531,-0.765625,0.37304688,-0.3046875,0.12597656,-1.4765625,-0.59375,-2.328125,0.22070312,1.046875,0.265625,-0.42773438,0.125,-2.15625,0.60546875,-0.6875,0.03149414,-1.0546875,0.5390625,3.1875,-0.7578125,-0.42382812,0.70703125,0.8359375
+4,0,6,0.9609375,1.21875,-0.484375,-1.4609375,1.328125,0.35351562,-1.1953125,0.75,1.1484375,-0.24414062,0.3515625,0.609375,-1.5,0.36523438,0.41601562,-0.72265625,-1.2109375,0.037841797,1.59375,1.2578125,0.42773438,-0.6875,0.34570312,0.28320312,-1.046875,0.036376953,-1.359375,-1.2578125,-2.578125,-0.6640625,-0.7265625,0.46289062,-1.1015625,1.015625,-0.31640625,-0.20214844,0.50390625,0.67578125,0.546875,1.0625,-0.068359375,0.58203125,0.9453125,0.875,0.35742188,-0.13476562,0.6328125,-1.640625,0.48828125,-0.31835938,1.0234375,-0.80078125,0.69921875,0.38085938,-2.140625,-0.21386719,0.40820312,-1.21875,0.32617188,-1.546875,0.19628906,0.78515625,1.359375,-1.0390625,0.625,-1.2265625,0.35546875,-1.7109375,0.15820312,0.93359375,-0.07519531,1.578125,-0.41210938,-1.25,0.11035156,0.671875,-0.048828125,0.5078125,-0.6171875,0.049316406,0.38867188,-1.03125,0.40820312,-0.625,0.20019531,-0.6953125,0.1328125,-1.96875,1.1796875,0.23535156,0.55859375,1.59375,-1.03125,1.75,-0.55078125,-0.6484375,-0.484375,0.65625,-0.88671875,0.014282227,0.05859375,1.3046875,1.421875,-1.453125,-0.53125,0.31835938,-0.3046875,0.75,0.515625,-0.30078125,0.51953125,0.66796875,-0.015625,0.76953125,1.3515625,-0.45507812,-0.39648438,-1.09375,0.18359375,-0.38671875,0.07324219,-0.10058594,-0.5625,0.16503906,-1.09375,1.921875,0.025268555,-1.9453125,1.109375,2.9375,1.2578125,-1.3515625,-1.5,1.3125,-2.0625,-0.78515625,-0.93359375,-0.81640625,0.47265625,-1.7265625,1.9921875,-1.9296875,-1.859375,2.171875,0.84375,-0.6640625,0.4453125,0.75,-0.034423828,1.4140625,0.28320312,-0.8203125,1.0625,-0.7890625,-0.38867188,1.4296875,1.796875,0.039794922,0.32226562,-1.7578125,0.69140625,1.6015625,0.8046875,2.265625,-0.095703125,0.62890625,-1.2890625,-1.828125,1.4609375,-0.65234375,0.9765625,-1.28125,-0.021972656,-0.3203125,-0.27148438,-0.8125,-0.6171875,-0.54296875,0.18359375,-0.030273438,-0.38867188,-0.90625,0.23632812,0.45703125,-0.78125,1.40625,-0.46679688,-0.87890625,1.2109375,-0.99609375,-1.046875,-0.6171875,-0.43164062,-0.53125,-0.75390625,2.140625,-0.45117188,0.6796875,-0.5234375,0.39453125
+4,0,7,1.734375,-0.51171875,-0.18066406,0.24902344,-0.5390625,-0.765625,-0.69140625,0.37695312,-0.92578125,-0.625,0.48632812,1.03125,0.5625,0.640625,-0.1328125,1.3984375,-0.018188477,-0.5078125,1.734375,-0.49804688,1.9140625,0.14355469,-1.1328125,0.060791016,2.578125,0.1640625,-0.5078125,-1.46875,-0.96484375,0.052734375,-0.79296875,1.1796875,0.25585938,0.03112793,-1.2890625,-0.546875,0.6796875,-0.9453125,-2.015625,-0.62890625,0.006866455,-0.82421875,-0.4453125,0.067871094,-0.17578125,0.088378906,0.11816406,-0.80078125,-1.296875,-1.1640625,-0.38671875,-0.84765625,-1.1640625,-0.60546875,-0.43945312,2.53125,1.2421875,-0.033447266,0.0020599365,-0.671875,1.25,-0.35351562,-0.23632812,1.6328125,0.39453125,-0.40234375,-0.19921875,-0.609375,2.9375,-0.4453125,-0.359375,-0.26757812,-0.99609375,0.359375,1.3671875,2.28125,-0.0013809204,0.28125,-1.5078125,-0.37304688,0.56640625,-0.42773438,-0.48828125,-0.50390625,1.6328125,-0.859375,0.82421875,1.328125,0.10888672,1.578125,0.49023438,-0.58984375,0.90234375,0.9296875,-1.921875,1.625,-0.94140625,0.5859375,0.4921875,0.15234375,0.22558594,0.020385742,-0.91796875,0.19726562,-0.34765625,-0.7265625,0.13085938,-2.34375,-0.86328125,0.056396484,0.703125,-0.6171875,0.6328125,-0.40039062,-0.27539062,0.11328125,-0.796875,0.578125,-0.6875,0.06689453,0.5234375,1.0546875,0.19238281,-1.6640625,-1.3828125,0.93359375,-1.34375,-0.45703125,-0.21289062,0.28515625,1.5078125,-0.44140625,0.005706787,-0.74609375,-0.953125,-0.16503906,2.21875,-0.4453125,-1.5703125,-2.234375,2.015625,0.78125,-1.625,0.068359375,-0.59765625,-0.234375,0.053710938,-0.35742188,0.9296875,1.8359375,-0.34765625,-0.80859375,1.34375,-2.359375,-1.2109375,-0.97265625,1.5,0.8671875,0.921875,-0.4609375,-0.5234375,-0.609375,0.0859375,-0.33984375,-0.0051574707,-0.092285156,0.265625,-1.2578125,1.2421875,-0.12695312,0.6875,-0.15332031,-1.234375,0.20019531,0.100097656,1.9453125,0.24414062,0.67578125,1.078125,-0.16699219,-1.0859375,-0.06689453,-0.7421875,-0.037841797,-0.55859375,-0.44726562,0.72265625,-0.78515625,1.8515625,-1.1875,-0.06298828,-1.3515625,0.29492188,-0.87109375,-0.34960938,0.85546875,-0.29296875,2.96875,2.421875,0.43945312
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
index 8e5eb006..c8ece4d5 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,5,0.23828125,-0.07763672,-0.24707031,-0.41210938,0.17773438,0.35742188,1.0,-0.984375,0.13183594,-0.008911133,-0.91796875,-3.046875,-1.890625,-0.22753906,0.40820312,-0.18945312,0.28515625,1.0546875,-0.25390625,-0.21484375,-0.12597656,-0.25585938,0.6015625,0.1875,0.017211914,-0.083496094,-0.057861328,-0.5859375,-1.0234375,-0.8984375,-0.5,0.51171875,-0.41601562,0.33789062,0.56640625,-0.8046875,1.0625,-0.98828125,-1.265625,-1.28125,0.35351562,0.84375,-0.44921875,-0.703125,-0.63671875,-0.40625,-0.20410156,0.26171875,-1.1171875,-0.65625,0.56640625,-0.10888672,-1.3984375,-0.017089844,2.71875,0.55859375,1.0,-0.90625,-0.24121094,0.58203125,-0.33203125,-0.36523438,-1.015625,0.765625,0.73046875,0.55859375,2.546875,-0.05102539,-0.70703125,-1.0703125,0.7421875,-0.23828125,0.24316406,-0.05029297,0.38671875,0.041748047,-1.0078125,0.52734375,0.020507812,-0.54296875,1.6171875,1.203125,-1.421875,-2.109375,0.46289062,0.609375,-1.2109375,0.08691406,-0.12207031,0.9765625,-1.8515625,-1.9765625,2.609375,1.4296875,-0.3515625,1.5703125,1.015625,0.4140625,-1.0546875,-0.14746094,0.044189453,-0.30859375,1.0625,0.7265625,1.703125,-1.15625,0.34765625,-1.359375,-1.3984375,-1.0,1.640625,-0.609375,-0.028686523,-0.58203125,0.69140625,-0.09423828,-0.44335938,0.578125,0.55859375,-0.94140625,1.296875,0.2265625,-0.6328125,2.953125,1.328125,-0.26757812,0.47460938,-0.75390625,-0.53515625,0.16894531,-0.29882812,-1.4140625,-1.6953125,0.34179688,-0.27929688,-0.34179688,1.1484375,-1.3359375,-0.53125,-0.10888672,0.05834961,-1.640625,-0.07519531,-0.39648438,-1.328125,-2.34375,-1.796875,0.20019531,0.24414062,-1.2578125,1.0625,-1.3984375,0.515625,1.578125,-0.25976562,-1.65625,0.12792969,0.65625,-0.13867188,-0.7109375,-0.39257812,-0.6015625,-0.43554688,1.0078125,0.875,-0.3515625,-0.9921875,-0.546875,0.22851562,0.87109375,-0.013793945,1.921875,-0.4140625,0.8203125,0.3203125,1.2890625,1.4921875,0.40234375,0.099609375,2.8125,0.34960938,0.625,-0.609375,-1.0859375,1.1953125,1.6484375,-0.29882812,-0.07861328,1.2265625,-0.4375,0.16113281,-1.3515625,1.328125,3.09375,-0.0008583069,0.74609375,1.4921875,0.21875,-0.41796875,0.05810547
-5,0,6,-0.68359375,0.19140625,0.78515625,1.40625,0.60546875,0.21484375,-1.390625,1.109375,0.9765625,-0.30078125,0.75390625,-0.061767578,-0.796875,-0.49023438,-0.6484375,-1.6015625,-1.5,-0.41992188,0.65625,1.8359375,0.5625,-0.93359375,2.203125,-0.421875,0.27148438,-1.2421875,0.91796875,-1.9921875,-0.81640625,0.97265625,-1.9609375,0.6875,0.0038146973,-1.078125,1.1875,0.578125,0.27929688,1.28125,0.09667969,-0.5703125,-0.4921875,0.64453125,-0.18066406,0.12890625,0.25976562,-0.9453125,0.734375,-1.421875,-2.34375,-0.48046875,-0.29101562,0.37890625,-0.953125,0.6640625,-0.34960938,1.328125,-0.43554688,-0.39453125,1.78125,1.90625,-0.103515625,-1.265625,0.95703125,-0.97265625,0.51953125,-0.50390625,-1.3828125,-0.50390625,0.23535156,-1.7890625,-0.375,1.203125,0.65625,-0.42382812,0.67578125,-0.62890625,-0.0035705566,-1.0546875,-1.296875,0.13378906,1.125,0.80078125,0.890625,-0.5078125,-0.107910156,-1.390625,-0.27539062,-1.3515625,0.69921875,0.80859375,-0.36328125,0.21386719,0.7890625,0.22851562,-0.44140625,-0.65625,0.62109375,-1.7890625,-0.9609375,1.2109375,0.099609375,-0.63671875,1.125,1.5546875,0.49414062,-0.71875,2.203125,0.029541016,-0.984375,0.29882812,-0.5546875,2.40625,-0.7734375,0.43945312,-0.083984375,1.421875,-0.88671875,0.64453125,-1.453125,-0.10253906,0.103515625,1.6875,0.0039367676,2.28125,0.067871094,-1.3515625,1.234375,1.4140625,0.96875,0.90234375,0.97265625,0.65234375,0.91015625,-0.625,-0.75390625,0.33984375,-0.06982422,0.6640625,-0.26757812,0.4453125,0.515625,-0.92578125,-0.24023438,1.0234375,-1.2421875,-0.6796875,-1.3203125,-0.890625,-0.123535156,0.6171875,0.375,-0.044921875,-1.03125,0.049072266,-1.8125,-0.50390625,-1.2421875,-0.4140625,1.1796875,-0.59765625,-0.765625,-1.0234375,-0.41796875,-1.125,-1.2578125,0.98046875,0.46679688,-1.34375,-1.03125,-1.6875,0.92578125,0.6328125,-0.53515625,1.296875,1.5390625,0.55859375,1.703125,0.578125,0.27929688,0.578125,-0.29296875,1.234375,-0.3046875,-1.46875,2.5,1.3828125,-0.36523438,-1.3125,2.171875,-2.109375,-0.3671875,0.29296875,0.15625,-1.1015625,0.30078125,-0.10986328,-1.171875,-0.328125,0.96875,-1.1875
-5,0,7,1.84375,0.55078125,0.58984375,1.203125,-0.5546875,1.4375,-0.88671875,0.75,2.875,-1.4375,0.48632812,1.21875,-1.21875,-0.33398438,-0.44140625,-1.140625,-1.234375,-1.2578125,-0.36132812,0.14746094,1.6484375,0.55078125,1.8203125,-0.27148438,-0.40429688,-0.828125,-0.51953125,-0.82421875,-0.78125,1.8203125,-1.03125,0.1875,1.4921875,-1.125,-0.49804688,-0.19726562,-0.953125,1.0078125,-0.72265625,1.265625,-0.25585938,-0.51171875,-1.71875,-0.61328125,0.08154297,-0.09765625,0.09716797,-0.4765625,-0.92578125,-0.20019531,-0.05493164,0.73046875,-2.25,-0.20507812,1.7265625,2.625,0.24902344,-0.46289062,0.31835938,-0.7734375,1.5703125,0.2890625,-0.40429688,-0.83203125,-0.578125,-0.88671875,0.51953125,-1.8046875,0.953125,-0.41796875,-0.49023438,0.8984375,-0.07324219,-0.15332031,0.51953125,-0.5703125,-0.17675781,0.3828125,-0.9765625,2.453125,0.43945312,-0.12597656,-0.13476562,-0.9453125,0.24511719,-0.65234375,0.096191406,-1.9609375,-0.546875,-0.095214844,-0.97265625,-0.34570312,-1.2734375,-0.29492188,-0.08251953,-0.63671875,0.19238281,-0.98828125,-0.8359375,-0.34960938,-1.890625,0.47851562,0.76171875,-1.546875,0.90625,0.21777344,1.4921875,-0.092285156,-0.33398438,0.5234375,0.77734375,0.7734375,-1.3671875,-0.94921875,1.8515625,0.39648438,-0.53125,0.3671875,1.2421875,-2.203125,-1.4765625,1.8125,-0.00079345703,-0.8125,-0.35351562,-1.6484375,0.00579834,-0.3203125,0.37695312,1.859375,-1.8203125,0.36328125,0.390625,0.025146484,-0.43554688,1.171875,1.203125,-0.43554688,-0.31054688,-0.73046875,1.4765625,-0.546875,-0.13574219,1.8984375,-0.48242188,-0.115234375,-0.51953125,-0.18847656,-0.44335938,0.37695312,0.038085938,1.4453125,-1.234375,0.96875,-1.6015625,-0.44921875,-0.42382812,0.9296875,0.096191406,-0.026977539,-1.5390625,0.3671875,-0.609375,-1.15625,-0.022338867,0.36523438,0.36523438,-0.70703125,-1.609375,-1.921875,0.77734375,0.3984375,0.5,0.6796875,1.484375,-0.24902344,-0.55859375,1.6171875,0.8203125,-0.8828125,0.53125,1.3046875,1.5546875,0.125,0.38085938,2.078125,0.15429688,-1.03125,-0.65234375,-1.1015625,0.546875,2.421875,0.39257812,1.546875,1.109375,-0.17089844,-0.48632812,0.1171875,-0.40234375,0.52734375
+5,0,5,1.890625,1.671875,-0.37304688,-0.3125,-0.74609375,-1.3984375,-0.87890625,0.22949219,0.5625,-1.1875,-0.19140625,-0.40039062,-2.421875,1.6015625,-0.6328125,1.2578125,0.546875,0.16992188,-1.4296875,0.5859375,1.7890625,0.5,-0.42773438,0.004333496,1.15625,-0.22167969,-1.609375,0.095703125,-2.140625,0.640625,-0.8828125,0.19726562,0.578125,0.75390625,0.29882812,0.24023438,1.046875,1.0078125,-0.62890625,0.55078125,-0.20605469,0.104003906,-0.33398438,0.8359375,-0.82421875,-0.03515625,1.0546875,-0.37304688,-0.46289062,-0.47460938,-0.33203125,-1.015625,-0.6328125,-0.390625,-0.5546875,2.375,1.6953125,-0.14257812,-0.15429688,-1.71875,0.56640625,0.7578125,0.38671875,0.37695312,0.98046875,0.4140625,-0.859375,1.0,1.640625,-0.6796875,-1.46875,-1.453125,-2.015625,-0.26757812,-0.19726562,0.47070312,1.0625,1.09375,0.107910156,-0.34765625,1.0234375,-1.3515625,0.9765625,-0.921875,0.45117188,-0.95703125,0.96875,0.42578125,-0.67578125,-0.66015625,2.84375,-0.4140625,1.5546875,1.1953125,-1.328125,0.72265625,-1.0234375,0.58203125,0.10253906,1.046875,-0.38671875,0.13378906,1.0703125,-1.6328125,0.546875,-0.578125,0.41601562,-0.44140625,-0.76171875,0.22265625,0.734375,-0.068847656,1.1953125,-0.05908203,0.53515625,-0.47460938,-0.17773438,0.07910156,0.18847656,-1.1015625,0.890625,-0.9921875,-0.4453125,-1.4140625,-1.3984375,2.484375,-0.40820312,-1.359375,-1.5078125,-1.203125,1.515625,-0.5546875,-1.5234375,0.28710938,-2.265625,1.5625,0.72265625,-0.4765625,-1.5390625,-2.3125,1.1640625,-2.234375,0.2109375,0.30273438,0.8203125,0.125,-1.0390625,0.34375,1.140625,1.828125,-1.2109375,-0.6953125,0.734375,-0.58203125,0.5625,0.15429688,0.41015625,0.06298828,0.21386719,0.49804688,-0.734375,-0.2734375,-0.765625,0.265625,-0.6796875,2.25,-0.107910156,0.19238281,1.515625,-0.57421875,-0.9609375,-1.125,-0.8515625,-1.3046875,0.33007812,1.2734375,1.15625,-0.33984375,-0.36328125,-1.6171875,0.69140625,1.34375,-0.5703125,-1.015625,0.234375,-1.109375,-0.06640625,1.0546875,0.6484375,-1.1171875,-0.1640625,0.34179688,0.10253906,-0.29882812,-0.52734375,1.203125,1.421875,0.23730469,0.73046875,1.1015625
+5,0,6,0.79296875,-0.24023438,-0.79296875,-1.0390625,0.19824219,-1.375,-1.296875,1.3125,-0.17578125,0.6328125,-0.9453125,-0.5078125,0.8359375,0.32421875,-0.8125,1.6328125,-0.18066406,-1.2109375,0.20214844,0.4140625,2.40625,-0.7109375,-1.265625,-0.02355957,0.77734375,-0.07763672,-1.640625,-1.1328125,-1.6640625,1.75,-0.28320312,1.4140625,-2.046875,1.1015625,-0.62890625,-0.13867188,-0.013183594,0.03491211,-0.3359375,-0.578125,0.5546875,0.03515625,-0.79296875,0.31835938,-0.33398438,1.5390625,-0.46484375,-0.71484375,1.0859375,-0.515625,-0.8671875,-3.375,-1.328125,0.2265625,-1.46875,2.03125,1.796875,-0.7421875,-0.65625,-2.03125,-1.0703125,-0.47851562,0.5,1.2734375,1.375,1.5859375,-1.3828125,1.0625,0.5546875,0.41015625,0.671875,-1.8515625,-1.5234375,-0.49023438,0.10595703,0.73046875,0.640625,0.08544922,0.59765625,-0.111328125,1.078125,0.11376953,0.103515625,0.67578125,0.22167969,-0.28320312,-0.23339844,-0.46875,0.060546875,-0.10595703,0.5390625,-0.47265625,2.765625,-0.26757812,-0.98046875,-0.28125,-1.625,1.0703125,-0.007293701,0.4140625,0.15820312,0.1640625,0.05029297,-0.104003906,0.8515625,0.38476562,-0.87109375,-1.6875,-1.1875,-0.91796875,0.60546875,1.0859375,-0.4453125,0.024414062,1.0078125,-2.609375,0.14648438,0.4921875,0.41992188,1.296875,0.24121094,-0.62109375,0.0070495605,-1.359375,-1.6875,0.9375,-1.5078125,-0.625,-0.78515625,0.75,2.328125,-1.546875,-0.76953125,0.5703125,-0.734375,0.1875,0.6484375,0.36132812,0.56640625,-2.015625,-0.12109375,2.46875,0.26367188,1.453125,0.6875,0.32421875,-1.1796875,1.3984375,0.7265625,1.40625,-0.122558594,-2.21875,0.68359375,-0.89453125,-0.38476562,0.40625,1.5234375,0.25390625,-0.07519531,0.59375,0.06347656,1.265625,0.15820312,0.65625,-0.11328125,1.296875,0.43945312,0.075683594,0.015197754,0.44140625,0.22851562,0.65625,-0.015258789,-0.34765625,-0.48632812,1.703125,-0.06982422,-1.2578125,-0.61328125,-0.75390625,-0.546875,0.83203125,-0.58984375,-0.5859375,-0.5390625,0.18554688,0.6953125,-0.15136719,0.14648438,-0.90625,-1.03125,-0.65625,0.53515625,-0.52734375,-0.048339844,1.171875,0.40820312,-0.19628906,1.8984375,1.5
+5,0,7,1.640625,0.83984375,0.90625,0.87109375,-1.1484375,-0.37109375,-1.3671875,-0.29492188,2.5625,-1.546875,1.96875,-0.04345703,-0.18457031,1.3125,0.07519531,-0.390625,-0.2734375,-0.016601562,2.171875,1.21875,0.044921875,-0.58984375,-0.4140625,0.37695312,-0.34765625,-0.1640625,1.078125,-0.48632812,0.4921875,0.013183594,-0.29882812,0.49023438,0.171875,-0.703125,0.9140625,0.42382812,1.296875,0.47070312,-0.104003906,1.34375,-1.7734375,-1.2109375,-1.4765625,1.015625,-0.49609375,-0.984375,0.59765625,-0.53125,0.8203125,-0.36914062,0.029541016,-0.33789062,-0.17578125,0.859375,0.8125,0.625,0.31054688,-0.66015625,-0.796875,0.92578125,1.2421875,-0.20507812,0.47070312,0.06640625,0.40429688,-0.06689453,0.31835938,-1.7890625,-0.75,0.16894531,-0.296875,1.40625,0.57421875,-0.8125,-0.53515625,0.6875,-0.34570312,0.7109375,-1.296875,0.65625,-2.53125,-0.71484375,0.984375,-1.8984375,-0.29492188,-0.22851562,0.17675781,-1.890625,0.37109375,0.16210938,0.69921875,1.015625,-0.31640625,1.0625,-0.53125,0.03930664,-1.6875,1.140625,-0.6015625,0.068359375,-1.265625,0.765625,0.6796875,-1.4921875,0.22070312,1.859375,-0.119140625,-0.064453125,-0.40429688,0.42382812,-0.22851562,0.5234375,1.734375,-0.31445312,0.89453125,0.34179688,-0.6953125,0.033691406,-0.71875,0.12988281,0.609375,-0.042236328,0.46679688,-1.3515625,-2.34375,0.921875,-0.072753906,0.13476562,-0.55859375,1.3203125,1.3203125,0.44335938,-0.8046875,0.22753906,-1.1484375,1.359375,0.26367188,-0.66015625,-0.27734375,-1.515625,1.75,-0.0234375,-1.1484375,1.625,-0.2265625,-0.11376953,1.1171875,-0.07910156,-0.375,0.26367188,-0.90625,-0.75390625,2.609375,-0.85546875,-1.296875,0.51171875,1.359375,-0.86328125,0.36914062,0.49023438,-0.796875,-0.14355469,1.28125,1.4765625,0.16503906,-0.59765625,-1.1015625,-0.7578125,0.50390625,-1.4921875,-1.703125,0.10205078,-3.078125,-0.75,-0.42578125,2.28125,0.875,-0.35351562,-0.62109375,-0.052001953,-0.21875,-0.1796875,-0.8203125,-1.359375,-1.703125,0.15527344,1.1015625,0.14453125,-0.37109375,-0.85546875,-1.5,-0.4453125,-0.18066406,-1.0390625,-2.609375,1.8984375,2.234375,0.2578125,0.69921875,1.2109375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
index a656e4e9..7fb3dcc2 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,5,-0.578125,0.38867188,-0.64453125,0.453125,0.37304688,0.51171875,-0.83984375,0.08105469,1.546875,-1.265625,1.140625,-0.53125,-0.546875,0.41992188,0.13183594,0.2421875,0.15722656,1.9375,1.2578125,0.5,0.828125,0.18261719,0.78125,0.93359375,1.359375,-0.66015625,0.46289062,0.85546875,0.45507812,0.100097656,-0.8515625,0.8046875,-0.5546875,-1.3125,0.11669922,-0.92578125,0.0546875,-0.53125,-1.5859375,0.44921875,-0.26953125,-1.484375,0.20507812,-0.74609375,0.36914062,-0.28320312,0.30078125,-0.65625,-0.71484375,-1.40625,2.0625,0.35742188,-0.90234375,-0.10595703,-1.0078125,2.1875,-1.2890625,-0.8515625,1.375,1.25,1.2890625,0.35351562,0.4453125,-0.44140625,0.85546875,-0.03100586,0.69140625,-0.047607422,-0.54296875,-2.59375,-2.09375,0.14453125,-0.30664062,-0.59375,0.25,-0.96484375,0.23144531,-0.71875,-1.515625,1.0,-0.24511719,2.25,-0.12402344,-1.640625,0.13964844,0.31835938,-1.546875,-1.28125,0.50390625,1.46875,0.84765625,1.234375,0.328125,-0.47070312,-0.82421875,-0.10546875,0.53125,-1.265625,-1.2890625,1.546875,-0.079589844,-0.41992188,-0.19433594,0.1328125,-0.9765625,1.1640625,1.2890625,-1.0546875,-0.44335938,-0.55078125,0.08300781,0.83203125,-1.2890625,-2.328125,0.5859375,-0.072265625,-1.671875,0.75390625,-0.05078125,-0.78125,-0.81640625,0.10253906,-1.1640625,0.73046875,-0.111328125,-0.9453125,-0.49804688,0.6796875,0.34765625,1.7578125,-0.87109375,-0.025512695,1.296875,0.54296875,0.734375,-2.046875,1.265625,-0.296875,-0.86328125,-0.02722168,2.390625,-0.25195312,0.55859375,1.40625,-0.25390625,-0.49414062,-1.9921875,-1.09375,-0.30859375,0.08691406,0.09375,-0.40429688,-0.29101562,0.2265625,-1.109375,1.296875,0.609375,-0.08935547,1.0,-1.9375,-1.9609375,-0.87890625,-1.3203125,-1.1640625,-0.45117188,-0.64453125,-0.671875,0.42382812,-1.6484375,-0.80859375,-0.061767578,0.92578125,1.4296875,1.078125,0.72265625,0.59375,1.6640625,-0.43164062,1.7109375,1.8671875,1.9609375,1.8828125,-1.015625,-1.71875,1.75,0.9453125,-0.001739502,-0.061035156,0.43945312,-1.3046875,0.7265625,-0.71484375,-0.13378906,1.9140625,0.43359375,0.3515625,0.53515625,-0.27539062,0.53125,-0.64453125
-6,0,6,0.51171875,0.107421875,-0.6796875,1.734375,-1.15625,1.6015625,-1.671875,0.1953125,2.953125,-1.34375,-0.010925293,-0.42773438,-1.6171875,-0.22851562,-0.421875,-0.35351562,0.26367188,0.9296875,0.095214844,-0.64453125,1.890625,1.1484375,0.17578125,-0.234375,0.8515625,-0.75390625,-1.46875,-0.66015625,-0.94140625,0.84375,-2.453125,0.58984375,-1.203125,-1.515625,-0.5625,-0.83203125,0.41015625,-0.546875,0.045898438,1.125,0.36523438,-0.15136719,0.4453125,-0.67578125,0.4296875,-0.13378906,-0.22167969,-0.00016403198,0.20117188,-0.80859375,0.90234375,1.7890625,-0.41210938,0.62109375,1.015625,2.125,0.47265625,-0.9765625,0.84765625,-0.06640625,1.9921875,0.013793945,1.2109375,0.11767578,-0.6484375,0.064941406,-0.2734375,-0.87109375,-0.24902344,-0.34765625,-0.39648438,1.5078125,-0.58203125,0.85546875,0.19921875,-1.6875,0.6015625,0.6640625,-1.4765625,0.828125,-0.15332031,2.234375,0.37695312,-0.3515625,0.34765625,0.13671875,-2.234375,-0.35742188,-0.97265625,1.2890625,1.6328125,-0.14257812,-1.875,-0.7578125,0.55078125,-0.640625,0.47070312,-1.234375,-0.08544922,0.59765625,0.578125,-0.27734375,0.41015625,-0.26171875,-0.44140625,0.123046875,0.890625,-0.859375,-0.73828125,2.453125,1.3984375,1.9453125,-0.42773438,-1.734375,-0.28320312,0.80859375,0.017578125,0.61328125,-0.546875,-1.5859375,-1.6015625,1.09375,-0.057861328,0.46289062,-0.23242188,-0.5078125,0.05078125,0.171875,0.8984375,1.0859375,-1.3984375,0.6484375,1.328125,0.31640625,-0.52734375,-0.6328125,1.0078125,0.5390625,-1.109375,-1.15625,0.55859375,0.70703125,-0.079589844,1.3515625,-1.6796875,-1.5703125,-1.3046875,-0.390625,-1.375,-0.09814453,-0.28710938,-0.14941406,-1.4921875,-0.31054688,-0.43164062,-0.9609375,-0.30664062,1.6328125,0.3125,-0.41601562,-0.42773438,0.35351562,0.13183594,-1.609375,-0.703125,1.84375,0.46679688,-0.54296875,-1.1796875,-1.640625,0.62109375,0.6640625,0.80078125,0.79296875,1.8515625,1.2421875,-0.14746094,-0.51953125,-0.33398438,0.890625,2.84375,0.53125,-1.1875,-1.59375,-0.016479492,0.03930664,0.45898438,-0.97265625,0.52734375,-0.19335938,-0.7421875,0.69921875,-0.921875,1.3046875,0.119628906,0.37109375,-1.484375,0.6484375,-1.6953125,0.609375
-6,0,7,1.3515625,-1.140625,1.8984375,0.6640625,-0.7578125,0.32226562,-0.9375,-0.578125,1.78125,-1.8203125,-1.1484375,0.059570312,-1.5703125,-1.0390625,-1.7421875,-0.796875,-1.0546875,-0.98828125,0.92578125,-0.28515625,1.625,-0.12207031,0.56640625,-0.25390625,-1.59375,0.5703125,0.82421875,-0.61328125,-0.6484375,0.68359375,-0.85546875,0.8515625,-1.1171875,-2.203125,-0.044189453,-0.49023438,-1.8359375,2.3125,-0.20410156,1.15625,0.9609375,-0.09326172,-1.2734375,0.4140625,0.88671875,-0.087402344,-1.2265625,0.765625,-1.5546875,-1.65625,-1.7578125,1.59375,-1.5703125,1.578125,1.5390625,1.0078125,-1.375,0.47460938,0.04663086,0.17871094,0.14160156,-0.36914062,1.4453125,-1.359375,-0.5,0.703125,-0.15136719,-0.38476562,1.5390625,1.3671875,0.359375,-0.46289062,0.0025634766,0.32421875,0.31445312,1.6953125,0.63671875,0.32226562,-0.5625,1.2578125,0.8359375,1.0703125,-0.14746094,-0.32617188,1.2421875,-0.91796875,0.73046875,-1.1640625,-0.87890625,-0.72265625,-0.82421875,-0.62890625,-0.9921875,-1.28125,0.52734375,-0.20019531,0.58203125,-0.13085938,0.7890625,-0.0028839111,-2.703125,0.66015625,0.05444336,1.3515625,1.59375,-0.17871094,1.078125,0.44140625,-0.70703125,-0.609375,1.0234375,0.16210938,1.5,-1.140625,1.2578125,0.45898438,0.3203125,-0.640625,-0.26953125,-2.625,0.21582031,0.67578125,0.27734375,0.65234375,-0.77734375,0.76953125,0.453125,1.3046875,1.75,-0.20507812,-1.28125,-0.96875,0.953125,0.671875,-0.26953125,0.6484375,0.44335938,0.05859375,-1.046875,-0.34570312,0.90234375,-0.13671875,-0.16894531,1.2734375,-0.8203125,-1.2734375,1.0546875,0.027709961,-1.671875,-0.140625,0.31445312,-0.45898438,-0.6015625,1.46875,-0.66015625,-1.109375,-0.11376953,1.390625,0.296875,0.0016860962,0.39648438,0.106933594,-0.796875,-0.796875,-0.05419922,0.41992188,1.5625,1.1640625,-2.09375,0.15625,-0.20996094,1.6171875,0.89453125,1.4921875,0.734375,-0.30664062,-0.115234375,1.0078125,0.54296875,0.5859375,-0.83203125,0.22949219,-0.375,-1.3046875,-0.15234375,-0.07324219,0.1484375,-0.38476562,-1.7421875,0.21289062,-0.18066406,1.5859375,-0.56640625,-1.2734375,0.47851562,1.515625,-1.15625,0.76953125,-1.078125,-1.328125
-7,0,5,2.015625,-1.09375,0.07714844,0.09716797,1.4140625,-0.8671875,-1.234375,-0.36328125,2.03125,-0.080566406,0.23632812,-0.48242188,-2.40625,-0.97265625,0.8984375,0.25390625,-0.2734375,0.34375,-0.07373047,0.14941406,2.28125,-0.1484375,1.6171875,0.96875,-0.91015625,0.16601562,-0.90625,-0.75390625,-0.203125,0.76953125,-1.125,0.609375,0.87109375,0.12158203,-0.118652344,0.20996094,0.69921875,0.8984375,-0.84375,-0.93359375,0.4296875,-0.15332031,0.37109375,-0.33398438,1.2734375,-1.46875,1.265625,-0.24121094,-0.921875,-0.9375,1.984375,0.7890625,-0.8515625,-0.96875,-0.625,0.58203125,-0.578125,-0.4296875,-0.41992188,0.515625,-0.6015625,-1.171875,0.33007812,-1.59375,1.4375,1.1328125,-0.31835938,-0.33007812,0.421875,-2.40625,1.1953125,1.2578125,-0.0703125,-0.02746582,-0.46484375,-0.6875,-1.578125,-0.58203125,-0.27734375,2.328125,0.8125,1.5859375,-0.032958984,-0.53515625,-0.09716797,-0.91796875,-0.734375,-0.7265625,-0.43554688,1.9609375,-1.140625,0.5234375,-0.13769531,-0.40429688,0.032714844,0.22265625,1.171875,-1.8984375,0.33984375,0.037597656,-0.421875,-0.625,-0.2265625,0.42773438,-1.453125,1.9140625,0.58984375,-0.51953125,0.36523438,-0.9296875,0.29101562,1.0859375,-0.48632812,-1.6484375,-0.23925781,-0.23144531,-1.359375,-0.13769531,0.29882812,-0.33007812,-0.87890625,-0.265625,1.328125,0.27734375,-0.14355469,-0.19628906,-0.7578125,0.032226562,0.375,1.0859375,0.390625,-0.96875,0.49609375,1.0625,-1.3515625,-1.0390625,1.2578125,0.96875,-0.07080078,0.8125,1.8828125,0.73046875,1.1953125,0.921875,-2.609375,-0.3359375,-2.90625,1.046875,2.0625,0.77734375,1.2265625,-0.51171875,-0.8203125,-0.5078125,-1.0,-1.3984375,-0.22070312,1.7734375,0.4140625,-0.515625,-2.046875,0.007171631,-0.984375,-1.546875,-1.828125,1.1640625,0.32617188,-0.09814453,-1.5,-0.45507812,0.69921875,0.33007812,-0.06640625,0.96875,0.06933594,-1.4453125,-0.51171875,0.65234375,1.2890625,1.7109375,1.1953125,0.7578125,-1.78125,-1.21875,1.390625,0.6328125,-0.25390625,-0.18652344,0.029052734,0.5078125,0.81640625,-0.77734375,-0.17578125,0.47070312,1.125,0.3203125,1.4765625,-0.4296875,0.15722656,-0.72265625
-7,0,6,0.0703125,-0.5078125,-0.34960938,0.5234375,0.58984375,-0.067871094,-1.5,0.4921875,1.109375,0.8515625,0.6796875,-0.5390625,-1.484375,-0.92578125,0.703125,-0.30273438,-0.008605957,-0.03540039,0.83984375,0.17871094,1.3828125,-0.013366699,0.6015625,1.1953125,1.171875,-0.109375,-0.9375,-0.578125,1.078125,1.578125,-1.0234375,-1.2421875,1.171875,-1.140625,-0.020385742,-1.09375,-0.76953125,1.1875,-1.421875,0.94140625,-0.075683594,-1.296875,-0.34960938,1.0234375,1.15625,-1.234375,1.65625,0.8671875,-0.93359375,-0.0046691895,0.9453125,0.99609375,-0.1484375,0.019165039,-0.421875,0.88671875,-0.4921875,-0.115234375,0.3984375,0.39648438,-0.765625,0.21972656,0.15917969,-0.78125,0.90234375,-0.859375,0.20019531,-0.8984375,0.6015625,-1.5625,1.1171875,-0.484375,-0.11425781,-0.65234375,0.69140625,-1.015625,-0.5078125,-0.15429688,-2.734375,0.94140625,0.08984375,1.9765625,0.90625,-0.89453125,-1.0078125,0.70703125,-2.140625,-1.7578125,-0.12988281,0.82421875,0.203125,1.7265625,0.98828125,-1.734375,-1.2421875,-0.8046875,0.44921875,-2.59375,0.3359375,1.7109375,0.40234375,-1.171875,-0.28320312,1.6875,-0.07519531,1.5234375,0.41015625,0.20703125,-1.0859375,-0.012207031,-0.328125,0.734375,0.12597656,-0.58203125,0.33007812,0.29101562,-1.0859375,-1.0546875,0.33789062,0.87890625,0.007873535,0.890625,-0.4375,-0.49023438,-1.8828125,-0.19921875,-0.25390625,0.056152344,-0.984375,1.15625,0.58984375,-1.8203125,0.04296875,-0.3359375,-0.26367188,-0.98828125,1.5234375,-0.57421875,-0.80859375,0.5078125,1.4140625,0.021484375,-0.036376953,0.953125,-1.2578125,-0.36523438,-2.0,-0.14160156,0.063964844,1.0,-0.703125,-1.4609375,-0.84765625,0.44335938,-1.265625,-1.96875,0.080078125,0.67578125,-0.015991211,0.14648438,-1.03125,0.32617188,-0.9296875,-0.71875,-2.390625,2.5,-1.3125,0.984375,0.6796875,1.4375,1.1328125,2.53125,1.4375,1.5625,0.5546875,-0.36132812,-0.578125,0.85546875,-0.06689453,0.8984375,2.515625,0.640625,-1.2734375,-1.34375,-0.16015625,1.21875,0.07470703,-1.1015625,1.0078125,-0.94921875,1.1796875,0.55078125,-0.028808594,0.76953125,-0.53125,0.36914062,0.86328125,-0.67578125,0.83984375,-0.72265625
-7,0,7,0.3046875,-0.51953125,0.060546875,1.0703125,0.26757812,1.125,0.16992188,-1.1484375,0.076171875,-2.203125,0.71484375,-1.765625,-1.3515625,0.6640625,-1.1171875,-0.81640625,2.03125,0.12597656,2.359375,-0.83984375,1.71875,-1.109375,-0.7578125,-0.075683594,1.625,-0.6875,-0.25195312,-0.1640625,-1.5,0.59765625,-0.88671875,1.65625,-0.65625,-1.046875,-0.25585938,-1.21875,0.26367188,-0.22460938,-0.19433594,-0.8984375,0.91015625,0.40429688,0.5078125,-0.36914062,0.18164062,0.110839844,-0.6484375,1.703125,-0.890625,-1.71875,0.65234375,0.69921875,-0.53515625,0.16894531,0.8203125,1.21875,-2.984375,-0.36914062,0.45117188,0.99609375,-0.85546875,0.5546875,0.9765625,-0.546875,0.40234375,1.1171875,1.46875,-0.5,-0.028442383,0.07470703,0.75,-0.33398438,0.6953125,1.3359375,0.022705078,0.68359375,2.25,0.099609375,-1.1328125,0.54296875,-0.24511719,1.5078125,0.203125,-1.0625,-1.03125,0.23925781,-1.8203125,-0.12792969,0.37695312,-0.12158203,-0.828125,-0.76953125,1.921875,0.16699219,-0.5546875,0.70703125,0.25585938,-0.1640625,-1.90625,-0.35546875,0.671875,-0.091796875,1.5,-0.080078125,0.0065307617,-0.9453125,-0.359375,-0.4296875,-0.07080078,-1.015625,0.640625,0.20898438,-0.64453125,0.4375,0.90234375,-0.076660156,-0.53125,-0.671875,-0.15429688,-0.042236328,0.46484375,-0.734375,0.032470703,1.171875,0.99609375,-1.140625,0.90234375,1.6796875,0.890625,-0.088378906,-0.22851562,-0.17578125,-0.19042969,-0.2109375,-0.83984375,-1.5390625,-0.11376953,-1.1328125,-0.7734375,-0.32226562,1.9140625,-0.94140625,-0.028686523,-1.0078125,-0.93359375,-1.8046875,-0.62890625,-1.4140625,-1.4453125,-0.32421875,0.06347656,-0.7578125,-0.6171875,0.15722656,-0.06201172,-1.2734375,-1.609375,1.265625,2.453125,-1.046875,0.7578125,0.17773438,0.11621094,-0.7734375,0.028686523,-0.22265625,0.3984375,-1.5625,-1.3828125,0.21484375,0.25390625,2.140625,1.03125,2.09375,-0.46289062,2.546875,1.2734375,-1.0625,1.2421875,1.2109375,2.1875,2.078125,-1.9921875,-0.32421875,0.20117188,0.7578125,-0.80859375,-1.234375,0.703125,-0.71484375,0.47460938,-0.19726562,0.95703125,-0.24609375,0.89453125,-0.43164062,-0.001914978,0.5390625,-0.0016326904,-1.2109375
+6,0,5,0.94921875,0.62890625,0.4921875,-1.28125,1.1796875,0.37304688,0.73828125,0.33007812,-0.4921875,-1.1328125,0.36328125,-0.44921875,0.33398438,0.9609375,-0.61328125,1.015625,-0.71484375,-2.4375,0.81640625,-1.0703125,-0.09716797,-0.78515625,0.71484375,0.34960938,-1.1015625,-0.48632812,-0.69140625,-0.8828125,-2.875,-0.049804688,-0.91015625,0.3828125,-0.671875,-0.04321289,-0.022094727,0.59375,1.21875,0.42578125,-1.2421875,2.390625,-0.48046875,-0.004211426,0.515625,0.58203125,0.18847656,-0.39453125,0.73046875,-0.83984375,0.5234375,0.55078125,-1.3671875,-0.89453125,-0.27734375,1.2734375,-2.90625,0.40039062,0.17871094,-0.84375,1.2890625,-2.046875,1.71875,0.19726562,-0.62109375,0.73828125,-0.87109375,-0.5078125,-0.87890625,-0.22851562,1.421875,1.4453125,-0.33789062,-0.62109375,0.01940918,-0.30859375,-0.56640625,2.203125,1.0546875,1.1796875,-1.796875,-0.6796875,0.81640625,0.75,-0.11621094,-0.7421875,0.08642578,-0.75390625,0.42382812,1.125,-0.8671875,0.6328125,0.51953125,-0.055419922,-1.015625,0.50390625,-1.6875,-0.12695312,-0.29882812,-0.041259766,0.77734375,-0.7734375,-1.6328125,-0.34375,2.109375,0.94921875,-0.15917969,-0.73046875,0.765625,-0.58203125,0.8359375,0.82421875,1.0234375,-0.13574219,1.0703125,-0.64453125,1.1796875,0.21777344,-0.4375,-1.0625,-0.64453125,-1.984375,0.42382812,-1.0703125,0.875,-1.0546875,-0.03881836,1.0234375,-1.640625,-2.359375,-0.41992188,0.33007812,-0.15332031,0.3125,0.49804688,-0.45507812,-1.078125,-0.111328125,-0.73046875,0.6875,0.73046875,-1.421875,3.28125,-0.60546875,-0.022705078,0.74609375,-1.2734375,1.15625,-0.9140625,0.421875,-1.7109375,1.0234375,-0.609375,-0.859375,1.21875,-0.9453125,-1.328125,-0.71875,0.140625,0.70703125,0.5390625,1.6953125,-0.12890625,-0.030151367,1.890625,1.734375,-0.51953125,1.453125,-1.4296875,-0.984375,1.6953125,0.15722656,0.9296875,-0.3203125,0.37695312,0.51953125,1.578125,-0.0546875,-0.80859375,-0.104003906,0.8984375,-1.03125,0.16113281,-1.2265625,-0.82421875,0.15917969,-0.33398438,0.5859375,-0.05419922,-0.64453125,1.484375,-2.15625,0.87109375,0.6796875,0.29882812,0.4375,1.140625,0.06933594,0.5,0.39453125,1.421875,0.18554688
+6,0,6,1.75,1.09375,-0.24414062,0.59765625,1.21875,0.36914062,-0.49414062,0.010375977,0.13769531,-1.6015625,1.46875,-0.21191406,-0.020385742,0.033447266,-0.34960938,-0.29492188,-0.95703125,-0.6875,1.0859375,1.3203125,0.06591797,-2.1875,-0.092285156,-0.032714844,-1.0078125,-0.74609375,-0.90234375,-0.36523438,-0.38476562,-0.59375,-0.40625,0.27539062,0.8828125,0.41015625,0.12109375,-0.57421875,0.5078125,0.40234375,-0.06542969,1.2890625,0.42773438,0.62109375,0.1875,1.859375,-0.640625,-0.19628906,-0.7890625,-1.34375,0.08886719,0.5703125,1.0625,-1.3125,0.13085938,-0.5625,-1.109375,-1.9609375,0.93359375,1.28125,-0.31445312,-1.0703125,1.6484375,1.734375,1.140625,-0.18066406,-1.2265625,-1.0546875,0.6171875,-2.5,0.12451172,0.40820312,-0.103027344,0.106933594,1.234375,-1.6484375,0.43359375,0.65234375,0.33984375,-0.41992188,-1.578125,0.3125,-0.8203125,0.58203125,0.13769531,-0.11425781,-0.20117188,0.55859375,-2.4375,0.048095703,1.1875,1.0859375,1.03125,0.8203125,1.25,0.44335938,-0.6640625,-1.0078125,-0.66796875,1.828125,-0.096191406,0.390625,0.061767578,-0.41992188,0.578125,0.02331543,-0.59375,0.30664062,-0.40625,0.21777344,-0.040039062,-0.43359375,0.49023438,1.8671875,0.30859375,-0.39648438,-0.13085938,1.0546875,-0.44726562,-0.29882812,-0.64453125,0.49609375,-0.016357422,-1.3125,-1.1171875,-0.12207031,-1.2421875,1.5859375,-0.75,-0.3046875,-0.6015625,0.76171875,0.010803223,-2.359375,-0.24902344,1.5859375,-1.2578125,2.703125,-0.47070312,0.55859375,2.25,-1.2890625,1.3671875,-0.625,-0.7734375,0.890625,0.71484375,-0.11230469,-0.28710938,0.5078125,0.31445312,-0.010009766,-0.01159668,-0.9765625,-1.640625,0.8359375,-1.296875,1.75,0.6875,-0.953125,-0.7109375,1.3515625,0.65625,1.0703125,0.87890625,3.484375,-1.84375,1.34375,-2.171875,0.57421875,-0.25195312,-0.022705078,-1.71875,-1.6953125,0.8203125,-0.78515625,0.92578125,0.20703125,-0.37890625,-1.546875,1.4453125,-0.515625,0.4296875,0.35546875,0.053710938,-0.7890625,-1.453125,1.6796875,-1.0078125,-0.73828125,0.6796875,-0.45117188,1.0390625,0.77734375,-1.1328125,-0.875,-0.6015625,-0.30078125,-1.5625,-0.29882812,0.23144531,0.37304688
+6,0,7,-0.34765625,0.27929688,0.10546875,0.8515625,1.1875,-0.13476562,-0.44726562,0.58203125,0.78515625,-0.734375,0.44726562,1.59375,-0.9296875,-0.20605469,-0.65625,0.22070312,0.03540039,-1.1171875,1.4765625,-0.43554688,-0.265625,-2.015625,-1.1328125,0.2421875,-0.052246094,-0.041748047,1.0546875,0.5703125,-0.17285156,-0.21191406,0.46289062,0.7734375,0.5,-0.29882812,2.796875,0.453125,1.2265625,-0.068359375,-0.9140625,0.9921875,0.44921875,-1.0625,-1.3125,-1.5390625,-0.97265625,0.38476562,-0.35742188,-1.046875,0.62890625,0.6875,1.140625,-0.27539062,-0.69921875,-0.0021362305,-0.18847656,-1.453125,0.8984375,1.34375,0.32226562,0.31054688,0.62890625,0.8515625,-0.9921875,1.578125,-1.0390625,0.08935547,-0.81640625,-2.203125,0.97265625,-0.49804688,0.36132812,1.4609375,1.640625,-1.6953125,0.78515625,0.9140625,0.080078125,-0.5078125,-0.08544922,0.80859375,-1.8203125,0.34375,-0.31054688,-1.4296875,0.55078125,-1.65625,-0.10498047,-0.5546875,1.0546875,0.67578125,0.78125,-0.56640625,0.1015625,1.03125,-0.25585938,-0.27929688,-0.75390625,-0.48828125,-0.828125,-0.003540039,0.80859375,0.36914062,0.75,1.2734375,-0.5625,0.75,1.25,0.53125,0.07128906,0.734375,1.2265625,0.80078125,2.65625,-1.4765625,1.1328125,-0.28320312,-1.65625,-0.1328125,-1.046875,-1.015625,1.1484375,0.58984375,-0.69140625,-0.44921875,-1.265625,2.046875,-0.033935547,1.25,0.6953125,-1.2421875,2.046875,-2.1875,0.52734375,0.8359375,-0.1640625,0.82421875,0.53125,0.14941406,0.4921875,-0.64453125,1.5234375,0.30859375,-1.984375,1.4140625,-0.47070312,-1.125,0.31054688,0.3671875,0.6015625,-1.40625,0.35742188,-0.27734375,2.796875,-0.765625,-0.859375,0.6640625,0.27734375,-2.046875,0.78125,0.296875,0.5234375,0.44335938,0.8984375,0.375,0.8671875,0.765625,0.006866455,0.68359375,-0.828125,0.026367188,-1.125,-1.1640625,-2.3125,-1.484375,-0.051513672,0.97265625,-0.18359375,-1.1015625,-1.671875,-1.0859375,1.1640625,0.33398438,-0.5,-1.765625,-0.94140625,-1.203125,-0.640625,-0.17285156,-1.0234375,1.046875,-0.92578125,-0.75390625,-1.6328125,-1.1640625,-1.296875,0.9609375,-0.984375,0.1640625,0.19726562,1.6484375
+7,0,5,1.5546875,1.359375,-0.57421875,-0.25390625,-0.94140625,-0.37890625,-0.95703125,0.921875,-0.9375,-0.21875,2.265625,-0.26757812,-0.33007812,-0.77734375,-0.72265625,1.9453125,0.22851562,-1.7734375,0.234375,-0.014099121,1.8984375,0.40429688,-0.16796875,0.6796875,0.875,-0.32226562,-0.33789062,-1.75,-2.203125,-0.8125,-1.0078125,2.953125,-0.625,2.265625,0.5078125,-0.14160156,0.41015625,0.140625,0.7734375,0.41015625,1.0703125,-0.119628906,0.6875,0.08886719,0.50390625,-0.14257812,0.04345703,-0.15820312,0.44335938,-2.25,-0.83984375,-0.30078125,-0.9921875,-1.4609375,-1.265625,1.0703125,1.1953125,-0.58984375,0.08642578,-0.3046875,0.014770508,-0.057861328,0.8828125,1.4609375,1.421875,0.19726562,-0.15820312,-0.6484375,0.81640625,-0.28320312,-0.4609375,-0.71875,-1.3203125,-0.48046875,0.16015625,0.6796875,0.20605469,0.20703125,-0.02368164,-0.20996094,-0.6015625,-0.40234375,1.0234375,-0.359375,1.59375,0.19238281,-0.16796875,0.37695312,0.91796875,-0.54296875,-0.040771484,-0.14648438,1.1796875,1.0625,0.06982422,-1.2421875,0.71484375,0.90234375,-1.109375,-0.10205078,0.024658203,0.74609375,1.734375,-1.7265625,-0.69140625,-2.140625,-1.71875,-0.53125,-0.76171875,0.37109375,0.63671875,-0.44921875,0.27539062,0.703125,1.421875,-0.13378906,-0.026855469,-1.2734375,-0.36523438,-0.10058594,-0.328125,-0.09765625,-0.6875,-1.6953125,-1.9609375,1.109375,-2.078125,0.119140625,0.41796875,0.94921875,1.1796875,-0.5859375,-0.09765625,0.024047852,-1.71875,0.53515625,1.796875,-0.008544922,-1.0625,-2.3125,1.046875,0.06591797,-0.6015625,0.68359375,-0.106933594,0.56640625,-1.125,-0.20996094,0.8359375,1.171875,-0.2890625,-0.81640625,1.2265625,-1.9296875,-0.08544922,0.234375,-0.080078125,-0.93359375,0.25585938,-0.3515625,1.015625,-0.28125,0.3125,-0.72265625,0.33789062,3.15625,-0.94140625,-0.41992188,0.41210938,-0.375,-0.012329102,-0.67578125,-0.83984375,-1.3125,0.20410156,1.3828125,0.24804688,-0.44921875,0.88671875,-1.3125,0.84765625,-0.12158203,-0.296875,-1.7890625,0.095214844,0.092285156,0.16210938,-0.032470703,0.48828125,-0.14453125,1.9921875,-1.7265625,-0.78515625,-2.0,0.45507812,1.59375,0.72265625,2.515625,1.328125,0.796875
+7,0,6,1.4296875,1.3984375,-0.46289062,-0.3515625,-1.6015625,-1.1328125,0.66015625,0.61328125,-0.21777344,0.21679688,0.2265625,0.37695312,-0.390625,-0.859375,0.28515625,1.875,-1.1875,-0.76953125,0.265625,0.82421875,1.0,-1.0546875,0.82421875,-0.076171875,1.4140625,0.0068969727,-1.4296875,-0.80078125,0.25585938,-0.5078125,-0.7109375,0.16503906,0.89453125,-0.63671875,-0.07714844,-0.5078125,0.5703125,-0.70703125,-0.76953125,-0.38476562,0.45703125,-0.084472656,0.20605469,-0.10205078,-1.2109375,-0.31640625,-0.056396484,0.34765625,-0.59375,-0.79296875,0.7890625,0.17382812,-3.15625,-0.18359375,1.1953125,2.140625,0.93359375,0.34570312,0.34179688,-0.025878906,-0.92578125,-0.328125,-0.07470703,0.5390625,1.390625,-1.078125,-0.052001953,-0.73046875,0.96484375,0.022583008,-0.7890625,1.2578125,-0.50390625,-1.0234375,2.65625,1.140625,1.2109375,0.1328125,-1.4296875,0.041503906,0.08496094,0.37109375,1.203125,-0.5390625,1.0546875,-1.0,-0.609375,1.0859375,1.203125,0.111328125,-0.27539062,-0.83984375,1.5859375,0.63671875,-1.640625,0.69921875,-0.9296875,1.7421875,-0.94140625,0.9140625,1.078125,1.0234375,-0.8828125,0.65625,-0.19921875,-0.58984375,-1.3515625,-1.421875,-0.38476562,-0.0058288574,-0.75390625,-2.0625,0.97265625,-0.13867188,-0.43164062,0.87890625,-0.45507812,0.7421875,-1.3828125,-0.9765625,-0.095214844,0.24414062,-0.93359375,-1.4453125,0.31640625,1.015625,-1.453125,0.5703125,-0.515625,-0.984375,1.3515625,-0.78125,0.5703125,-1.6171875,-0.39453125,-0.97265625,2.53125,0.09082031,-0.90625,-2.40625,1.140625,0.41796875,-1.1484375,-1.0546875,-0.4296875,-0.51171875,0.20605469,1.2421875,2.734375,1.90625,-0.625,-2.125,-0.07470703,-0.73046875,-1.0078125,-0.765625,1.0,-0.15136719,0.65234375,0.828125,-0.5390625,-0.16699219,-0.171875,-0.43164062,-0.3046875,0.33789062,-0.40820312,-0.8125,0.29492188,1.15625,1.5,0.31445312,-0.40625,0.76171875,-1.53125,1.078125,2.890625,-0.53515625,1.3515625,1.140625,-0.46484375,0.49804688,-0.13574219,-1.5390625,-0.890625,-1.0,0.58203125,-0.4375,1.0625,-0.35742188,-0.2734375,-0.04736328,-0.43554688,-0.36914062,-1.15625,0.52734375,-0.6484375,2.21875,1.7890625,0.079589844
+7,0,7,1.65625,0.68359375,-0.87890625,-1.0234375,0.41015625,-0.9375,-0.44335938,1.4765625,0.6015625,0.35351562,-1.234375,1.53125,-0.46679688,0.9296875,0.6171875,0.7890625,0.609375,-0.5546875,0.87890625,1.640625,0.36914062,-0.31054688,-0.6953125,-0.55859375,-0.8671875,1.75,-0.14941406,-0.041015625,0.3359375,-2.265625,0.7890625,2.140625,-0.20117188,0.049072266,-0.009765625,-0.81640625,0.21972656,1.9609375,0.009460449,0.118652344,-0.38476562,-0.18554688,-0.14355469,0.98046875,-0.10058594,-0.40625,-0.53125,-0.65625,-0.44140625,-0.34570312,0.11328125,-1.4375,0.17578125,-0.953125,-0.115722656,0.6640625,-0.94140625,0.13964844,0.19238281,-0.62109375,-1.0390625,-0.20605469,0.62890625,-0.45703125,0.33789062,0.095214844,0.48828125,-0.38085938,0.4140625,0.953125,-0.31054688,2.125,-0.875,-0.65625,0.734375,1.6796875,0.50390625,-0.5859375,-2.5,0.33203125,0.87890625,0.7421875,1.640625,1.609375,0.043945312,0.06542969,-0.04296875,-0.8984375,-0.859375,1.5234375,0.4921875,0.58984375,0.016113281,1.5078125,-0.0390625,0.000541687,-1.25,0.9765625,0.24023438,0.87109375,1.21875,-0.24609375,-0.4765625,-0.27929688,-0.33398438,0.609375,0.08642578,-0.58984375,-1.4296875,-0.921875,0.33789062,-0.20410156,2.09375,-0.796875,0.64453125,-0.91796875,0.16210938,-0.15625,0.31835938,0.40234375,1.125,0.025390625,-0.96875,-0.6015625,-0.44335938,0.70703125,0.040771484,0.16308594,0.99609375,-0.14160156,0.98046875,-1.9453125,0.95703125,-0.234375,-1.203125,0.55859375,-1.34375,0.3984375,0.58984375,-1.96875,0.053710938,0.16601562,-1.359375,1.5078125,-0.43164062,0.12890625,-0.83984375,2.265625,0.8125,-1.90625,1.2734375,-2.71875,0.19824219,0.09765625,-0.12597656,1.0859375,1.1015625,0.6328125,-1.0078125,0.5546875,-1.7421875,1.875,-1.0390625,0.83203125,1.890625,-0.953125,-0.40039062,-2.625,-0.027954102,1.3359375,-0.31445312,0.7265625,-0.99609375,1.1953125,0.2890625,1.2890625,0.18261719,-1.34375,-0.8828125,0.39257812,-0.31054688,0.43945312,-1.0625,0.14453125,-2.125,0.29492188,-1.6484375,-2.40625,1.34375,-1.2421875,-1.390625,-1.6640625,-2.125,0.39453125,0.059326172,0.625,-0.66015625,-0.23730469,-0.70703125,1.0546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
index d4afd773..28d6a225 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,5,0.6484375,-1.109375,-1.046875,0.515625,0.6015625,0.3203125,-1.4609375,1.8203125,1.2421875,0.6640625,0.265625,-0.5390625,-1.53125,-0.049560547,-0.8828125,0.06225586,0.59765625,-0.43164062,1.046875,0.7421875,0.91015625,-0.40820312,1.0234375,1.2421875,0.85546875,-1.4921875,-0.34375,-0.010437012,0.48046875,1.9453125,-1.1875,-1.46875,1.3671875,-0.42773438,0.30664062,0.57421875,-0.44726562,0.55859375,0.14941406,-0.18066406,-0.828125,-0.17089844,-0.5078125,0.75,0.64453125,-0.92578125,1.40625,-0.546875,-0.984375,-1.7265625,0.953125,0.71484375,-0.91015625,-1.046875,-0.43164062,1.046875,0.66796875,-0.35546875,0.59765625,1.0703125,-0.87890625,0.390625,1.359375,-0.48632812,0.1484375,-0.74609375,1.171875,-0.36523438,0.47070312,-1.0078125,0.62109375,1.2109375,1.0078125,0.17480469,0.19433594,-1.0546875,-0.66796875,0.4453125,-1.90625,0.42382812,0.51171875,1.4296875,0.06933594,-0.14160156,-0.66015625,0.36328125,-1.859375,-1.6875,0.14746094,0.015136719,-0.18945312,1.390625,0.19628906,-0.73828125,0.007385254,-1.1796875,1.4921875,-3.28125,-0.38476562,2.265625,0.1484375,-0.95703125,0.72265625,1.578125,-0.7109375,1.6171875,0.044677734,0.80859375,-0.41015625,0.47265625,0.91015625,1.625,-0.76953125,-1.4921875,0.62109375,-0.9296875,-0.57421875,-0.41796875,0.03955078,0.97265625,-0.22363281,0.44726562,-0.71875,0.98046875,-0.6171875,-0.609375,0.13183594,-0.26757812,-1.0078125,1.9140625,0.58203125,-2.53125,0.14160156,0.453125,-0.78515625,0.23828125,2.046875,-1.890625,-0.84765625,0.88671875,0.7421875,0.93359375,1.0703125,0.5546875,-2.734375,0.16992188,-2.53125,-0.115722656,0.08984375,0.80078125,-0.47265625,-1.2578125,-1.65625,-0.26757812,-0.4921875,-0.71875,0.90234375,-0.16015625,1.078125,0.00089263916,-1.2421875,-1.7734375,-0.3359375,-1.8203125,-2.65625,1.1171875,-0.69140625,-0.3984375,-0.12792969,1.078125,1.3359375,1.2578125,0.66015625,1.5,0.31445312,-0.55078125,0.84375,-0.19042969,-0.37695312,0.99609375,0.27929688,1.109375,-0.54296875,-1.7109375,-0.46484375,0.96484375,-0.11279297,-0.41210938,0.01361084,0.05908203,1.578125,-0.02758789,-0.19042969,0.57421875,1.2578125,-0.30078125,0.38476562,-0.75390625,-0.8671875,0.055419922
-8,0,6,-1.03125,-0.16796875,-0.26757812,-0.25585938,1.21875,0.37695312,0.5078125,0.85546875,0.2578125,-1.6796875,-0.23828125,-1.640625,-1.875,-0.28320312,-0.72265625,-0.33984375,1.1484375,-0.72265625,1.2421875,0.025146484,0.72265625,0.19335938,0.40820312,0.9296875,-0.578125,-0.44726562,-1.109375,0.75,-0.087890625,1.390625,0.063964844,-0.58984375,-0.12988281,-2.1875,-0.0004119873,0.22070312,-0.05419922,0.34960938,-0.40820312,0.40039062,0.87890625,0.17871094,-1.21875,0.7109375,0.65234375,0.34960938,0.16210938,0.6171875,-0.41210938,-2.3125,0.08984375,0.64453125,-1.265625,0.20410156,1.3828125,1.1953125,-0.5625,0.5,0.008605957,0.85546875,-1.6328125,-0.453125,1.1015625,0.012939453,0.10205078,0.5625,2.078125,-0.96484375,1.671875,-0.3515625,-0.08935547,-0.90234375,0.049560547,0.40234375,-0.6484375,0.4296875,0.671875,0.921875,-1.2734375,-0.296875,-0.19433594,1.75,-0.5078125,-0.5703125,1.1171875,-1.0078125,-0.39648438,-2.09375,-0.0023651123,-0.14355469,-1.0625,-0.59375,1.0,-0.796875,-0.87109375,0.53125,0.8203125,-2.546875,-1.5625,0.37890625,-0.12597656,-1.1875,-0.765625,-0.110839844,1.28125,0.93359375,1.09375,-0.3203125,-2.765625,-0.7578125,1.8515625,0.765625,1.25,0.060302734,2.015625,0.875,0.31445312,0.984375,1.296875,0.6875,0.82421875,0.27929688,-0.5078125,0.49414062,0.78125,0.55859375,-0.14746094,-0.87109375,0.6015625,1.0078125,-1.484375,-1.015625,0.42578125,0.66796875,-0.15429688,-1.1953125,0.032470703,-1.875,-0.6328125,0.38867188,0.004699707,0.045166016,0.51953125,0.40234375,-1.28125,-1.078125,-0.4765625,1.6328125,-0.7265625,0.49609375,-0.18652344,-1.5625,-0.29101562,0.55078125,0.035888672,-0.41796875,1.4921875,0.8828125,-0.20507812,0.453125,-1.21875,-0.9375,-1.5234375,-0.73046875,0.25976562,0.5,-0.66796875,-0.40234375,-3.71875,2.515625,0.43359375,1.0,1.9453125,1.1328125,1.3828125,0.056640625,0.90625,-0.13183594,0.96875,1.28125,1.984375,0.75390625,-2.359375,-1.0390625,-0.4296875,1.0234375,0.5625,-1.1484375,0.10839844,-1.171875,-0.80078125,-1.015625,0.07373047,1.84375,-0.030883789,-0.27539062,0.19628906,-1.203125,0.35742188,0.12109375
-8,0,7,1.34375,-1.4296875,-0.57421875,1.3046875,-0.796875,0.16308594,1.0,-0.24804688,3.359375,0.55859375,1.046875,-0.46875,-1.6953125,0.09863281,1.1484375,-1.8125,1.7578125,-1.078125,-1.3359375,-0.12695312,1.3515625,1.03125,0.78515625,-0.58984375,0.22363281,-0.20410156,0.12695312,-0.22070312,-0.13964844,1.7890625,-0.34960938,0.009216309,2.53125,-0.34375,1.109375,0.48046875,-0.390625,1.6875,0.046142578,0.045410156,-1.15625,0.6796875,-1.90625,-1.1875,-0.076171875,-0.33203125,1.0078125,0.030761719,-2.046875,-1.0078125,-0.76953125,0.13964844,-1.5,-0.69140625,1.15625,1.40625,0.5390625,-0.734375,-0.42382812,0.053955078,-0.22558594,0.064453125,0.59375,-0.4765625,0.19824219,-0.07080078,3.515625,-0.8671875,-0.008911133,-0.46875,-0.44140625,1.7109375,0.98828125,0.20117188,-0.6171875,-0.03955078,-0.61328125,0.9375,-1.9765625,2.84375,0.72265625,0.77734375,0.51953125,-0.52734375,-0.6328125,-1.4921875,0.4140625,-1.265625,-0.68359375,0.95703125,-0.24316406,0.110839844,0.28125,-0.53515625,0.671875,-0.2734375,1.1328125,-1.0390625,0.61328125,-0.640625,-1.46875,-0.84765625,0.7265625,-2.265625,0.049804688,-1.0625,0.40625,-1.3984375,-0.21777344,-0.71484375,1.15625,-0.33203125,-0.76171875,-1.2421875,1.0078125,-0.5546875,0.875,0.6484375,0.013244629,-1.0234375,-1.328125,-0.19726562,0.97265625,0.66015625,-1.3203125,-0.81640625,-0.31054688,-1.21875,-0.53125,1.3125,-0.53125,-0.18554688,0.04321289,1.015625,-0.80078125,0.34765625,1.046875,-0.0234375,-0.71484375,0.98828125,2.8125,0.50390625,0.23632812,0.11767578,-1.25,-0.44140625,-0.97265625,0.546875,0.83984375,0.3828125,1.2265625,0.859375,-0.65234375,0.10498047,-0.15332031,-0.1640625,-0.17773438,0.7421875,-0.14746094,-0.16210938,-0.25976562,0.55078125,-0.15625,-1.3359375,-1.375,1.5859375,-0.6796875,-0.671875,-2.265625,-0.765625,0.31054688,-0.23046875,-1.1328125,-0.057128906,-0.43359375,0.3515625,0.40039062,0.8671875,0.80859375,-0.095703125,0.103027344,1.375,0.33398438,-0.65234375,0.71484375,0.88671875,-0.24121094,-1.8125,-1.4765625,-1.0859375,1.0078125,0.47851562,0.86328125,1.4765625,1.0078125,-0.36328125,-0.390625,-0.296875,-1.609375,0.049560547
+8,0,5,1.7890625,0.8515625,0.26367188,-0.18652344,-0.059570312,-0.56640625,-1.6015625,0.59375,-0.08984375,0.57421875,0.33789062,0.31445312,-0.14648438,-0.13574219,-0.07714844,-0.35351562,-1.2109375,-2.171875,0.78515625,0.74609375,1.9453125,-1.28125,-1.265625,-0.16308594,1.609375,-0.068359375,-0.010253906,-0.359375,-0.515625,-0.36328125,-1.2109375,0.22558594,0.65234375,-0.60546875,0.52734375,-1.5234375,0.31640625,-0.52734375,-2.140625,-0.005493164,1.4609375,-0.8359375,-0.59375,0.43164062,-1.7734375,-0.20800781,-0.359375,0.34179688,1.953125,0.328125,0.14746094,-0.33984375,-1.484375,1.1171875,-0.78515625,0.8125,0.44726562,1.390625,0.7890625,-0.640625,-1.125,-1.0390625,-0.087402344,1.1875,1.2109375,0.4140625,-0.41992188,-2.0625,1.6796875,0.765625,0.28125,1.328125,-0.79296875,-1.109375,2.734375,1.625,-0.16113281,0.22949219,-0.35351562,-0.067871094,-1.0390625,0.3984375,-0.012390137,1.078125,0.27734375,-1.5390625,0.049804688,2.453125,0.109375,1.2734375,-0.296875,0.21386719,1.7578125,-0.41601562,-0.9921875,-0.30664062,0.70703125,1.46875,0.578125,-0.19824219,1.4609375,1.046875,-0.91015625,0.15527344,0.27734375,-0.08642578,-0.8125,-0.6796875,0.14453125,0.4609375,-0.24023438,0.10253906,0.008056641,-1.703125,-0.92578125,0.51171875,-0.875,0.58203125,-1.5625,0.203125,0.21582031,0.38476562,-2.171875,-0.984375,-1.1875,0.625,-0.36914062,1.0859375,0.42773438,-0.19824219,1.984375,-0.98828125,0.58984375,-0.546875,-0.65234375,0.23535156,1.015625,0.18164062,-0.21875,-1.953125,1.3515625,2.703125,-0.33007812,-0.671875,-1.484375,0.8203125,0.21777344,0.86328125,1.4921875,0.96484375,0.40039062,-0.24414062,0.95703125,-0.69921875,-1.7890625,-0.38085938,-0.045410156,-0.7890625,0.5703125,-0.20214844,0.37890625,0.33789062,-0.41210938,-1.2265625,1.7265625,0.119140625,0.29101562,-0.18261719,-1.828125,1.734375,0.32226562,0.8359375,-0.703125,-0.9140625,-1.2578125,1.8203125,0.25,-0.578125,0.35742188,-0.8828125,1.515625,0.734375,-0.19628906,-1.765625,-1.671875,-0.96484375,-0.21289062,-0.48046875,0.7890625,-0.14941406,-0.61328125,-1.421875,-0.49023438,-1.2890625,-0.16992188,-0.65625,-1.7109375,1.203125,1.2421875,0.9375
+8,0,6,1.46875,1.421875,0.018798828,0.24511719,0.045898438,-0.7265625,0.734375,0.81640625,1.0546875,-0.8203125,1.5703125,1.234375,-0.8359375,-0.89453125,0.21972656,0.265625,-0.671875,-0.91796875,0.71875,0.515625,-0.05517578,-0.74609375,-0.20214844,1.4765625,-0.029174805,0.58203125,-2.265625,-0.14355469,-0.2578125,0.51953125,-0.84375,-0.21972656,0.4375,-0.796875,1.2734375,0.09326172,0.6328125,-0.38867188,0.025390625,1.40625,-0.45117188,1.0859375,-0.17675781,-1.90625,-1.234375,-1.03125,0.1953125,-0.53515625,1.390625,0.65234375,0.734375,-0.7421875,-0.703125,0.61328125,0.3828125,0.48632812,0.7890625,-1.046875,-1.1796875,0.064453125,1.4453125,0.6484375,-0.18847656,0.96484375,0.44140625,-1.3125,0.51953125,0.21972656,-0.8203125,0.36914062,-0.5703125,0.6484375,0.94140625,-0.62109375,1.703125,0.703125,1.6171875,1.3125,-0.125,0.20117188,-0.4921875,-1.03125,0.7109375,-2.234375,0.31445312,0.020019531,0.46679688,-0.37890625,1.25,0.21484375,0.625,-0.76171875,-0.80078125,1.171875,-1.2109375,-0.65625,-2.71875,0.93359375,-0.73046875,-0.016357422,-0.8125,1.4375,0.13476562,0.091308594,-0.27929688,1.3828125,-0.49804688,0.46289062,-1.1328125,0.32617188,-0.20117188,-0.19433594,1.390625,-0.111816406,1.078125,2.03125,-0.5546875,-0.09765625,-0.46289062,-0.765625,0.328125,-0.78125,-0.35546875,-0.87109375,-1.6796875,2.5,-1.0859375,-0.68359375,0.33007812,-0.890625,1.5234375,-1.546875,-0.16894531,0.029541016,-1.53125,0.45507812,1.4140625,-0.29296875,0.22167969,-2.078125,1.96875,-0.765625,-1.6875,1.625,0.6875,-0.296875,0.83984375,-0.28515625,-0.25585938,1.078125,-1.5078125,-1.546875,0.92578125,-1.421875,0.5625,1.171875,1.4375,-1.3203125,0.7578125,0.092285156,-0.20605469,1.546875,2.0625,1.71875,-0.74609375,0.17480469,-0.56640625,0.34179688,0.31640625,0.23730469,-0.8203125,-0.0859375,-1.890625,-0.3203125,-1.3203125,0.94140625,1.3125,-0.96875,0.44726562,-0.14941406,-1.1328125,-1.4296875,-0.33203125,-1.859375,-0.8046875,0.33007812,-0.22753906,0.20117188,-1.4296875,-2.015625,-0.44335938,1.484375,-1.2890625,-1.1953125,-1.3046875,2.15625,0.875,0.9453125,-0.375,-0.4375
+8,0,7,1.640625,1.9765625,0.36132812,0.13574219,-1.796875,-0.22851562,-1.4921875,1.5546875,1.421875,-2.71875,2.03125,-0.84765625,-0.49609375,1.3125,-0.22363281,0.5234375,0.36523438,0.37304688,1.71875,0.08935547,-0.72265625,-0.111328125,-0.54296875,-0.234375,-0.703125,-0.5859375,-0.35742188,-0.31835938,-0.19824219,-0.6171875,-0.4765625,0.46679688,0.5703125,0.75,0.80078125,-0.14648438,1.7265625,0.68359375,-0.62109375,1.8515625,-1.578125,-0.22265625,0.19042969,2.328125,0.033447266,0.62890625,-0.71875,0.453125,0.052734375,-0.037841797,-0.56640625,-0.14257812,-0.22851562,-0.359375,-0.53125,-0.16796875,0.7265625,0.5546875,-1.1953125,0.33789062,0.38867188,0.875,1.1484375,-0.33984375,1.0703125,-0.11035156,0.3203125,-1.640625,-0.60546875,-0.15039062,-0.44921875,-0.27148438,0.38476562,-0.14746094,-0.94921875,0.734375,0.14453125,0.35546875,-1.6484375,1.2265625,-1.125,-0.75390625,1.078125,-1.8125,0.23242188,-0.12207031,0.48632812,-1.546875,-0.28515625,0.47265625,1.34375,1.53125,-0.12988281,1.0,-0.89453125,-0.23535156,-0.63671875,0.31835938,-0.04248047,-0.16503906,-1.875,0.83984375,1.5,-2.171875,0.53515625,1.3828125,-1.0390625,-0.12695312,-0.107910156,-0.0014190674,0.2265625,-0.15820312,1.6640625,-0.42578125,0.53515625,0.20214844,-0.51171875,-0.17089844,-0.48632812,0.65234375,1.3359375,-0.828125,-0.0040893555,-1.59375,-0.9765625,0.48632812,0.5390625,-1.328125,-0.421875,1.6953125,-0.67578125,1.0234375,-0.9921875,1.0859375,-2.21875,1.390625,0.98828125,-0.4609375,-0.66796875,-2.046875,1.5703125,-1.125,-0.9453125,0.09326172,-0.16699219,0.031982422,0.22851562,0.0035095215,0.0031433105,1.28125,-1.1875,-0.33789062,0.98046875,-0.375,-0.9453125,0.5625,1.0546875,-0.052734375,0.3359375,0.20703125,-0.71875,0.15820312,0.14355469,0.33007812,-0.41015625,0.8125,-0.7421875,0.7265625,0.5390625,-1.9609375,-1.984375,-1.0859375,-2.328125,-0.75,0.5625,2.296875,0.578125,-0.23632812,0.032470703,-0.76171875,0.69140625,-0.19628906,-0.8671875,-0.4140625,-0.734375,0.22460938,0.66796875,1.2109375,-0.23046875,-0.14550781,-0.18359375,-0.53515625,0.22753906,-1.546875,-3.125,2.109375,3.0625,0.58984375,-0.25195312,0.73046875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
index d999f205..4477b4bf 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,5,0.21875,0.12402344,0.19824219,-0.34765625,0.9453125,-0.2890625,-2.46875,-0.33203125,1.125,-1.0546875,-0.1171875,-1.78125,-0.98046875,-1.171875,0.94921875,1.578125,0.60546875,1.2421875,0.12207031,-1.046875,2.109375,-0.578125,-1.21875,1.59375,0.32226562,0.56640625,-0.39648438,1.0859375,-0.52734375,-0.47460938,-1.4921875,-0.10058594,0.33007812,0.19140625,-1.3671875,-0.107421875,1.046875,-0.86328125,-0.96484375,-0.5546875,-0.10986328,-0.421875,0.828125,0.061523438,0.5390625,0.14355469,0.22363281,-0.5234375,-0.4296875,-0.765625,2.703125,0.24316406,-0.13476562,1.9609375,-0.484375,0.91015625,-0.14355469,-0.31835938,-1.390625,-0.5625,0.50390625,-0.34960938,-1.296875,-0.9296875,0.48828125,0.21191406,-0.53125,-1.3125,0.15429688,0.33398438,-0.33398438,0.38867188,-0.5859375,-0.032958984,-1.5078125,-1.4765625,0.6015625,-1.296875,0.3671875,0.70703125,1.2421875,1.0859375,-0.42382812,-1.140625,0.21972656,-0.103027344,-1.1328125,-0.5234375,-0.69140625,0.578125,0.640625,0.84375,-1.0234375,0.68359375,-0.28320312,0.0015869141,1.3125,-0.41992188,-0.84765625,0.123046875,0.51171875,0.05493164,-1.515625,1.3359375,-1.9375,1.40625,-0.44921875,1.4375,0.52734375,0.796875,1.21875,0.42578125,-2.109375,0.62109375,0.55859375,0.66796875,-2.046875,-0.80859375,0.3828125,-0.71484375,-2.28125,2.328125,0.22851562,0.91015625,-0.15234375,0.049560547,0.18457031,0.24902344,-0.118652344,1.1328125,-0.048095703,0.50390625,0.13574219,-0.53515625,-0.40429688,0.28320312,2.375,0.8125,0.06738281,-0.22070312,0.20019531,-0.46289062,1.484375,-0.00014972687,-0.28710938,-1.3984375,-1.3046875,-0.99609375,0.83203125,-0.12597656,0.59765625,-2.84375,0.953125,-0.9453125,0.95703125,-1.25,-0.28515625,-0.390625,1.0078125,0.18847656,-0.37695312,0.16308594,-0.20410156,-0.96484375,-0.06689453,1.7734375,1.1640625,0.49609375,0.29296875,-1.5234375,0.7265625,0.6171875,1.7890625,0.0002937317,-0.953125,0.69140625,0.99609375,-0.453125,2.171875,2.234375,1.890625,-0.30078125,-2.078125,-1.234375,0.10839844,-0.096191406,1.3046875,-1.0078125,-0.40429688,-0.453125,0.41210938,-1.203125,0.47070312,1.734375,-1.9921875,-0.83203125,-1.078125,0.421875,1.234375,-0.03149414
-9,0,6,0.021850586,1.2890625,-0.671875,1.25,-2.15625,1.1953125,-2.15625,0.33007812,3.390625,-1.0390625,0.4609375,0.034423828,-0.92578125,-0.23925781,0.3671875,-0.078125,-0.004180908,1.265625,-1.5546875,-0.26757812,1.5703125,1.3671875,1.5625,-0.85546875,0.765625,-1.765625,-0.9765625,-0.87109375,0.38085938,-0.14355469,-1.734375,0.35351562,-0.88671875,-0.546875,-0.69921875,-0.23339844,0.15625,-0.6484375,-0.5703125,0.55078125,-0.64453125,0.16015625,-0.060058594,-1.671875,2.265625,-0.9609375,0.33398438,-0.33789062,-0.38671875,-0.12402344,0.46679688,1.65625,-0.49804688,-0.640625,1.234375,1.2421875,1.3046875,0.1953125,-0.03466797,-0.41796875,1.8359375,-0.14941406,1.28125,0.625,-0.22265625,-0.5625,1.6484375,-0.29101562,-0.8203125,-1.1328125,-1.0078125,1.8203125,-0.18457031,0.8203125,0.28320312,-1.828125,-0.53515625,0.17773438,-1.25,1.6015625,0.703125,1.078125,0.1015625,-0.39453125,0.37890625,0.90234375,-0.66796875,-0.98046875,-0.77734375,1.0703125,0.18066406,0.55078125,-1.4921875,0.052734375,0.15527344,-1.265625,0.45507812,0.25585938,0.00030136108,1.6640625,-0.44335938,-0.21679688,0.22070312,-1.0,0.20605469,0.484375,0.87109375,0.095703125,-0.37695312,0.39453125,1.359375,0.18164062,-1.6953125,-2.015625,0.078125,-0.26953125,-0.52734375,-0.86328125,0.50390625,-1.671875,-1.15625,0.49609375,0.32617188,0.671875,-0.7890625,0.123535156,0.62109375,-0.96875,-0.5078125,1.453125,-0.85546875,-0.028320312,1.7265625,1.6875,0.15527344,-0.45507812,0.09814453,-0.14648438,-0.76953125,-1.1015625,-0.20996094,-0.25,-0.14257812,0.57421875,-0.8046875,-1.0234375,-1.640625,-0.029296875,-0.49414062,0.38867188,-0.11425781,-0.51953125,-0.3359375,0.640625,-0.9921875,-0.5859375,-0.5234375,2.046875,-0.37304688,-0.86328125,-0.265625,0.18359375,0.25,-1.4296875,-0.12207031,3.171875,-0.053955078,0.60546875,-1.609375,-0.7578125,0.83984375,1.25,0.23046875,1.140625,1.03125,0.91015625,0.265625,1.3359375,-0.88671875,0.85546875,-0.12890625,0.011962891,1.1171875,-1.78125,0.42773438,1.125,-0.87109375,-0.57421875,-0.49609375,-0.55078125,-1.2890625,0.002532959,0.2890625,0.86328125,1.3984375,1.453125,-2.515625,1.25,-1.9453125,0.12792969
-9,0,7,-0.58984375,2.0,0.61328125,1.984375,1.2421875,0.34179688,-1.3671875,-0.65234375,1.5625,-0.8671875,0.5625,0.28515625,-1.09375,-0.29492188,-0.26367188,0.099121094,0.11767578,0.38085938,-0.703125,0.51171875,0.84375,0.65234375,0.17675781,-0.34765625,2.34375,0.24023438,-2.0625,1.046875,2.3125,0.62890625,-1.5390625,0.13964844,1.703125,-0.24804688,-1.1875,0.53125,-0.49023438,-0.109375,-0.44726562,0.24121094,-1.5078125,-0.54296875,0.33203125,-1.96875,1.5546875,-0.15136719,1.015625,-0.328125,-0.69921875,1.46875,0.076171875,1.0234375,-0.43945312,-0.22265625,-0.90625,2.515625,-0.5,-0.033447266,0.7578125,0.103515625,0.9375,0.14746094,-0.40429688,-0.82421875,-0.16503906,0.11621094,2.3125,-0.2578125,-0.38476562,-0.80859375,-1.90625,0.65625,0.16503906,-1.0,0.41210938,-1.2421875,1.21875,-0.73046875,-1.8671875,1.6328125,1.1640625,1.734375,0.025512695,-0.012268066,0.36914062,1.0078125,-1.25,-0.7734375,0.25195312,-0.15527344,1.2421875,2.703125,0.51171875,-1.171875,-1.75,-2.390625,-1.1484375,-1.1171875,-0.25,0.87890625,1.2265625,0.3046875,-0.828125,0.71875,-0.40039062,0.79296875,0.73046875,-0.23535156,-0.7421875,-0.7734375,0.828125,0.46484375,-1.90625,-0.6484375,0.48632812,1.328125,-1.3046875,0.421875,-0.47070312,0.08300781,-0.83203125,0.70703125,-0.35351562,0.5234375,-1.4375,-1.8671875,0.032470703,-0.921875,0.041503906,1.125,0.06347656,0.66015625,1.171875,0.6953125,0.87890625,-2.0625,0.8671875,-1.3515625,-0.95703125,-0.9296875,0.16894531,-0.703125,0.38476562,0.87890625,-0.7265625,0.44140625,-0.08300781,0.20410156,-1.6171875,1.140625,-0.6796875,-0.65625,1.25,0.8828125,-2.296875,-0.421875,-1.4765625,0.099609375,0.88671875,-0.046875,-0.6171875,-0.03491211,0.47460938,-1.1640625,-1.375,0.98046875,-0.25976562,0.58984375,-1.578125,-1.3203125,0.15722656,1.09375,0.98828125,-0.036376953,-0.34570312,-0.7265625,-0.014892578,1.6328125,0.66015625,0.25,0.68359375,0.78515625,0.5234375,-0.059326172,1.0859375,-0.39453125,-0.21875,-0.53515625,-0.69921875,-1.1796875,-0.74609375,1.1484375,-0.13085938,0.734375,-0.5234375,0.75390625,-0.65625,-1.1015625,0.8125,-1.125
+9,0,5,1.7421875,0.86328125,0.08642578,0.15820312,-0.040771484,-0.63671875,0.12792969,1.234375,0.028320312,-2.0,1.1875,0.027832031,-1.6875,-0.94921875,0.26953125,0.6015625,-1.03125,0.66796875,-0.765625,0.12792969,-0.055419922,0.2890625,-0.49804688,-0.484375,-0.30664062,-0.3125,-1.6875,0.110839844,-1.2265625,-0.2265625,-1.5390625,-0.38867188,-1.2421875,0.58984375,-0.061035156,-0.70703125,0.6015625,-0.100097656,0.34570312,0.9609375,0.8046875,0.328125,1.0859375,2.4375,-0.20410156,-0.33007812,-0.09033203,0.026245117,-2.28125,-0.984375,0.48242188,-1.203125,-1.53125,-0.26171875,-1.1171875,0.032226562,1.4609375,1.3046875,-0.7578125,-1.5546875,-0.27734375,1.453125,-0.9140625,0.33984375,0.984375,-0.46875,0.66796875,-0.63671875,0.74609375,0.21972656,-0.40429688,-1.7578125,-0.018676758,-1.15625,-0.053222656,1.4453125,2.109375,-0.07470703,0.27148438,-0.050048828,-0.24707031,0.68359375,-0.078125,-1.703125,0.9765625,0.640625,0.6015625,-2.0,-0.07421875,0.84765625,0.12060547,0.94921875,1.0546875,0.953125,-0.890625,-0.5390625,-1.046875,0.35546875,0.30664062,0.86328125,2.109375,0.41992188,-1.0703125,-0.53515625,0.9765625,-1.0859375,-0.484375,-0.36523438,0.73046875,-0.671875,-0.44140625,1.96875,0.41992188,1.0234375,0.63671875,0.47265625,-0.068359375,0.16503906,-0.51953125,0.875,0.92578125,-1.6953125,-0.74609375,-0.43554688,0.20117188,0.28320312,0.29492188,-0.71484375,-1.515625,-0.009338379,0.76953125,-1.0390625,-0.1640625,1.4453125,-0.64453125,1.0546875,0.12597656,-0.40039062,0.9375,-2.34375,1.6328125,-0.73046875,-1.2890625,0.15332031,1.59375,0.39453125,-1.0234375,0.76171875,2.90625,3.0,-0.89453125,-1.6015625,-2.421875,1.65625,-0.5078125,1.171875,0.0107421875,0.5625,-0.8359375,0.36132812,-0.94140625,0.828125,0.23046875,1.5625,-0.9609375,0.25195312,-0.82421875,0.37304688,0.56640625,-0.98046875,0.23339844,-1.265625,1.640625,-0.86328125,0.026611328,-0.091308594,0.55078125,-0.16699219,0.890625,-0.6875,-0.24023438,0.390625,0.58203125,-0.43945312,-0.34375,1.0703125,0.578125,0.36328125,1.5078125,-2.296875,-0.3203125,-0.671875,-0.24804688,-1.7578125,-1.265625,2.09375,0.09326172,-0.671875,-0.21484375,-0.4140625
+9,0,6,1.1328125,1.78125,0.05908203,-0.7734375,0.39453125,-0.114746094,0.078125,-1.1640625,2.09375,-1.5390625,2.453125,-1.359375,-1.0234375,0.3359375,0.5703125,-0.26171875,-0.88671875,-0.73046875,0.6015625,0.29101562,0.28320312,-0.859375,0.9296875,-0.28710938,-1.1015625,-1.28125,0.2578125,0.033447266,-0.8203125,0.43554688,0.43945312,0.041992188,0.087890625,1.015625,0.65625,-0.37890625,0.9453125,1.9140625,-0.4296875,0.35546875,0.65234375,0.07861328,1.015625,1.28125,-1.3046875,0.17480469,0.1875,-1.4140625,-1.515625,0.6328125,0.40234375,-0.49414062,-0.05517578,0.49609375,0.20214844,-0.32421875,0.84375,1.0078125,0.08691406,-0.32226562,0.83203125,2.09375,-0.84765625,0.87109375,1.2421875,-0.15332031,-0.27539062,-1.125,0.84375,0.875,-1.484375,0.5703125,0.052490234,-0.73046875,-0.44921875,-0.08935547,-0.16601562,-0.67578125,-0.2578125,1.0,0.053955078,0.23535156,0.57421875,1.625,1.0625,-0.6640625,-1.7109375,-0.9609375,0.8828125,-0.20996094,1.6875,-0.62890625,1.5859375,-0.111328125,-0.36914062,-0.19140625,-1.0625,0.47265625,-0.86328125,-1.078125,0.13476562,-0.70703125,1.46875,-0.5859375,0.10498047,0.15039062,-1.5390625,0.55859375,-0.0859375,-0.45507812,0.19628906,2.296875,-0.46484375,-2.0,0.14941406,0.8359375,-1.265625,-0.17773438,-0.36328125,-0.38085938,0.075683594,-1.3125,-0.26757812,-1.578125,-1.71875,2.421875,-1.5546875,-0.27539062,-0.66796875,-0.22851562,1.1796875,-0.032958984,-0.32421875,0.60546875,-1.40625,2.03125,-0.65234375,1.109375,1.046875,-0.8671875,1.359375,0.30859375,-0.55078125,2.234375,0.15234375,-1.5703125,-0.3125,0.029663086,-0.3359375,0.22753906,-0.6796875,-1.5390625,-0.1796875,0.0045166016,-1.140625,0.66015625,0.58984375,-0.8203125,-1.09375,2.15625,-0.28515625,-0.546875,1.3203125,1.6875,-0.94140625,1.890625,-1.0546875,1.4140625,-0.4921875,-0.076171875,-1.3203125,0.107910156,-0.26757812,-1.390625,1.0703125,0.1484375,0.14648438,-2.40625,-0.66015625,-2.03125,1.5234375,0.55078125,-0.48632812,0.1953125,-1.4765625,1.9140625,-1.734375,-1.28125,-0.296875,0.68359375,0.84765625,0.265625,-0.62890625,0.30664062,-0.022216797,-1.59375,-0.73828125,-0.08886719,1.5234375,0.953125
+9,0,7,0.546875,0.875,0.71484375,-2.0625,-0.20117188,0.33203125,1.6484375,-0.9140625,1.4296875,-0.85546875,1.375,-1.3515625,0.859375,-0.30664062,0.2578125,0.890625,-0.08251953,-1.1015625,2.09375,-1.171875,-0.91015625,-1.328125,0.96484375,-0.3828125,-2.203125,-1.1953125,0.83203125,-1.0234375,-1.078125,1.2578125,0.37695312,-0.57421875,-0.53125,0.34570312,1.6796875,0.55078125,0.21582031,0.54296875,-1.7890625,1.4765625,-0.9453125,0.18945312,1.0,1.34375,0.6796875,0.70703125,-0.83984375,-0.88671875,-0.75,0.37695312,-1.8984375,0.703125,-0.8828125,1.8046875,-0.6875,0.6171875,0.90625,0.17480469,-0.27929688,-0.41796875,0.31054688,0.20214844,-1.671875,0.875,0.51171875,1.0234375,-0.29882812,-1.40625,-0.35351562,2.125,-1.40625,-0.7890625,-0.29492188,0.6875,-0.4609375,1.109375,1.7421875,0.0859375,0.03100586,-0.052490234,0.40625,0.8359375,0.28320312,0.7109375,0.59375,-1.40625,-0.20703125,-1.5234375,-0.46875,-0.5625,0.41796875,-0.24121094,-0.78125,-0.58203125,-1.765625,0.40039062,-0.546875,-0.34765625,-0.50390625,-0.30859375,-0.796875,-0.029785156,2.40625,1.171875,0.94921875,1.0390625,-0.609375,-0.022583008,0.4609375,-0.80859375,0.890625,1.2734375,1.0859375,-2.203125,1.15625,-0.27734375,0.12792969,-0.48046875,-0.73828125,0.011169434,1.421875,-0.41015625,1.234375,-1.703125,0.47460938,0.33007812,-2.4375,-1.578125,-0.67578125,0.703125,0.8046875,0.40820312,0.8125,-1.2578125,0.734375,-0.08300781,-0.40820312,0.9375,0.15332031,-0.671875,2.453125,1.96875,-0.984375,1.6015625,-0.83203125,-0.75,0.1015625,0.6953125,-1.7890625,0.609375,-0.53125,-1.015625,0.22851562,-1.25,-1.796875,-0.34179688,0.68359375,-0.67578125,0.029663086,2.375,-1.8125,-0.24316406,1.296875,0.52734375,-0.57421875,-0.16992188,-0.359375,0.98046875,-0.3046875,0.65625,0.36523438,0.65625,-1.21875,-0.66015625,0.49414062,-0.22460938,1.0078125,-1.1015625,-0.25976562,-0.69921875,-0.40234375,-1.6015625,0.3203125,0.28320312,-1.109375,-0.018920898,0.09863281,-0.890625,-0.76171875,-0.020874023,-0.34570312,0.765625,0.13183594,0.0146484375,0.26953125,-0.15625,1.046875,-0.85546875,2.546875,0.60546875
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
index 645bdbe4..3841e07f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
@@ -4,7 +4,7 @@ sequenceId,itemPosition,itemId
 1,10,124
 1,11,124
 1,12,124
-1,13,124
+1,13,113
 1,14,124
 1,15,124
 1,16,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
index a040fdee..58d912fa 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
-2,8,113
-2,9,113
-2,10,113
-2,11,113
-2,12,113
-2,13,113
-2,14,113
-2,15,113
-2,16,113
-2,17,113
-2,18,113
-2,19,113
-2,20,113
-2,21,113
-2,22,113
-2,23,113
-2,24,113
-2,25,113
-2,26,113
-2,27,113
+2,8,103
+2,9,124
+2,10,103
+2,11,124
+2,12,103
+2,13,124
+2,14,124
+2,15,124
+2,16,124
+2,17,124
+2,18,124
+2,19,124
+2,20,124
+2,21,124
+2,22,124
+2,23,124
+2,24,124
+2,25,124
+2,26,124
+2,27,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
index c64d261b..081f39f9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
@@ -19,8 +19,8 @@ sequenceId,itemPosition,itemId
 3,25,124
 3,26,124
 3,27,124
-4,8,113
-4,9,113
+4,8,124
+4,9,124
 4,10,113
 4,11,113
 4,12,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
index f86b83ee..da076443 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
-5,8,113
-5,9,113
-5,10,113
-5,11,113
-5,12,113
-5,13,113
+5,8,124
+5,9,103
+5,10,124
+5,11,124
+5,12,124
+5,13,124
 5,14,113
-5,15,113
-5,16,113
-5,17,113
-5,18,113
-5,19,113
-5,20,113
-5,21,113
-5,22,113
-5,23,113
-5,24,113
-5,25,113
-5,26,113
-5,27,113
+5,15,124
+5,16,124
+5,17,124
+5,18,124
+5,19,124
+5,20,124
+5,21,124
+5,22,124
+5,23,124
+5,24,124
+5,25,124
+5,26,124
+5,27,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
index 0acf1d86..541bd2be 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
@@ -1,41 +1,41 @@
 sequenceId,itemPosition,itemId
-6,8,113
-6,9,113
-6,10,113
-6,11,113
-6,12,113
-6,13,113
-6,14,113
-6,15,113
-6,16,113
-6,17,113
-6,18,113
-6,19,113
-6,20,113
-6,21,113
-6,22,113
-6,23,113
-6,24,113
-6,25,113
-6,26,113
-6,27,113
-7,8,113
-7,9,113
-7,10,113
-7,11,113
-7,12,113
-7,13,113
+6,8,124
+6,9,124
+6,10,124
+6,11,124
+6,12,124
+6,13,124
+6,14,124
+6,15,124
+6,16,124
+6,17,124
+6,18,124
+6,19,124
+6,20,124
+6,21,124
+6,22,124
+6,23,124
+6,24,124
+6,25,124
+6,26,124
+6,27,124
+7,8,124
+7,9,124
+7,10,124
+7,11,124
+7,12,124
+7,13,124
 7,14,113
-7,15,113
-7,16,113
-7,17,113
-7,18,113
-7,19,113
-7,20,113
-7,21,113
-7,22,113
-7,23,113
-7,24,113
-7,25,113
-7,26,113
-7,27,113
+7,15,124
+7,16,124
+7,17,124
+7,18,124
+7,19,124
+7,20,124
+7,21,124
+7,22,124
+7,23,124
+7,24,124
+7,25,124
+7,26,124
+7,27,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
index 236a8361..79dc2ec9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
-8,8,113
-8,9,113
-8,10,113
-8,11,113
-8,12,113
-8,13,113
-8,14,113
-8,15,113
-8,16,113
-8,17,113
-8,18,113
-8,19,113
-8,20,113
-8,21,113
-8,22,113
-8,23,113
-8,24,113
-8,25,113
-8,26,113
-8,27,113
+8,8,103
+8,9,124
+8,10,103
+8,11,122
+8,12,122
+8,13,122
+8,14,122
+8,15,122
+8,16,122
+8,17,122
+8,18,122
+8,19,122
+8,20,122
+8,21,122
+8,22,122
+8,23,122
+8,24,122
+8,25,122
+8,26,122
+8,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
index c97879eb..31db896c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId
-9,8,113
-9,9,113
+9,8,103
+9,9,124
 9,10,113
 9,11,113
 9,12,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
index 7e8daf90..460f1499 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,113
+2,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
index fc6d272e..d279d22f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 3,8,124
-4,8,113
+4,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
index 4e9004da..81899090 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,113
+5,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
index 953b3bbc..25ad0933 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,113
-7,8,113
+6,8,124
+7,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
index 346e8242..bc91b117 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,113
+8,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
index ba964099..b949bc17 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,113
+9,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
index 76b336da..8892a7bd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-1,6,102
+1,6,122
 1,7,111
-1,8,121
+1,8,117
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
index 613c7045..8812ffc8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-2,6,121
+2,6,122
 2,7,102
 2,8,126
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
index 2d77f92b..60c4449f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId
-3,6,121
+3,6,122
 3,7,109
 3,8,124
-4,6,103
+4,6,127
 4,7,119
 4,8,120
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
index 18bdec24..285c20cb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId
 6,6,102
 6,7,118
-6,8,118
+6,8,113
 7,6,109
-7,7,121
-7,8,105
+7,7,103
+7,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
index c63e7d33..6475d519 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-8,6,121
-8,7,121
+8,6,104
+8,7,122
 8,8,112
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
index 268a2feb..d566f05c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-9,6,110
+9,6,118
 9,7,118
 9,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
index 18e88ed1..52479655 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 3,8,113
-4,8,113
+4,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
index dd4911cc..799db1cf 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-0,6,124,3,7
-0,7,118,4,7
-0,8,121,6,1
+0,6,119,5,7
+0,7,103,4,5
+0,8,111,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
index 056092e7..cbe044a8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-1,6,109,4,3
-1,7,104,9,1
-1,8,124,6,2
+1,6,122,6,1
+1,7,121,9,1
+1,8,108,9,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
index 735dfc03..c9b5b93b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
 2,6,121,6,1
-2,7,102,3,0
-2,8,121,6,1
+2,7,102,6,5
+2,8,112,3,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
index 041a7dbb..c39abddc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
 3,6,121,6,1
 3,7,109,3,1
-3,8,102,0,0
-4,6,124,2,1
-4,7,111,3,1
-4,8,102,4,8
+3,8,110,2,6
+4,6,102,6,1
+4,7,119,1,1
+4,8,102,0,6
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
index 8631d3ed..f6da44a0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
 5,6,126,6,1
-5,7,104,4,0
-5,8,112,4,5
+5,7,113,6,0
+5,8,121,4,5
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
index 1aa02270..71855096 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-6,6,102,4,8
-6,7,118,4,5
-6,8,102,4,6
+6,6,102,5,5
+6,7,118,5,1
+6,8,109,5,6
 7,6,109,3,1
-7,7,121,4,8
-7,8,102,1,0
+7,7,121,2,6
+7,8,119,5,7
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
index 1f26c457..03e17fbc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-8,6,121,4,1
+8,6,102,8,6
 8,7,121,6,1
-8,8,109,4,1
+8,8,121,3,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
index ff81b9fb..59439ca4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-9,6,124,3,1
-9,7,118,4,8
+9,6,122,6,1
+9,7,118,4,4
 9,8,103,4,8
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
index 7acc0370..5aa808f6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,122
+0,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
index dea53fa9..c3d900b3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,122
+1,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
index 8302df2f..7e8daf90 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,122
+2,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
index c780b2f6..18e88ed1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,122
-4,8,122
+3,8,113
+4,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
index c71ccfa4..953b3bbc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 6,8,113
-7,8,122
+7,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
index f3ae7b5b..346e8242 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,122
+8,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
index 405b8711..9576f383 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,121
+0,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
index ac1c42b4..72fb79cf 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,117
+1,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
index 04dd0078..598df3ba 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,121
+2,8,110
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
index 6aeaeb13..04c6c551 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,119
-4,8,102
+3,8,115
+4,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
index 431d8b95..3e9b1443 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,119
+5,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
index b8bead8a..f97711ff 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,119
-7,8,119
+6,8,115
+7,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
index 13695e6e..41144ce1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,119
+9,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
index 405b8711..9576f383 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,121
+0,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
index 04dd0078..598df3ba 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,121
+2,8,110
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
index 6aeaeb13..04c6c551 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,119
-4,8,102
+3,8,115
+4,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
index 431d8b95..3e9b1443 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,119
+5,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
index 8a664645..f97711ff 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,121
-7,8,122
+6,8,115
+7,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
index 13695e6e..41144ce1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,119
+9,8,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
index 4e261efe..a7c99704 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,8,110,4,-0.057408422
+0,8,109,6,0.035670765
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
index 0eefb672..d440d67e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,8,103,7,-0.21608433
+1,8,102,1,-0.094851345
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
index aa696c55..741217c7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,100,7,-0.20284554
+2,8,mask,9,-0.12064132
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
index 3ff84f45..fc5d931d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,8,103,0,0.074112006
-4,8,110,0,0.012103111
+3,8,103,4,0.08533255
+4,8,110,4,0.072973825
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
index 2003e2b9..7d6c2949 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,8,111,7,-0.1722543
+5,8,110,5,-0.04847882
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
index 45c54f06..9c8622d1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,8,122,3,-0.034481153
-7,8,114,4,-0.022850327
+6,8,122,7,-0.012625184
+7,8,113,4,0.0941027
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
index 7a2374cb..d71ac4b8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,111,3,-0.088609315
+8,8,110,7,-0.03440542
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
index 24299a0c..9a89327f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,8,115,4,-0.001879178
+9,8,114,6,0.0070434585
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
index 0c925c0a..b87c3693 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,8,113,4,-0.057408422
+0,8,102,6,0.035670765
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
index 97d2b3a3..d672e857 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,8,117,7,-0.21608433
+1,8,100,1,-0.094851345
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
index f91f4a65..cff9a2a5 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,117,7,-0.20284554
+2,8,100,9,-0.12064132
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
index 38f40f6a..e3a13c6f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,8,108,0,0.074112006
-4,8,103,0,0.012103111
+3,8,111,4,0.08533255
+4,8,109,4,0.072973825
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
index 7413eb80..dcae774d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,8,104,7,-0.1722543
+5,8,102,5,-0.04847882
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
index 1ca260e9..0ab95c0d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,8,116,3,-0.034481153
-7,8,108,4,-0.022850327
+6,8,110,7,-0.012625184
+7,8,102,4,0.0941027
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
index b6ad1aac..ded1b8f0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,116,3,-0.088609315
+8,8,mask,7,-0.03440542
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
index fbb74fe9..b1aaa3c1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,8,113,4,-0.001879178
+9,8,111,6,0.0070434585
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index fb7b7172..08aef492 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.16013545
-0,22,0.40992618
-0,23,0.29611352
-0,24,0.63355803
-0,25,0.34203795
-0,26,0.21724334
-0,27,0.080967665
-0,28,-0.18309766
-0,29,0.05850465
-0,30,-0.018368812
-0,31,0.057256702
-0,32,0.11041919
-0,33,-0.06628993
-0,34,0.10592658
-0,35,-0.11471023
-0,36,-0.0600502
-0,37,-0.09773817
-0,38,0.24320062
-0,39,0.05201533
-0,40,-0.08925214
-1,19,0.027680388
-1,20,0.28413326
-1,21,0.091450416
-1,22,0.21424827
-1,23,0.2252302
-1,24,-0.029475529
-1,25,-0.039583888
-1,26,-0.05680554
-1,27,-0.0011637344
-1,28,0.11441261
-1,29,-0.03658882
-1,30,0.06798903
-1,31,-0.014063398
-1,32,-0.0034246612
-1,33,-0.16412888
-1,34,0.1493551
-1,35,0.1333814
-1,36,0.02555888
-1,37,-0.058053486
-1,38,-0.17211573
-2,17,0.29611352
-2,18,0.079470135
-2,19,0.02830436
-2,20,0.0124554485
-2,21,-0.044326082
-2,22,0.058255058
-2,23,-0.013876207
-2,24,0.1653288
-2,25,0.06798903
-2,26,0.0435293
-2,27,0.11541097
-2,28,-0.037836764
-2,29,-0.33484784
-2,30,0.067489855
-2,31,-0.07976775
-2,32,0.16832387
-2,33,-0.090749666
-2,34,0.07447835
-2,35,0.13837317
-2,36,-0.16712394
-3,17,0.07697424
-3,18,0.27215296
-3,19,0.21025485
-3,20,0.086957805
-3,21,-0.03658882
-3,22,0.26616284
-3,23,-0.041081425
-3,24,-0.026480459
-3,25,0.1079233
-3,26,0.02393655
-3,27,-0.12818804
-3,28,-0.052562527
-3,29,0.10542741
-3,30,0.16233373
-3,31,0.08346356
-3,32,0.23421541
-3,33,0.09644219
-3,34,0.0709841
-3,35,-0.13916996
-3,36,0.045526013
-4,14,0.1493551
-4,15,-0.08725542
-4,16,0.1353781
-4,17,0.00095874834
-4,18,0.23521377
-4,19,0.035792034
-4,20,-0.019242374
-4,21,-0.1501519
-4,22,0.053263277
-4,23,-0.27095303
-4,24,0.0021754955
-4,25,0.021815043
-4,26,0.16233373
-4,27,-0.16113381
-4,28,0.15035345
-4,29,0.07747342
-4,30,-0.04582362
-4,31,0.25018913
-4,32,-0.03409293
-4,33,-0.04282855
-5,14,0.24619569
-5,15,0.34203795
-5,16,0.122898646
-5,17,0.35601494
-5,18,0.33205438
-5,19,0.10592658
-5,20,0.11990357
-5,21,0.16732551
-5,22,-0.013065042
-5,23,-0.1521486
-5,24,-0.07327843
-5,25,0.090951234
-5,26,0.13737482
-5,27,-0.06054938
-5,28,-0.19208287
-5,29,-0.05156417
-5,30,-0.014312988
-5,31,0.063496426
-5,32,-0.061547734
-5,33,0.0939463
-6,18,0.3999426
-6,19,0.43987688
-6,20,0.20426472
-6,21,0.216245
-6,22,0.19328278
-6,23,0.039036695
-6,24,-0.007917265
-6,25,0.023187783
-6,26,0.085460275
-6,27,-0.0057957578
-6,28,-0.03359375
-6,29,-0.05430965
-6,30,-0.13517654
-6,31,-0.23401384
-6,32,-0.023111006
-6,33,0.035542447
-6,34,0.011644284
-6,35,-0.05430965
-6,36,-0.1221979
-6,37,0.20725977
-7,15,0.2182417
-7,16,0.06449478
-7,17,0.12339783
-7,18,0.21424827
-7,19,0.016074492
-7,20,0.15035345
-7,21,0.2322187
-7,22,-0.003292067
-7,23,-0.075275145
-7,24,0.15235017
-7,25,-0.14915353
-7,26,0.024560522
-7,27,-0.0133770285
-7,28,0.18629429
-7,29,-0.016122509
-7,30,-0.12369544
-7,31,-0.03658882
-7,32,-0.15314695
-7,33,-0.22103521
-7,34,-0.15115024
-8,20,0.02805477
-8,21,0.050018616
-8,22,-0.02435895
-8,23,0.062498074
-8,24,-0.0053745764
-8,25,0.10592658
-8,26,-0.013189836
-8,27,0.09594302
-8,28,-0.13517654
-8,29,-0.12369544
-8,30,0.12140111
-8,31,-0.040332656
-8,32,-0.10722256
-8,33,-0.008229252
-8,34,-0.013439426
-8,35,0.11890522
-8,36,-0.019491963
-8,37,-0.014749769
-8,38,-0.12369544
-8,39,0.17032059
-9,16,0.22722691
-9,17,0.36999193
-9,18,0.22223513
-9,19,0.36999193
-9,20,0.21524663
-9,21,0.18729265
-9,22,0.28812668
-9,23,0.0040006163
-9,24,-0.034841694
-9,25,-0.06504199
-9,26,0.06499396
-9,27,-0.009227608
-9,28,0.069486566
-9,29,-0.055557594
-9,30,0.11391344
-9,31,-0.10173159
-9,32,-0.15813874
-9,33,-0.07777103
-9,34,0.012954627
-9,35,0.011644284
+0,21,-0.22103521
+0,22,0.22123677
+0,23,0.31408396
+0,24,0.17531237
+0,25,0.18030415
+0,26,0.23421541
+0,27,-0.01312744
+0,28,0.17032059
+0,29,0.092947945
+0,30,-0.12020119
+0,31,0.013141819
+0,32,-0.08226364
+0,33,-0.096739806
+0,34,-0.10522584
+0,35,0.056258347
+0,36,0.14336495
+0,37,-0.044076495
+0,38,0.014576957
+0,39,-0.008042061
+0,40,-0.053061705
+1,19,0.17930579
+1,20,0.2741497
+1,21,0.30809382
+1,22,-0.013938604
+1,23,0.25418255
+1,24,0.053762455
+1,25,0.017072849
+1,26,-0.13018475
+1,27,-0.06404363
+1,28,0.014452162
+1,29,0.053762455
+1,30,0.038537517
+1,31,0.10343069
+1,32,-0.03608964
+1,33,-0.19008616
+1,34,0.084961094
+1,35,0.053512864
+1,36,-0.0055773673
+1,37,-0.0017945322
+1,38,-0.14016832
+2,17,0.43189004
+2,18,-0.035340875
+2,19,0.018819973
+2,20,0.114911795
+2,21,0.036041625
+2,22,-0.04806992
+2,23,0.029552305
+2,24,0.13837317
+2,25,-0.13717325
+2,26,0.12938796
+2,27,0.033046555
+2,28,0.026307646
+2,29,-0.083262
+2,30,0.28413326
+2,31,-0.034342516
+2,32,0.19128607
+2,33,-0.009321204
+2,34,-0.08575789
+2,35,0.17730908
+2,36,0.03778875
+3,17,0.059503004
+3,18,0.22423184
+3,19,0.5217421
+3,20,0.15235017
+3,21,0.15933867
+3,22,0.11441261
+3,23,0.0342945
+3,24,0.067489855
+3,25,0.2481924
+3,26,-0.11620776
+3,27,-0.10372831
+3,28,0.069486566
+3,29,0.042031765
+3,30,0.09594302
+3,31,0.01657367
+3,32,-0.1661256
+3,33,-0.021613471
+3,34,0.16033702
+3,35,0.06599232
+3,36,-0.02435895
+4,14,0.07847178
+4,15,0.13637646
+4,16,-0.03272019
+4,17,0.069486566
+4,18,0.062248483
+4,19,-0.098736525
+4,20,0.052264918
+4,21,0.16333209
+4,22,-0.1521486
+4,23,-0.15714039
+4,24,0.15534523
+4,25,-0.013751412
+4,26,0.04602519
+4,27,0.047522724
+4,28,-0.16512723
+4,29,-0.27694318
+4,30,-0.16113381
+4,31,-0.0625461
+4,32,0.04377889
+4,33,-0.08725542
+5,14,0.06798903
+5,15,0.42190647
+5,16,0.22622855
+5,17,0.18729265
+5,18,0.34203795
+5,19,0.41591632
+5,20,-0.03633923
+5,21,0.04677396
+5,22,-0.016871277
+5,23,0.033795323
+5,24,-0.19507794
+5,25,0.17830744
+5,26,-0.025107719
+5,27,0.07198246
+5,28,0.008586817
+5,29,0.07647506
+5,30,-0.1661256
+5,31,0.28213653
+5,32,-0.17411245
+5,33,-0.080766104
+6,18,0.45385388
+6,19,0.18529594
+6,20,0.36000836
+6,21,0.073479995
+6,22,0.22822526
+6,23,-0.06678911
+6,24,0.22722691
+6,25,0.06798903
+6,26,-0.029101145
+6,27,-0.019616757
+6,28,0.069486566
+6,29,0.045526013
+6,30,-0.21804014
+6,31,0.041532584
+6,32,0.06449478
+6,33,0.04951944
+6,34,-0.10522584
+6,35,-0.0037522472
+6,36,-0.16912067
+6,37,-0.06579076
+7,15,0.20526306
+7,16,0.2781431
+7,17,0.02693162
+7,18,0.042031765
+7,19,-0.023610184
+7,20,0.11591015
+7,21,-0.055557594
+7,22,0.07198246
+7,23,-0.18010259
+7,24,-0.1177053
+7,25,-0.08026692
+7,26,-0.0762735
+7,27,-0.07976775
+7,28,-0.16013545
+7,29,-0.01774484
+7,30,-0.08575789
+7,31,-0.0047038053
+7,32,-0.046821974
+7,33,-0.065291576
+7,34,0.2322187
+8,20,0.11291508
+8,21,0.025683673
+8,22,0.009335584
+8,23,0.08296438
+8,24,-0.038835123
+8,25,0.02693162
+8,26,0.08995288
+8,27,0.08795616
+8,28,-0.10722256
+8,29,-0.04282855
+8,30,-0.15614203
+8,31,-0.05156417
+8,32,-0.1431634
+8,33,-0.015685728
+8,34,-0.22802371
+8,35,-0.007917265
+8,36,-0.071281716
+8,37,-0.063294865
+8,38,-0.31687742
+8,39,0.062498074
+9,16,0.43987688
+9,17,0.43388674
+9,18,0.34203795
+9,19,0.22423184
+9,20,0.11740769
+9,21,0.15135181
+9,22,0.42789662
+9,23,-0.12269708
+9,24,-0.13916996
+9,25,0.13038632
+9,26,0.022439014
+9,27,0.1653288
+9,28,-0.047570743
+9,29,-0.082762815
+9,30,0.026557237
+9,31,0.10992001
+9,32,0.07597589
+9,33,-0.11371187
+9,34,-0.14615846
+9,35,0.09843891
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
index 9c9a70fe..39c6e464 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,127,6,2,0,1
-0,9,118,6,2,0,1
-0,10,118,6,2,2,1
-0,11,118,6,2,2,1
-0,12,118,6,7,2,1
-0,13,118,6,3,2,1
-0,14,118,6,3,6,1
-0,15,118,6,3,6,1
-0,16,127,6,3,other,0
-0,17,127,6,3,0,1
-0,18,127,6,3,0,1
-0,19,127,6,3,0,1
-0,20,118,6,3,0,1
-0,21,118,6,3,0,1
-0,22,118,6,3,2,1
-0,23,118,6,3,2,1
-0,24,118,6,3,2,1
-0,25,124,6,3,0,1
-0,26,124,6,3,0,1
-0,27,124,6,3,0,1
-0,28,104,6,3,0,1
-0,29,104,6,3,1,1
-0,30,109,6,3,8,1
-0,31,109,6,3,1,1
-0,32,109,4,3,8,0
-0,33,113,3,3,0,1
-0,34,122,3,5,8,0
-0,35,118,3,5,0,1
-0,36,118,6,5,8,0
-0,37,118,6,5,2,1
+0,8,102,6,3,8,1
+0,9,121,6,3,8,1
+0,10,104,6,3,8,1
+0,11,121,6,3,8,1
+0,12,121,6,3,8,1
+0,13,104,6,3,8,0
+0,14,104,3,3,0,0
+0,15,104,3,3,0,0
+0,16,104,3,3,0,0
+0,17,104,3,3,0,0
+0,18,104,3,3,0,0
+0,19,104,3,3,0,0
+0,20,104,3,3,0,1
+0,21,103,6,3,0,1
+0,22,113,6,3,8,1
+0,23,113,6,3,8,1
+0,24,113,6,3,8,1
+0,25,104,6,3,8,1
+0,26,104,6,3,8,1
+0,27,104,6,3,8,1
+0,28,104,6,3,8,1
+0,29,104,6,3,8,1
+0,30,104,9,3,0,0
+0,31,104,3,3,0,1
+0,32,104,9,3,0,1
+0,33,104,9,3,0,1
+0,34,104,3,3,8,1
+0,35,104,6,3,8,0
+0,36,121,6,3,8,0
+0,37,121,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
index 18c70481..b1f5c970 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,118,6,2,2,1
-1,9,118,6,9,0,1
-1,10,118,6,9,2,1
-1,11,118,6,9,2,1
-1,12,118,6,9,2,1
-1,13,118,6,3,2,1
-1,14,124,6,3,2,1
-1,15,104,6,3,8,1
-1,16,114,6,3,5,1
-1,17,114,6,3,5,1
-1,18,114,6,3,5,1
-1,19,114,6,3,5,1
-1,20,114,6,3,5,1
-1,21,109,6,3,5,1
-1,22,127,6,3,5,1
-1,23,127,6,9,5,1
-1,24,127,6,3,5,1
-1,25,118,6,3,5,1
-1,26,118,6,3,5,1
-1,27,118,6,3,5,1
-1,28,118,6,3,5,1
-1,29,118,6,3,9,1
-1,30,118,6,3,5,1
-1,31,127,6,3,5,1
-1,32,127,6,3,5,1
-1,33,127,6,3,5,1
-1,34,127,6,3,5,1
-1,35,127,6,3,5,1
-1,36,118,6,3,5,1
-1,37,118,6,3,0,1
+1,8,121,6,3,8,1
+1,9,121,6,3,8,0
+1,10,121,6,3,8,0
+1,11,121,6,3,8,0
+1,12,121,6,3,8,0
+1,13,104,3,3,8,0
+1,14,104,3,3,0,0
+1,15,104,3,3,0,0
+1,16,104,3,3,0,0
+1,17,104,3,3,0,0
+1,18,104,3,3,0,0
+1,19,104,3,3,0,0
+1,20,104,3,3,0,0
+1,21,121,6,3,0,1
+1,22,113,6,3,8,1
+1,23,121,6,3,8,1
+1,24,104,6,3,8,1
+1,25,104,6,3,8,0
+1,26,104,6,3,0,0
+1,27,104,9,3,0,1
+1,28,104,3,3,0,1
+1,29,104,9,3,0,1
+1,30,104,3,3,0,1
+1,31,104,3,3,0,1
+1,32,103,6,3,8,1
+1,33,113,6,3,8,1
+1,34,121,6,3,8,1
+1,35,121,6,3,8,1
+1,36,104,6,3,8,1
+1,37,104,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
index cec8e864..b32ec715 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,127,6,2,0,1
-2,9,127,6,3,0,1
-2,10,127,6,3,0,1
-2,11,118,6,3,0,1
-2,12,118,6,5,2,1
-2,13,118,6,7,2,1
-2,14,118,6,7,2,1
-2,15,124,6,3,2,1
-2,16,124,6,3,2,1
-2,17,124,6,3,2,1
-2,18,124,6,3,0,1
-2,19,104,6,3,0,1
-2,20,104,6,3,1,1
-2,21,109,6,3,0,1
-2,22,109,6,3,0,1
-2,23,104,3,3,8,1
-2,24,113,6,3,8,1
-2,25,113,6,3,8,1
-2,26,113,4,3,8,0
-2,27,113,3,4,8,0
-2,28,unknown,4,4,8,2
-2,29,unknown,4,4,8,0
-2,30,109,4,4,8,2
-2,31,109,4,3,8,2
-2,32,127,3,6,8,1
-2,33,unknown,3,5,4,1
-2,34,unknown,3,9,4,1
-2,35,unknown,0,9,8,1
-2,36,127,4,3,8,0
-2,37,118,4,9,3,1
+2,8,102,6,3,8,1
+2,9,121,6,3,8,1
+2,10,104,6,3,8,1
+2,11,104,6,3,8,0
+2,12,104,6,3,8,0
+2,13,104,6,3,8,0
+2,14,104,6,3,0,0
+2,15,104,3,3,0,0
+2,16,104,3,3,0,0
+2,17,104,3,3,0,0
+2,18,104,3,3,0,1
+2,19,103,4,3,0,1
+2,20,102,6,3,8,1
+2,21,121,6,3,8,1
+2,22,104,6,3,8,1
+2,23,104,6,3,8,1
+2,24,104,6,3,8,0
+2,25,104,6,3,8,0
+2,26,104,6,3,0,0
+2,27,104,9,3,0,1
+2,28,104,3,3,0,0
+2,29,104,3,3,0,0
+2,30,104,3,3,0,1
+2,31,103,4,3,0,1
+2,32,102,6,3,8,1
+2,33,121,6,3,8,1
+2,34,121,6,3,8,1
+2,35,104,6,3,8,1
+2,36,104,6,3,8,0
+2,37,104,6,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
index cd3c4c28..03dc6b54 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,127,2,2,2,1
-3,9,127,6,2,0,1
-3,10,127,6,3,2,1
-3,11,118,6,3,0,1
-3,12,118,6,3,2,1
-3,13,118,6,3,2,1
-3,14,118,6,3,2,1
-3,15,118,6,3,6,1
-3,16,109,6,3,8,1
-3,17,109,6,3,0,1
-3,18,109,6,3,0,1
-3,19,127,6,3,0,0
-3,20,109,6,3,0,1
-3,21,127,6,3,0,1
-3,22,118,6,3,0,1
-3,23,118,4,5,2,1
-3,24,118,6,5,2,1
-3,25,118,6,3,2,1
-3,26,118,6,3,2,1
-3,27,118,6,3,2,1
-3,28,118,6,3,0,1
-3,29,104,6,3,0,1
-3,30,104,6,3,1,1
-3,31,104,6,3,5,1
-3,32,104,6,3,1,1
-3,33,104,6,3,8,0
-3,34,109,6,3,5,1
-3,35,127,6,3,5,0
-3,36,118,6,9,8,0
-3,37,118,6,9,5,1
-4,8,127,6,2,0,1
-4,9,127,6,3,0,1
-4,10,127,6,3,0,1
-4,11,127,6,3,0,1
-4,12,118,6,3,0,1
-4,13,118,6,7,2,1
-4,14,118,1,7,2,1
-4,15,118,1,7,2,1
-4,16,118,6,5,2,1
-4,17,124,6,3,2,1
-4,18,124,6,3,2,1
-4,19,124,6,3,0,1
-4,20,109,6,3,0,1
-4,21,109,6,3,0,1
-4,22,109,6,3,0,1
-4,23,109,6,3,0,1
-4,24,109,6,3,8,1
-4,25,unknown,3,3,0,1
-4,26,124,3,5,8,1
-4,27,124,4,5,2,0
-4,28,124,6,7,2,1
-4,29,unknown,6,5,2,1
-4,30,124,6,3,2,1
-4,31,unknown,6,3,8,1
-4,32,unknown,6,3,8,1
-4,33,unknown,6,3,8,1
-4,34,104,6,3,8,1
-4,35,104,6,3,8,1
-4,36,104,6,3,8,1
-4,37,104,6,3,8,0
+3,8,121,6,3,8,1
+3,9,104,4,3,8,0
+3,10,121,3,3,8,0
+3,11,104,3,3,8,0
+3,12,121,3,3,8,0
+3,13,121,3,3,8,0
+3,14,104,3,3,8,0
+3,15,104,3,3,0,0
+3,16,104,3,3,0,0
+3,17,104,3,3,0,0
+3,18,104,3,3,0,0
+3,19,104,3,3,0,0
+3,20,104,3,3,0,0
+3,21,104,3,3,0,0
+3,22,121,6,3,0,1
+3,23,113,6,3,8,1
+3,24,121,6,3,8,1
+3,25,104,6,3,8,1
+3,26,104,6,3,8,0
+3,27,104,6,3,0,0
+3,28,104,9,3,0,1
+3,29,104,3,3,0,1
+3,30,104,9,3,0,1
+3,31,104,3,3,0,1
+3,32,104,3,3,0,1
+3,33,103,6,3,8,1
+3,34,113,6,3,8,1
+3,35,121,6,3,8,1
+3,36,121,6,3,8,1
+3,37,104,6,3,8,1
+4,8,102,6,9,8,1
+4,9,121,6,3,8,1
+4,10,121,6,3,8,1
+4,11,121,6,3,8,0
+4,12,121,6,3,8,0
+4,13,121,6,3,8,0
+4,14,104,3,3,8,0
+4,15,104,3,3,0,0
+4,16,104,3,3,0,0
+4,17,104,3,3,0,0
+4,18,104,3,3,0,0
+4,19,104,3,3,0,0
+4,20,104,3,3,0,0
+4,21,104,3,3,0,0
+4,22,121,6,3,0,1
+4,23,113,6,3,8,1
+4,24,121,6,3,8,1
+4,25,104,6,3,8,1
+4,26,104,6,3,8,0
+4,27,104,6,3,0,0
+4,28,104,9,3,0,1
+4,29,104,3,3,0,1
+4,30,104,9,3,0,1
+4,31,104,3,3,0,1
+4,32,104,3,3,0,1
+4,33,103,6,3,8,1
+4,34,113,6,3,8,1
+4,35,121,6,3,8,1
+4,36,121,6,3,8,1
+4,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
index 16a159f1..549102b2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,127,6,2,2,0
-5,9,118,6,2,0,1
-5,10,127,6,2,0,1
-5,11,127,6,3,0,1
-5,12,118,6,3,0,1
-5,13,118,6,5,2,1
-5,14,118,6,3,2,1
-5,15,124,6,3,8,1
-5,16,118,6,3,2,1
-5,17,127,6,3,8,other
-5,18,109,6,3,0,1
-5,19,109,6,3,0,other
-5,20,109,6,3,0,1
-5,21,112,6,3,0,1
-5,22,118,6,3,8,1
-5,23,124,6,3,8,1
-5,24,124,4,3,2,1
-5,25,124,6,5,2,1
-5,26,124,6,3,2,1
-5,27,unknown,6,3,2,1
-5,28,104,6,3,8,1
-5,29,104,6,3,8,1
-5,30,104,6,3,8,1
-5,31,104,6,3,8,1
-5,32,104,6,3,5,1
-5,33,104,6,3,8,1
-5,34,127,6,3,5,0
-5,35,109,1,9,8,0
-5,36,127,1,9,5,2
-5,37,127,9,9,5,1
+5,8,113,4,9,8,1
+5,9,113,6,3,8,0
+5,10,104,6,3,8,0
+5,11,104,6,3,0,0
+5,12,104,9,3,0,0
+5,13,121,3,3,8,1
+5,14,121,6,3,8,0
+5,15,104,3,3,8,0
+5,16,104,3,3,0,0
+5,17,104,3,3,0,0
+5,18,104,3,3,0,0
+5,19,121,3,3,0,0
+5,20,104,3,3,0,0
+5,21,104,3,3,0,0
+5,22,104,3,3,0,0
+5,23,121,4,3,0,1
+5,24,121,6,3,8,1
+5,25,104,6,3,8,1
+5,26,104,6,3,8,0
+5,27,104,3,3,0,0
+5,28,104,3,3,0,1
+5,29,104,4,3,0,1
+5,30,104,9,3,0,1
+5,31,104,3,3,0,1
+5,32,104,3,3,0,1
+5,33,104,6,3,0,1
+5,34,113,6,3,8,1
+5,35,113,6,3,8,1
+5,36,113,6,3,8,1
+5,37,113,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
index 07028e2d..68e5ec32 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,127,6,2,0,1
-6,9,127,6,3,0,1
-6,10,127,6,3,0,1
-6,11,127,4,3,0,1
-6,12,127,4,5,0,0
-6,13,112,1,7,2,1
-6,14,118,1,5,2,1
-6,15,118,1,7,2,1
-6,16,118,1,5,2,1
-6,17,118,6,3,2,1
-6,18,118,6,3,2,1
-6,19,118,6,3,2,1
-6,20,118,6,3,2,1
-6,21,109,6,3,0,1
-6,22,109,6,3,0,1
-6,23,109,6,3,0,1
-6,24,109,6,3,0,1
-6,25,109,6,3,0,1
-6,26,127,6,3,0,1
-6,27,118,6,3,0,1
-6,28,118,6,5,0,1
-6,29,118,4,5,other,0
-6,30,118,6,7,0,1
-6,31,118,4,5,2,0
-6,32,118,6,7,2,1
-6,33,118,6,3,2,1
-6,34,118,6,3,0,1
-6,35,127,6,3,0,1
-6,36,127,6,3,0,1
-6,37,127,6,3,2,1
-7,8,127,6,2,2,1
-7,9,127,6,3,0,1
-7,10,127,6,3,0,1
-7,11,118,6,3,0,1
-7,12,118,6,3,2,1
-7,13,118,6,3,2,1
-7,14,118,6,3,2,1
-7,15,118,6,3,2,1
-7,16,118,6,3,0,1
-7,17,124,6,3,8,1
-7,18,unknown,6,3,0,1
-7,19,109,6,3,1,1
-7,20,109,6,3,1,0
-7,21,109,6,3,0,1
-7,22,109,6,3,0,0
-7,23,109,6,5,0,1
-7,24,118,6,5,0,0
-7,25,118,3,7,2,1
-7,26,118,1,5,2,0
-7,27,118,6,7,2,1
-7,28,118,6,5,2,1
-7,29,118,6,3,2,1
-7,30,112,6,3,2,1
-7,31,112,6,3,0,1
-7,32,112,6,3,8,1
-7,33,109,6,3,8,1
-7,34,109,6,3,5,1
-7,35,109,6,3,1,1
-7,36,109,6,3,8,0
-7,37,109,6,3,0,1
+6,8,102,6,2,8,1
+6,9,113,6,3,8,1
+6,10,121,6,3,8,1
+6,11,121,6,3,8,1
+6,12,121,6,3,8,1
+6,13,121,6,3,8,0
+6,14,121,6,3,8,0
+6,15,104,3,3,8,0
+6,16,104,3,3,0,0
+6,17,104,3,3,0,0
+6,18,104,3,3,0,0
+6,19,104,3,3,0,0
+6,20,104,3,3,0,0
+6,21,104,3,3,0,0
+6,22,104,3,3,0,0
+6,23,121,6,3,0,1
+6,24,113,6,3,8,1
+6,25,121,6,3,8,1
+6,26,104,6,3,8,1
+6,27,104,6,3,8,0
+6,28,104,6,3,0,0
+6,29,104,9,3,0,1
+6,30,104,3,3,0,1
+6,31,104,9,3,0,1
+6,32,104,3,3,0,1
+6,33,104,3,3,0,1
+6,34,103,6,3,8,1
+6,35,113,6,3,8,1
+6,36,121,6,3,8,1
+6,37,121,6,3,8,1
+7,8,121,6,9,8,1
+7,9,104,6,3,8,1
+7,10,104,6,3,8,0
+7,11,104,6,3,0,0
+7,12,104,6,3,0,1
+7,13,121,6,3,8,1
+7,14,104,6,3,8,0
+7,15,104,3,3,0,0
+7,16,104,3,3,0,0
+7,17,104,3,3,0,1
+7,18,104,4,3,0,1
+7,19,103,6,3,8,1
+7,20,104,6,3,8,1
+7,21,104,9,3,0,0
+7,22,121,6,3,8,1
+7,23,104,6,3,8,1
+7,24,104,6,3,8,1
+7,25,104,6,3,8,1
+7,26,104,6,3,8,0
+7,27,104,3,3,0,0
+7,28,104,3,3,0,0
+7,29,104,3,3,0,0
+7,30,104,3,3,0,0
+7,31,104,3,3,0,1
+7,32,104,3,3,0,1
+7,33,103,4,3,0,1
+7,34,113,6,3,8,1
+7,35,113,6,3,8,1
+7,36,113,6,3,8,1
+7,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
index bc2b4d05..84a9c524 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,127,6,2,0,1
-8,9,127,6,3,0,1
-8,10,127,6,3,0,1
-8,11,118,6,9,2,1
-8,12,118,6,7,2,1
-8,13,118,6,9,6,1
-8,14,127,6,3,8,1
-8,15,127,6,3,5,1
-8,16,118,6,3,0,1
-8,17,124,6,3,0,1
-8,18,124,6,3,2,1
-8,19,126,6,3,5,1
-8,20,127,6,3,5,1
-8,21,118,6,3,0,1
-8,22,118,6,3,8,1
-8,23,118,6,3,0,1
-8,24,118,6,3,8,1
-8,25,118,6,3,0,1
-8,26,104,6,3,0,1
-8,27,126,6,3,5,0
-8,28,109,6,3,0,1
-8,29,127,6,3,0,0
-8,30,109,6,3,0,1
-8,31,127,6,3,0,0
-8,32,118,3,5,0,1
-8,33,118,6,5,2,0
-8,34,118,6,7,2,1
-8,35,118,6,5,2,1
-8,36,118,6,3,2,1
-8,37,112,6,3,2,1
+8,8,113,6,3,8,1
+8,9,121,6,3,8,1
+8,10,104,6,3,8,1
+8,11,104,6,3,8,1
+8,12,121,6,3,8,1
+8,13,121,6,3,8,1
+8,14,104,6,3,8,0
+8,15,104,9,3,0,0
+8,16,104,3,3,0,0
+8,17,104,3,3,0,0
+8,18,104,3,3,0,0
+8,19,104,3,3,0,0
+8,20,104,3,3,0,0
+8,21,104,3,3,0,1
+8,22,103,6,3,8,1
+8,23,121,6,3,8,1
+8,24,104,6,3,8,1
+8,25,104,6,3,8,1
+8,26,104,6,3,8,0
+8,27,104,6,3,0,0
+8,28,104,9,3,0,1
+8,29,104,9,3,0,1
+8,30,104,3,3,0,1
+8,31,104,3,3,0,1
+8,32,104,6,3,0,1
+8,33,113,6,3,8,1
+8,34,121,6,3,8,1
+8,35,121,6,3,8,1
+8,36,121,6,3,8,1
+8,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
index 940e8923..ad971442 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,127,6,2,other,1
-9,9,127,6,3,0,1
-9,10,127,6,3,0,1
-9,11,118,6,3,0,1
-9,12,118,6,3,other,1
-9,13,126,6,3,8,1
-9,14,113,6,3,8,0
-9,15,113,4,3,0,1
-9,16,118,6,3,0,1
-9,17,118,6,3,8,1
-9,18,118,6,3,0,1
-9,19,118,6,3,0,0
-9,20,118,6,3,0,1
-9,21,127,6,3,8,0
-9,22,118,6,3,0,1
-9,23,118,6,3,8,0
-9,24,109,6,3,0,1
-9,25,127,6,3,0,0
-9,26,112,6,5,0,1
-9,27,112,6,3,2,1
-9,28,118,6,3,8,1
-9,29,118,4,3,6,1
-9,30,124,6,3,8,1
-9,31,124,4,3,2,1
-9,32,124,6,3,2,1
-9,33,124,6,3,2,1
-9,34,104,6,3,1,1
-9,35,104,6,3,1,1
+9,8,113,6,3,0,0
+9,9,121,6,3,8,1
+9,10,104,6,3,8,1
+9,11,104,6,3,8,1
+9,12,104,6,3,8,1
+9,13,104,6,3,8,1
+9,14,104,6,3,8,0
+9,15,104,6,3,0,0
+9,16,104,9,3,0,2
+9,17,104,3,3,0,0
+9,18,104,3,3,0,1
+9,19,104,9,3,0,1
+9,20,103,6,3,8,1
+9,21,121,6,3,8,1
+9,22,104,6,3,8,1
+9,23,121,6,3,8,1
+9,24,121,6,3,8,1
+9,25,104,6,3,8,0
+9,26,104,9,3,0,0
+9,27,104,3,3,0,0
+9,28,104,3,3,0,0
+9,29,104,3,3,0,0
+9,30,104,3,3,0,0
+9,31,104,3,3,0,0
+9,32,104,3,3,0,1
+9,33,103,6,3,8,1
+9,34,121,6,3,8,1
+9,35,104,6,3,8,1
 9,36,104,6,3,8,1
-9,37,104,6,3,5,1
+9,37,104,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index c5dc282f..4d68f099 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,102,6,3,6,0
-0,9,102,6,3,6,0
-0,10,102,6,3,6,0
-0,11,102,6,3,6,0
-0,12,121,6,3,6,0
-0,13,102,6,3,6,0
-0,14,102,6,3,6,0
-0,15,102,6,3,6,0
-0,16,102,6,3,6,0
-0,17,102,6,3,6,0
-0,18,102,6,3,6,0
-0,19,102,6,3,6,0
-0,20,102,6,3,6,0
-0,21,102,6,3,6,0
-0,22,102,6,3,6,0
-0,23,102,6,3,6,0
-0,24,102,6,3,6,0
-0,25,102,6,3,6,0
-0,26,102,6,3,6,0
-0,27,102,6,3,6,0
-0,28,102,6,3,6,0
-0,29,102,6,3,6,0
-0,30,102,6,3,6,0
-0,31,102,6,3,6,0
-0,32,102,6,3,6,0
-0,33,102,6,3,6,0
-0,34,102,6,3,6,0
-0,35,102,6,3,6,0
-0,36,102,6,3,6,0
-0,37,102,6,3,6,0
+0,8,121,1,1,2,1
+0,9,113,0,7,8,2
+0,10,121,1,1,6,0
+0,11,113,9,0,8,1
+0,12,121,1,0,2,0
+0,13,127,4,0,7,0
+0,14,127,4,0,7,1
+0,15,127,1,0,2,1
+0,16,127,4,7,7,1
+0,17,102,1,1,7,1
+0,18,105,8,7,7,1
+0,19,105,1,7,2,1
+0,20,113,mask,7,7,1
+0,21,113,1,7,2,1
+0,22,113,1,7,2,1
+0,23,113,1,7,7,1
+0,24,113,1,5,2,1
+0,25,113,1,7,2,1
+0,26,113,1,1,7,1
+0,27,113,1,0,7,1
+0,28,113,1,0,2,1
+0,29,113,1,0,7,1
+0,30,113,1,0,7,1
+0,31,113,1,7,7,1
+0,32,113,1,7,7,1
+0,33,113,1,7,7,1
+0,34,113,1,7,7,1
+0,35,113,6,7,7,1
+0,36,113,1,7,2,1
+0,37,113,6,7,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index e842b43e..1241eea1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,6,3,6,0
-1,9,121,6,3,6,0
-1,10,121,6,3,6,0
-1,11,121,6,3,6,0
-1,12,121,4,3,6,0
-1,13,121,6,3,8,0
-1,14,121,6,9,6,0
-1,15,121,6,9,6,0
-1,16,102,6,9,6,0
-1,17,102,6,3,6,0
-1,18,121,6,3,6,0
-1,19,121,6,9,6,0
-1,20,102,6,3,6,0
-1,21,121,6,3,6,0
-1,22,121,6,3,6,0
-1,23,121,6,3,6,0
-1,24,121,6,3,6,0
-1,25,121,6,3,6,0
-1,26,121,6,9,6,0
-1,27,102,6,3,6,0
-1,28,102,6,3,6,0
-1,29,102,6,9,6,0
-1,30,102,6,3,6,0
-1,31,102,6,3,6,0
-1,32,102,6,3,6,0
-1,33,102,6,3,6,0
-1,34,102,6,3,6,0
-1,35,102,6,3,6,0
-1,36,102,6,3,6,0
-1,37,102,6,3,6,0
+1,8,113,0,7,6,0
+1,9,121,8,9,2,1
+1,10,113,9,0,8,0
+1,11,121,3,0,2,1
+1,12,127,9,0,7,0
+1,13,121,6,0,7,1
+1,14,127,1,0,7,1
+1,15,127,1,0,7,1
+1,16,108,1,7,7,1
+1,17,105,8,7,7,1
+1,18,105,1,7,2,1
+1,19,113,2,7,2,1
+1,20,113,2,7,2,1
+1,21,113,1,7,2,1
+1,22,113,1,7,7,1
+1,23,113,1,7,7,1
+1,24,113,1,5,2,1
+1,25,113,1,0,8,1
+1,26,113,1,0,2,1
+1,27,113,1,0,2,1
+1,28,113,1,0,7,1
+1,29,108,1,0,7,1
+1,30,108,1,7,7,1
+1,31,113,1,7,7,1
+1,32,113,1,7,7,1
+1,33,113,6,7,7,1
+1,34,113,1,7,2,1
+1,35,113,6,7,7,1
+1,36,113,1,7,2,1
+1,37,113,6,7,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index 5ee56466..46b4436b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,102,6,3,6,0
-2,9,121,6,3,6,0
-2,10,102,6,3,6,0
-2,11,102,6,3,6,0
-2,12,102,6,3,6,0
-2,13,121,6,3,6,0
-2,14,102,6,3,6,0
-2,15,102,6,3,6,0
-2,16,102,6,3,6,0
-2,17,102,6,3,6,0
-2,18,102,6,3,6,0
-2,19,102,6,3,6,0
-2,20,102,6,3,6,0
-2,21,102,6,3,6,0
-2,22,102,6,3,6,0
-2,23,102,6,3,6,0
-2,24,102,6,3,6,0
-2,25,102,6,3,6,0
-2,26,102,6,3,6,0
-2,27,102,6,3,6,0
-2,28,102,6,3,6,0
-2,29,102,6,3,6,0
-2,30,102,6,3,6,0
-2,31,102,6,3,6,0
-2,32,102,6,3,6,0
-2,33,102,6,3,6,0
-2,34,102,6,3,6,0
-2,35,102,6,3,6,0
-2,36,102,6,3,6,0
-2,37,102,6,3,6,0
+2,8,121,1,7,6,0
+2,9,113,0,7,7,2
+2,10,121,8,9,6,1
+2,11,102,1,1,6,2
+2,12,105,1,7,8,0
+2,13,127,1,1,0,1
+2,14,127,1,7,0,1
+2,15,127,1,1,0,1
+2,16,127,1,7,7,1
+2,17,127,1,7,0,1
+2,18,113,1,7,0,1
+2,19,113,1,7,0,1
+2,20,113,1,7,7,1
+2,21,113,1,7,0,1
+2,22,113,1,7,7,1
+2,23,113,1,7,0,1
+2,24,113,1,7,2,1
+2,25,113,1,7,2,1
+2,26,113,1,1,7,1
+2,27,113,1,5,2,1
+2,28,113,1,2,8,1
+2,29,113,1,1,7,1
+2,30,113,1,0,7,1
+2,31,113,1,0,7,1
+2,32,113,1,0,7,1
+2,33,113,1,7,7,1
+2,34,113,1,7,7,1
+2,35,113,1,7,7,1
+2,36,113,1,7,7,1
+2,37,113,1,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index d549003f..f0150155 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,6,3,6,0
-3,9,121,6,3,6,0
-3,10,121,6,3,6,0
-3,11,102,6,3,6,0
-3,12,102,6,3,6,0
-3,13,121,6,3,6,0
-3,14,102,6,9,6,0
-3,15,102,6,3,6,0
-3,16,102,6,3,6,0
-3,17,102,6,3,6,0
-3,18,102,6,3,6,0
-3,19,102,6,3,6,0
-3,20,102,6,3,6,0
-3,21,102,6,3,6,0
-3,22,102,6,3,6,0
-3,23,102,6,3,6,0
-3,24,102,6,3,6,0
-3,25,102,6,3,6,0
-3,26,102,6,3,6,0
-3,27,102,6,3,6,0
-3,28,102,6,3,6,0
-3,29,102,6,3,6,0
-3,30,102,6,3,6,0
-3,31,102,6,3,6,0
-3,32,102,6,3,6,0
-3,33,102,6,3,6,0
-3,34,102,6,3,6,0
-3,35,102,6,3,6,0
-3,36,102,6,3,6,0
-3,37,102,6,3,6,0
-4,8,102,6,3,6,0
-4,9,121,6,3,6,0
-4,10,121,6,3,6,0
-4,11,121,6,3,6,0
-4,12,121,6,3,6,0
-4,13,121,6,3,6,0
-4,14,102,6,3,6,0
-4,15,102,6,3,6,0
-4,16,102,6,3,6,0
-4,17,102,6,3,6,0
-4,18,102,6,3,6,0
-4,19,102,6,3,6,0
-4,20,102,6,3,6,0
-4,21,102,6,3,6,0
-4,22,102,6,3,6,0
-4,23,102,6,3,6,0
-4,24,102,6,3,6,0
-4,25,102,6,3,6,0
-4,26,102,6,3,6,0
-4,27,102,6,3,6,0
-4,28,102,6,3,6,0
-4,29,102,6,3,6,0
-4,30,102,6,3,6,0
-4,31,102,6,3,6,0
-4,32,102,6,3,6,0
-4,33,102,6,3,6,0
-4,34,102,6,3,6,0
-4,35,102,6,3,6,0
-4,36,102,6,3,6,0
-4,37,102,6,3,6,0
+3,8,109,3,0,6,0
+3,9,121,3,9,8,0
+3,10,121,3,1,8,0
+3,11,121,0,1,8,0
+3,12,127,0,0,6,0
+3,13,127,4,8,7,0
+3,14,127,4,1,7,1
+3,15,127,1,1,7,1
+3,16,127,0,7,0,1
+3,17,127,5,1,0,1
+3,18,127,1,7,0,1
+3,19,113,1,7,0,1
+3,20,113,1,7,7,1
+3,21,113,1,7,0,1
+3,22,113,1,7,0,1
+3,23,113,1,7,7,1
+3,24,113,1,7,0,1
+3,25,113,1,7,2,1
+3,26,113,1,7,2,1
+3,27,113,1,7,7,1
+3,28,113,1,5,7,1
+3,29,113,1,7,2,1
+3,30,113,1,1,7,1
+3,31,113,1,0,7,1
+3,32,113,1,7,7,1
+3,33,113,1,7,7,1
+3,34,113,1,5,7,1
+3,35,113,1,7,2,1
+3,36,113,1,7,7,1
+3,37,113,1,5,7,1
+4,8,113,3,7,4,1
+4,9,113,1,7,2,1
+4,10,113,3,7,2,1
+4,11,113,1,1,2,1
+4,12,127,1,0,2,1
+4,13,127,1,0,2,1
+4,14,127,4,0,7,1
+4,15,102,1,0,2,1
+4,16,113,1,0,7,1
+4,17,108,1,7,7,1
+4,18,108,1,7,7,1
+4,19,113,1,7,7,1
+4,20,113,8,7,7,1
+4,21,113,1,7,2,1
+4,22,113,6,7,2,1
+4,23,113,1,7,2,1
+4,24,113,6,7,2,1
+4,25,113,1,1,2,1
+4,26,113,1,0,8,1
+4,27,113,1,0,2,1
+4,28,113,1,0,2,1
+4,29,113,1,0,7,1
+4,30,108,1,0,7,1
+4,31,108,1,0,7,1
+4,32,108,1,0,7,1
+4,33,108,1,7,7,1
+4,34,113,6,7,7,1
+4,35,113,1,7,7,1
+4,36,113,6,7,7,1
+4,37,113,6,7,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index ee14eff8..f2cf6b66 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,102,6,3,6,0
-5,9,102,6,3,6,0
-5,10,102,6,3,6,0
-5,11,102,6,3,6,0
-5,12,102,6,3,6,0
-5,13,102,6,3,8,0
-5,14,102,6,3,6,0
-5,15,102,6,3,6,0
-5,16,102,6,3,6,0
-5,17,102,6,3,6,0
-5,18,102,6,3,6,0
-5,19,102,6,3,6,0
-5,20,102,6,3,6,0
-5,21,102,6,3,6,0
-5,22,102,6,3,6,0
-5,23,102,6,3,6,0
-5,24,102,6,3,6,0
-5,25,102,6,3,6,0
-5,26,102,6,3,6,0
-5,27,102,6,3,6,0
-5,28,102,6,3,6,0
-5,29,102,6,3,6,0
-5,30,102,6,3,6,0
-5,31,102,6,3,6,0
-5,32,102,6,3,6,0
-5,33,102,6,3,6,0
-5,34,102,6,3,6,0
-5,35,102,6,3,6,0
-5,36,102,6,3,6,0
-5,37,102,6,3,6,0
+5,8,113,3,7,4,1
+5,9,113,1,9,6,1
+5,10,113,1,7,0,1
+5,11,113,1,7,2,1
+5,12,113,1,1,2,1
+5,13,113,1,0,2,1
+5,14,113,1,0,2,1
+5,15,113,1,0,2,1
+5,16,113,1,8,7,1
+5,17,113,1,8,7,1
+5,18,113,1,7,7,1
+5,19,113,1,7,7,1
+5,20,113,1,7,7,1
+5,21,113,1,7,7,1
+5,22,113,1,7,7,1
+5,23,113,1,7,2,1
+5,24,113,1,7,2,1
+5,25,113,1,7,2,1
+5,26,113,1,1,7,1
+5,27,113,1,5,2,1
+5,28,113,1,2,8,1
+5,29,113,1,1,7,1
+5,30,113,1,0,7,1
+5,31,113,1,0,7,1
+5,32,113,1,0,7,1
+5,33,113,1,7,7,1
+5,34,113,1,7,7,1
+5,35,113,1,7,7,1
+5,36,113,1,7,7,1
+5,37,113,1,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index 89e13f1d..650c1470 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,121,6,3,6,0
-6,9,121,6,3,6,0
-6,10,121,6,3,6,0
-6,11,121,6,3,6,0
-6,12,121,6,3,6,0
-6,13,121,6,9,6,0
-6,14,121,6,9,6,0
-6,15,102,6,9,6,0
-6,16,102,6,3,6,0
-6,17,121,6,3,6,0
-6,18,121,6,9,6,0
-6,19,102,6,3,6,0
-6,20,121,6,3,6,0
-6,21,121,6,3,6,0
-6,22,121,6,3,6,0
-6,23,121,6,3,6,0
-6,24,121,6,3,6,0
-6,25,121,6,9,6,0
-6,26,102,6,3,6,0
-6,27,102,6,3,6,0
-6,28,102,6,9,6,0
-6,29,102,6,3,6,0
-6,30,102,6,3,6,0
-6,31,102,6,3,6,0
-6,32,102,6,3,6,0
-6,33,102,6,3,6,0
-6,34,102,6,3,6,0
-6,35,102,6,3,6,0
-6,36,102,6,3,6,0
-6,37,102,6,3,6,0
-7,8,121,6,3,6,0
-7,9,121,6,3,6,0
-7,10,121,6,9,6,0
-7,11,102,6,9,6,0
-7,12,102,6,3,6,0
-7,13,102,6,9,6,0
-7,14,102,6,3,6,0
-7,15,102,6,3,6,0
-7,16,121,6,3,6,0
-7,17,121,6,3,6,0
-7,18,121,6,3,6,0
-7,19,121,6,3,6,0
-7,20,121,6,3,6,0
-7,21,121,6,3,6,0
-7,22,102,6,9,6,0
-7,23,102,6,3,6,0
-7,24,102,6,9,6,0
-7,25,102,6,3,6,0
-7,26,102,6,3,6,0
-7,27,102,6,3,6,0
-7,28,102,6,3,6,0
-7,29,102,6,3,6,0
-7,30,121,6,3,6,0
-7,31,102,6,3,6,0
-7,32,102,6,3,6,0
-7,33,102,6,3,6,0
-7,34,102,6,3,6,0
-7,35,102,6,3,6,0
-7,36,102,6,3,6,0
-7,37,102,6,3,6,0
+6,8,122,4,7,4,1
+6,9,113,1,7,2,0
+6,10,113,3,7,4,1
+6,11,113,1,7,2,1
+6,12,113,6,7,7,1
+6,13,102,1,1,2,1
+6,14,127,4,7,8,1
+6,15,102,1,1,0,1
+6,16,113,1,7,7,1
+6,17,102,1,1,7,1
+6,18,105,1,7,7,1
+6,19,113,1,7,0,1
+6,20,113,1,7,7,1
+6,21,113,1,5,7,1
+6,22,113,1,5,2,1
+6,23,113,1,7,2,1
+6,24,113,1,7,7,1
+6,25,113,1,5,7,1
+6,26,113,1,7,8,1
+6,27,113,1,1,7,1
+6,28,113,1,5,7,1
+6,29,113,1,7,7,1
+6,30,113,1,5,7,1
+6,31,113,1,5,7,1
+6,32,113,1,2,7,1
+6,33,113,1,5,7,1
+6,34,113,1,2,7,1
+6,35,113,1,5,7,1
+6,36,113,1,2,8,1
+6,37,113,1,1,7,1
+7,8,121,1,9,6,0
+7,9,105,0,9,1,2
+7,10,121,0,9,6,0
+7,11,121,0,1,6,0
+7,12,121,0,3,8,0
+7,13,127,0,0,6,0
+7,14,127,4,0,7,0
+7,15,127,4,0,2,1
+7,16,127,4,0,2,0
+7,17,127,4,0,8,0
+7,18,127,4,1,2,0
+7,19,127,4,1,8,1
+7,20,127,4,1,8,1
+7,21,127,1,1,8,1
+7,22,113,1,7,0,1
+7,23,113,1,7,7,1
+7,24,113,1,9,0,1
+7,25,113,1,7,7,1
+7,26,113,1,9,0,1
+7,27,113,1,7,7,1
+7,28,113,1,5,0,1
+7,29,113,1,2,2,1
+7,30,113,1,7,7,1
+7,31,113,1,5,7,1
+7,32,113,1,2,2,1
+7,33,113,1,5,7,1
+7,34,113,1,2,7,1
+7,35,113,1,5,7,1
+7,36,108,1,2,8,1
+7,37,113,1,0,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index 0988fb6a..2cdb2be9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,102,6,3,6,0
-8,9,102,6,3,6,0
-8,10,102,6,3,6,0
-8,11,102,6,3,6,0
-8,12,102,6,3,6,0
-8,13,102,6,3,6,0
-8,14,102,6,3,6,0
-8,15,102,6,3,6,0
-8,16,102,6,3,6,0
-8,17,102,6,3,6,0
-8,18,102,6,3,6,0
-8,19,102,6,3,6,0
-8,20,102,6,3,6,0
-8,21,102,6,3,6,0
-8,22,102,6,3,6,0
-8,23,102,6,3,6,0
-8,24,102,6,3,6,0
-8,25,102,6,3,6,0
-8,26,102,6,3,6,0
-8,27,102,6,3,6,0
-8,28,102,6,3,6,0
-8,29,102,6,3,6,0
-8,30,102,6,3,6,0
-8,31,102,6,3,6,0
-8,32,102,6,3,6,0
-8,33,102,6,3,6,0
-8,34,102,6,3,6,0
-8,35,102,6,3,6,0
-8,36,102,6,3,6,0
-8,37,102,6,3,6,0
+8,8,113,1,3,2,0
+8,9,113,3,7,7,1
+8,10,121,1,7,2,1
+8,11,113,4,7,7,2
+8,12,121,1,1,2,1
+8,13,127,1,0,8,0
+8,14,109,1,0,8,1
+8,15,127,1,0,7,0
+8,16,127,1,0,7,1
+8,17,127,1,7,7,1
+8,18,127,1,7,7,1
+8,19,113,1,7,0,1
+8,20,113,1,7,7,1
+8,21,113,8,7,7,1
+8,22,113,1,5,2,1
+8,23,113,1,7,8,1
+8,24,113,1,7,0,1
+8,25,113,1,7,7,1
+8,26,113,1,5,7,1
+8,27,113,1,2,2,1
+8,28,113,1,1,7,1
+8,29,108,1,0,2,1
+8,30,113,1,0,7,1
+8,31,113,1,0,7,1
+8,32,113,1,7,7,1
+8,33,113,1,7,7,1
+8,34,113,1,7,7,1
+8,35,113,1,7,7,1
+8,36,113,6,7,7,1
+8,37,113,6,7,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index f032d896..1c08eb2a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,113,6,3,6,0
-9,9,102,6,3,6,0
-9,10,102,6,3,6,0
-9,11,102,6,3,6,0
-9,12,102,6,3,6,0
-9,13,121,6,3,8,0
-9,14,102,6,3,6,0
-9,15,102,6,3,6,0
-9,16,102,6,3,6,0
-9,17,102,6,3,6,0
-9,18,102,6,3,6,0
-9,19,102,6,3,6,0
-9,20,102,6,3,6,0
-9,21,102,6,3,6,0
-9,22,102,6,3,6,0
-9,23,102,6,3,6,0
-9,24,102,6,3,6,0
-9,25,102,6,3,6,0
-9,26,102,6,3,6,0
-9,27,102,6,3,6,0
-9,28,102,6,3,6,0
-9,29,102,6,3,6,0
-9,30,102,6,3,6,0
-9,31,102,6,3,6,0
-9,32,102,6,3,6,0
-9,33,102,6,3,6,0
-9,34,102,6,3,6,0
-9,35,102,6,3,6,0
-9,36,102,6,3,6,0
-9,37,102,6,3,6,0
+9,8,121,1,7,2,0
+9,9,113,3,7,1,2
+9,10,113,1,9,6,1
+9,11,113,1,7,7,1
+9,12,113,1,7,7,1
+9,13,113,1,7,7,1
+9,14,113,1,7,7,1
+9,15,113,1,7,2,1
+9,16,113,1,1,2,1
+9,17,113,1,0,2,1
+9,18,113,1,0,2,1
+9,19,113,1,0,2,1
+9,20,113,1,0,7,1
+9,21,113,1,0,7,1
+9,22,108,1,0,7,1
+9,23,108,1,7,7,1
+9,24,113,1,7,7,1
+9,25,113,1,7,7,1
+9,26,113,6,7,7,1
+9,27,113,1,7,2,1
+9,28,113,6,7,7,1
+9,29,113,1,7,2,1
+9,30,113,6,7,2,1
+9,31,113,1,1,2,1
+9,32,113,1,0,2,1
+9,33,113,1,0,2,1
+9,34,113,1,0,2,1
+9,35,113,1,0,7,1
+9,36,108,1,0,2,1
+9,37,113,1,0,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index 261bca34..105792b4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,115,7,0,3,0
-0,9,107,1,2,5,1
-0,10,103,6,2,3,1
-0,11,126,6,2,0,0
-0,12,112,4,9,5,0
-0,13,112,6,3,0,0
-0,14,112,6,2,0,0
-0,15,112,6,2,0,0
-0,16,100,6,2,0,0
-0,17,110,6,2,0,2
-0,18,115,6,1,0,2
-0,19,107,6,2,0,2
-0,20,100,6,2,6,2
-0,21,112,6,2,6,2
-0,22,107,6,2,0,2
-0,23,107,6,2,0,2
-0,24,100,6,2,0,2
-0,25,100,9,2,0,2
-0,26,115,6,2,0,1
-0,27,107,9,2,0,2
-0,28,115,5,2,0,1
-0,29,107,6,2,0,1
-0,30,110,6,2,0,2
-0,31,113,9,2,0,1
-0,32,115,6,2,0,1
-0,33,103,0,2,0,1
-0,34,113,0,2,0,0
-0,35,113,4,2,0,1
-0,36,110,6,2,0,1
-0,37,110,5,2,0,1
+0,8,113,6,2,0,1
+0,9,102,6,2,0,1
+0,10,102,6,2,0,1
+0,11,121,6,2,0,1
+0,12,121,6,5,6,1
+0,13,121,6,2,6,1
+0,14,121,6,2,6,1
+0,15,121,6,2,6,1
+0,16,121,6,2,6,1
+0,17,121,6,5,6,1
+0,18,121,3,5,6,1
+0,19,121,3,5,6,1
+0,20,121,3,5,6,1
+0,21,121,3,5,6,1
+0,22,121,3,5,6,1
+0,23,121,3,5,6,1
+0,24,121,3,5,6,1
+0,25,121,3,5,6,0
+0,26,121,3,5,6,0
+0,27,121,3,5,6,0
+0,28,121,3,5,6,0
+0,29,121,3,5,6,0
+0,30,121,3,5,6,0
+0,31,121,3,5,6,0
+0,32,121,3,5,6,0
+0,33,121,3,5,6,0
+0,34,121,3,5,6,0
+0,35,121,3,5,6,0
+0,36,121,3,5,6,0
+0,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index cdaed10b..350e2f2f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,107,7,2,7,0
-1,9,107,7,2,0,0
-1,10,110,9,2,0,0
-1,11,121,6,2,0,1
-1,12,121,4,9,8,0
-1,13,112,4,9,5,0
-1,14,112,6,3,5,0
-1,15,112,4,3,5,0
-1,16,112,6,2,4,0
-1,17,112,4,9,0,0
-1,18,112,6,9,0,0
-1,19,112,6,9,5,0
-1,20,112,6,9,5,0
-1,21,112,6,9,0,0
-1,22,100,6,9,6,0
-1,23,100,6,9,5,0
-1,24,100,6,9,5,0
-1,25,100,6,9,5,0
-1,26,100,6,9,5,0
-1,27,100,1,9,5,0
-1,28,100,6,9,5,0
-1,29,100,4,9,5,0
-1,30,100,1,9,5,0
-1,31,100,6,9,5,0
-1,32,100,4,9,5,0
-1,33,112,4,9,5,0
-1,34,112,4,9,5,0
-1,35,112,4,9,5,0
-1,36,112,4,9,5,0
-1,37,112,4,9,5,0
+1,8,113,6,2,0,0
+1,9,121,6,5,0,1
+1,10,121,4,9,6,0
+1,11,121,4,5,6,1
+1,12,121,4,5,6,1
+1,13,121,3,2,6,1
+1,14,121,3,2,6,1
+1,15,121,4,5,6,1
+1,16,121,3,5,6,1
+1,17,121,3,5,6,1
+1,18,121,3,2,6,1
+1,19,121,3,2,6,1
+1,20,121,3,2,6,1
+1,21,121,3,5,6,1
+1,22,121,3,5,6,1
+1,23,121,3,5,6,1
+1,24,121,3,5,6,1
+1,25,121,3,5,6,1
+1,26,121,3,5,6,1
+1,27,121,3,5,6,1
+1,28,121,3,5,6,1
+1,29,121,3,5,6,0
+1,30,121,3,5,6,0
+1,31,121,3,5,6,0
+1,32,121,3,5,6,0
+1,33,121,3,5,6,0
+1,34,121,3,5,6,0
+1,35,121,3,5,6,0
+1,36,121,3,5,6,0
+1,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index f74eed86..7cb7ee53 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,107,1,0,9,1
-2,9,108,0,2,5,1
-2,10,110,0,2,8,1
-2,11,110,0,2,0,1
-2,12,110,0,2,0,1
-2,13,103,6,8,0,0
-2,14,112,4,8,8,0
-2,15,112,6,2,8,0
-2,16,112,4,2,0,0
-2,17,112,6,2,0,0
-2,18,110,4,2,0,0
-2,19,102,6,1,0,0
-2,20,110,6,1,0,0
-2,21,122,6,9,0,0
-2,22,112,6,9,6,0
-2,23,110,6,2,0,0
-2,24,110,6,2,0,0
-2,25,110,4,2,0,0
-2,26,110,4,2,0,0
-2,27,102,4,0,0,1
-2,28,108,6,0,5,1
-2,29,108,6,0,5,0
-2,30,108,6,9,4,0
-2,31,112,6,9,5,0
-2,32,108,6,9,0,0
-2,33,112,6,9,5,0
-2,34,112,6,9,0,0
-2,35,100,6,9,5,0
-2,36,100,6,9,5,0
-2,37,100,6,9,5,0
+2,8,105,5,3,0,0
+2,9,102,6,1,0,1
+2,10,121,6,2,0,1
+2,11,121,6,2,6,1
+2,12,121,6,2,6,1
+2,13,121,6,2,6,1
+2,14,121,6,2,6,1
+2,15,121,6,2,6,1
+2,16,121,6,5,6,1
+2,17,121,6,5,6,1
+2,18,121,3,5,6,1
+2,19,121,3,5,6,1
+2,20,121,3,5,6,1
+2,21,121,3,5,6,1
+2,22,121,3,5,6,1
+2,23,121,3,5,6,1
+2,24,121,3,5,6,1
+2,25,121,3,5,6,0
+2,26,121,3,5,6,0
+2,27,121,3,5,6,0
+2,28,121,3,5,6,0
+2,29,121,3,5,6,0
+2,30,121,3,5,6,0
+2,31,121,3,5,6,0
+2,32,121,3,5,6,0
+2,33,121,3,5,6,0
+2,34,121,3,5,6,0
+2,35,121,3,5,6,0
+2,36,121,3,5,6,0
+2,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index 8ce16f14..e82f6f8a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,110,1,2,7,1
-3,9,107,1,2,3,1
-3,10,110,0,2,3,1
-3,11,110,1,7,0,1
-3,12,108,0,7,5,1
-3,13,112,6,7,8,1
-3,14,112,6,2,0,0
-3,15,113,6,2,0,0
-3,16,110,6,2,0,0
-3,17,110,6,2,0,1
-3,18,102,6,2,0,1
-3,19,110,6,2,0,1
-3,20,102,6,2,0,1
-3,21,110,6,2,0,1
-3,22,110,6,2,0,1
-3,23,110,6,2,0,1
-3,24,110,6,2,0,1
-3,25,102,6,2,0,1
-3,26,110,6,2,0,1
-3,27,102,6,2,0,1
-3,28,110,6,2,0,1
-3,29,102,6,2,0,1
-3,30,110,6,2,0,1
-3,31,102,6,2,0,1
-3,32,110,6,2,0,1
-3,33,110,6,2,0,1
-3,34,102,6,2,0,1
-3,35,110,6,2,0,1
-3,36,102,6,2,0,1
-3,37,110,6,2,0,1
-4,8,107,7,0,3,0
-4,9,110,1,2,9,0
-4,10,119,7,0,4,0
-4,11,107,7,0,5,0
-4,12,103,6,9,5,0
-4,13,112,6,9,5,0
-4,14,112,6,9,0,0
-4,15,112,4,9,0,0
-4,16,112,6,9,5,0
-4,17,112,6,9,0,0
-4,18,100,4,9,5,0
-4,19,112,6,9,5,0
-4,20,100,6,9,0,0
-4,21,100,4,9,5,0
-4,22,112,6,9,5,0
-4,23,100,6,9,5,0
-4,24,100,4,9,5,0
-4,25,112,6,9,5,0
-4,26,100,4,9,5,0
-4,27,112,6,9,5,0
-4,28,100,4,9,5,0
-4,29,112,9,9,5,0
-4,30,112,6,9,5,0
-4,31,100,4,9,5,0
-4,32,112,9,9,5,0
-4,33,112,6,9,5,0
-4,34,112,9,9,5,0
-4,35,112,6,9,5,0
-4,36,112,9,9,5,0
-4,37,112,6,9,5,0
+3,8,113,6,2,0,0
+3,9,102,6,2,0,1
+3,10,121,6,2,0,1
+3,11,121,6,5,6,1
+3,12,121,6,5,6,1
+3,13,121,4,5,6,1
+3,14,121,3,5,6,1
+3,15,121,4,5,6,1
+3,16,121,3,5,6,1
+3,17,121,3,2,6,1
+3,18,121,3,2,6,1
+3,19,121,3,2,6,1
+3,20,121,3,5,6,1
+3,21,121,3,5,6,1
+3,22,121,3,5,6,1
+3,23,121,3,5,6,1
+3,24,121,3,5,6,1
+3,25,121,3,5,6,1
+3,26,121,3,5,6,1
+3,27,121,3,5,6,1
+3,28,121,3,5,6,0
+3,29,121,3,5,6,0
+3,30,121,3,5,6,0
+3,31,121,3,5,6,0
+3,32,121,3,5,6,0
+3,33,121,3,5,6,0
+3,34,121,3,5,6,0
+3,35,121,3,5,6,0
+3,36,121,3,5,6,0
+3,37,121,3,5,6,0
+4,8,105,5,8,2,0
+4,9,102,6,3,0,1
+4,10,121,6,3,0,1
+4,11,121,4,9,6,1
+4,12,121,6,5,6,1
+4,13,121,4,2,6,1
+4,14,121,4,2,6,1
+4,15,121,3,2,6,1
+4,16,121,6,2,6,1
+4,17,121,6,2,6,1
+4,18,121,6,5,6,1
+4,19,121,6,5,6,1
+4,20,121,3,5,6,1
+4,21,121,3,2,6,1
+4,22,121,3,5,6,1
+4,23,121,3,5,6,1
+4,24,121,3,5,6,1
+4,25,121,3,5,6,1
+4,26,121,3,5,6,1
+4,27,121,3,5,6,1
+4,28,121,3,5,6,1
+4,29,121,3,5,6,1
+4,30,121,3,5,6,0
+4,31,121,3,5,6,0
+4,32,121,3,5,6,0
+4,33,121,3,5,6,0
+4,34,121,3,5,6,0
+4,35,121,3,5,6,0
+4,36,121,3,5,6,0
+4,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index e2d9465c..7ee664d5 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,107,7,2,5,0
-5,9,119,7,2,9,0
-5,10,119,7,2,0,0
-5,11,119,9,0,0,0
-5,12,103,6,0,5,0
-5,13,103,4,9,0,0
-5,14,112,4,9,5,0
-5,15,112,6,9,5,0
-5,16,112,6,9,0,0
-5,17,112,4,9,5,0
-5,18,112,6,9,5,0
-5,19,112,6,9,5,0
-5,20,100,6,9,0,0
-5,21,100,4,9,5,0
-5,22,112,6,9,5,0
-5,23,100,6,9,0,0
-5,24,100,4,9,5,0
-5,25,112,6,9,5,0
-5,26,100,4,9,5,0
-5,27,112,6,9,5,0
-5,28,100,4,9,5,0
-5,29,112,4,9,5,0
-5,30,112,6,9,5,0
-5,31,112,4,9,5,0
-5,32,112,9,9,5,0
-5,33,112,6,9,5,0
-5,34,112,9,9,5,0
-5,35,112,6,9,5,0
-5,36,112,9,9,5,0
-5,37,112,6,9,5,0
+5,8,122,6,8,2,0
+5,9,113,6,2,0,1
+5,10,113,6,3,0,1
+5,11,121,6,2,0,1
+5,12,121,6,2,6,1
+5,13,121,6,2,6,1
+5,14,121,6,2,6,1
+5,15,121,6,5,6,1
+5,16,121,6,5,6,1
+5,17,121,6,5,6,1
+5,18,121,3,5,6,1
+5,19,121,3,5,6,1
+5,20,121,3,5,6,1
+5,21,121,3,5,6,1
+5,22,121,3,5,6,1
+5,23,121,3,5,6,1
+5,24,121,3,5,6,1
+5,25,121,3,5,6,0
+5,26,121,3,5,6,0
+5,27,121,3,5,6,0
+5,28,121,3,5,6,0
+5,29,121,3,5,6,0
+5,30,121,3,5,6,0
+5,31,121,3,5,6,0
+5,32,121,3,5,6,0
+5,33,121,3,5,6,0
+5,34,121,3,5,6,0
+5,35,121,3,5,6,0
+5,36,121,3,5,6,0
+5,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index 328eac11..e3ab2a3d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,107,1,7,8,0
-6,9,110,6,7,5,1
-6,10,107,6,2,0,0
-6,11,110,6,9,0,0
-6,12,108,6,9,5,0
-6,13,112,6,9,8,0
-6,14,108,6,9,0,0
-6,15,112,6,9,0,0
-6,16,108,6,0,0,0
-6,17,110,6,9,0,0
-6,18,108,6,0,5,0
-6,19,110,6,9,5,0
-6,20,108,6,9,5,0
-6,21,112,6,9,5,0
-6,22,112,1,9,5,0
-6,23,112,6,0,5,0
-6,24,112,1,9,5,0
-6,25,112,6,0,5,0
-6,26,112,1,9,0,0
-6,27,112,6,0,5,0
-6,28,100,6,9,0,0
-6,29,100,4,9,5,0
-6,30,112,6,9,5,0
-6,31,100,6,9,5,0
-6,32,112,6,9,5,0
-6,33,100,6,9,5,0
-6,34,100,6,9,5,0
-6,35,100,4,9,5,0
-6,36,112,6,9,5,0
-6,37,100,9,9,5,0
-7,8,110,7,2,7,0
-7,9,119,1,2,0,0
-7,10,119,7,0,5,1
-7,11,107,6,0,5,0
-7,12,103,6,9,5,0
-7,13,112,6,9,8,0
-7,14,112,6,9,5,0
-7,15,112,6,9,0,0
-7,16,112,6,9,0,0
-7,17,100,4,9,0,0
-7,18,100,9,9,5,0
-7,19,112,6,9,5,0
-7,20,100,6,9,5,0
-7,21,100,6,9,5,0
-7,22,100,6,9,5,0
-7,23,100,6,9,5,0
-7,24,100,4,9,5,0
-7,25,112,4,9,5,0
-7,26,112,9,9,5,0
-7,27,112,6,9,5,0
-7,28,100,4,9,5,0
-7,29,112,9,9,5,0
-7,30,112,6,9,5,0
-7,31,112,9,9,5,0
-7,32,112,6,9,5,0
-7,33,112,6,9,5,0
-7,34,112,6,9,5,0
-7,35,100,6,9,5,0
-7,36,112,6,9,5,0
-7,37,100,6,9,8,0
+6,8,121,6,3,0,0
+6,9,121,4,9,6,1
+6,10,121,6,5,6,0
+6,11,121,4,5,6,1
+6,12,121,4,5,6,1
+6,13,121,4,5,6,1
+6,14,121,4,5,6,0
+6,15,121,3,5,6,1
+6,16,121,3,5,6,1
+6,17,121,3,5,6,1
+6,18,121,3,5,6,1
+6,19,121,3,2,6,1
+6,20,121,3,2,6,1
+6,21,121,3,5,6,1
+6,22,121,3,5,6,0
+6,23,121,3,5,6,1
+6,24,121,3,5,6,0
+6,25,121,3,5,6,1
+6,26,121,3,5,6,0
+6,27,121,3,5,6,1
+6,28,121,3,5,6,0
+6,29,121,3,5,6,0
+6,30,121,3,5,6,0
+6,31,121,3,5,6,0
+6,32,121,3,5,6,0
+6,33,121,3,5,6,0
+6,34,121,3,5,6,0
+6,35,121,3,5,6,0
+6,36,121,3,5,6,0
+6,37,121,3,5,6,0
+7,8,121,4,9,5,0
+7,9,121,4,3,6,0
+7,10,121,4,5,6,0
+7,11,121,4,5,6,0
+7,12,121,4,9,6,0
+7,13,121,4,9,6,0
+7,14,121,4,5,6,0
+7,15,121,3,5,6,1
+7,16,121,3,5,6,1
+7,17,121,3,5,6,1
+7,18,121,3,5,6,0
+7,19,121,3,5,6,1
+7,20,121,3,5,6,0
+7,21,121,3,5,6,0
+7,22,121,3,5,6,0
+7,23,121,3,5,6,0
+7,24,121,3,5,6,0
+7,25,121,3,5,6,0
+7,26,121,3,5,6,0
+7,27,121,3,5,6,0
+7,28,121,3,5,6,0
+7,29,121,3,5,6,0
+7,30,121,3,5,6,0
+7,31,121,3,5,6,0
+7,32,121,3,5,6,0
+7,33,121,3,5,6,0
+7,34,121,3,5,6,0
+7,35,121,3,5,6,0
+7,36,121,3,5,6,0
+7,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index 648373ce..7306d67d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,107,1,2,0,0
-8,9,119,6,2,7,1
-8,10,107,7,2,0,0
-8,11,103,6,2,0,0
-8,12,121,4,9,0,0
-8,13,112,6,9,5,0
-8,14,108,6,9,0,0
-8,15,108,6,9,0,0
-8,16,108,6,0,0,0
-8,17,110,6,0,0,0
-8,18,110,6,9,0,0
-8,19,108,6,9,5,0
-8,20,112,6,9,5,0
-8,21,112,6,9,4,0
-8,22,112,6,9,5,0
-8,23,112,6,9,5,0
-8,24,112,6,9,5,0
-8,25,112,1,9,5,0
-8,26,112,6,9,5,0
-8,27,100,1,9,0,0
-8,28,100,6,9,5,0
-8,29,100,6,9,5,0
-8,30,100,1,9,5,0
-8,31,100,6,9,5,0
-8,32,100,4,9,5,0
-8,33,112,4,9,5,0
-8,34,112,4,9,5,0
-8,35,112,4,9,5,0
-8,36,112,9,9,5,0
-8,37,112,6,9,5,0
+8,8,102,6,3,0,0
+8,9,121,6,3,0,0
+8,10,121,4,5,6,0
+8,11,121,3,5,6,0
+8,12,121,4,5,6,1
+8,13,121,4,9,6,0
+8,14,121,3,5,6,1
+8,15,121,3,5,6,1
+8,16,121,3,5,6,0
+8,17,121,3,5,6,0
+8,18,121,3,5,6,0
+8,19,121,3,5,6,0
+8,20,121,3,5,6,0
+8,21,121,3,5,6,0
+8,22,121,3,5,6,0
+8,23,121,3,5,6,0
+8,24,121,3,5,6,0
+8,25,121,3,5,6,0
+8,26,121,3,5,6,0
+8,27,121,3,5,6,0
+8,28,121,3,5,6,0
+8,29,121,3,5,6,0
+8,30,121,3,5,6,0
+8,31,121,3,5,6,0
+8,32,121,3,5,6,0
+8,33,121,3,5,6,0
+8,34,121,3,5,6,0
+8,35,121,3,5,6,0
+8,36,121,3,5,6,0
+8,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index fe16cefa..b3e95b3e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,107,1,2,7,1
-9,9,107,1,2,3,1
-9,10,110,0,2,3,1
-9,11,110,1,2,0,1
-9,12,103,0,7,0,1
-9,13,103,6,7,0,0
-9,14,112,6,9,0,0
-9,15,113,6,7,0,0
-9,16,110,6,2,0,0
-9,17,110,6,2,0,0
-9,18,102,6,2,0,0
-9,19,110,6,2,0,0
-9,20,102,6,2,0,1
-9,21,110,6,2,0,1
-9,22,110,6,2,0,1
-9,23,110,6,2,0,1
-9,24,110,6,2,0,1
-9,25,110,6,2,0,1
-9,26,110,6,2,0,1
-9,27,102,6,2,0,1
-9,28,110,6,2,0,1
-9,29,102,6,2,0,1
-9,30,110,6,2,0,1
-9,31,102,6,2,0,1
-9,32,110,6,2,0,1
-9,33,102,6,2,0,1
-9,34,110,6,2,0,1
-9,35,110,6,2,0,1
-9,36,102,6,2,0,1
-9,37,110,6,2,0,1
+9,8,113,6,3,2,0
+9,9,102,6,3,0,0
+9,10,121,6,2,0,1
+9,11,121,4,5,6,0
+9,12,121,3,5,6,1
+9,13,121,4,5,6,1
+9,14,121,3,5,6,1
+9,15,121,3,5,6,1
+9,16,121,3,5,6,1
+9,17,121,3,5,6,1
+9,18,121,3,5,6,1
+9,19,121,3,5,6,0
+9,20,121,3,5,6,0
+9,21,121,3,5,6,0
+9,22,121,3,5,6,0
+9,23,121,3,5,6,0
+9,24,121,3,5,6,0
+9,25,121,3,5,6,0
+9,26,121,3,5,6,0
+9,27,121,3,5,6,0
+9,28,121,3,5,6,0
+9,29,121,3,5,6,0
+9,30,121,3,5,6,0
+9,31,121,3,5,6,0
+9,32,121,3,5,6,0
+9,33,121,3,5,6,0
+9,34,121,3,5,6,0
+9,35,121,3,5,6,0
+9,36,121,3,5,6,0
+9,37,121,3,5,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
index 9c9a70fe..39c6e464 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,127,6,2,0,1
-0,9,118,6,2,0,1
-0,10,118,6,2,2,1
-0,11,118,6,2,2,1
-0,12,118,6,7,2,1
-0,13,118,6,3,2,1
-0,14,118,6,3,6,1
-0,15,118,6,3,6,1
-0,16,127,6,3,other,0
-0,17,127,6,3,0,1
-0,18,127,6,3,0,1
-0,19,127,6,3,0,1
-0,20,118,6,3,0,1
-0,21,118,6,3,0,1
-0,22,118,6,3,2,1
-0,23,118,6,3,2,1
-0,24,118,6,3,2,1
-0,25,124,6,3,0,1
-0,26,124,6,3,0,1
-0,27,124,6,3,0,1
-0,28,104,6,3,0,1
-0,29,104,6,3,1,1
-0,30,109,6,3,8,1
-0,31,109,6,3,1,1
-0,32,109,4,3,8,0
-0,33,113,3,3,0,1
-0,34,122,3,5,8,0
-0,35,118,3,5,0,1
-0,36,118,6,5,8,0
-0,37,118,6,5,2,1
+0,8,102,6,3,8,1
+0,9,121,6,3,8,1
+0,10,104,6,3,8,1
+0,11,121,6,3,8,1
+0,12,121,6,3,8,1
+0,13,104,6,3,8,0
+0,14,104,3,3,0,0
+0,15,104,3,3,0,0
+0,16,104,3,3,0,0
+0,17,104,3,3,0,0
+0,18,104,3,3,0,0
+0,19,104,3,3,0,0
+0,20,104,3,3,0,1
+0,21,103,6,3,0,1
+0,22,113,6,3,8,1
+0,23,113,6,3,8,1
+0,24,113,6,3,8,1
+0,25,104,6,3,8,1
+0,26,104,6,3,8,1
+0,27,104,6,3,8,1
+0,28,104,6,3,8,1
+0,29,104,6,3,8,1
+0,30,104,9,3,0,0
+0,31,104,3,3,0,1
+0,32,104,9,3,0,1
+0,33,104,9,3,0,1
+0,34,104,3,3,8,1
+0,35,104,6,3,8,0
+0,36,121,6,3,8,0
+0,37,121,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
index 18c70481..b1f5c970 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,118,6,2,2,1
-1,9,118,6,9,0,1
-1,10,118,6,9,2,1
-1,11,118,6,9,2,1
-1,12,118,6,9,2,1
-1,13,118,6,3,2,1
-1,14,124,6,3,2,1
-1,15,104,6,3,8,1
-1,16,114,6,3,5,1
-1,17,114,6,3,5,1
-1,18,114,6,3,5,1
-1,19,114,6,3,5,1
-1,20,114,6,3,5,1
-1,21,109,6,3,5,1
-1,22,127,6,3,5,1
-1,23,127,6,9,5,1
-1,24,127,6,3,5,1
-1,25,118,6,3,5,1
-1,26,118,6,3,5,1
-1,27,118,6,3,5,1
-1,28,118,6,3,5,1
-1,29,118,6,3,9,1
-1,30,118,6,3,5,1
-1,31,127,6,3,5,1
-1,32,127,6,3,5,1
-1,33,127,6,3,5,1
-1,34,127,6,3,5,1
-1,35,127,6,3,5,1
-1,36,118,6,3,5,1
-1,37,118,6,3,0,1
+1,8,121,6,3,8,1
+1,9,121,6,3,8,0
+1,10,121,6,3,8,0
+1,11,121,6,3,8,0
+1,12,121,6,3,8,0
+1,13,104,3,3,8,0
+1,14,104,3,3,0,0
+1,15,104,3,3,0,0
+1,16,104,3,3,0,0
+1,17,104,3,3,0,0
+1,18,104,3,3,0,0
+1,19,104,3,3,0,0
+1,20,104,3,3,0,0
+1,21,121,6,3,0,1
+1,22,113,6,3,8,1
+1,23,121,6,3,8,1
+1,24,104,6,3,8,1
+1,25,104,6,3,8,0
+1,26,104,6,3,0,0
+1,27,104,9,3,0,1
+1,28,104,3,3,0,1
+1,29,104,9,3,0,1
+1,30,104,3,3,0,1
+1,31,104,3,3,0,1
+1,32,103,6,3,8,1
+1,33,113,6,3,8,1
+1,34,121,6,3,8,1
+1,35,121,6,3,8,1
+1,36,104,6,3,8,1
+1,37,104,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
index cec8e864..b32ec715 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,127,6,2,0,1
-2,9,127,6,3,0,1
-2,10,127,6,3,0,1
-2,11,118,6,3,0,1
-2,12,118,6,5,2,1
-2,13,118,6,7,2,1
-2,14,118,6,7,2,1
-2,15,124,6,3,2,1
-2,16,124,6,3,2,1
-2,17,124,6,3,2,1
-2,18,124,6,3,0,1
-2,19,104,6,3,0,1
-2,20,104,6,3,1,1
-2,21,109,6,3,0,1
-2,22,109,6,3,0,1
-2,23,104,3,3,8,1
-2,24,113,6,3,8,1
-2,25,113,6,3,8,1
-2,26,113,4,3,8,0
-2,27,113,3,4,8,0
-2,28,unknown,4,4,8,2
-2,29,unknown,4,4,8,0
-2,30,109,4,4,8,2
-2,31,109,4,3,8,2
-2,32,127,3,6,8,1
-2,33,unknown,3,5,4,1
-2,34,unknown,3,9,4,1
-2,35,unknown,0,9,8,1
-2,36,127,4,3,8,0
-2,37,118,4,9,3,1
+2,8,102,6,3,8,1
+2,9,121,6,3,8,1
+2,10,104,6,3,8,1
+2,11,104,6,3,8,0
+2,12,104,6,3,8,0
+2,13,104,6,3,8,0
+2,14,104,6,3,0,0
+2,15,104,3,3,0,0
+2,16,104,3,3,0,0
+2,17,104,3,3,0,0
+2,18,104,3,3,0,1
+2,19,103,4,3,0,1
+2,20,102,6,3,8,1
+2,21,121,6,3,8,1
+2,22,104,6,3,8,1
+2,23,104,6,3,8,1
+2,24,104,6,3,8,0
+2,25,104,6,3,8,0
+2,26,104,6,3,0,0
+2,27,104,9,3,0,1
+2,28,104,3,3,0,0
+2,29,104,3,3,0,0
+2,30,104,3,3,0,1
+2,31,103,4,3,0,1
+2,32,102,6,3,8,1
+2,33,121,6,3,8,1
+2,34,121,6,3,8,1
+2,35,104,6,3,8,1
+2,36,104,6,3,8,0
+2,37,104,6,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
index cd3c4c28..03dc6b54 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,127,2,2,2,1
-3,9,127,6,2,0,1
-3,10,127,6,3,2,1
-3,11,118,6,3,0,1
-3,12,118,6,3,2,1
-3,13,118,6,3,2,1
-3,14,118,6,3,2,1
-3,15,118,6,3,6,1
-3,16,109,6,3,8,1
-3,17,109,6,3,0,1
-3,18,109,6,3,0,1
-3,19,127,6,3,0,0
-3,20,109,6,3,0,1
-3,21,127,6,3,0,1
-3,22,118,6,3,0,1
-3,23,118,4,5,2,1
-3,24,118,6,5,2,1
-3,25,118,6,3,2,1
-3,26,118,6,3,2,1
-3,27,118,6,3,2,1
-3,28,118,6,3,0,1
-3,29,104,6,3,0,1
-3,30,104,6,3,1,1
-3,31,104,6,3,5,1
-3,32,104,6,3,1,1
-3,33,104,6,3,8,0
-3,34,109,6,3,5,1
-3,35,127,6,3,5,0
-3,36,118,6,9,8,0
-3,37,118,6,9,5,1
-4,8,127,6,2,0,1
-4,9,127,6,3,0,1
-4,10,127,6,3,0,1
-4,11,127,6,3,0,1
-4,12,118,6,3,0,1
-4,13,118,6,7,2,1
-4,14,118,1,7,2,1
-4,15,118,1,7,2,1
-4,16,118,6,5,2,1
-4,17,124,6,3,2,1
-4,18,124,6,3,2,1
-4,19,124,6,3,0,1
-4,20,109,6,3,0,1
-4,21,109,6,3,0,1
-4,22,109,6,3,0,1
-4,23,109,6,3,0,1
-4,24,109,6,3,8,1
-4,25,unknown,3,3,0,1
-4,26,124,3,5,8,1
-4,27,124,4,5,2,0
-4,28,124,6,7,2,1
-4,29,unknown,6,5,2,1
-4,30,124,6,3,2,1
-4,31,unknown,6,3,8,1
-4,32,unknown,6,3,8,1
-4,33,unknown,6,3,8,1
-4,34,104,6,3,8,1
-4,35,104,6,3,8,1
-4,36,104,6,3,8,1
-4,37,104,6,3,8,0
+3,8,121,6,3,8,1
+3,9,104,4,3,8,0
+3,10,121,3,3,8,0
+3,11,104,3,3,8,0
+3,12,121,3,3,8,0
+3,13,121,3,3,8,0
+3,14,104,3,3,8,0
+3,15,104,3,3,0,0
+3,16,104,3,3,0,0
+3,17,104,3,3,0,0
+3,18,104,3,3,0,0
+3,19,104,3,3,0,0
+3,20,104,3,3,0,0
+3,21,104,3,3,0,0
+3,22,121,6,3,0,1
+3,23,113,6,3,8,1
+3,24,121,6,3,8,1
+3,25,104,6,3,8,1
+3,26,104,6,3,8,0
+3,27,104,6,3,0,0
+3,28,104,9,3,0,1
+3,29,104,3,3,0,1
+3,30,104,9,3,0,1
+3,31,104,3,3,0,1
+3,32,104,3,3,0,1
+3,33,103,6,3,8,1
+3,34,113,6,3,8,1
+3,35,121,6,3,8,1
+3,36,121,6,3,8,1
+3,37,104,6,3,8,1
+4,8,102,6,9,8,1
+4,9,121,6,3,8,1
+4,10,121,6,3,8,1
+4,11,121,6,3,8,0
+4,12,121,6,3,8,0
+4,13,121,6,3,8,0
+4,14,104,3,3,8,0
+4,15,104,3,3,0,0
+4,16,104,3,3,0,0
+4,17,104,3,3,0,0
+4,18,104,3,3,0,0
+4,19,104,3,3,0,0
+4,20,104,3,3,0,0
+4,21,104,3,3,0,0
+4,22,121,6,3,0,1
+4,23,113,6,3,8,1
+4,24,121,6,3,8,1
+4,25,104,6,3,8,1
+4,26,104,6,3,8,0
+4,27,104,6,3,0,0
+4,28,104,9,3,0,1
+4,29,104,3,3,0,1
+4,30,104,9,3,0,1
+4,31,104,3,3,0,1
+4,32,104,3,3,0,1
+4,33,103,6,3,8,1
+4,34,113,6,3,8,1
+4,35,121,6,3,8,1
+4,36,121,6,3,8,1
+4,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
index 16a159f1..549102b2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,127,6,2,2,0
-5,9,118,6,2,0,1
-5,10,127,6,2,0,1
-5,11,127,6,3,0,1
-5,12,118,6,3,0,1
-5,13,118,6,5,2,1
-5,14,118,6,3,2,1
-5,15,124,6,3,8,1
-5,16,118,6,3,2,1
-5,17,127,6,3,8,other
-5,18,109,6,3,0,1
-5,19,109,6,3,0,other
-5,20,109,6,3,0,1
-5,21,112,6,3,0,1
-5,22,118,6,3,8,1
-5,23,124,6,3,8,1
-5,24,124,4,3,2,1
-5,25,124,6,5,2,1
-5,26,124,6,3,2,1
-5,27,unknown,6,3,2,1
-5,28,104,6,3,8,1
-5,29,104,6,3,8,1
-5,30,104,6,3,8,1
-5,31,104,6,3,8,1
-5,32,104,6,3,5,1
-5,33,104,6,3,8,1
-5,34,127,6,3,5,0
-5,35,109,1,9,8,0
-5,36,127,1,9,5,2
-5,37,127,9,9,5,1
+5,8,113,4,9,8,1
+5,9,113,6,3,8,0
+5,10,104,6,3,8,0
+5,11,104,6,3,0,0
+5,12,104,9,3,0,0
+5,13,121,3,3,8,1
+5,14,121,6,3,8,0
+5,15,104,3,3,8,0
+5,16,104,3,3,0,0
+5,17,104,3,3,0,0
+5,18,104,3,3,0,0
+5,19,121,3,3,0,0
+5,20,104,3,3,0,0
+5,21,104,3,3,0,0
+5,22,104,3,3,0,0
+5,23,121,4,3,0,1
+5,24,121,6,3,8,1
+5,25,104,6,3,8,1
+5,26,104,6,3,8,0
+5,27,104,3,3,0,0
+5,28,104,3,3,0,1
+5,29,104,4,3,0,1
+5,30,104,9,3,0,1
+5,31,104,3,3,0,1
+5,32,104,3,3,0,1
+5,33,104,6,3,0,1
+5,34,113,6,3,8,1
+5,35,113,6,3,8,1
+5,36,113,6,3,8,1
+5,37,113,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
index 07028e2d..68e5ec32 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,127,6,2,0,1
-6,9,127,6,3,0,1
-6,10,127,6,3,0,1
-6,11,127,4,3,0,1
-6,12,127,4,5,0,0
-6,13,112,1,7,2,1
-6,14,118,1,5,2,1
-6,15,118,1,7,2,1
-6,16,118,1,5,2,1
-6,17,118,6,3,2,1
-6,18,118,6,3,2,1
-6,19,118,6,3,2,1
-6,20,118,6,3,2,1
-6,21,109,6,3,0,1
-6,22,109,6,3,0,1
-6,23,109,6,3,0,1
-6,24,109,6,3,0,1
-6,25,109,6,3,0,1
-6,26,127,6,3,0,1
-6,27,118,6,3,0,1
-6,28,118,6,5,0,1
-6,29,118,4,5,other,0
-6,30,118,6,7,0,1
-6,31,118,4,5,2,0
-6,32,118,6,7,2,1
-6,33,118,6,3,2,1
-6,34,118,6,3,0,1
-6,35,127,6,3,0,1
-6,36,127,6,3,0,1
-6,37,127,6,3,2,1
-7,8,127,6,2,2,1
-7,9,127,6,3,0,1
-7,10,127,6,3,0,1
-7,11,118,6,3,0,1
-7,12,118,6,3,2,1
-7,13,118,6,3,2,1
-7,14,118,6,3,2,1
-7,15,118,6,3,2,1
-7,16,118,6,3,0,1
-7,17,124,6,3,8,1
-7,18,unknown,6,3,0,1
-7,19,109,6,3,1,1
-7,20,109,6,3,1,0
-7,21,109,6,3,0,1
-7,22,109,6,3,0,0
-7,23,109,6,5,0,1
-7,24,118,6,5,0,0
-7,25,118,3,7,2,1
-7,26,118,1,5,2,0
-7,27,118,6,7,2,1
-7,28,118,6,5,2,1
-7,29,118,6,3,2,1
-7,30,112,6,3,2,1
-7,31,112,6,3,0,1
-7,32,112,6,3,8,1
-7,33,109,6,3,8,1
-7,34,109,6,3,5,1
-7,35,109,6,3,1,1
-7,36,109,6,3,8,0
-7,37,109,6,3,0,1
+6,8,102,6,2,8,1
+6,9,113,6,3,8,1
+6,10,121,6,3,8,1
+6,11,121,6,3,8,1
+6,12,121,6,3,8,1
+6,13,121,6,3,8,0
+6,14,121,6,3,8,0
+6,15,104,3,3,8,0
+6,16,104,3,3,0,0
+6,17,104,3,3,0,0
+6,18,104,3,3,0,0
+6,19,104,3,3,0,0
+6,20,104,3,3,0,0
+6,21,104,3,3,0,0
+6,22,104,3,3,0,0
+6,23,121,6,3,0,1
+6,24,113,6,3,8,1
+6,25,121,6,3,8,1
+6,26,104,6,3,8,1
+6,27,104,6,3,8,0
+6,28,104,6,3,0,0
+6,29,104,9,3,0,1
+6,30,104,3,3,0,1
+6,31,104,9,3,0,1
+6,32,104,3,3,0,1
+6,33,104,3,3,0,1
+6,34,103,6,3,8,1
+6,35,113,6,3,8,1
+6,36,121,6,3,8,1
+6,37,121,6,3,8,1
+7,8,121,6,9,8,1
+7,9,104,6,3,8,1
+7,10,104,6,3,8,0
+7,11,104,6,3,0,0
+7,12,104,6,3,0,1
+7,13,121,6,3,8,1
+7,14,104,6,3,8,0
+7,15,104,3,3,0,0
+7,16,104,3,3,0,0
+7,17,104,3,3,0,1
+7,18,104,4,3,0,1
+7,19,103,6,3,8,1
+7,20,104,6,3,8,1
+7,21,104,9,3,0,0
+7,22,121,6,3,8,1
+7,23,104,6,3,8,1
+7,24,104,6,3,8,1
+7,25,104,6,3,8,1
+7,26,104,6,3,8,0
+7,27,104,3,3,0,0
+7,28,104,3,3,0,0
+7,29,104,3,3,0,0
+7,30,104,3,3,0,0
+7,31,104,3,3,0,1
+7,32,104,3,3,0,1
+7,33,103,4,3,0,1
+7,34,113,6,3,8,1
+7,35,113,6,3,8,1
+7,36,113,6,3,8,1
+7,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
index bc2b4d05..84a9c524 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,127,6,2,0,1
-8,9,127,6,3,0,1
-8,10,127,6,3,0,1
-8,11,118,6,9,2,1
-8,12,118,6,7,2,1
-8,13,118,6,9,6,1
-8,14,127,6,3,8,1
-8,15,127,6,3,5,1
-8,16,118,6,3,0,1
-8,17,124,6,3,0,1
-8,18,124,6,3,2,1
-8,19,126,6,3,5,1
-8,20,127,6,3,5,1
-8,21,118,6,3,0,1
-8,22,118,6,3,8,1
-8,23,118,6,3,0,1
-8,24,118,6,3,8,1
-8,25,118,6,3,0,1
-8,26,104,6,3,0,1
-8,27,126,6,3,5,0
-8,28,109,6,3,0,1
-8,29,127,6,3,0,0
-8,30,109,6,3,0,1
-8,31,127,6,3,0,0
-8,32,118,3,5,0,1
-8,33,118,6,5,2,0
-8,34,118,6,7,2,1
-8,35,118,6,5,2,1
-8,36,118,6,3,2,1
-8,37,112,6,3,2,1
+8,8,113,6,3,8,1
+8,9,121,6,3,8,1
+8,10,104,6,3,8,1
+8,11,104,6,3,8,1
+8,12,121,6,3,8,1
+8,13,121,6,3,8,1
+8,14,104,6,3,8,0
+8,15,104,9,3,0,0
+8,16,104,3,3,0,0
+8,17,104,3,3,0,0
+8,18,104,3,3,0,0
+8,19,104,3,3,0,0
+8,20,104,3,3,0,0
+8,21,104,3,3,0,1
+8,22,103,6,3,8,1
+8,23,121,6,3,8,1
+8,24,104,6,3,8,1
+8,25,104,6,3,8,1
+8,26,104,6,3,8,0
+8,27,104,6,3,0,0
+8,28,104,9,3,0,1
+8,29,104,9,3,0,1
+8,30,104,3,3,0,1
+8,31,104,3,3,0,1
+8,32,104,6,3,0,1
+8,33,113,6,3,8,1
+8,34,121,6,3,8,1
+8,35,121,6,3,8,1
+8,36,121,6,3,8,1
+8,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
index 940e8923..ad971442 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,127,6,2,other,1
-9,9,127,6,3,0,1
-9,10,127,6,3,0,1
-9,11,118,6,3,0,1
-9,12,118,6,3,other,1
-9,13,126,6,3,8,1
-9,14,113,6,3,8,0
-9,15,113,4,3,0,1
-9,16,118,6,3,0,1
-9,17,118,6,3,8,1
-9,18,118,6,3,0,1
-9,19,118,6,3,0,0
-9,20,118,6,3,0,1
-9,21,127,6,3,8,0
-9,22,118,6,3,0,1
-9,23,118,6,3,8,0
-9,24,109,6,3,0,1
-9,25,127,6,3,0,0
-9,26,112,6,5,0,1
-9,27,112,6,3,2,1
-9,28,118,6,3,8,1
-9,29,118,4,3,6,1
-9,30,124,6,3,8,1
-9,31,124,4,3,2,1
-9,32,124,6,3,2,1
-9,33,124,6,3,2,1
-9,34,104,6,3,1,1
-9,35,104,6,3,1,1
+9,8,113,6,3,0,0
+9,9,121,6,3,8,1
+9,10,104,6,3,8,1
+9,11,104,6,3,8,1
+9,12,104,6,3,8,1
+9,13,104,6,3,8,1
+9,14,104,6,3,8,0
+9,15,104,6,3,0,0
+9,16,104,9,3,0,2
+9,17,104,3,3,0,0
+9,18,104,3,3,0,1
+9,19,104,9,3,0,1
+9,20,103,6,3,8,1
+9,21,121,6,3,8,1
+9,22,104,6,3,8,1
+9,23,121,6,3,8,1
+9,24,121,6,3,8,1
+9,25,104,6,3,8,0
+9,26,104,9,3,0,0
+9,27,104,3,3,0,0
+9,28,104,3,3,0,0
+9,29,104,3,3,0,0
+9,30,104,3,3,0,0
+9,31,104,3,3,0,0
+9,32,104,3,3,0,1
+9,33,103,6,3,8,1
+9,34,121,6,3,8,1
+9,35,104,6,3,8,1
 9,36,104,6,3,8,1
-9,37,104,6,3,5,1
+9,37,104,6,3,8,0
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
index cd526dfe..498e4491 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0063816966,0.006468037,0.024559338,0.0068818387,0.03805276,0.034742553,0.018915702,0.036322363,0.0075609423,0.0050840396,0.031691603,0.021121172,0.022060119,0.035620775,0.009137776,0.27302954,0.0035919647,0.005177132,0.012216523,0.008537154,0.039849125,0.022662228,0.0051864795,0.06108719,0.03615953,0.0095728375,0.13449506,0.005642359,0.020200068,0.019690767,0.010418236,0.027883159
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0091290055,0.0077892784,0.007239608,0.020136122,0.006804843,0.065107465,0.09113526,0.03846739,0.046789855,0.011856737,0.010647341,0.0430088,0.01229634,0.024512613,0.043958448,0.015431894,0.10510967,0.008920799,0.012439459,0.013178185,0.0115483245,0.042840227,0.020539861,0.013582593,0.07060103,0.024588121,0.021217003,0.10432629,0.008402498,0.025339192,0.025642872,0.0106556,0.026757332
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
index fde48406..2a1876ce 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0049449983,0.0042691776,0.021052632,0.0058146087,0.030396372,0.032892935,0.013441515,0.028407525,0.0050566397,0.0035540466,0.026184268,0.023003984,0.023714056,0.02700193,0.010255454,0.193688,0.003523743,0.0065743346,0.0075657065,0.0084641315,0.036537226,0.017473338,0.0043345136,0.08128023,0.0313194,0.007980294,0.2656846,0.0042771567,0.018264784,0.017538315,0.00946817,0.0260359
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.008205366,0.008331864,0.0074459263,0.021188624,0.0073211472,0.061644685,0.07545811,0.034722853,0.043902382,0.010114443,0.010213885,0.042470828,0.012390661,0.022267908,0.032764185,0.01806368,0.12040593,0.009458063,0.014117386,0.011872736,0.013330289,0.041174386,0.020099496,0.013925595,0.072014526,0.024408026,0.018947104,0.13367255,0.00828295,0.024368336,0.02391859,0.011060069,0.022437433
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
index 9d1b4695..dd67b48c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0070600286,0.005355665,0.024419116,0.006750209,0.03974409,0.060156815,0.021477934,0.029202912,0.006069319,0.0048991265,0.030470513,0.029163782,0.02428838,0.040785775,0.008904983,0.2596093,0.0037373006,0.006169808,0.0080467705,0.010179635,0.037312083,0.021897098,0.005871768,0.067005344,0.024129355,0.008555806,0.117676124,0.00547475,0.022196412,0.01896278,0.011333607,0.033093367
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.012586267,0.009460248,0.01056256,0.024452873,0.0086881975,0.052435704,0.10992195,0.0706047,0.03934652,0.017093472,0.012070613,0.038141258,0.016003482,0.024356524,0.05110281,0.022539876,0.057360444,0.0130562065,0.014306477,0.011672565,0.013660459,0.019611526,0.033125043,0.01840493,0.062889695,0.025482355,0.023796566,0.06978374,0.011446344,0.029403456,0.027195802,0.014777219,0.034660116
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
index 1502f2fd..426a7d74 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0073428494,0.0054249666,0.023946963,0.0076724873,0.05550431,0.041024055,0.02343562,0.020106466,0.0058254926,0.005979295,0.028401894,0.027254183,0.03406939,0.03580471,0.0151072955,0.1438499,0.0050315387,0.008578817,0.010699981,0.009593445,0.039330103,0.022077711,0.0072748065,0.116141096,0.03376495,0.009846874,0.17126702,0.0052832607,0.029783443,0.016034205,0.010827838,0.023715047
-0.0064484207,0.0047859964,0.030502057,0.006627743,0.040947184,0.053503774,0.015772693,0.036306743,0.0063552163,0.0056825983,0.029170575,0.01872254,0.034786753,0.037946172,0.016469195,0.19048862,0.004163757,0.0076852962,0.012417914,0.010849618,0.02091185,0.017051233,0.0060045645,0.041049585,0.118274905,0.009881875,0.1420341,0.0046904692,0.015074657,0.022749325,0.013102149,0.019542495
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.008913519,0.007365497,0.007884704,0.019221375,0.008245324,0.061635274,0.10626261,0.04613362,0.039076947,0.013314558,0.01237192,0.04430893,0.014824947,0.02413848,0.04149791,0.018489841,0.07907649,0.010472187,0.011848482,0.012690502,0.011933226,0.034617245,0.024047974,0.015420446,0.064966045,0.021884238,0.020587962,0.12284686,0.008948034,0.029838102,0.020556517,0.010585136,0.025995124
+0.0094871465,0.011149254,0.009741706,0.023295518,0.012869805,0.05367749,0.048896357,0.031161776,0.05537468,0.012018335,0.013214679,0.028068008,0.016052,0.03157924,0.03311481,0.021816775,0.052957285,0.012617017,0.022129932,0.015422226,0.02169244,0.031480502,0.015010904,0.02252592,0.07512524,0.07089812,0.022424813,0.11010506,0.011601552,0.021417523,0.044051338,0.016005058,0.02301758
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
index 0842a970..10cd194d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0070105107,0.005817882,0.02375718,0.0067135287,0.030807294,0.05775498,0.019987397,0.0336207,0.007363106,0.005212145,0.029612567,0.028933655,0.021020675,0.036609236,0.009110481,0.2528482,0.003447206,0.0057708235,0.008732559,0.009616723,0.03541558,0.02171815,0.0055170287,0.059915405,0.022731164,0.009284548,0.14595954,0.006151366,0.020992719,0.020696605,0.011107996,0.036763098
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.00905651,0.008449699,0.007513583,0.01789081,0.009958172,0.059586458,0.09060577,0.05163472,0.05080496,0.015027517,0.0131017715,0.046035837,0.0129659865,0.021924116,0.043815948,0.022697417,0.07993372,0.0115594845,0.015695909,0.011922975,0.012971503,0.034338184,0.025293143,0.016621767,0.06711588,0.022368787,0.02659395,0.0939604,0.010572614,0.026793035,0.024672363,0.012390149,0.026126813
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
index b1ea5679..e7d6329d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0061376016,0.0055368184,0.02686486,0.006603274,0.05257738,0.04243002,0.019543372,0.03439729,0.0066835904,0.0054204813,0.027758658,0.018920917,0.026187373,0.03562787,0.0103454115,0.21988945,0.0038413862,0.0058459127,0.011070427,0.008652827,0.04451155,0.019691685,0.0054005976,0.073532715,0.056998175,0.008044991,0.14140613,0.00463186,0.019962309,0.017421842,0.010648626,0.023414658
-0.006748481,0.0056581046,0.025895737,0.0065028435,0.036626477,0.04927831,0.01836768,0.033150148,0.006620763,0.0048519536,0.030104272,0.023016918,0.02434799,0.040022057,0.008816339,0.259297,0.0034051202,0.005711451,0.00943816,0.009649961,0.0344164,0.01987012,0.0051980447,0.06372621,0.030300407,0.008548483,0.14363155,0.005186423,0.022001177,0.019676577,0.010164736,0.029770095
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.008020919,0.0066487696,0.0072186273,0.023239477,0.009595765,0.059273914,0.062329326,0.027587652,0.0486198,0.008902664,0.009877493,0.03435876,0.017895322,0.022803104,0.0360042,0.016689297,0.112537004,0.008584821,0.014853362,0.013560702,0.015122532,0.043928403,0.0197575,0.01412689,0.07697556,0.037536968,0.014428177,0.14790235,0.008241942,0.016693683,0.024688615,0.013138984,0.018857364
+0.0086295465,0.007917115,0.007522475,0.023743981,0.0073922593,0.06037666,0.08741754,0.043700412,0.04167852,0.012650199,0.01019312,0.043437473,0.013766954,0.021962289,0.04165744,0.018287549,0.10325626,0.009215165,0.01214539,0.011779306,0.012235232,0.045917273,0.018463248,0.014126799,0.068317495,0.021482795,0.016241094,0.12552734,0.008585244,0.025119003,0.02262137,0.01049176,0.02414174
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
index 179c80ec..0efa80de 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.006577232,0.005941025,0.025182042,0.0067693973,0.04229883,0.07847597,0.0240807,0.0301303,0.0069031618,0.0057024322,0.03144487,0.028760783,0.024837973,0.04591679,0.012943257,0.22671296,0.0036830874,0.005482532,0.009327667,0.0106034735,0.033890318,0.023517158,0.005874784,0.059042808,0.030655574,0.009653993,0.10965886,0.005722431,0.023167748,0.021588275,0.0129337795,0.032519735
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.00940564,0.0063802544,0.0065206103,0.027256832,0.0066316505,0.05138203,0.10203898,0.048445128,0.049608555,0.00873864,0.010182529,0.026134778,0.014573924,0.02411956,0.038605724,0.036600724,0.07013269,0.011340922,0.020680869,0.011217838,0.01796068,0.016984288,0.026266541,0.013434889,0.07714,0.08002693,0.012619701,0.06227914,0.007597093,0.02409692,0.043419037,0.014213739,0.0239632
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
index 462d3218..30a706e1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0062551964,0.005907708,0.02252796,0.006632302,0.0375025,0.042552833,0.01901747,0.029515766,0.0066615692,0.0049623456,0.028662894,0.025402665,0.024885334,0.03338471,0.00894907,0.23237301,0.0036392827,0.0054352307,0.00970244,0.008878192,0.044655215,0.022057772,0.004962897,0.08231816,0.02434962,0.008952783,0.16011275,0.0052076653,0.023783736,0.018114585,0.01071558,0.03192075
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.009661113,0.007831191,0.009208629,0.018906532,0.008831174,0.065691166,0.10545667,0.038684368,0.039912082,0.01415859,0.013239993,0.03936315,0.0131430235,0.025695173,0.04284285,0.015051846,0.08823259,0.009918131,0.012693808,0.015639171,0.01375921,0.053180333,0.021613518,0.016188124,0.064060025,0.024109922,0.022105852,0.08915296,0.01004957,0.026042204,0.028148724,0.010506249,0.026922056
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
index 030ea209..93284692 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0056786733,0.0056396397,0.019625373,0.00989916,0.028172797,0.06280763,0.018479379,0.03159086,0.00802512,0.004560256,0.02441183,0.030483417,0.024728071,0.034464385,0.02173319,0.23811956,0.009270334,0.010733643,0.013194141,0.012353504,0.027120981,0.01968651,0.007962703,0.08780438,0.07940018,0.006885157,0.066043444,0.0059645167,0.032393124,0.024882888,0.012979289,0.014905925
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0042463774,0.005140092,0.0060796365,0.016366256,0.0067424467,0.03495936,0.06623093,0.020122634,0.050006244,0.00649412,0.0041204235,0.02871717,0.018642396,0.028757636,0.033106435,0.01599839,0.23275402,0.007276528,0.008622721,0.017800169,0.010982657,0.028370103,0.02974473,0.0034585185,0.06776658,0.046286162,0.0064784237,0.120085396,0.0039499593,0.021950865,0.015566807,0.018041762,0.015134025
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
index 14379442..d5bd2f61 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.007233303,0.006430847,0.02017487,0.009407937,0.03021497,0.063631795,0.015276678,0.030304415,0.009666988,0.0051930123,0.022986863,0.028146453,0.02071075,0.03240998,0.026094265,0.2046195,0.012769067,0.010396391,0.013358195,0.012718733,0.019840602,0.020281564,0.011074808,0.09205814,0.100550726,0.008028014,0.06728822,0.0068513094,0.025007999,0.030172698,0.01773199,0.019368906
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0041903867,0.0054771546,0.0058246264,0.015669782,0.0071306545,0.030696195,0.06473787,0.018447235,0.05242684,0.0068101883,0.004801957,0.029178623,0.022633398,0.02907579,0.033386257,0.018674878,0.19909294,0.006572294,0.009357265,0.01596702,0.013374633,0.026702495,0.032056045,0.003591021,0.07047866,0.053875104,0.0050279247,0.13526632,0.0042218207,0.021093741,0.021248914,0.019235754,0.013676274
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
index f54d9b81..ae9e3e1a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.007526629,0.0073912153,0.014185694,0.012860334,0.025072798,0.06497633,0.019334141,0.030267734,0.010712494,0.005902856,0.025737327,0.036715742,0.025636889,0.027740806,0.023948258,0.27562156,0.012211119,0.012502682,0.012710197,0.011621473,0.015836276,0.020809514,0.009763865,0.0765267,0.053563163,0.007314694,0.053682905,0.008438264,0.03344781,0.028250687,0.015079057,0.014610726
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0054918397,0.0063162544,0.0061463765,0.016365495,0.0076878374,0.034990918,0.0610308,0.020188866,0.0626216,0.008143705,0.0056989286,0.028557653,0.01842186,0.026081735,0.02933535,0.018272739,0.24774212,0.008434298,0.009743969,0.017181823,0.013515602,0.028113525,0.035016388,0.004349286,0.05375894,0.048733603,0.0074889436,0.08502142,0.0040941467,0.018917236,0.024357485,0.024139129,0.014040137
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
index b5ae3c31..71f47b47 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.007322969,0.005654382,0.019285746,0.01174128,0.028390037,0.051265758,0.017005999,0.023170853,0.00963273,0.0050818776,0.019823872,0.027385317,0.024844429,0.025241857,0.024961751,0.27105013,0.010011087,0.008763377,0.012323763,0.010727558,0.01852274,0.019683188,0.010597888,0.06964346,0.08186063,0.0069930656,0.08686133,0.0072649037,0.02479653,0.024108676,0.017185295,0.018797552
-0.0073998687,0.0062193214,0.028439596,0.010987616,0.04028016,0.06468185,0.016779186,0.03044594,0.008741515,0.0064253476,0.019445894,0.021537095,0.028751887,0.024736227,0.023209155,0.2096015,0.013735791,0.012793584,0.016223645,0.0155749805,0.02702452,0.021603411,0.012366309,0.06640046,0.08558532,0.0070591993,0.07619427,0.007099011,0.026309213,0.02994487,0.015465732,0.018937428
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.004381987,0.006287049,0.006620055,0.019419348,0.006867307,0.03929928,0.07064888,0.017149057,0.060524255,0.006789356,0.0050974856,0.027337514,0.024331838,0.03136498,0.029390393,0.020899236,0.15376608,0.0074551413,0.009935729,0.015628591,0.013709809,0.03817771,0.03494091,0.0051154993,0.06018743,0.075591475,0.0067811315,0.11117232,0.0047863433,0.015857743,0.028808644,0.024059782,0.017617658
+0.004469706,0.006364652,0.0068032146,0.020612448,0.007166283,0.039386127,0.08379896,0.017597469,0.06735702,0.007070649,0.005259082,0.027966166,0.024157904,0.03855127,0.026747128,0.019213978,0.11608029,0.0063566845,0.010583058,0.01697354,0.012141066,0.043337125,0.031579476,0.004105956,0.061916213,0.07502441,0.0060634622,0.11869589,0.005418634,0.019128878,0.027781855,0.025369624,0.01692181
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
index 0c5b5d15..0e9d5181 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008149271,0.0075202077,0.018574426,0.013677846,0.027515505,0.07347045,0.022090634,0.027623143,0.009446033,0.004922327,0.02111737,0.029754518,0.027266419,0.030263303,0.022770183,0.26600966,0.013034052,0.0148931295,0.014900886,0.014568352,0.01946197,0.021304643,0.010476352,0.08244683,0.05353372,0.00757325,0.033484895,0.008503189,0.039844204,0.028181866,0.012981479,0.014639842
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.005170072,0.0052365907,0.00694314,0.017285356,0.008234528,0.037991844,0.07294639,0.022531565,0.05857518,0.0065963003,0.005045354,0.026125679,0.020520791,0.035622887,0.0324971,0.01749203,0.16650967,0.007892656,0.009556492,0.017534178,0.011642951,0.03663207,0.033574145,0.0035568238,0.07068761,0.048016,0.0072974605,0.12339532,0.0045464407,0.022555495,0.018934185,0.020668464,0.01818533
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
index 0759929f..87181f3b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0058703404,0.005491598,0.018997362,0.009865255,0.027035879,0.053124264,0.01636745,0.030759059,0.009746491,0.005269951,0.030089295,0.031095663,0.022614336,0.024479996,0.023500303,0.25250036,0.010035898,0.009619795,0.013611923,0.01075333,0.020063184,0.019546557,0.009071543,0.081420965,0.08086878,0.0082148705,0.07848736,0.0068253702,0.028736375,0.02457363,0.015248171,0.016114665
-0.008826475,0.00650084,0.03350764,0.011784676,0.035665788,0.057461075,0.017939368,0.026917169,0.009738677,0.006538719,0.024049368,0.028872915,0.029989129,0.031923886,0.02343755,0.1983992,0.013674415,0.0132244,0.015470912,0.01583799,0.026933711,0.020640487,0.011796321,0.09157396,0.064802006,0.007862755,0.07190949,0.0074055316,0.029487757,0.027891217,0.013090327,0.016846225
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0050350777,0.005766118,0.006457709,0.019448206,0.007035608,0.036554355,0.06908976,0.016720336,0.073623985,0.007380201,0.0050186585,0.027139133,0.019853646,0.03107149,0.027048368,0.018990275,0.18321688,0.0086597875,0.00932425,0.017520232,0.013038053,0.03883165,0.028723942,0.0043549896,0.060425516,0.060644113,0.0076746633,0.10625886,0.0049037784,0.016128076,0.024719888,0.024399208,0.014943133
+0.0042376784,0.005924756,0.0069834692,0.018832779,0.0078081987,0.04281036,0.07998875,0.020728823,0.057859186,0.006507186,0.005319783,0.035538692,0.020944735,0.03357084,0.033523228,0.017411817,0.14398022,0.0064492146,0.010395881,0.018382689,0.010860411,0.03802423,0.034201827,0.003840388,0.06658509,0.059049625,0.0067133494,0.10930205,0.00455577,0.024278667,0.023504738,0.026207874,0.015677737
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
index cecbf4eb..5f3dd3e1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0066114813,0.0059746164,0.015643341,0.009955672,0.025352968,0.06201794,0.017243594,0.029164374,0.009917572,0.0053090053,0.026186187,0.028928613,0.02072396,0.0292177,0.023453875,0.29139313,0.011147256,0.010212424,0.012978361,0.010813403,0.0158392,0.019197216,0.008891929,0.07571251,0.06993627,0.0069566243,0.05914112,0.0064757746,0.029883983,0.025710123,0.015420712,0.01458909
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.005213573,0.005612019,0.005980635,0.014423169,0.008148407,0.035215456,0.08324456,0.021167181,0.05202259,0.006690955,0.0051461984,0.03389285,0.022059022,0.028906152,0.04061394,0.018913923,0.16657288,0.005920637,0.008955144,0.018136969,0.010939499,0.027240243,0.03896024,0.0037672839,0.06677831,0.05034332,0.006739948,0.11629482,0.004037235,0.02923933,0.021340776,0.020757355,0.016725339
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
index 917cbb24..57dae23d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.006754691,0.006674459,0.016557487,0.0103658335,0.02666079,0.058823474,0.017140431,0.028328251,0.010410326,0.0055948445,0.026582949,0.03277368,0.021883957,0.027258111,0.022582687,0.2764412,0.0118644545,0.01098305,0.013153517,0.010666833,0.01935101,0.021501819,0.009382378,0.08465579,0.058386862,0.007471711,0.063156575,0.007812191,0.031189099,0.025576143,0.01507718,0.0149383
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.004618134,0.006132982,0.006672614,0.015618401,0.0074373814,0.03397381,0.07677221,0.019593101,0.05670967,0.0069386126,0.0054081944,0.030513486,0.021875117,0.02945475,0.029283365,0.017094098,0.20585725,0.007897137,0.009288681,0.016525654,0.011750388,0.03215573,0.031738274,0.0035157073,0.06421899,0.043070234,0.00695176,0.11497121,0.004500923,0.020835267,0.023099473,0.021389142,0.014138295
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index c39d520b..67d3a1aa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.017302375,0.01482844,0.014713044,0.015540072,0.033318657,0.022022286,0.029224692,0.023904966,0.012149988,0.009064443,0.015359024,0.049968407,0.0245197,0.020110352,0.019914918,0.056400876,0.0115936,0.012584733,0.04621322,0.035850577,0.08206279,0.09154765,0.017336203,0.032738093,0.013035035,0.021850908,0.124157034,0.013448811,0.060038418,0.022390943,0.013344151,0.023465507
-0.009482484,0.0060274336,0.046130657,0.009335471,0.06764614,0.08857251,0.05499592,0.017070245,0.009937557,0.0076791495,0.04376102,0.053096063,0.028434156,0.034147743,0.05993116,0.033161834,0.009860223,0.01984055,0.01256225,0.02168444,0.098040834,0.050862912,0.01778495,0.040789753,0.016320387,0.015451832,0.026600784,0.010373869,0.010914272,0.027834028,0.028337134,0.023332264
-0.010967831,0.011053853,0.02335534,0.016658744,0.05860075,0.048015736,0.028505739,0.027371904,0.016789401,0.013281509,0.045727536,0.03805783,0.022198936,0.065886594,0.057020094,0.06262433,0.016529106,0.017458213,0.021432064,0.024096718,0.019590493,0.026042046,0.012923263,0.09111785,0.032283496,0.028794264,0.050714724,0.009529016,0.025877263,0.026908096,0.021348508,0.029238785
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.016444147,0.017504703,0.010868917,0.024397992,0.02378631,0.05476737,0.018488651,0.01870659,0.03153483,0.01417574,0.018273251,0.04281981,0.028775947,0.030504966,0.017300766,0.020425139,0.019642664,0.01798995,0.01798995,0.06529241,0.026927112,0.06377991,0.09389312,0.034339547,0.03991245,0.021912975,0.010130945,0.04549251,0.012461275,0.0647843,0.021114701,0.01657312,0.03898788
+0.018649058,0.011534323,0.009525009,0.04894217,0.016076498,0.027988719,0.09323943,0.032794107,0.024086585,0.014523906,0.023028513,0.028638465,0.056332096,0.02776413,0.029872134,0.031101944,0.02357462,0.010298991,0.0192035,0.01811063,0.018979773,0.07148915,0.060672056,0.015280495,0.074919984,0.019054057,0.016522154,0.039403025,0.012767332,0.009750889,0.041294016,0.035528205,0.019054057
+0.009243722,0.015240321,0.012585413,0.012245942,0.011020156,0.044706393,0.04126598,0.022088086,0.022789238,0.016350478,0.017748276,0.052165084,0.025299398,0.037463184,0.07560385,0.03431107,0.060041644,0.014372992,0.013344871,0.014485721,0.019722441,0.03816479,0.025200764,0.012536347,0.06316938,0.05865078,0.021873433,0.038314164,0.036381554,0.024521017,0.025949989,0.020953465,0.062190026
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index c159c246..38cf6ada 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.017744087,0.006948684,0.00858048,0.016410597,0.055733632,0.034286235,0.036515333,0.028605288,0.013186285,0.011912816,0.023324167,0.059328143,0.045488466,0.033475645,0.042482805,0.052356903,0.014595887,0.031864792,0.020503258,0.021953864,0.05863695,0.04702418,0.0390225,0.034219332,0.057502806,0.014538983,0.023737786,0.014883765,0.034403622,0.0456665,0.022873037,0.032193214
-0.009073278,0.009965046,0.01715093,0.014555804,0.03801386,0.058020845,0.079927094,0.033448976,0.017974019,0.01410797,0.0699865,0.025571192,0.030388791,0.047355235,0.055798102,0.054081377,0.013673913,0.023626404,0.014785025,0.04074311,0.020208733,0.03655757,0.013098806,0.049145572,0.024543593,0.020248242,0.051604807,0.012114416,0.017285446,0.031925026,0.027879516,0.027140763
-0.014087621,0.012287442,0.03534718,0.01643796,0.031933516,0.06289021,0.026609892,0.083969444,0.02233677,0.018736115,0.04442197,0.026389923,0.010843629,0.01895697,0.027663156,0.03439375,0.017669836,0.021168692,0.018195163,0.06071764,0.021065582,0.020497372,0.022821855,0.041567888,0.020678319,0.021883357,0.15086576,0.016183114,0.022206273,0.032690797,0.008919726,0.015563149
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.018813502,0.0097793965,0.019639514,0.023574388,0.030255334,0.08317135,0.027166266,0.015935654,0.042812943,0.011081507,0.016963415,0.03456955,0.039287366,0.025708733,0.03456955,0.021297682,0.02741194,0.017777506,0.02591037,0.016123498,0.03373579,0.01538515,0.036263976,0.033180345,0.039828185,0.11821033,0.015935654,0.05716278,0.01116842,0.03830239,0.035701755,0.008766194,0.014509578
+0.013969102,0.009713985,0.01286894,0.013969102,0.015045315,0.06638303,0.04526929,0.020847589,0.039484575,0.014639493,0.013645508,0.046162147,0.04162265,0.021467391,0.05503346,0.02815642,0.026431143,0.014356339,0.012231754,0.020206176,0.039484575,0.038720876,0.0321792,0.0149866585,0.092163704,0.023808694,0.018834226,0.04049995,0.016524045,0.019508144,0.04734924,0.046797603,0.03763967
+0.017729616,0.014929833,0.020768544,0.022854377,0.015403757,0.04781904,0.025558222,0.03895261,0.058589116,0.019701704,0.013646972,0.08070991,0.04781904,0.021261055,0.05070465,0.018726205,0.047077674,0.024115471,0.022086421,0.015646331,0.028528795,0.032546125,0.01435788,0.021136845,0.0401891,0.06310296,0.017626034,0.050903104,0.02488098,0.021302622,0.018873077,0.016655432,0.025796438
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 37799880..ac0e5c1a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.018528365,0.011348305,0.026592543,0.016287476,0.10766971,0.032645565,0.019116517,0.02492655,0.019955833,0.011527015,0.02259085,0.027787218,0.03303038,0.031579427,0.01690325,0.07815983,0.013164263,0.017135935,0.020429073,0.034548,0.031517807,0.024956997,0.0139586525,0.11551274,0.068975806,0.023149172,0.035436425,0.009975799,0.044534132,0.014743276,0.015941283,0.017371824
-0.011191083,0.005224706,0.020623794,0.012053271,0.12365075,0.039831124,0.028327364,0.024147004,0.008681646,0.007843214,0.024527263,0.03959842,0.034102652,0.030959982,0.070454076,0.05114417,0.009098287,0.013604898,0.020304052,0.028584411,0.07921365,0.023622217,0.0178833,0.04620491,0.05705545,0.011323,0.041417815,0.0121953515,0.02816187,0.03845514,0.009098287,0.031416837
-0.009739021,0.0067989323,0.029710405,0.013416068,0.049846753,0.05431991,0.024651846,0.03325377,0.013156578,0.0067989323,0.031200917,0.028736616,0.030504202,0.02744746,0.040842965,0.032546956,0.011078949,0.017293995,0.013896115,0.041975174,0.037369967,0.020577284,0.021480734,0.09096873,0.022776976,0.013363764,0.080279626,0.018699272,0.08748377,0.02619055,0.0155629115,0.048030872
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.014180237,0.016257722,0.02075337,0.015035935,0.023724249,0.09210627,0.033686046,0.016321354,0.029331319,0.01988051,0.02091614,0.022904685,0.020075608,0.043507922,0.036637343,0.019610556,0.02790456,0.0151538635,0.01619434,0.025217365,0.032018133,0.042417135,0.02589112,0.017960545,0.12688175,0.066083014,0.015881114,0.024741694,0.022048742,0.031521738,0.026837192,0.013960393,0.02435811
+0.010987974,0.009621391,0.008102024,0.010004664,0.01264708,0.09947699,0.039184675,0.038127735,0.026918061,0.008229612,0.013890101,0.029495217,0.02808619,0.054296385,0.039108217,0.046989605,0.02911958,0.009696852,0.010649909,0.018258192,0.032708064,0.031547327,0.04019234,0.021597609,0.04019234,0.045543887,0.023859728,0.069717936,0.013890101,0.030031558,0.07900077,0.007976414,0.02085151
+0.016185524,0.016569352,0.011432509,0.008799916,0.01716223,0.0705819,0.029802466,0.025813933,0.0328878,0.008075258,0.016122423,0.023230024,0.018376462,0.034974713,0.0608454,0.10472156,0.020986577,0.009367462,0.015996957,0.017637983,0.0325523,0.024179006,0.022515312,0.025738416,0.0861416,0.026348786,0.020620897,0.04803876,0.019523472,0.043314718,0.05390604,0.013629578,0.02392066
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 8b6aa288..58c7fbc8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.01669491,0.010389542,0.026836602,0.01910696,0.12638526,0.04060527,0.023563892,0.02014098,0.014828836,0.010469111,0.022223387,0.025744986,0.03159677,0.02423024,0.015737114,0.06580435,0.010926759,0.014931464,0.018498765,0.041775487,0.035436004,0.022769153,0.012573552,0.13099544,0.051511902,0.019180572,0.056185905,0.010143914,0.028734636,0.014651826,0.018792372,0.018534083
-0.016519701,0.007162946,0.045259815,0.022103863,0.038290083,0.053124834,0.06819331,0.036727816,0.018657679,0.011355199,0.043989643,0.13988414,0.022740448,0.019094305,0.023814064,0.021139339,0.018106835,0.014832587,0.014998272,0.04460881,0.027002994,0.03233611,0.028538257,0.04015461,0.018103521,0.01528009,0.052740376,0.013800199,0.026066186,0.028910367,0.012520822,0.023942681
-0.011194919,0.010561037,0.03043257,0.029184885,0.16755833,0.032587912,0.021776065,0.02243084,0.018040746,0.011284099,0.022678632,0.04638041,0.06720232,0.021271471,0.02572572,0.031782538,0.011916202,0.008333079,0.020233061,0.030141652,0.019565124,0.017642204,0.020604435,0.042393837,0.04602931,0.016611801,0.08245273,0.01454134,0.023293085,0.027460624,0.024097573,0.024591517
-0.01608547,0.010368459,0.031037511,0.011805071,0.03809329,0.04238475,0.043802626,0.035902377,0.012958213,0.009891841,0.02051157,0.070250556,0.014836899,0.022624696,0.038268067,0.056611795,0.014446229,0.016333792,0.02795824,0.023835989,0.049290363,0.06045451,0.016494084,0.02983232,0.015678363,0.023880403,0.1506435,0.015196918,0.023785125,0.031565513,0.008330273,0.016841229
-0.0137396315,0.010914814,0.02667995,0.0098691415,0.03460263,0.034310307,0.03730518,0.018527672,0.008636964,0.009481218,0.04571112,0.032784116,0.023355061,0.099395365,0.03462587,0.048355766,0.008171825,0.0144227585,0.03637852,0.022415577,0.082270995,0.057494033,0.013951741,0.05162547,0.023641635,0.018418867,0.04009568,0.008299499,0.050505035,0.025156023,0.022480657,0.036376856
-0.011915657,0.0036924249,0.04062848,0.012023047,0.110737495,0.049395066,0.031113395,0.026186328,0.011969596,0.006231006,0.030099835,0.051332965,0.03739974,0.026679186,0.053064372,0.040329844,0.0105973575,0.011433326,0.018081808,0.02057271,0.08053413,0.015506836,0.046564084,0.03711733,0.023726063,0.011824371,0.031816028,0.013342459,0.01617631,0.02257506,0.055539753,0.041793946
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.013940462,0.019203799,0.019620843,0.012943281,0.026376959,0.08539569,0.034201056,0.016171219,0.036371298,0.017553693,0.018832363,0.023096437,0.022320347,0.02289433,0.035982683,0.026197264,0.0309963,0.018396111,0.017349185,0.024706373,0.039096992,0.03601784,0.021845924,0.017417088,0.139698,0.046610545,0.013511559,0.03353955,0.018649349,0.01886918,0.03457043,0.012447432,0.04517649
+0.018725712,0.011787114,0.01107297,0.033447456,0.019894524,0.08055031,0.03819819,0.035431203,0.03186907,0.018291932,0.019207258,0.02245556,0.13702068,0.054503065,0.023191012,0.02562002,0.01916978,0.013834468,0.016687555,0.02840035,0.030843345,0.021892576,0.026087089,0.01579946,0.04885623,0.023418596,0.015923375,0.042116705,0.01549387,0.026875785,0.03695072,0.019433666,0.016950345
+0.020445274,0.016654389,0.013200464,0.028482636,0.033642784,0.083429486,0.028041055,0.019777574,0.025619086,0.021657871,0.030393722,0.014611581,0.022830635,0.123300776,0.01745365,0.037973646,0.017728506,0.027734673,0.01872503,0.024196517,0.019738983,0.021280492,0.021657871,0.019509016,0.022542626,0.05344693,0.025725637,0.07596341,0.020049827,0.038271476,0.015432905,0.013780033,0.026701493
+0.016903484,0.014345801,0.01919161,0.02922801,0.017169677,0.073139146,0.028308017,0.035229895,0.035195507,0.019079488,0.01275942,0.023376677,0.053509746,0.026592921,0.03432988,0.015330795,0.024594432,0.016969644,0.017645637,0.023376677,0.03652613,0.047130045,0.024142297,0.023931038,0.06951778,0.054352403,0.024307897,0.06207239,0.022835156,0.02310433,0.034129318,0.010332893,0.03134177
+0.012535897,0.012634218,0.007906292,0.027009122,0.012584961,0.057965416,0.026101558,0.024281856,0.028694961,0.0115486095,0.016803129,0.074138895,0.03044326,0.030756325,0.042741016,0.019302549,0.026152587,0.010392624,0.011822477,0.047309995,0.01809768,0.06803367,0.085000664,0.020993682,0.06697891,0.02416358,0.010151879,0.041750923,0.010474134,0.058648694,0.01960652,0.01673762,0.028236296
+0.020117404,0.02431366,0.01404421,0.041117303,0.020796577,0.08436723,0.027712893,0.025282206,0.0328136,0.018245693,0.022224486,0.02491456,0.034054168,0.03721901,0.018496858,0.032352228,0.03961943,0.016101766,0.03537617,0.02481743,0.02498766,0.016743189,0.031564236,0.05642065,0.06519394,0.03134155,0.0072011556,0.046229426,0.02007815,0.036965452,0.027564444,0.01680872,0.02491456
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 0150b2b0..daeca86b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012020789,0.0114255985,0.013781902,0.011336684,0.030397858,0.024943626,0.019588038,0.027308403,0.0142750405,0.007791581,0.024653023,0.027215222,0.019322054,0.05623878,0.032358356,0.05221594,0.013253926,0.02056822,0.01688594,0.029197672,0.020131039,0.03461389,0.02210962,0.06372689,0.028759044,0.0143309105,0.03210654,0.013048443,0.2367741,0.017320175,0.009849482,0.04245126
-0.00968035,0.0060578086,0.02164524,0.011229403,0.049160875,0.04091531,0.09700724,0.020897496,0.006810977,0.012527305,0.029767573,0.033377804,0.042296633,0.04004565,0.060468912,0.05906815,0.01140624,0.013975218,0.029484136,0.020215027,0.053155582,0.022861931,0.017122796,0.05636322,0.05212746,0.013077319,0.04449998,0.0103855,0.009987637,0.0559246,0.02224527,0.026211333
-0.010586183,0.007959717,0.014697459,0.012869525,0.055902667,0.026143547,0.028727092,0.017867567,0.012769373,0.009161578,0.022964824,0.09621457,0.035395365,0.022852965,0.108176954,0.040501803,0.013699538,0.028454864,0.0163962,0.01607907,0.098496236,0.06434358,0.015342754,0.026869912,0.020049846,0.020049846,0.04634494,0.011268932,0.009676554,0.020686297,0.038122308,0.031327948
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.015050883,0.01172164,0.012093725,0.022309033,0.010756382,0.094016306,0.022882726,0.03638838,0.029932264,0.010672675,0.015347736,0.027256964,0.022027602,0.037543472,0.08695087,0.023957483,0.07613672,0.010714447,0.015407805,0.012674115,0.016659811,0.021770975,0.033035073,0.019553527,0.0648693,0.03820925,0.009755615,0.032081287,0.012924089,0.10570537,0.0215594,0.0116304215,0.018404752
+0.01221852,0.020105632,0.011523162,0.04298132,0.016506167,0.056940995,0.022168143,0.083172925,0.026531791,0.014566641,0.0132113695,0.06253745,0.035355467,0.036727987,0.024466112,0.017230876,0.11821256,0.01090991,0.019639885,0.026805244,0.014284897,0.02955142,0.036549088,0.017708533,0.03152267,0.038228378,0.0096279625,0.02415747,0.01929767,0.024778698,0.03875462,0.014008602,0.029717823
+0.0134996595,0.008414936,0.008818777,0.03387192,0.012195926,0.014370312,0.033657596,0.01964189,0.031905293,0.0134996595,0.019757316,0.038307022,0.07952743,0.028943907,0.039215446,0.05639339,0.046569496,0.01007136,0.015844474,0.011457012,0.019000426,0.048709184,0.071847044,0.01907479,0.08498784,0.01888942,0.009314485,0.045937136,0.01634743,0.01713196,0.024065744,0.06390209,0.024829673
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 2efab20a..eb9b1110 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010618431,0.00532009,0.031241473,0.010025754,0.12621595,0.03801969,0.03556635,0.03164376,0.010138364,0.0072703697,0.0275396,0.05265158,0.03624807,0.020274848,0.062158115,0.04025004,0.010453169,0.011500197,0.021514071,0.021920329,0.09396764,0.020422643,0.019221632,0.03632739,0.034134757,0.013216465,0.06821319,0.010370556,0.018837776,0.03664193,0.012034463,0.026041366
-0.011248999,0.007763945,0.020370394,0.013426363,0.052157905,0.02627952,0.029951978,0.030582616,0.009331442,0.007465146,0.03328639,0.038111825,0.01810378,0.015171406,0.036365446,0.036902063,0.011336534,0.022397265,0.027198253,0.029846832,0.23694438,0.030660179,0.018613372,0.04210234,0.021134239,0.013412438,0.06735742,0.007580391,0.008907657,0.029101651,0.021942623,0.024945194
-0.014820769,0.016931016,0.03134662,0.017758721,0.07756528,0.030280596,0.027700873,0.026437817,0.011194134,0.014254735,0.025730982,0.038801637,0.034535516,0.03318171,0.038952313,0.064535856,0.014877869,0.03107635,0.022293469,0.017415768,0.03861853,0.074516445,0.018766109,0.05231379,0.03215885,0.030682446,0.0475016,0.010968913,0.011374928,0.018307073,0.054861203,0.020238126
-0.012825078,0.0054109595,0.04036408,0.014380072,0.06844456,0.030411042,0.0412312,0.025175504,0.0114584975,0.008511076,0.035235465,0.17975047,0.029292798,0.02909468,0.031851053,0.026281644,0.013975125,0.009772897,0.020356039,0.021315373,0.06219029,0.024319902,0.021852175,0.06849053,0.020569606,0.013285601,0.041427664,0.007992975,0.013719467,0.018521097,0.022934454,0.029558716
-0.014430656,0.004168151,0.06305564,0.021247871,0.09074368,0.07415783,0.03126866,0.020979187,0.017970206,0.010274172,0.0257828,0.059320774,0.028706122,0.025938272,0.037631854,0.03580352,0.011777239,0.01175139,0.014962524,0.015961098,0.07775944,0.011435114,0.018415455,0.09703935,0.028552026,0.013933843,0.04162163,0.008481792,0.013676437,0.024435634,0.02378487,0.024932796
-0.008010261,0.0074686627,0.023822267,0.013119117,0.074442916,0.04628194,0.04254105,0.058279544,0.0090823565,0.005861148,0.038926944,0.022764165,0.046605628,0.03756041,0.03390609,0.0453148,0.0128496755,0.017593464,0.024916999,0.030104289,0.038540415,0.029914714,0.023720885,0.05801514,0.032298174,0.012752013,0.07088667,0.011338414,0.037955638,0.02446528,0.02521379,0.035447173
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.013031122,0.011321634,0.010389316,0.012531906,0.014766195,0.104928374,0.035284165,0.027871214,0.020864384,0.010389316,0.017811432,0.024512332,0.020182995,0.08139971,0.029885026,0.043062516,0.016154267,0.010552924,0.013817477,0.024947023,0.036833324,0.051740803,0.04085048,0.033833183,0.039362326,0.033146404,0.022692353,0.050938636,0.017263431,0.053175114,0.046470832,0.012385906,0.017603923
+0.013867004,0.0117917815,0.01430719,0.011837933,0.013572258,0.038663756,0.02143049,0.018033074,0.02613567,0.012749956,0.014475838,0.04330135,0.019403422,0.029014427,0.033526108,0.025405882,0.021388676,0.013921278,0.016086007,0.024153845,0.020156674,0.32372767,0.016957058,0.01336184,0.04830615,0.016403275,0.014877124,0.025381085,0.016825097,0.016661588,0.018156769,0.017359182,0.028760536
+0.02343217,0.018681755,0.01668093,0.015979351,0.02779923,0.038860146,0.042846564,0.01560919,0.039012242,0.023295274,0.038633116,0.022228505,0.060541473,0.026786525,0.037848905,0.03317399,0.03386134,0.021544605,0.028099464,0.034529194,0.031596933,0.040725086,0.03131662,0.013299196,0.04855152,0.04276296,0.021798568,0.028569853,0.029247366,0.014549365,0.03781196,0.029795108,0.03053148
+0.015153667,0.017171355,0.015880907,0.04367764,0.013012349,0.09067712,0.021834187,0.05186848,0.02257132,0.019533848,0.019572038,0.028627018,0.09138831,0.06738551,0.024596829,0.029257633,0.03824542,0.014573139,0.02051132,0.022926766,0.016708186,0.023174377,0.017306032,0.017855383,0.055864554,0.033130992,0.013425405,0.041758977,0.02423915,0.028627018,0.018748865,0.0150945885,0.025601646
+0.020325784,0.03276724,0.017183015,0.036877196,0.0409791,0.060563177,0.054928415,0.013097705,0.02611157,0.020930074,0.021573389,0.027016183,0.022258202,0.041219912,0.019700425,0.035533786,0.029733112,0.020991484,0.033840418,0.018255536,0.018219914,0.018835029,0.020807795,0.015075367,0.14931092,0.027404804,0.014328933,0.0464354,0.020167608,0.028804252,0.016817788,0.013252097,0.01665435
+0.014279534,0.026315438,0.013732494,0.018192599,0.034430902,0.022774111,0.018624028,0.013572506,0.040204696,0.016825402,0.018770097,0.065004244,0.019747883,0.029950626,0.025983466,0.04906778,0.041724697,0.02481782,0.01630774,0.040957645,0.027930688,0.04398407,0.07167262,0.020374749,0.04398407,0.026008852,0.0155002875,0.031533923,0.028012637,0.030362919,0.0260597,0.021021511,0.062270254
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 1259f535..74d4cc81 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.014694284,0.00484566,0.038039703,0.01950482,0.078818366,0.048180703,0.04508512,0.019164959,0.01751819,0.007865283,0.041291706,0.043697998,0.030476343,0.028685847,0.052504353,0.059263274,0.009195448,0.0085710995,0.020361185,0.016456816,0.048748642,0.016618315,0.017794061,0.0893129,0.04169692,0.01588831,0.03295278,0.010792644,0.014636996,0.03708601,0.046154283,0.024097003
-0.011956437,0.007511431,0.03889856,0.009203197,0.098943695,0.040606413,0.027962914,0.014880026,0.011863391,0.0074529764,0.039050806,0.030681275,0.035141923,0.045833707,0.02887874,0.050142225,0.0061787344,0.0146493325,0.011364432,0.027826708,0.039821014,0.049172387,0.013131577,0.13418755,0.050733287,0.021128137,0.03740838,0.008713412,0.023766605,0.01837438,0.023593169,0.020943254
-0.013887067,0.0071764276,0.0151924975,0.018254206,0.07991457,0.04781246,0.023301933,0.017418284,0.01525196,0.008724318,0.026781103,0.08342324,0.027746145,0.042691726,0.06127263,0.03917613,0.018254206,0.018218588,0.014782706,0.020644369,0.04244231,0.036910545,0.021932937,0.05985325,0.017282734,0.021299692,0.021175254,0.008931208,0.025170805,0.02839026,0.069430985,0.04725543
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.01910463,0.02436371,0.021312755,0.02668002,0.029352391,0.070567615,0.022909949,0.013841413,0.03352967,0.023545036,0.025808707,0.034057684,0.031127442,0.056261405,0.02002148,0.035972085,0.04840573,0.025758348,0.038366944,0.02652415,0.030842822,0.029618738,0.034543376,0.017982226,0.04520731,0.032403,0.018123262,0.04100125,0.022118514,0.023962572,0.035972085,0.020257486,0.020456282
+0.013497088,0.009385752,0.010718865,0.022892157,0.017195737,0.047109466,0.042228654,0.023229958,0.034365196,0.015841454,0.01488167,0.039592747,0.037339516,0.025289863,0.04841539,0.025228195,0.036049604,0.009608328,0.015474488,0.012289238,0.024380473,0.03366762,0.052554406,0.010148417,0.2136187,0.02663978,0.014536936,0.043314595,0.011321378,0.008347861,0.026646286,0.017845849,0.016344316
+0.009544797,0.0067682746,0.008827494,0.02225743,0.00889673,0.03658895,0.043281052,0.023995617,0.03151086,0.010858004,0.017283395,0.049428515,0.06729738,0.028021205,0.06756077,0.07025208,0.047720984,0.008758798,0.011423627,0.010320387,0.020766476,0.04203145,0.031113345,0.019470233,0.094904505,0.027490968,0.008489318,0.034540378,0.013618987,0.026697252,0.028558182,0.05139752,0.02032508
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index a4ae747f..d407a31f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.021899408,0.01375122,0.06875178,0.01996887,0.025484513,0.04252252,0.03902086,0.08623405,0.015073292,0.016425943,0.052713815,0.047996476,0.014524183,0.009053567,0.018360289,0.018869255,0.023249278,0.015858501,0.025478292,0.022211714,0.030988853,0.030687701,0.028464755,0.0355289,0.016980492,0.014411155,0.1662205,0.013224419,0.022462567,0.021867353,0.007360498,0.014354971
-0.02203823,0.012158787,0.019582076,0.019851638,0.061747413,0.028869862,0.03577158,0.022407154,0.016234111,0.010899066,0.018905604,0.054705177,0.01843162,0.026778596,0.03470488,0.05385705,0.018467655,0.027081026,0.022649152,0.018252501,0.13860744,0.05000458,0.029917473,0.035182618,0.025070326,0.025365846,0.045087226,0.0137239965,0.015131854,0.035354827,0.016880805,0.02627989
-0.01170118,0.0052330913,0.030678341,0.019595789,0.058046952,0.14593004,0.036824893,0.02589769,0.017668778,0.0141143175,0.025014492,0.04591892,0.020697279,0.014505581,0.07934078,0.029359365,0.017807355,0.026954826,0.011655562,0.028372828,0.08185932,0.015562217,0.02015868,0.029821709,0.016991898,0.014791678,0.0247291,0.02443503,0.010245896,0.056261037,0.010949386,0.028875977
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.015481911,0.021790396,0.019840382,0.03492312,0.020390378,0.059813716,0.027535558,0.028744513,0.04444978,0.020833192,0.014430728,0.048249777,0.029595483,0.019685984,0.057973444,0.018601947,0.035025585,0.018277813,0.016224904,0.025851479,0.029047865,0.061471812,0.022220174,0.02191845,0.041189697,0.06392057,0.01972447,0.050961923,0.024214173,0.033323877,0.018894885,0.01597336,0.019418672
+0.013983444,0.010351101,0.01371298,0.024782542,0.011960674,0.073268,0.036626223,0.02346937,0.02346364,0.022808464,0.020010805,0.03494899,0.019855078,0.04996443,0.029229501,0.023452187,0.026665796,0.013343096,0.015357809,0.02130139,0.01657335,0.20548517,0.04207427,0.009028391,0.04006921,0.015752748,0.028000148,0.020666188,0.018927356,0.024217322,0.020828273,0.027378248,0.02244389
+0.012367314,0.014458855,0.009707215,0.026529063,0.018349245,0.059930302,0.13718182,0.03231419,0.018601837,0.015572799,0.02426137,0.023676222,0.026946833,0.040948816,0.021202514,0.05785998,0.014515445,0.011172936,0.013424592,0.019570857,0.021961171,0.045237765,0.03869391,0.013635999,0.041593663,0.022746976,0.036456212,0.028863892,0.020113382,0.012033726,0.06530834,0.032015786,0.022746976
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 0ed4b179..887fec91 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.030829223,0.02203265,0.07899471,0.067452386,0.13132587,0.15056586,0.08587365,0.12193194,0.0947754,0.11232897,0.057428077,0.046461277
-0.014666316,0.022189427,0.059498083,0.06742016,0.039249096,0.14300752,0.37238505,0.06761797,0.044562005,0.027187055,0.042771243,0.09944612
-0.025165679,0.016633498,0.03282223,0.10289244,0.07257083,0.060428306,0.11959067,0.06701881,0.24063799,0.099726774,0.026998831,0.13551399
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.021211134,0.028320579,0.023478584,0.058909986,0.13095303,0.09981858,0.10119266,0.05229339,0.18758154,0.118306205,0.10178733,0.036364343,0.039782714
+0.016987843,0.024144594,0.015347238,0.05002766,0.08610305,0.046630908,0.095680416,0.25555444,0.118380524,0.08460274,0.08410848,0.03582306,0.08660904
+0.03188173,0.04349213,0.038607225,0.03515228,0.05744916,0.060088728,0.064544536,0.08375145,0.061064795,0.27327284,0.060530446,0.032006513,0.15815818
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index c4f4b91a..3b040be2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.042400617,0.021825965,0.08374923,0.15631144,0.031142365,0.090377815,0.19151676,0.0473013,0.07689015,0.11127522,0.05892494,0.08828422
-0.015833566,0.013437772,0.041961044,0.10164713,0.037466995,0.06844288,0.11837423,0.034047525,0.14617266,0.08426843,0.057690993,0.28065673
-0.03181457,0.05401281,0.1453952,0.045572266,0.11890011,0.059962142,0.07022276,0.047480304,0.17675562,0.0464711,0.050443903,0.15296924
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.036800325,0.042522695,0.031600185,0.123525076,0.04186344,0.10062151,0.12113588,0.06976658,0.029225392,0.18399003,0.06321378,0.057444494,0.098290615
+0.023137681,0.02156669,0.017740281,0.07512827,0.09349864,0.04173359,0.115230136,0.07210918,0.043226883,0.15873621,0.05558578,0.03740975,0.24489695
+0.03727073,0.06427239,0.031631254,0.054867815,0.053701587,0.09143873,0.115476556,0.026950166,0.027481709,0.1778087,0.052355234,0.033020034,0.23372512
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 8d04302d..a9e05bb0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.022884028,0.023426706,0.13376229,0.06100784,0.048445486,0.04047772,0.04047772,0.064007714,0.46505603,0.03600163,0.033296064,0.031156814
-0.01766297,0.016592825,0.08159303,0.11076467,0.043122653,0.16369955,0.1496335,0.073605254,0.1455974,0.11250896,0.039494358,0.045724865
-0.024478791,0.02061321,0.15053765,0.09872501,0.050129212,0.07090094,0.15837957,0.047740296,0.19254059,0.044152558,0.0707626,0.07103955
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.020409178,0.02520197,0.032233812,0.12212525,0.036383294,0.0602501,0.0626177,0.04644414,0.08024828,0.33719864,0.04397243,0.08328219,0.04963303
+0.018363692,0.022676133,0.023579445,0.082380764,0.027245997,0.11293152,0.13254708,0.07497192,0.07754087,0.18692133,0.097163096,0.057175048,0.08650309
+0.015567869,0.014854965,0.02337014,0.12197339,0.0134203425,0.0694305,0.18382038,0.11868336,0.12584524,0.1414916,0.04014407,0.09117559,0.04022255
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 26f8fed2..e5aef4b6 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.02404165,0.024330985,0.16612503,0.0601377,0.06399806,0.049517293,0.04449823,0.070558906,0.37823856,0.040439203,0.046034217,0.03208023
-0.027730746,0.02667491,0.066027075,0.05778757,0.0820242,0.27731502,0.11232613,0.03457894,0.06187734,0.090703204,0.050964195,0.11199069
-0.022751989,0.019618228,0.2622568,0.091747545,0.100092806,0.15698826,0.09324552,0.050576176,0.095965005,0.04294408,0.031872693,0.031940855
-0.032131083,0.03407833,0.06330054,0.047110375,0.13916276,0.113678426,0.113963254,0.09667804,0.104930215,0.11553822,0.07334705,0.066081785
-0.020889135,0.015167588,0.021204498,0.12352331,0.04426551,0.2877678,0.10738013,0.10587679,0.09809948,0.04156298,0.02445021,0.10981257
-0.014412873,0.019248405,0.102152325,0.07133191,0.055629157,0.1822007,0.25063998,0.08880837,0.04430273,0.06214931,0.067607224,0.041517083
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.021790942,0.02875576,0.023932666,0.11462324,0.033882566,0.074223466,0.087884694,0.03663579,0.07831908,0.31525108,0.041190725,0.09521178,0.048298124
+0.019185683,0.01774385,0.017605767,0.11346654,0.04976387,0.07859597,0.28414276,0.10032964,0.06620061,0.09333503,0.042648513,0.07882658,0.038155284
+0.049818195,0.04616438,0.07234362,0.14387244,0.037678268,0.15077703,0.11031084,0.13836077,0.054394927,0.069640145,0.04336742,0.054607823,0.028664112
+0.022936853,0.028434087,0.022405522,0.04789797,0.036652986,0.1330277,0.16882095,0.031969305,0.054701194,0.24659441,0.06426521,0.067086786,0.07520705
+0.01597938,0.02019983,0.017826287,0.037008327,0.15703289,0.073062755,0.15339524,0.051783957,0.12617949,0.14410149,0.08930007,0.02306893,0.09106136
+0.028527373,0.050263055,0.0540826,0.15242085,0.032834806,0.13770077,0.070279166,0.104757205,0.073994376,0.10133605,0.056182116,0.12104672,0.016574996
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 7d06ac54..f5f64a91 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.036718734,0.027393846,0.054266773,0.24703673,0.033826884,0.041445564,0.09358162,0.030679652,0.26920536,0.06927283,0.038181443,0.058390565
-0.0143717965,0.020748112,0.026745372,0.053462926,0.02802891,0.068698056,0.29321107,0.06816344,0.0434969,0.24498703,0.029374044,0.10871233
-0.021368038,0.02411876,0.05199026,0.04653555,0.04023385,0.16546689,0.30434018,0.08805062,0.03627733,0.0858431,0.037355866,0.0984196
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.021908142,0.024061386,0.026947476,0.05529263,0.035699606,0.055726297,0.103704534,0.12558115,0.045839198,0.3591476,0.039055485,0.076020196,0.031016354
+0.031836722,0.03900717,0.033890016,0.036358677,0.06507003,0.07140441,0.04242466,0.11265075,0.088420585,0.23159625,0.14606592,0.054314982,0.04695982
+0.0118632335,0.019789722,0.010469266,0.04605744,0.101288736,0.03852003,0.21526752,0.2169559,0.096273586,0.13005732,0.03852003,0.046919998,0.02801721
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index dc13ad4c..849903e2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.0132217435,0.015582825,0.09537193,0.05511677,0.04859985,0.201613,0.21431485,0.1044041,0.09673497,0.06951364,0.040298074,0.045228217
-0.02334472,0.015194411,0.06716989,0.035730656,0.055619556,0.11289076,0.21729738,0.16236128,0.06862233,0.08559951,0.039524514,0.11664504
-0.062737055,0.03526971,0.066480264,0.083916724,0.07574927,0.04294516,0.17295846,0.14586803,0.07443847,0.07331404,0.10477662,0.061546147
-0.02023797,0.019773984,0.026585264,0.04649983,0.08008181,0.31102097,0.18837911,0.0946825,0.04079707,0.068564005,0.03542132,0.067956254
-0.018193766,0.02195123,0.10250693,0.042961787,0.05208291,0.19067423,0.23882356,0.05122539,0.06386787,0.06638725,0.051385093,0.09994003
-0.019719949,0.01622515,0.09795474,0.22302167,0.03292809,0.054548316,0.14749242,0.117770776,0.15120222,0.051777475,0.046444554,0.040914692
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.019913744,0.02274221,0.032832213,0.12439244,0.026381426,0.079455666,0.09334777,0.11459569,0.13293332,0.11731324,0.09190055,0.060269855,0.08392191
+0.017521279,0.033905335,0.016459718,0.045982454,0.0826155,0.044134606,0.0711495,0.08046563,0.19895993,0.1259736,0.060649738,0.052003276,0.17017946
+0.025699943,0.03512766,0.01985938,0.048484996,0.0704764,0.038580187,0.07725216,0.15866601,0.17562728,0.14334281,0.07823914,0.10631546,0.0223285
+0.021763925,0.03864713,0.032927766,0.06163743,0.033446305,0.11904154,0.23581937,0.08023351,0.070469536,0.1138123,0.043452136,0.103829876,0.0449191
+0.035014756,0.06755865,0.06371388,0.10998025,0.038008157,0.13370198,0.07059374,0.07311969,0.049186207,0.12958841,0.064717226,0.113471396,0.05134574
+0.044106077,0.03992442,0.04333755,0.04750411,0.14632298,0.071626194,0.04958979,0.079012446,0.19158807,0.16007784,0.05232613,0.05032155,0.024262724
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 01f1cf46..31f376e9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.01664073,0.014233571,0.055206023,0.044014078,0.064416446,0.14152527,0.23886913,0.112393886,0.10025849,0.08669238,0.050659895,0.07509014
-0.01838463,0.01896822,0.10095148,0.10294256,0.028586118,0.078009255,0.123206116,0.07873634,0.2921128,0.026854172,0.048531745,0.08271667
-0.029839043,0.023057824,0.048528235,0.08008771,0.04224449,0.15695731,0.255764,0.12810484,0.07081965,0.06270827,0.027275093,0.07461351
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.044508155,0.08047588,0.044508155,0.05337343,0.06501244,0.10866292,0.057261333,0.064758986,0.11499529,0.079421885,0.09061411,0.14939743,0.047009982
+0.012853926,0.02102762,0.016504768,0.088702664,0.063192055,0.041492984,0.12681279,0.11774168,0.07601974,0.21071875,0.06582174,0.07075447,0.088356845
+0.011193676,0.014945519,0.013086734,0.07242391,0.087019324,0.05896654,0.2649157,0.11393881,0.05971991,0.20471099,0.019190425,0.03407726,0.045811202
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index cb64c3b7..be998e4d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.039938595,0.043183915,0.118075885,0.05954648,0.12374247,0.15490499,0.08697871,0.045789823,0.083974,0.07239011,0.036364503,0.13511053
-0.046179265,0.02545285,0.071640715,0.04229379,0.04180105,0.1114293,0.23961064,0.08378333,0.059353285,0.15230578,0.036175795,0.08997421
-0.02781237,0.0368454,0.15094069,0.03213713,0.0700912,0.1276023,0.24983314,0.03163889,0.04967784,0.062363237,0.035433877,0.125624
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.03258783,0.04750771,0.034689564,0.0454208,0.056416802,0.106019855,0.053991236,0.023748815,0.043766156,0.28041837,0.04988499,0.026701512,0.1988464
+0.02455211,0.0374375,0.024744675,0.051875398,0.11622004,0.027604679,0.038625892,0.22227663,0.1474909,0.11622004,0.06943205,0.04445813,0.079061925
+0.027025884,0.03122835,0.031596463,0.08698541,0.055997707,0.056547236,0.067414336,0.21009848,0.09963192,0.10380335,0.10002187,0.03098533,0.09866369
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 083b09c1..a629f483 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.023895286,0.01832115,0.059314918,0.18721822,0.06704871,0.028936027,0.040173694,0.14868146,0.0849647,0.20481735,0.042764675,0.09386384
-0.013502704,0.015786262,0.1363725,0.077173345,0.068605885,0.14125212,0.054298226,0.09013671,0.08873927,0.15881403,0.12416856,0.031150412
-0.025655951,0.017495839,0.15114097,0.31996542,0.08954807,0.031434327,0.07864097,0.07727069,0.04591569,0.03618069,0.084534384,0.04221698
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.017289048,0.016052337,0.023816664,0.022026813,0.09165594,0.034517944,0.036103908,0.02730594,0.11362295,0.14877227,0.39813527,0.0136234425,0.05707744
+0.019879099,0.019879099,0.028475493,0.0552105,0.080095805,0.04910524,0.08279988,0.06850958,0.17240135,0.12712078,0.11001386,0.107885994,0.07862336
+0.028068636,0.036751684,0.030827366,0.19407536,0.14006053,0.1583997,0.07188607,0.0728045,0.052336797,0.053473387,0.031681933,0.09230354,0.037330437
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index d7a54c48..0cfd51a0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.028366823,0.027280107,0.08306996,0.08640018,0.054999802,0.1901026,0.15698618,0.10560158,0.07615487,0.03628172,0.08560232,0.06915384
-0.01920806,0.017903887,0.17525947,0.30758995,0.064238675,0.062589936,0.11315601,0.053346824,0.053817756,0.045363374,0.06276591,0.02476017
-0.039392862,0.044812668,0.17967735,0.1542877,0.18250686,0.05657941,0.05983002,0.052983012,0.036575034,0.07682679,0.089294866,0.027233396
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.028672202,0.035683133,0.04295809,0.05304616,0.20614591,0.06991537,0.099297024,0.038885273,0.108631276,0.13206205,0.030881178,0.055105437,0.09871691
+0.01782257,0.03189652,0.03448836,0.090687424,0.24411833,0.09002563,0.051571313,0.08114297,0.06220688,0.11419271,0.09202563,0.054790262,0.03503147
+0.038860124,0.03483399,0.03840739,0.11969761,0.24464603,0.101586044,0.07508794,0.03708059,0.077282585,0.058908433,0.050584264,0.046418726,0.07660632
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 3cfad144..d1779af8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.042999305,0.035647675,0.075098544,0.21165,0.11240646,0.044537887,0.04333655,0.11529708,0.13665152,0.056853674,0.059698623,0.06582273
-0.0210577,0.013177572,0.25059265,0.12749122,0.0773651,0.04726925,0.043631613,0.1461694,0.05774607,0.052835755,0.1179101,0.04475363
-0.017115196,0.01850594,0.13727663,0.41466334,0.06151393,0.040539283,0.028108178,0.060975652,0.062421646,0.09324955,0.050047114,0.0155835645
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.020406444,0.035674796,0.030395307,0.051704157,0.2998705,0.12500462,0.033382714,0.056234058,0.076600574,0.12598504,0.067698866,0.03051427,0.046528704
+0.015827617,0.024323527,0.020968191,0.17016296,0.15676147,0.063647375,0.036371585,0.03074782,0.1898305,0.057951596,0.07677339,0.11029547,0.0463386
+0.017605642,0.022783393,0.029254457,0.20951222,0.16063859,0.06966567,0.046384685,0.0307784,0.1224463,0.08047888,0.09829233,0.08552304,0.026636485
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index d6845334..02fe2c42 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.041693583,0.031973436,0.10998214,0.16479066,0.1252256,0.043118283,0.03657029,0.09519903,0.14785722,0.09291999,0.05095015,0.059719656
-0.022328043,0.022683812,0.04283925,0.41126367,0.050769344,0.05469367,0.059796266,0.040055063,0.084618196,0.088319905,0.086037345,0.03659542
-0.038595863,0.036979344,0.16525516,0.08796479,0.14711224,0.04366683,0.09374597,0.060864523,0.09171209,0.040690783,0.15224949,0.041162893
-0.028332265,0.021141028,0.07007169,0.28262228,0.09273151,0.036750935,0.060434613,0.124088004,0.047553435,0.09717809,0.08297382,0.05612234
-0.012886485,0.011553485,0.10148387,0.28855413,0.030025361,0.045084067,0.02283268,0.114905074,0.054962385,0.22098571,0.051132176,0.045594625
-0.025287734,0.01807564,0.09002567,0.12390211,0.06067184,0.12525174,0.10664663,0.096130595,0.07915938,0.07539393,0.1290897,0.070365
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.019000068,0.031819124,0.030243652,0.07024996,0.21407309,0.13396353,0.028858693,0.026585873,0.06696755,0.16285823,0.15180056,0.022563139,0.041016474
+0.017298587,0.02785987,0.026171926,0.022916906,0.29258338,0.047437537,0.034001548,0.03626522,0.08464918,0.20266658,0.11324245,0.06333742,0.03156937
+0.0686808,0.047946926,0.05949635,0.06376796,0.06336447,0.15089422,0.07095185,0.051540107,0.055402566,0.20384508,0.026530573,0.08195969,0.055619407
+0.028476672,0.036995783,0.024548426,0.05917686,0.32845432,0.06365823,0.01987992,0.022703582,0.13638632,0.07755896,0.11989127,0.025526324,0.056743342
+0.016698256,0.012903412,0.01798473,0.037851244,0.29368037,0.023183238,0.032629814,0.026476089,0.12052618,0.07571856,0.26531813,0.021524908,0.055505183
+0.045183532,0.035257764,0.031236647,0.11381345,0.10505485,0.08457898,0.07918349,0.04026563,0.08683875,0.17422402,0.07487796,0.07225598,0.057228997
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 1cc44686..a5de7340 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.021187872,0.01605604,0.054263867,0.5659949,0.045957677,0.034796704,0.03555238,0.069608174,0.0645029,0.0374044,0.031313643,0.023361403
-0.024042008,0.015890798,0.19133413,0.07945869,0.039024033,0.08096902,0.16238318,0.1394373,0.051296122,0.07202193,0.1016168,0.04252598
-0.017251918,0.010221423,0.046895575,0.07552666,0.05393696,0.050017677,0.11118601,0.32807288,0.07731773,0.04220145,0.096978165,0.09039358
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.022081608,0.018594578,0.018740417,0.103307895,0.47212556,0.04692973,0.030657277,0.045264293,0.059411697,0.078746155,0.02350575,0.05239225,0.028242808
+0.027631663,0.025856346,0.021603787,0.16926447,0.12096786,0.07964316,0.043301236,0.1349494,0.09815427,0.05714117,0.099311285,0.0701477,0.052027624
+0.026684716,0.020863418,0.016504321,0.07736575,0.10974434,0.03814951,0.06553134,0.049032796,0.21570995,0.21236569,0.061913315,0.043144707,0.062990114
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 7ac72445..d58192d2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.018481381,0.015086826,0.15493174,0.15046775,0.08715649,0.04693676,0.06383222,0.1263513,0.05064553,0.06657839,0.17764094,0.04189063
-0.016217614,0.012148656,0.08499495,0.09927823,0.077306755,0.047871754,0.055010583,0.21122654,0.062933914,0.10617896,0.112716906,0.11411521
-0.04758948,0.026389102,0.04877754,0.053574916,0.05382935,0.09806836,0.07419494,0.1068721,0.23706995,0.060697217,0.050034944,0.14290217
-0.022031544,0.017568294,0.057664018,0.3444,0.03500623,0.048234522,0.055807646,0.08084915,0.07480748,0.12028488,0.08741878,0.05592742
-0.026553975,0.020047663,0.12833178,0.15571688,0.07450259,0.064350314,0.083343506,0.10243373,0.08976791,0.038867597,0.17135589,0.044728138
-0.01926051,0.017816322,0.20410872,0.16115807,0.08684079,0.086861335,0.11499798,0.060789067,0.0708855,0.097253375,0.055153485,0.024874803
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.016777907,0.029446134,0.027233222,0.09903301,0.15368521,0.065074414,0.049991827,0.071156874,0.1554968,0.06006656,0.10398861,0.11969016,0.048359197
+0.023548124,0.02840446,0.027638298,0.0627111,0.20662825,0.07437815,0.030832749,0.047290612,0.15904744,0.048412077,0.11411094,0.045656934,0.13134089
+0.020972665,0.028554477,0.028778432,0.029231627,0.05453193,0.057009615,0.037976593,0.024233868,0.18987098,0.32806766,0.0779229,0.04914536,0.073703915
+0.02953646,0.036472656,0.021108754,0.053275105,0.39211807,0.06938212,0.028074007,0.041734517,0.065625556,0.09798792,0.073174864,0.06043496,0.031075094
+0.054990366,0.03739047,0.034580532,0.07439615,0.102434985,0.16691916,0.07954307,0.051911525,0.051206626,0.2151673,0.025698068,0.05061005,0.05515171
+0.035944857,0.03580472,0.038714133,0.05932095,0.06886392,0.0958184,0.08983742,0.06345591,0.0639536,0.14272109,0.23808104,0.028769901,0.038714133
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 9c317eff..19b076a5 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.024021462,0.016381215,0.08735352,0.24451095,0.042655937,0.053293943,0.0703621,0.16674198,0.0637845,0.048809845,0.12784553,0.054239035
-0.019587502,0.01649433,0.1322968,0.19324411,0.07915247,0.08095043,0.03358079,0.15167905,0.1481654,0.07876693,0.03941358,0.026668562
-0.025625726,0.014773212,0.05668684,0.21879344,0.057635788,0.052529253,0.06815239,0.20714946,0.06667327,0.03907457,0.14292884,0.04997713
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.04036501,0.0526459,0.035483122,0.057146583,0.056979407,0.10676387,0.06257964,0.06425163,0.12384826,0.17465405,0.050137047,0.100884765,0.07426074
+0.0159961,0.026167873,0.02203556,0.084553465,0.22782892,0.074536234,0.040371586,0.03837265,0.13551247,0.18814044,0.07248592,0.03346922,0.040529598
+0.018406145,0.016499165,0.01400266,0.088499516,0.33661324,0.033329222,0.051798053,0.042049877,0.14087047,0.096441306,0.039967842,0.07855951,0.042963065
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 080e0d42..51ade0ca 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.032925528,0.04885123,0.104025796,0.2016933,0.14358173,0.054073382,0.07711691,0.057899076,0.04113686,0.09965061,0.09252246,0.046523213
-0.022180002,0.017545827,0.026339404,0.06680181,0.056901958,0.041035004,0.08519086,0.20880033,0.061623402,0.029155063,0.22053711,0.16388926
-0.017088706,0.023911403,0.14086375,0.034385562,0.080261745,0.05502852,0.046999488,0.06905494,0.028506633,0.050992366,0.4205414,0.0323654
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.04502345,0.037545096,0.032875586,0.1290136,0.23822105,0.09531464,0.04755424,0.04746145,0.110351175,0.038585886,0.04271068,0.039655525,0.09568769
+0.024061177,0.0339317,0.034197826,0.027478727,0.10804553,0.092235915,0.03138169,0.15936935,0.15446608,0.06424972,0.027264884,0.13900442,0.104313046
+0.01398243,0.020826768,0.033673387,0.059273835,0.023971463,0.071147464,0.054311454,0.12708203,0.15815614,0.043747153,0.033022083,0.31823865,0.04256715
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
index 9bf90eea..47225f8b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0040172087,0.0053837127,0.01794233,0.008481664,0.041915163,0.060892984,0.015276147,0.03922581,0.0074369167,0.0065139355,0.020977434,0.020048343,0.019622974,0.037060324,0.008304551,0.23707671,0.00647016,0.01106858,0.011167494,0.012748863,0.033062108,0.018191332,0.0066894284,0.047790576,0.10322021,0.0052910694,0.11262705,0.00659916,0.017055837,0.026257407,0.015436073,0.016148428
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0051731854,0.006115724,0.004452979,0.02984497,0.0035049068,0.023042496,0.04243705,0.017752096,0.043330517,0.005744317,0.005712806,0.02443036,0.03076382,0.016793936,0.020566238,0.014915136,0.23716973,0.006960561,0.016510267,0.008906468,0.016893843,0.032925818,0.022828635,0.0059587103,0.08213512,0.101522684,0.009693471,0.08036983,0.0065862564,0.016128413,0.025845902,0.011681713,0.02330209
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
index a5f8bc58..aab7381f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004360183,0.006277071,0.017062904,0.009641751,0.043229353,0.05727278,0.014225212,0.038007263,0.007835306,0.006881622,0.023389304,0.016517002,0.020064484,0.039711393,0.008007171,0.2261134,0.007357157,0.012218177,0.012470042,0.014183982,0.030792018,0.019355278,0.0073876814,0.04283245,0.12202452,0.0056234123,0.10076162,0.0072904723,0.015891792,0.029221293,0.018248018,0.015745874
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0051706634,0.006392545,0.004375827,0.030993352,0.0034347752,0.021954894,0.041624684,0.017873196,0.04168018,0.00586722,0.005894553,0.022671,0.0292719,0.01654101,0.019867422,0.013532966,0.2661518,0.007395097,0.01631813,0.008761997,0.016077364,0.033230375,0.020873664,0.0064080334,0.07242437,0.097736195,0.008801166,0.08053392,0.0065339063,0.015795996,0.02301082,0.012271813,0.02052921
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
index 59ae4a55..abcdbcaa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004300562,0.006707173,0.0156384,0.009617707,0.04338564,0.06583099,0.014432732,0.035469737,0.008234734,0.00742003,0.021091163,0.015702654,0.019683242,0.038769204,0.008150284,0.23311892,0.008400009,0.011828063,0.012611091,0.01447467,0.03342824,0.019685786,0.007448794,0.041252162,0.11632072,0.006234854,0.091513716,0.007690376,0.015090434,0.031744882,0.018613188,0.016109822
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0064157196,0.006527345,0.0056665037,0.027466085,0.0044602095,0.026517421,0.049759857,0.01652433,0.041261666,0.006874849,0.0067948974,0.023469295,0.029153131,0.016635327,0.019607816,0.016634526,0.17611296,0.007703006,0.023564544,0.010035094,0.022443883,0.03051713,0.025578223,0.0072342893,0.08143584,0.14751312,0.013761597,0.055317592,0.0074805734,0.01573484,0.03309507,0.01149615,0.027207157
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
index 3be3b2c1..8abda85f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0042441166,0.006550821,0.014524397,0.008744351,0.049152624,0.060611952,0.017833132,0.039753407,0.007960037,0.007296914,0.020477273,0.01763899,0.020569323,0.03703013,0.009169324,0.21462496,0.0073043634,0.011937346,0.012409705,0.016439967,0.026413875,0.018571155,0.00883063,0.052616574,0.12547207,0.0054714065,0.08710683,0.0072533595,0.016600486,0.03165694,0.01850874,0.017224781
-0.0044898638,0.006292895,0.019563716,0.009088499,0.045074668,0.059290387,0.013071808,0.038151007,0.007435785,0.0067792335,0.023250628,0.017980209,0.020187698,0.033060163,0.007900615,0.21634111,0.007439023,0.0124958875,0.012788248,0.014472153,0.030570945,0.019401966,0.0072201015,0.041807372,0.12621097,0.005592084,0.10920363,0.007921556,0.014996767,0.028330075,0.018433904,0.015157071
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0055406922,0.006386817,0.004625329,0.02794532,0.0038382933,0.023122994,0.047041234,0.017717674,0.043066762,0.0060848948,0.006028181,0.024094954,0.031411633,0.017373681,0.020232791,0.016327478,0.21710728,0.0073115835,0.017294787,0.008706568,0.019171692,0.034296166,0.022187935,0.0064854,0.08433365,0.10816069,0.01044296,0.07764751,0.0066119726,0.016789336,0.028201265,0.012045506,0.022366948
+0.005273,0.006350921,0.0045870915,0.029855195,0.003674753,0.022280036,0.04056523,0.018330282,0.042522132,0.0058698533,0.005789554,0.023899432,0.030028421,0.016782813,0.018927349,0.014220322,0.25088266,0.0070974855,0.017512802,0.008408013,0.01705966,0.032813244,0.022579813,0.0065027904,0.08215564,0.09919083,0.009322651,0.072932474,0.0066912686,0.015727002,0.02790005,0.011463137,0.022804117
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
index 7f4da896..1adcaec0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004009504,0.00491257,0.020144317,0.008090519,0.03827003,0.05458871,0.012453571,0.032190487,0.0065081106,0.0065404586,0.024827298,0.0168541,0.021103444,0.02935654,0.00813333,0.23077916,0.006991138,0.008845747,0.011959966,0.010493611,0.03519008,0.017726563,0.005951777,0.04237902,0.09315022,0.005446867,0.17138276,0.0067939414,0.01596632,0.020403966,0.0153063685,0.013249553
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.005243044,0.0068628322,0.004119111,0.027635053,0.0037787466,0.021621196,0.03984544,0.020376597,0.036851183,0.005732252,0.0054553016,0.02128327,0.026677908,0.019110823,0.01986349,0.013093736,0.30414927,0.006873597,0.014688124,0.007742951,0.014003141,0.031778734,0.020403402,0.006862341,0.07797771,0.064194806,0.0077092745,0.088538475,0.0069644405,0.016042879,0.022522885,0.012781177,0.019216869
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
index c99b1ffa..c03ca665 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0038721918,0.005179969,0.01615901,0.008171162,0.043776285,0.065734304,0.017611375,0.04358564,0.0065292907,0.0068701566,0.021048758,0.019796705,0.021642718,0.036108084,0.008354264,0.22899278,0.0068489937,0.010631369,0.011964392,0.013607925,0.035942245,0.016927836,0.0066868635,0.05139611,0.11702679,0.0061801723,0.0860707,0.006092417,0.017596524,0.026235523,0.015397529,0.017961878
-0.004533815,0.005865554,0.01902702,0.008466478,0.04793634,0.056333646,0.01547706,0.043305673,0.00786583,0.0073028216,0.02227584,0.020118108,0.02158854,0.033464324,0.007674004,0.21536033,0.00774036,0.011820359,0.011987151,0.0145265665,0.03601687,0.01974236,0.0073520797,0.049199916,0.10748087,0.0053152973,0.11075006,0.0071330904,0.016156215,0.025402343,0.01668626,0.0160948
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.005483725,0.0066012214,0.005198228,0.033509348,0.0042043817,0.02317797,0.03937999,0.017307758,0.037636165,0.005854237,0.0058676037,0.02215927,0.024298837,0.016519763,0.01830959,0.013106523,0.27282035,0.0074146762,0.019438406,0.0094330795,0.016409066,0.027353697,0.023160962,0.0076946104,0.074612305,0.11091272,0.0100175105,0.06297743,0.007235825,0.013821495,0.024065388,0.011177126,0.022840766
+0.005132123,0.006258132,0.004523277,0.028863266,0.0038300497,0.02196964,0.043445766,0.018415095,0.051318515,0.0059415693,0.0059717335,0.024573747,0.03365483,0.017032608,0.022367323,0.015106722,0.21536613,0.0069446526,0.016772965,0.008130857,0.016136188,0.037611797,0.023627339,0.005972406,0.0856897,0.09184303,0.009807569,0.08422559,0.0065450855,0.01732655,0.02972546,0.012266841,0.023603454
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
index 6866d118..8fa8b147 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0042001447,0.0057136347,0.016083458,0.008578314,0.041375857,0.061581593,0.015413483,0.03713672,0.0070986184,0.006767118,0.021731189,0.01628298,0.020695891,0.036857005,0.008185106,0.23843908,0.0069048796,0.01146324,0.011551858,0.014451719,0.030510709,0.018468594,0.0077012125,0.041727073,0.1173028,0.0058830674,0.101249166,0.0075246836,0.015359337,0.030536193,0.018259723,0.014965526
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0055701155,0.0063301595,0.0042756563,0.029668182,0.003665929,0.023308653,0.041472793,0.01839364,0.03860736,0.0060253735,0.0056214742,0.020402266,0.027661264,0.016783247,0.01859294,0.013065602,0.28502858,0.006916266,0.01649864,0.008324767,0.015853647,0.026072072,0.02126262,0.007229569,0.07821424,0.10320244,0.009021552,0.0650905,0.006879877,0.015151082,0.022952529,0.012675275,0.020181743
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
index 283f60ec..c0b34494 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.003893989,0.0049262834,0.016242828,0.008228639,0.040873,0.06588315,0.017734595,0.03817924,0.0072236313,0.006978905,0.020142669,0.020148635,0.020427244,0.038231753,0.007943124,0.24722922,0.0073402775,0.010124343,0.009805043,0.010841479,0.035839595,0.018256433,0.006397891,0.058549624,0.081516184,0.0055939215,0.11330463,0.0059981574,0.018389849,0.022836696,0.013435556,0.017483389
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.005370935,0.0065400284,0.004214182,0.028507866,0.004052519,0.023884999,0.0431287,0.01866619,0.038549002,0.0058642467,0.005612742,0.022007838,0.028794203,0.018062135,0.018920997,0.014034251,0.28207308,0.00724066,0.01554973,0.008246547,0.015637344,0.029362565,0.020395592,0.0072190757,0.08529724,0.08158563,0.008495348,0.07213983,0.0073112454,0.015740085,0.024467451,0.01239826,0.020629534
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
index dd2ce90f..d1eb6c78 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.003755447,0.005919478,0.021355113,0.0037806085,0.03258467,0.05897691,0.010546966,0.025433974,0.003790851,0.0039917994,0.024652747,0.02598865,0.024072263,0.03375422,0.015235061,0.1802716,0.004973748,0.009942133,0.011008963,0.019226586,0.0240091,0.023200186,0.0065351687,0.05277045,0.20373958,0.004230297,0.07716799,0.008635033,0.014409094,0.030130265,0.01123739,0.024673715
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.003818576,0.0031525614,0.0051886565,0.027662326,0.006760011,0.041391376,0.062559254,0.016534436,0.041413,0.006456904,0.0038409191,0.031356983,0.025992455,0.02640143,0.038692255,0.022995137,0.17438106,0.00587134,0.01310105,0.012426717,0.017008489,0.037873317,0.023261664,0.0044478937,0.07316778,0.08910169,0.00722368,0.09950632,0.0060560848,0.021839175,0.01798219,0.013436582,0.01909868
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
index 7b7003a3..57d78954 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004108263,0.0065284492,0.02557481,0.0044105044,0.0301069,0.056982104,0.009312343,0.028644396,0.0036156578,0.0050784764,0.023851369,0.023929859,0.024968062,0.034262866,0.015073786,0.18393585,0.005726918,0.010663445,0.011374998,0.020710837,0.024605207,0.022978598,0.0071213874,0.046844285,0.20492582,0.0045672082,0.06766794,0.009786485,0.015642509,0.027877597,0.012733039,0.026390014
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0038112525,0.0032467076,0.005103138,0.028848037,0.006253271,0.040058028,0.06397161,0.015401709,0.039240718,0.007229387,0.0042264964,0.031109253,0.022743622,0.026395943,0.03867059,0.0224054,0.16184604,0.006153928,0.014256049,0.012480883,0.018035192,0.033755783,0.022141878,0.0048214206,0.06857576,0.11654172,0.0072527346,0.091533765,0.0067206784,0.023300141,0.021024033,0.014561617,0.018283268
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
index fd6efa28..f09487aa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004564731,0.006814505,0.024392659,0.004759701,0.032980636,0.049138952,0.009276454,0.02643774,0.0042945277,0.0050275098,0.024946013,0.024413325,0.025301483,0.029949466,0.014343448,0.16849588,0.0060743075,0.011278828,0.012514742,0.020937728,0.026535869,0.027676357,0.00742597,0.047013473,0.20633397,0.0054930253,0.075621,0.0104807485,0.014071454,0.030229164,0.013769736,0.029406633
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0038847511,0.0034304704,0.0054487633,0.030906213,0.0065313987,0.04176309,0.0651858,0.015840625,0.039366536,0.00712093,0.0041720322,0.030338436,0.021227272,0.027650725,0.03322719,0.02358945,0.17575502,0.0062058195,0.015167547,0.013033766,0.018510986,0.03583581,0.021288542,0.0049735317,0.066214904,0.10693726,0.0075933044,0.08376561,0.007044815,0.021833733,0.02074671,0.015919521,0.019489419
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
index ff6923be..9faa7fb9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0041145687,0.005967945,0.019461952,0.004325763,0.0340289,0.056127317,0.01061059,0.025928281,0.0038670604,0.0041770483,0.022350615,0.030672995,0.022840748,0.030872822,0.017933363,0.15256037,0.0054863794,0.010823059,0.011521493,0.021566499,0.01982184,0.023499597,0.008011047,0.048069283,0.23439504,0.004496008,0.0691574,0.0085690105,0.014794546,0.035597444,0.012133661,0.026217332
-0.0041103526,0.006220995,0.024224846,0.004720488,0.03338627,0.055826817,0.008586799,0.028634403,0.0038456598,0.004600682,0.025519812,0.026524976,0.024987198,0.030848276,0.015773777,0.17763434,0.006000569,0.011141409,0.011032916,0.021263443,0.025327949,0.024359219,0.0073631583,0.050158545,0.18946855,0.00479376,0.07836646,0.010019454,0.014274488,0.030027406,0.012687519,0.028269513
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0038600573,0.0032016537,0.0050423536,0.030087678,0.0062640845,0.042983707,0.06558176,0.016441714,0.037939113,0.0068796654,0.0041524945,0.033378396,0.02355689,0.02613971,0.03884025,0.022489678,0.16319337,0.005882346,0.0136809675,0.012107426,0.018530237,0.035350807,0.02193522,0.0048262407,0.069117285,0.106256664,0.007452489,0.08964618,0.0068337126,0.02311564,0.021159057,0.014839026,0.019234167
+0.0038384346,0.0032803454,0.005382596,0.028020917,0.0065062335,0.040472195,0.059231486,0.01510768,0.044317365,0.00677788,0.004063074,0.030269165,0.023738174,0.028254144,0.035682805,0.02283463,0.17437491,0.0062798257,0.014255069,0.013157966,0.018651582,0.03610014,0.021770932,0.0047950475,0.06709478,0.10704565,0.007234076,0.09272612,0.0065334365,0.020849966,0.019035984,0.013944705,0.018372733
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
index ee563925..4062c3bf 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0039387504,0.0065258797,0.024669716,0.004433713,0.030984672,0.056557093,0.008888177,0.028441487,0.003681258,0.0049070055,0.023858456,0.020984871,0.022434939,0.033650003,0.011881462,0.21709673,0.005855772,0.010228604,0.011613099,0.017855505,0.02831009,0.021852583,0.0060879337,0.04023303,0.16921678,0.00440954,0.09614935,0.009410114,0.01360511,0.0240166,0.012136324,0.026085403
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.004071277,0.0036563596,0.005473332,0.026431864,0.0070418348,0.040103372,0.058523994,0.014499132,0.048744615,0.0068927244,0.003979709,0.024812536,0.02332055,0.028595718,0.034942232,0.022658326,0.19808845,0.0063588833,0.014898518,0.014481609,0.017180959,0.036216,0.021954877,0.004474714,0.07242044,0.09361984,0.0072108386,0.08443967,0.0062276307,0.019533861,0.01759654,0.013579895,0.01796969
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
index fcf89d60..717cd470 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004052816,0.006258594,0.023345444,0.00449139,0.030517468,0.05268502,0.00887704,0.026460174,0.004128396,0.004363119,0.022904512,0.023697617,0.024568599,0.032895457,0.012510463,0.19101547,0.005496111,0.010627602,0.011373989,0.01724674,0.027330758,0.024636947,0.0059708157,0.04202071,0.18282118,0.0045206607,0.10554287,0.008928312,0.014136884,0.02655903,0.011113563,0.02890221
-0.0059897667,0.008406405,0.026541125,0.0054274616,0.03536363,0.04665471,0.0095773535,0.02762081,0.005177147,0.0062065427,0.025016924,0.032284416,0.029640928,0.028332105,0.019341344,0.11613233,0.0074259285,0.013591332,0.0128795,0.027375998,0.021657849,0.031640794,0.01288579,0.055171557,0.22059953,0.0076504927,0.040872708,0.011520912,0.017828679,0.04107078,0.014468098,0.03564709
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0042564487,0.0033090666,0.0054760864,0.029484103,0.0063959234,0.032995798,0.058816995,0.013607776,0.048283372,0.0057379017,0.0038245237,0.021225223,0.023702899,0.026598837,0.030306822,0.02262228,0.23025683,0.0063128057,0.014794539,0.013137006,0.019704955,0.031384137,0.019589942,0.004760695,0.06250796,0.106131054,0.006412059,0.074764796,0.005834489,0.019974858,0.01842272,0.012555495,0.016811622
+0.0038192424,0.0033686042,0.0054370016,0.02769945,0.006779793,0.04220823,0.05973473,0.015520086,0.04562838,0.0071383878,0.00422071,0.03299786,0.025394814,0.0281485,0.038514096,0.023156043,0.16337366,0.0063498737,0.014283524,0.012739055,0.01869462,0.036976255,0.023111258,0.004908983,0.07237423,0.09736417,0.0072591524,0.08847638,0.006704482,0.022873966,0.021077238,0.0138563765,0.019810876
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
index 52fec462..887cda4e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004102211,0.0063269846,0.02072375,0.0048008375,0.034570824,0.059226204,0.0096925255,0.028749216,0.0039989166,0.004183866,0.023596365,0.025296958,0.025585536,0.031098658,0.013962489,0.17769668,0.005346446,0.010870259,0.01055643,0.02107274,0.026247509,0.023033429,0.007330994,0.052349277,0.19195877,0.0043321247,0.08064205,0.009575042,0.016013324,0.02888181,0.012639321,0.02553845
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.0038691815,0.0031717422,0.005064685,0.029039107,0.0061659142,0.037487313,0.06612843,0.01513539,0.03811665,0.0066099614,0.0039828895,0.026459182,0.02207245,0.029161863,0.032705896,0.02051133,0.18691592,0.0063597304,0.0139315445,0.013663895,0.017307293,0.030300051,0.019515662,0.004841311,0.059204914,0.1291775,0.0065799546,0.08632184,0.006249234,0.022190582,0.020693045,0.014265866,0.016799755
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
index cffd8dd7..34ba1b0b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0038631586,0.006477673,0.024767213,0.003959401,0.030919716,0.063857056,0.009385453,0.025800953,0.0038557237,0.004502775,0.024877692,0.022792442,0.027524808,0.033570554,0.013489485,0.1976561,0.0054532574,0.010784638,0.010051554,0.020462193,0.026673099,0.021765215,0.006219106,0.048690844,0.1782489,0.0051354156,0.07887633,0.009734921,0.015730219,0.027250536,0.012058978,0.025564646
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.004423313,0.004286555,0.00660057,0.023772126,0.008373217,0.040369432,0.054351408,0.01465567,0.05245067,0.008132533,0.0042896164,0.029554067,0.024817849,0.03035214,0.039045297,0.025620734,0.17074145,0.0072791767,0.015768386,0.016019871,0.01698007,0.041483015,0.025339559,0.00460742,0.08207413,0.08653425,0.008505664,0.07673737,0.00643838,0.020246439,0.016483322,0.014442152,0.01922423
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index bc35a828..f2414b5b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02987246,0.0291298,0.034083534,0.026629673,0.032076977,0.030450517,0.035012987,0.029515885,0.035601407,0.02849371,0.034817744,0.030076707,0.029692352,0.030673426,0.029644514,0.038891632,0.030484164,0.032383278,0.02580936,0.035029277,0.03223794,0.033890463,0.03001799,0.031699885,0.033890437,0.025421664,0.032209013,0.029906737,0.028936805,0.03221391,0.02733048,0.033875357
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.029515171,0.029384783,0.02892698,0.031414278,0.030240053,0.03790003,0.031263754,0.030703781,0.02985053,0.03231135,0.030141735,0.028332692,0.033920184,0.027992733,0.031994548,0.032312855,0.032003276,0.029009493,0.025961198,0.028668242,0.031744353,0.028004343,0.0351939,0.028472656,0.033312727,0.029293254,0.025253193,0.03191485,0.031521086,0.030837052,0.027421221,0.028424587,0.026759079
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index 9e8c6ade..ce3129a4 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027755681,0.033875797,0.035090785,0.026336204,0.027495425,0.027336687,0.03491952,0.032417938,0.034913428,0.030351402,0.027904775,0.03262825,0.034914885,0.034497607,0.032371227,0.031979073,0.028421713,0.03136423,0.030974774,0.036030304,0.030301245,0.028893707,0.031219127,0.035561685,0.032045122,0.02882453,0.030305024,0.030230341,0.029906748,0.030945972,0.02952125,0.030665599
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.02660504,0.030707078,0.032001417,0.035039,0.029511122,0.034290012,0.03181159,0.0294397,0.0321122,0.03097334,0.0332632,0.027464185,0.030202633,0.028321788,0.025869891,0.03017357,0.031728413,0.033907395,0.02845779,0.031313535,0.028817244,0.02832817,0.033035055,0.026313312,0.030338703,0.028560834,0.03160956,0.031486206,0.030627396,0.029206958,0.030443136,0.027497737,0.030542677
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index 69d54ab8..db2db85d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.028207576,0.03346115,0.03521835,0.026430761,0.02787494,0.02663613,0.03487845,0.031793576,0.035627346,0.03034215,0.02791883,0.032285154,0.03488344,0.03348667,0.031713963,0.033340506,0.028663984,0.03198072,0.029641826,0.03631048,0.030261524,0.029211927,0.0310106,0.035286736,0.032737605,0.028236382,0.031136692,0.030599257,0.030495808,0.030660555,0.029510591,0.030156327
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.027332451,0.031302933,0.030766888,0.034912553,0.029892711,0.03336584,0.031323344,0.029848998,0.03191401,0.031425472,0.033242542,0.027521798,0.029320594,0.027383283,0.025944008,0.03078899,0.032123405,0.03406133,0.028520554,0.031947363,0.0294212,0.028340716,0.032867085,0.02619513,0.02996483,0.028789517,0.03081592,0.031394005,0.03228332,0.029442273,0.02991017,0.026885256,0.03075153
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index 98cb886d..b2366f0b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.033209767,0.027763555,0.02949299,0.033150658,0.034999833,0.0368729,0.02978079,0.030827697,0.02852166,0.028758308,0.037743535,0.03003544,0.027150638,0.030506711,0.031755105,0.03137323,0.03211672,0.029910667,0.030969618,0.028884642,0.033850152,0.03439652,0.030612124,0.028037883,0.030710937,0.031090267,0.030988533,0.029567717,0.030128382,0.032672625,0.029851323,0.034269065
-0.031579662,0.029196938,0.029836416,0.033177998,0.032533493,0.037377365,0.029789915,0.032062903,0.028400185,0.02797571,0.035992093,0.030920217,0.02906526,0.032699972,0.033772778,0.028461806,0.031120755,0.029319655,0.034247786,0.02993261,0.03364331,0.032858204,0.031222675,0.029189978,0.029043447,0.032523304,0.029388353,0.028918054,0.029303428,0.03236885,0.030073373,0.0340035
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.03185726,0.028593035,0.029586729,0.02717517,0.029036317,0.03239331,0.030127635,0.029756432,0.028826356,0.03008631,0.027377693,0.03182702,0.03495427,0.032569427,0.03616851,0.031138953,0.02905974,0.026557622,0.028117852,0.02744148,0.032749053,0.030639345,0.031042855,0.03220263,0.033083834,0.030625483,0.027325926,0.030158056,0.028103651,0.03071186,0.029626483,0.033320524,0.027759206
+0.031184139,0.0276112,0.032012727,0.027526684,0.029155925,0.032260306,0.030398509,0.029150296,0.029186625,0.029938763,0.02798222,0.03192229,0.034919117,0.034535985,0.03453919,0.0301626,0.02843686,0.026707487,0.02883699,0.027758451,0.03170132,0.030358283,0.030214984,0.032069325,0.033027902,0.029928787,0.02964113,0.029945217,0.026400918,0.029637858,0.030964117,0.033538274,0.028345458
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index 457f32d0..5c1cac56 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.028416378,0.031858474,0.035576776,0.026082192,0.029002782,0.027531331,0.035973124,0.031896308,0.035577897,0.030908158,0.029429477,0.031626973,0.033014473,0.03156573,0.03071435,0.035173185,0.029379372,0.031848334,0.027813641,0.03572987,0.030105326,0.031015579,0.030929359,0.034231957,0.032853644,0.02719569,0.032351676,0.03100664,0.029942783,0.03149756,0.028656607,0.031094389
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.027090317,0.031356793,0.031175932,0.034752592,0.028963773,0.035918295,0.032366708,0.029567668,0.031858996,0.031686503,0.031844135,0.027539104,0.03215706,0.02786867,0.028114032,0.030691825,0.03100698,0.032229245,0.026652839,0.030048568,0.02923739,0.02832744,0.035018757,0.0267673,0.031383105,0.028060874,0.028454317,0.03174636,0.030990787,0.030642902,0.030124784,0.02802195,0.028333979
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index 2175edbc..3ebe548e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.030749116,0.031062534,0.029946074,0.03304623,0.03177652,0.036106378,0.029489571,0.0326022,0.027348582,0.029303638,0.032161407,0.03134503,0.031155298,0.032429844,0.034513615,0.027003089,0.031428523,0.028919302,0.037302244,0.029815737,0.032031134,0.030656634,0.032251637,0.029612571,0.028429309,0.03463406,0.02905775,0.029093673,0.030223982,0.032136854,0.032039963,0.032327514
-0.029171862,0.032260604,0.031479053,0.027849672,0.031409796,0.033327498,0.031761337,0.028732147,0.03429961,0.02492806,0.036731984,0.030338226,0.03251292,0.033941925,0.031204814,0.03442956,0.030674377,0.033014756,0.031726107,0.03608367,0.033338442,0.031844705,0.030129898,0.030746242,0.031700723,0.027094038,0.029256253,0.027283229,0.028538387,0.030670708,0.028805278,0.034714192
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.031245146,0.026523279,0.03429213,0.028590597,0.030950105,0.030549498,0.030602774,0.029434195,0.02976708,0.030371303,0.02923278,0.03193279,0.032057486,0.036185876,0.031233333,0.028832216,0.028311577,0.027465537,0.031915396,0.028459525,0.029657632,0.029486306,0.02885231,0.031842597,0.032040454,0.029984664,0.032361843,0.029286683,0.025559865,0.028563855,0.03226416,0.032237228,0.029909858
+0.031189607,0.027768781,0.03171535,0.0286404,0.02989582,0.03686473,0.030307468,0.029793784,0.029219873,0.03136719,0.029169736,0.030565001,0.035063583,0.033113644,0.033308994,0.03176959,0.030947847,0.027378758,0.026388302,0.028127484,0.03166254,0.02842469,0.033268318,0.029070573,0.033865992,0.028430037,0.027634714,0.031596266,0.02863526,0.028753258,0.02869571,0.030206194,0.027160522
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index deba9442..ef108ae3 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.028799655,0.031856664,0.03283534,0.030231427,0.03015927,0.035772216,0.030314315,0.03224368,0.02983013,0.02746894,0.033414993,0.03132566,0.032006927,0.034900524,0.034404952,0.0279653,0.029755965,0.029660394,0.036329567,0.032125212,0.032573488,0.032185875,0.032224666,0.03076007,0.02825598,0.031999446,0.028718766,0.026965003,0.029454451,0.031563707,0.030569414,0.033328008
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.03022107,0.026185002,0.035542563,0.030422375,0.030111654,0.032666244,0.03029986,0.028772637,0.029862503,0.031565245,0.030188533,0.031473093,0.03380565,0.034408633,0.031200787,0.029270852,0.029954033,0.026347158,0.029255822,0.028777694,0.031132765,0.028987736,0.03171347,0.029104978,0.032444056,0.027695423,0.03135624,0.030321337,0.025290275,0.027976021,0.03161053,0.03318183,0.028853968
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index a882e828..a93e39ea 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.032289907,0.028264519,0.03246752,0.029104413,0.033277128,0.030759139,0.032730892,0.029931597,0.033015575,0.031707276,0.03496927,0.030008726,0.027751597,0.029273633,0.02851361,0.03743072,0.03143402,0.0318968,0.024980046,0.03104891,0.032157212,0.03514057,0.029877376,0.030226672,0.03398293,0.027080398,0.034191467,0.031496957,0.03116081,0.03200099,0.029164206,0.032665085
+unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.030450879,0.031356543,0.026286202,0.030208621,0.029305283,0.033798173,0.029997792,0.031071521,0.029736698,0.031199083,0.029278152,0.029431423,0.03262524,0.026415547,0.03385641,0.032870974,0.031922586,0.029316304,0.026414368,0.029405521,0.03292572,0.03013779,0.033768725,0.029333936,0.031706,0.030111287,0.025230156,0.03123331,0.033786975,0.032116424,0.027622495,0.029445931,0.027633844
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index 4dbf410d..7b307493 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.0748158,0.0772802,0.08608035,0.08609827,0.07993774,0.07705543,0.09423849,0.086710036,0.086828426,0.08284977,0.088354506,0.079750955
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.069280244,0.08278568,0.072002776,0.068465054,0.073495515,0.069683656,0.08337737,0.08428103,0.079714045,0.08516558,0.07238791,0.07722001,0.08214111
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index 6d89b556..f5c9cd40 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.07326655,0.08353559,0.07533537,0.08782462,0.08211486,0.08560346,0.08289929,0.08048919,0.08408942,0.0929573,0.089654244,0.082230076
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.08120824,0.0770729,0.080374785,0.07006595,0.08169089,0.06985054,0.077308096,0.06746512,0.08158482,0.078970395,0.07628951,0.077936,0.08018279
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index bf13de6f..f76ecc72 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.073459856,0.08505141,0.07544217,0.08925692,0.08299525,0.0825825,0.082352445,0.081173785,0.085358195,0.091914274,0.09012994,0.08028323
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.081267715,0.07711225,0.080673225,0.07033195,0.08024122,0.06911172,0.07555922,0.06642563,0.0804719,0.08202005,0.07461091,0.079976365,0.08219792
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index eb1f1644..a99204aa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.088295944,0.07419422,0.09625292,0.075507574,0.07814089,0.08647674,0.09078418,0.089773305,0.08116939,0.07382207,0.078370996,0.08721177
-0.0871225,0.0723557,0.095413424,0.07266789,0.07581121,0.09429607,0.089163,0.08606107,0.07906439,0.076575436,0.0773663,0.09410305
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.06782719,0.078334294,0.07045661,0.0793704,0.073115356,0.08177368,0.08440978,0.09057371,0.076156095,0.07454853,0.078214675,0.071796,0.07342377
+0.07023328,0.0774746,0.07357778,0.07948175,0.07633355,0.08361607,0.08418787,0.08768755,0.07620843,0.06998042,0.0815577,0.06901032,0.0706507
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index e1753ff4..e0a22959 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.07255601,0.0840692,0.07761021,0.089467965,0.08273568,0.07989225,0.08776099,0.08379931,0.08662724,0.08953262,0.08896252,0.07698605
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.075564295,0.077890314,0.07621962,0.06734453,0.07971224,0.06832646,0.07854212,0.07214813,0.08339084,0.082671195,0.07736374,0.0789834,0.08184321
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index 4704ea71..d2469e90 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.08670176,0.07764295,0.09074383,0.07478624,0.07692155,0.09733425,0.08432601,0.08585503,0.08131818,0.07754218,0.07455162,0.09227636
-0.078247,0.06804814,0.09125708,0.07758486,0.07693072,0.08519664,0.09631906,0.07838798,0.0837765,0.080676705,0.08734935,0.09622603
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.07543901,0.07629552,0.079985805,0.07938394,0.08105448,0.08392259,0.08113814,0.08250335,0.07510063,0.06616155,0.08535827,0.06484661,0.068810105
+0.0684685,0.08285618,0.07265628,0.07221835,0.074672826,0.07720729,0.087464236,0.08804019,0.07745041,0.072739474,0.07616589,0.07260381,0.07745654
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index 037661f9..11d4ea08 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.083966985,0.073211156,0.091535,0.07379018,0.071819365,0.09759158,0.086949356,0.0855146,0.08113236,0.08044534,0.07800937,0.096034706
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.07194322,0.077352665,0.08031595,0.07728969,0.08099176,0.080330044,0.08255197,0.079121925,0.07681379,0.069612056,0.08324984,0.06879602,0.07163119
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index fbf283d5..072240da 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,0,1,2,3,4,5,6,7,8,9
-0.081276454,0.08213866,0.08511532,0.08706703,0.08304337,0.07355978,0.0899685,0.08830506,0.08529962,0.08146648,0.087632634,0.07512704
+unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+0.068312116,0.08032282,0.06863922,0.07357371,0.06984851,0.07086958,0.0803648,0.0826013,0.079164565,0.08935935,0.06972866,0.0837885,0.083426945
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index ac79cf10..fac968d1 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -10,34 +10,36 @@
 
 def test_construct_index_maps_string():
     """Tests reversing a string-to-int map, ensuring 0 maps to 'unknown'."""
-    id_maps: dict[str, dict[str | int, int]] = {"itemId": {"apple": 2, "banana": 3}}
+    id_maps: dict[str, dict[str | int, int]] = {"itemId": {"apple": 3, "banana": 4}}
     target_cols = ["itemId"]
 
     result = construct_index_maps(id_maps, target_cols, map_to_id=True)
 
     # Check standard reversal
-    assert result["itemId"][2] == "apple"
-    assert result["itemId"][3] == "banana"
+    assert result["itemId"][3] == "apple"
+    assert result["itemId"][4] == "banana"
 
     # Check the special 0 index for strings
     assert result["itemId"][0] == "unknown"
     assert result["itemId"][1] == "other"
+    assert result["itemId"][2] == "mask"
 
 
 def test_construct_index_maps_integer():
     """Tests reversing an int-to-int map, ensuring 0 maps to min_id - 1."""
-    id_maps: dict[str, dict[str | int, int]] = {"storeId": {100: 2, 101: 3}}
+    id_maps: dict[str, dict[str | int, int]] = {"storeId": {100: 3, 101: 4}}
     target_cols = ["storeId"]
 
     result = construct_index_maps(id_maps, target_cols, map_to_id=True)
 
-    assert result["storeId"][2] == 100
-    assert result["storeId"][3] == 101
+    assert result["storeId"][3] == 100
+    assert result["storeId"][4] == 101
 
     # Check the special 0 index for integers (min value - 2)
     # Min value is 100, so 2 -> 98, 1 -> 99
-    assert result["storeId"][0] == 98
-    assert result["storeId"][1] == 99
+    assert result["storeId"][0] == 97
+    assert result["storeId"][1] == 98
+    assert result["storeId"][2] == 99
 
 
 def test_construct_index_maps_flag_false():
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 98191bad..359919df 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -198,6 +198,6 @@ def test_create_id_map():
     mapping = create_id_map(df, "A")
 
     # Sorted unique values: x, y, z -> 2, 3, 4
-    assert mapping["x"] == 2
-    assert mapping["y"] == 3
-    assert mapping["z"] == 4
+    assert mapping["x"] == 3
+    assert mapping["y"] == 4
+    assert mapping["z"] == 5

From 709d9f1d8a844f028ae9d5072080b8f3d9fa0ed9 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 4 Jun 2026 10:38:54 +0200
Subject: [PATCH 03/81] Move training_objective to TrainingSpecModel

---
 src/sequifier/config/train_config.py | 123 ++++++++++++++-------------
 src/sequifier/make.py                |   2 +-
 tests/unit/test_train.py             |   2 +-
 3 files changed, 64 insertions(+), 63 deletions(-)

diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index eb095ab3..4ddb8067 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -111,6 +111,27 @@ def __setstate__(self, state):
         self.update(state)
 
 
+class ReplacementDistribution(BaseModel):
+    masked: float = Field(..., gt=0.0, le=1.0)
+    random: float = Field(..., gt=0.0, le=1.0)
+    identical: float = Field(..., gt=0.0, le=1.0)
+
+    @model_validator(mode="after")
+    def validate_sum(self):
+        total = self.masked + self.random + self.identical
+        if not math.isclose(total, 1.0, abs_tol=1e-5):
+            raise ValueError(
+                f"Replacement distribution probabilities must sum to 1.0, got {total}"
+            )
+        return self
+
+
+class BERTSpecModel(BaseModel):
+    masking_probability: float = Field(..., gt=0.0, le=1.0)
+    replacement_distribution: ReplacementDistribution
+    span_masking: ProbabilityDistribution
+
+
 class TrainingSpecModel(BaseModel):
     """Pydantic model for training specifications.
 
@@ -151,6 +172,7 @@ class TrainingSpecModel(BaseModel):
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
+    training_objective: str = "causal"
     device: str
     device_max_concat_length: int = 12
     epochs: int
@@ -176,6 +198,8 @@ class TrainingSpecModel(BaseModel):
         )
     )
     scheduler_step_on: str = "epoch"
+    bert_spec: Optional[BERTSpecModel] = None
+
     continue_training: bool = True
     enforce_determinism: bool = False
     distributed: bool = False
@@ -293,6 +317,27 @@ def validate_scheduler_step_on(cls, v):
             )
         return v
 
+    @field_validator("training_objective")
+    @classmethod
+    def validate_training_objective(cls, v):
+        if v not in ["causal", "bert"]:
+            raise ValueError(f"Only 'causal' and 'bert' are allowed, found {v}")
+        return v
+
+    @field_validator("bert_spec")
+    @classmethod
+    def validate_bert_spec(cls, v, info):
+        training_objective = info.data.get("training_objective")
+        if v and not training_objective == "bert":
+            raise ValueError(
+                "The BERT hyperparameters should only be configured if the training objective is 'bert'"
+            )
+        if not v and training_objective == "bert":
+            raise ValueError(
+                "If the training_objective is 'bert', the BERT hyperparameters must be set"
+            )
+        return v
+
     @field_validator("sampling_strategy")
     @classmethod
     def validate_sampling_strategy(cls, v):
@@ -312,27 +357,6 @@ def validate_data_parallelism(cls, v):
         return v
 
 
-class ReplacementDistribution(BaseModel):
-    masked: float = Field(..., gt=0.0, le=1.0)
-    random: float = Field(..., gt=0.0, le=1.0)
-    identical: float = Field(..., gt=0.0, le=1.0)
-
-    @model_validator(mode="after")
-    def validate_sum(self):
-        total = self.masked + self.random + self.identical
-        if not math.isclose(total, 1.0, abs_tol=1e-5):
-            raise ValueError(
-                f"Replacement distribution probabilities must sum to 1.0, got {total}"
-            )
-        return self
-
-
-class BERTSpecModel(BaseModel):
-    masking_probability: float = Field(..., gt=0.0, le=1.0)
-    replacement_distribution: ReplacementDistribution
-    span_masking: ProbabilityDistribution
-
-
 class ModelSpecModel(BaseModel):
     """Pydantic model for model specifications.
 
@@ -347,7 +371,6 @@ class ModelSpecModel(BaseModel):
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
-    training_objective: str = "causal"
     initial_embedding_dim: int
     feature_embedding_dims: Optional[dict[str, int]] = None
     joint_embedding_dim: Optional[int] = None
@@ -368,28 +391,6 @@ class ModelSpecModel(BaseModel):
     rope_theta: float = 10000.0
 
     prediction_length: int
-    bert_spec: Optional[BERTSpecModel] = None
-
-    @field_validator("training_objective")
-    @classmethod
-    def validate_training_objective(cls, v):
-        if v not in ["causal", "bert"]:
-            raise ValueError(f"Only 'causal' and 'bert' are allowed, found {v}")
-        return v
-
-    @field_validator("bert_spec")
-    @classmethod
-    def validate_bert_spec(cls, v, info):
-        training_objective = info.data.get("training_objective")
-        if v and not training_objective == "bert":
-            raise ValueError(
-                "The BERT hyperparameters should only be configured if the training objective is 'bert'"
-            )
-        if not v and training_objective == "bert":
-            raise ValueError(
-                "If the training_objective is 'bert', the BERT hyperparameters must be set"
-            )
-        return v
 
     @field_validator("dim_model")
     @classmethod
@@ -660,6 +661,23 @@ def validate_training_spec(cls, v, info):
                 "If 'distributed' is True, data_parallelism cannot be 'None'"
             )
 
+        export_generative_model = info.data.get("export_generative_model")
+        export_embedding_model = info.data.get("export_embedding_model")
+        if v.training_objective == "bert":
+            if export_generative_model:
+                raise ValueError(
+                    "'training_objective: bert' is incompatible with generative model export"
+                )
+            if not export_embedding_model:
+                raise ValueError(
+                    "'training_objective: bert' requires embedding model export"
+                )
+        if v.training_objective == "causal":
+            if not export_generative_model and not export_embedding_model:
+                warnings.warn(
+                    "At least one of 'export_generative_model' and 'export_embedding_model' should be true"
+                )
+
         return v
 
     @field_validator("column_types")
@@ -713,21 +731,4 @@ def validate_model_spec(cls, v, info):
                     f"initial_embedding_dim ({embedding_size}) must be a multiple of the number of categorical variables ({n_categorical}: {categorical_columns})."
                 )
 
-        export_generative_model = info.data.get("export_generative_model")
-        export_embedding_model = info.data.get("export_embedding_model")
-        if v.training_objective == "bert":
-            if export_generative_model:
-                raise ValueError(
-                    "'training_objective: bert' is incompatible with generative model export"
-                )
-            if not export_embedding_model:
-                raise ValueError(
-                    "'training_objective: bert' requires embedding model export"
-                )
-        if v.training_objective == "causal":
-            if not export_generative_model and not export_embedding_model:
-                warnings.warn(
-                    "At least one of 'export_generative_model' and 'export_embedding_model' should be true"
-                )
-
         return v
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 8943d861..6858053f 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -36,7 +36,6 @@
 export_onnx: true
 
 model_spec:
-  training_objective: causal
   initial_embedding_dim: 128
   feature_embedding_dims: # the size of the embedding of individual variables, must sum to dim_model
     EXAMPLE_INPUT_COLUMN_NAME: # can be left out if either all input variables are real or all are categorical
@@ -47,6 +46,7 @@
   num_layers: 3
   prediction_length: 1
 training_spec:
+  training_objective: causal
   device: cuda
   epochs: 10
   save_interval_epochs: 10
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index cf140984..e296979b 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -14,7 +14,6 @@ def model_config(tmp_path):
     (tmp_path / "logs").mkdir(exist_ok=True)
 
     model_spec = ModelSpecModel(
-        training_objective="causal",
         initial_embedding_dim=16,
         dim_model=16,
         n_head=4,
@@ -26,6 +25,7 @@ def model_config(tmp_path):
     )
 
     training_spec = TrainingSpecModel(
+        training_objective="causal",
         device="cpu",
         epochs=1,
         save_interval_epochs=1,

From 7525affd953808b17caf93955cdd9f9f2e031048 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 4 Jun 2026 11:59:21 +0200
Subject: [PATCH 04/81] WIP

---
 documentation/configs/preprocess.md   |   2 +-
 documentation/consolidated-docs.md    |   2 +-
 src/sequifier/config/probabilities.py |  48 +++++-----
 src/sequifier/helpers.py              | 133 ++++++++++++++++++++++++--
 src/sequifier/train.py                |  31 ++++--
 tests/integration/test_inference.py   |  18 ++--
 tests/unit/test_helpers.py            |   8 +-
 7 files changed, 191 insertions(+), 51 deletions(-)

diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index 9b42f592..f58c542c 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -116,7 +116,7 @@ After running `preprocess`, the following are generated:
 
 ## 5\. Advanced: Custom ID Mapping
 
-By default, Sequifier automatically generates integer IDs for categorical columns starting from index 2 (indices 0 and 1 are reserved for system use, such as "unknown" values).
+By default, Sequifier automatically generates integer IDs for categorical columns starting from index 2 (indices 0 and 1 are reserved for system use, such as "[unknown]" values).
 
 If you need to enforce specific integer mappings (e.g., to maintain consistency across different training runs or datasets), you can provide **precomputed ID maps**.
 
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 0b776e6a..65315961 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -324,7 +324,7 @@ After running `preprocess`, the following are generated:
 
 ## 5\. Advanced: Custom ID Mapping
 
-By default, Sequifier automatically generates integer IDs for categorical columns starting from index 2 (indices 0 and 1 are reserved for system use, such as "unknown" values).
+By default, Sequifier automatically generates integer IDs for categorical columns starting from index 2 (indices 0 and 1 are reserved for system use, such as "[unknown]" values).
 
 If you need to enforce specific integer mappings (e.g., to maintain consistency across different training runs or datasets), you can provide **precomputed ID maps**.
 
diff --git a/src/sequifier/config/probabilities.py b/src/sequifier/config/probabilities.py
index 2591cc97..d18a8de4 100644
--- a/src/sequifier/config/probabilities.py
+++ b/src/sequifier/config/probabilities.py
@@ -1,12 +1,9 @@
-import math
-import random
 from abc import ABC, abstractmethod
 from typing import Annotated, Literal, Union
 
+import torch
 from pydantic import BaseModel, Field
 
-epsilon = 1e-20
-
 
 class ProbabilityDistributionBaseClass(ABC):
     """
@@ -14,7 +11,7 @@ class ProbabilityDistributionBaseClass(ABC):
     """
 
     @abstractmethod
-    def sample(self) -> Union[int, float]:
+    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
         pass
 
 
@@ -22,10 +19,14 @@ class GeometricDistribution(BaseModel, ProbabilityDistributionBaseClass):
     type: Literal["GeometricDistribution"] = "GeometricDistribution"
     p: float = Field(..., gt=0.0, le=1.0)
 
-    def sample(self) -> int:
+    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
         if self.p == 1.0:
-            return 1
-        return math.ceil(math.log(random.random() + epsilon) / math.log(1.0 - self.p))
+            return torch.ones(shape, device=device, dtype=torch.long)
+
+        # torch.distributions.Geometric models the number of failures before the first success.
+        # Adding 1 converts it to the total number of trials (span length).
+        m = torch.distributions.Geometric(probs=torch.tensor([self.p], device=device))
+        return (m.sample(shape).squeeze(-1) + 1).long()
 
 
 class NormalDistributionDiscretizedFloor(BaseModel, ProbabilityDistributionBaseClass):
@@ -35,9 +36,11 @@ class NormalDistributionDiscretizedFloor(BaseModel, ProbabilityDistributionBaseC
     mean: float
     standard_deviation: float = Field(..., gt=0.0)
 
-    def sample(self) -> int:
-        val = random.gauss(self.mean, self.standard_deviation)
-        return max(round(val), 0) + 1
+    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
+        val = torch.normal(
+            mean=self.mean, std=self.standard_deviation, size=shape, device=device
+        )
+        return torch.clamp(torch.round(val), min=0).long() + 1
 
 
 class LogNormalDistributionDistcretizedFloor(
@@ -49,25 +52,22 @@ class LogNormalDistributionDistcretizedFloor(
     mean: float
     standard_deviation: float = Field(..., gt=0.0)
 
-    def sample(self) -> int:
-        val = random.lognormvariate(self.mean, self.standard_deviation)
-        return round(val) + 1
+    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
+        m = torch.distributions.LogNormal(
+            loc=torch.tensor([self.mean], device=device),
+            scale=torch.tensor([self.standard_deviation], device=device),
+        )
+        val = m.sample(shape).squeeze(-1)
+        return torch.round(val).long() + 1
 
 
 class PoissonDistributionFloor(BaseModel, ProbabilityDistributionBaseClass):
     type: Literal["PoissonDistributionFloor"] = "PoissonDistributionFloor"
     rate: float = Field(..., gt=0.0)
 
-    def sample(self) -> int:
-        L = math.exp(-self.rate)
-        k = 0
-        p = 1.0
-
-        while p > L:
-            k += 1
-            p *= random.random()
-
-        return k
+    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
+        m = torch.distributions.Poisson(rate=torch.tensor([self.rate], device=device))
+        return m.sample(shape).squeeze(-1).long()
 
 
 ProbabilityDistribution = Annotated[
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 84e39b0c..c57606eb 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -4,7 +4,7 @@
 import re
 import sys
 from datetime import datetime
-from typing import Any, Optional, Union
+from typing import Any, Dict, Optional, Union
 
 import numpy as np
 import polars as pl
@@ -84,9 +84,9 @@ def construct_index_maps(
     operation if `map_to_id` is True and `id_maps` is provided.
 
     A special mapping for these indices are added:
-    - If original IDs are strings, 0 maps to "unknown".
-    - If original IDs are strings, 1 maps to "other".
-    - If original IDs are strings, 2 maps to "unknown".
+    - If original IDs are strings, 0 maps to "[unknown]".
+    - If original IDs are strings, 1 maps to "[other]".
+    - If original IDs are strings, 2 maps to "[unknown]".
     - If original IDs are integers, 0 maps to (minimum original ID) - 3.
     - If original IDs are integers, 1 maps to (minimum original ID) - 2.
     - If original IDs are integers, 2 maps to (minimum original ID) - 1.
@@ -120,9 +120,9 @@ def construct_index_maps(
             map_ = {v: k for k, v in id_maps[target_column].items()}
             val = next(iter(map_.values()))
             if isinstance(val, str):
-                map_[0] = "unknown"
-                map_[1] = "other"
-                map_[2] = "mask"
+                map_[0] = "[unknown]"
+                map_[1] = "[other]"
+                map_[2] = "[mask]"
             else:
                 if not isinstance(val, int):
                     raise TypeError(f"Expected integer ID in map, got {type(val)}")
@@ -511,3 +511,122 @@ def get_last_training_batch_timedelta(
     t1, t2 = timestamps[-2], timestamps[-1]
 
     return (t2 - t1).total_seconds()
+
+
+def apply_bert_masking(
+    data_batch: Dict[str, torch.Tensor],
+    targets_batch: Dict[str, torch.Tensor],
+    config: Any,  # TrainConfig
+) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
+    """
+    Applies BERT-style span corruption to the input data using custom distributions.
+    Explicitly passes the boolean prediction mask via the targets dictionary.
+    """
+    batch_size, seq_len = config.training_spec.batch_size, config.seq_length
+
+    # 1. Identify valid tokens (Renamed from padding_mask to valid_mask for clarity)
+    if len(config.categorical_columns):
+        ref_col = config.categorical_columns[0]
+        valid_mask = data_batch[ref_col] != 0
+    else:
+        ref_col = list(data_batch.keys())[0]
+        valid_mask = (data_batch[ref_col] != 0.0).long().cumsum(dim=1) > 0
+
+    device = valid_mask.device
+
+    # Calculate exact number of tokens to mask per sequence based on valid length
+    masking_prob = config.training_spec.bert_spec.masking_probability
+    budgets = (valid_mask.sum(dim=1) * masking_prob).long()
+
+    bert_mask = torch.zeros((batch_size, seq_len), dtype=torch.bool, device=device)
+
+    # 2. Pre-sample lengths and start positions to avoid massive GPU-CPU sync overhead
+    # We sample more spans than needed to account for overlaps
+    max_spans = int(seq_len * masking_prob) + 10
+
+    sampled_lengths = config.training_spec.bert_spec.span_masking.sample(
+        (batch_size, max_spans), device=device
+    )
+    sampled_starts_pct = torch.rand((batch_size, max_spans), device=device)
+
+    # 3. Span Masking Loop
+    for i in range(batch_size):
+        budget = budgets[i].item()
+        valid_len = valid_mask[i].sum().item()
+
+        if budget == 0 or valid_len == 0:
+            continue
+
+        current_masked = 0
+        span_idx = 0
+
+        # Apply spans until budget is hit
+        while current_masked < budget and span_idx < max_spans:
+            span_len = sampled_lengths[i, span_idx].item()
+            start_idx = int(sampled_starts_pct[i, span_idx].item() * valid_len)
+            end_idx = min(start_idx + span_len, valid_len)
+
+            span_view = bert_mask[i, start_idx:end_idx]
+            newly_masked = (~span_view).sum().item()
+
+            # Truncate the span if adding it exceeds our exact masking budget
+            if current_masked + newly_masked > budget:
+                allowance = budget - current_masked
+                unmasked_indices = (~span_view).nonzero(as_tuple=True)[0]
+                if len(unmasked_indices) > allowance:
+                    end_idx = start_idx + unmasked_indices[allowance - 1].item() + 1
+
+            bert_mask[i, start_idx:end_idx] = True
+            current_masked = bert_mask[i].sum().item()
+            span_idx += 1
+
+        # Fallback: if heavy overlaps exhausted our max_spans, uniformly mask the remainder
+        if current_masked < budget:
+            remaining = budget - current_masked
+            unmasked_valid = (valid_mask[i] & ~bert_mask[i]).nonzero(as_tuple=True)[0]
+            if len(unmasked_valid) > 0:
+                idx = torch.randperm(len(unmasked_valid), device=device)[:remaining]
+                bert_mask[i, unmasked_valid[idx]] = True
+
+    # 4. Create the replacement split masks (e.g., 80% Mask, 10% Random, 10% Identical)
+    replacement_probs = torch.rand((batch_size, seq_len), device=device)
+
+    p_masked = config.training_spec.bert_spec.replacement_distribution.masked
+    p_random = config.training_spec.bert_spec.replacement_distribution.random
+
+    mask_token_mask = bert_mask & (replacement_probs < p_masked)
+    random_token_mask = (
+        bert_mask
+        & (replacement_probs >= p_masked)
+        & (replacement_probs < (p_masked + p_random))
+    )
+
+    # 5. Apply corruption to data_batch
+    for col, tensor in data_batch.items():
+        if col in config.categorical_columns:
+            mask_id = 2
+
+            random_tokens = torch.randint(
+                low=3,
+                high=config.n_classes[col],
+                size=(batch_size, seq_len),
+                device=device,
+                dtype=tensor.dtype,
+            )
+
+            tensor[mask_token_mask] = mask_id
+            tensor[random_token_mask] = random_tokens[random_token_mask]
+
+        elif col in config.real_columns:
+            mask_val = 0.0
+            random_noise = torch.randn(
+                (batch_size, seq_len), device=device, dtype=tensor.dtype
+            )
+
+            tensor[mask_token_mask] = mask_val
+            tensor[random_token_mask] = random_noise[random_token_mask]
+
+    # 6. Append the explicit prediction mask
+    targets_batch["_bert_mask"] = bert_mask
+
+    return data_batch, targets_batch
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index e958f78c..5e148c51 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -54,13 +54,14 @@
 
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
-from sequifier.helpers import normalize_path  # noqa: E402
 from sequifier.helpers import (  # noqa: E402
+    apply_bert_masking,
     conditional_beartype,
     configure_determinism,
     configure_logger,
     construct_index_maps,
     get_torch_dtype,
+    normalize_path,
 )
 from sequifier.io.sequifier_dataset_from_file import (  # noqa: E402
     SequifierDatasetFromFile,
@@ -709,11 +710,20 @@ def __init__(
         self.batch_size = hparams.training_spec.batch_size
         self.accumulation_steps = hparams.training_spec.accumulation_steps
 
-        self.register_buffer(
-            "src_mask",
-            self._generate_square_subsequent_mask(self.seq_length),
-            persistent=False,  # Optional: prevents the mask from being saved in your checkpoints
-        )
+        if hparams.training_spec.training_objective == "causal":
+            self.register_buffer(
+                "src_mask",
+                self._generate_square_subsequent_mask(self.seq_length),
+                persistent=False,  # Optional: prevents the mask from being saved in your checkpoints
+            )
+        elif hparams.training_spec.training_objective == "bert":
+            self.register_buffer(
+                "src_mask",
+                torch.zeros(self.seq_length, self.seq_length),
+                persistent=False,
+            )
+        else:
+            pass
 
         self._init_weights()
 
@@ -1432,6 +1442,8 @@ def _train_epoch(
                     for k, v in targets.items()
                     if k in self.target_column_types
                 }
+                if self.hparams.training_spec.training_objective == "bert":
+                    data, targets = apply_bert_masking(data, targets, self.hparams)
 
                 # Only use standard torch.autocast if FSDP MixedPrecision is NOT handling it natively
                 if (
@@ -1613,10 +1625,13 @@ def _calculate_loss(
         else:
             seq_mask_2d = targets[mask_col] != 0
 
+        if "_bert_mask" in targets:
+            seq_mask_2d = seq_mask_2d & targets["_bert_mask"]
+
         mask = seq_mask_2d.T.contiguous().reshape(-1)
 
         losses = {}
-        for target_column in targets.keys():
+        for target_column in [k for k in targets.keys() if k != "_bert_mask"]:
             target_column_type = self.target_column_types[target_column]
             if target_column_type == "categorical":
                 output[target_column] = (
@@ -1755,6 +1770,8 @@ def _evaluate(
                     for k, v in targets.items()
                     if k in self.target_column_types
                 }
+                if self.hparams.training_spec.training_objective == "bert":
+                    data, targets = apply_bert_masking(data, targets, self.hparams)
 
                 if (
                     self.hparams.training_spec.layer_autocast
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 540c8e1f..52bd0c7c 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -26,7 +26,11 @@ def test_predictions_real(predictions):
 
 
 def test_predictions_cat(predictions):
-    valid_values = [str(x) for x in np.arange(100, 130)] + ["unknown", "other", "mask"]
+    valid_values = [str(x) for x in np.arange(100, 130)] + [
+        "[unknown]",
+        "[other]",
+        "[mask]",
+    ]
     for model_name, model_predictions in predictions.items():
         if "categorical" in model_name or "multitarget" in model_name:
             assert np.all(
@@ -40,9 +44,9 @@ def test_predictions_cat(predictions):
 
             if "multitarget" in model_name:
                 admssible_vals = [str(x) for x in np.arange(0, 10)] + [
-                    "unknown",
-                    "other",
-                    "mask",
+                    "[unknown]",
+                    "[other]",
+                    "[mask]",
                 ]
                 assert np.all(
                     [
@@ -103,9 +107,9 @@ def test_multi_pred(predictions):
         assert preds.shape[1] == 5, f"{model_name} should have 5 columns"
 
         admssible_vals = [str(x) for x in np.arange(0, 10)] + [
-            "unknown",
-            "other",
-            "mask",
+            "[unknown]",
+            "[other]",
+            "[mask]",
         ]
 
         assert np.all(
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index fac968d1..6ad3b56c 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -9,7 +9,7 @@
 
 
 def test_construct_index_maps_string():
-    """Tests reversing a string-to-int map, ensuring 0 maps to 'unknown'."""
+    """Tests reversing a string-to-int map, ensuring 0 maps to '[unknown]'."""
     id_maps: dict[str, dict[str | int, int]] = {"itemId": {"apple": 3, "banana": 4}}
     target_cols = ["itemId"]
 
@@ -20,9 +20,9 @@ def test_construct_index_maps_string():
     assert result["itemId"][4] == "banana"
 
     # Check the special 0 index for strings
-    assert result["itemId"][0] == "unknown"
-    assert result["itemId"][1] == "other"
-    assert result["itemId"][2] == "mask"
+    assert result["itemId"][0] == "[unknown]"
+    assert result["itemId"][1] == "[other]"
+    assert result["itemId"][2] == "[mask]"
 
 
 def test_construct_index_maps_integer():

From 3a179be8d5bb2ce63f3607a50569b72006f6aa00 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 4 Jun 2026 13:37:46 +0200
Subject: [PATCH 05/81] Add curly brackets

---
 src/sequifier/config/infer_config.py          |  16 +-
 src/sequifier/helpers.py                      |   4 +-
 src/sequifier/infer.py                        |   9 +-
 src/sequifier/make.py                         |   1 +
 ...cal-multitarget-5-best-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-6-predictions.csv |   2 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...custom-eval-run-0-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-1-predictions.csv |  58 +--
 ...custom-eval-run-3-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-7-predictions.csv |  60 +--
 ...l-categorical-1-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-3-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-5-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-7-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-3-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-5-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-7-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-3-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-5-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-7-probabilities.csv |   2 +-
 ...-categorical-50-best-3-0-probabilities.csv |   2 +-
 ...-categorical-50-best-3-1-probabilities.csv |   2 +-
 ...-categorical-50-best-3-2-probabilities.csv |   2 +-
 ...-categorical-50-best-3-3-probabilities.csv |   2 +-
 ...-categorical-50-best-3-4-probabilities.csv |   2 +-
 ...-categorical-50-best-3-5-probabilities.csv |   2 +-
 ...-categorical-50-best-3-6-probabilities.csv |   2 +-
 ...-categorical-50-best-3-7-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   2 +-
 tests/unit/test_infer.py                      |   2 +
 112 files changed, 1500 insertions(+), 1478 deletions(-)

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 8377fb9e..be60da2c 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -111,6 +111,7 @@ class InfererModel(BaseModel):
     metadata_config_path: str
     model_path: Union[str, list[str]]
     model_type: str
+    training_objective: str = "causal"
     data_path: str
     training_config_path: str = Field(default="configs/train.yaml")
     read_format: str = Field(default="parquet")
@@ -129,7 +130,7 @@ class InfererModel(BaseModel):
     seed: int
     device: str
     seq_length: int
-    prediction_length: int = Field(default=1)
+    prediction_length: Optional[int] = None
     inference_batch_size: int
 
     sample_from_distribution_columns: Optional[list[str]] = Field(default=None)
@@ -137,6 +138,13 @@ class InfererModel(BaseModel):
     autoregression: bool = Field(default=False)
     autoregression_total_steps: Optional[int] = Field(default=None)
 
+    @field_validator("training_objective")
+    @classmethod
+    def validate_training_objective(cls, v):
+        if v not in ["causal", "bert"]:
+            raise ValueError(f"Only 'causal' and 'bert' are allowed, found {v}")
+        return v
+
     @field_validator("model_type")
     @classmethod
     def validate_model_type(cls, v: str) -> str:
@@ -197,7 +205,11 @@ def validate_autoregression_total_steps(
     def validate_autoregression(cls, v: bool, info: ValidationInfo):
         if v and info.data.get("model_type") == "embedding":
             raise ValueError("Autoregression is not possible for embedding models")
-        if v and info.data.get("prediction_length") > 1:
+        if (
+            v
+            and info.data.get("prediction_length") is not None
+            and info.data.get("prediction_length") > 1
+        ):
             raise ValueError(
                 "Autoregressive inference is not possible for models with prediction_length > 1"
             )
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index c57606eb..d3e57f9e 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -515,13 +515,13 @@ def get_last_training_batch_timedelta(
 
 def apply_bert_masking(
     data_batch: Dict[str, torch.Tensor],
-    targets_batch: Dict[str, torch.Tensor],
     config: Any,  # TrainConfig
 ) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
     """
     Applies BERT-style span corruption to the input data using custom distributions.
     Explicitly passes the boolean prediction mask via the targets dictionary.
     """
+    targets_batch = {k: tensor.clone().detach() for k, tensor in data_batch.items()}
     batch_size, seq_len = config.training_spec.batch_size, config.seq_length
 
     # 1. Identify valid tokens (Renamed from padding_mask to valid_mask for clarity)
@@ -554,7 +554,7 @@ def apply_bert_masking(
         budget = budgets[i].item()
         valid_len = valid_mask[i].sum().item()
 
-        if budget == 0 or valid_len == 0:
+        if budget < 1 or valid_len < 1:
             continue
 
         current_masked = 0
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index dc08ec9b..d87b77d4 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -213,6 +213,13 @@ def infer_worker(
                 f"Unsupported input type or read format: {config.read_format}"
             )
 
+        default_prediction_length = {"causal": 1, "bert": config.seq_length}
+        prediction_length = (
+            config.prediction_length
+            if config.prediction_length is not None
+            else default_prediction_length[config.training_objective]
+        )
+
         inferer = Inferer(
             config.model_type,
             model_path,
@@ -227,7 +234,7 @@ def infer_worker(
             config.target_column_types,
             config.sample_from_distribution_columns,
             config.infer_with_dropout,
-            config.prediction_length,
+            prediction_length,
             config.inference_batch_size,
             config.device,
             args_config=args_config,
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 6858053f..9ebf9383 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -82,6 +82,7 @@
 target_column_types:
   EXAMPLE_TARGET_COLUMN_NAME: real
 
+training_objective: causal
 output_probabilities: false
 map_to_id: true
 device: cpu
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
index 741217c7..6c270219 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,mask,9,-0.12064132
+2,8,[mask],9,-0.12064132
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
index ded1b8f0..edf48cd7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,mask,7,-0.03440542
+8,8,[mask],7,-0.03440542
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index 08aef492..314cd804 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.22103521
-0,22,0.22123677
-0,23,0.31408396
-0,24,0.17531237
-0,25,0.18030415
-0,26,0.23421541
-0,27,-0.01312744
-0,28,0.17032059
-0,29,0.092947945
-0,30,-0.12020119
-0,31,0.013141819
-0,32,-0.08226364
-0,33,-0.096739806
-0,34,-0.10522584
-0,35,0.056258347
-0,36,0.14336495
-0,37,-0.044076495
-0,38,0.014576957
-0,39,-0.008042061
-0,40,-0.053061705
-1,19,0.17930579
-1,20,0.2741497
-1,21,0.30809382
-1,22,-0.013938604
-1,23,0.25418255
-1,24,0.053762455
-1,25,0.017072849
-1,26,-0.13018475
-1,27,-0.06404363
-1,28,0.014452162
-1,29,0.053762455
-1,30,0.038537517
-1,31,0.10343069
-1,32,-0.03608964
-1,33,-0.19008616
-1,34,0.084961094
-1,35,0.053512864
-1,36,-0.0055773673
-1,37,-0.0017945322
-1,38,-0.14016832
-2,17,0.43189004
-2,18,-0.035340875
-2,19,0.018819973
-2,20,0.114911795
-2,21,0.036041625
-2,22,-0.04806992
-2,23,0.029552305
-2,24,0.13837317
-2,25,-0.13717325
-2,26,0.12938796
-2,27,0.033046555
-2,28,0.026307646
-2,29,-0.083262
-2,30,0.28413326
-2,31,-0.034342516
-2,32,0.19128607
-2,33,-0.009321204
-2,34,-0.08575789
-2,35,0.17730908
-2,36,0.03778875
-3,17,0.059503004
-3,18,0.22423184
-3,19,0.5217421
-3,20,0.15235017
-3,21,0.15933867
-3,22,0.11441261
-3,23,0.0342945
-3,24,0.067489855
-3,25,0.2481924
-3,26,-0.11620776
-3,27,-0.10372831
-3,28,0.069486566
-3,29,0.042031765
-3,30,0.09594302
-3,31,0.01657367
-3,32,-0.1661256
-3,33,-0.021613471
-3,34,0.16033702
-3,35,0.06599232
-3,36,-0.02435895
-4,14,0.07847178
-4,15,0.13637646
-4,16,-0.03272019
-4,17,0.069486566
-4,18,0.062248483
-4,19,-0.098736525
-4,20,0.052264918
-4,21,0.16333209
-4,22,-0.1521486
-4,23,-0.15714039
-4,24,0.15534523
-4,25,-0.013751412
-4,26,0.04602519
-4,27,0.047522724
-4,28,-0.16512723
-4,29,-0.27694318
-4,30,-0.16113381
-4,31,-0.0625461
-4,32,0.04377889
-4,33,-0.08725542
-5,14,0.06798903
-5,15,0.42190647
-5,16,0.22622855
-5,17,0.18729265
-5,18,0.34203795
-5,19,0.41591632
-5,20,-0.03633923
-5,21,0.04677396
-5,22,-0.016871277
-5,23,0.033795323
-5,24,-0.19507794
-5,25,0.17830744
-5,26,-0.025107719
-5,27,0.07198246
-5,28,0.008586817
-5,29,0.07647506
-5,30,-0.1661256
-5,31,0.28213653
-5,32,-0.17411245
-5,33,-0.080766104
-6,18,0.45385388
-6,19,0.18529594
-6,20,0.36000836
-6,21,0.073479995
-6,22,0.22822526
-6,23,-0.06678911
-6,24,0.22722691
-6,25,0.06798903
-6,26,-0.029101145
-6,27,-0.019616757
-6,28,0.069486566
-6,29,0.045526013
-6,30,-0.21804014
-6,31,0.041532584
-6,32,0.06449478
-6,33,0.04951944
-6,34,-0.10522584
-6,35,-0.0037522472
-6,36,-0.16912067
-6,37,-0.06579076
-7,15,0.20526306
-7,16,0.2781431
-7,17,0.02693162
-7,18,0.042031765
-7,19,-0.023610184
-7,20,0.11591015
-7,21,-0.055557594
-7,22,0.07198246
-7,23,-0.18010259
-7,24,-0.1177053
-7,25,-0.08026692
-7,26,-0.0762735
-7,27,-0.07976775
-7,28,-0.16013545
-7,29,-0.01774484
-7,30,-0.08575789
-7,31,-0.0047038053
-7,32,-0.046821974
-7,33,-0.065291576
-7,34,0.2322187
-8,20,0.11291508
-8,21,0.025683673
-8,22,0.009335584
-8,23,0.08296438
-8,24,-0.038835123
-8,25,0.02693162
-8,26,0.08995288
-8,27,0.08795616
-8,28,-0.10722256
-8,29,-0.04282855
-8,30,-0.15614203
-8,31,-0.05156417
-8,32,-0.1431634
-8,33,-0.015685728
-8,34,-0.22802371
-8,35,-0.007917265
-8,36,-0.071281716
-8,37,-0.063294865
-8,38,-0.31687742
-8,39,0.062498074
-9,16,0.43987688
-9,17,0.43388674
-9,18,0.34203795
-9,19,0.22423184
-9,20,0.11740769
-9,21,0.15135181
-9,22,0.42789662
-9,23,-0.12269708
-9,24,-0.13916996
-9,25,0.13038632
-9,26,0.022439014
-9,27,0.1653288
-9,28,-0.047570743
-9,29,-0.082762815
-9,30,0.026557237
-9,31,0.10992001
-9,32,0.07597589
-9,33,-0.11371187
-9,34,-0.14615846
-9,35,0.09843891
+0,21,0.045526013
+0,22,0.20925649
+0,23,0.28213653
+0,24,0.33005765
+0,25,0.3220708
+0,26,0.13737482
+0,27,0.20725977
+0,28,0.09843891
+0,29,-0.060299788
+0,30,-0.003611853
+0,31,0.045026835
+0,32,0.029302716
+0,33,-0.05430965
+0,34,-0.06728829
+0,35,0.009834763
+0,36,0.23521377
+0,37,0.08845534
+0,38,0.040534228
+0,39,0.050268207
+0,40,-0.2969103
+1,19,0.25617927
+1,20,0.21324992
+1,21,0.38596562
+1,22,0.37797877
+1,23,0.07298081
+1,24,-0.05381047
+1,25,0.03778875
+1,26,0.098938085
+1,27,-0.020615114
+1,28,-0.037088
+1,29,0.020941481
+1,30,-0.024234157
+1,31,0.1423666
+1,32,0.09594302
+1,33,0.07198246
+1,34,-0.112713516
+1,35,0.042281352
+1,36,-0.053560883
+1,37,-0.032595392
+1,38,0.102931514
+2,17,0.38596562
+2,18,-0.12868722
+2,19,0.10043562
+2,20,0.014514559
+2,21,0.15434688
+2,22,0.08995288
+2,23,0.14536166
+2,24,0.06599232
+2,25,0.15235017
+2,26,0.0074636657
+2,27,0.047772314
+2,28,0.24220227
+2,29,-0.00005520758
+2,30,0.07148328
+2,31,-0.028352378
+2,32,-0.10422748
+2,33,-0.070782535
+2,34,0.0664915
+2,35,-0.12369544
+2,36,-0.27095303
+3,17,0.13038632
+3,18,0.14036989
+3,19,0.37398535
+3,20,-0.042578958
+3,21,0.34603137
+3,22,0.26216942
+3,23,-0.1177053
+3,24,0.25617927
+3,25,-0.029849913
+3,26,-0.0600502
+3,27,0.10343069
+3,28,0.07048492
+3,29,-0.07228007
+3,30,-0.015623331
+3,31,-0.0010243154
+3,32,0.12589371
+3,33,0.19228444
+3,34,0.086957805
+3,35,0.003563835
+3,36,-0.004672607
+4,14,0.13038632
+4,15,0.04877067
+4,16,0.0894537
+4,17,0.054261632
+4,18,0.18928936
+4,19,0.053263277
+4,20,-0.10322913
+4,21,-0.14116667
+4,22,-0.25897273
+4,23,-0.039583888
+4,24,-0.068785824
+4,25,-0.112713516
+4,26,0.09943727
+4,27,-0.0018198809
+4,28,0.14036989
+4,29,-0.15813874
+4,30,0.09494466
+4,31,-0.07976775
+4,32,0.24719404
+4,33,0.0709841
+5,14,0.23321705
+5,15,0.37198862
+5,16,0.21524663
+5,17,0.26416612
+5,18,0.40193933
+5,19,0.07697424
+5,20,0.18729265
+5,21,0.22722691
+5,22,-0.043577317
+5,23,0.0550104
+5,24,-0.08925214
+5,25,0.00047516936
+5,26,-0.09324556
+5,27,-0.14216504
+5,28,-0.09973488
+5,29,-0.09374474
+5,30,-0.043826904
+5,31,-0.14216504
+5,32,0.04278053
+5,33,-0.17211573
+6,18,0.4438703
+6,19,0.022938194
+6,20,0.23022199
+6,21,0.4119229
+6,22,0.19927293
+6,23,0.12539454
+6,24,-0.11371187
+6,25,-0.021613471
+6,26,0.16033702
+6,27,0.23022199
+6,28,-0.1501519
+6,29,-0.058802254
+6,30,-0.028976351
+6,31,-0.1591371
+6,32,-0.106224194
+6,33,-0.022861416
+6,34,-0.03384334
+6,35,0.09344713
+6,36,-0.021987854
+6,37,-0.14016832
+7,15,0.12339783
+7,16,0.3041004
+7,17,-0.0067005185
+7,18,0.00015538326
+7,19,-0.09374474
+7,20,0.07198246
+7,21,0.034544088
+7,22,0.23321705
+7,23,-0.06778747
+7,24,0.0054045552
+7,25,0.020067917
+7,26,-0.032221008
+7,27,0.033545732
+7,28,-0.04806992
+7,29,-0.034342516
+7,30,-0.13417818
+7,31,-0.094743095
+7,32,-0.118204474
+7,33,0.16233373
+7,34,-0.09524228
+8,20,0.18329921
+8,21,0.116409324
+8,22,0.2921201
+8,23,0.0100843515
+8,24,-0.052812114
+8,25,0.07198246
+8,26,0.08595945
+8,27,-0.26096946
+8,28,-0.00040619232
+8,29,0.09993644
+8,30,-0.312884
+8,31,-0.12070037
+8,32,-0.13417818
+8,33,-0.18709108
+8,34,0.06898739
+8,35,0.010333941
+8,36,-0.00027359813
+8,37,0.011956271
+8,38,-0.12519297
+8,39,0.086957805
+9,16,0.25817597
+9,17,0.3939525
+9,18,0.12489536
+9,19,0.20126964
+9,20,0.40393606
+9,21,0.27215296
+9,22,0.14835674
+9,23,-0.006326135
+9,24,-0.14715682
+9,25,-0.058053486
+9,26,-0.058552664
+9,27,0.22023842
+9,28,-0.00056998525
+9,29,0.024810111
+9,30,0.10992001
+9,31,0.010833119
+9,32,0.038287926
+9,33,0.020567097
+9,34,0.12739125
+9,35,-0.18709108
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
index 39c6e464..923104fa 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,102,6,3,8,1
-0,9,121,6,3,8,1
-0,10,104,6,3,8,1
-0,11,121,6,3,8,1
-0,12,121,6,3,8,1
-0,13,104,6,3,8,0
-0,14,104,3,3,0,0
-0,15,104,3,3,0,0
-0,16,104,3,3,0,0
-0,17,104,3,3,0,0
-0,18,104,3,3,0,0
-0,19,104,3,3,0,0
-0,20,104,3,3,0,1
-0,21,103,6,3,0,1
-0,22,113,6,3,8,1
-0,23,113,6,3,8,1
-0,24,113,6,3,8,1
-0,25,104,6,3,8,1
-0,26,104,6,3,8,1
-0,27,104,6,3,8,1
-0,28,104,6,3,8,1
-0,29,104,6,3,8,1
-0,30,104,9,3,0,0
-0,31,104,3,3,0,1
-0,32,104,9,3,0,1
-0,33,104,9,3,0,1
-0,34,104,3,3,8,1
-0,35,104,6,3,8,0
-0,36,121,6,3,8,0
-0,37,121,6,3,8,0
+0,8,121,4,9,8,1
+0,9,121,4,3,8,0
+0,10,121,4,3,6,1
+0,11,121,6,3,6,1
+0,12,121,6,3,6,1
+0,13,121,6,3,6,1
+0,14,121,4,3,6,1
+0,15,121,4,3,6,1
+0,16,121,4,3,6,1
+0,17,121,4,3,6,1
+0,18,121,4,3,8,1
+0,19,121,4,5,6,1
+0,20,121,4,3,8,1
+0,21,121,4,3,6,1
+0,22,121,4,3,6,0
+0,23,121,4,3,8,0
+0,24,121,4,3,6,0
+0,25,121,4,3,6,0
+0,26,121,4,3,6,0
+0,27,121,4,3,6,0
+0,28,121,4,3,6,0
+0,29,121,4,3,6,0
+0,30,121,4,3,6,0
+0,31,121,4,3,6,0
+0,32,121,4,3,6,1
+0,33,121,4,3,6,0
+0,34,121,4,3,6,1
+0,35,121,4,3,6,0
+0,36,121,4,3,6,1
+0,37,121,4,3,6,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
index b1f5c970..aa4c0f4a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,6,3,8,1
-1,9,121,6,3,8,0
-1,10,121,6,3,8,0
-1,11,121,6,3,8,0
-1,12,121,6,3,8,0
-1,13,104,3,3,8,0
-1,14,104,3,3,0,0
-1,15,104,3,3,0,0
-1,16,104,3,3,0,0
-1,17,104,3,3,0,0
-1,18,104,3,3,0,0
-1,19,104,3,3,0,0
-1,20,104,3,3,0,0
-1,21,121,6,3,0,1
-1,22,113,6,3,8,1
-1,23,121,6,3,8,1
-1,24,104,6,3,8,1
-1,25,104,6,3,8,0
-1,26,104,6,3,0,0
-1,27,104,9,3,0,1
-1,28,104,3,3,0,1
-1,29,104,9,3,0,1
-1,30,104,3,3,0,1
-1,31,104,3,3,0,1
-1,32,103,6,3,8,1
-1,33,113,6,3,8,1
-1,34,121,6,3,8,1
-1,35,121,6,3,8,1
-1,36,104,6,3,8,1
-1,37,104,6,3,8,0
+1,8,103,4,9,8,1
+1,9,121,4,3,6,1
+1,10,121,4,3,6,1
+1,11,121,4,3,6,1
+1,12,121,6,3,6,1
+1,13,121,4,3,6,0
+1,14,121,4,3,6,0
+1,15,121,4,3,6,0
+1,16,121,4,3,6,0
+1,17,121,4,3,6,1
+1,18,121,4,3,6,1
+1,19,121,4,3,6,1
+1,20,121,4,3,6,1
+1,21,121,4,3,6,1
+1,22,121,4,3,6,1
+1,23,121,4,3,6,1
+1,24,121,4,3,6,1
+1,25,121,4,3,8,1
+1,26,121,4,3,6,1
+1,27,121,4,3,6,1
+1,28,121,4,3,6,1
+1,29,121,4,3,6,1
+1,30,121,4,3,6,1
+1,31,121,4,3,6,0
+1,32,121,4,3,8,1
+1,33,121,4,3,6,0
+1,34,121,4,3,8,1
+1,35,121,4,3,6,1
+1,36,121,4,3,6,0
+1,37,121,4,3,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
index b32ec715..a00ae412 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,102,6,3,8,1
-2,9,121,6,3,8,1
-2,10,104,6,3,8,1
-2,11,104,6,3,8,0
-2,12,104,6,3,8,0
-2,13,104,6,3,8,0
-2,14,104,6,3,0,0
-2,15,104,3,3,0,0
-2,16,104,3,3,0,0
-2,17,104,3,3,0,0
-2,18,104,3,3,0,1
-2,19,103,4,3,0,1
-2,20,102,6,3,8,1
-2,21,121,6,3,8,1
-2,22,104,6,3,8,1
-2,23,104,6,3,8,1
-2,24,104,6,3,8,0
-2,25,104,6,3,8,0
-2,26,104,6,3,0,0
-2,27,104,9,3,0,1
-2,28,104,3,3,0,0
-2,29,104,3,3,0,0
-2,30,104,3,3,0,1
-2,31,103,4,3,0,1
-2,32,102,6,3,8,1
-2,33,121,6,3,8,1
-2,34,121,6,3,8,1
-2,35,104,6,3,8,1
-2,36,104,6,3,8,0
-2,37,104,6,3,0,0
+2,8,103,4,9,8,1
+2,9,103,4,9,5,1
+2,10,103,4,3,6,0
+2,11,121,4,3,6,0
+2,12,121,4,3,6,0
+2,13,121,6,3,6,0
+2,14,121,4,3,6,0
+2,15,121,4,3,6,0
+2,16,121,4,3,6,0
+2,17,121,4,3,6,0
+2,18,121,4,3,6,0
+2,19,121,4,3,6,1
+2,20,121,4,3,6,1
+2,21,121,4,3,6,1
+2,22,121,4,3,6,0
+2,23,121,4,3,6,1
+2,24,121,4,3,6,1
+2,25,121,4,3,6,1
+2,26,121,4,3,6,1
+2,27,121,4,3,6,1
+2,28,121,4,3,6,1
+2,29,121,4,3,6,1
+2,30,121,4,3,6,1
+2,31,121,4,3,8,1
+2,32,121,4,3,6,1
+2,33,121,4,3,6,1
+2,34,121,4,3,6,1
+2,35,121,4,3,6,1
+2,36,121,4,3,6,1
+2,37,121,4,3,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
index 03dc6b54..48a9f496 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,6,3,8,1
-3,9,104,4,3,8,0
-3,10,121,3,3,8,0
-3,11,104,3,3,8,0
-3,12,121,3,3,8,0
-3,13,121,3,3,8,0
-3,14,104,3,3,8,0
-3,15,104,3,3,0,0
-3,16,104,3,3,0,0
-3,17,104,3,3,0,0
-3,18,104,3,3,0,0
-3,19,104,3,3,0,0
-3,20,104,3,3,0,0
-3,21,104,3,3,0,0
-3,22,121,6,3,0,1
-3,23,113,6,3,8,1
-3,24,121,6,3,8,1
-3,25,104,6,3,8,1
-3,26,104,6,3,8,0
-3,27,104,6,3,0,0
-3,28,104,9,3,0,1
-3,29,104,3,3,0,1
-3,30,104,9,3,0,1
-3,31,104,3,3,0,1
-3,32,104,3,3,0,1
-3,33,103,6,3,8,1
-3,34,113,6,3,8,1
-3,35,121,6,3,8,1
-3,36,121,6,3,8,1
-3,37,104,6,3,8,1
-4,8,102,6,9,8,1
-4,9,121,6,3,8,1
-4,10,121,6,3,8,1
-4,11,121,6,3,8,0
-4,12,121,6,3,8,0
-4,13,121,6,3,8,0
-4,14,104,3,3,8,0
-4,15,104,3,3,0,0
-4,16,104,3,3,0,0
-4,17,104,3,3,0,0
-4,18,104,3,3,0,0
-4,19,104,3,3,0,0
-4,20,104,3,3,0,0
-4,21,104,3,3,0,0
-4,22,121,6,3,0,1
-4,23,113,6,3,8,1
-4,24,121,6,3,8,1
-4,25,104,6,3,8,1
-4,26,104,6,3,8,0
-4,27,104,6,3,0,0
-4,28,104,9,3,0,1
-4,29,104,3,3,0,1
-4,30,104,9,3,0,1
-4,31,104,3,3,0,1
-4,32,104,3,3,0,1
-4,33,103,6,3,8,1
-4,34,113,6,3,8,1
-4,35,121,6,3,8,1
-4,36,121,6,3,8,1
-4,37,104,6,3,8,1
+3,8,103,4,9,8,1
+3,9,103,4,9,8,1
+3,10,121,4,3,6,1
+3,11,121,4,3,6,0
+3,12,121,4,3,6,1
+3,13,121,6,3,6,1
+3,14,121,4,3,6,1
+3,15,121,6,3,6,0
+3,16,121,4,3,6,0
+3,17,121,4,3,6,0
+3,18,121,4,3,6,1
+3,19,121,4,3,6,1
+3,20,121,4,3,6,1
+3,21,121,4,3,6,1
+3,22,121,4,3,6,1
+3,23,121,4,3,6,1
+3,24,121,4,3,6,1
+3,25,121,4,3,6,1
+3,26,121,4,3,8,1
+3,27,121,4,3,6,1
+3,28,121,4,3,6,1
+3,29,121,4,3,6,1
+3,30,121,4,3,6,1
+3,31,121,4,3,6,1
+3,32,121,4,3,6,0
+3,33,121,4,3,8,1
+3,34,121,4,3,6,0
+3,35,121,4,3,8,1
+3,36,121,4,3,6,1
+3,37,121,4,3,6,0
+4,8,103,4,9,8,1
+4,9,103,4,9,8,1
+4,10,121,4,3,6,1
+4,11,121,4,3,6,0
+4,12,121,4,3,6,1
+4,13,121,6,3,6,0
+4,14,121,4,3,6,1
+4,15,121,6,3,6,0
+4,16,121,4,3,6,0
+4,17,121,4,3,6,0
+4,18,121,4,3,6,1
+4,19,121,4,3,6,1
+4,20,121,4,3,6,1
+4,21,121,4,3,6,1
+4,22,121,4,3,6,1
+4,23,121,4,3,6,1
+4,24,121,4,3,6,1
+4,25,121,4,3,6,1
+4,26,121,4,3,8,1
+4,27,121,4,3,6,1
+4,28,121,4,3,6,1
+4,29,121,4,3,6,1
+4,30,121,4,3,6,1
+4,31,121,4,3,6,1
+4,32,121,4,3,6,0
+4,33,121,4,3,8,1
+4,34,121,4,3,6,0
+4,35,121,4,3,8,1
+4,36,121,4,3,6,1
+4,37,121,4,3,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
index 549102b2..551fdd40 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,4,9,8,1
-5,9,113,6,3,8,0
-5,10,104,6,3,8,0
-5,11,104,6,3,0,0
-5,12,104,9,3,0,0
-5,13,121,3,3,8,1
-5,14,121,6,3,8,0
-5,15,104,3,3,8,0
-5,16,104,3,3,0,0
-5,17,104,3,3,0,0
-5,18,104,3,3,0,0
-5,19,121,3,3,0,0
-5,20,104,3,3,0,0
-5,21,104,3,3,0,0
-5,22,104,3,3,0,0
-5,23,121,4,3,0,1
-5,24,121,6,3,8,1
-5,25,104,6,3,8,1
-5,26,104,6,3,8,0
-5,27,104,3,3,0,0
-5,28,104,3,3,0,1
-5,29,104,4,3,0,1
-5,30,104,9,3,0,1
-5,31,104,3,3,0,1
-5,32,104,3,3,0,1
-5,33,104,6,3,0,1
-5,34,113,6,3,8,1
-5,35,113,6,3,8,1
-5,36,113,6,3,8,1
-5,37,113,6,3,8,1
+5,8,103,3,9,8,1
+5,9,103,4,3,8,1
+5,10,121,4,3,6,1
+5,11,121,4,3,6,0
+5,12,121,4,3,6,1
+5,13,121,6,3,6,1
+5,14,121,6,3,6,1
+5,15,121,4,3,6,1
+5,16,121,4,3,6,0
+5,17,121,4,3,6,1
+5,18,121,4,3,6,1
+5,19,121,4,3,6,1
+5,20,121,4,3,6,1
+5,21,121,4,3,6,1
+5,22,121,4,3,6,1
+5,23,121,4,3,6,1
+5,24,121,4,3,6,1
+5,25,121,4,3,8,1
+5,26,121,4,3,6,1
+5,27,121,4,3,6,1
+5,28,121,4,3,6,1
+5,29,121,4,3,6,1
+5,30,121,4,3,6,1
+5,31,121,4,3,6,0
+5,32,121,4,3,8,1
+5,33,121,4,3,6,0
+5,34,121,4,3,8,1
+5,35,121,4,3,6,1
+5,36,121,4,3,6,0
+5,37,121,4,3,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
index 68e5ec32..d875f86b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,102,6,2,8,1
-6,9,113,6,3,8,1
-6,10,121,6,3,8,1
-6,11,121,6,3,8,1
-6,12,121,6,3,8,1
-6,13,121,6,3,8,0
-6,14,121,6,3,8,0
-6,15,104,3,3,8,0
-6,16,104,3,3,0,0
-6,17,104,3,3,0,0
-6,18,104,3,3,0,0
-6,19,104,3,3,0,0
-6,20,104,3,3,0,0
-6,21,104,3,3,0,0
-6,22,104,3,3,0,0
-6,23,121,6,3,0,1
-6,24,113,6,3,8,1
-6,25,121,6,3,8,1
-6,26,104,6,3,8,1
-6,27,104,6,3,8,0
-6,28,104,6,3,0,0
-6,29,104,9,3,0,1
-6,30,104,3,3,0,1
-6,31,104,9,3,0,1
-6,32,104,3,3,0,1
-6,33,104,3,3,0,1
-6,34,103,6,3,8,1
-6,35,113,6,3,8,1
-6,36,121,6,3,8,1
-6,37,121,6,3,8,1
-7,8,121,6,9,8,1
-7,9,104,6,3,8,1
-7,10,104,6,3,8,0
-7,11,104,6,3,0,0
-7,12,104,6,3,0,1
-7,13,121,6,3,8,1
-7,14,104,6,3,8,0
-7,15,104,3,3,0,0
-7,16,104,3,3,0,0
-7,17,104,3,3,0,1
-7,18,104,4,3,0,1
-7,19,103,6,3,8,1
-7,20,104,6,3,8,1
-7,21,104,9,3,0,0
-7,22,121,6,3,8,1
-7,23,104,6,3,8,1
-7,24,104,6,3,8,1
-7,25,104,6,3,8,1
-7,26,104,6,3,8,0
-7,27,104,3,3,0,0
-7,28,104,3,3,0,0
-7,29,104,3,3,0,0
-7,30,104,3,3,0,0
-7,31,104,3,3,0,1
-7,32,104,3,3,0,1
-7,33,103,4,3,0,1
-7,34,113,6,3,8,1
-7,35,113,6,3,8,1
-7,36,113,6,3,8,1
-7,37,104,6,3,8,1
+6,8,121,4,9,8,1
+6,9,121,4,9,8,1
+6,10,121,4,3,6,1
+6,11,121,6,3,6,1
+6,12,121,6,3,6,1
+6,13,121,4,3,6,1
+6,14,121,4,3,6,1
+6,15,121,4,3,6,1
+6,16,121,4,3,6,0
+6,17,121,4,3,8,1
+6,18,121,4,3,6,1
+6,19,121,4,3,6,1
+6,20,121,4,3,6,1
+6,21,121,4,3,6,0
+6,22,121,4,3,6,1
+6,23,121,4,3,6,0
+6,24,121,4,3,6,0
+6,25,121,4,3,6,0
+6,26,121,4,3,6,1
+6,27,121,4,3,6,1
+6,28,121,4,3,6,1
+6,29,121,4,3,6,1
+6,30,121,4,3,6,1
+6,31,121,4,3,6,1
+6,32,121,4,3,6,1
+6,33,121,4,3,6,1
+6,34,121,4,3,8,1
+6,35,121,4,3,6,1
+6,36,121,4,3,6,1
+6,37,121,4,3,6,1
+7,8,121,4,9,8,1
+7,9,103,4,9,5,1
+7,10,121,4,3,6,1
+7,11,121,4,3,6,1
+7,12,121,6,3,6,1
+7,13,121,6,3,6,1
+7,14,121,4,3,6,1
+7,15,121,4,3,6,1
+7,16,121,4,3,6,0
+7,17,121,4,3,6,1
+7,18,121,4,3,6,1
+7,19,121,4,3,6,1
+7,20,121,4,3,6,1
+7,21,121,4,3,6,1
+7,22,121,4,3,6,1
+7,23,121,4,3,6,1
+7,24,121,4,3,6,1
+7,25,121,4,3,8,1
+7,26,121,4,3,6,1
+7,27,121,4,3,6,1
+7,28,121,4,3,6,1
+7,29,121,4,3,6,1
+7,30,121,4,3,6,1
+7,31,121,4,3,6,0
+7,32,121,4,3,8,1
+7,33,121,4,3,6,0
+7,34,121,4,3,8,1
+7,35,121,4,3,6,1
+7,36,121,4,3,6,0
+7,37,121,4,3,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
index 84a9c524..c203f13c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,6,3,8,1
-8,9,121,6,3,8,1
-8,10,104,6,3,8,1
-8,11,104,6,3,8,1
-8,12,121,6,3,8,1
-8,13,121,6,3,8,1
-8,14,104,6,3,8,0
-8,15,104,9,3,0,0
-8,16,104,3,3,0,0
-8,17,104,3,3,0,0
-8,18,104,3,3,0,0
-8,19,104,3,3,0,0
-8,20,104,3,3,0,0
-8,21,104,3,3,0,1
-8,22,103,6,3,8,1
-8,23,121,6,3,8,1
-8,24,104,6,3,8,1
-8,25,104,6,3,8,1
-8,26,104,6,3,8,0
-8,27,104,6,3,0,0
-8,28,104,9,3,0,1
-8,29,104,9,3,0,1
-8,30,104,3,3,0,1
-8,31,104,3,3,0,1
-8,32,104,6,3,0,1
-8,33,113,6,3,8,1
-8,34,121,6,3,8,1
-8,35,121,6,3,8,1
-8,36,121,6,3,8,1
-8,37,104,6,3,8,1
+8,8,103,4,9,8,1
+8,9,121,4,3,6,1
+8,10,121,4,3,6,1
+8,11,121,6,3,6,1
+8,12,121,6,3,6,1
+8,13,121,6,3,6,1
+8,14,121,4,3,6,1
+8,15,121,4,3,6,1
+8,16,121,4,3,6,1
+8,17,121,4,3,8,1
+8,18,121,4,5,8,1
+8,19,121,4,9,6,1
+8,20,121,4,3,6,1
+8,21,121,4,3,6,1
+8,22,121,4,3,6,0
+8,23,121,4,3,6,0
+8,24,121,4,3,6,0
+8,25,121,4,3,6,0
+8,26,121,4,3,6,0
+8,27,121,4,3,6,1
+8,28,121,4,3,6,1
+8,29,121,4,3,6,1
+8,30,121,4,3,6,0
+8,31,121,4,3,6,1
+8,32,121,4,3,6,1
+8,33,121,4,3,6,1
+8,34,121,4,3,6,1
+8,35,121,4,3,6,1
+8,36,121,4,3,6,1
+8,37,121,4,3,6,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
index ad971442..629630a6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,113,6,3,0,0
-9,9,121,6,3,8,1
-9,10,104,6,3,8,1
-9,11,104,6,3,8,1
-9,12,104,6,3,8,1
-9,13,104,6,3,8,1
-9,14,104,6,3,8,0
-9,15,104,6,3,0,0
-9,16,104,9,3,0,2
-9,17,104,3,3,0,0
-9,18,104,3,3,0,1
-9,19,104,9,3,0,1
-9,20,103,6,3,8,1
-9,21,121,6,3,8,1
-9,22,104,6,3,8,1
-9,23,121,6,3,8,1
-9,24,121,6,3,8,1
-9,25,104,6,3,8,0
-9,26,104,9,3,0,0
-9,27,104,3,3,0,0
-9,28,104,3,3,0,0
-9,29,104,3,3,0,0
-9,30,104,3,3,0,0
-9,31,104,3,3,0,0
-9,32,104,3,3,0,1
-9,33,103,6,3,8,1
-9,34,121,6,3,8,1
-9,35,104,6,3,8,1
-9,36,104,6,3,8,1
-9,37,104,6,3,8,0
+9,8,103,4,9,8,1
+9,9,103,4,3,8,0
+9,10,121,4,3,7,0
+9,11,121,4,3,7,0
+9,12,121,4,3,6,0
+9,13,121,4,3,6,0
+9,14,121,4,3,6,0
+9,15,121,4,3,6,0
+9,16,121,4,3,6,0
+9,17,121,4,3,6,0
+9,18,121,4,3,6,1
+9,19,121,4,3,6,1
+9,20,121,4,3,6,0
+9,21,121,4,3,6,1
+9,22,121,4,3,6,1
+9,23,121,4,3,6,1
+9,24,121,4,3,6,1
+9,25,121,4,3,6,1
+9,26,121,4,3,6,1
+9,27,121,4,3,6,1
+9,28,121,4,3,6,1
+9,29,121,4,3,8,1
+9,30,121,4,3,6,1
+9,31,121,4,3,6,1
+9,32,121,4,3,6,1
+9,33,121,4,3,6,1
+9,34,121,4,3,6,1
+9,35,121,4,3,6,0
+9,36,121,4,3,8,1
+9,37,121,4,3,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index 4d68f099..2a8d05f6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,121,1,1,2,1
-0,9,113,0,7,8,2
-0,10,121,1,1,6,0
-0,11,113,9,0,8,1
-0,12,121,1,0,2,0
-0,13,127,4,0,7,0
-0,14,127,4,0,7,1
-0,15,127,1,0,2,1
-0,16,127,4,7,7,1
-0,17,102,1,1,7,1
-0,18,105,8,7,7,1
-0,19,105,1,7,2,1
-0,20,113,mask,7,7,1
-0,21,113,1,7,2,1
-0,22,113,1,7,2,1
-0,23,113,1,7,7,1
-0,24,113,1,5,2,1
-0,25,113,1,7,2,1
-0,26,113,1,1,7,1
-0,27,113,1,0,7,1
-0,28,113,1,0,2,1
-0,29,113,1,0,7,1
-0,30,113,1,0,7,1
-0,31,113,1,7,7,1
-0,32,113,1,7,7,1
-0,33,113,1,7,7,1
-0,34,113,1,7,7,1
-0,35,113,6,7,7,1
-0,36,113,1,7,2,1
-0,37,113,6,7,2,1
+0,8,113,6,2,0,1
+0,9,113,6,2,0,1
+0,10,113,6,2,0,1
+0,11,113,6,2,0,1
+0,12,113,6,2,0,1
+0,13,113,6,2,0,2
+0,14,113,6,2,0,2
+0,15,113,6,2,0,2
+0,16,113,6,2,0,2
+0,17,113,6,2,0,2
+0,18,113,6,2,0,2
+0,19,113,6,2,0,2
+0,20,113,6,2,0,2
+0,21,113,6,2,0,2
+0,22,113,6,2,0,2
+0,23,113,6,2,0,2
+0,24,113,6,2,0,2
+0,25,113,6,2,0,2
+0,26,113,6,2,0,2
+0,27,113,6,2,0,2
+0,28,113,6,2,0,2
+0,29,113,6,2,0,2
+0,30,113,6,2,0,2
+0,31,113,6,2,0,2
+0,32,113,6,2,0,2
+0,33,113,6,2,0,2
+0,34,113,6,2,0,2
+0,35,113,6,2,0,2
+0,36,113,6,2,0,2
+0,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index 1241eea1..fed4cb3b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,113,0,7,6,0
-1,9,121,8,9,2,1
-1,10,113,9,0,8,0
-1,11,121,3,0,2,1
-1,12,127,9,0,7,0
-1,13,121,6,0,7,1
-1,14,127,1,0,7,1
-1,15,127,1,0,7,1
-1,16,108,1,7,7,1
-1,17,105,8,7,7,1
-1,18,105,1,7,2,1
-1,19,113,2,7,2,1
-1,20,113,2,7,2,1
-1,21,113,1,7,2,1
-1,22,113,1,7,7,1
-1,23,113,1,7,7,1
-1,24,113,1,5,2,1
-1,25,113,1,0,8,1
-1,26,113,1,0,2,1
-1,27,113,1,0,2,1
-1,28,113,1,0,7,1
-1,29,108,1,0,7,1
-1,30,108,1,7,7,1
-1,31,113,1,7,7,1
-1,32,113,1,7,7,1
-1,33,113,6,7,7,1
-1,34,113,1,7,2,1
-1,35,113,6,7,7,1
-1,36,113,1,7,2,1
-1,37,113,6,7,2,1
+1,8,121,4,2,6,0
+1,9,109,4,9,6,1
+1,10,121,6,2,0,0
+1,11,121,6,2,8,0
+1,12,121,6,0,8,0
+1,13,109,6,0,8,0
+1,14,121,6,0,8,0
+1,15,109,4,9,8,0
+1,16,109,4,9,5,0
+1,17,120,6,0,9,0
+1,18,109,4,0,5,0
+1,19,109,4,0,5,0
+1,20,109,4,9,5,0
+1,21,127,4,0,9,0
+1,22,120,4,0,5,0
+1,23,127,4,0,9,0
+1,24,127,4,0,5,0
+1,25,109,4,9,5,0
+1,26,109,4,3,5,0
+1,27,109,4,3,5,0
+1,28,109,4,3,5,0
+1,29,109,4,9,6,0
+1,30,109,4,3,8,0
+1,31,109,4,9,8,0
+1,32,109,4,0,8,0
+1,33,109,4,0,8,0
+1,34,109,4,9,8,0
+1,35,109,4,0,5,0
+1,36,109,4,0,5,0
+1,37,109,4,0,5,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index 46b4436b..c4fe8a8e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,121,1,7,6,0
-2,9,113,0,7,7,2
-2,10,121,8,9,6,1
-2,11,102,1,1,6,2
-2,12,105,1,7,8,0
-2,13,127,1,1,0,1
-2,14,127,1,7,0,1
-2,15,127,1,1,0,1
-2,16,127,1,7,7,1
-2,17,127,1,7,0,1
-2,18,113,1,7,0,1
-2,19,113,1,7,0,1
-2,20,113,1,7,7,1
-2,21,113,1,7,0,1
-2,22,113,1,7,7,1
-2,23,113,1,7,0,1
-2,24,113,1,7,2,1
-2,25,113,1,7,2,1
-2,26,113,1,1,7,1
-2,27,113,1,5,2,1
-2,28,113,1,2,8,1
-2,29,113,1,1,7,1
-2,30,113,1,0,7,1
-2,31,113,1,0,7,1
-2,32,113,1,0,7,1
-2,33,113,1,7,7,1
-2,34,113,1,7,7,1
-2,35,113,1,7,7,1
-2,36,113,1,7,7,1
-2,37,113,1,7,7,1
+2,8,103,4,2,6,1
+2,9,113,6,2,0,1
+2,10,113,6,2,0,1
+2,11,113,6,2,0,1
+2,12,113,6,2,0,1
+2,13,113,6,2,0,2
+2,14,113,6,2,0,2
+2,15,113,6,2,0,2
+2,16,113,6,2,0,2
+2,17,113,6,2,0,2
+2,18,113,6,2,0,2
+2,19,113,6,2,0,2
+2,20,113,6,2,0,2
+2,21,113,6,2,0,2
+2,22,113,6,2,0,2
+2,23,113,6,2,0,2
+2,24,113,6,2,0,2
+2,25,113,6,2,0,2
+2,26,113,6,2,0,2
+2,27,113,6,2,0,2
+2,28,113,6,2,0,2
+2,29,113,6,2,0,2
+2,30,113,6,2,0,2
+2,31,113,6,2,0,2
+2,32,113,6,2,0,2
+2,33,113,6,2,0,2
+2,34,113,6,2,0,2
+2,35,113,6,2,0,2
+2,36,113,6,2,0,2
+2,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index f0150155..6052a5e2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,109,3,0,6,0
-3,9,121,3,9,8,0
-3,10,121,3,1,8,0
-3,11,121,0,1,8,0
-3,12,127,0,0,6,0
-3,13,127,4,8,7,0
-3,14,127,4,1,7,1
-3,15,127,1,1,7,1
-3,16,127,0,7,0,1
-3,17,127,5,1,0,1
-3,18,127,1,7,0,1
-3,19,113,1,7,0,1
-3,20,113,1,7,7,1
-3,21,113,1,7,0,1
-3,22,113,1,7,0,1
-3,23,113,1,7,7,1
-3,24,113,1,7,0,1
-3,25,113,1,7,2,1
-3,26,113,1,7,2,1
-3,27,113,1,7,7,1
-3,28,113,1,5,7,1
-3,29,113,1,7,2,1
-3,30,113,1,1,7,1
-3,31,113,1,0,7,1
-3,32,113,1,7,7,1
-3,33,113,1,7,7,1
-3,34,113,1,5,7,1
-3,35,113,1,7,2,1
-3,36,113,1,7,7,1
-3,37,113,1,5,7,1
-4,8,113,3,7,4,1
-4,9,113,1,7,2,1
-4,10,113,3,7,2,1
-4,11,113,1,1,2,1
-4,12,127,1,0,2,1
-4,13,127,1,0,2,1
-4,14,127,4,0,7,1
-4,15,102,1,0,2,1
-4,16,113,1,0,7,1
-4,17,108,1,7,7,1
-4,18,108,1,7,7,1
-4,19,113,1,7,7,1
-4,20,113,8,7,7,1
-4,21,113,1,7,2,1
-4,22,113,6,7,2,1
-4,23,113,1,7,2,1
-4,24,113,6,7,2,1
-4,25,113,1,1,2,1
-4,26,113,1,0,8,1
-4,27,113,1,0,2,1
-4,28,113,1,0,2,1
-4,29,113,1,0,7,1
-4,30,108,1,0,7,1
-4,31,108,1,0,7,1
-4,32,108,1,0,7,1
-4,33,108,1,7,7,1
-4,34,113,6,7,7,1
-4,35,113,1,7,7,1
-4,36,113,6,7,7,1
-4,37,113,6,7,2,1
+3,8,109,4,2,6,1
+3,9,113,4,2,0,1
+3,10,113,6,2,0,0
+3,11,113,6,2,0,0
+3,12,113,6,2,0,0
+3,13,113,6,2,0,0
+3,14,113,6,2,0,0
+3,15,113,6,2,0,0
+3,16,113,6,2,0,1
+3,17,113,6,2,0,1
+3,18,113,6,2,0,1
+3,19,113,6,2,0,2
+3,20,113,6,2,0,2
+3,21,113,6,2,0,2
+3,22,113,6,2,0,2
+3,23,113,6,2,0,2
+3,24,113,6,2,0,2
+3,25,113,6,2,0,2
+3,26,113,6,2,0,2
+3,27,113,6,2,0,2
+3,28,113,6,2,0,2
+3,29,113,6,2,0,2
+3,30,113,6,2,0,2
+3,31,113,6,2,0,2
+3,32,113,6,2,0,2
+3,33,113,6,2,0,2
+3,34,113,6,2,0,2
+3,35,113,6,2,0,2
+3,36,113,6,2,0,2
+3,37,113,6,2,0,2
+4,8,121,4,9,6,1
+4,9,113,4,2,6,1
+4,10,113,6,2,0,0
+4,11,113,6,2,0,0
+4,12,113,6,2,0,0
+4,13,113,6,2,0,1
+4,14,113,6,2,0,1
+4,15,113,6,2,0,1
+4,16,113,6,2,0,2
+4,17,113,6,2,0,2
+4,18,113,6,2,0,2
+4,19,113,6,2,0,2
+4,20,113,6,2,0,2
+4,21,113,6,2,0,2
+4,22,113,6,2,0,2
+4,23,113,6,2,0,2
+4,24,113,6,2,0,2
+4,25,113,6,2,0,2
+4,26,113,6,2,0,2
+4,27,113,6,2,0,2
+4,28,113,6,2,0,2
+4,29,113,6,2,0,2
+4,30,113,6,2,0,2
+4,31,113,6,2,0,2
+4,32,113,6,2,0,2
+4,33,113,6,2,0,2
+4,34,113,6,2,0,2
+4,35,113,6,2,0,2
+4,36,113,6,2,0,2
+4,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index f2cf6b66..77123cc4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,3,7,4,1
-5,9,113,1,9,6,1
-5,10,113,1,7,0,1
-5,11,113,1,7,2,1
-5,12,113,1,1,2,1
-5,13,113,1,0,2,1
-5,14,113,1,0,2,1
-5,15,113,1,0,2,1
-5,16,113,1,8,7,1
-5,17,113,1,8,7,1
-5,18,113,1,7,7,1
-5,19,113,1,7,7,1
-5,20,113,1,7,7,1
-5,21,113,1,7,7,1
-5,22,113,1,7,7,1
-5,23,113,1,7,2,1
-5,24,113,1,7,2,1
-5,25,113,1,7,2,1
-5,26,113,1,1,7,1
-5,27,113,1,5,2,1
-5,28,113,1,2,8,1
-5,29,113,1,1,7,1
-5,30,113,1,0,7,1
-5,31,113,1,0,7,1
-5,32,113,1,0,7,1
-5,33,113,1,7,7,1
-5,34,113,1,7,7,1
-5,35,113,1,7,7,1
-5,36,113,1,7,7,1
-5,37,113,1,7,7,1
+5,8,103,4,2,0,1
+5,9,113,6,2,0,1
+5,10,113,6,2,0,1
+5,11,113,6,2,0,1
+5,12,113,6,2,0,2
+5,13,113,6,2,0,2
+5,14,113,6,2,0,2
+5,15,113,6,2,0,2
+5,16,113,6,2,0,2
+5,17,113,6,2,0,2
+5,18,113,6,2,0,2
+5,19,113,6,2,0,2
+5,20,113,6,2,0,2
+5,21,113,6,2,0,2
+5,22,113,6,2,0,2
+5,23,113,6,2,0,2
+5,24,113,6,2,0,2
+5,25,113,6,2,0,2
+5,26,113,6,2,0,2
+5,27,113,6,2,0,2
+5,28,113,6,2,0,2
+5,29,113,6,2,0,2
+5,30,113,6,2,0,2
+5,31,113,6,2,0,2
+5,32,113,6,2,0,2
+5,33,113,6,2,0,2
+5,34,113,6,2,0,2
+5,35,113,6,2,0,2
+5,36,113,6,2,0,2
+5,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index 650c1470..f7596c81 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,122,4,7,4,1
-6,9,113,1,7,2,0
-6,10,113,3,7,4,1
-6,11,113,1,7,2,1
-6,12,113,6,7,7,1
-6,13,102,1,1,2,1
-6,14,127,4,7,8,1
-6,15,102,1,1,0,1
-6,16,113,1,7,7,1
-6,17,102,1,1,7,1
-6,18,105,1,7,7,1
-6,19,113,1,7,0,1
-6,20,113,1,7,7,1
-6,21,113,1,5,7,1
-6,22,113,1,5,2,1
-6,23,113,1,7,2,1
-6,24,113,1,7,7,1
-6,25,113,1,5,7,1
-6,26,113,1,7,8,1
-6,27,113,1,1,7,1
-6,28,113,1,5,7,1
-6,29,113,1,7,7,1
-6,30,113,1,5,7,1
-6,31,113,1,5,7,1
-6,32,113,1,2,7,1
-6,33,113,1,5,7,1
-6,34,113,1,2,7,1
-6,35,113,1,5,7,1
-6,36,113,1,2,8,1
-6,37,113,1,1,7,1
-7,8,121,1,9,6,0
-7,9,105,0,9,1,2
-7,10,121,0,9,6,0
-7,11,121,0,1,6,0
-7,12,121,0,3,8,0
-7,13,127,0,0,6,0
-7,14,127,4,0,7,0
-7,15,127,4,0,2,1
-7,16,127,4,0,2,0
-7,17,127,4,0,8,0
-7,18,127,4,1,2,0
-7,19,127,4,1,8,1
-7,20,127,4,1,8,1
-7,21,127,1,1,8,1
-7,22,113,1,7,0,1
-7,23,113,1,7,7,1
-7,24,113,1,9,0,1
-7,25,113,1,7,7,1
-7,26,113,1,9,0,1
-7,27,113,1,7,7,1
-7,28,113,1,5,0,1
-7,29,113,1,2,2,1
-7,30,113,1,7,7,1
-7,31,113,1,5,7,1
-7,32,113,1,2,2,1
-7,33,113,1,5,7,1
-7,34,113,1,2,7,1
-7,35,113,1,5,7,1
-7,36,108,1,2,8,1
-7,37,113,1,0,7,1
+6,8,121,4,2,6,0
+6,9,121,4,2,6,1
+6,10,113,4,2,0,1
+6,11,113,6,2,0,0
+6,12,113,6,2,0,0
+6,13,113,6,2,0,0
+6,14,113,6,2,0,0
+6,15,113,6,2,0,0
+6,16,113,6,2,0,0
+6,17,113,6,2,0,1
+6,18,113,6,2,0,1
+6,19,113,6,2,0,1
+6,20,113,6,2,0,2
+6,21,113,6,2,0,2
+6,22,113,6,2,0,2
+6,23,113,6,2,0,2
+6,24,113,6,2,0,2
+6,25,113,6,2,0,2
+6,26,113,6,2,0,2
+6,27,113,6,2,0,2
+6,28,113,6,2,0,2
+6,29,113,6,2,0,2
+6,30,113,6,2,0,2
+6,31,113,6,2,0,2
+6,32,113,6,2,0,2
+6,33,113,6,2,0,2
+6,34,113,6,2,0,2
+6,35,113,6,2,0,2
+6,36,113,6,2,0,2
+6,37,113,6,2,0,2
+7,8,109,4,9,6,0
+7,9,109,4,9,6,0
+7,10,121,4,9,6,0
+7,11,121,4,2,8,1
+7,12,109,4,9,8,0
+7,13,109,4,9,8,0
+7,14,109,4,0,5,0
+7,15,109,6,0,5,0
+7,16,109,6,0,8,0
+7,17,109,4,0,5,0
+7,18,109,4,0,5,0
+7,19,127,4,0,5,0
+7,20,127,4,0,5,0
+7,21,109,4,9,5,0
+7,22,109,4,3,0,0
+7,23,109,4,3,0,0
+7,24,109,4,9,5,0
+7,25,109,4,3,0,0
+7,26,109,4,3,0,0
+7,27,121,4,3,0,0
+7,28,109,4,9,8,0
+7,29,109,4,0,8,0
+7,30,109,4,9,8,0
+7,31,109,4,9,8,0
+7,32,109,4,0,5,0
+7,33,109,4,9,5,0
+7,34,109,4,0,5,0
+7,35,109,4,0,5,0
+7,36,127,4,0,5,0
+7,37,109,4,0,5,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index 2cdb2be9..49dff757 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,1,3,2,0
-8,9,113,3,7,7,1
-8,10,121,1,7,2,1
-8,11,113,4,7,7,2
-8,12,121,1,1,2,1
-8,13,127,1,0,8,0
-8,14,109,1,0,8,1
-8,15,127,1,0,7,0
-8,16,127,1,0,7,1
-8,17,127,1,7,7,1
-8,18,127,1,7,7,1
-8,19,113,1,7,0,1
-8,20,113,1,7,7,1
-8,21,113,8,7,7,1
-8,22,113,1,5,2,1
-8,23,113,1,7,8,1
-8,24,113,1,7,0,1
-8,25,113,1,7,7,1
-8,26,113,1,5,7,1
-8,27,113,1,2,2,1
-8,28,113,1,1,7,1
-8,29,108,1,0,2,1
-8,30,113,1,0,7,1
-8,31,113,1,0,7,1
-8,32,113,1,7,7,1
-8,33,113,1,7,7,1
-8,34,113,1,7,7,1
-8,35,113,1,7,7,1
-8,36,113,6,7,7,1
-8,37,113,6,7,2,1
+8,8,103,4,2,0,0
+8,9,113,6,2,0,0
+8,10,113,6,2,0,0
+8,11,113,6,2,0,0
+8,12,113,6,2,0,0
+8,13,113,6,2,0,0
+8,14,113,6,7,0,0
+8,15,113,6,7,0,0
+8,16,113,6,2,0,1
+8,17,113,6,2,0,1
+8,18,113,6,2,0,1
+8,19,113,6,2,0,2
+8,20,113,6,2,0,2
+8,21,113,6,2,0,2
+8,22,113,6,2,0,2
+8,23,113,6,2,0,2
+8,24,113,6,2,0,2
+8,25,113,6,2,0,2
+8,26,113,6,2,0,2
+8,27,113,6,2,0,2
+8,28,113,6,2,0,2
+8,29,113,6,2,0,2
+8,30,113,6,2,0,2
+8,31,113,6,2,0,2
+8,32,113,6,2,0,2
+8,33,113,6,2,0,2
+8,34,113,6,2,0,2
+8,35,113,6,2,0,2
+8,36,113,6,2,0,2
+8,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index 1c08eb2a..4a0c0385 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,121,1,7,2,0
-9,9,113,3,7,1,2
-9,10,113,1,9,6,1
-9,11,113,1,7,7,1
-9,12,113,1,7,7,1
-9,13,113,1,7,7,1
-9,14,113,1,7,7,1
-9,15,113,1,7,2,1
-9,16,113,1,1,2,1
-9,17,113,1,0,2,1
-9,18,113,1,0,2,1
-9,19,113,1,0,2,1
-9,20,113,1,0,7,1
-9,21,113,1,0,7,1
-9,22,108,1,0,7,1
-9,23,108,1,7,7,1
-9,24,113,1,7,7,1
-9,25,113,1,7,7,1
-9,26,113,6,7,7,1
-9,27,113,1,7,2,1
-9,28,113,6,7,7,1
-9,29,113,1,7,2,1
-9,30,113,6,7,2,1
-9,31,113,1,1,2,1
-9,32,113,1,0,2,1
-9,33,113,1,0,2,1
-9,34,113,1,0,2,1
-9,35,113,1,0,7,1
-9,36,108,1,0,2,1
-9,37,113,1,0,7,1
+9,8,103,4,3,0,0
+9,9,103,4,2,0,0
+9,10,121,6,2,0,0
+9,11,121,6,2,0,0
+9,12,121,6,2,0,1
+9,13,113,6,2,8,0
+9,14,121,6,2,8,0
+9,15,121,6,2,8,0
+9,16,121,4,2,8,0
+9,17,109,4,9,8,0
+9,18,109,4,0,5,0
+9,19,109,4,0,5,0
+9,20,120,4,0,5,0
+9,21,109,4,0,5,0
+9,22,109,4,9,5,0
+9,23,109,4,0,5,0
+9,24,109,4,9,5,0
+9,25,127,4,0,9,0
+9,26,109,4,9,5,0
+9,27,109,4,3,9,0
+9,28,109,4,0,5,0
+9,29,109,4,3,9,0
+9,30,120,4,0,5,0
+9,31,109,4,3,9,0
+9,32,120,4,0,5,0
+9,33,109,4,9,5,0
+9,34,127,4,0,9,0
+9,35,109,4,9,5,0
+9,36,109,4,9,5,0
+9,37,109,4,0,5,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index 105792b4..68b4bc7d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,113,6,2,0,1
-0,9,102,6,2,0,1
-0,10,102,6,2,0,1
-0,11,121,6,2,0,1
-0,12,121,6,5,6,1
-0,13,121,6,2,6,1
-0,14,121,6,2,6,1
-0,15,121,6,2,6,1
-0,16,121,6,2,6,1
-0,17,121,6,5,6,1
-0,18,121,3,5,6,1
-0,19,121,3,5,6,1
-0,20,121,3,5,6,1
-0,21,121,3,5,6,1
-0,22,121,3,5,6,1
-0,23,121,3,5,6,1
-0,24,121,3,5,6,1
-0,25,121,3,5,6,0
-0,26,121,3,5,6,0
-0,27,121,3,5,6,0
-0,28,121,3,5,6,0
-0,29,121,3,5,6,0
-0,30,121,3,5,6,0
-0,31,121,3,5,6,0
-0,32,121,3,5,6,0
-0,33,121,3,5,6,0
-0,34,121,3,5,6,0
-0,35,121,3,5,6,0
-0,36,121,3,5,6,0
-0,37,121,3,5,6,0
+0,8,104,4,2,8,1
+0,9,102,6,2,8,1
+0,10,102,6,2,8,0
+0,11,102,6,2,8,1
+0,12,102,6,2,8,1
+0,13,102,6,2,8,1
+0,14,102,6,2,6,1
+0,15,102,6,2,8,1
+0,16,102,6,2,6,1
+0,17,102,6,2,8,1
+0,18,102,6,2,6,1
+0,19,102,6,2,0,1
+0,20,102,6,2,0,1
+0,21,102,6,2,0,1
+0,22,102,6,2,0,1
+0,23,102,6,2,0,1
+0,24,102,6,2,0,1
+0,25,102,6,2,0,1
+0,26,102,6,2,0,1
+0,27,102,6,2,0,1
+0,28,102,6,2,0,1
+0,29,102,6,2,0,1
+0,30,102,6,2,0,1
+0,31,102,6,2,0,1
+0,32,102,6,2,0,1
+0,33,102,6,2,0,1
+0,34,102,6,2,0,1
+0,35,102,6,2,0,1
+0,36,102,6,2,0,1
+0,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index 350e2f2f..875d7e7d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,113,6,2,0,0
-1,9,121,6,5,0,1
-1,10,121,4,9,6,0
-1,11,121,4,5,6,1
-1,12,121,4,5,6,1
-1,13,121,3,2,6,1
-1,14,121,3,2,6,1
-1,15,121,4,5,6,1
-1,16,121,3,5,6,1
-1,17,121,3,5,6,1
-1,18,121,3,2,6,1
-1,19,121,3,2,6,1
-1,20,121,3,2,6,1
-1,21,121,3,5,6,1
-1,22,121,3,5,6,1
-1,23,121,3,5,6,1
-1,24,121,3,5,6,1
-1,25,121,3,5,6,1
-1,26,121,3,5,6,1
-1,27,121,3,5,6,1
-1,28,121,3,5,6,1
-1,29,121,3,5,6,0
-1,30,121,3,5,6,0
-1,31,121,3,5,6,0
-1,32,121,3,5,6,0
-1,33,121,3,5,6,0
-1,34,121,3,5,6,0
-1,35,121,3,5,6,0
-1,36,121,3,5,6,0
-1,37,121,3,5,6,0
+1,8,104,4,7,6,0
+1,9,103,4,9,8,1
+1,10,102,6,2,8,0
+1,11,102,6,2,8,1
+1,12,102,6,2,8,1
+1,13,102,6,2,8,1
+1,14,102,6,2,8,1
+1,15,102,6,2,8,1
+1,16,102,6,2,8,1
+1,17,102,6,2,6,1
+1,18,102,6,2,8,1
+1,19,102,6,2,0,1
+1,20,102,6,2,0,1
+1,21,102,6,2,0,1
+1,22,102,6,2,0,1
+1,23,102,6,2,0,1
+1,24,102,6,2,0,1
+1,25,102,6,2,0,1
+1,26,102,6,2,0,1
+1,27,102,6,2,0,1
+1,28,102,6,2,0,1
+1,29,102,6,2,0,1
+1,30,102,6,2,0,1
+1,31,102,6,2,0,1
+1,32,102,6,2,0,1
+1,33,102,6,2,0,1
+1,34,102,6,2,0,1
+1,35,102,6,2,0,1
+1,36,102,6,2,0,1
+1,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index 7cb7ee53..aaa36998 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,105,5,3,0,0
-2,9,102,6,1,0,1
-2,10,121,6,2,0,1
-2,11,121,6,2,6,1
-2,12,121,6,2,6,1
-2,13,121,6,2,6,1
-2,14,121,6,2,6,1
-2,15,121,6,2,6,1
-2,16,121,6,5,6,1
-2,17,121,6,5,6,1
-2,18,121,3,5,6,1
-2,19,121,3,5,6,1
-2,20,121,3,5,6,1
-2,21,121,3,5,6,1
-2,22,121,3,5,6,1
-2,23,121,3,5,6,1
-2,24,121,3,5,6,1
-2,25,121,3,5,6,0
-2,26,121,3,5,6,0
-2,27,121,3,5,6,0
-2,28,121,3,5,6,0
-2,29,121,3,5,6,0
-2,30,121,3,5,6,0
-2,31,121,3,5,6,0
-2,32,121,3,5,6,0
-2,33,121,3,5,6,0
-2,34,121,3,5,6,0
-2,35,121,3,5,6,0
-2,36,121,3,5,6,0
-2,37,121,3,5,6,0
+2,8,110,4,2,6,0
+2,9,110,4,2,8,1
+2,10,110,6,2,8,0
+2,11,102,6,2,8,0
+2,12,102,6,2,8,1
+2,13,102,6,2,8,1
+2,14,102,6,2,8,1
+2,15,102,6,2,8,1
+2,16,102,6,2,8,1
+2,17,102,6,2,6,1
+2,18,102,6,2,8,1
+2,19,102,6,2,6,1
+2,20,102,6,2,0,1
+2,21,102,6,2,0,1
+2,22,102,6,2,0,1
+2,23,102,6,2,0,1
+2,24,102,6,2,0,1
+2,25,102,6,2,0,1
+2,26,102,6,2,0,1
+2,27,102,6,2,0,1
+2,28,102,6,2,0,1
+2,29,102,6,2,0,1
+2,30,102,6,2,0,1
+2,31,102,6,2,0,1
+2,32,102,6,2,0,1
+2,33,102,6,2,0,1
+2,34,102,6,2,0,1
+2,35,102,6,2,0,1
+2,36,102,6,2,0,1
+2,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index e82f6f8a..9d5383b8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,113,6,2,0,0
-3,9,102,6,2,0,1
-3,10,121,6,2,0,1
-3,11,121,6,5,6,1
-3,12,121,6,5,6,1
-3,13,121,4,5,6,1
-3,14,121,3,5,6,1
-3,15,121,4,5,6,1
-3,16,121,3,5,6,1
-3,17,121,3,2,6,1
-3,18,121,3,2,6,1
-3,19,121,3,2,6,1
-3,20,121,3,5,6,1
-3,21,121,3,5,6,1
-3,22,121,3,5,6,1
-3,23,121,3,5,6,1
-3,24,121,3,5,6,1
-3,25,121,3,5,6,1
-3,26,121,3,5,6,1
-3,27,121,3,5,6,1
-3,28,121,3,5,6,0
-3,29,121,3,5,6,0
-3,30,121,3,5,6,0
-3,31,121,3,5,6,0
-3,32,121,3,5,6,0
-3,33,121,3,5,6,0
-3,34,121,3,5,6,0
-3,35,121,3,5,6,0
-3,36,121,3,5,6,0
-3,37,121,3,5,6,0
-4,8,105,5,8,2,0
-4,9,102,6,3,0,1
-4,10,121,6,3,0,1
-4,11,121,4,9,6,1
-4,12,121,6,5,6,1
-4,13,121,4,2,6,1
-4,14,121,4,2,6,1
-4,15,121,3,2,6,1
-4,16,121,6,2,6,1
-4,17,121,6,2,6,1
-4,18,121,6,5,6,1
-4,19,121,6,5,6,1
-4,20,121,3,5,6,1
-4,21,121,3,2,6,1
-4,22,121,3,5,6,1
-4,23,121,3,5,6,1
-4,24,121,3,5,6,1
-4,25,121,3,5,6,1
-4,26,121,3,5,6,1
-4,27,121,3,5,6,1
-4,28,121,3,5,6,1
-4,29,121,3,5,6,1
-4,30,121,3,5,6,0
-4,31,121,3,5,6,0
-4,32,121,3,5,6,0
-4,33,121,3,5,6,0
-4,34,121,3,5,6,0
-4,35,121,3,5,6,0
-4,36,121,3,5,6,0
-4,37,121,3,5,6,0
+3,8,104,4,2,6,0
+3,9,103,6,5,8,1
+3,10,102,4,2,6,0
+3,11,102,6,2,8,1
+3,12,102,6,2,8,1
+3,13,102,6,2,8,1
+3,14,102,6,2,8,1
+3,15,102,6,2,8,1
+3,16,102,6,2,8,1
+3,17,102,6,2,8,1
+3,18,102,6,2,6,1
+3,19,102,6,2,0,1
+3,20,102,6,2,0,1
+3,21,102,6,2,0,1
+3,22,102,6,2,0,1
+3,23,102,6,2,0,1
+3,24,102,6,2,0,1
+3,25,102,6,2,0,1
+3,26,102,6,2,0,1
+3,27,102,6,2,0,1
+3,28,102,6,2,0,1
+3,29,102,6,2,0,1
+3,30,102,6,2,0,1
+3,31,102,6,2,0,1
+3,32,102,6,2,0,1
+3,33,102,6,2,0,1
+3,34,102,6,2,0,1
+3,35,102,6,2,0,1
+3,36,102,6,2,0,1
+3,37,102,6,2,0,1
+4,8,110,4,2,6,0
+4,9,110,4,2,8,1
+4,10,102,6,2,6,0
+4,11,102,6,2,8,1
+4,12,102,6,2,8,1
+4,13,102,6,2,0,1
+4,14,102,6,2,8,1
+4,15,102,6,2,6,1
+4,16,102,6,2,0,1
+4,17,102,6,2,0,1
+4,18,102,6,2,0,1
+4,19,102,6,2,0,1
+4,20,102,6,2,0,1
+4,21,102,6,2,0,1
+4,22,102,6,2,0,1
+4,23,102,6,2,0,1
+4,24,102,6,2,0,1
+4,25,102,6,2,0,1
+4,26,102,6,2,0,1
+4,27,102,6,2,0,1
+4,28,102,6,2,0,1
+4,29,102,6,2,0,1
+4,30,102,6,2,0,1
+4,31,102,6,2,0,1
+4,32,102,6,2,0,1
+4,33,102,6,2,0,1
+4,34,102,6,2,0,1
+4,35,102,6,2,0,1
+4,36,102,6,2,0,1
+4,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index 7ee664d5..956ddf9b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,122,6,8,2,0
-5,9,113,6,2,0,1
-5,10,113,6,3,0,1
-5,11,121,6,2,0,1
-5,12,121,6,2,6,1
-5,13,121,6,2,6,1
-5,14,121,6,2,6,1
-5,15,121,6,5,6,1
-5,16,121,6,5,6,1
-5,17,121,6,5,6,1
-5,18,121,3,5,6,1
-5,19,121,3,5,6,1
-5,20,121,3,5,6,1
-5,21,121,3,5,6,1
-5,22,121,3,5,6,1
-5,23,121,3,5,6,1
-5,24,121,3,5,6,1
-5,25,121,3,5,6,0
-5,26,121,3,5,6,0
-5,27,121,3,5,6,0
-5,28,121,3,5,6,0
-5,29,121,3,5,6,0
-5,30,121,3,5,6,0
-5,31,121,3,5,6,0
-5,32,121,3,5,6,0
-5,33,121,3,5,6,0
-5,34,121,3,5,6,0
-5,35,121,3,5,6,0
-5,36,121,3,5,6,0
-5,37,121,3,5,6,0
+5,8,110,4,9,6,0
+5,9,110,4,2,8,1
+5,10,110,4,2,6,0
+5,11,113,6,2,8,2
+5,12,113,6,2,0,1
+5,13,113,6,2,0,1
+5,14,102,6,2,0,1
+5,15,102,6,2,0,1
+5,16,102,6,2,0,1
+5,17,102,6,2,0,1
+5,18,102,6,2,0,1
+5,19,102,6,2,0,1
+5,20,102,6,2,0,1
+5,21,102,6,2,0,1
+5,22,102,6,2,0,1
+5,23,102,6,2,0,1
+5,24,102,6,2,0,1
+5,25,102,6,2,0,1
+5,26,102,6,2,0,1
+5,27,102,6,2,0,1
+5,28,102,6,2,0,1
+5,29,102,6,2,0,1
+5,30,102,6,2,0,1
+5,31,102,6,2,0,1
+5,32,102,6,2,0,1
+5,33,102,6,2,0,1
+5,34,102,6,2,0,1
+5,35,102,6,2,0,1
+5,36,102,6,2,0,1
+5,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index e3ab2a3d..b4ef4bd3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,121,6,3,0,0
-6,9,121,4,9,6,1
-6,10,121,6,5,6,0
-6,11,121,4,5,6,1
-6,12,121,4,5,6,1
-6,13,121,4,5,6,1
-6,14,121,4,5,6,0
-6,15,121,3,5,6,1
-6,16,121,3,5,6,1
-6,17,121,3,5,6,1
-6,18,121,3,5,6,1
-6,19,121,3,2,6,1
-6,20,121,3,2,6,1
-6,21,121,3,5,6,1
-6,22,121,3,5,6,0
-6,23,121,3,5,6,1
-6,24,121,3,5,6,0
-6,25,121,3,5,6,1
-6,26,121,3,5,6,0
-6,27,121,3,5,6,1
-6,28,121,3,5,6,0
-6,29,121,3,5,6,0
-6,30,121,3,5,6,0
-6,31,121,3,5,6,0
-6,32,121,3,5,6,0
-6,33,121,3,5,6,0
-6,34,121,3,5,6,0
-6,35,121,3,5,6,0
-6,36,121,3,5,6,0
-6,37,121,3,5,6,0
-7,8,121,4,9,5,0
-7,9,121,4,3,6,0
-7,10,121,4,5,6,0
-7,11,121,4,5,6,0
-7,12,121,4,9,6,0
-7,13,121,4,9,6,0
-7,14,121,4,5,6,0
-7,15,121,3,5,6,1
-7,16,121,3,5,6,1
-7,17,121,3,5,6,1
-7,18,121,3,5,6,0
-7,19,121,3,5,6,1
-7,20,121,3,5,6,0
-7,21,121,3,5,6,0
-7,22,121,3,5,6,0
-7,23,121,3,5,6,0
-7,24,121,3,5,6,0
-7,25,121,3,5,6,0
-7,26,121,3,5,6,0
-7,27,121,3,5,6,0
-7,28,121,3,5,6,0
-7,29,121,3,5,6,0
-7,30,121,3,5,6,0
-7,31,121,3,5,6,0
-7,32,121,3,5,6,0
-7,33,121,3,5,6,0
-7,34,121,3,5,6,0
-7,35,121,3,5,6,0
-7,36,121,3,5,6,0
-7,37,121,3,5,6,0
+6,8,110,4,2,8,0
+6,9,102,6,2,8,0
+6,10,102,6,2,8,1
+6,11,102,6,2,8,1
+6,12,102,6,2,8,1
+6,13,102,6,2,8,1
+6,14,102,6,2,8,1
+6,15,102,6,3,6,1
+6,16,102,6,2,8,1
+6,17,102,6,2,6,1
+6,18,102,6,2,0,1
+6,19,102,6,2,0,1
+6,20,102,6,2,0,1
+6,21,102,6,2,0,1
+6,22,102,6,2,0,1
+6,23,102,6,2,0,1
+6,24,102,6,2,0,1
+6,25,102,6,2,0,1
+6,26,102,6,2,0,1
+6,27,102,6,2,0,1
+6,28,102,6,2,0,1
+6,29,102,6,2,0,1
+6,30,102,6,2,0,1
+6,31,102,6,2,0,1
+6,32,102,6,2,0,1
+6,33,102,6,2,0,1
+6,34,102,6,2,0,1
+6,35,102,6,2,0,1
+6,36,102,6,2,0,1
+6,37,102,6,2,0,1
+7,8,104,4,7,6,0
+7,9,124,4,9,8,1
+7,10,110,6,2,8,0
+7,11,102,4,2,6,0
+7,12,102,6,2,8,1
+7,13,102,6,2,8,1
+7,14,102,6,2,6,1
+7,15,102,6,2,8,1
+7,16,102,6,2,8,1
+7,17,102,6,2,6,1
+7,18,102,6,2,8,1
+7,19,102,6,2,0,1
+7,20,102,6,2,0,1
+7,21,102,6,2,0,1
+7,22,102,6,2,0,1
+7,23,102,6,2,0,1
+7,24,102,6,2,0,1
+7,25,102,6,2,0,1
+7,26,102,6,2,0,1
+7,27,102,6,2,0,1
+7,28,102,6,2,0,1
+7,29,102,6,2,0,1
+7,30,102,6,2,0,1
+7,31,102,6,2,0,1
+7,32,102,6,2,0,1
+7,33,102,6,2,0,1
+7,34,102,6,2,0,1
+7,35,102,6,2,0,1
+7,36,102,6,2,0,1
+7,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index 7306d67d..11f63b7e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,102,6,3,0,0
-8,9,121,6,3,0,0
-8,10,121,4,5,6,0
-8,11,121,3,5,6,0
-8,12,121,4,5,6,1
-8,13,121,4,9,6,0
-8,14,121,3,5,6,1
-8,15,121,3,5,6,1
-8,16,121,3,5,6,0
-8,17,121,3,5,6,0
-8,18,121,3,5,6,0
-8,19,121,3,5,6,0
-8,20,121,3,5,6,0
-8,21,121,3,5,6,0
-8,22,121,3,5,6,0
-8,23,121,3,5,6,0
-8,24,121,3,5,6,0
-8,25,121,3,5,6,0
-8,26,121,3,5,6,0
-8,27,121,3,5,6,0
-8,28,121,3,5,6,0
-8,29,121,3,5,6,0
-8,30,121,3,5,6,0
-8,31,121,3,5,6,0
-8,32,121,3,5,6,0
-8,33,121,3,5,6,0
-8,34,121,3,5,6,0
-8,35,121,3,5,6,0
-8,36,121,3,5,6,0
-8,37,121,3,5,6,0
+8,8,110,4,2,6,0
+8,9,102,4,2,8,1
+8,10,102,6,2,8,0
+8,11,102,6,2,8,1
+8,12,102,6,2,8,1
+8,13,102,6,2,8,1
+8,14,102,6,2,8,1
+8,15,102,6,2,6,1
+8,16,102,6,2,8,1
+8,17,102,6,2,6,1
+8,18,102,6,2,8,1
+8,19,102,6,2,0,1
+8,20,102,6,2,0,1
+8,21,102,6,2,0,1
+8,22,102,6,2,0,1
+8,23,102,6,2,0,1
+8,24,102,6,2,0,1
+8,25,102,6,2,0,1
+8,26,102,6,2,0,1
+8,27,102,6,2,0,1
+8,28,102,6,2,0,1
+8,29,102,6,2,0,1
+8,30,102,6,2,0,1
+8,31,102,6,2,0,1
+8,32,102,6,2,0,1
+8,33,102,6,2,0,1
+8,34,102,6,2,0,1
+8,35,102,6,2,0,1
+8,36,102,6,2,0,1
+8,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index b3e95b3e..c00f4102 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,113,6,3,2,0
-9,9,102,6,3,0,0
-9,10,121,6,2,0,1
-9,11,121,4,5,6,0
-9,12,121,3,5,6,1
-9,13,121,4,5,6,1
-9,14,121,3,5,6,1
-9,15,121,3,5,6,1
-9,16,121,3,5,6,1
-9,17,121,3,5,6,1
-9,18,121,3,5,6,1
-9,19,121,3,5,6,0
-9,20,121,3,5,6,0
-9,21,121,3,5,6,0
-9,22,121,3,5,6,0
-9,23,121,3,5,6,0
-9,24,121,3,5,6,0
-9,25,121,3,5,6,0
-9,26,121,3,5,6,0
-9,27,121,3,5,6,0
-9,28,121,3,5,6,0
-9,29,121,3,5,6,0
-9,30,121,3,5,6,0
-9,31,121,3,5,6,0
-9,32,121,3,5,6,0
-9,33,121,3,5,6,0
-9,34,121,3,5,6,0
-9,35,121,3,5,6,0
-9,36,121,3,5,6,0
-9,37,121,3,5,6,0
+9,8,110,4,8,6,0
+9,9,110,4,2,8,0
+9,10,102,4,2,8,0
+9,11,102,6,2,8,1
+9,12,102,6,2,8,0
+9,13,102,6,2,0,1
+9,14,102,6,2,0,1
+9,15,102,6,2,0,1
+9,16,102,6,2,8,1
+9,17,102,6,2,8,1
+9,18,102,6,2,6,1
+9,19,102,6,2,8,1
+9,20,102,6,2,6,1
+9,21,102,6,2,0,1
+9,22,102,6,2,0,1
+9,23,102,6,2,0,1
+9,24,102,6,2,0,1
+9,25,102,6,2,0,1
+9,26,102,6,2,0,1
+9,27,102,6,2,0,1
+9,28,102,6,2,0,1
+9,29,102,6,2,0,1
+9,30,102,6,2,0,1
+9,31,102,6,2,0,1
+9,32,102,6,2,0,1
+9,33,102,6,2,0,1
+9,34,102,6,2,0,1
+9,35,102,6,2,0,1
+9,36,102,6,2,0,1
+9,37,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
index 39c6e464..0920b6c1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,102,6,3,8,1
-0,9,121,6,3,8,1
-0,10,104,6,3,8,1
-0,11,121,6,3,8,1
-0,12,121,6,3,8,1
-0,13,104,6,3,8,0
-0,14,104,3,3,0,0
-0,15,104,3,3,0,0
-0,16,104,3,3,0,0
-0,17,104,3,3,0,0
-0,18,104,3,3,0,0
-0,19,104,3,3,0,0
-0,20,104,3,3,0,1
-0,21,103,6,3,0,1
-0,22,113,6,3,8,1
-0,23,113,6,3,8,1
-0,24,113,6,3,8,1
-0,25,104,6,3,8,1
-0,26,104,6,3,8,1
-0,27,104,6,3,8,1
-0,28,104,6,3,8,1
-0,29,104,6,3,8,1
-0,30,104,9,3,0,0
-0,31,104,3,3,0,1
-0,32,104,9,3,0,1
-0,33,104,9,3,0,1
-0,34,104,3,3,8,1
-0,35,104,6,3,8,0
-0,36,121,6,3,8,0
-0,37,121,6,3,8,0
+0,8,121,4,9,2,1
+0,9,113,9,7,8,1
+0,10,113,3,1,6,1
+0,11,113,9,3,2,1
+0,12,127,6,1,7,1
+0,13,121,6,1,7,1
+0,14,121,9,1,7,1
+0,15,121,6,0,7,1
+0,16,121,6,0,7,1
+0,17,121,6,7,7,1
+0,18,113,6,7,7,1
+0,19,121,6,7,2,1
+0,20,113,6,7,2,0
+0,21,113,4,7,2,0
+0,22,121,6,1,2,0
+0,23,127,4,0,8,1
+0,24,127,6,0,2,0
+0,25,127,4,0,8,1
+0,26,127,4,1,8,1
+0,27,127,4,1,8,0
+0,28,127,4,1,8,1
+0,29,127,4,1,7,1
+0,30,127,0,1,0,1
+0,31,113,1,4,0,1
+0,32,113,1,9,7,1
+0,33,113,1,7,7,1
+0,34,113,6,9,0,1
+0,35,113,1,7,7,1
+0,36,113,6,7,7,1
+0,37,113,6,7,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
index b1f5c970..a73055c6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,6,3,8,1
+1,8,121,0,7,6,0
 1,9,121,6,3,8,0
-1,10,121,6,3,8,0
-1,11,121,6,3,8,0
-1,12,121,6,3,8,0
-1,13,104,3,3,8,0
-1,14,104,3,3,0,0
-1,15,104,3,3,0,0
-1,16,104,3,3,0,0
-1,17,104,3,3,0,0
-1,18,104,3,3,0,0
-1,19,104,3,3,0,0
-1,20,104,3,3,0,0
-1,21,121,6,3,0,1
-1,22,113,6,3,8,1
-1,23,121,6,3,8,1
-1,24,104,6,3,8,1
-1,25,104,6,3,8,0
-1,26,104,6,3,0,0
-1,27,104,9,3,0,1
-1,28,104,3,3,0,1
-1,29,104,9,3,0,1
-1,30,104,3,3,0,1
-1,31,104,3,3,0,1
-1,32,103,6,3,8,1
-1,33,113,6,3,8,1
-1,34,121,6,3,8,1
-1,35,121,6,3,8,1
-1,36,104,6,3,8,1
-1,37,104,6,3,8,0
+1,10,121,9,0,8,0
+1,11,121,4,0,8,0
+1,12,127,6,0,7,0
+1,13,121,6,0,7,1
+1,14,127,6,7,7,1
+1,15,127,6,7,7,1
+1,16,127,4,1,7,1
+1,17,127,4,1,2,1
+1,18,127,4,1,2,1
+1,19,127,6,7,2,1
+1,20,113,6,7,0,1
+1,21,113,1,7,0,1
+1,22,113,1,7,0,1
+1,23,113,1,7,0,1
+1,24,113,1,7,0,1
+1,25,113,1,7,0,1
+1,26,113,6,7,2,1
+1,27,113,6,1,2,1
+1,28,113,6,0,2,1
+1,29,113,6,0,2,1
+1,30,113,6,0,2,1
+1,31,113,6,0,2,1
+1,32,113,1,8,2,1
+1,33,113,1,1,8,1
+1,34,113,1,0,8,1
+1,35,113,1,8,8,1
+1,36,113,1,0,7,1
+1,37,113,1,8,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
index b32ec715..087f11a3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,102,6,3,8,1
-2,9,121,6,3,8,1
-2,10,104,6,3,8,1
-2,11,104,6,3,8,0
-2,12,104,6,3,8,0
-2,13,104,6,3,8,0
-2,14,104,6,3,0,0
-2,15,104,3,3,0,0
-2,16,104,3,3,0,0
-2,17,104,3,3,0,0
-2,18,104,3,3,0,1
-2,19,103,4,3,0,1
-2,20,102,6,3,8,1
-2,21,121,6,3,8,1
-2,22,104,6,3,8,1
-2,23,104,6,3,8,1
-2,24,104,6,3,8,0
-2,25,104,6,3,8,0
-2,26,104,6,3,0,0
-2,27,104,9,3,0,1
-2,28,104,3,3,0,0
-2,29,104,3,3,0,0
-2,30,104,3,3,0,1
-2,31,103,4,3,0,1
-2,32,102,6,3,8,1
-2,33,121,6,3,8,1
-2,34,121,6,3,8,1
-2,35,104,6,3,8,1
-2,36,104,6,3,8,0
-2,37,104,6,3,0,0
+2,8,121,4,7,8,0
+2,9,121,0,3,7,0
+2,10,121,0,7,6,1
+2,11,113,0,7,6,0
+2,12,121,4,1,6,1
+2,13,127,0,7,7,1
+2,14,127,6,7,0,1
+2,15,127,6,1,0,1
+2,16,127,6,7,2,1
+2,17,127,6,7,0,1
+2,18,127,6,1,0,1
+2,19,127,6,1,0,1
+2,20,113,6,7,0,1
+2,21,113,1,7,0,1
+2,22,113,1,7,0,1
+2,23,113,1,7,0,1
+2,24,113,1,7,0,1
+2,25,113,6,7,2,1
+2,26,113,6,1,2,1
+2,27,113,6,0,2,1
+2,28,113,6,0,2,1
+2,29,113,6,0,2,1
+2,30,113,6,0,2,1
+2,31,113,1,8,2,1
+2,32,113,1,1,8,1
+2,33,113,1,0,8,1
+2,34,113,1,8,8,1
+2,35,113,1,0,7,1
+2,36,113,1,8,7,1
+2,37,121,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
index 03dc6b54..c6ba0686 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,6,3,8,1
-3,9,104,4,3,8,0
-3,10,121,3,3,8,0
-3,11,104,3,3,8,0
-3,12,121,3,3,8,0
-3,13,121,3,3,8,0
-3,14,104,3,3,8,0
-3,15,104,3,3,0,0
-3,16,104,3,3,0,0
-3,17,104,3,3,0,0
-3,18,104,3,3,0,0
-3,19,104,3,3,0,0
-3,20,104,3,3,0,0
-3,21,104,3,3,0,0
-3,22,121,6,3,0,1
-3,23,113,6,3,8,1
-3,24,121,6,3,8,1
-3,25,104,6,3,8,1
-3,26,104,6,3,8,0
-3,27,104,6,3,0,0
-3,28,104,9,3,0,1
-3,29,104,3,3,0,1
-3,30,104,9,3,0,1
-3,31,104,3,3,0,1
-3,32,104,3,3,0,1
-3,33,103,6,3,8,1
-3,34,113,6,3,8,1
-3,35,121,6,3,8,1
-3,36,121,6,3,8,1
-3,37,104,6,3,8,1
-4,8,102,6,9,8,1
-4,9,121,6,3,8,1
-4,10,121,6,3,8,1
-4,11,121,6,3,8,0
-4,12,121,6,3,8,0
-4,13,121,6,3,8,0
-4,14,104,3,3,8,0
-4,15,104,3,3,0,0
-4,16,104,3,3,0,0
-4,17,104,3,3,0,0
-4,18,104,3,3,0,0
-4,19,104,3,3,0,0
-4,20,104,3,3,0,0
-4,21,104,3,3,0,0
-4,22,121,6,3,0,1
-4,23,113,6,3,8,1
-4,24,121,6,3,8,1
-4,25,104,6,3,8,1
-4,26,104,6,3,8,0
-4,27,104,6,3,0,0
-4,28,104,9,3,0,1
-4,29,104,3,3,0,1
-4,30,104,9,3,0,1
-4,31,104,3,3,0,1
-4,32,104,3,3,0,1
-4,33,103,6,3,8,1
-4,34,113,6,3,8,1
-4,35,121,6,3,8,1
-4,36,121,6,3,8,1
-4,37,104,6,3,8,1
+3,8,121,4,9,6,0
+3,9,121,0,3,8,1
+3,10,121,0,7,8,0
+3,11,121,4,0,8,0
+3,12,127,4,0,8,0
+3,13,127,4,0,7,0
+3,14,127,6,0,7,1
+3,15,127,6,7,7,1
+3,16,127,6,7,7,1
+3,17,127,6,1,7,1
+3,18,127,6,1,7,1
+3,19,124,6,7,2,1
+3,20,113,6,7,2,1
+3,21,113,6,7,2,1
+3,22,113,6,7,2,1
+3,23,113,6,1,2,1
+3,24,113,1,2,8,1
+3,25,113,1,1,8,1
+3,26,113,1,0,8,1
+3,27,113,1,0,8,1
+3,28,113,1,0,7,1
+3,29,121,1,8,7,1
+3,30,121,6,0,7,1
+3,31,121,6,0,7,1
+3,32,121,6,7,7,1
+3,33,113,6,7,7,1
+3,34,121,6,7,7,1
+3,35,113,6,7,7,1
+3,36,113,6,7,2,1
+3,37,113,6,7,2,1
+4,8,113,4,9,8,1
+4,9,121,1,7,2,0
+4,10,113,3,7,8,0
+4,11,121,6,9,6,0
+4,12,121,9,1,7,1
+4,13,127,6,9,7,1
+4,14,127,6,1,7,1
+4,15,127,6,1,7,1
+4,16,127,9,1,7,1
+4,17,127,4,1,2,1
+4,18,127,6,1,2,1
+4,19,127,4,1,2,1
+4,20,113,6,2,2,1
+4,21,113,1,3,2,1
+4,22,113,6,7,8,1
+4,23,113,1,1,0,1
+4,24,113,1,2,8,1
+4,25,113,1,1,7,1
+4,26,113,1,8,2,1
+4,27,113,6,7,7,1
+4,28,121,6,0,2,1
+4,29,113,6,0,7,1
+4,30,121,6,8,2,1
+4,31,113,6,7,7,1
+4,32,121,6,7,2,1
+4,33,113,6,7,7,1
+4,34,121,6,7,2,1
+4,35,113,6,7,2,1
+4,36,113,4,1,2,1
+4,37,113,6,0,2,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
index 549102b2..67fdd593 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,4,9,8,1
-5,9,113,6,3,8,0
-5,10,104,6,3,8,0
-5,11,104,6,3,0,0
-5,12,104,9,3,0,0
-5,13,121,3,3,8,1
-5,14,121,6,3,8,0
-5,15,104,3,3,8,0
-5,16,104,3,3,0,0
-5,17,104,3,3,0,0
-5,18,104,3,3,0,0
-5,19,121,3,3,0,0
-5,20,104,3,3,0,0
-5,21,104,3,3,0,0
-5,22,104,3,3,0,0
-5,23,121,4,3,0,1
-5,24,121,6,3,8,1
-5,25,104,6,3,8,1
-5,26,104,6,3,8,0
-5,27,104,3,3,0,0
-5,28,104,3,3,0,1
-5,29,104,4,3,0,1
-5,30,104,9,3,0,1
-5,31,104,3,3,0,1
-5,32,104,3,3,0,1
-5,33,104,6,3,0,1
-5,34,113,6,3,8,1
-5,35,113,6,3,8,1
-5,36,113,6,3,8,1
-5,37,113,6,3,8,1
+5,8,113,0,7,4,1
+5,9,113,1,9,6,1
+5,10,113,1,9,2,1
+5,11,113,1,7,2,1
+5,12,113,6,9,7,1
+5,13,121,1,3,2,1
+5,14,113,6,7,7,1
+5,15,121,6,3,2,1
+5,16,113,6,0,7,1
+5,17,121,1,8,2,1
+5,18,113,6,7,7,1
+5,19,121,6,1,2,1
+5,20,113,6,0,2,1
+5,21,113,4,7,2,1
+5,22,113,6,1,2,1
+5,23,113,1,0,2,1
+5,24,113,1,7,2,1
+5,25,113,1,7,7,1
+5,26,113,1,7,2,1
+5,27,113,1,1,7,1
+5,28,121,6,9,2,1
+5,29,113,6,0,7,1
+5,30,121,1,8,2,1
+5,31,113,6,7,7,1
+5,32,121,6,7,7,1
+5,33,113,6,7,7,1
+5,34,121,6,7,2,1
+5,35,113,6,7,7,1
+5,36,121,6,1,2,1
+5,37,113,6,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
index 68e5ec32..f223e485 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,102,6,2,8,1
-6,9,113,6,3,8,1
-6,10,121,6,3,8,1
-6,11,121,6,3,8,1
-6,12,121,6,3,8,1
-6,13,121,6,3,8,0
-6,14,121,6,3,8,0
-6,15,104,3,3,8,0
-6,16,104,3,3,0,0
-6,17,104,3,3,0,0
-6,18,104,3,3,0,0
-6,19,104,3,3,0,0
-6,20,104,3,3,0,0
-6,21,104,3,3,0,0
-6,22,104,3,3,0,0
-6,23,121,6,3,0,1
-6,24,113,6,3,8,1
-6,25,121,6,3,8,1
-6,26,104,6,3,8,1
-6,27,104,6,3,8,0
-6,28,104,6,3,0,0
-6,29,104,9,3,0,1
-6,30,104,3,3,0,1
-6,31,104,9,3,0,1
-6,32,104,3,3,0,1
-6,33,104,3,3,0,1
-6,34,103,6,3,8,1
-6,35,113,6,3,8,1
-6,36,121,6,3,8,1
-6,37,121,6,3,8,1
-7,8,121,6,9,8,1
-7,9,104,6,3,8,1
-7,10,104,6,3,8,0
-7,11,104,6,3,0,0
-7,12,104,6,3,0,1
-7,13,121,6,3,8,1
-7,14,104,6,3,8,0
-7,15,104,3,3,0,0
-7,16,104,3,3,0,0
-7,17,104,3,3,0,1
-7,18,104,4,3,0,1
-7,19,103,6,3,8,1
-7,20,104,6,3,8,1
-7,21,104,9,3,0,0
-7,22,121,6,3,8,1
-7,23,104,6,3,8,1
-7,24,104,6,3,8,1
-7,25,104,6,3,8,1
-7,26,104,6,3,8,0
-7,27,104,3,3,0,0
-7,28,104,3,3,0,0
-7,29,104,3,3,0,0
-7,30,104,3,3,0,0
-7,31,104,3,3,0,1
-7,32,104,3,3,0,1
-7,33,103,4,3,0,1
-7,34,113,6,3,8,1
-7,35,113,6,3,8,1
-7,36,113,6,3,8,1
-7,37,104,6,3,8,1
+6,8,121,4,7,8,0
+6,9,121,6,7,8,0
+6,10,121,6,7,7,0
+6,11,121,6,7,7,1
+6,12,127,6,7,7,1
+6,13,127,4,7,0,1
+6,14,127,4,1,0,1
+6,15,127,4,1,0,1
+6,16,127,4,1,0,1
+6,17,127,4,1,2,1
+6,18,127,6,7,8,1
+6,19,113,6,1,0,1
+6,20,113,1,7,0,1
+6,21,113,6,7,0,1
+6,22,113,1,7,0,1
+6,23,113,1,7,0,1
+6,24,113,1,7,0,1
+6,25,113,6,7,0,1
+6,26,113,6,1,2,1
+6,27,113,6,0,2,1
+6,28,113,6,0,2,1
+6,29,113,6,0,2,1
+6,30,113,6,0,2,1
+6,31,113,6,8,2,1
+6,32,113,1,1,2,1
+6,33,113,1,8,8,1
+6,34,113,1,0,8,1
+6,35,113,1,8,7,1
+6,36,121,1,8,2,1
+6,37,113,6,7,7,1
+7,8,121,4,9,6,0
+7,9,121,0,3,8,0
+7,10,121,0,3,6,0
+7,11,121,0,3,7,0
+7,12,121,0,9,7,0
+7,13,127,0,9,7,1
+7,14,127,6,7,7,1
+7,15,127,6,1,7,1
+7,16,127,6,7,2,1
+7,17,127,4,1,8,1
+7,18,127,6,1,8,1
+7,19,127,6,1,8,1
+7,20,127,6,1,0,1
+7,21,113,1,7,0,1
+7,22,113,1,7,0,1
+7,23,113,1,7,0,1
+7,24,113,1,1,0,1
+7,25,113,1,4,0,1
+7,26,113,1,7,7,1
+7,27,113,6,7,2,1
+7,28,113,6,7,2,1
+7,29,113,6,7,7,1
+7,30,121,6,7,2,1
+7,31,113,6,7,7,1
+7,32,121,6,7,2,1
+7,33,113,6,7,7,1
+7,34,121,6,1,2,1
+7,35,127,6,0,8,1
+7,36,127,6,0,2,1
+7,37,127,4,0,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
index 84a9c524..c09630f3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,6,3,8,1
-8,9,121,6,3,8,1
-8,10,104,6,3,8,1
-8,11,104,6,3,8,1
-8,12,121,6,3,8,1
-8,13,121,6,3,8,1
-8,14,104,6,3,8,0
-8,15,104,9,3,0,0
-8,16,104,3,3,0,0
-8,17,104,3,3,0,0
-8,18,104,3,3,0,0
-8,19,104,3,3,0,0
-8,20,104,3,3,0,0
-8,21,104,3,3,0,1
-8,22,103,6,3,8,1
-8,23,121,6,3,8,1
-8,24,104,6,3,8,1
-8,25,104,6,3,8,1
-8,26,104,6,3,8,0
-8,27,104,6,3,0,0
-8,28,104,9,3,0,1
-8,29,104,9,3,0,1
-8,30,104,3,3,0,1
-8,31,104,3,3,0,1
-8,32,104,6,3,0,1
-8,33,113,6,3,8,1
-8,34,121,6,3,8,1
-8,35,121,6,3,8,1
-8,36,121,6,3,8,1
-8,37,104,6,3,8,1
+8,8,113,4,3,8,0
+8,9,121,1,3,8,0
+8,10,121,9,0,7,0
+8,11,121,4,9,7,1
+8,12,127,6,7,7,1
+8,13,127,6,7,7,1
+8,14,127,6,7,0,1
+8,15,127,6,1,0,1
+8,16,127,4,1,2,1
+8,17,127,4,1,2,1
+8,18,127,6,7,2,1
+8,19,113,6,7,0,1
+8,20,113,1,1,0,1
+8,21,113,1,2,8,1
+8,22,113,1,1,0,1
+8,23,113,1,2,8,1
+8,24,113,1,9,7,1
+8,25,113,6,7,7,1
+8,26,121,6,1,2,1
+8,27,113,6,0,7,1
+8,28,121,6,7,2,1
+8,29,113,6,0,7,1
+8,30,121,6,7,2,1
+8,31,113,6,0,7,1
+8,32,121,1,7,2,1
+8,33,113,6,7,8,1
+8,34,113,1,1,2,1
+8,35,113,6,7,8,1
+8,36,102,1,1,2,1
+8,37,113,6,7,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
index ad971442..ac971d29 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,113,6,3,0,0
-9,9,121,6,3,8,1
-9,10,104,6,3,8,1
-9,11,104,6,3,8,1
-9,12,104,6,3,8,1
-9,13,104,6,3,8,1
-9,14,104,6,3,8,0
-9,15,104,6,3,0,0
-9,16,104,9,3,0,2
-9,17,104,3,3,0,0
-9,18,104,3,3,0,1
-9,19,104,9,3,0,1
-9,20,103,6,3,8,1
-9,21,121,6,3,8,1
-9,22,104,6,3,8,1
-9,23,121,6,3,8,1
-9,24,121,6,3,8,1
-9,25,104,6,3,8,0
-9,26,104,9,3,0,0
-9,27,104,3,3,0,0
-9,28,104,3,3,0,0
-9,29,104,3,3,0,0
-9,30,104,3,3,0,0
-9,31,104,3,3,0,0
-9,32,104,3,3,0,1
-9,33,103,6,3,8,1
-9,34,121,6,3,8,1
-9,35,104,6,3,8,1
-9,36,104,6,3,8,1
-9,37,104,6,3,8,0
+9,8,121,0,9,2,0
+9,9,121,3,3,8,0
+9,10,121,3,0,8,0
+9,11,121,3,0,8,0
+9,12,121,3,0,7,0
+9,13,121,6,8,7,1
+9,14,127,0,7,7,1
+9,15,127,0,7,7,1
+9,16,127,4,7,7,1
+9,17,127,6,7,7,1
+9,18,127,6,7,7,1
+9,19,127,6,7,2,1
+9,20,113,6,7,2,1
+9,21,113,6,7,2,1
+9,22,113,6,7,2,1
+9,23,113,6,1,2,1
+9,24,113,1,2,8,1
+9,25,113,1,1,8,1
+9,26,113,1,0,8,1
+9,27,113,1,0,8,1
+9,28,113,1,0,7,1
+9,29,121,1,8,7,1
+9,30,121,6,0,7,1
+9,31,121,6,0,7,1
+9,32,121,6,7,7,1
+9,33,113,6,7,7,1
+9,34,121,6,7,7,1
+9,35,113,6,7,7,1
+9,36,113,6,7,2,1
+9,37,113,6,7,2,1
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
index 498e4491..feb89b42 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0091290055,0.0077892784,0.007239608,0.020136122,0.006804843,0.065107465,0.09113526,0.03846739,0.046789855,0.011856737,0.010647341,0.0430088,0.01229634,0.024512613,0.043958448,0.015431894,0.10510967,0.008920799,0.012439459,0.013178185,0.0115483245,0.042840227,0.020539861,0.013582593,0.07060103,0.024588121,0.021217003,0.10432629,0.008402498,0.025339192,0.025642872,0.0106556,0.026757332
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
index 2a1876ce..d7938c94 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.008205366,0.008331864,0.0074459263,0.021188624,0.0073211472,0.061644685,0.07545811,0.034722853,0.043902382,0.010114443,0.010213885,0.042470828,0.012390661,0.022267908,0.032764185,0.01806368,0.12040593,0.009458063,0.014117386,0.011872736,0.013330289,0.041174386,0.020099496,0.013925595,0.072014526,0.024408026,0.018947104,0.13367255,0.00828295,0.024368336,0.02391859,0.011060069,0.022437433
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
index dd67b48c..06ba23c9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.012586267,0.009460248,0.01056256,0.024452873,0.0086881975,0.052435704,0.10992195,0.0706047,0.03934652,0.017093472,0.012070613,0.038141258,0.016003482,0.024356524,0.05110281,0.022539876,0.057360444,0.0130562065,0.014306477,0.011672565,0.013660459,0.019611526,0.033125043,0.01840493,0.062889695,0.025482355,0.023796566,0.06978374,0.011446344,0.029403456,0.027195802,0.014777219,0.034660116
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
index 426a7d74..d6044728 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.008913519,0.007365497,0.007884704,0.019221375,0.008245324,0.061635274,0.10626261,0.04613362,0.039076947,0.013314558,0.01237192,0.04430893,0.014824947,0.02413848,0.04149791,0.018489841,0.07907649,0.010472187,0.011848482,0.012690502,0.011933226,0.034617245,0.024047974,0.015420446,0.064966045,0.021884238,0.020587962,0.12284686,0.008948034,0.029838102,0.020556517,0.010585136,0.025995124
 0.0094871465,0.011149254,0.009741706,0.023295518,0.012869805,0.05367749,0.048896357,0.031161776,0.05537468,0.012018335,0.013214679,0.028068008,0.016052,0.03157924,0.03311481,0.021816775,0.052957285,0.012617017,0.022129932,0.015422226,0.02169244,0.031480502,0.015010904,0.02252592,0.07512524,0.07089812,0.022424813,0.11010506,0.011601552,0.021417523,0.044051338,0.016005058,0.02301758
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
index 10cd194d..d211d4e4 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.00905651,0.008449699,0.007513583,0.01789081,0.009958172,0.059586458,0.09060577,0.05163472,0.05080496,0.015027517,0.0131017715,0.046035837,0.0129659865,0.021924116,0.043815948,0.022697417,0.07993372,0.0115594845,0.015695909,0.011922975,0.012971503,0.034338184,0.025293143,0.016621767,0.06711588,0.022368787,0.02659395,0.0939604,0.010572614,0.026793035,0.024672363,0.012390149,0.026126813
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
index e7d6329d..8e0cfcaa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.008020919,0.0066487696,0.0072186273,0.023239477,0.009595765,0.059273914,0.062329326,0.027587652,0.0486198,0.008902664,0.009877493,0.03435876,0.017895322,0.022803104,0.0360042,0.016689297,0.112537004,0.008584821,0.014853362,0.013560702,0.015122532,0.043928403,0.0197575,0.01412689,0.07697556,0.037536968,0.014428177,0.14790235,0.008241942,0.016693683,0.024688615,0.013138984,0.018857364
 0.0086295465,0.007917115,0.007522475,0.023743981,0.0073922593,0.06037666,0.08741754,0.043700412,0.04167852,0.012650199,0.01019312,0.043437473,0.013766954,0.021962289,0.04165744,0.018287549,0.10325626,0.009215165,0.01214539,0.011779306,0.012235232,0.045917273,0.018463248,0.014126799,0.068317495,0.021482795,0.016241094,0.12552734,0.008585244,0.025119003,0.02262137,0.01049176,0.02414174
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
index 0efa80de..e2701732 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.00940564,0.0063802544,0.0065206103,0.027256832,0.0066316505,0.05138203,0.10203898,0.048445128,0.049608555,0.00873864,0.010182529,0.026134778,0.014573924,0.02411956,0.038605724,0.036600724,0.07013269,0.011340922,0.020680869,0.011217838,0.01796068,0.016984288,0.026266541,0.013434889,0.07714,0.08002693,0.012619701,0.06227914,0.007597093,0.02409692,0.043419037,0.014213739,0.0239632
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
index 30a706e1..99ee0e82 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.009661113,0.007831191,0.009208629,0.018906532,0.008831174,0.065691166,0.10545667,0.038684368,0.039912082,0.01415859,0.013239993,0.03936315,0.0131430235,0.025695173,0.04284285,0.015051846,0.08823259,0.009918131,0.012693808,0.015639171,0.01375921,0.053180333,0.021613518,0.016188124,0.064060025,0.024109922,0.022105852,0.08915296,0.01004957,0.026042204,0.028148724,0.010506249,0.026922056
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
index 93284692..a2fc7cd2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0042463774,0.005140092,0.0060796365,0.016366256,0.0067424467,0.03495936,0.06623093,0.020122634,0.050006244,0.00649412,0.0041204235,0.02871717,0.018642396,0.028757636,0.033106435,0.01599839,0.23275402,0.007276528,0.008622721,0.017800169,0.010982657,0.028370103,0.02974473,0.0034585185,0.06776658,0.046286162,0.0064784237,0.120085396,0.0039499593,0.021950865,0.015566807,0.018041762,0.015134025
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
index d5bd2f61..609b1f0f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0041903867,0.0054771546,0.0058246264,0.015669782,0.0071306545,0.030696195,0.06473787,0.018447235,0.05242684,0.0068101883,0.004801957,0.029178623,0.022633398,0.02907579,0.033386257,0.018674878,0.19909294,0.006572294,0.009357265,0.01596702,0.013374633,0.026702495,0.032056045,0.003591021,0.07047866,0.053875104,0.0050279247,0.13526632,0.0042218207,0.021093741,0.021248914,0.019235754,0.013676274
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
index ae9e3e1a..a7ac31d6 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0054918397,0.0063162544,0.0061463765,0.016365495,0.0076878374,0.034990918,0.0610308,0.020188866,0.0626216,0.008143705,0.0056989286,0.028557653,0.01842186,0.026081735,0.02933535,0.018272739,0.24774212,0.008434298,0.009743969,0.017181823,0.013515602,0.028113525,0.035016388,0.004349286,0.05375894,0.048733603,0.0074889436,0.08502142,0.0040941467,0.018917236,0.024357485,0.024139129,0.014040137
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
index 71f47b47..942890e0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.004381987,0.006287049,0.006620055,0.019419348,0.006867307,0.03929928,0.07064888,0.017149057,0.060524255,0.006789356,0.0050974856,0.027337514,0.024331838,0.03136498,0.029390393,0.020899236,0.15376608,0.0074551413,0.009935729,0.015628591,0.013709809,0.03817771,0.03494091,0.0051154993,0.06018743,0.075591475,0.0067811315,0.11117232,0.0047863433,0.015857743,0.028808644,0.024059782,0.017617658
 0.004469706,0.006364652,0.0068032146,0.020612448,0.007166283,0.039386127,0.08379896,0.017597469,0.06735702,0.007070649,0.005259082,0.027966166,0.024157904,0.03855127,0.026747128,0.019213978,0.11608029,0.0063566845,0.010583058,0.01697354,0.012141066,0.043337125,0.031579476,0.004105956,0.061916213,0.07502441,0.0060634622,0.11869589,0.005418634,0.019128878,0.027781855,0.025369624,0.01692181
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
index 0e9d5181..7f7b97c7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.005170072,0.0052365907,0.00694314,0.017285356,0.008234528,0.037991844,0.07294639,0.022531565,0.05857518,0.0065963003,0.005045354,0.026125679,0.020520791,0.035622887,0.0324971,0.01749203,0.16650967,0.007892656,0.009556492,0.017534178,0.011642951,0.03663207,0.033574145,0.0035568238,0.07068761,0.048016,0.0072974605,0.12339532,0.0045464407,0.022555495,0.018934185,0.020668464,0.01818533
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
index 87181f3b..220a5080 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0050350777,0.005766118,0.006457709,0.019448206,0.007035608,0.036554355,0.06908976,0.016720336,0.073623985,0.007380201,0.0050186585,0.027139133,0.019853646,0.03107149,0.027048368,0.018990275,0.18321688,0.0086597875,0.00932425,0.017520232,0.013038053,0.03883165,0.028723942,0.0043549896,0.060425516,0.060644113,0.0076746633,0.10625886,0.0049037784,0.016128076,0.024719888,0.024399208,0.014943133
 0.0042376784,0.005924756,0.0069834692,0.018832779,0.0078081987,0.04281036,0.07998875,0.020728823,0.057859186,0.006507186,0.005319783,0.035538692,0.020944735,0.03357084,0.033523228,0.017411817,0.14398022,0.0064492146,0.010395881,0.018382689,0.010860411,0.03802423,0.034201827,0.003840388,0.06658509,0.059049625,0.0067133494,0.10930205,0.00455577,0.024278667,0.023504738,0.026207874,0.015677737
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
index 5f3dd3e1..83108e29 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.005213573,0.005612019,0.005980635,0.014423169,0.008148407,0.035215456,0.08324456,0.021167181,0.05202259,0.006690955,0.0051461984,0.03389285,0.022059022,0.028906152,0.04061394,0.018913923,0.16657288,0.005920637,0.008955144,0.018136969,0.010939499,0.027240243,0.03896024,0.0037672839,0.06677831,0.05034332,0.006739948,0.11629482,0.004037235,0.02923933,0.021340776,0.020757355,0.016725339
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
index 57dae23d..d3b4ccb7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.004618134,0.006132982,0.006672614,0.015618401,0.0074373814,0.03397381,0.07677221,0.019593101,0.05670967,0.0069386126,0.0054081944,0.030513486,0.021875117,0.02945475,0.029283365,0.017094098,0.20585725,0.007897137,0.009288681,0.016525654,0.011750388,0.03215573,0.031738274,0.0035157073,0.06421899,0.043070234,0.00695176,0.11497121,0.004500923,0.020835267,0.023099473,0.021389142,0.014138295
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 67d3a1aa..7e5bfc90 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.016444147,0.017504703,0.010868917,0.024397992,0.02378631,0.05476737,0.018488651,0.01870659,0.03153483,0.01417574,0.018273251,0.04281981,0.028775947,0.030504966,0.017300766,0.020425139,0.019642664,0.01798995,0.01798995,0.06529241,0.026927112,0.06377991,0.09389312,0.034339547,0.03991245,0.021912975,0.010130945,0.04549251,0.012461275,0.0647843,0.021114701,0.01657312,0.03898788
 0.018649058,0.011534323,0.009525009,0.04894217,0.016076498,0.027988719,0.09323943,0.032794107,0.024086585,0.014523906,0.023028513,0.028638465,0.056332096,0.02776413,0.029872134,0.031101944,0.02357462,0.010298991,0.0192035,0.01811063,0.018979773,0.07148915,0.060672056,0.015280495,0.074919984,0.019054057,0.016522154,0.039403025,0.012767332,0.009750889,0.041294016,0.035528205,0.019054057
 0.009243722,0.015240321,0.012585413,0.012245942,0.011020156,0.044706393,0.04126598,0.022088086,0.022789238,0.016350478,0.017748276,0.052165084,0.025299398,0.037463184,0.07560385,0.03431107,0.060041644,0.014372992,0.013344871,0.014485721,0.019722441,0.03816479,0.025200764,0.012536347,0.06316938,0.05865078,0.021873433,0.038314164,0.036381554,0.024521017,0.025949989,0.020953465,0.062190026
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 38cf6ada..c1bad420 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.018813502,0.0097793965,0.019639514,0.023574388,0.030255334,0.08317135,0.027166266,0.015935654,0.042812943,0.011081507,0.016963415,0.03456955,0.039287366,0.025708733,0.03456955,0.021297682,0.02741194,0.017777506,0.02591037,0.016123498,0.03373579,0.01538515,0.036263976,0.033180345,0.039828185,0.11821033,0.015935654,0.05716278,0.01116842,0.03830239,0.035701755,0.008766194,0.014509578
 0.013969102,0.009713985,0.01286894,0.013969102,0.015045315,0.06638303,0.04526929,0.020847589,0.039484575,0.014639493,0.013645508,0.046162147,0.04162265,0.021467391,0.05503346,0.02815642,0.026431143,0.014356339,0.012231754,0.020206176,0.039484575,0.038720876,0.0321792,0.0149866585,0.092163704,0.023808694,0.018834226,0.04049995,0.016524045,0.019508144,0.04734924,0.046797603,0.03763967
 0.017729616,0.014929833,0.020768544,0.022854377,0.015403757,0.04781904,0.025558222,0.03895261,0.058589116,0.019701704,0.013646972,0.08070991,0.04781904,0.021261055,0.05070465,0.018726205,0.047077674,0.024115471,0.022086421,0.015646331,0.028528795,0.032546125,0.01435788,0.021136845,0.0401891,0.06310296,0.017626034,0.050903104,0.02488098,0.021302622,0.018873077,0.016655432,0.025796438
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index ac0e5c1a..4bb17616 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.014180237,0.016257722,0.02075337,0.015035935,0.023724249,0.09210627,0.033686046,0.016321354,0.029331319,0.01988051,0.02091614,0.022904685,0.020075608,0.043507922,0.036637343,0.019610556,0.02790456,0.0151538635,0.01619434,0.025217365,0.032018133,0.042417135,0.02589112,0.017960545,0.12688175,0.066083014,0.015881114,0.024741694,0.022048742,0.031521738,0.026837192,0.013960393,0.02435811
 0.010987974,0.009621391,0.008102024,0.010004664,0.01264708,0.09947699,0.039184675,0.038127735,0.026918061,0.008229612,0.013890101,0.029495217,0.02808619,0.054296385,0.039108217,0.046989605,0.02911958,0.009696852,0.010649909,0.018258192,0.032708064,0.031547327,0.04019234,0.021597609,0.04019234,0.045543887,0.023859728,0.069717936,0.013890101,0.030031558,0.07900077,0.007976414,0.02085151
 0.016185524,0.016569352,0.011432509,0.008799916,0.01716223,0.0705819,0.029802466,0.025813933,0.0328878,0.008075258,0.016122423,0.023230024,0.018376462,0.034974713,0.0608454,0.10472156,0.020986577,0.009367462,0.015996957,0.017637983,0.0325523,0.024179006,0.022515312,0.025738416,0.0861416,0.026348786,0.020620897,0.04803876,0.019523472,0.043314718,0.05390604,0.013629578,0.02392066
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 58c7fbc8..b9714301 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.013940462,0.019203799,0.019620843,0.012943281,0.026376959,0.08539569,0.034201056,0.016171219,0.036371298,0.017553693,0.018832363,0.023096437,0.022320347,0.02289433,0.035982683,0.026197264,0.0309963,0.018396111,0.017349185,0.024706373,0.039096992,0.03601784,0.021845924,0.017417088,0.139698,0.046610545,0.013511559,0.03353955,0.018649349,0.01886918,0.03457043,0.012447432,0.04517649
 0.018725712,0.011787114,0.01107297,0.033447456,0.019894524,0.08055031,0.03819819,0.035431203,0.03186907,0.018291932,0.019207258,0.02245556,0.13702068,0.054503065,0.023191012,0.02562002,0.01916978,0.013834468,0.016687555,0.02840035,0.030843345,0.021892576,0.026087089,0.01579946,0.04885623,0.023418596,0.015923375,0.042116705,0.01549387,0.026875785,0.03695072,0.019433666,0.016950345
 0.020445274,0.016654389,0.013200464,0.028482636,0.033642784,0.083429486,0.028041055,0.019777574,0.025619086,0.021657871,0.030393722,0.014611581,0.022830635,0.123300776,0.01745365,0.037973646,0.017728506,0.027734673,0.01872503,0.024196517,0.019738983,0.021280492,0.021657871,0.019509016,0.022542626,0.05344693,0.025725637,0.07596341,0.020049827,0.038271476,0.015432905,0.013780033,0.026701493
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index daeca86b..c35131db 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.015050883,0.01172164,0.012093725,0.022309033,0.010756382,0.094016306,0.022882726,0.03638838,0.029932264,0.010672675,0.015347736,0.027256964,0.022027602,0.037543472,0.08695087,0.023957483,0.07613672,0.010714447,0.015407805,0.012674115,0.016659811,0.021770975,0.033035073,0.019553527,0.0648693,0.03820925,0.009755615,0.032081287,0.012924089,0.10570537,0.0215594,0.0116304215,0.018404752
 0.01221852,0.020105632,0.011523162,0.04298132,0.016506167,0.056940995,0.022168143,0.083172925,0.026531791,0.014566641,0.0132113695,0.06253745,0.035355467,0.036727987,0.024466112,0.017230876,0.11821256,0.01090991,0.019639885,0.026805244,0.014284897,0.02955142,0.036549088,0.017708533,0.03152267,0.038228378,0.0096279625,0.02415747,0.01929767,0.024778698,0.03875462,0.014008602,0.029717823
 0.0134996595,0.008414936,0.008818777,0.03387192,0.012195926,0.014370312,0.033657596,0.01964189,0.031905293,0.0134996595,0.019757316,0.038307022,0.07952743,0.028943907,0.039215446,0.05639339,0.046569496,0.01007136,0.015844474,0.011457012,0.019000426,0.048709184,0.071847044,0.01907479,0.08498784,0.01888942,0.009314485,0.045937136,0.01634743,0.01713196,0.024065744,0.06390209,0.024829673
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index eb9b1110..6fdccfca 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.013031122,0.011321634,0.010389316,0.012531906,0.014766195,0.104928374,0.035284165,0.027871214,0.020864384,0.010389316,0.017811432,0.024512332,0.020182995,0.08139971,0.029885026,0.043062516,0.016154267,0.010552924,0.013817477,0.024947023,0.036833324,0.051740803,0.04085048,0.033833183,0.039362326,0.033146404,0.022692353,0.050938636,0.017263431,0.053175114,0.046470832,0.012385906,0.017603923
 0.013867004,0.0117917815,0.01430719,0.011837933,0.013572258,0.038663756,0.02143049,0.018033074,0.02613567,0.012749956,0.014475838,0.04330135,0.019403422,0.029014427,0.033526108,0.025405882,0.021388676,0.013921278,0.016086007,0.024153845,0.020156674,0.32372767,0.016957058,0.01336184,0.04830615,0.016403275,0.014877124,0.025381085,0.016825097,0.016661588,0.018156769,0.017359182,0.028760536
 0.02343217,0.018681755,0.01668093,0.015979351,0.02779923,0.038860146,0.042846564,0.01560919,0.039012242,0.023295274,0.038633116,0.022228505,0.060541473,0.026786525,0.037848905,0.03317399,0.03386134,0.021544605,0.028099464,0.034529194,0.031596933,0.040725086,0.03131662,0.013299196,0.04855152,0.04276296,0.021798568,0.028569853,0.029247366,0.014549365,0.03781196,0.029795108,0.03053148
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 74d4cc81..74de4f73 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.01910463,0.02436371,0.021312755,0.02668002,0.029352391,0.070567615,0.022909949,0.013841413,0.03352967,0.023545036,0.025808707,0.034057684,0.031127442,0.056261405,0.02002148,0.035972085,0.04840573,0.025758348,0.038366944,0.02652415,0.030842822,0.029618738,0.034543376,0.017982226,0.04520731,0.032403,0.018123262,0.04100125,0.022118514,0.023962572,0.035972085,0.020257486,0.020456282
 0.013497088,0.009385752,0.010718865,0.022892157,0.017195737,0.047109466,0.042228654,0.023229958,0.034365196,0.015841454,0.01488167,0.039592747,0.037339516,0.025289863,0.04841539,0.025228195,0.036049604,0.009608328,0.015474488,0.012289238,0.024380473,0.03366762,0.052554406,0.010148417,0.2136187,0.02663978,0.014536936,0.043314595,0.011321378,0.008347861,0.026646286,0.017845849,0.016344316
 0.009544797,0.0067682746,0.008827494,0.02225743,0.00889673,0.03658895,0.043281052,0.023995617,0.03151086,0.010858004,0.017283395,0.049428515,0.06729738,0.028021205,0.06756077,0.07025208,0.047720984,0.008758798,0.011423627,0.010320387,0.020766476,0.04203145,0.031113345,0.019470233,0.094904505,0.027490968,0.008489318,0.034540378,0.013618987,0.026697252,0.028558182,0.05139752,0.02032508
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index d407a31f..b406953d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.015481911,0.021790396,0.019840382,0.03492312,0.020390378,0.059813716,0.027535558,0.028744513,0.04444978,0.020833192,0.014430728,0.048249777,0.029595483,0.019685984,0.057973444,0.018601947,0.035025585,0.018277813,0.016224904,0.025851479,0.029047865,0.061471812,0.022220174,0.02191845,0.041189697,0.06392057,0.01972447,0.050961923,0.024214173,0.033323877,0.018894885,0.01597336,0.019418672
 0.013983444,0.010351101,0.01371298,0.024782542,0.011960674,0.073268,0.036626223,0.02346937,0.02346364,0.022808464,0.020010805,0.03494899,0.019855078,0.04996443,0.029229501,0.023452187,0.026665796,0.013343096,0.015357809,0.02130139,0.01657335,0.20548517,0.04207427,0.009028391,0.04006921,0.015752748,0.028000148,0.020666188,0.018927356,0.024217322,0.020828273,0.027378248,0.02244389
 0.012367314,0.014458855,0.009707215,0.026529063,0.018349245,0.059930302,0.13718182,0.03231419,0.018601837,0.015572799,0.02426137,0.023676222,0.026946833,0.040948816,0.021202514,0.05785998,0.014515445,0.011172936,0.013424592,0.019570857,0.021961171,0.045237765,0.03869391,0.013635999,0.041593663,0.022746976,0.036456212,0.028863892,0.020113382,0.012033726,0.06530834,0.032015786,0.022746976
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 887fec91..c5305b93 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.021211134,0.028320579,0.023478584,0.058909986,0.13095303,0.09981858,0.10119266,0.05229339,0.18758154,0.118306205,0.10178733,0.036364343,0.039782714
 0.016987843,0.024144594,0.015347238,0.05002766,0.08610305,0.046630908,0.095680416,0.25555444,0.118380524,0.08460274,0.08410848,0.03582306,0.08660904
 0.03188173,0.04349213,0.038607225,0.03515228,0.05744916,0.060088728,0.064544536,0.08375145,0.061064795,0.27327284,0.060530446,0.032006513,0.15815818
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 3b040be2..5dc72934 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.036800325,0.042522695,0.031600185,0.123525076,0.04186344,0.10062151,0.12113588,0.06976658,0.029225392,0.18399003,0.06321378,0.057444494,0.098290615
 0.023137681,0.02156669,0.017740281,0.07512827,0.09349864,0.04173359,0.115230136,0.07210918,0.043226883,0.15873621,0.05558578,0.03740975,0.24489695
 0.03727073,0.06427239,0.031631254,0.054867815,0.053701587,0.09143873,0.115476556,0.026950166,0.027481709,0.1778087,0.052355234,0.033020034,0.23372512
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index a9e05bb0..dc7b23c0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.020409178,0.02520197,0.032233812,0.12212525,0.036383294,0.0602501,0.0626177,0.04644414,0.08024828,0.33719864,0.04397243,0.08328219,0.04963303
 0.018363692,0.022676133,0.023579445,0.082380764,0.027245997,0.11293152,0.13254708,0.07497192,0.07754087,0.18692133,0.097163096,0.057175048,0.08650309
 0.015567869,0.014854965,0.02337014,0.12197339,0.0134203425,0.0694305,0.18382038,0.11868336,0.12584524,0.1414916,0.04014407,0.09117559,0.04022255
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index e5aef4b6..b1f3057f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.021790942,0.02875576,0.023932666,0.11462324,0.033882566,0.074223466,0.087884694,0.03663579,0.07831908,0.31525108,0.041190725,0.09521178,0.048298124
 0.019185683,0.01774385,0.017605767,0.11346654,0.04976387,0.07859597,0.28414276,0.10032964,0.06620061,0.09333503,0.042648513,0.07882658,0.038155284
 0.049818195,0.04616438,0.07234362,0.14387244,0.037678268,0.15077703,0.11031084,0.13836077,0.054394927,0.069640145,0.04336742,0.054607823,0.028664112
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index f5f64a91..7a6d6645 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.021908142,0.024061386,0.026947476,0.05529263,0.035699606,0.055726297,0.103704534,0.12558115,0.045839198,0.3591476,0.039055485,0.076020196,0.031016354
 0.031836722,0.03900717,0.033890016,0.036358677,0.06507003,0.07140441,0.04242466,0.11265075,0.088420585,0.23159625,0.14606592,0.054314982,0.04695982
 0.0118632335,0.019789722,0.010469266,0.04605744,0.101288736,0.03852003,0.21526752,0.2169559,0.096273586,0.13005732,0.03852003,0.046919998,0.02801721
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 849903e2..f5d1ffa1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.019913744,0.02274221,0.032832213,0.12439244,0.026381426,0.079455666,0.09334777,0.11459569,0.13293332,0.11731324,0.09190055,0.060269855,0.08392191
 0.017521279,0.033905335,0.016459718,0.045982454,0.0826155,0.044134606,0.0711495,0.08046563,0.19895993,0.1259736,0.060649738,0.052003276,0.17017946
 0.025699943,0.03512766,0.01985938,0.048484996,0.0704764,0.038580187,0.07725216,0.15866601,0.17562728,0.14334281,0.07823914,0.10631546,0.0223285
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 31f376e9..81f95111 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.044508155,0.08047588,0.044508155,0.05337343,0.06501244,0.10866292,0.057261333,0.064758986,0.11499529,0.079421885,0.09061411,0.14939743,0.047009982
 0.012853926,0.02102762,0.016504768,0.088702664,0.063192055,0.041492984,0.12681279,0.11774168,0.07601974,0.21071875,0.06582174,0.07075447,0.088356845
 0.011193676,0.014945519,0.013086734,0.07242391,0.087019324,0.05896654,0.2649157,0.11393881,0.05971991,0.20471099,0.019190425,0.03407726,0.045811202
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index be998e4d..55254927 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.03258783,0.04750771,0.034689564,0.0454208,0.056416802,0.106019855,0.053991236,0.023748815,0.043766156,0.28041837,0.04988499,0.026701512,0.1988464
 0.02455211,0.0374375,0.024744675,0.051875398,0.11622004,0.027604679,0.038625892,0.22227663,0.1474909,0.11622004,0.06943205,0.04445813,0.079061925
 0.027025884,0.03122835,0.031596463,0.08698541,0.055997707,0.056547236,0.067414336,0.21009848,0.09963192,0.10380335,0.10002187,0.03098533,0.09866369
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index a629f483..6a547fcd 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.017289048,0.016052337,0.023816664,0.022026813,0.09165594,0.034517944,0.036103908,0.02730594,0.11362295,0.14877227,0.39813527,0.0136234425,0.05707744
 0.019879099,0.019879099,0.028475493,0.0552105,0.080095805,0.04910524,0.08279988,0.06850958,0.17240135,0.12712078,0.11001386,0.107885994,0.07862336
 0.028068636,0.036751684,0.030827366,0.19407536,0.14006053,0.1583997,0.07188607,0.0728045,0.052336797,0.053473387,0.031681933,0.09230354,0.037330437
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 0cfd51a0..97f72b85 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.028672202,0.035683133,0.04295809,0.05304616,0.20614591,0.06991537,0.099297024,0.038885273,0.108631276,0.13206205,0.030881178,0.055105437,0.09871691
 0.01782257,0.03189652,0.03448836,0.090687424,0.24411833,0.09002563,0.051571313,0.08114297,0.06220688,0.11419271,0.09202563,0.054790262,0.03503147
 0.038860124,0.03483399,0.03840739,0.11969761,0.24464603,0.101586044,0.07508794,0.03708059,0.077282585,0.058908433,0.050584264,0.046418726,0.07660632
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index d1779af8..be8b8095 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.020406444,0.035674796,0.030395307,0.051704157,0.2998705,0.12500462,0.033382714,0.056234058,0.076600574,0.12598504,0.067698866,0.03051427,0.046528704
 0.015827617,0.024323527,0.020968191,0.17016296,0.15676147,0.063647375,0.036371585,0.03074782,0.1898305,0.057951596,0.07677339,0.11029547,0.0463386
 0.017605642,0.022783393,0.029254457,0.20951222,0.16063859,0.06966567,0.046384685,0.0307784,0.1224463,0.08047888,0.09829233,0.08552304,0.026636485
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 02fe2c42..0c828e54 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.019000068,0.031819124,0.030243652,0.07024996,0.21407309,0.13396353,0.028858693,0.026585873,0.06696755,0.16285823,0.15180056,0.022563139,0.041016474
 0.017298587,0.02785987,0.026171926,0.022916906,0.29258338,0.047437537,0.034001548,0.03626522,0.08464918,0.20266658,0.11324245,0.06333742,0.03156937
 0.0686808,0.047946926,0.05949635,0.06376796,0.06336447,0.15089422,0.07095185,0.051540107,0.055402566,0.20384508,0.026530573,0.08195969,0.055619407
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index a5de7340..15afec14 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.022081608,0.018594578,0.018740417,0.103307895,0.47212556,0.04692973,0.030657277,0.045264293,0.059411697,0.078746155,0.02350575,0.05239225,0.028242808
 0.027631663,0.025856346,0.021603787,0.16926447,0.12096786,0.07964316,0.043301236,0.1349494,0.09815427,0.05714117,0.099311285,0.0701477,0.052027624
 0.026684716,0.020863418,0.016504321,0.07736575,0.10974434,0.03814951,0.06553134,0.049032796,0.21570995,0.21236569,0.061913315,0.043144707,0.062990114
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index d58192d2..bf39a092 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.016777907,0.029446134,0.027233222,0.09903301,0.15368521,0.065074414,0.049991827,0.071156874,0.1554968,0.06006656,0.10398861,0.11969016,0.048359197
 0.023548124,0.02840446,0.027638298,0.0627111,0.20662825,0.07437815,0.030832749,0.047290612,0.15904744,0.048412077,0.11411094,0.045656934,0.13134089
 0.020972665,0.028554477,0.028778432,0.029231627,0.05453193,0.057009615,0.037976593,0.024233868,0.18987098,0.32806766,0.0779229,0.04914536,0.073703915
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 19b076a5..4e43aa2e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.04036501,0.0526459,0.035483122,0.057146583,0.056979407,0.10676387,0.06257964,0.06425163,0.12384826,0.17465405,0.050137047,0.100884765,0.07426074
 0.0159961,0.026167873,0.02203556,0.084553465,0.22782892,0.074536234,0.040371586,0.03837265,0.13551247,0.18814044,0.07248592,0.03346922,0.040529598
 0.018406145,0.016499165,0.01400266,0.088499516,0.33661324,0.033329222,0.051798053,0.042049877,0.14087047,0.096441306,0.039967842,0.07855951,0.042963065
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 51ade0ca..d35903cd 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.04502345,0.037545096,0.032875586,0.1290136,0.23822105,0.09531464,0.04755424,0.04746145,0.110351175,0.038585886,0.04271068,0.039655525,0.09568769
 0.024061177,0.0339317,0.034197826,0.027478727,0.10804553,0.092235915,0.03138169,0.15936935,0.15446608,0.06424972,0.027264884,0.13900442,0.104313046
 0.01398243,0.020826768,0.033673387,0.059273835,0.023971463,0.071147464,0.054311454,0.12708203,0.15815614,0.043747153,0.033022083,0.31823865,0.04256715
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
index 47225f8b..a0bbf05d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0051731854,0.006115724,0.004452979,0.02984497,0.0035049068,0.023042496,0.04243705,0.017752096,0.043330517,0.005744317,0.005712806,0.02443036,0.03076382,0.016793936,0.020566238,0.014915136,0.23716973,0.006960561,0.016510267,0.008906468,0.016893843,0.032925818,0.022828635,0.0059587103,0.08213512,0.101522684,0.009693471,0.08036983,0.0065862564,0.016128413,0.025845902,0.011681713,0.02330209
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
index aab7381f..68d35a3b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0051706634,0.006392545,0.004375827,0.030993352,0.0034347752,0.021954894,0.041624684,0.017873196,0.04168018,0.00586722,0.005894553,0.022671,0.0292719,0.01654101,0.019867422,0.013532966,0.2661518,0.007395097,0.01631813,0.008761997,0.016077364,0.033230375,0.020873664,0.0064080334,0.07242437,0.097736195,0.008801166,0.08053392,0.0065339063,0.015795996,0.02301082,0.012271813,0.02052921
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
index abcdbcaa..6ae5b497 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0064157196,0.006527345,0.0056665037,0.027466085,0.0044602095,0.026517421,0.049759857,0.01652433,0.041261666,0.006874849,0.0067948974,0.023469295,0.029153131,0.016635327,0.019607816,0.016634526,0.17611296,0.007703006,0.023564544,0.010035094,0.022443883,0.03051713,0.025578223,0.0072342893,0.08143584,0.14751312,0.013761597,0.055317592,0.0074805734,0.01573484,0.03309507,0.01149615,0.027207157
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
index 8abda85f..99c55a45 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0055406922,0.006386817,0.004625329,0.02794532,0.0038382933,0.023122994,0.047041234,0.017717674,0.043066762,0.0060848948,0.006028181,0.024094954,0.031411633,0.017373681,0.020232791,0.016327478,0.21710728,0.0073115835,0.017294787,0.008706568,0.019171692,0.034296166,0.022187935,0.0064854,0.08433365,0.10816069,0.01044296,0.07764751,0.0066119726,0.016789336,0.028201265,0.012045506,0.022366948
 0.005273,0.006350921,0.0045870915,0.029855195,0.003674753,0.022280036,0.04056523,0.018330282,0.042522132,0.0058698533,0.005789554,0.023899432,0.030028421,0.016782813,0.018927349,0.014220322,0.25088266,0.0070974855,0.017512802,0.008408013,0.01705966,0.032813244,0.022579813,0.0065027904,0.08215564,0.09919083,0.009322651,0.072932474,0.0066912686,0.015727002,0.02790005,0.011463137,0.022804117
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
index 1adcaec0..ac00fb4a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.005243044,0.0068628322,0.004119111,0.027635053,0.0037787466,0.021621196,0.03984544,0.020376597,0.036851183,0.005732252,0.0054553016,0.02128327,0.026677908,0.019110823,0.01986349,0.013093736,0.30414927,0.006873597,0.014688124,0.007742951,0.014003141,0.031778734,0.020403402,0.006862341,0.07797771,0.064194806,0.0077092745,0.088538475,0.0069644405,0.016042879,0.022522885,0.012781177,0.019216869
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
index c03ca665..545f567e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.005483725,0.0066012214,0.005198228,0.033509348,0.0042043817,0.02317797,0.03937999,0.017307758,0.037636165,0.005854237,0.0058676037,0.02215927,0.024298837,0.016519763,0.01830959,0.013106523,0.27282035,0.0074146762,0.019438406,0.0094330795,0.016409066,0.027353697,0.023160962,0.0076946104,0.074612305,0.11091272,0.0100175105,0.06297743,0.007235825,0.013821495,0.024065388,0.011177126,0.022840766
 0.005132123,0.006258132,0.004523277,0.028863266,0.0038300497,0.02196964,0.043445766,0.018415095,0.051318515,0.0059415693,0.0059717335,0.024573747,0.03365483,0.017032608,0.022367323,0.015106722,0.21536613,0.0069446526,0.016772965,0.008130857,0.016136188,0.037611797,0.023627339,0.005972406,0.0856897,0.09184303,0.009807569,0.08422559,0.0065450855,0.01732655,0.02972546,0.012266841,0.023603454
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
index 8fa8b147..981d5e28 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0055701155,0.0063301595,0.0042756563,0.029668182,0.003665929,0.023308653,0.041472793,0.01839364,0.03860736,0.0060253735,0.0056214742,0.020402266,0.027661264,0.016783247,0.01859294,0.013065602,0.28502858,0.006916266,0.01649864,0.008324767,0.015853647,0.026072072,0.02126262,0.007229569,0.07821424,0.10320244,0.009021552,0.0650905,0.006879877,0.015151082,0.022952529,0.012675275,0.020181743
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
index c0b34494..af6571aa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.005370935,0.0065400284,0.004214182,0.028507866,0.004052519,0.023884999,0.0431287,0.01866619,0.038549002,0.0058642467,0.005612742,0.022007838,0.028794203,0.018062135,0.018920997,0.014034251,0.28207308,0.00724066,0.01554973,0.008246547,0.015637344,0.029362565,0.020395592,0.0072190757,0.08529724,0.08158563,0.008495348,0.07213983,0.0073112454,0.015740085,0.024467451,0.01239826,0.020629534
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
index d1eb6c78..9391aa2a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.003818576,0.0031525614,0.0051886565,0.027662326,0.006760011,0.041391376,0.062559254,0.016534436,0.041413,0.006456904,0.0038409191,0.031356983,0.025992455,0.02640143,0.038692255,0.022995137,0.17438106,0.00587134,0.01310105,0.012426717,0.017008489,0.037873317,0.023261664,0.0044478937,0.07316778,0.08910169,0.00722368,0.09950632,0.0060560848,0.021839175,0.01798219,0.013436582,0.01909868
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
index 57d78954..79957644 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0038112525,0.0032467076,0.005103138,0.028848037,0.006253271,0.040058028,0.06397161,0.015401709,0.039240718,0.007229387,0.0042264964,0.031109253,0.022743622,0.026395943,0.03867059,0.0224054,0.16184604,0.006153928,0.014256049,0.012480883,0.018035192,0.033755783,0.022141878,0.0048214206,0.06857576,0.11654172,0.0072527346,0.091533765,0.0067206784,0.023300141,0.021024033,0.014561617,0.018283268
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
index f09487aa..fc35a03f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0038847511,0.0034304704,0.0054487633,0.030906213,0.0065313987,0.04176309,0.0651858,0.015840625,0.039366536,0.00712093,0.0041720322,0.030338436,0.021227272,0.027650725,0.03322719,0.02358945,0.17575502,0.0062058195,0.015167547,0.013033766,0.018510986,0.03583581,0.021288542,0.0049735317,0.066214904,0.10693726,0.0075933044,0.08376561,0.007044815,0.021833733,0.02074671,0.015919521,0.019489419
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
index 9faa7fb9..dfae95aa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0038600573,0.0032016537,0.0050423536,0.030087678,0.0062640845,0.042983707,0.06558176,0.016441714,0.037939113,0.0068796654,0.0041524945,0.033378396,0.02355689,0.02613971,0.03884025,0.022489678,0.16319337,0.005882346,0.0136809675,0.012107426,0.018530237,0.035350807,0.02193522,0.0048262407,0.069117285,0.106256664,0.007452489,0.08964618,0.0068337126,0.02311564,0.021159057,0.014839026,0.019234167
 0.0038384346,0.0032803454,0.005382596,0.028020917,0.0065062335,0.040472195,0.059231486,0.01510768,0.044317365,0.00677788,0.004063074,0.030269165,0.023738174,0.028254144,0.035682805,0.02283463,0.17437491,0.0062798257,0.014255069,0.013157966,0.018651582,0.03610014,0.021770932,0.0047950475,0.06709478,0.10704565,0.007234076,0.09272612,0.0065334365,0.020849966,0.019035984,0.013944705,0.018372733
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
index 4062c3bf..4800fc39 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.004071277,0.0036563596,0.005473332,0.026431864,0.0070418348,0.040103372,0.058523994,0.014499132,0.048744615,0.0068927244,0.003979709,0.024812536,0.02332055,0.028595718,0.034942232,0.022658326,0.19808845,0.0063588833,0.014898518,0.014481609,0.017180959,0.036216,0.021954877,0.004474714,0.07242044,0.09361984,0.0072108386,0.08443967,0.0062276307,0.019533861,0.01759654,0.013579895,0.01796969
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
index 717cd470..bda5f04e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0042564487,0.0033090666,0.0054760864,0.029484103,0.0063959234,0.032995798,0.058816995,0.013607776,0.048283372,0.0057379017,0.0038245237,0.021225223,0.023702899,0.026598837,0.030306822,0.02262228,0.23025683,0.0063128057,0.014794539,0.013137006,0.019704955,0.031384137,0.019589942,0.004760695,0.06250796,0.106131054,0.006412059,0.074764796,0.005834489,0.019974858,0.01842272,0.012555495,0.016811622
 0.0038192424,0.0033686042,0.0054370016,0.02769945,0.006779793,0.04220823,0.05973473,0.015520086,0.04562838,0.0071383878,0.00422071,0.03299786,0.025394814,0.0281485,0.038514096,0.023156043,0.16337366,0.0063498737,0.014283524,0.012739055,0.01869462,0.036976255,0.023111258,0.004908983,0.07237423,0.09736417,0.0072591524,0.08847638,0.006704482,0.022873966,0.021077238,0.0138563765,0.019810876
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
index 887cda4e..0c133e09 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.0038691815,0.0031717422,0.005064685,0.029039107,0.0061659142,0.037487313,0.06612843,0.01513539,0.03811665,0.0066099614,0.0039828895,0.026459182,0.02207245,0.029161863,0.032705896,0.02051133,0.18691592,0.0063597304,0.0139315445,0.013663895,0.017307293,0.030300051,0.019515662,0.004841311,0.059204914,0.1291775,0.0065799546,0.08632184,0.006249234,0.022190582,0.020693045,0.014265866,0.016799755
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
index 34ba1b0b..1ffc4204 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.004423313,0.004286555,0.00660057,0.023772126,0.008373217,0.040369432,0.054351408,0.01465567,0.05245067,0.008132533,0.0042896164,0.029554067,0.024817849,0.03035214,0.039045297,0.025620734,0.17074145,0.0072791767,0.015768386,0.016019871,0.01698007,0.041483015,0.025339559,0.00460742,0.08207413,0.08653425,0.008505664,0.07673737,0.00643838,0.020246439,0.016483322,0.014442152,0.01922423
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index f2414b5b..35ddcd10 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.029515171,0.029384783,0.02892698,0.031414278,0.030240053,0.03790003,0.031263754,0.030703781,0.02985053,0.03231135,0.030141735,0.028332692,0.033920184,0.027992733,0.031994548,0.032312855,0.032003276,0.029009493,0.025961198,0.028668242,0.031744353,0.028004343,0.0351939,0.028472656,0.033312727,0.029293254,0.025253193,0.03191485,0.031521086,0.030837052,0.027421221,0.028424587,0.026759079
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index ce3129a4..b475e5bf 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.02660504,0.030707078,0.032001417,0.035039,0.029511122,0.034290012,0.03181159,0.0294397,0.0321122,0.03097334,0.0332632,0.027464185,0.030202633,0.028321788,0.025869891,0.03017357,0.031728413,0.033907395,0.02845779,0.031313535,0.028817244,0.02832817,0.033035055,0.026313312,0.030338703,0.028560834,0.03160956,0.031486206,0.030627396,0.029206958,0.030443136,0.027497737,0.030542677
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index db2db85d..ca893131 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.027332451,0.031302933,0.030766888,0.034912553,0.029892711,0.03336584,0.031323344,0.029848998,0.03191401,0.031425472,0.033242542,0.027521798,0.029320594,0.027383283,0.025944008,0.03078899,0.032123405,0.03406133,0.028520554,0.031947363,0.0294212,0.028340716,0.032867085,0.02619513,0.02996483,0.028789517,0.03081592,0.031394005,0.03228332,0.029442273,0.02991017,0.026885256,0.03075153
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index b2366f0b..736eedb1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.03185726,0.028593035,0.029586729,0.02717517,0.029036317,0.03239331,0.030127635,0.029756432,0.028826356,0.03008631,0.027377693,0.03182702,0.03495427,0.032569427,0.03616851,0.031138953,0.02905974,0.026557622,0.028117852,0.02744148,0.032749053,0.030639345,0.031042855,0.03220263,0.033083834,0.030625483,0.027325926,0.030158056,0.028103651,0.03071186,0.029626483,0.033320524,0.027759206
 0.031184139,0.0276112,0.032012727,0.027526684,0.029155925,0.032260306,0.030398509,0.029150296,0.029186625,0.029938763,0.02798222,0.03192229,0.034919117,0.034535985,0.03453919,0.0301626,0.02843686,0.026707487,0.02883699,0.027758451,0.03170132,0.030358283,0.030214984,0.032069325,0.033027902,0.029928787,0.02964113,0.029945217,0.026400918,0.029637858,0.030964117,0.033538274,0.028345458
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index 5c1cac56..136902ac 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.027090317,0.031356793,0.031175932,0.034752592,0.028963773,0.035918295,0.032366708,0.029567668,0.031858996,0.031686503,0.031844135,0.027539104,0.03215706,0.02786867,0.028114032,0.030691825,0.03100698,0.032229245,0.026652839,0.030048568,0.02923739,0.02832744,0.035018757,0.0267673,0.031383105,0.028060874,0.028454317,0.03174636,0.030990787,0.030642902,0.030124784,0.02802195,0.028333979
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index 3ebe548e..7b1ad181 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.031245146,0.026523279,0.03429213,0.028590597,0.030950105,0.030549498,0.030602774,0.029434195,0.02976708,0.030371303,0.02923278,0.03193279,0.032057486,0.036185876,0.031233333,0.028832216,0.028311577,0.027465537,0.031915396,0.028459525,0.029657632,0.029486306,0.02885231,0.031842597,0.032040454,0.029984664,0.032361843,0.029286683,0.025559865,0.028563855,0.03226416,0.032237228,0.029909858
 0.031189607,0.027768781,0.03171535,0.0286404,0.02989582,0.03686473,0.030307468,0.029793784,0.029219873,0.03136719,0.029169736,0.030565001,0.035063583,0.033113644,0.033308994,0.03176959,0.030947847,0.027378758,0.026388302,0.028127484,0.03166254,0.02842469,0.033268318,0.029070573,0.033865992,0.028430037,0.027634714,0.031596266,0.02863526,0.028753258,0.02869571,0.030206194,0.027160522
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index ef108ae3..5bd5387d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.03022107,0.026185002,0.035542563,0.030422375,0.030111654,0.032666244,0.03029986,0.028772637,0.029862503,0.031565245,0.030188533,0.031473093,0.03380565,0.034408633,0.031200787,0.029270852,0.029954033,0.026347158,0.029255822,0.028777694,0.031132765,0.028987736,0.03171347,0.029104978,0.032444056,0.027695423,0.03135624,0.030321337,0.025290275,0.027976021,0.03161053,0.03318183,0.028853968
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index a93e39ea..211a7d66 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
 0.030450879,0.031356543,0.026286202,0.030208621,0.029305283,0.033798173,0.029997792,0.031071521,0.029736698,0.031199083,0.029278152,0.029431423,0.03262524,0.026415547,0.03385641,0.032870974,0.031922586,0.029316304,0.026414368,0.029405521,0.03292572,0.03013779,0.033768725,0.029333936,0.031706,0.030111287,0.025230156,0.03123331,0.033786975,0.032116424,0.027622495,0.029445931,0.027633844
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index 7b307493..488a103e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.069280244,0.08278568,0.072002776,0.068465054,0.073495515,0.069683656,0.08337737,0.08428103,0.079714045,0.08516558,0.07238791,0.07722001,0.08214111
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index f5c9cd40..287c3c11 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.08120824,0.0770729,0.080374785,0.07006595,0.08169089,0.06985054,0.077308096,0.06746512,0.08158482,0.078970395,0.07628951,0.077936,0.08018279
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index f76ecc72..6cd732d3 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.081267715,0.07711225,0.080673225,0.07033195,0.08024122,0.06911172,0.07555922,0.06642563,0.0804719,0.08202005,0.07461091,0.079976365,0.08219792
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index a99204aa..0caabb64 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.06782719,0.078334294,0.07045661,0.0793704,0.073115356,0.08177368,0.08440978,0.09057371,0.076156095,0.07454853,0.078214675,0.071796,0.07342377
 0.07023328,0.0774746,0.07357778,0.07948175,0.07633355,0.08361607,0.08418787,0.08768755,0.07620843,0.06998042,0.0815577,0.06901032,0.0706507
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index e0a22959..75244a2c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.075564295,0.077890314,0.07621962,0.06734453,0.07971224,0.06832646,0.07854212,0.07214813,0.08339084,0.082671195,0.07736374,0.0789834,0.08184321
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index d2469e90..e8c6c744 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.07543901,0.07629552,0.079985805,0.07938394,0.08105448,0.08392259,0.08113814,0.08250335,0.07510063,0.06616155,0.08535827,0.06484661,0.068810105
 0.0684685,0.08285618,0.07265628,0.07221835,0.074672826,0.07720729,0.087464236,0.08804019,0.07745041,0.072739474,0.07616589,0.07260381,0.07745654
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index 11d4ea08..0a26093e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.07194322,0.077352665,0.08031595,0.07728969,0.08099176,0.080330044,0.08255197,0.079121925,0.07681379,0.069612056,0.08324984,0.06879602,0.07163119
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index 072240da..fdcbac81 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
-unknown,other,mask,0,1,2,3,4,5,6,7,8,9
+[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
 0.068312116,0.08032282,0.06863922,0.07357371,0.06984851,0.07086958,0.0803648,0.0826013,0.079164565,0.08935935,0.06972866,0.0837885,0.083426945
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 46ecc58f..ecb3628b 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -163,6 +163,7 @@ def ar_config():
         metadata_config_path="dummy.json",
         model_path="dummy.onnx",
         model_type="generative",
+        training_objective="causal",
         data_path="tests/unit/data/dummy.parquet",
         input_columns=["target_col"],
         categorical_columns=[],
@@ -172,6 +173,7 @@ def ar_config():
         target_column_types={"target_col": "real"},
         seed=42,
         device="cpu",
+        prediction_length=1,
         seq_length=3,
         inference_batch_size=2,
         output_probabilities=False,

From 857b86e0847aab5772cce9572e7d9a9ff508658f Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 4 Jun 2026 14:13:12 +0200
Subject: [PATCH 06/81] Fix index alignment

---
 src/sequifier/infer.py | 90 +++++++++++++++++++++++++++---------------
 src/sequifier/train.py |  4 +-
 2 files changed, 61 insertions(+), 33 deletions(-)

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index d87b77d4..fd2d97a2 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -259,6 +259,41 @@ def infer_worker(
     logger.info("--- Inference Complete ---")
 
 
+def calculate_item_positions(
+    start_positions: np.ndarray,
+    seq_length: int,
+    prediction_length: int,
+    training_objective: str,
+) -> np.ndarray:
+    """
+    Calculates absolute item positions for inference outputs based on the training objective.
+
+    Args:
+        start_positions: 1D array of base start positions for each sequence in the batch.
+        seq_length: The length of the input sequence window.
+        prediction_length: The total number of predicted tokens per sequence.
+        training_objective: Either "causal" or "bert".
+
+    Returns:
+        A 1D array of absolute item positions mapped to every flattened prediction row.
+    """
+    if training_objective == "bert":
+        # Anchor positions to the start of the input sequence and tile forwards
+        base_positions = start_positions
+        position_offsets = np.arange(0, prediction_length)
+    else:
+        # Anchor positions to the future token step and tile backwards
+        base_positions = start_positions + seq_length
+        position_offsets = np.arange(-prediction_length + 1, 1)
+
+    # Repeat base anchors to match the number of predictions per sequence window
+    repeated_bases = np.repeat(base_positions, prediction_length)
+    # Tile the relative step offsets across all sequences in the batch chunk
+    tiled_offsets = np.tile(position_offsets, len(start_positions))
+
+    return repeated_bases + tiled_offsets
+
+
 @beartype
 def infer_embedding(
     config: "InfererModel",
@@ -468,26 +503,23 @@ def infer_generative(
 
                 # Get base IDs and positions (shape: batch_size)
                 sequence_ids_for_preds_base = data.get_column("sequenceId").filter(mask)
-                item_positions_for_preds_base = (
+                item_positions_base_raw = (
                     data.get_column("startItemPosition").filter(mask).to_numpy()
-                    + config.seq_length
                 )
-
                 prediction_length = inferer.prediction_length
 
-                # Expand IDs and positions to match model output shape (batch_size * prediction_length)
+                # Expand IDs to match model output shape
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
                 )
 
-                item_positions_repeated = np.repeat(
-                    item_positions_for_preds_base, prediction_length
+                # Invoke the unified positioning engine
+                item_positions_for_preds = calculate_item_positions(
+                    item_positions_base_raw,
+                    config.seq_length,
+                    prediction_length,
+                    config.training_objective,
                 )
-                position_offsets = np.tile(
-                    np.arange(-prediction_length + 1, 1),
-                    len(item_positions_for_preds_base),
-                )
-                item_positions_for_preds = item_positions_repeated + position_offsets
 
             else:
                 if inferer.prediction_length != 1:
@@ -520,23 +552,22 @@ def infer_generative(
                 sequence_ids_for_preds_base = (
                     data.get_column("sequenceId").filter(mask).to_numpy()
                 )
-                item_positions_for_preds_base = (
+                item_positions_base_raw = (
                     data.get_column("startItemPosition").filter(mask).to_numpy()
-                    + config.seq_length
                 )
-
                 prediction_length = inferer.prediction_length
+
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
                 )
-                item_positions_repeated = np.repeat(
-                    item_positions_for_preds_base, prediction_length
-                )
-                position_offsets = np.tile(
-                    np.arange(-prediction_length + 1, 1),
-                    len(item_positions_for_preds_base),
+
+                # Invoke the unified positioning engine
+                item_positions_for_preds = calculate_item_positions(
+                    item_positions_base_raw,
+                    config.seq_length,
+                    prediction_length,
+                    config.training_objective,
                 )
-                item_positions_for_preds = item_positions_repeated + position_offsets
             else:
                 if inferer.prediction_length != 1:
                     raise ValueError(
@@ -571,22 +602,19 @@ def infer_generative(
             if total_steps == 1:
                 # Non-autoregressive path: Apply prediction_length logic
                 sequence_ids_for_preds_base = sequence_ids_tensor.numpy()
-                item_positions_for_preds_base = (
-                    start_positions_tensor.numpy() + config.seq_length
-                )
+                item_positions_base_raw = start_positions_tensor.numpy()
 
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
                 )
 
-                item_positions_repeated = np.repeat(
-                    item_positions_for_preds_base, prediction_length
-                )
-                position_offsets = np.tile(
-                    np.arange(-prediction_length + 1, 1),
-                    len(item_positions_for_preds_base),
+                # Invoke the unified positioning engine
+                item_positions_for_preds = calculate_item_positions(
+                    item_positions_base_raw,
+                    config.seq_length,
+                    prediction_length,
+                    config.training_objective,
                 )
-                item_positions_for_preds = item_positions_repeated + position_offsets
 
             else:
                 sequence_ids_for_preds = np.repeat(
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 5e148c51..50cdbd2a 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1443,7 +1443,7 @@ def _train_epoch(
                     if k in self.target_column_types
                 }
                 if self.hparams.training_spec.training_objective == "bert":
-                    data, targets = apply_bert_masking(data, targets, self.hparams)
+                    data, targets = apply_bert_masking(data, self.hparams)
 
                 # Only use standard torch.autocast if FSDP MixedPrecision is NOT handling it natively
                 if (
@@ -1771,7 +1771,7 @@ def _evaluate(
                     if k in self.target_column_types
                 }
                 if self.hparams.training_spec.training_objective == "bert":
-                    data, targets = apply_bert_masking(data, targets, self.hparams)
+                    data, targets = apply_bert_masking(data, self.hparams)
 
                 if (
                     self.hparams.training_spec.layer_autocast

From 82cc41f11fc0311687d1f265bd698e8c3c137491 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 4 Jun 2026 18:24:02 +0200
Subject: [PATCH 07/81] Fix tests

---
 src/sequifier/config/infer_config.py          |  8 +++
 src/sequifier/config/train_config.py          | 41 ++++++++-----
 src/sequifier/helpers.py                      | 24 ++++++--
 src/sequifier/infer.py                        | 13 ++++-
 .../io/sequifier_dataset_from_file.py         |  4 ++
 .../sequifier_dataset_from_folder_parquet.py  | 19 ++++++-
 ...uifier_dataset_from_folder_parquet_lazy.py | 57 +++++++++++--------
 .../io/sequifier_dataset_from_folder_pt.py    | 16 +++++-
 .../sequifier_dataset_from_folder_pt_lazy.py  | 16 +++++-
 src/sequifier/train.py                        |  4 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  5 ++
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  4 +-
 tests/unit/test_helpers.py                    | 18 +++++-
 13 files changed, 168 insertions(+), 61 deletions(-)

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index be60da2c..d67b04c6 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -221,6 +221,14 @@ def validate_autoregression(cls, v: bool, info: ValidationInfo):
                 "Autoregressive inference with non-identical 'input_columns' and 'target_columns' is possible but should not be performed"
             )
 
+        if (
+            info.data.get("training_objective") is not None
+            and info.data.get("training_objective") == "bert"
+        ):
+            raise ValueError(
+                "Autoregressive inference is not possible with BERT-style models."
+            )
+
         return v
 
     @field_validator("data_path")
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 4ddb8067..92117551 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -11,7 +11,14 @@
 import yaml
 from beartype import beartype
 from loguru import logger
-from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validator
+from pydantic import (
+    BaseModel,
+    ConfigDict,
+    Field,
+    computed_field,
+    field_validator,
+    model_validator,
+)
 
 import sequifier
 from sequifier.config.probabilities import ProbabilityDistribution
@@ -226,6 +233,16 @@ def __init__(self, **kwargs):
         self.validate_scheduler_config(kwargs["scheduler"], kwargs)
         self.scheduler = DotDict(kwargs["scheduler"])
 
+    @computed_field
+    @property
+    def data_offset(self) -> int:
+        return 1
+
+    @computed_field
+    @property
+    def target_offset(self) -> int:
+        return 0 if self.training_objective == "causal" else 1
+
     @field_validator("layer_type_dtypes")
     @classmethod
     def validate_layer_type_dtypes(cls, v):
@@ -663,20 +680,14 @@ def validate_training_spec(cls, v, info):
 
         export_generative_model = info.data.get("export_generative_model")
         export_embedding_model = info.data.get("export_embedding_model")
-        if v.training_objective == "bert":
-            if export_generative_model:
-                raise ValueError(
-                    "'training_objective: bert' is incompatible with generative model export"
-                )
-            if not export_embedding_model:
-                raise ValueError(
-                    "'training_objective: bert' requires embedding model export"
-                )
-        if v.training_objective == "causal":
-            if not export_generative_model and not export_embedding_model:
-                warnings.warn(
-                    "At least one of 'export_generative_model' and 'export_embedding_model' should be true"
-                )
+        if (
+            not export_generative_model
+            and not export_embedding_model
+            and os.getenv("SEQUIFIER_PREVENT_EXPORT") is None
+        ):
+            raise ValueError(
+                "At least one of 'export_generative_model' and 'export_embedding_model' must be true. If you want to override this, set the env variable 'SEQUIFIER_PREVENT_EXPORT' to any value"
+            )
 
         return v
 
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index d3e57f9e..d09f75f8 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -248,6 +248,8 @@ def numpy_to_pytorch(
     column_types: dict[str, torch.dtype],
     all_columns: list[str],
     seq_length: int,
+    data_offset: int,
+    target_offset: int,
 ) -> dict[str, Tensor]:
     """Converts a long-format Polars DataFrame to a dict of sequence tensors.
 
@@ -281,8 +283,12 @@ def numpy_to_pytorch(
         tensors. Target tensors are stored with a `_target` suffix
         (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
     """
-    input_seq_cols = [str(c) for c in range(seq_length, 0, -1)]
-    target_seq_cols = [str(c) for c in range(seq_length - 1, -1, -1)]
+    input_seq_cols = [
+        str(c) for c in range(seq_length - 1 + data_offset, (-1 + data_offset), -1)
+    ]
+    target_seq_cols = [
+        str(c) for c in range(seq_length - 1 + target_offset, (-1 + target_offset), -1)
+    ]
 
     # We will create a unified dictionary
     unified_tensors = {}
@@ -515,13 +521,14 @@ def get_last_training_batch_timedelta(
 
 def apply_bert_masking(
     data_batch: Dict[str, torch.Tensor],
+    targets_batch: Dict[str, torch.Tensor],
     config: Any,  # TrainConfig
 ) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
     """
     Applies BERT-style span corruption to the input data using custom distributions.
     Explicitly passes the boolean prediction mask via the targets dictionary.
     """
-    targets_batch = {k: tensor.clone().detach() for k, tensor in data_batch.items()}
+    targets_batch = {k: tensor.clone().detach() for k, tensor in targets_batch.items()}
     batch_size, seq_len = config.training_spec.batch_size, config.seq_length
 
     # 1. Identify valid tokens (Renamed from padding_mask to valid_mask for clarity)
@@ -547,8 +554,8 @@ def apply_bert_masking(
     sampled_lengths = config.training_spec.bert_spec.span_masking.sample(
         (batch_size, max_spans), device=device
     )
-    sampled_starts_pct = torch.rand((batch_size, max_spans), device=device)
 
+    sampled_starts_pct = torch.rand((batch_size, max_spans), device=device)
     # 3. Span Masking Loop
     for i in range(batch_size):
         budget = budgets[i].item()
@@ -557,13 +564,18 @@ def apply_bert_masking(
         if budget < 1 or valid_len < 1:
             continue
 
+        sampled_starts = (sampled_starts_pct[i] * valid_len).long().tolist()
+
+        sampled_lengths_list = sampled_lengths[i].tolist()
+
         current_masked = 0
         span_idx = 0
 
         # Apply spans until budget is hit
         while current_masked < budget and span_idx < max_spans:
-            span_len = sampled_lengths[i, span_idx].item()
-            start_idx = int(sampled_starts_pct[i, span_idx].item() * valid_len)
+            span_len = sampled_lengths_list[span_idx]
+            start_idx = sampled_starts[span_idx]
+
             end_idx = min(start_idx + span_len, valid_len)
 
             span_view = bert_mask[i, start_idx:end_idx]
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index fd2d97a2..ba94086a 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -879,7 +879,10 @@ def get_embeddings(
         A NumPy array containing the computed embeddings for the batch.
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
-    X = numpy_to_pytorch(data, column_types, all_columns, config.seq_length)
+    target_offset = 0 if config.training_objective == "causal" else 1
+    X = numpy_to_pytorch(
+        data, column_types, all_columns, config.seq_length, 1, target_offset
+    )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     del data
 
@@ -919,7 +922,11 @@ def get_probs_preds_from_df(
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
 
-    X = numpy_to_pytorch(data, column_types, all_columns, config.seq_length)
+    target_offset = 0 if config.training_objective == "causal" else 1
+
+    X = numpy_to_pytorch(
+        data, column_types, all_columns, config.seq_length, 1, target_offset
+    )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     del data
 
@@ -1041,6 +1048,8 @@ def get_probs_preds_autoregression(
         column_types,
         config.input_columns,
         seq_length,
+        data_offset=1,
+        target_offset=0,
     )
 
     # Run the autoregressive PyTorch inference
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index ec590953..c2f0916c 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -40,12 +40,16 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in config.column_types
         }
 
+        target_offset = 0 if config.training_spec.training_objective == "causal" else 1
+
         # self.all_tensors now holds both inputs and targets
         all_tensors = numpy_to_pytorch(
             data=data_df,
             column_types=column_types,
             all_columns=all_columns,
             seq_length=config.seq_length,
+            data_offset=1,
+            target_offset=target_offset,
         )
         self.n_samples = all_tensors[all_columns[0]].shape[0]
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 5beb3347..d37e11ab 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -54,9 +54,22 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
 
         # Sequence formatting structures matching long-format schema boundaries
         train_seq_len = self.config.seq_length
-        input_seq_cols = [str(c) for c in range(train_seq_len, 0, -1)]
-        target_seq_cols = [str(c) for c in range(train_seq_len - 1, -1, -1)]
-
+        input_seq_cols = [
+            str(c)
+            for c in range(
+                train_seq_len - 1 + self.config.training_spec.data_offset,
+                (-1 + self.config.training_spec.data_offset),
+                -1,
+            )
+        ]
+        target_seq_cols = [
+            str(c)
+            for c in range(
+                train_seq_len - 1 + self.config.training_spec.target_offset,
+                (-1 + self.config.training_spec.target_offset),
+                -1,
+            )
+        ]
         all_sequences: Dict[str, list[torch.Tensor]] = {
             col: [] for col in config.input_columns
         }
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index cc000e1c..adda1a6e 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -209,9 +209,22 @@ def __iter__(
         train_seq_len = self.config.seq_length
         global_file_start_sample = 0
 
-        # Sequence formatting configurations
-        input_seq_cols = [str(c) for c in range(train_seq_len, 0, -1)]
-        target_seq_cols = [str(c) for c in range(train_seq_len - 1, -1, -1)]
+        input_seq_cols = [
+            str(c)
+            for c in range(
+                train_seq_len - 1 + self.config.training_spec.data_offset,
+                (-1 + self.config.training_spec.data_offset),
+                -1,
+            )
+        ]
+        target_seq_cols = [
+            str(c)
+            for c in range(
+                train_seq_len - 1 + self.config.training_spec.target_offset,
+                (-1 + self.config.training_spec.target_offset),
+                -1,
+            )
+        ]
 
         # Initialize cross-file buffers
         seq_buffer: Dict[str, torch.Tensor] = {}
@@ -234,7 +247,6 @@ def __iter__(
             # This file overlaps with our worker's assigned boundary. Load it.
             file_path = os.path.join(self.data_dir, self.batch_files_info[f_id]["path"])
             df = pl.read_parquet(file_path)
-            feature_names = df["inputCol"].unique().to_list()
 
             # Generate indices for the whole file using torch (matching pt_lazy)
             indices = torch.arange(file_samples)
@@ -264,44 +276,39 @@ def __iter__(
             }
 
             # Dynamic feature names fallback to gracefully handle MagicMock objects in unit tests
-            feature_names = list(feature_partitions.keys())
-            cols_to_process = (
-                self.config.input_columns
-                if isinstance(getattr(self.config, "input_columns", None), list)
-                else feature_names
-            )
 
             # Process Long format data structures into PyTorch Tensors
             new_seq, new_tgt = {}, {}
             expected_samples = len(worker_indices_np)
 
             # 2. Iterate over the expected config columns, not the dynamically found ones
-            for col_name in cols_to_process:
-                # Gracefully handle MagicMock column_torch_types dictionaries
-                torch_type = self.column_torch_types[col_name]
-
+            # Process inputs
+            for col_name in self.config.input_columns:
                 if col_name in feature_partitions:
-                    # Positional advanced selection using the coordinated indices
                     feature_chunk = feature_partitions[col_name][worker_indices_np]
-
                     new_seq[col_name] = torch.tensor(
                         feature_chunk.select(input_seq_cols).to_numpy(),
-                        dtype=torch_type,
+                        dtype=self.column_torch_types[col_name],
                     )
+                else:
+                    new_seq[col_name] = torch.zeros(
+                        (expected_samples, train_seq_len),
+                        dtype=self.column_torch_types[col_name],
+                    )
+
+            # Process targets
+            for col_name in self.config.target_columns:
+                if col_name in feature_partitions:
+                    feature_chunk = feature_partitions[col_name][worker_indices_np]
                     new_tgt[col_name] = torch.tensor(
                         feature_chunk.select(target_seq_cols).to_numpy(),
-                        dtype=torch_type,
+                        dtype=self.column_torch_types[col_name],
                     )
                 else:
-                    # 3. Graceful fallback: Pad with zeros if a chunk is mysteriously missing a feature
-                    new_seq[col_name] = torch.zeros(
-                        (expected_samples, train_seq_len), dtype=torch_type
-                    )
-                    new_tgt[col_name] = torch.zeros(
-                        (expected_samples, train_seq_len), dtype=torch_type
+                    raise RuntimeError(
+                        f"Missing required column {col_name} in Parquet partition"
                     )
 
-            # Free the DataFrame immediately to keep RAM down
             del df
 
             # Append the new slice to the cross-file buffer
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 182ad841..9c29d8b5 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -149,13 +149,25 @@ def __iter__(
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
+            data_offset = self.config.training_spec.data_offset
+            target_offset = self.config.training_spec.target_offset
             data_batch = {
-                key: tensor[batch_indices, -(train_seq_len + 1) : -1]
+                key: tensor[
+                    batch_indices,
+                    -(train_seq_len + data_offset) : (
+                        -data_offset if data_offset > 0 else None
+                    ),
+                ]
                 for key, tensor in self.sequences.items()
                 if key in self.config.input_columns
             }
             targets_batch = {
-                key: tensor[batch_indices, -train_seq_len:]
+                key: tensor[
+                    batch_indices,
+                    -(train_seq_len + target_offset) : (
+                        -target_offset if target_offset > 0 else None
+                    ),
+                ]
                 for key, tensor in self.sequences.items()
                 if key in self.config.target_columns
             }
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index c6d1e238..e07924e0 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -246,13 +246,25 @@ def __iter__(
                 continue
 
             # Extract the data subset for this worker (Advanced indexing copies the data)
+            data_offset = self.config.training_spec.data_offset
+            target_offset = self.config.training_spec.target_offset
             new_seq = {
-                k: v[worker_indices, -(train_seq_len + 1) : -1]
+                k: v[
+                    worker_indices,
+                    -(train_seq_len + data_offset) : (
+                        -data_offset if data_offset > 0 else None
+                    ),
+                ]
                 for k, v in sequences_batch.items()
                 if k in self.config.input_columns
             }
             new_tgt = {
-                k: v[worker_indices, -train_seq_len:]
+                k: v[
+                    worker_indices,
+                    -(train_seq_len + target_offset) : (
+                        -target_offset if target_offset > 0 else None
+                    ),
+                ]
                 for k, v in sequences_batch.items()
                 if k in self.config.target_columns
             }
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 50cdbd2a..5e148c51 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1443,7 +1443,7 @@ def _train_epoch(
                     if k in self.target_column_types
                 }
                 if self.hparams.training_spec.training_objective == "bert":
-                    data, targets = apply_bert_masking(data, self.hparams)
+                    data, targets = apply_bert_masking(data, targets, self.hparams)
 
                 # Only use standard torch.autocast if FSDP MixedPrecision is NOT handling it natively
                 if (
@@ -1771,7 +1771,7 @@ def _evaluate(
                     if k in self.target_column_types
                 }
                 if self.hparams.training_spec.training_objective == "bert":
-                    data, targets = apply_bert_masking(data, self.hparams)
+                    data, targets = apply_bert_masking(data, targets, self.hparams)
 
                 if (
                     self.hparams.training_spec.layer_autocast
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 1d565fca..a47bbfd4 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -20,6 +20,11 @@ def mock_config():
     config.training_spec.batch_size = 5
     config.training_spec.num_workers = 0
     config.training_spec.sampling_strategy = "exact"
+    config.training_spec.data_offset = 1
+    config.training_spec.target_offset = 0
+    config.input_columns = ["item"]
+    config.target_columns = ["item"]
+
     return config
 
 
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 8d7abb2f..29cd1af6 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -20,8 +20,10 @@ def mock_config(tmp_path):
     config.training_spec.num_workers = 0
     config.seed = 42
     config.seq_length = 5
-    config.input_columns = ["col1"]  # <-- ADD THIS LINE
+    config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
+    config.training_spec.data_offset = 1
+    config.training_spec.target_offset = 0
     return config
 
 
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index 6ad3b56c..cae5c87f 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -80,7 +80,9 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     all_columns = ["A"]
     seq_length = 3
 
-    tensors = numpy_to_pytorch(data, column_types, all_columns, seq_length)
+    tensors = numpy_to_pytorch(
+        data, column_types, all_columns, seq_length, data_offset=1, target_offset=0
+    )
 
     # 1. Check Keys
     assert "A" in tensors
@@ -110,13 +112,23 @@ def test_numpy_to_pytorch_dtypes():
     # Case 1: Integer
     data_int = pl.DataFrame({"inputCol": ["int_col"], "1": [10], "0": [20]})
     tensors_int = numpy_to_pytorch(
-        data_int, {"int_col": torch.int64}, ["int_col"], seq_length=1
+        data_int,
+        {"int_col": torch.int64},
+        ["int_col"],
+        seq_length=1,
+        data_offset=1,
+        target_offset=0,
     )
     assert tensors_int["int_col"].dtype == torch.int64
 
     # Case 2: Float
     data_float = pl.DataFrame({"inputCol": ["float_col"], "1": [10.5], "0": [20.5]})
     tensors_float = numpy_to_pytorch(
-        data_float, {"float_col": torch.float32}, ["float_col"], seq_length=1
+        data_float,
+        {"float_col": torch.float32},
+        ["float_col"],
+        seq_length=1,
+        data_offset=1,
+        target_offset=0,
     )
     assert tensors_float["float_col"].dtype == torch.float32

From 6df557a3c522fe7f0518081c77a980411663a48e Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 5 Jun 2026 13:01:09 +0200
Subject: [PATCH 08/81] Add padding mask

---
 src/sequifier/helpers.py                      |   3 +
 src/sequifier/preprocess.py                   |   3 -
 src/sequifier/train.py                        |  72 +++-
 ...orical-1-best-embedding-3-0-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-1-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-2-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-3-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-4-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-5-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-6-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-7-embeddings.csv |   2 +-
 ...inf-size-best-embedding-3-0-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-1-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-2-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-3-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-4-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-5-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-6-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-7-embeddings.csv |   6 +-
 ...-1-best-3-autoregression-1-predictions.csv |  38 +-
 ...-1-best-3-autoregression-2-predictions.csv |  38 +-
 ...-1-best-3-autoregression-3-predictions.csv |  42 +-
 ...-1-best-3-autoregression-4-predictions.csv |  40 +-
 ...-1-best-3-autoregression-5-predictions.csv |  78 ++--
 ...-1-best-3-autoregression-6-predictions.csv |  22 +-
 ...-1-best-3-autoregression-7-predictions.csv |   4 +-
 ...del-categorical-1-best-3-1-predictions.csv |   2 +-
 ...del-categorical-1-best-3-3-predictions.csv |   4 +-
 ...del-categorical-1-best-3-4-predictions.csv |   2 +-
 ...del-categorical-1-best-3-5-predictions.csv |   4 +-
 ...del-categorical-1-best-3-6-predictions.csv |   2 +-
 ...del-categorical-1-best-3-7-predictions.csv |   2 +-
 ...orical-1-inf-size-best-3-1-predictions.csv |   4 +-
 ...orical-1-inf-size-best-3-2-predictions.csv |   2 +-
 ...orical-1-inf-size-best-3-3-predictions.csv |   4 +-
 ...orical-1-inf-size-best-3-6-predictions.csv |   2 +-
 ...orical-1-inf-size-best-3-7-predictions.csv |   2 +-
 ...del-categorical-3-best-3-1-predictions.csv |   2 +-
 ...del-categorical-3-best-3-3-predictions.csv |   4 +-
 ...del-categorical-3-best-3-4-predictions.csv |   2 +-
 ...del-categorical-3-best-3-5-predictions.csv |   4 +-
 ...del-categorical-3-best-3-6-predictions.csv |   2 +-
 ...del-categorical-3-best-3-7-predictions.csv |   2 +-
 ...orical-3-inf-size-best-3-1-predictions.csv |   2 +-
 ...orical-3-inf-size-best-3-2-predictions.csv |   4 +-
 ...orical-3-inf-size-best-3-3-predictions.csv |   8 +-
 ...orical-3-inf-size-best-3-4-predictions.csv |   4 +-
 ...orical-3-inf-size-best-3-5-predictions.csv |  10 +-
 ...orical-3-inf-size-best-3-6-predictions.csv |   4 +-
 ...orical-3-inf-size-best-3-7-predictions.csv |   4 +-
 ...del-categorical-5-best-3-2-predictions.csv |   2 +-
 ...del-categorical-5-best-3-3-predictions.csv |   2 +-
 ...del-categorical-5-best-3-5-predictions.csv |   2 +-
 ...el-categorical-50-best-3-0-predictions.csv |   2 +-
 ...el-categorical-50-best-3-1-predictions.csv |   2 +-
 ...el-categorical-50-best-3-2-predictions.csv |   2 +-
 ...el-categorical-50-best-3-3-predictions.csv |   4 +-
 ...el-categorical-50-best-3-4-predictions.csv |   2 +-
 ...el-categorical-50-best-3-5-predictions.csv |   2 +-
 ...el-categorical-50-best-3-7-predictions.csv |   2 +-
 ...rical-distributed-best-3-0-predictions.csv |   2 +-
 ...rical-distributed-best-3-2-predictions.csv |   2 +-
 ...rical-distributed-best-3-3-predictions.csv |   2 +-
 ...rical-distributed-best-3-4-predictions.csv |   2 +-
 ...rical-distributed-best-3-5-predictions.csv |   2 +-
 ...rical-distributed-best-3-6-predictions.csv |   2 +-
 ...rical-distributed-best-3-7-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-0-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-2-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-3-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-4-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-5-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-6-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-7-predictions.csv |   2 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...uifier-model-real-1-best-3-predictions.csv |  48 +--
 ...uifier-model-real-3-best-3-predictions.csv |  48 +--
 ...uifier-model-real-5-best-3-predictions.csv |  48 +--
 ...ifier-model-real-50-best-3-predictions.csv |  48 +--
 ...custom-eval-run-0-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |  12 +-
 ...custom-eval-run-2-best-1-1-predictions.csv |  12 +-
 ...custom-eval-run-2-best-1-2-predictions.csv |  20 +-
 ...custom-eval-run-2-best-1-3-predictions.csv |  24 +-
 ...custom-eval-run-2-best-1-4-predictions.csv |   8 +-
 ...custom-eval-run-2-best-1-5-predictions.csv |  26 +-
 ...custom-eval-run-2-best-1-6-predictions.csv |  16 +-
 ...custom-eval-run-2-best-1-7-predictions.csv |  20 +-
 ...custom-eval-run-3-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-7-predictions.csv |  60 +--
 ...l-categorical-1-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-3-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-5-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-7-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-3-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-5-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-7-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...l-categorical-5-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-3-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-5-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-7-probabilities.csv |   2 +-
 ...-categorical-50-best-3-0-probabilities.csv |   2 +-
 ...-categorical-50-best-3-1-probabilities.csv |   2 +-
 ...-categorical-50-best-3-2-probabilities.csv |   2 +-
 ...-categorical-50-best-3-3-probabilities.csv |   4 +-
 ...-categorical-50-best-3-4-probabilities.csv |   2 +-
 ...-categorical-50-best-3-5-probabilities.csv |   4 +-
 ...-categorical-50-best-3-6-probabilities.csv |   2 +-
 ...-categorical-50-best-3-7-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   2 +-
 tests/unit/test_train.py                      | 108 +++++
 200 files changed, 1854 insertions(+), 1676 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index d09f75f8..fb6c2541 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -528,6 +528,7 @@ def apply_bert_masking(
     Applies BERT-style span corruption to the input data using custom distributions.
     Explicitly passes the boolean prediction mask via the targets dictionary.
     """
+    data_batch = {k: tensor.clone() for k, tensor in data_batch.items()}
     targets_batch = {k: tensor.clone().detach() for k, tensor in targets_batch.items()}
     batch_size, seq_len = config.training_spec.batch_size, config.seq_length
 
@@ -641,4 +642,6 @@ def apply_bert_masking(
     # 6. Append the explicit prediction mask
     targets_batch["_bert_mask"] = bert_mask
 
+    data_batch["_attention_valid_mask"] = valid_mask.detach()
+
     return data_batch, targets_batch
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index dacb92f6..54ca8f14 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -1267,9 +1267,6 @@ def _process_batches_multiple_files_inner(
                         n_rows_running_count += data.shape[0]
                     continue
 
-            data = _load_and_preprocess_data(
-                path, read_format, selected_columns, max_rows_inner
-            )
             data = _load_and_preprocess_data(
                 path, read_format, selected_columns, max_rows_inner
             )
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 5e148c51..bd1086cf 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -973,6 +973,70 @@ def _recursive_concat(self, srcs: list[Tensor]):
                 srcs_inner.append(src)
             return self._recursive_concat(srcs_inner)
 
+    @conditional_beartype
+    def _infer_attention_valid_mask(self, src: dict[str, Tensor]) -> Tensor:
+        """
+        Returns a boolean mask of shape (batch_size, seq_length).
+
+        True means valid/non-padding token.
+        False means padding token.
+        """
+        if "_attention_valid_mask" in src:
+            return src["_attention_valid_mask"].bool()
+
+        # Prefer categorical columns because padding is unambiguously encoded as 0.
+        for col in self.categorical_columns:
+            if col in src:
+                return src[col] != 0
+
+        # Fallback for real-only data.
+        # This matches the existing heuristic used in BERT masking.
+        ref_col = next(col for col in self.input_columns if col in src)
+        return (src[ref_col] != 0.0).long().cumsum(dim=1) > 0
+
+    @conditional_beartype
+    def _build_attention_mask(self, valid_mask: Tensor, dtype: torch.dtype) -> Tensor:
+        batch_size, seq_length = valid_mask.shape
+        device = valid_mask.device
+
+        expected_seq_length = self.src_mask.shape[-1]
+        if seq_length != expected_seq_length:
+            raise ValueError(
+                f"valid_mask sequence length ({seq_length}) must match "
+                f"model sequence length ({expected_seq_length})."
+            )
+
+        base_mask = self.src_mask.to(device=device, dtype=dtype)
+        base_mask = base_mask.view(1, 1, seq_length, seq_length)
+
+        invalid_keys = ~valid_mask.bool()
+
+        padding_mask = torch.zeros(
+            batch_size,
+            1,
+            1,
+            seq_length,
+            device=device,
+            dtype=dtype,
+        )
+
+        padding_mask = padding_mask.masked_fill(
+            invalid_keys[:, None, None, :],
+            torch.finfo(dtype).min,
+        )
+
+        return base_mask + padding_mask
+
+    @conditional_beartype
+    def _zero_padding_positions(self, x: Tensor, valid_mask: Tensor) -> Tensor:
+        """
+        x shape:
+            (batch_size, seq_length, dim_model)
+
+        Zeroes padded query positions so padded rows do not keep evolving.
+        """
+        return x * valid_mask[:, :, None].to(dtype=x.dtype)
+
     @conditional_beartype
     def forward_inner(self, src: dict[str, Tensor]) -> Tensor:
         """The inner forward pass of the model.
@@ -1043,11 +1107,17 @@ def forward_inner(self, src: dict[str, Tensor]) -> Tensor:
         if self.joint_embedding_layer is not None:
             src2 = self.joint_embedding_layer(src2)
 
-        mask = self.src_mask.to(dtype=src2.dtype)
+        valid_mask = self._infer_attention_valid_mask(src).to(device=src2.device)
+        src2 = self._zero_padding_positions(src2, valid_mask)
+
+        mask = self._build_attention_mask(valid_mask, dtype=src2.dtype)
+
         for layer in self.layers:
             src2 = layer(src2, src_mask=mask)
+            src2 = self._zero_padding_positions(src2, valid_mask)
 
         src2 = self.final_norm(src2)
+        src2 = self._zero_padding_positions(src2, valid_mask)
 
         return src2.transpose(0, 1)
 
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
index d63e16f0..69fa9fda 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,7,0.41149214,-0.7849847,-0.28549865,0.4135578,-1.5136797,0.28504756,-0.2592171,0.3586638,0.0096122995,0.68446684,0.27280936,1.1332684,-0.9290369,1.6734341,0.96415323,1.5616682,0.6196785,-0.47138083,-1.7291825,0.6455711,-0.30171967,0.43178305,0.0013747738,-0.076876126,-0.26705164,0.09099751,-1.0307256,0.8569853,0.077619076,0.80527884,-0.3927059,-0.48262763,0.77014446,-1.2891862,-0.6874516,-0.18790592,1.2064557,-0.4788146,1.7160684,-0.37874106,-1.8563964,0.94303226,0.9052338,-0.8621848,-0.46681213,-2.1950614,-0.69286275,0.12579149,0.017325861,-1.8529755,-1.4810988,-1.7104719,-0.908732,1.3428756,1.6590891,1.4576144,-2.0680587,2.074148,0.1254711,-1.3453596,-0.150765,-0.31348786,-0.03713879,0.7493652,-0.09301235,-0.70159227,0.36488333,-1.3442415,-0.74384296,-1.0073787,0.5603107,1.5509155,-1.1920458,0.18350546,-0.26176476,-0.46722165,-0.17439422,0.7131231,0.48848873,-0.995511,0.1133322,-1.4447079,-0.022810481,-0.91964275,0.9241803,-0.87918293,0.80785894,-0.8586497,0.55474204,-0.09638581,-0.28760654,0.24903235,0.24263471,-1.3331411,2.1252306,-1.6045873,0.092501745,1.4959875,-0.7847978,-0.07687911,-1.3790759,-1.2054987,-0.4129643,1.175199,1.6125137,-0.15670183,-0.0052074674,-2.0629342,0.41843984,1.343122,1.3168796,0.4892252,0.8831178,-0.56325597,-0.40071782,-0.2586007,0.8234577,0.3491471,-2.5417411,1.2765507,1.3083528,0.13020971,0.56978697,-1.4101067,-0.31448993,-0.09979597,0.82165474,0.16443604,1.0280895,-0.22518575,-0.7565461,0.22468717,-1.430905,0.5723812,1.408025,-0.66557765,-1.2709581,-0.47722962,-1.0509661,2.1456828,0.35895014,-0.3888584,-2.3072958,0.28834087,0.9454994,1.2007508,0.13654324,-1.0512546,1.1556263,0.28248435,1.1180621,-2.4063637,-0.049587123,-0.4283151,0.34162042,1.1140897,0.74316895,-0.07375317,1.7339555,0.06241791,-0.14042361,-2.4541376,0.76050717,1.1560489,-0.81901497,-1.2859453,-0.049847722,1.0361446,-1.1722547,1.2732593,0.5962597,-0.30993164,-0.031384904,0.20975727,-0.23721394,0.710139,-1.6749394,-0.05069577,1.013451,-0.03550218,0.873963,-1.3086079,-0.043819394,-0.0068593933,0.054935377,2.020619,0.013108201,-1.2887985,-0.85200655,0.054864895,-0.44489822,-1.6583155,-0.63917583,1.483576,-0.23919883,1.4165703,-0.97365326,-1.2111417,0.5758731,0.2508152
+0,0,7,0.129906,-0.5716,0.18384615,0.48382017,-0.95800894,-0.06879192,-1.5230852,0.32024476,-0.032266557,-0.2324689,1.2026423,1.2646347,-1.6631967,0.8923384,1.0081543,2.146518,1.260503,-0.024123436,-0.4261092,0.71811444,0.6665925,-0.83603954,0.7274643,0.081591725,-0.46056688,-0.7607759,-0.5974563,0.5428709,0.04133952,0.6946457,0.54585093,-1.3706633,0.9921805,-1.1147716,-1.1846336,-0.6535198,1.6928166,-0.6169712,1.1777917,-0.33404595,-1.2148309,0.16786538,1.4557567,-0.7030484,0.40415078,-2.0602548,-0.44774872,-0.9443908,-0.37846982,-1.4340259,-1.0351791,-0.54817986,-1.3410672,1.057497,0.6667623,1.8221252,-1.484842,1.5296896,0.2569867,-2.118917,0.27047974,-0.802363,-0.34508198,-0.7845012,0.29995131,-1.5465921,0.3353841,-0.85972494,0.9633959,0.1057641,0.77337104,1.4520825,-2.3267972,0.8137759,-0.9976677,0.25013983,-0.27935335,0.8346441,-0.33092296,-1.0873228,0.49370176,-0.4727684,-0.7608272,-0.8022482,1.3239352,-2.2216249,0.55208087,-1.1424916,1.1881708,-0.4595066,-0.09527821,0.9202227,0.57056504,-2.0961475,1.643413,-1.046783,-0.116270036,-0.3729071,-0.61354154,-0.8894295,-1.1429787,-0.78733057,0.57443416,1.4698808,0.43613648,-0.075808406,-0.14003257,-2.9505506,-0.016151994,0.6177055,0.68450046,0.5607075,0.87257403,-1.3993292,0.35658643,-0.0063092983,-0.25653496,-0.31633425,-3.5138404,0.88777286,1.6519107,0.40932092,-0.027439872,-0.79843307,-1.1049063,0.07724614,0.56875324,-1.0056511,0.48330793,-0.157245,-1.6178644,0.1934022,-1.9760464,0.0589674,0.5110818,0.089602806,-0.968479,-1.0943018,-0.7229412,1.4727569,0.47599703,-0.6288432,-2.055541,-1.011629,0.9921429,0.5582884,0.50531685,-0.097561255,0.731814,0.6783084,0.3472921,-1.7636684,0.27717423,0.28878227,0.5271399,1.5502092,0.1620326,-0.50224847,1.5970831,-0.6694094,-0.53928703,-1.924573,-0.013257886,-0.10813448,-2.0357018,-1.6700076,0.2103319,1.2394764,-1.1461264,0.036453854,0.40960938,-0.83274007,-0.008255889,-0.30468202,0.5700774,0.6194162,-1.6036439,0.2115395,0.6522949,-0.39073604,0.21115232,-0.46797746,-0.018942848,0.09517513,-0.61526686,0.67094487,-0.45691407,-0.42524102,-0.8716745,0.07637033,-0.72497505,-1.2465461,-0.4429114,1.5433344,-0.8214027,1.8114822,-0.90004414,-0.6661404,0.8775225,0.91526055
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
index 3e39464b..31fbd49b 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,7,0.42257375,-0.17903732,-0.20059693,0.29945794,-1.3389122,0.03707557,-0.299859,0.40720436,0.15333666,0.64613265,-0.47270918,0.8683621,-1.0405228,1.8252618,1.1951929,0.98608077,0.6486402,-0.27345127,-1.749177,0.3916777,-0.06327997,0.5837565,-0.029801903,-0.4157889,-0.8207295,0.56711143,-0.9630651,0.7541865,0.040734746,0.6285889,-0.0074708797,-0.26125398,0.5846216,-1.6676134,-0.6075296,0.39818704,1.1361198,-0.14915793,1.784923,-0.2994867,-1.1560175,1.0586625,0.8207005,-1.4820725,-0.59732884,-2.1600444,-1.0302081,-0.14026557,-0.16139339,-1.6454352,-2.011675,-1.9869673,-0.7221869,1.3151264,1.5895944,1.3244375,-1.8021666,1.6869603,0.08559678,-1.2588723,-0.07605097,0.048324257,-0.35193312,0.9454642,0.104382835,-0.9171362,0.35889798,-0.6452473,-0.6342578,-1.0011637,0.2077998,1.4437941,-0.9567251,0.14908887,-0.24294905,-0.687203,0.05336324,1.0594568,-0.07940463,-1.1028723,-0.08218197,-1.3812248,0.22032668,-0.9392466,1.3229952,-1.090876,1.0943706,-0.5451269,0.12716077,0.06487062,0.0940708,0.37863347,-0.3223036,-1.1197766,1.7777491,-1.5668588,0.25907642,1.4157779,-0.7196445,-0.38355565,-1.5430949,-1.3738754,-0.7312654,1.3590227,1.652726,-0.31939238,-0.023165464,-2.2086487,-0.1569172,1.3741649,0.9475051,0.48269206,0.6875285,-0.534931,-0.24520622,0.24692726,1.1467029,-0.025945317,-2.077082,1.548747,1.3998401,-0.09173701,0.60546964,-1.7211899,-0.32520935,-0.6385531,0.58607954,0.63349676,1.2241516,-0.115974545,-1.2130051,0.3455885,-1.230523,0.3379228,1.3327022,-0.7980125,-1.4078081,-0.5755934,-1.0106425,1.8490473,0.6007134,-0.49850124,-2.4873343,0.3709469,1.0519376,1.3123273,-0.014704171,-1.0918032,0.4931872,0.14534308,1.087156,-2.4626598,0.4272686,-0.5852703,0.11401402,1.3979368,0.96870464,0.11613599,2.0814247,0.08643781,-0.25604638,-2.6740973,0.7021289,1.4160581,-0.9518831,-0.9351058,0.3818723,0.64308923,-1.4009758,0.878512,0.3460209,-0.6776031,0.16450398,-0.31286997,-0.18456456,0.3714644,-1.7571229,0.3403873,1.3923695,-0.17252812,0.6139845,-1.0345956,-0.1693556,-0.044252478,-0.30749786,1.657718,0.78084534,-1.4596986,-1.1058352,0.1094069,-0.053861905,-1.9468046,-0.2616004,1.6512417,-0.49316752,0.82700384,-0.6377279,-0.76203316,0.432102,0.4528831
+1,0,7,0.6310362,0.334674,0.16728853,0.36186692,-0.9478092,-0.3926634,-1.3003972,-0.044596065,-0.20526434,-0.026398035,0.19380486,0.8118285,-2.0392873,1.2787219,1.2542659,1.4919833,1.088667,-0.010551525,-0.8448257,0.43219802,0.56950766,-0.8026797,0.74732757,-0.42725757,-1.0895562,-0.21274805,-0.7626764,0.026918262,-0.5983657,0.28117272,0.73443246,-0.9515915,0.81785864,-1.6458837,-0.71132445,0.23738682,1.6034894,0.019240104,1.3241774,-0.20347075,-0.7043628,0.38013214,1.2755842,-1.4004871,0.46275398,-1.9880224,-0.9198009,-1.0473716,-0.65675557,-0.84835905,-1.3224105,-0.70712143,-0.7341193,1.1134466,0.6568125,1.3994768,-0.954168,1.3808511,0.251891,-1.9545913,0.2308793,-0.26487717,-0.64339316,-0.2622317,0.603964,-1.4150456,0.6588127,0.08482099,1.1321739,-0.099473916,0.010979951,1.831232,-1.8722236,0.88451487,-0.76108027,0.080021404,-0.06698817,1.07159,-0.9351867,-1.0435191,0.026436348,-0.41959018,-0.52513933,-0.5412578,2.0294697,-2.7480896,0.53006375,-1.0863824,0.33880177,-0.14623298,0.6506472,1.0939751,-0.48882413,-1.810358,1.3499824,-0.89643645,-0.008765998,-0.06027395,-0.48046795,-1.2652203,-1.4560691,-0.946743,-0.27704963,1.9105397,0.20772824,-0.3049073,-0.34610513,-3.3094442,-0.46867853,0.65010065,0.4332434,0.5593448,0.8796511,-0.8775561,0.56195873,0.2979048,0.54411083,-0.49651542,-2.4436908,1.2963794,1.9646289,0.41595295,0.077845775,-0.8488377,-1.2093579,-0.2594095,0.12893897,-0.5105848,0.64491534,-0.017900532,-2.1271238,0.27657434,-1.7817888,-0.6613088,0.2365224,-0.07435218,-1.0712526,-0.80908316,-0.77087396,1.1399101,0.7127807,-0.94809335,-2.0319622,-0.7065952,1.1998774,0.56364894,0.62384087,-0.25050914,-0.12355562,0.011005514,0.4171477,-2.2401168,1.0053111,0.10580753,0.5745275,1.6763166,0.3952029,-0.3734366,2.0440457,-0.5067878,-0.59976774,-2.560222,0.28748652,-0.068856396,-1.7209245,-1.5318912,0.7257471,0.8677974,-0.9443491,0.32327485,0.09680152,-1.0634658,0.2343228,-0.73152024,0.6149891,0.50479555,-1.6433326,0.36874586,1.5886227,-0.48504025,-0.5036868,-0.24184056,-0.20931806,-0.046914995,-0.6832404,0.3294942,0.5522635,-0.77527386,-1.4231143,0.22487257,-0.16668221,-1.7301935,0.025722886,1.6323115,-1.0889417,1.2082901,-0.3673339,0.19397202,0.32195246,1.1962756
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
index 891aca89..c085fdb1 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,7,0.34137836,-0.54560816,-0.34839255,0.5934387,-1.4561145,-0.3552921,1.236413,0.2230994,0.52624667,0.35157225,1.0009108,1.453514,-0.99844396,0.52108085,0.99309534,0.9884845,-0.49107695,-0.44872406,-2.3321922,1.2099328,0.02559154,-0.15125763,-0.25792414,1.148591,0.34748212,-0.7123179,-0.9918085,0.72194755,-0.57125664,-0.5386358,-1.4469051,-1.0329736,0.2649161,-1.8303887,-0.22947203,-0.22510229,0.4087416,-0.6991921,0.27918908,0.2699201,-2.0308623,0.73651797,1.1014013,0.28801918,-0.043545906,-1.3224466,-0.728169,1.3165932,-0.38451204,-1.2290868,-0.3865898,-0.82043016,-0.15159693,0.95450556,0.720509,0.58943474,-0.96818477,2.4339142,-0.045729626,-2.2731185,-0.69900125,-0.72985935,-1.0451107,0.16168666,1.275655,0.0921365,-1.3980165,-1.8175794,-1.2656915,-1.2789646,0.65161574,2.2280807,-1.0717951,0.7397777,-0.35529038,0.13295835,-1.1914927,0.17240041,1.2711916,-1.2801683,-0.7685431,-0.809741,-0.24176663,-0.520219,1.2797531,-1.700323,0.7044165,-0.28107253,-0.3459173,-0.61043745,0.52076197,0.051599972,-0.45293874,-0.06818946,0.8897772,0.097269796,0.41846913,1.6724628,-0.749013,0.34057495,-0.7477783,-0.06767678,-1.2013714,0.9092441,1.5504543,0.3554491,1.1866004,-1.1053133,0.1902895,1.1757743,1.115244,0.25665316,0.61796343,-0.47255665,-1.0932264,-1.2121068,0.3602891,1.3848262,-2.0418153,-0.25206277,0.64594483,0.06378719,0.63946444,-1.0031189,0.08267042,0.7659137,0.8319407,-0.17579815,0.7439447,-1.1327,-0.6398448,0.55752283,-1.4105185,0.6037283,0.882441,0.2301522,0.016749382,-0.097922415,-1.7987541,-0.056111883,0.7459678,-0.4149461,-1.682275,0.67691326,-0.4379247,0.76354676,1.4371464,-0.36294943,2.097733,0.2694331,1.5530957,-1.4040655,-0.9152303,-0.5327291,0.41197965,1.040547,-1.1637155,-0.59128654,1.9143434,0.75515777,-0.5771012,-1.1852001,-0.2920808,0.5541353,-1.050848,-0.79349184,0.107663386,0.108831875,0.8791816,2.1749172,-0.9475491,-0.28542468,-0.5886931,0.441553,-0.010180896,0.54911983,-0.48679575,0.46939653,0.29680255,-0.33886543,1.8130943,-1.2549541,-1.0724515,0.56288785,-0.12585218,2.9909089,-1.5593657,-0.92594236,-0.9341071,-0.14739718,-1.5810376,-0.5024728,-1.707612,2.2963934,1.5836017,2.0804906,-0.6040425,0.7326207,0.4728579,-0.083124615
+2,0,7,0.14896812,-0.7527326,0.098616265,1.5360665,-0.83212125,-0.8661572,0.87993956,0.7707245,0.8990906,0.35979807,1.5049816,0.9981651,-2.2045355,0.46189427,1.3522851,0.81136656,-0.3787586,-0.41562492,-1.6569216,0.9181141,0.8893558,-0.6614972,0.35581687,1.0705252,0.61813873,-1.6933419,-1.6523572,0.4721115,-0.6415743,-0.8052874,-1.1905364,-1.3483986,0.7520653,-1.4446422,0.032763932,-0.77299255,0.8698881,-0.9012206,-0.105312526,-0.03295452,-1.7241203,0.40647608,1.3450831,0.00073365227,0.23433386,-0.9771891,-0.1076481,1.2662536,-0.7311288,-0.6664411,0.006385311,-0.31623417,0.108964175,0.43619278,0.284846,0.34632137,0.13046862,1.7500304,-0.31573698,-2.1822793,-0.073760144,-0.63938767,-2.4171443,-1.3825034,1.4659554,0.061223067,-2.0762868,-1.1751351,-0.081808135,-1.6416469,0.61133504,1.4318693,-1.5806884,1.4651628,-0.42543516,0.27195463,-1.9691073,0.5124442,0.7785031,-1.2194265,-1.2140163,-0.21277447,-0.64433247,-0.34592688,1.480311,-1.6828288,0.5201631,-0.45541105,-0.17800705,-0.29835147,0.06779409,0.17683165,-0.5690362,-0.4795994,0.24317117,0.7038561,0.47528592,0.38569337,-0.92215085,0.12652376,-0.5222174,0.79115415,0.1062228,1.2142566,0.57600933,0.3640244,1.4705232,-1.8675983,-0.151888,0.84477943,0.5522926,0.25622973,0.50753284,-0.87316334,-1.7905984,-0.8826149,-0.80747306,1.2090178,-2.0752993,-0.18883346,0.35716313,0.12954345,-0.5863821,-0.3407679,-0.8663318,0.8249107,1.2127439,-1.0448643,-0.53810304,-0.6317458,-0.50353926,0.4781922,-0.91658026,-0.21077591,0.49530765,0.66140443,-0.061057545,-0.011890421,-1.4830853,-0.5170973,1.2860987,-0.48179677,-1.2840593,-0.12392463,-0.5612262,0.5283732,1.6894128,-0.015182488,1.6194154,0.6165103,0.846802,-1.259089,-0.19867773,0.05881237,0.87158996,1.7141718,-2.1852305,-0.17253563,1.7481165,0.7287171,-0.9333205,-0.5228718,-1.1484395,-1.0014697,-1.5860218,-0.9670557,0.5662043,-0.6358826,0.6285491,1.4611297,-1.5053117,-0.73541224,-1.1782357,0.4759477,0.56338716,0.29015008,-0.5925466,0.4408311,-0.23582521,-1.1164836,1.0048937,-0.39400414,-0.7058568,0.99132824,-1.3044466,2.0163782,-1.8661258,0.013690879,-0.98518753,-0.2840235,-1.4385029,-0.70052123,-1.5944667,1.7985247,0.9840921,1.857999,-0.74875927,1.5471251,0.73758185,0.14302142
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
index d8417f89..961e9b5e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,7,0.13949223,-0.8288385,-0.011691158,0.6978597,-1.7423092,-0.41049773,0.4738875,0.43795598,-0.13211526,0.83174855,0.18503393,0.8696095,-0.98063534,1.320197,1.2542927,0.804681,0.4159498,-0.5308098,-2.063883,0.7713375,-0.040647052,0.50756997,-0.47521272,-0.050390225,-0.41869515,0.15082894,-0.91570437,1.1039559,-0.119457014,0.630291,-0.78385514,-0.26605347,0.30441046,-1.7381796,-0.3145166,-0.09518902,0.98777574,-0.5672071,1.5451479,-0.5920584,-1.8862137,0.73404765,0.76614827,-1.2776103,-0.5546785,-1.6537995,-1.1388252,0.5387951,-0.04569923,-1.8089854,-1.3872888,-2.010136,-0.32747167,1.414386,1.5741073,1.1150281,-1.6234998,2.3318863,0.09676073,-1.2265372,-0.32222414,0.23237921,-0.34041348,0.7922307,0.6937952,-0.47161567,0.18344484,-1.5619589,-1.089482,-1.4129832,0.28963518,1.6352048,-1.1390606,0.18842995,-0.38225302,-0.65850925,-0.33246985,0.6151895,0.6373732,-1.5749953,-0.327212,-1.2576468,0.2955655,-0.5931339,1.2506938,-1.0565152,0.96414435,-0.8558795,0.34775218,-0.5049455,0.26373684,-0.20436549,-0.080223046,-0.94886553,1.4435066,-0.92269295,0.47257257,1.7526731,-0.9398457,-0.03447294,-1.2984382,-1.0528736,-1.0664232,0.9039165,1.928867,-0.4283627,0.48781595,-1.5944821,0.19360305,1.8330988,1.3015158,-0.05350088,0.53769165,-0.29242146,-0.8438974,-0.069205664,0.82746214,0.84352636,-1.9790401,0.94773203,1.0971557,0.038646087,0.88122386,-1.4750324,-0.11020492,-0.09284775,0.32929987,-0.14679804,1.1929734,-0.6901023,-0.6029133,0.21264303,-1.3289111,0.8272518,1.2592579,-0.38540855,-1.034611,-0.0327922,-1.550678,1.1788012,0.53695285,-0.36779112,-1.9978998,0.6491437,0.62098545,1.3861,0.43899333,-1.0762482,1.1092504,0.23858577,1.405184,-2.151006,-0.51262474,-0.6709533,0.47459316,1.2390981,0.41524518,0.045998752,2.0420544,0.5617183,-0.0899196,-2.7423296,-0.106221266,1.0869207,-0.7299839,-1.148185,0.08723943,0.6347005,-0.4051242,1.4006757,0.099802785,-0.47955215,-0.47176704,0.27028126,-0.8404873,0.59273887,-1.5649644,0.14975634,0.9384124,-0.36595413,1.3495317,-1.2204468,-0.33850634,0.27997792,-0.07943602,2.6751914,-0.09228743,-1.481454,-0.97814053,0.20486094,-0.731338,-1.5517998,-0.83328533,2.0616615,-0.18377517,1.3525189,-0.7911948,-1.0217361,0.5704211,0.053546384
-4,0,7,1.1079531,0.56694883,-0.51460046,-0.44087544,-0.43843356,0.25955358,0.13465126,-0.19686213,-0.0069119474,1.2750784,-0.36285907,0.29798138,0.2409469,1.3927422,0.2650472,0.2537195,-0.099152684,1.1273077,-1.4502947,-0.10494784,-1.1681588,1.3679986,0.028042292,-0.14520136,-0.9369158,1.5287392,-0.46746886,1.6174921,0.15947264,1.0661743,0.06785149,0.16602738,0.6872166,-0.90826625,-1.1785315,0.49854565,-0.70833915,-0.57204914,2.0476813,-1.371946,-1.3191279,0.64045453,1.1497461,-1.2700158,-0.91564953,-3.0730312,-0.0084550865,1.1021385,-0.20811193,-0.9027882,-1.3729584,-2.1781166,-0.82236534,0.14685307,0.39046445,-0.108870916,-0.63499683,1.680812,1.4627557,0.48326835,-0.4139974,0.048960343,0.16166344,1.8954678,-0.5226412,0.20060515,-0.04435112,-0.754301,-0.71545064,-1.4038323,0.069040895,1.1398026,-1.154028,-0.43103975,0.54638636,-0.74899,-0.14519033,0.9119781,0.7352537,-0.1921951,-0.4378232,-1.6286504,1.354772,-1.0531448,1.8048595,-1.1451042,0.80384815,0.22715564,-0.24490066,0.5841813,-1.1601851,0.73505855,-0.39013195,-0.18186606,1.2028245,-1.022498,1.2703432,2.3782692,-0.9626023,0.44522825,-1.4234884,-1.6156629,-0.5665078,0.95502675,1.4593245,0.47318575,-0.8122251,-1.0151421,-0.82546866,0.6082922,1.1691772,-0.5170128,0.067346595,-0.6556871,-0.5575009,-0.0006836925,2.0677729,-0.63626707,-1.0719508,0.08163928,1.2308236,0.3763118,0.33964872,-1.3674697,0.8556785,-0.69404805,0.6134989,1.0646461,0.9832764,-0.34057733,-0.12112485,0.23851065,-0.026012735,0.8402205,1.8853765,0.90379107,-1.7303153,-0.17408463,-0.7569175,1.1544394,0.10548308,-0.58828104,-2.2133024,0.76151407,0.9882251,-0.33799013,-0.36300513,-0.32815427,-0.16083471,-0.7817147,-0.15756014,-2.7104619,0.73015183,-1.8245586,0.12698413,2.2169824,0.95323753,0.43615663,1.2318543,0.45343706,-0.06706062,-1.7847573,1.080589,2.519145,-0.86268,-0.17361444,1.168513,-0.91562283,-1.569328,1.259442,0.037228383,0.30334857,-0.23501681,-0.49459893,0.3947516,-0.39016256,-0.7743366,0.74605846,0.21725878,1.8476028,0.54618216,-1.5716995,0.9664981,0.12692592,0.3307412,1.9256802,1.494507,-1.164124,-0.9523818,-0.61173975,0.45403442,-1.0929524,-0.046263948,1.1769654,-0.009714122,-0.049784996,-0.15999958,-0.46489516,0.8143899,-0.30355412
+3,0,7,0.38537663,-0.76236117,0.54453766,1.1489445,-0.93219095,-1.0841125,-0.279618,0.190426,0.066999875,0.60205054,0.7307121,0.975402,-2.318475,0.97368747,1.6034338,0.94340336,1.2762036,-0.30378067,-0.9700833,0.62717986,1.3900626,-0.94647413,0.48373678,-0.030570993,-0.42904085,-1.1276957,-0.7923079,0.563961,-0.40978292,-0.2334528,-0.27136517,-0.8576837,0.69629365,-1.6003033,0.11132377,-0.5791753,1.9399358,-0.9257662,0.5349056,-0.49527618,-1.3456327,-0.17603794,1.3494829,-1.1282389,0.20037875,-0.9559931,-0.8822725,-0.07519408,-0.77170074,-0.9350284,-0.5532877,-0.94019765,-0.10448791,1.1707094,0.6575927,1.3083346,-0.61123514,1.6713382,0.16979483,-2.0898654,0.009840045,-0.44183388,-1.1921828,-1.265742,1.6432867,-1.1305618,-0.46437404,-0.8430221,0.7453205,-0.65720564,0.35852164,1.6237742,-2.0383017,1.1427602,-0.8917945,0.2076105,-1.3489536,0.6207456,-0.44713888,-1.8621222,-0.5244267,-0.45581636,-0.3975574,-0.4345815,1.8005577,-2.0100255,0.3588848,-1.2393867,0.95991385,-0.65878433,0.7419609,0.43640068,-0.37811753,-1.3537853,1.0124421,0.12569557,0.45669934,-0.062139753,-0.68999034,-0.7770923,-1.26833,-0.42139238,-0.41232172,1.722412,0.38184437,-0.50226235,0.19698814,-2.7483191,-0.18802643,1.3159317,0.4502178,-0.26388723,0.5703611,-0.85277057,-0.46127146,0.2382723,-0.42989612,0.4115149,-2.6187692,0.5646321,1.1139393,-0.032928452,0.10219407,-0.6601092,-1.3054988,0.093954325,0.43201578,-1.4673975,0.08933542,-0.7154099,-1.6057762,0.40757596,-1.6277497,0.05748511,-0.1892006,0.10566295,-0.8918468,-0.17913234,-1.4436157,0.07747513,0.6109399,-0.8347807,-1.3056538,-0.8713027,0.28275943,0.87232286,0.8705629,-0.14861424,0.91572547,0.6586051,0.67494464,-1.7701943,0.16297512,0.4728508,1.0638494,1.7032028,-0.6542731,-0.017882321,2.1696196,-0.04748722,-0.77344686,-2.259672,-1.1815661,-0.58230156,-1.6384532,-1.8893389,0.917642,0.3281549,0.40393066,0.178792,-0.32735556,-1.1835171,-1.0626796,-0.12413608,-0.13082595,0.59983134,-1.7939893,0.5362146,1.2261001,-1.4384091,0.4796257,-0.45061558,-0.10099141,0.51506233,-1.2538948,1.3633755,-0.588958,-0.21253218,-1.3333261,0.5079261,-0.97876984,-1.5163478,-0.66665024,1.8272513,-1.481488,2.3199954,-0.635016,0.18672499,0.83364874,0.9136821
+4,0,7,1.832777,0.9652215,-0.2531473,0.2162134,1.9031242,0.70369786,-0.9306243,-0.88834244,0.55274016,0.5570371,1.0132977,-0.9646427,-0.5195061,-0.055798862,-0.37479177,0.4407284,0.16671644,0.5790386,0.41375777,0.80011296,-1.5359619,-0.22593135,1.7854866,-0.71542126,-1.0921632,-0.04137182,-0.21795602,-0.33691144,-1.2671196,-0.7473329,0.18677165,-1.8774186,2.0318227,0.273334,-0.9697613,-0.5860969,-0.36416003,0.2273136,0.47210863,-1.3540341,-0.25599045,0.6586222,1.5556432,-0.8154356,1.168724,-2.2140863,0.42116508,0.4787695,-0.64788985,0.4423342,0.20755024,-0.55014634,-0.595621,-0.6086162,-1.3033568,-0.92190075,1.9484235,-0.028180895,0.9555864,-0.610766,1.3185904,0.42626792,-1.1561954,-0.5161907,-0.83520865,0.22098021,-0.32627407,-1.1956958,1.73692,-1.1766446,0.07778312,-1.4228519,-2.1649313,0.6194682,-0.44674367,-0.010885017,-0.43634287,1.7373236,0.6609659,-0.20005386,-0.5650337,0.29500005,0.115850784,-0.054047436,1.7766252,-1.624335,0.4819138,1.0377665,-1.2305183,0.80048895,-1.9219421,1.6361128,0.13642035,-1.2401028,0.56233996,-0.15820792,1.5966988,-0.2590406,-0.5514963,0.43703985,0.40010962,-0.12492114,1.6866367,2.0976324,-0.49593025,0.8581704,0.21792285,-0.63466465,-1.7050067,-1.6924849,-0.08375951,-0.38175452,-0.3269363,-0.96400464,-0.90264994,-0.5941345,0.1949616,-0.8319494,-0.5528981,-0.80228716,1.6503075,1.2991755,-1.0835283,0.3950482,0.2928661,-0.13108586,-0.38219264,-0.5976724,-1.3788452,-0.2650713,0.34214625,1.046247,1.518125,-0.5940697,0.589522,2.7725646,-1.0286809,1.1897798,0.33696088,-1.0014184,-0.029821016,-0.5898694,-0.9996754,0.0104257725,1.4441028,-0.87508523,0.57642156,2.500753,-1.5390791,-1.6458074,-1.7197642,-0.74723464,1.16423,-0.3134553,-0.005579411,1.5905838,-0.9592407,-0.41740352,-0.6339657,0.8530593,1.299676,-0.11439452,0.39143115,0.46125954,-3.0907333,-0.27514967,0.20107317,-1.4204426,-1.7638286,-0.4874675,-1.2546072,-0.23415713,-0.10110022,-0.24766764,1.7387943,-0.75891596,0.75030476,0.5836463,0.10834095,0.7989047,-0.48462987,-0.62584233,0.5027705,1.1866,-0.2813258,-0.8015657,0.4489704,1.1347668,0.07279327,-0.6131345,0.88922334,0.71706647,0.65837896,0.33229914,1.1745048,-0.74355304,0.5451966,0.36238006,1.4138216,0.37747213
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
index 5144f594..1ede8ed4 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,7,0.30457106,-1.0499812,-0.2708557,0.6033656,-2.1559174,-0.014660421,0.40023395,0.35113475,0.0902287,0.7888887,-0.34383535,0.93346864,-1.0638552,1.6174538,1.5091573,0.9228686,0.086696744,-0.50717336,-2.0618286,0.39436528,-0.021834815,0.6010914,-0.47710696,0.17810024,-0.91974175,-0.12123473,-1.365223,0.79932564,0.19411209,1.1538398,-0.63365114,-0.39145467,0.2746975,-1.130279,-0.6916063,-0.12119007,1.0064207,-0.330205,1.6648016,-0.45908698,-1.8615019,1.1402794,0.75518245,-1.5022097,-0.9118628,-2.2375493,-0.902504,0.6392769,-0.21022926,-1.8474433,-1.5057714,-1.960203,-0.3685194,1.2326003,1.5652041,1.0757546,-1.4903946,2.2756448,-0.32394972,-1.1672122,-0.059360582,-0.04470684,-0.8703109,1.0302415,-0.0895821,-0.403575,-0.37002468,-0.8056304,-1.1433159,-1.2743354,0.85718685,1.1835376,-0.6666402,-0.14671929,-0.04834712,-0.98218584,-0.13421746,0.78738457,0.38280028,-1.2351733,-0.48965338,-1.6121022,0.42098215,-0.85432166,0.9951741,-0.12262745,0.9710562,-0.5171932,0.16173087,0.17513205,-0.07830988,-0.34751597,-0.012498501,-0.66930187,1.9089013,-1.4641144,0.5140048,1.7203206,-0.9557444,0.13077603,-1.1026406,-0.9473788,-0.31690073,0.97551376,1.9448526,-0.476878,0.7957327,-1.8465495,0.67249143,1.7720664,1.2585933,0.45854238,0.5997015,-0.6044482,-0.76524645,-0.30329454,0.5679286,0.5396677,-1.7696911,1.5391197,1.026936,-0.31885958,0.50461066,-1.7645643,-0.50854623,-0.4742418,0.77532613,0.09777384,0.54422975,-0.1796674,-0.8478994,0.0031664053,-1.1256933,0.77754414,1.3871409,-0.26337987,-0.8610575,-0.51069754,-1.2064705,1.7107098,0.7312288,-0.28927425,-2.1887672,0.9429601,0.9586188,1.8643379,0.19833568,-1.2076133,1.2674617,0.3626933,1.0161105,-2.1542332,0.07333201,-0.60430324,0.2051994,0.90605885,0.19795726,0.60270673,2.1178296,0.3766683,-0.5104417,-2.3974814,0.124425165,1.0825484,-0.41810575,-1.0645667,0.04239827,0.48415333,-1.4814224,1.1013521,-0.101820946,-0.6659215,-0.4296166,0.30863544,0.20719278,0.43899494,-1.6712543,0.00065493764,0.7002628,-0.41546488,1.3948044,-1.0820502,-0.11341275,-0.15781793,-0.2200905,1.9745494,-0.07587735,-1.6396667,-0.65765786,-0.13210705,-0.30101702,-1.8123761,-0.7385888,1.5914155,-0.37091118,0.9381928,-1.189511,-0.54202545,0.30190322,-0.34896195
+5,0,7,0.8678485,-1.1492084,-0.29577827,0.9786946,-1.9542456,-0.07664234,-0.03973026,-0.09718513,-0.46550712,0.088307075,0.014937765,1.1039448,-1.9457144,0.7994327,1.8773035,1.5537525,0.39788535,-0.4953131,-1.4370357,1.0003624,0.841088,-0.99132866,0.15745488,0.4423288,-1.511852,-1.9748083,-1.3705088,0.117527954,0.0471839,0.72272456,0.15624413,-1.3321223,0.8274153,-0.50209063,-0.85796344,-0.82082087,1.3705729,-0.32510358,0.92136246,-0.4567902,-2.0167284,0.778191,1.3592294,-1.6879605,0.3511618,-1.6433352,-0.8796016,0.4710184,-0.49356312,-1.540319,-0.5620308,-0.7479297,0.11485024,0.5929443,1.0806792,0.8520335,-0.28449163,2.3426988,-0.7176778,-2.077216,0.44693425,-0.51118195,-1.8134435,-0.3199498,0.5275097,-0.7924806,-0.90600455,-0.5945148,0.62720704,-0.76678073,1.4406856,0.83508265,-1.3691335,0.43675628,-0.93423057,-0.587019,-0.5398221,1.3742958,-0.36840782,-1.3804276,-1.3538328,-1.2029364,-0.6361642,-0.20616663,1.1179503,-0.86969507,0.46374834,-0.9316529,0.36905083,0.18406925,0.57265496,-0.09701405,-0.3084512,-1.0851078,1.2294931,-0.43448195,0.4714254,0.2976226,-0.6675889,-0.24110621,-0.86843914,-0.15433894,0.3217168,1.7343627,0.64216596,-0.39611214,1.2057974,-3.0774295,0.8041049,1.0713865,0.31967443,0.8850117,0.5413772,-1.8904046,-0.64331496,-0.7131405,-0.8982892,0.49162266,-1.7892696,1.266022,1.2143359,0.16972244,-0.42226094,-0.920911,-1.287436,-0.30553937,0.78019506,-1.3130422,-0.54638034,-0.06269855,-1.2766076,0.3067186,-1.1979859,-0.5041536,0.24134868,0.9985403,-0.20058866,-0.19003277,-0.91562223,0.7388305,0.60628974,-0.39632004,-1.401826,0.266171,1.2859257,1.7502289,1.1291436,-0.2593596,1.0141768,-0.0024154119,-0.012608729,-1.6622946,0.84452707,0.4089601,0.24172992,0.8491719,-1.2459073,0.44094768,2.3070037,-0.20285222,-0.9055848,-1.9905902,-0.62915814,-0.7049501,-1.5236275,-1.6411532,0.28731275,0.23528811,-0.8580845,0.87269217,-0.8224613,-1.3814896,-0.8996288,0.0075937505,1.4490063,0.6224928,-1.3583354,-0.39259246,0.55505186,-1.1764414,1.2858897,-0.51201284,-0.36201963,0.028119631,-1.0583313,0.626196,-0.93444836,-0.6852238,-0.55748594,0.1277266,-1.1170981,-1.7487799,-0.62091154,1.1936655,-0.5524272,1.6852114,-1.5539126,0.93077433,0.54523623,-0.3293535
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
index 9f0978c3..541113f3 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,7,0.7273449,0.009315487,-0.5659161,0.0539636,-1.1747422,0.11731785,-0.24878183,-0.34799668,-0.39417046,1.5134698,0.180432,0.45988694,-0.48523727,1.8745505,1.0106969,0.83485544,0.46109983,0.18401664,-1.8427206,0.5820249,-0.9933856,0.57772356,-0.27879953,-0.88057303,-0.9495125,0.27947068,-0.98555374,0.65584594,-0.7251935,0.63879484,0.27707863,-0.23204361,0.8994904,-1.6550843,-0.41345245,-0.13837196,0.25872716,0.28248805,1.929113,-0.9793597,-1.2198358,1.0012894,0.7334509,-1.2398807,-0.88356876,-2.096589,-0.55407935,0.06886668,-0.776487,-0.90201944,-2.001808,-2.458255,-1.0559742,1.7568047,1.9512438,1.0935278,-1.6113944,1.89258,1.1544857,-0.70098376,-0.571621,0.7649165,-0.5899035,1.3507531,-0.7255975,0.17241018,0.50311905,-0.6263059,-0.36380523,-0.38990656,-0.27967066,1.416065,-1.308391,-0.352508,0.26046076,-0.18396786,0.50395614,0.2636389,0.86980325,-0.877985,0.18548113,-1.2215562,0.33346045,-0.9510769,1.7036265,-1.5579641,0.6795726,-0.18158613,0.011276826,-0.1970394,-0.26797757,0.3612404,-0.5695245,-1.4026842,1.758898,-1.2169626,0.73240304,1.9216541,-0.5449494,-0.14847618,-1.4680748,-1.6136966,-0.1330753,2.083131,2.207608,-0.20060927,-1.1179813,-1.3244749,-0.3105317,1.1410857,1.0294605,0.12662466,0.8479376,-0.43145418,0.2132454,-0.025244722,1.6760103,-0.2850787,-1.5536851,0.5495985,1.1164563,0.2283902,1.3229661,-1.758217,-0.4295333,-0.5159716,0.14204884,0.31548834,1.3842486,-0.54636705,-0.62329435,0.13254096,-0.53141403,0.2421158,1.3180143,0.27180845,-0.8231482,0.23262133,-1.0528578,1.495885,0.19461124,-1.0369684,-2.0902848,-0.4222444,1.0143939,0.5824076,-0.20892456,-0.4718485,0.06580811,-0.58581954,0.8422832,-2.566362,1.0881084,-0.64552516,0.3792994,2.2656064,1.8036302,0.2566192,1.7502713,0.65841585,0.25055784,-2.5194345,0.84616184,1.1300479,-0.60688585,-0.82479155,0.24982473,0.5697576,-0.6672031,1.3321275,0.23418252,0.32546026,0.52113616,0.41371918,-0.2717294,0.47879305,-1.3640656,-0.10326316,1.3061558,0.70007694,0.5445003,-1.2608218,0.51818657,-0.36812475,0.9051562,1.1091714,0.82054716,-1.0394655,-1.0836576,-0.22977057,0.17767808,-1.7726516,0.3174601,1.4640669,-0.17567706,1.0871913,0.17054728,-1.0595478,0.054115366,0.44907287
-7,0,7,0.19220954,-0.49153456,-0.04641067,0.18849212,-1.5853618,0.12341867,-0.11363969,0.13727675,-0.06847534,0.76124924,-0.28678593,0.8900653,-1.0943688,1.5146927,1.2307183,1.5759242,0.80081123,-1.0552121,-1.5437708,0.8671235,-0.33835396,0.6237686,0.060264308,-0.4205532,-0.569102,0.31872287,-1.1516503,0.56125844,0.025306754,0.6017618,-0.29687238,-0.3951651,0.7847005,-1.504742,-0.2827757,-0.21072185,1.2361257,-0.2929404,1.6530923,-0.54373467,-1.6885573,0.8691418,0.6955893,-1.2405534,-0.5988692,-1.744112,-1.3133477,0.17552854,0.058676194,-1.7747884,-1.4241207,-1.7917409,-0.62875485,1.7578604,1.5654361,1.3629228,-1.6896278,1.6548164,-0.14070357,-1.9311712,-0.15432832,-0.25670385,-0.4815931,0.8380533,0.28672034,-0.9617282,0.51135164,-1.1442822,-0.4330058,-0.7627599,0.6438905,1.7754638,-0.89877915,0.2844743,-0.36010924,-0.62290066,0.05098487,0.43144348,0.37151685,-1.350735,-0.3010244,-1.2662656,-0.017898949,-0.89459026,1.0812829,-1.0964411,0.66727376,-0.4757698,0.45214677,-0.045206342,0.10654671,-0.018967189,-0.5759011,-0.76332796,1.7140834,-1.397187,0.37946275,1.4883934,-0.9071629,-0.39879188,-1.280823,-1.1647131,-1.0890014,1.5997056,1.4125903,-0.36653465,0.25216198,-2.067858,0.7692397,1.5507644,1.0313568,0.47055215,1.0971524,-0.27837852,-0.19586825,-0.053837664,0.94147635,0.6782325,-1.8499793,1.3899486,1.4842832,-0.20198905,0.9118752,-1.3156699,-0.32807803,-0.46155733,0.59480095,-0.001722204,0.80457634,-0.21580535,-1.2434112,0.56428885,-1.0542293,0.7155026,1.2230334,-0.9761847,-1.017852,-0.1672411,-1.4611741,2.0070598,0.71912134,-0.6476812,-2.0907934,0.44325376,1.0635188,1.3988785,0.15533331,-0.9266868,1.1053526,0.25385013,1.2198808,-2.3114061,-0.13443123,-0.2354869,0.25447476,0.91357034,0.8403027,-0.059212927,1.9575791,0.24892993,-0.31545094,-2.6089191,0.80578184,1.2054881,-0.7192824,-1.2654837,-0.09570098,0.94349724,-0.8900375,1.2061056,-0.01412794,-0.66192263,-0.2441078,0.041381236,-0.2976183,0.9883735,-1.9825552,0.07066407,1.6886215,-0.68948716,0.5837726,-1.1841415,-0.10145782,-0.008195412,-0.1493982,1.5905204,0.13115445,-1.1599777,-1.202373,0.2164183,-0.3614637,-2.1209643,-0.1951186,1.9769924,-0.68308234,1.2002243,-0.90931207,-0.6301795,0.1339641,0.01462584
+6,0,7,0.47420257,1.2638454,-0.29428783,0.6063462,0.3250649,-0.011651794,-1.383389,-1.234404,-0.45170534,1.0377884,1.8840407,-0.33119473,-1.0696625,0.9726884,0.80814546,0.9354395,0.71671236,0.428451,-0.015223929,0.81312406,-1.2162952,-1.1050556,0.6366257,-1.6387161,-0.9819109,-0.4953308,0.12158297,-0.67469645,-1.717451,-0.21087128,1.0931989,-1.6311557,1.6655924,-1.1295466,-0.8290883,-0.27834374,0.3571047,0.7679167,1.286279,-1.5655476,-0.47575822,0.3837134,0.82579845,-0.65413463,-0.060213037,-1.6886673,0.31571612,-1.3439902,-1.3430762,0.0015742916,-0.8718748,-1.2474332,-0.8262217,0.5390429,0.43257564,0.7165555,-0.013134941,1.0078024,1.2916026,-1.6230772,0.18352468,1.4117193,-0.99339473,-0.17660125,-0.8539951,0.17099175,1.0725634,-0.23263906,1.7940855,0.49845868,-0.58279014,0.9100562,-2.1336784,-0.08051058,-0.15371843,0.53641343,0.036809027,0.27225932,0.43129542,-0.51692873,0.86170065,0.3777755,-0.3340567,-0.5066735,2.3349266,-3.2664793,0.23907657,0.13636138,-0.13565235,-0.29753345,-0.56251806,1.5495608,-0.25081062,-2.0128417,1.4139545,-0.16621536,0.784448,0.5935694,-0.85539657,-0.5690536,-0.5415991,-0.8971023,1.8330036,2.7651486,1.039496,-0.23426558,-0.57880366,-1.1803632,-1.0954609,-0.30837312,0.2056301,-0.45784757,0.79928094,-0.58580524,0.8425376,0.11244702,0.72019076,-1.2291323,-1.666961,-0.4446293,2.2034054,1.0664692,0.81338817,-0.4559747,-1.2628947,-0.2225234,-1.3892788,-0.7460107,0.097924024,-0.91731095,-1.1128799,0.40236434,-0.7147696,-0.7736082,0.39257404,2.3405876,-0.27565396,0.40534735,-0.81300443,-0.097918786,0.61959285,-1.4504018,-1.5624175,-1.1680082,1.2857004,-1.2578357,0.41585726,1.0423983,-0.53722256,-0.38153195,-0.51949257,-1.5859617,1.3256052,-0.097611226,0.5610492,2.6576302,1.24299,-0.7566358,0.91004014,0.7399129,-0.1096572,-2.0155513,-0.009600659,-0.6858703,-1.7821821,-1.058989,-0.13436024,0.42209816,-1.1379089,0.4846043,-0.6162597,-0.33495048,0.5386396,0.18685773,0.26678535,-0.20175369,-0.72045624,0.21726054,1.1120163,0.79319596,-0.36998135,-0.67134833,0.7267623,-0.18856248,0.4676755,-1.0685669,0.32332313,-0.06708168,-1.2248515,-0.7845494,0.5460868,-0.6168076,1.5106831,1.7673893,-0.6841837,0.44412613,0.7936944,-0.8888349,0.039817397,1.6578218
+7,0,7,-0.08779775,-0.7206459,0.39195007,0.30277744,-1.0405563,-0.8779835,-0.7976403,0.071234435,-0.29967046,0.27734953,-0.105664976,0.77108765,-1.7671883,0.824929,1.5373951,1.5010271,1.4846263,-0.7896349,-0.8870087,0.69192314,0.6787518,-0.4330691,0.6862604,-0.2904696,-0.9619547,-0.3613372,-0.92799646,0.18862405,-0.13149612,0.33348697,0.35246965,-0.86275643,0.7647196,-1.5118189,-0.24293661,-0.62797713,1.9434764,-0.1523092,1.2640053,-0.53051686,-1.5717334,-0.23882772,1.2569783,-1.5155659,0.316097,-1.4623227,-1.5276753,-0.6301773,-0.15530248,-1.2590177,-1.1136596,-0.8283695,-0.3316041,1.6386757,0.7145627,1.4172399,-0.92403394,1.1554213,-0.2054311,-2.2690709,0.032772172,-0.60741836,-0.840842,-0.43909273,1.1674483,-1.5343083,0.39456716,-0.6319559,0.694979,-0.17052399,0.6638801,1.6770401,-1.6856104,0.8501758,-1.0100726,-0.06956186,-0.49753255,0.45815906,-0.8491857,-1.5388988,-0.21994215,-0.6159065,-0.43685117,-0.5483946,1.8440413,-1.9056607,0.239437,-0.8901972,1.0178717,-0.41829032,0.47147453,0.2824819,-0.4255798,-1.3693498,1.3788786,-0.8324858,0.2925801,0.038576633,-0.9931528,-0.88978964,-1.3541443,-0.3884408,-0.5666751,2.0802114,0.20410398,-0.5688041,-0.25536698,-2.8470824,0.22910744,1.3398811,0.41405335,0.37236086,1.0213393,-0.612114,0.20765163,0.22666231,-0.052260194,0.20032823,-2.3570495,1.430299,1.4743409,-0.2883624,0.44256932,-0.7141606,-0.87645125,-0.20459764,0.5313735,-1.1576004,0.2882263,-0.31856775,-2.041501,0.55573547,-1.3157725,0.2728312,0.18704545,-0.34522098,-0.7096739,-0.432244,-1.6015022,1.0732054,0.9189849,-0.6834102,-1.5872613,-0.5969934,0.90725124,1.1939424,0.6279301,-0.4908578,0.9603637,0.6682552,0.66679686,-1.698892,0.26019415,0.5773275,0.53784454,1.2200284,0.32803503,-0.15348214,2.141097,-0.0141839655,-1.0378202,-2.6077065,-0.3066732,-0.06714368,-1.3892244,-1.7354015,0.5355618,0.9041617,-0.3285599,0.12410717,-0.4072241,-1.3432459,-0.7320013,-0.25400367,0.17272092,1.1034756,-2.2850537,0.62561816,1.75507,-1.3452936,-0.017986165,-0.54053295,0.05288172,0.2870577,-0.98898715,0.76313764,0.15638581,-0.6830085,-1.3387133,0.6244851,-0.7181842,-1.9862365,0.07094988,1.840984,-1.5815738,1.6396283,-0.85405743,0.19265085,0.44211808,0.6777788
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
index d965bd46..84ef00be 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,7,1.8829988,0.5603056,-0.6805401,0.873459,-0.75400966,0.6241306,-0.022826796,-0.65359944,0.45811382,1.1624973,-0.36711055,1.1964695,-0.9203239,1.9030054,0.92825997,1.2234915,0.3935566,-0.10803351,-1.7747352,1.0153017,-0.22650294,0.07317609,0.7041025,0.083895385,-0.5787242,-1.1430625,-1.0159122,0.3586008,-0.81441945,0.12919727,-0.23688294,-0.8359273,1.0575047,-0.58928937,-0.19002041,-0.26165125,0.6974495,-0.35162425,1.4806678,-0.91830134,-1.5502017,1.193903,1.0963876,-1.9676751,0.075405344,-2.0842352,-0.2591765,1.1972393,-0.45678777,-1.2595507,0.0008469863,-1.209172,-0.04429603,0.6055306,1.2668537,-0.2412063,0.19914478,1.9480219,1.0576845,-1.475916,0.17119397,-0.6035175,-1.7098999,-0.37294888,0.419368,0.268539,-0.70104766,-1.7786356,0.5031316,-1.5002457,0.9026701,0.80984163,-1.2359964,0.22069247,-0.42576674,-0.39630565,-1.4970282,2.616495,0.6003689,-1.1450349,-2.02723,-1.7124385,0.30198652,-0.6123571,1.4489877,-1.4337479,0.3514351,-0.39667892,-0.68725485,0.8674336,-0.02994205,1.0586603,-0.8096146,-1.0183178,0.82786477,0.15899718,1.4175017,1.6423237,-0.7518763,0.46193466,-1.7427615,-0.89581615,0.25311762,1.9719808,0.84027904,0.5098649,0.69561577,-2.01775,0.16369472,0.09517597,0.4123239,0.43099934,-0.5587974,-2.099149,-1.5859907,-0.90635055,-0.023114711,0.15737575,-1.1646839,-0.39253628,1.4398996,1.0370032,-0.44213772,-0.64347506,0.49100915,-0.8488655,0.52995205,-0.12177242,0.4132956,-0.117767714,-0.041891925,1.1423306,-0.9095124,-0.7704257,0.6956402,1.2622151,-1.6017178,1.253087,-0.88926196,0.024840193,-0.6669459,-0.84649396,-1.5089233,0.5503707,1.0900313,0.52742964,1.4928386,1.0869912,0.35689026,-0.5618133,-0.1295253,-2.3059578,0.9978885,-0.25662556,0.12965643,1.9835927,-1.0982833,-0.09509576,1.6303843,-0.04004063,0.85300845,-2.1637661,0.115972474,0.8912982,-1.8220121,-0.67457515,0.4750627,-0.7254808,-1.0370337,1.8635073,-0.7830165,-0.7419083,-0.59229213,0.36369795,0.009352995,-0.652557,-0.9893492,0.13321461,0.05142843,-0.173441,1.0776938,-1.6504616,-0.21089926,0.84927285,-0.2011813,1.2290883,-0.25737432,0.17806257,-0.74700785,-0.22532028,-0.73679703,-1.1095785,-1.5106932,1.1296397,0.6030286,0.907326,-0.8425243,0.72134006,1.3058224,0.4379919
+8,0,7,1.6896031,0.29085124,-0.23292182,1.3335884,0.45459226,0.34372637,-0.4165882,-0.42207715,0.60733074,1.2122438,0.8359352,0.5889716,-1.7851549,1.4177666,0.7491194,0.5095495,0.26941702,0.21018945,-0.88023686,0.56112635,0.4965653,-0.8069379,1.1553066,-0.012635032,-0.1019034,-1.7098439,-0.778535,0.08457131,-1.0872964,-0.46575657,0.1705186,-1.0981392,1.2801948,-0.2028321,-0.3777968,-1.0450279,0.83502,-0.4449086,0.3772813,-1.3267674,-1.3647468,0.4980426,1.0723178,-1.55848,0.6211332,-1.7105299,0.93637806,0.962023,-0.5543043,-0.2215327,0.51130736,-0.5170269,0.11910182,-0.16186877,0.3342436,-0.8031384,1.6559254,0.4785719,0.5240285,-0.85450006,1.1488028,-0.123829335,-2.314366,-1.7480177,0.34907198,0.33689144,-0.96712893,-1.1980494,1.7140707,-1.4368652,0.665594,-0.2777855,-1.740489,1.0585523,-0.34576628,-0.026606081,-1.963539,2.3253458,0.718786,-0.63990796,-2.111302,-0.76395285,-0.37198,-0.00446455,1.4715577,-1.4121685,0.046070743,0.04943243,-0.6720886,0.77220064,-0.78486633,1.6651013,-0.5048683,-1.0169151,0.19552472,0.60602754,1.0623702,0.15507461,-0.32467106,0.2723276,-0.9548157,-0.0605602,1.6082473,1.8419511,-0.6165263,0.5530546,0.5313712,-1.6408414,-0.0837646,-0.62233794,0.053751737,0.17250709,-0.5837834,-1.9890919,-2.1878119,-0.6350772,-1.3977064,-0.09614897,-0.99611235,-0.87689996,0.8914697,1.3250514,-1.3865021,0.07022134,0.14449094,-0.4635786,0.22937033,-1.0663302,-1.124088,0.17105857,0.38474694,1.002527,0.06262806,-1.439073,-0.079469405,2.1103234,-1.7293836,1.1202102,-0.19353467,-1.0497702,-0.18288192,-0.8191539,-0.91999054,-0.0011520431,1.4552093,0.0285639,1.5781827,1.5662957,-0.37651268,-0.7598728,-0.96915853,-1.9078715,1.1331103,-0.11489686,0.537274,2.1304922,-2.0829363,0.2657547,0.73497164,-0.30037352,0.74554676,-1.2468303,-0.9414173,-0.96966434,-2.6400287,-1.1223296,0.87630385,-0.9678277,-0.9243955,1.2501684,-1.4089605,-0.80661803,-0.8789456,0.509323,0.86616665,-1.0657727,-0.44514504,0.16784553,-0.82460415,-0.6202602,0.22096892,-0.5744518,-0.051139973,0.95496964,-0.73176867,0.21573815,-0.73088634,1.3564851,-0.56419003,-0.3550247,-1.0255396,-0.57220864,-0.90364045,0.36711964,-0.09569923,0.6653445,-0.5859866,1.2525154,1.7926838,0.031228097
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
index ea9dd586..69bb2f81 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,7,0.1209258,-0.7836567,-0.3029543,0.2785893,-1.0025321,0.037851073,0.048571438,0.33848727,-0.3773915,0.7844221,0.24281782,1.148343,-0.13137017,1.279799,0.97056156,1.0246092,0.80062383,-0.3923224,-1.8497859,1.1537666,-0.38878572,0.5604603,-0.58034015,-0.02841155,-0.48242685,0.8719435,-0.9715767,0.86573464,-0.0033297834,1.0148878,-0.93299323,-0.20949924,0.95323825,-1.5853611,-0.5756929,0.1248061,0.5185648,-0.11291537,1.8827963,-0.75403863,-1.9923172,0.3773292,0.76919514,-1.0892999,-0.5601977,-2.0243154,-0.80850774,0.68828875,0.8130654,-1.7597806,-1.9047588,-2.6262157,-1.1613489,1.2342267,1.8630753,0.9635703,-1.9320338,1.985816,-0.055163626,-0.7479937,-0.36713076,0.07263066,-0.35597026,0.9809213,0.099624224,-0.22642994,0.59534913,-1.3628237,-1.0601852,-1.4001173,0.26098126,1.0985968,-1.100743,0.23194666,-0.3807315,-1.1726305,-0.24944744,0.57557213,0.2954017,-1.1652223,0.08526788,-1.5450522,0.21617737,-0.8069991,0.8448825,-0.76840484,0.887578,-0.55467606,0.28331083,0.06474368,-0.70855623,-0.25272217,0.19233815,-1.6127638,1.925872,-1.4058347,0.109641865,2.0278912,-1.168832,-0.40527755,-1.4968951,-0.9538667,-0.92621404,1.25343,1.4133275,0.07042062,-0.17381254,-1.4111978,0.36360025,1.3456023,1.6989193,0.42588353,0.9820415,0.12705263,-0.5005522,0.054737598,0.76003623,0.84157014,-1.7930596,1.0273554,0.44291124,0.082684554,0.61543924,-1.5382718,-0.26630726,-0.56344163,0.47257867,0.62037295,1.3090107,-0.49892786,-0.3358481,-0.06688593,-0.927906,0.33645353,1.4333607,-0.5287842,-1.4776187,0.023554774,-1.1187627,1.6363778,0.19606885,-0.41932362,-2.4530096,0.52753305,1.0463974,1.1490204,0.4760272,-1.2210376,0.7896005,-0.003464025,0.8981069,-2.112449,0.16817431,-0.64763904,0.200444,1.2383525,0.66687185,-0.045646884,1.3314408,0.10086164,0.30638647,-2.6051254,0.5737982,1.2559868,-0.6507682,-1.2285036,0.194903,0.7337301,-0.36549726,1.3863009,0.23165633,-0.05751545,-0.071033075,0.66757375,-1.0468653,0.9908146,-1.5330824,-0.010820814,0.87953395,0.21151483,0.9051806,-1.0364068,-0.45686463,0.33169436,0.61652184,2.6955714,0.38302016,-1.3782058,-0.8940241,0.33452165,-0.84705555,-1.7417383,-0.18060239,1.1929514,-0.091192424,1.401481,-1.0531416,-1.0770215,0.70313454,-0.3485591
+9,0,7,-0.16720565,-0.95600957,0.28606528,0.6061069,-0.38458014,-0.20699821,-1.1377771,0.20950352,-0.13692231,-0.39809802,1.1865551,1.4303648,-1.0548229,0.47550696,0.9458503,2.1207294,2.1566584,-0.24435447,-0.64201504,1.7221577,0.931517,-0.7352646,0.34909317,0.7247565,-0.5341214,-0.38585183,-0.7807762,0.55336505,0.0011219969,1.0769404,0.17883597,-1.319804,1.3214653,-1.3146976,-0.71112573,-0.4876438,1.6924084,-0.5158393,1.3672885,-0.55709517,-0.8911706,-0.76211774,1.6342849,-1.2171471,0.21610762,-2.0004432,-0.90466094,-0.66080266,0.099175125,-1.5156087,-1.408444,-1.4207674,-1.6178052,1.158086,0.932027,1.5196209,-1.0533147,1.268581,0.0013256798,-1.9838752,-0.02037545,-0.6469344,-1.380692,-1.059281,0.5494309,-1.3961787,-0.15649295,-0.5929407,0.9686789,-0.40543148,1.0557822,0.8631479,-2.2666907,0.67865163,-0.9861262,-0.4633933,-0.5806146,1.3590292,-0.97610635,-1.6902065,0.015555903,-1.0601696,-0.6836848,-0.48810723,1.0633672,-2.1536193,0.27614263,-0.9428519,0.8829793,-0.13327233,-0.69135016,0.057943318,0.29546723,-1.8476412,1.6050954,-0.77256197,-0.34552777,-0.3999227,-0.91350764,-1.4197913,-1.4342732,0.054446727,0.38957366,1.6343316,-0.018683603,0.20149976,-0.51843685,-2.7988346,-0.23812824,0.3947416,0.7759951,0.26147768,0.7695459,-0.8783734,0.24983662,0.22944754,-0.50273734,0.17921257,-2.96453,0.84934795,0.8776484,0.17855342,-0.3165259,-0.49537578,-1.4034133,-0.5197445,0.65165275,-1.1936586,0.4376411,-0.3197974,-1.1554341,-0.09437472,-1.4360906,-0.666431,0.19690531,0.22096209,-0.77558017,-0.6480959,-1.0999116,1.1237246,0.32265925,-0.71968275,-2.2545595,-1.4951689,0.5562803,1.1043609,1.1391095,0.12772344,0.39196163,0.68036157,-0.48576477,-1.285276,0.81914747,0.30907387,-0.07296551,1.534054,0.30991098,-0.007137307,1.7125996,-0.8783491,-0.079967216,-2.1059916,-0.31304887,-0.31512117,-1.8248315,-1.5771934,0.5107034,0.7182178,-0.25824896,-0.45623496,-0.26445267,-0.9527643,-0.7283463,0.17178623,0.2119341,0.9493435,-1.3720604,0.15322246,0.70129305,-0.55172205,-0.15245205,0.06241885,-0.6320953,0.5529451,-0.4287735,1.0122429,-0.0774352,-0.28508043,-1.175873,0.21220754,-1.3312136,-1.4357239,0.21157016,0.5917681,-1.1742185,1.9612663,-1.2883048,0.06013697,1.073086,0.5143654
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
index 04665ca6..9a165d17 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,5,0.99609375,0.23242188,-0.5625,-0.53515625,1.359375,0.578125,-0.7109375,0.4609375,1.2265625,-0.52734375,0.734375,2.5625,0.026000977,0.6171875,-0.11767578,-0.09326172,-0.80859375,0.23925781,2.96875,1.59375,0.6640625,-1.09375,0.49414062,0.060791016,-0.17675781,1.203125,-0.1171875,-0.61328125,-1.765625,-1.46875,-0.15820312,0.8203125,-0.796875,1.34375,-0.6953125,-0.18359375,0.28125,0.734375,0.33203125,0.16992188,0.74609375,-0.34765625,0.31835938,-0.27734375,-0.08642578,0.15820312,0.080566406,-1.7890625,-0.34375,-0.67578125,0.7578125,-1.1875,0.41992188,0.13671875,-1.59375,0.30859375,0.3046875,-0.44921875,1.1484375,-1.109375,-0.12695312,-0.15039062,1.46875,-0.7734375,-0.6171875,-1.25,0.20898438,-1.78125,0.53515625,0.71875,-0.52734375,1.3515625,-0.73828125,-0.76953125,0.81640625,2.46875,-0.80078125,-0.06591797,-1.9296875,-0.19042969,-0.033935547,-0.19824219,0.45117188,-0.26367188,0.83203125,-0.73046875,-0.35351562,-1.9375,0.12988281,1.2421875,0.59765625,0.42578125,-1.46875,1.65625,-0.24609375,-0.08496094,0.008178711,1.2890625,-0.5703125,0.091796875,0.50390625,0.28125,0.6953125,-1.0546875,-1.0703125,-0.13476562,0.060546875,0.8515625,-0.09277344,-0.11279297,1.2734375,0.6875,0.28320312,-0.09716797,1.875,-0.8984375,-0.90234375,-0.94140625,0.27539062,-0.15917969,-0.421875,0.37695312,0.38671875,0.33398438,-1.40625,2.15625,-0.26367188,-1.1953125,1.7734375,1.921875,0.8671875,-2.078125,-1.140625,0.47851562,-1.9140625,0.171875,-1.421875,-0.24023438,0.35351562,-0.984375,1.3515625,-1.40625,-1.671875,2.078125,-0.47070312,-0.41992188,-0.55859375,1.5703125,0.21777344,0.47265625,0.83203125,-1.375,0.6484375,-0.8671875,-1.578125,0.76953125,1.1640625,0.18359375,-0.05810547,-1.171875,0.20117188,1.4375,0.484375,2.296875,1.046875,0.13085938,-0.9921875,-2.859375,1.09375,0.14160156,0.76171875,-0.63671875,-0.19140625,0.546875,-0.5546875,-0.49609375,-0.92578125,-0.52734375,-0.4375,0.76953125,-0.4765625,-0.51171875,-0.3515625,0.7734375,-1.4765625,1.2109375,-1.0,-1.921875,1.875,-1.0546875,-0.859375,-1.03125,-0.6875,-0.38867188,0.021728516,1.46875,-1.6328125,0.27734375,0.15917969,1.0859375
-0,0,6,1.1015625,0.88671875,1.1875,-0.51171875,0.19824219,-0.25,0.55078125,-1.09375,0.76953125,-0.87890625,0.91796875,-0.039794922,1.1015625,-0.43359375,0.59765625,-0.4453125,-0.24804688,-0.140625,2.328125,-0.010864258,-0.47070312,-1.296875,0.25,-0.12890625,-0.6484375,-0.734375,0.859375,-0.5859375,-0.9921875,1.265625,0.45898438,-0.6875,0.78515625,-1.3671875,0.70703125,-0.0072021484,0.11621094,-0.5703125,-0.625,1.109375,-1.421875,0.032470703,-0.5625,-0.08251953,-0.22167969,-1.296875,0.5703125,-1.109375,1.234375,-0.3671875,-0.78515625,0.58203125,-0.71875,1.34375,0.16210938,-0.043945312,0.75390625,-0.73046875,-1.421875,-0.71484375,1.1953125,0.7109375,-0.84375,1.140625,-0.21679688,0.051757812,0.35742188,-2.1875,-0.6640625,1.40625,-1.0546875,0.39257812,0.4453125,0.025634766,0.17382812,1.4609375,1.0390625,1.2109375,-0.828125,-0.265625,-1.3984375,-0.48242188,0.734375,-0.65625,0.11425781,-1.0546875,-0.6640625,-2.46875,1.546875,-1.03125,0.49804688,0.8203125,-0.73828125,0.5859375,-1.21875,-0.65625,-1.9765625,-0.046142578,-1.421875,0.43164062,-0.59765625,1.1015625,1.6015625,0.029663086,1.140625,2.15625,-0.515625,0.72265625,0.40625,-0.64453125,0.95703125,1.484375,0.76953125,0.33203125,0.83203125,0.46484375,-0.140625,-0.30273438,-0.40039062,1.0234375,0.29492188,-0.74609375,1.4140625,-0.640625,-1.5546875,1.3359375,-1.65625,-1.203125,-0.64453125,0.29882812,1.5625,-0.59375,-0.30273438,-0.984375,0.38671875,0.01586914,0.40625,-0.10205078,0.48828125,-1.7421875,2.421875,0.796875,-1.21875,2.296875,0.040771484,-1.84375,2.09375,-0.58203125,-0.61328125,0.98828125,-1.1640625,-0.28125,1.5390625,-1.3125,-0.8984375,0.71875,1.5703125,-1.3046875,0.58203125,0.43359375,-0.35546875,-0.28320312,2.484375,2.078125,0.060058594,-0.87890625,-1.2578125,0.69140625,0.5,-0.24121094,-0.24414062,-0.22265625,-1.7109375,-1.40625,-0.5859375,0.234375,1.3203125,-0.9921875,0.6015625,0.020507812,-1.8828125,-1.2734375,0.546875,-0.35742188,-1.890625,-0.375,0.26953125,-0.79296875,-0.57421875,-0.1875,-1.53125,1.3046875,0.24316406,-0.58984375,-0.90234375,1.3046875,1.2421875,-0.609375,1.5546875,1.1484375
-0,0,7,1.390625,1.453125,-0.890625,-1.4140625,-0.546875,-2.125,0.17285156,1.1171875,-0.43359375,-1.171875,0.104003906,-1.3359375,-2.234375,-0.20605469,-0.25390625,1.15625,-0.24511719,0.6171875,-0.15625,0.0019226074,0.6640625,-0.234375,-0.23242188,1.046875,-1.7109375,-0.34765625,-2.0,0.5859375,1.34375,0.51953125,0.5,1.875,-2.34375,-0.18847656,1.2578125,-0.25390625,1.1640625,2.046875,0.111328125,0.8125,-1.46875,-0.38476562,0.4453125,2.171875,0.5390625,1.6171875,0.43164062,0.63671875,-0.13183594,0.34375,0.36914062,-0.10253906,-0.8984375,-1.234375,-0.30078125,-0.296875,-0.34375,-0.01574707,-0.14160156,-0.546875,-0.24609375,0.88671875,-0.4375,-0.35742188,1.390625,0.0025939941,0.46484375,-0.1328125,-0.26757812,0.4140625,-0.515625,-0.24414062,0.13769531,-0.7734375,0.53125,0.07421875,2.234375,-0.078125,-0.265625,-0.19824219,0.88671875,0.7734375,1.6796875,0.48632812,0.67578125,0.59375,-0.77734375,0.34765625,-1.984375,0.25195312,0.484375,1.8046875,2.078125,-0.26367188,-0.05493164,-1.46875,-0.0390625,0.26953125,1.2265625,1.5234375,1.484375,-0.14941406,0.8515625,-0.28515625,-0.76171875,-0.91796875,-1.015625,-1.28125,-0.6796875,-0.032470703,-0.15234375,-0.25585938,2.6875,-0.6875,-0.076171875,-0.703125,-0.34179688,-1.5703125,0.7265625,0.48046875,1.90625,-2.046875,-0.85546875,-2.578125,-0.41601562,0.08105469,-0.80078125,1.375,0.04638672,-1.734375,0.97265625,-0.48242188,1.0,-0.45117188,-2.046875,0.2578125,0.56640625,-0.20996094,0.29101562,-2.34375,0.32421875,0.9609375,1.296875,1.109375,-0.6171875,-0.44921875,-1.921875,1.6796875,0.46484375,0.10546875,-0.26367188,-1.8828125,-0.58203125,1.5390625,-0.049316406,-0.071777344,-0.49023438,0.84375,-0.7265625,2.09375,-2.0625,0.94921875,-0.59375,0.0859375,-0.51953125,-0.25195312,-0.21875,-0.032714844,0.04711914,-0.5546875,-0.25976562,0.5078125,-1.1953125,0.9140625,1.265625,-0.15917969,1.1953125,-1.1796875,0.052246094,-0.42578125,1.125,0.5234375,-0.20507812,-1.375,-0.45703125,-0.12792969,-0.4140625,-0.37304688,-0.31445312,-0.953125,-0.890625,-0.2890625,-0.74609375,-0.7890625,1.4140625,0.53515625,1.1796875,-0.3984375,0.0047912598,0.24609375
+0,0,5,0.24902344,-0.003112793,-1.1171875,-0.2890625,1.21875,0.76953125,-0.56640625,1.03125,0.9609375,-0.14257812,0.66796875,2.828125,0.4453125,0.28320312,-0.91015625,-0.048583984,-0.67578125,0.5625,3.265625,1.6875,1.0078125,-0.90234375,0.63671875,0.2578125,0.028686523,0.7890625,-0.5703125,-0.44140625,-0.91796875,-0.875,0.095703125,0.05810547,0.16601562,1.1328125,-1.125,0.0028686523,0.2890625,0.22851562,0.19140625,0.015991211,-0.37695312,-0.42578125,-0.25195312,-1.078125,0.11767578,0.7578125,0.14355469,-1.484375,0.13183594,0.13085938,0.49804688,-1.546875,0.0703125,0.119140625,-1.3515625,0.052490234,-0.33789062,-0.44726562,1.2421875,-1.2578125,-0.06982422,-0.359375,2.21875,-0.40820312,-0.6953125,-1.15625,0.02746582,-1.4140625,-0.25976562,0.8125,0.044921875,1.40625,-0.53125,-0.78125,1.0859375,2.875,-0.91015625,-0.13671875,-2.328125,0.02319336,0.49609375,-0.06640625,0.546875,0.3125,0.044189453,-0.29492188,-1.15625,-1.34375,0.38671875,1.65625,1.421875,0.66796875,-1.625,1.640625,-0.014099121,-0.024658203,-0.32617188,0.546875,-0.76953125,-0.25195312,-0.16210938,0.24316406,1.0078125,-1.2421875,-1.4765625,0.27148438,-0.63671875,0.6953125,-0.45507812,-0.14746094,1.2109375,0.49804688,0.13867188,-0.24902344,1.4296875,-1.53125,-1.0,-0.4140625,0.81640625,0.18847656,-0.21484375,0.39648438,0.4765625,-0.12402344,-1.4765625,2.421875,-0.49804688,-1.265625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40820312,-1.6796875,0.546875,-1.3671875,-0.609375,-0.083496094,-0.7890625,1.4140625,-1.640625,-0.87890625,2.03125,-0.66796875,-0.37695312,-0.14160156,1.0,-0.76953125,0.390625,1.09375,-0.9375,0.54296875,-0.87890625,-1.0390625,-0.0859375,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.375,0.73046875,2.28125,0.064941406,0.07910156,-1.28125,-2.359375,1.7734375,0.033691406,0.37890625,-0.78515625,-0.3359375,0.7421875,-0.95703125,-0.1953125,-0.90234375,-1.109375,-0.69140625,0.6484375,-0.76953125,-0.053222656,0.41210938,1.109375,-1.4453125,0.9453125,-0.9609375,-1.2890625,1.796875,-0.93359375,-1.4296875,-0.5390625,0.43359375,-0.578125,-0.20507812,1.375,-0.421875,0.31054688,0.33398438,1.21875
+0,0,6,0.29882812,0.5,0.65625,-0.48046875,0.25976562,-0.025512695,0.6796875,-0.6171875,0.58984375,-0.6640625,0.68359375,0.095703125,1.6015625,-0.73046875,0.007873535,-0.3125,-0.19628906,-0.005218506,2.625,0.16503906,-0.15234375,-1.3984375,0.6640625,0.057373047,-0.703125,-1.0,0.29296875,-0.3125,-0.37890625,1.609375,0.64453125,-1.0546875,1.5078125,-1.4140625,0.19238281,0.20996094,0.06982422,-1.046875,-0.84375,0.8203125,-2.140625,0.29101562,-0.82421875,-0.6484375,-0.28125,-0.51953125,0.31445312,-0.984375,1.3515625,0.671875,-0.78515625,0.34179688,-0.9296875,1.5390625,0.20117188,-0.30273438,0.140625,-0.421875,-1.1484375,-0.83203125,1.3359375,0.50390625,-0.051757812,1.25,-0.19140625,0.4453125,-0.05908203,-1.796875,-1.265625,1.5390625,-0.6640625,0.578125,0.640625,0.064453125,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.012451172,-0.80078125,-0.66796875,0.5625,-0.18066406,-0.27734375,-0.609375,-1.2421875,-1.953125,1.484375,-0.5,1.1953125,0.953125,-0.9609375,0.328125,-1.109375,-0.74609375,-1.8984375,-0.75,-1.296875,0.04248047,-1.375,1.125,1.796875,-0.017578125,0.44335938,2.609375,-1.03125,0.58984375,0.28515625,-0.625,0.69921875,1.1484375,0.640625,-0.078125,0.46484375,0.30859375,-0.21484375,0.24707031,-0.06347656,1.296875,0.5390625,-0.6171875,1.1875,-0.90625,-1.1171875,1.1171875,-1.8203125,-1.40625,-0.18847656,-0.10107422,1.1171875,-0.609375,-0.54296875,-1.3515625,0.38671875,0.40234375,0.40234375,-0.40820312,0.21582031,-1.8828125,2.484375,0.57421875,-0.671875,2.34375,-0.24902344,-1.5234375,2.21875,-1.0703125,-1.34375,0.82421875,-0.9609375,0.20019531,1.0390625,-1.4140625,-0.32617188,-0.030151367,1.2578125,-0.8203125,1.0859375,0.32421875,-0.390625,-0.53515625,2.390625,1.7734375,-0.94921875,-0.7734375,-1.1484375,1.015625,1.1015625,-0.39453125,-0.30664062,-0.50390625,-1.734375,-0.9140625,-0.88671875,0.265625,0.98046875,-1.4453125,0.5078125,0.2578125,-1.9296875,-1.015625,1.171875,0.26953125,-1.5390625,-0.41992188,0.43359375,-0.20898438,-0.65625,-0.05883789,-1.7109375,1.7890625,1.1875,-0.53125,-1.0859375,0.8984375,1.796875,-0.34570312,1.359375,1.0703125
+0,0,7,0.66796875,1.21875,-1.203125,-1.3125,-0.51171875,-1.8984375,0.31445312,1.1875,-0.53125,-0.6796875,-0.05419922,-1.1796875,-1.828125,-0.33203125,-0.9375,1.0703125,-0.27929688,0.91796875,0.1640625,0.022216797,0.90234375,-0.23339844,-0.03466797,1.328125,-1.7265625,-0.96484375,-2.390625,0.6640625,1.984375,0.9765625,0.76953125,1.3984375,-1.5234375,-0.38671875,0.73828125,-0.020141602,0.9453125,1.3671875,-0.044921875,0.6171875,-2.15625,-0.27734375,0.072265625,1.75,0.7109375,2.03125,0.43359375,0.5859375,0.18847656,1.25,0.28515625,-0.23242188,-1.3515625,-0.80078125,-0.1328125,-0.43164062,-0.9765625,0.012329102,-0.064941406,-0.609375,-0.1640625,0.53515625,0.34765625,-0.3828125,1.3359375,0.23632812,0.1640625,0.06542969,-1.0625,0.59375,0.047851562,-0.16796875,0.390625,-0.828125,0.6796875,0.09082031,2.09375,-0.044677734,-0.42578125,0.03955078,1.2734375,0.609375,1.578125,0.9140625,-0.08691406,0.9765625,-1.328125,0.515625,-1.7421875,0.35351562,0.9609375,2.203125,1.7890625,-0.40820312,-0.068359375,-1.421875,-0.33203125,0.012512207,0.96484375,1.03125,0.87890625,-0.18457031,1.0234375,-0.17285156,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0068969727,-0.37890625,-0.22851562,2.546875,-0.88671875,-0.47851562,-0.890625,-0.35546875,-1.03125,0.99609375,0.6875,2.0625,-1.8828125,-0.86328125,-2.96875,-0.21582031,0.19042969,-0.98046875,1.2421875,0.17089844,-1.9765625,0.90234375,-0.37695312,0.67578125,-1.078125,-1.875,0.58984375,0.53515625,-0.546875,0.111816406,-2.1875,0.32226562,0.89453125,1.703125,1.21875,-0.62109375,-0.49414062,-1.625,1.078125,-0.37695312,0.12890625,-0.041015625,-1.359375,-0.7890625,1.4296875,0.19238281,-0.7109375,-0.6015625,1.0625,-0.37695312,1.90625,-2.0625,0.96875,-0.33007812,0.079589844,-1.296875,-0.1484375,-0.26953125,0.29296875,0.46679688,-0.58203125,-0.32421875,0.47265625,-1.109375,0.9921875,0.953125,0.25,0.984375,-1.546875,0.041015625,-0.103515625,1.0,0.53125,0.62109375,-0.87890625,-0.49804688,-0.32421875,-0.24511719,0.0625,-0.54296875,-1.046875,-1.2421875,0.31640625,-0.036132812,-0.86328125,1.2890625,0.119140625,1.921875,-0.23046875,0.09375,0.42382812
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
index 855bd8e8..f802b0fc 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,5,1.703125,1.15625,-0.33398438,0.65234375,-1.03125,0.07519531,1.015625,1.7890625,0.23632812,-1.8828125,0.053466797,0.5625,0.203125,0.107421875,-0.71875,0.7890625,-0.37304688,0.578125,0.546875,-0.578125,0.45507812,-0.86328125,-0.21875,-0.5390625,0.24707031,-0.18554688,-1.09375,-0.265625,-2.171875,-0.69921875,0.24316406,-0.71484375,0.796875,0.765625,0.66796875,-1.671875,0.546875,-0.765625,-0.55078125,-0.3203125,-0.421875,-1.0234375,-0.89453125,0.75,-2.4375,0.4140625,0.8203125,0.671875,1.203125,0.39257812,-0.95703125,-0.34179688,-0.17578125,1.5078125,-1.0859375,0.66796875,1.7265625,0.42578125,0.071777344,-0.49609375,-0.083984375,0.91796875,-0.41992188,0.15039062,0.359375,-1.0390625,-0.62890625,-0.78125,0.15136719,-0.35351562,0.69140625,-2.09375,0.035888672,-0.76953125,1.8515625,2.703125,1.734375,1.03125,-0.6484375,-0.9375,-0.90234375,-0.984375,-0.74609375,-2.125,1.75,-1.015625,1.8046875,0.09765625,-0.96875,1.7109375,1.6953125,-1.4921875,0.53125,0.921875,-1.2109375,0.42773438,0.48632812,-0.22363281,0.23925781,1.09375,-0.63671875,1.1875,0.90234375,-0.3828125,2.484375,-0.78125,0.47460938,1.1171875,0.9765625,1.4453125,0.58984375,0.515625,0.27929688,0.04711914,1.9296875,-0.033203125,-1.5078125,-0.2578125,-0.5078125,-0.6484375,-0.38867188,-0.9765625,0.8515625,-0.20117188,-1.1328125,0.6484375,0.66015625,-0.44921875,-0.6328125,-0.61328125,0.09326172,-0.18847656,-0.25,-0.37109375,-1.9609375,-1.015625,0.1328125,0.48046875,-0.359375,-1.9296875,2.171875,-0.71484375,-0.640625,-1.7265625,-0.011352539,-0.47070312,-0.890625,-0.46484375,1.2890625,4.03125,-0.69140625,0.15820312,-1.8046875,0.69921875,-0.75390625,-0.61328125,-0.30078125,0.515625,-0.31640625,-0.55859375,0.15625,0.087402344,-0.40820312,0.24316406,-0.41992188,0.9140625,0.14355469,1.03125,1.6875,-1.6796875,-0.21191406,0.107910156,-0.39453125,-0.3671875,0.546875,0.78515625,1.0078125,1.0078125,-0.90234375,-0.54296875,-0.66015625,1.265625,-0.029785156,-0.515625,-0.28320312,-1.203125,1.109375,0.7109375,-0.7890625,-0.30273438,-0.53125,-1.609375,0.30273438,-0.38476562,-1.5,1.859375,-0.35546875,-0.04272461,-0.30273438,-0.111328125
-1,0,6,-0.032958984,1.6171875,-1.140625,-1.5625,0.23144531,-1.4921875,0.41796875,1.671875,1.5625,-0.859375,1.0078125,1.0390625,-1.4609375,0.54296875,-0.5078125,0.70703125,0.020874023,0.046875,-0.18945312,-0.17480469,-0.01928711,-0.37890625,1.4609375,0.07910156,-0.7578125,0.007659912,-1.953125,0.796875,0.59765625,-0.7109375,0.63671875,0.7265625,-1.9140625,0.05029297,1.96875,-1.171875,1.296875,1.734375,-0.33984375,1.6328125,-0.828125,0.625,0.29492188,0.15527344,-0.70703125,2.390625,0.11230469,1.375,1.1015625,0.24609375,-0.59375,1.046875,-0.099609375,-0.40429688,-1.3828125,0.0078125,0.08984375,0.21777344,-0.953125,1.2265625,0.23632812,1.1953125,-0.82421875,0.67578125,0.35351562,-1.421875,0.27539062,0.29101562,-0.85546875,0.7734375,-0.2890625,0.609375,1.6875,-0.57421875,0.20117188,2.078125,0.7578125,0.625,-1.15625,-0.21972656,0.026000977,-0.16992188,0.50390625,-0.78125,0.890625,1.1328125,-0.9921875,-0.734375,-1.3984375,-0.49414062,0.07324219,1.421875,1.234375,0.0016174316,0.43945312,-1.9296875,-0.24121094,0.052490234,0.6796875,-0.07373047,0.9375,0.76953125,0.84375,-0.1484375,-0.58203125,-0.46679688,-0.7890625,0.0003528595,0.008300781,0.49023438,-1.078125,0.29101562,1.625,-1.6796875,1.34375,0.02709961,-1.484375,-0.90625,0.52734375,-0.41015625,1.328125,-2.6875,0.53515625,-1.5546875,-0.072753906,0.921875,-0.33203125,-1.328125,0.73828125,-0.15527344,-1.015625,-1.3125,-0.35546875,1.3828125,-1.9140625,0.94140625,1.671875,0.19921875,1.0,-2.046875,1.3828125,0.37304688,0.38671875,1.0546875,-1.0546875,-1.4765625,-0.94140625,-0.734375,-0.68359375,-0.69921875,0.41015625,-1.734375,-0.25390625,0.71875,-1.421875,2.234375,0.1640625,-0.123046875,0.00065612793,0.515625,-0.30664062,0.8671875,0.3984375,0.78125,0.51171875,-0.49804688,-0.33398438,0.66796875,-0.75390625,-1.71875,-0.9375,-0.15136719,-1.671875,-0.64453125,0.9609375,-0.55859375,1.0625,0.75,-0.57421875,-0.734375,0.3359375,-1.3046875,-0.515625,-0.51953125,-0.032226562,0.21875,-1.140625,-1.21875,-1.0625,-1.1015625,0.31640625,0.23535156,-1.375,0.32421875,-0.66015625,3.34375,1.5625,0.119140625,-0.84765625,-0.39453125
-1,0,7,1.5703125,0.671875,-0.91015625,0.052734375,-1.5078125,0.123046875,0.4375,1.0703125,-1.25,-1.6796875,0.25976562,-0.296875,-0.80078125,0.23828125,-1.171875,-0.7734375,-1.1640625,0.115234375,-0.97265625,-0.46875,0.49023438,-0.42578125,-0.58984375,-0.4296875,-0.02746582,-1.4453125,-1.921875,-0.65234375,-0.81640625,0.055664062,-0.91796875,-1.125,-0.36914062,0.94921875,0.049072266,-0.515625,0.34960938,-1.7578125,0.59765625,0.50390625,1.2109375,0.0015792847,-0.39648438,1.6015625,0.67578125,-0.06225586,-0.82421875,0.3046875,-0.14453125,-0.09082031,0.26171875,-0.33203125,-1.0625,0.40039062,-1.2890625,-0.58203125,0.1640625,1.1875,-1.0078125,-1.3828125,-0.4453125,0.98046875,0.6953125,0.953125,1.6484375,-0.39648438,0.16210938,-0.953125,-1.34375,0.10498047,0.54296875,-0.5078125,1.328125,-0.66796875,0.78515625,0.9140625,2.75,1.3125,0.51171875,0.076660156,-0.98046875,-0.1640625,0.24902344,-1.859375,0.43359375,1.171875,0.47070312,-0.828125,0.099121094,1.1953125,0.114746094,1.1953125,0.58984375,0.5703125,-1.234375,-1.328125,-0.609375,-0.0012664795,1.359375,1.4609375,0.953125,1.3515625,-0.29296875,-1.625,1.2890625,-1.4921875,-0.43945312,1.1015625,0.4921875,-0.40429688,-0.3046875,1.2421875,1.6015625,0.515625,0.72265625,0.46484375,-0.30078125,-0.016967773,-0.13769531,1.609375,0.36914062,-2.265625,-0.81640625,-0.58984375,-0.33203125,-1.3046875,0.21777344,-0.875,-0.044921875,-0.18847656,0.90234375,-0.9296875,-0.018676758,0.58203125,-1.171875,2.265625,-0.47460938,-1.3359375,1.4609375,-2.28125,0.7421875,-0.47265625,-0.49609375,-0.625,1.078125,0.359375,0.28320312,0.77734375,2.71875,3.46875,0.13476562,-0.58203125,-1.4609375,0.8984375,-0.5078125,0.51953125,0.44140625,0.671875,-1.3671875,0.828125,0.8359375,1.53125,0.4140625,-0.40234375,0.080566406,-0.4375,-0.17382812,1.6484375,0.24414062,-1.5703125,-0.53515625,-1.15625,0.09277344,-0.6640625,0.8671875,1.15625,-0.30078125,1.1328125,1.1015625,-0.859375,0.096191406,0.20605469,0.6875,-1.078125,0.20996094,-0.064941406,0.31445312,1.40625,0.5546875,-1.4296875,0.16601562,-2.234375,0.13574219,-2.125,-0.78125,1.84375,-0.037841797,-1.140625,-1.0625,-0.1796875
+1,0,5,1.0546875,0.89453125,-0.546875,0.82421875,-0.65234375,-0.37695312,0.7734375,1.6484375,0.55859375,-1.6171875,-0.53515625,0.546875,0.7578125,-0.09765625,-1.4453125,0.83203125,-0.057617188,0.5859375,0.33789062,-0.37304688,0.83984375,-0.93359375,-0.035888672,-0.9140625,0.16015625,-0.9453125,-1.5078125,-0.14550781,-1.1484375,-0.8828125,1.3203125,-1.328125,1.71875,0.7578125,0.22753906,-1.578125,-0.12597656,-1.7265625,-0.7265625,-1.1328125,-1.2421875,-1.3046875,-1.28125,0.015991211,-1.875,0.92578125,0.48046875,1.015625,1.125,1.546875,-1.3125,-0.12792969,-0.73046875,1.7578125,-0.671875,0.5078125,1.15625,0.66796875,-0.15429688,-0.22753906,0.05419922,0.51171875,0.08251953,-0.12109375,0.7265625,-0.578125,-0.8984375,-0.40625,-0.703125,0.25195312,1.59375,-1.8125,0.57421875,-1.046875,2.296875,2.59375,1.53125,0.82421875,-0.34375,-1.125,-0.38671875,-1.25,-0.42578125,-1.2734375,0.9375,-1.1875,1.28125,0.59765625,-0.30664062,1.4921875,2.234375,-1.34375,0.16992188,0.734375,-0.4609375,0.5703125,0.18554688,-0.6796875,-0.21484375,0.58203125,-1.5703125,1.15625,1.03125,-0.029785156,2.0,-0.55859375,0.40234375,0.57421875,0.8828125,1.5859375,-0.31835938,0.014892578,0.296875,-0.609375,1.0,-0.29101562,-1.6875,0.17382812,-0.15136719,-0.28515625,-0.66796875,-0.41015625,0.22363281,-0.024902344,-0.578125,0.58984375,0.57421875,-0.58984375,-0.5,-0.515625,-0.37890625,0.2109375,-0.25585938,-1.28125,-1.75,-0.46679688,-0.609375,0.37890625,-0.4921875,-2.34375,2.171875,-0.66796875,-0.51953125,-1.9375,-0.30664062,-0.19238281,-0.7265625,-1.09375,0.609375,3.640625,-0.44726562,1.0546875,-1.796875,0.828125,0.19726562,-0.984375,0.14550781,1.203125,0.29492188,-0.8515625,0.2578125,0.36914062,-0.029174805,0.12451172,-0.8359375,0.53125,0.1484375,0.83203125,1.96875,-2.34375,-0.33398438,0.29882812,-0.4140625,-0.25195312,0.27734375,1.15625,1.0078125,1.203125,-0.42773438,0.3828125,-0.828125,1.4375,0.87109375,0.13476562,0.32226562,-1.203125,1.4375,0.91796875,-1.3203125,-0.19238281,-0.40429688,-0.828125,0.8125,0.29296875,-1.6796875,1.109375,0.052001953,-0.037109375,-0.35351562,-0.3828125
+1,0,6,-0.5546875,1.2734375,-1.359375,-1.5078125,0.31054688,-1.9609375,0.34179688,1.6328125,1.59375,-0.58203125,0.21679688,1.03125,-0.953125,0.515625,-1.2109375,0.53515625,0.26953125,0.22167969,-0.37304688,-0.07128906,0.33984375,-0.390625,1.328125,-0.036621094,-0.73828125,-0.9296875,-2.421875,1.1015625,1.578125,-0.47851562,1.375,0.20800781,-1.125,0.1484375,1.5703125,-1.09375,0.6640625,0.87109375,-0.77734375,1.0546875,-1.5546875,0.43164062,-0.21777344,-0.265625,-0.3203125,2.609375,-0.060058594,1.8125,1.15625,1.6015625,-1.0234375,1.078125,-0.63671875,-0.09033203,-0.828125,-0.088378906,-0.51953125,0.484375,-1.109375,1.3984375,0.39453125,0.9453125,-0.15917969,0.609375,0.546875,-1.1953125,-0.12451172,0.71875,-1.6484375,1.2578125,0.53125,0.78125,2.0625,-0.5390625,0.5625,2.078125,0.5078125,0.703125,-0.8359375,-0.40429688,0.48242188,-0.36328125,0.765625,-0.13183594,-0.12011719,0.9296875,-1.421875,-0.26757812,-1.15625,-0.68359375,0.58203125,1.5390625,0.99609375,-0.25976562,0.78515625,-1.875,-0.5390625,-0.25976562,0.41015625,-0.5625,0.14648438,0.4921875,0.6953125,0.15722656,-0.765625,0.06347656,-1.046875,-0.40039062,0.044677734,0.5703125,-1.6484375,-0.067871094,1.6640625,-2.3125,0.64453125,-0.14941406,-1.578125,-0.31054688,0.921875,0.083496094,1.1796875,-2.359375,0.028686523,-1.3671875,0.37890625,0.97265625,-0.25976562,-1.359375,1.03125,-0.24902344,-1.3203125,-0.91015625,-0.5390625,0.5859375,-1.90625,1.5390625,1.1953125,-0.19335938,0.9140625,-2.03125,1.3046875,0.12695312,0.6484375,0.83203125,-1.234375,-1.3359375,-0.81640625,-1.3046875,-1.4609375,-0.6953125,0.60546875,-0.61328125,-0.48242188,0.87109375,-0.73828125,1.8203125,0.5625,0.39453125,0.4609375,0.29492188,-0.26757812,1.1171875,0.7265625,0.36523438,0.02722168,-0.7265625,-0.38671875,0.76171875,-0.49414062,-2.0,-1.03125,-0.060791016,-1.5859375,-0.6953125,0.8515625,-0.12402344,1.0703125,0.84375,-0.31445312,0.020874023,0.08251953,-1.1171875,0.30664062,0.16894531,0.3359375,0.06591797,-0.83203125,-0.84765625,-1.7734375,-0.99609375,0.29101562,0.8203125,-0.7265625,0.6796875,-0.81640625,2.625,2.03125,0.100097656,-0.8984375,-0.48632812
+1,0,7,1.2578125,0.5703125,-0.98828125,0.10546875,-1.359375,-0.1171875,0.61328125,1.0078125,-1.1640625,-1.4609375,-0.21484375,-0.38085938,-0.4296875,0.453125,-1.75,-0.8203125,-1.0625,0.2890625,-0.953125,-0.53125,0.59375,-0.22070312,-0.578125,-0.46289062,-0.10107422,-2.1875,-2.390625,-0.43554688,-0.014282227,0.19238281,-0.3828125,-1.4453125,0.28710938,0.87890625,-0.32617188,-0.546875,-0.09814453,-2.375,0.44921875,-0.083496094,0.41210938,-0.21386719,-0.86328125,1.2421875,0.94140625,0.34960938,-1.0234375,0.6015625,-0.019897461,1.171875,0.16308594,-0.20117188,-1.65625,0.70703125,-0.90625,-0.671875,-0.640625,1.265625,-1.0546875,-1.0234375,-0.28125,0.765625,1.2734375,0.7734375,1.921875,-0.3515625,-0.28125,-0.5078125,-1.9375,0.49804688,1.2109375,-0.36523438,1.65625,-0.69140625,0.890625,0.82421875,2.515625,1.2265625,0.60546875,0.10253906,-0.4140625,-0.38476562,0.375,-1.3359375,-0.28125,1.0,0.32421875,-0.60546875,0.23828125,0.984375,0.5390625,1.34375,0.22363281,0.578125,-0.9609375,-1.2890625,-0.7421875,-0.13378906,1.1484375,1.0859375,0.390625,1.25,-0.30078125,-1.2421875,1.2734375,-1.078125,-0.7109375,0.69140625,0.41210938,-0.33398438,-0.9609375,1.0078125,1.578125,0.26953125,0.118652344,0.47070312,-0.30859375,0.515625,0.12451172,1.921875,0.30664062,-2.09375,-1.1484375,-0.6015625,-0.055664062,-1.25,0.17285156,-1.046875,0.0017166138,-0.16894531,0.421875,-0.5703125,-0.19140625,-0.022949219,-1.046875,2.453125,-0.91015625,-1.703125,1.4140625,-2.3125,0.64453125,-0.671875,-0.28320312,-0.65234375,0.875,0.31054688,0.375,0.20019531,2.1875,3.578125,0.41992188,0.265625,-1.6171875,1.0234375,-0.008728027,0.29492188,0.66015625,1.140625,-1.359375,0.62890625,0.9765625,1.734375,0.8046875,-0.62109375,-0.5078125,-0.546875,-0.34765625,1.6796875,0.51953125,-1.7578125,-0.61328125,-0.93359375,0.23242188,-0.55078125,0.703125,1.390625,-0.3125,1.1171875,1.328125,-0.14257812,0.067871094,0.123535156,1.359375,-0.640625,0.48828125,-0.2890625,0.64453125,1.4140625,-0.0038909912,-1.5234375,0.21582031,-1.6015625,0.54296875,-1.7265625,-0.80859375,1.2109375,0.30664062,-1.0859375,-1.2421875,-0.35742188
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
index a4049a56..df728e04 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,5,1.7734375,0.96875,-0.4609375,-0.78125,1.203125,-0.75,-0.22753906,1.953125,-0.38671875,-1.609375,0.796875,0.68359375,-1.3359375,-0.5234375,-0.109375,2.40625,-0.0047912598,-1.6796875,-0.3046875,0.029785156,-1.03125,-0.9765625,-0.04248047,1.375,0.21289062,1.390625,-2.5,-0.6953125,-1.6796875,-0.48828125,-0.22753906,1.8203125,-0.025268555,1.0703125,0.97265625,-1.3515625,1.40625,0.859375,0.80859375,1.3515625,1.1796875,0.34179688,1.1640625,-0.32226562,-0.65234375,0.056396484,-0.5546875,0.24707031,-0.20703125,0.06982422,0.10498047,-1.9375,-0.5859375,-0.42773438,-0.95703125,0.44921875,1.625,-0.022583008,0.6796875,-0.51953125,-0.21679688,1.4296875,-0.9375,0.83203125,-0.13964844,-0.95703125,0.58984375,1.0546875,0.77734375,-0.73828125,-1.0390625,-0.16503906,0.5078125,-0.53125,1.875,1.7578125,1.2890625,0.7109375,-0.36914062,-0.28125,0.55859375,0.44726562,0.16210938,-1.296875,1.0859375,-1.53125,1.1015625,1.015625,0.24414062,1.4765625,0.40234375,-0.69140625,-0.84375,1.046875,-0.42773438,-0.13574219,-1.6484375,-0.06640625,-0.83984375,-0.076171875,0.012207031,-0.3671875,0.6640625,0.22363281,0.083984375,-0.60546875,-1.546875,-0.053466797,-0.30664062,0.022338867,-0.203125,-1.546875,1.8515625,-1.1640625,1.4609375,0.75,-2.125,-0.013977051,-0.484375,-1.578125,0.57421875,-1.0234375,-0.52734375,-0.875,-0.4140625,2.46875,-0.95703125,0.13671875,0.1875,-1.3984375,-0.6953125,-1.59375,0.29882812,0.17285156,-1.3203125,-0.26367188,1.3984375,0.890625,0.27539062,-1.234375,1.5625,-1.734375,-0.90234375,0.7734375,-0.65234375,-0.072265625,-0.64453125,1.125,0.7734375,0.53125,-0.84375,-2.234375,0.98046875,-0.59765625,-0.42773438,0.26171875,0.30859375,-0.89453125,0.62109375,1.1328125,-0.76171875,1.515625,1.03125,1.5546875,-0.092285156,1.4765625,0.53125,0.90625,0.2890625,0.5,-0.69140625,-0.19824219,-0.041503906,0.890625,-0.38671875,0.39453125,0.44726562,-0.46289062,0.083984375,-1.09375,0.83984375,-0.265625,-1.015625,-1.9375,-0.45703125,0.36132812,-1.890625,-0.81640625,0.31054688,-2.046875,0.39648438,1.0234375,-1.4765625,-0.38671875,0.071777344,2.359375,-0.5703125,0.671875,-0.88671875,0.515625
-2,0,6,0.92578125,0.53125,-0.47851562,-0.29296875,0.23242188,-0.30859375,0.5390625,0.94921875,-0.23144531,-0.5078125,-0.17089844,-1.015625,0.13867188,0.93359375,-1.0546875,0.89453125,-0.515625,-2.171875,0.359375,-1.046875,0.8359375,-0.67578125,-0.52734375,-0.27539062,-1.203125,-1.0859375,-1.5546875,-0.37890625,-3.03125,0.26757812,-1.09375,-0.5390625,0.010498047,0.46875,1.2734375,1.3984375,1.078125,-0.65625,-1.84375,2.125,-0.41210938,0.68359375,0.0035552979,0.734375,-0.08642578,0.31835938,0.74609375,-0.76953125,0.74609375,0.013366699,-1.71875,-1.03125,-0.86328125,0.875,-2.921875,0.82421875,0.81640625,-1.09375,-0.03466797,-2.0625,1.15625,-0.29882812,-0.4609375,0.13964844,0.014038086,-0.9296875,-0.61328125,-0.53515625,0.38867188,1.3671875,0.5390625,-1.96875,-1.1796875,-0.87109375,0.020263672,2.078125,1.484375,1.3125,-0.84375,-0.8125,1.0234375,0.41601562,0.09814453,-1.484375,0.21191406,-0.13671875,0.5546875,0.625,-0.98828125,0.6796875,1.1484375,-0.19433594,0.515625,0.3828125,-1.859375,-0.52734375,-0.122558594,0.20019531,1.5234375,-0.19921875,-1.3984375,0.104003906,1.6796875,0.98046875,0.15136719,-0.9921875,2.109375,-0.31054688,0.4296875,0.71484375,0.8046875,0.81640625,1.125,-0.66796875,1.15625,0.020874023,0.2578125,-1.09375,-0.640625,-1.5,0.77734375,-1.0625,0.8671875,-1.25,-0.71484375,1.375,-0.95703125,-2.3125,-0.73046875,0.40820312,1.2109375,-0.31640625,0.47265625,-0.20996094,-1.5546875,-0.5546875,-0.24121094,-0.18457031,0.49414062,-1.6328125,3.125,-0.421875,-0.22460938,0.10107422,-0.106933594,1.796875,-1.671875,0.86328125,-0.91796875,2.140625,0.053710938,-0.34765625,1.078125,-0.9765625,-0.73828125,-0.14257812,0.10839844,0.62890625,0.96484375,1.09375,-0.36523438,1.4453125,1.4296875,1.1484375,-0.48046875,0.83203125,-1.015625,-0.734375,1.78125,-0.048583984,0.99609375,-0.09814453,-0.5703125,-0.609375,1.1796875,0.6015625,-1.0,-0.375,1.0625,-0.93359375,0.00065612793,-0.8828125,0.109375,-0.28125,-0.13085938,0.23925781,1.078125,0.2578125,0.2578125,-1.6875,-0.20703125,-0.40429688,0.8515625,-0.35351562,1.4921875,0.71484375,0.53125,-0.33398438,1.515625,-0.057128906
-2,0,7,1.2734375,0.94140625,-0.05908203,0.140625,-0.060302734,-0.80078125,0.24023438,1.0390625,0.39648438,-0.9609375,0.75390625,1.3359375,-0.6328125,1.140625,-1.40625,1.1015625,0.48632812,-1.375,-0.375,-0.15234375,1.2265625,-0.71875,-0.68359375,0.59765625,0.03466797,-0.06298828,-0.3125,-0.041748047,-1.671875,-0.47460938,-1.1953125,0.118652344,1.53125,-0.3359375,2.125,1.171875,1.3984375,0.625,-2.0,2.359375,-0.875,0.69140625,-0.39257812,0.5703125,0.26367188,0.15429688,-0.17089844,-0.4296875,0.047607422,0.76171875,-1.0703125,0.15625,-0.30078125,-0.0859375,-1.28125,0.9609375,1.015625,0.41015625,0.03540039,-1.0859375,1.71875,-0.24121094,1.3203125,-0.1796875,-2.28125,-0.06689453,-0.6015625,0.16796875,0.28320312,0.93359375,-0.7421875,-1.2578125,-0.6953125,0.3203125,-0.21484375,2.046875,1.7265625,0.40429688,-1.3125,-1.609375,1.234375,-0.43359375,0.65625,-2.203125,-0.89453125,-0.73828125,0.44921875,1.2890625,-1.4609375,0.29492188,1.359375,-0.18261719,-0.6640625,0.515625,-2.03125,0.8203125,0.16308594,1.5078125,0.7109375,1.4140625,-1.765625,-0.05859375,1.4140625,0.8515625,-0.86328125,-0.014404297,1.1875,-1.09375,-1.4609375,0.13476562,0.953125,-0.053955078,2.1875,-0.875,0.27734375,-0.9140625,1.0234375,-0.7890625,-0.31835938,-2.078125,1.8984375,0.23339844,-0.35742188,-1.046875,0.484375,0.56640625,-1.0703125,-2.078125,-0.4765625,-0.11376953,0.49609375,-1.5390625,-0.24707031,0.044433594,-1.90625,1.4921875,0.3046875,-0.48046875,-0.25976562,-0.875,1.765625,-1.6953125,-1.3359375,-0.15332031,-0.25976562,1.25,-1.0703125,1.34375,-1.8125,-0.1875,0.60546875,-0.01953125,0.84375,-1.5,-0.9296875,-0.19042969,0.61328125,-0.44140625,0.6953125,0.8359375,-0.25976562,0.32226562,-0.39257812,1.375,-0.921875,1.5703125,-1.296875,-0.67578125,1.0390625,0.5390625,-1.1484375,-0.6484375,-0.62109375,-0.2890625,0.96484375,0.421875,0.94140625,-0.0024261475,0.265625,-0.96484375,-0.41601562,-0.45117188,-0.18652344,-1.3671875,0.54296875,-1.2890625,0.14648438,0.22949219,0.8515625,-1.4609375,1.1484375,0.6953125,-0.37304688,-0.87890625,-0.41796875,0.12402344,1.265625,1.7421875,1.4921875,-0.45117188
+2,0,5,1.2109375,-0.096191406,-1.3359375,-0.6015625,0.9609375,-1.484375,-0.8203125,3.140625,-0.5234375,-1.6640625,0.0234375,1.5859375,-0.53515625,-0.75,-0.5703125,1.75,-0.27148438,-1.40625,0.059570312,0.48632812,-1.375,-0.6796875,-0.44921875,1.2421875,-0.095703125,0.33007812,-2.484375,-0.15527344,-0.890625,-0.06933594,-0.020141602,0.7890625,0.6953125,0.7734375,-0.07373047,-1.0546875,1.25,-0.171875,0.515625,0.6953125,-0.06201172,-0.008300781,0.11279297,-0.78125,-0.115234375,0.65625,-0.46484375,0.96484375,-0.43945312,2.21875,-0.609375,-2.234375,-1.3359375,-0.048339844,-0.15039062,-0.33398438,0.921875,0.29882812,0.40234375,-1.046875,-0.3671875,1.4765625,-0.12988281,0.53515625,-0.26953125,-0.030273438,-0.00074005127,1.4765625,0.16796875,-0.11328125,-0.4140625,-0.2890625,0.9140625,-0.67578125,1.6328125,2.140625,0.62890625,0.26953125,-0.24414062,-0.75390625,1.671875,0.7578125,0.53515625,-0.6328125,-0.44140625,-1.1484375,1.328125,1.84375,0.50390625,1.65625,1.734375,-0.7734375,-1.046875,1.375,-0.28320312,0.033691406,-1.7265625,-0.73828125,-1.0859375,-0.21875,-1.0703125,-0.24804688,1.4296875,0.94140625,-0.46875,0.25390625,-1.640625,-0.62109375,-0.28515625,0.016479492,-1.140625,-1.9453125,1.421875,-2.265625,0.84765625,0.27539062,-2.0625,1.2734375,-0.30273438,-1.21875,1.2578125,-0.47265625,-0.92578125,-0.36523438,0.4140625,2.71875,-0.43945312,-0.41210938,0.4453125,-1.8671875,-1.484375,-1.640625,0.3671875,-1.1015625,-0.8984375,-0.012451172,1.3203125,0.34960938,0.43359375,-1.3515625,1.453125,-1.8203125,-0.6328125,-0.15917969,0.16210938,0.47070312,-0.32226562,0.83984375,-0.73828125,0.828125,-0.6171875,-0.019165039,0.80859375,-0.453125,0.046142578,-0.76171875,0.31835938,0.014770508,1.6640625,1.015625,-1.1484375,1.859375,1.2890625,0.8203125,-0.57421875,0.77734375,0.17285156,1.3203125,1.2578125,0.7734375,-1.2890625,-0.16796875,0.06640625,1.09375,-1.28125,0.43359375,0.35351562,-0.6953125,-0.056884766,-1.171875,0.46679688,0.38476562,0.3046875,-1.0859375,0.22265625,-0.24023438,-1.40625,0.0134887695,-0.84375,-1.34375,0.21386719,1.4375,-0.11621094,-0.42773438,0.123046875,1.1953125,0.49804688,0.6015625,-1.0703125,0.45703125
+2,0,6,0.31445312,-0.33398438,-0.72265625,-0.44140625,0.40625,-0.546875,-0.005645752,1.578125,-0.13769531,-0.8359375,-0.5390625,-0.37695312,1.2265625,0.5,-1.6484375,0.4453125,-0.51953125,-1.984375,0.42773438,-0.6953125,0.5625,-0.328125,-0.27734375,-0.06689453,-1.328125,-1.8125,-1.625,-0.088378906,-2.0625,0.48828125,-0.68359375,-1.1171875,0.55859375,0.076660156,0.546875,1.671875,0.625,-1.453125,-1.8671875,1.1875,-1.4765625,0.34960938,-0.54296875,-0.04272461,0.23144531,1.0625,0.5546875,-0.55078125,0.2421875,1.734375,-2.109375,-1.3203125,-1.4140625,0.7265625,-1.96875,0.6171875,0.12792969,-0.671875,-0.19238281,-2.171875,1.1875,-0.16015625,0.42773438,-0.07324219,0.17675781,0.05517578,-1.0859375,0.18652344,-0.67578125,2.0,0.90625,-1.609375,-0.45898438,-0.5,0.045410156,1.921875,1.109375,1.1015625,-0.61328125,-0.8828125,1.5546875,0.32226562,0.27539062,-0.5078125,-0.671875,-0.028320312,0.28125,0.9609375,-0.6640625,0.6796875,2.109375,-0.17285156,0.1484375,0.41992188,-1.3984375,-0.19824219,-0.3046875,-0.53125,1.421875,-0.33007812,-2.25,0.140625,2.21875,1.15625,-0.46875,-0.22851562,1.4765625,-0.9296875,0.40429688,0.7109375,0.27539062,0.35742188,0.859375,-1.4375,0.30859375,-0.47070312,0.18945312,-0.26953125,-0.66015625,-0.65625,1.3359375,-0.55078125,0.4140625,-1.1875,-0.24414062,1.2734375,-0.44921875,-2.3125,-0.006286621,-0.04321289,0.6796875,-0.061035156,0.19140625,-1.3046875,-1.328125,-0.20019531,-0.35351562,-0.46875,-0.01928711,-1.8046875,3.09375,-0.765625,0.390625,-0.052246094,-0.067871094,1.8359375,-1.21875,-0.016967773,-1.9296875,1.921875,0.23535156,1.140625,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.43359375,1.625,1.4453125,0.66796875,-0.72265625,1.34375,1.3671875,0.49609375,-1.3359375,0.53515625,-1.1171875,-0.33789062,2.375,-0.25390625,0.22949219,-0.1484375,-0.66796875,-0.31445312,0.12011719,0.453125,-0.94140625,-0.91796875,0.9296875,-0.5859375,-0.17871094,-0.107421875,1.2734375,0.49023438,0.66015625,-0.30078125,1.3203125,0.8203125,-0.5390625,-1.046875,-0.23730469,0.53515625,2.15625,0.02746582,0.828125,-0.15429688,1.4140625,-0.2109375,1.2265625,0.26953125
+2,0,7,0.7734375,0.110839844,-0.36914062,-0.22363281,0.2109375,-0.76171875,-0.26953125,1.4140625,0.5546875,-1.09375,0.5859375,1.421875,0.44335938,0.55078125,-1.59375,0.796875,0.18847656,-1.203125,-0.11621094,0.1171875,0.9921875,-0.53515625,-0.3984375,0.57421875,-0.29492188,-0.7265625,-0.51171875,0.061035156,-1.0078125,-0.061035156,-0.68359375,-0.51171875,1.796875,-0.453125,1.28125,1.59375,0.828125,-0.15820312,-2.21875,1.7265625,-1.8359375,0.37695312,-0.59765625,-0.049316406,0.62890625,0.81640625,-0.3359375,-0.1328125,-0.4453125,1.90625,-1.3671875,-0.10986328,-0.6875,-0.08984375,-0.796875,0.71484375,0.6171875,0.60546875,-0.079589844,-1.25,1.609375,-0.21484375,1.8671875,-0.34375,-1.7421875,0.8515625,-1.171875,0.53125,-0.52734375,1.265625,-0.10205078,-1.171875,-0.099609375,0.609375,-0.15625,1.9375,1.390625,0.38867188,-0.99609375,-1.46875,1.390625,-0.20019531,0.59765625,-1.2890625,-1.375,-0.33789062,0.19042969,1.40625,-0.8125,0.359375,1.9609375,-0.03515625,-0.8203125,0.62109375,-1.5390625,0.94140625,-0.17578125,0.88671875,0.73828125,0.9921875,-2.46875,0.011230469,2.0,1.03125,-1.390625,0.35351562,0.90625,-1.6640625,-1.0703125,0.18066406,0.578125,-0.36328125,1.546875,-1.5078125,-0.36132812,-1.21875,0.953125,-0.04736328,-0.3046875,-1.296875,2.125,0.44140625,-0.6015625,-1.2578125,0.60546875,0.41992188,-0.73046875,-2.171875,0.26367188,-0.36523438,0.1796875,-1.171875,-0.54296875,-0.79296875,-1.8515625,1.6640625,0.30664062,-0.625,-0.5390625,-1.3515625,1.8125,-1.6953125,-0.6875,-0.3828125,-0.17773438,1.3125,-0.5859375,0.5234375,-2.6875,-0.1171875,0.2578125,1.1328125,0.69921875,-1.03125,-0.23046875,-0.77734375,0.8984375,0.4765625,1.1171875,0.55078125,-0.41015625,0.36523438,-0.26171875,0.91796875,-1.5625,1.2734375,-1.1015625,-0.578125,1.3984375,0.091308594,-1.6796875,-0.7890625,-0.640625,-0.34960938,0.16503906,0.36523438,0.80078125,-0.46875,0.30859375,-0.8359375,-0.49804688,-0.031982422,0.640625,-0.49023438,1.234375,-1.5625,0.46875,0.61328125,0.24804688,-0.86328125,1.09375,1.546875,0.859375,-0.74609375,-0.7109375,-0.66796875,2.21875,1.6484375,1.2109375,-0.24121094
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
index f64c514e..f0aaa556 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,5,1.5078125,0.8671875,-0.5859375,0.064941406,0.94140625,-0.8125,0.20703125,2.078125,-0.27734375,-0.875,0.91015625,1.6640625,-0.78515625,-0.12988281,-0.36914062,1.921875,0.6875,-1.421875,0.34179688,0.89453125,-0.060791016,-1.1328125,-0.36328125,0.92578125,0.8359375,1.8515625,-2.640625,-0.26953125,-1.1328125,-1.2890625,-0.10253906,1.203125,0.67578125,1.40625,1.828125,-0.8359375,0.7734375,0.37304688,-0.0008163452,0.87109375,0.890625,0.95703125,0.5546875,-0.6015625,-0.625,0.5078125,-1.5234375,-0.29882812,0.119628906,-0.55078125,0.087402344,-1.5625,-0.5703125,-1.40625,-0.5625,0.49609375,1.4296875,-0.033691406,0.051757812,-0.6015625,0.546875,0.640625,0.390625,0.5546875,-1.1875,-1.25,0.85546875,-0.20507812,0.15917969,-0.9453125,-0.6171875,-0.38867188,0.22949219,-0.55859375,2.25,2.390625,1.5859375,0.4375,-0.74609375,-0.39648438,0.53515625,0.515625,0.24902344,-2.125,1.21875,-0.66015625,0.21777344,0.65234375,0.099121094,2.078125,0.86328125,-0.875,-0.31835938,0.8359375,-1.1015625,-0.07080078,-1.3203125,0.71875,-0.1640625,1.2109375,0.51953125,0.103027344,-0.38867188,0.62109375,-0.5859375,-0.78125,-0.91015625,0.3046875,-1.3671875,-0.013793945,0.011352539,-1.0703125,2.109375,-0.97265625,1.375,0.60546875,-1.328125,-0.40429688,-0.44726562,-0.71875,0.6875,-0.40429688,-0.55859375,-0.7265625,-0.78125,2.546875,-1.0,-0.041748047,0.91015625,-1.3125,-0.38671875,-2.96875,0.04345703,0.296875,-1.5390625,-0.099121094,1.28125,0.19238281,0.12890625,-0.82421875,1.140625,-0.94921875,-1.1953125,0.546875,-0.55078125,0.34179688,-1.0,1.1875,0.75,-0.08935547,0.05078125,-2.125,0.18359375,-1.375,-0.60546875,0.84375,0.75,-0.609375,0.44335938,0.98046875,-0.20800781,2.03125,0.1640625,1.2265625,-0.0625,1.15625,0.4375,0.36132812,0.27539062,1.015625,-0.59375,-0.37695312,-1.078125,0.79296875,-0.59375,0.53515625,0.3984375,-0.5234375,0.32617188,-0.27148438,-0.13183594,-0.55859375,-0.57421875,-2.328125,-0.32617188,0.27539062,-1.3671875,-1.234375,-0.0390625,-1.8203125,0.65625,0.24609375,-1.65625,-0.90625,0.064941406,2.84375,-1.171875,0.5859375,-1.140625,-0.11230469
-3,0,6,0.0043640137,1.171875,-0.86328125,-0.65234375,-0.55078125,0.42773438,0.005432129,1.484375,0.734375,-0.37695312,2.28125,0.69921875,0.921875,0.040283203,0.37304688,1.125,0.16015625,-1.109375,0.5078125,-0.5234375,0.7421875,-0.23046875,0.7109375,0.90625,0.19238281,0.6796875,0.3515625,-0.44921875,-1.84375,-2.296875,-0.6953125,2.15625,0.02709961,2.4375,1.78125,-0.52734375,0.6484375,0.9375,0.91796875,0.765625,0.828125,-0.62109375,0.6015625,-1.2734375,-0.017578125,0.45507812,0.16210938,-0.0625,0.44726562,-1.7734375,-1.5859375,1.2109375,-0.6640625,0.328125,-1.4453125,-0.68359375,0.18847656,-1.015625,0.65234375,0.5859375,-0.09472656,1.109375,0.028198242,1.546875,1.421875,-0.71875,-0.640625,0.15039062,-0.17578125,0.359375,-1.0078125,0.14257812,0.20214844,-0.11816406,-0.33203125,2.109375,-0.42382812,0.31054688,-0.21289062,0.024169922,-1.6015625,-0.099121094,0.91015625,-1.3046875,1.8515625,0.030395508,-0.578125,-0.89453125,0.67578125,-0.49609375,0.099121094,-0.22460938,-0.5234375,0.515625,-0.26757812,-1.0546875,0.53125,-0.3359375,-1.2890625,-0.66015625,-0.47851562,0.6328125,2.671875,-1.015625,-0.54296875,-1.6875,-0.5078125,0.8203125,-0.32226562,0.8671875,0.23144531,0.26171875,1.375,-0.33984375,2.390625,-0.044189453,-0.67578125,-1.3203125,-0.51171875,-1.015625,-0.49023438,-0.98828125,0.81640625,-0.79296875,-1.7109375,1.421875,-1.28125,-1.484375,0.98828125,0.83984375,-0.22363281,-0.70703125,0.5234375,-0.10986328,-1.9609375,1.21875,0.76171875,0.5625,-0.47070312,-1.609375,2.234375,-1.046875,-0.96875,0.63671875,-0.5,-0.39257812,-0.32617188,-1.2734375,-0.265625,0.016723633,0.34765625,0.028930664,1.609375,-2.4375,-0.703125,0.4296875,0.21582031,-1.3984375,1.0,-0.41015625,0.70703125,-0.44921875,0.65234375,-0.37109375,1.015625,2.53125,-1.0390625,0.1796875,1.25,-0.22167969,-0.8984375,-0.8046875,-2.03125,-1.875,-0.026367188,0.609375,0.09277344,-0.08203125,-1.4921875,-1.5078125,0.41992188,-0.8671875,-0.19726562,-0.54296875,-0.21875,-0.66015625,-0.6484375,-0.42773438,-0.3828125,0.671875,1.671875,-0.94921875,-0.36328125,-0.68359375,-0.96484375,1.6015625,0.95703125,1.859375,0.7109375,1.515625
-3,0,7,1.09375,1.171875,-0.28320312,0.38476562,-1.5703125,-1.2421875,0.7890625,0.83203125,-0.3515625,-0.40039062,1.984375,0.8515625,0.39257812,0.55078125,0.7421875,0.9296875,-0.90234375,-0.44335938,0.81640625,-0.69140625,1.0625,-0.34570312,2.046875,1.96875,0.110839844,0.91796875,-0.15820312,-1.2421875,-0.82421875,-0.45703125,-0.06542969,0.84375,0.056152344,1.8125,0.57421875,-0.8125,-0.04638672,1.203125,0.80859375,-1.234375,-0.087402344,-1.8046875,1.28125,0.040039062,0.38085938,-0.3671875,0.33398438,-0.58203125,-2.578125,-1.2734375,0.36914062,0.55078125,-1.265625,0.90234375,1.1640625,0.016357422,-0.1796875,-0.18164062,-0.50390625,0.5078125,0.47070312,0.45898438,-2.203125,-1.40625,1.15625,0.59765625,-0.578125,-0.30859375,1.5546875,-0.064941406,-1.890625,0.8828125,-1.0234375,0.32226562,1.1484375,0.49414062,1.5390625,1.0234375,-0.6015625,0.38671875,-1.4453125,0.47265625,1.671875,-0.34570312,1.7578125,-0.9375,-0.16894531,1.1796875,-0.32226562,0.86328125,-0.012878418,-1.046875,-0.90625,0.43945312,-0.59765625,-0.73828125,1.5,0.33203125,-0.26171875,0.037841797,0.0011062622,-1.421875,1.9140625,-0.33789062,-1.4609375,-2.0625,1.140625,-1.328125,0.83984375,0.546875,0.55078125,-1.03125,1.3359375,-0.43554688,0.78515625,-0.32617188,-0.22753906,-0.34375,-0.10498047,-1.6796875,0.7890625,0.43359375,-0.03491211,-2.375,-0.18359375,1.859375,-1.0078125,0.18066406,1.2578125,-1.1328125,1.03125,1.2109375,1.546875,-1.8125,-2.453125,-0.5078125,-1.15625,0.5546875,0.03881836,-0.9921875,0.47070312,0.21875,-0.31054688,0.96875,-0.62109375,0.3125,-1.5859375,0.13574219,0.071777344,0.42773438,-0.06933594,-1.0234375,-0.60546875,-0.07373047,-0.9765625,-1.671875,-0.8671875,-0.73046875,-0.27929688,1.6484375,-0.3515625,-1.5625,-0.26757812,-0.296875,1.9296875,0.110839844,-2.140625,-1.21875,1.5390625,-0.6875,0.16699219,0.87890625,-1.171875,-0.609375,0.27539062,-0.17675781,1.6171875,0.07519531,-0.34570312,-0.17578125,0.8671875,1.3515625,-0.33007812,-0.66015625,-0.1328125,-0.5625,-0.4453125,-0.32226562,1.1953125,1.1796875,0.56640625,0.012329102,-1.3359375,-0.734375,0.34960938,-0.86328125,-0.10205078,0.33007812,1.8828125,0.09667969
-4,0,5,2.625,0.2265625,-0.76171875,1.46875,0.4453125,-0.65234375,-0.095214844,1.1875,-0.06591797,-1.1875,1.6171875,0.85546875,-0.3828125,-0.008972168,-0.828125,0.9140625,0.1875,-0.16210938,0.22851562,0.36328125,0.65234375,-0.25976562,-0.32617188,0.171875,0.54296875,1.0234375,-0.88671875,0.69140625,-2.078125,-0.24414062,-0.671875,1.0078125,-0.05029297,1.625,0.89453125,0.43554688,0.8671875,-0.39257812,0.48632812,1.1484375,1.3203125,0.40234375,0.25195312,0.2421875,0.010009766,-0.99609375,0.74609375,-0.66015625,-0.73046875,-1.9296875,-0.072753906,-0.8671875,-1.8125,-1.09375,-1.9453125,0.046142578,1.5859375,-0.20703125,0.07763672,-0.42773438,0.053710938,0.119140625,-0.79296875,0.9921875,-0.14453125,-2.046875,0.51171875,-0.21777344,0.47460938,-0.3671875,0.008972168,-2.671875,-0.48632812,-0.45898438,1.3125,1.703125,0.984375,0.0074157715,-0.26367188,-1.6015625,-0.9453125,0.484375,-0.33398438,-2.078125,1.9296875,0.43164062,0.27539062,-0.9140625,0.5859375,0.1875,0.84765625,0.5390625,0.5625,0.71875,-0.05444336,-1.59375,-0.78515625,-0.12011719,-0.57421875,0.22167969,-0.25390625,1.1484375,0.1953125,-0.32421875,0.20605469,-2.15625,-0.71484375,1.046875,-0.35351562,1.0234375,0.16894531,1.0,0.025390625,0.74609375,1.78125,1.421875,-0.42773438,-1.90625,-0.30664062,-0.49414062,-0.70703125,-0.90234375,-0.4296875,0.18261719,-1.5703125,1.5390625,-0.796875,-0.703125,0.22753906,-0.36523438,1.3984375,-0.9609375,0.018798828,0.44921875,-1.6328125,0.9765625,0.66015625,-0.2578125,0.58203125,-2.078125,1.625,-1.125,-1.2578125,1.1171875,0.8515625,0.73046875,-2.15625,1.4140625,1.484375,2.03125,-0.34179688,-0.35742188,0.27148438,-0.13964844,0.59375,1.6015625,-1.546875,-0.08251953,0.8046875,-0.43945312,-0.50390625,1.7421875,0.68359375,1.2265625,0.061279297,0.828125,-0.01977539,0.26953125,0.12597656,-0.94921875,-0.3671875,-1.1484375,-0.1015625,-1.75,-0.68359375,-0.3203125,0.23144531,-0.765625,0.37304688,-0.3046875,0.12597656,-1.4765625,-0.59375,-2.328125,0.22070312,1.046875,0.265625,-0.42773438,0.125,-2.15625,0.60546875,-0.6875,0.03149414,-1.0546875,0.5390625,3.1875,-0.7578125,-0.42382812,0.70703125,0.8359375
-4,0,6,0.9609375,1.21875,-0.484375,-1.4609375,1.328125,0.35351562,-1.1953125,0.75,1.1484375,-0.24414062,0.3515625,0.609375,-1.5,0.36523438,0.41601562,-0.72265625,-1.2109375,0.037841797,1.59375,1.2578125,0.42773438,-0.6875,0.34570312,0.28320312,-1.046875,0.036376953,-1.359375,-1.2578125,-2.578125,-0.6640625,-0.7265625,0.46289062,-1.1015625,1.015625,-0.31640625,-0.20214844,0.50390625,0.67578125,0.546875,1.0625,-0.068359375,0.58203125,0.9453125,0.875,0.35742188,-0.13476562,0.6328125,-1.640625,0.48828125,-0.31835938,1.0234375,-0.80078125,0.69921875,0.38085938,-2.140625,-0.21386719,0.40820312,-1.21875,0.32617188,-1.546875,0.19628906,0.78515625,1.359375,-1.0390625,0.625,-1.2265625,0.35546875,-1.7109375,0.15820312,0.93359375,-0.07519531,1.578125,-0.41210938,-1.25,0.11035156,0.671875,-0.048828125,0.5078125,-0.6171875,0.049316406,0.38867188,-1.03125,0.40820312,-0.625,0.20019531,-0.6953125,0.1328125,-1.96875,1.1796875,0.23535156,0.55859375,1.59375,-1.03125,1.75,-0.55078125,-0.6484375,-0.484375,0.65625,-0.88671875,0.014282227,0.05859375,1.3046875,1.421875,-1.453125,-0.53125,0.31835938,-0.3046875,0.75,0.515625,-0.30078125,0.51953125,0.66796875,-0.015625,0.76953125,1.3515625,-0.45507812,-0.39648438,-1.09375,0.18359375,-0.38671875,0.07324219,-0.10058594,-0.5625,0.16503906,-1.09375,1.921875,0.025268555,-1.9453125,1.109375,2.9375,1.2578125,-1.3515625,-1.5,1.3125,-2.0625,-0.78515625,-0.93359375,-0.81640625,0.47265625,-1.7265625,1.9921875,-1.9296875,-1.859375,2.171875,0.84375,-0.6640625,0.4453125,0.75,-0.034423828,1.4140625,0.28320312,-0.8203125,1.0625,-0.7890625,-0.38867188,1.4296875,1.796875,0.039794922,0.32226562,-1.7578125,0.69140625,1.6015625,0.8046875,2.265625,-0.095703125,0.62890625,-1.2890625,-1.828125,1.4609375,-0.65234375,0.9765625,-1.28125,-0.021972656,-0.3203125,-0.27148438,-0.8125,-0.6171875,-0.54296875,0.18359375,-0.030273438,-0.38867188,-0.90625,0.23632812,0.45703125,-0.78125,1.40625,-0.46679688,-0.87890625,1.2109375,-0.99609375,-1.046875,-0.6171875,-0.43164062,-0.53125,-0.75390625,2.140625,-0.45117188,0.6796875,-0.5234375,0.39453125
-4,0,7,1.734375,-0.51171875,-0.18066406,0.24902344,-0.5390625,-0.765625,-0.69140625,0.37695312,-0.92578125,-0.625,0.48632812,1.03125,0.5625,0.640625,-0.1328125,1.3984375,-0.018188477,-0.5078125,1.734375,-0.49804688,1.9140625,0.14355469,-1.1328125,0.060791016,2.578125,0.1640625,-0.5078125,-1.46875,-0.96484375,0.052734375,-0.79296875,1.1796875,0.25585938,0.03112793,-1.2890625,-0.546875,0.6796875,-0.9453125,-2.015625,-0.62890625,0.006866455,-0.82421875,-0.4453125,0.067871094,-0.17578125,0.088378906,0.11816406,-0.80078125,-1.296875,-1.1640625,-0.38671875,-0.84765625,-1.1640625,-0.60546875,-0.43945312,2.53125,1.2421875,-0.033447266,0.0020599365,-0.671875,1.25,-0.35351562,-0.23632812,1.6328125,0.39453125,-0.40234375,-0.19921875,-0.609375,2.9375,-0.4453125,-0.359375,-0.26757812,-0.99609375,0.359375,1.3671875,2.28125,-0.0013809204,0.28125,-1.5078125,-0.37304688,0.56640625,-0.42773438,-0.48828125,-0.50390625,1.6328125,-0.859375,0.82421875,1.328125,0.10888672,1.578125,0.49023438,-0.58984375,0.90234375,0.9296875,-1.921875,1.625,-0.94140625,0.5859375,0.4921875,0.15234375,0.22558594,0.020385742,-0.91796875,0.19726562,-0.34765625,-0.7265625,0.13085938,-2.34375,-0.86328125,0.056396484,0.703125,-0.6171875,0.6328125,-0.40039062,-0.27539062,0.11328125,-0.796875,0.578125,-0.6875,0.06689453,0.5234375,1.0546875,0.19238281,-1.6640625,-1.3828125,0.93359375,-1.34375,-0.45703125,-0.21289062,0.28515625,1.5078125,-0.44140625,0.005706787,-0.74609375,-0.953125,-0.16503906,2.21875,-0.4453125,-1.5703125,-2.234375,2.015625,0.78125,-1.625,0.068359375,-0.59765625,-0.234375,0.053710938,-0.35742188,0.9296875,1.8359375,-0.34765625,-0.80859375,1.34375,-2.359375,-1.2109375,-0.97265625,1.5,0.8671875,0.921875,-0.4609375,-0.5234375,-0.609375,0.0859375,-0.33984375,-0.0051574707,-0.092285156,0.265625,-1.2578125,1.2421875,-0.12695312,0.6875,-0.15332031,-1.234375,0.20019531,0.100097656,1.9453125,0.24414062,0.67578125,1.078125,-0.16699219,-1.0859375,-0.06689453,-0.7421875,-0.037841797,-0.55859375,-0.44726562,0.72265625,-0.78515625,1.8515625,-1.1875,-0.06298828,-1.3515625,0.29492188,-0.87109375,-0.34960938,0.85546875,-0.29296875,2.96875,2.421875,0.43945312
+3,0,5,1.09375,-0.07519531,-1.4921875,0.24121094,0.8515625,-1.703125,-0.57421875,2.953125,-0.059326172,-0.76171875,0.22753906,2.6875,0.265625,-0.36328125,-0.703125,1.734375,0.27929688,-1.328125,0.359375,0.921875,-0.24414062,-0.734375,-0.65234375,0.55078125,0.94140625,1.328125,-2.703125,0.38085938,-0.60546875,-1.0234375,0.34570312,-0.042236328,1.6953125,1.3515625,0.55078125,-0.62890625,0.99609375,-0.015625,-0.40429688,0.10839844,0.044433594,0.30859375,-0.18164062,-1.65625,-0.13867188,0.875,-1.5703125,0.44921875,0.072265625,0.7890625,-0.8515625,-1.421875,-1.1796875,-1.046875,-0.18945312,-0.20800781,0.74609375,0.29882812,-0.24804688,-1.0703125,0.21289062,0.71484375,1.3125,0.80859375,-1.265625,-0.32421875,0.3125,0.28320312,-0.578125,-0.40234375,0.4140625,-0.42578125,0.50390625,-0.46679688,2.265625,3.09375,1.0390625,0.024169922,-0.80859375,-0.84375,1.4453125,0.6875,0.82421875,-1.6171875,0.03930664,-0.3984375,0.47460938,1.3125,0.578125,2.203125,1.8671875,-0.765625,-1.0078125,1.359375,-0.6171875,-0.091308594,-1.390625,0.0067749023,-0.1484375,0.6796875,-0.51171875,0.16601562,0.056640625,1.265625,-1.0390625,-0.18261719,-1.109375,0.03173828,-1.3828125,-0.35742188,-1.1328125,-1.53125,1.765625,-1.8828125,0.61328125,-0.036865234,-1.4609375,0.6953125,-0.0051879883,-0.39257812,1.2734375,0.11425781,-0.8828125,-0.10644531,-0.045166016,2.421875,-0.7578125,-0.62890625,1.40625,-1.5390625,-1.1640625,-2.8125,0.10205078,-0.89453125,-1.203125,0.33984375,1.1640625,-0.27929688,0.18359375,-1.1875,0.92578125,-1.2734375,-1.3984375,-0.55078125,-0.20117188,0.53515625,-0.8203125,0.765625,-0.34960938,0.02746582,0.41796875,-0.54296875,0.21386719,-1.2109375,0.28125,-0.12988281,0.96875,0.044189453,1.328125,0.58984375,-0.18164062,2.46875,0.42578125,0.94921875,-0.064453125,0.3984375,0.18652344,0.48046875,1.2734375,1.2578125,-1.03125,-0.640625,-0.88671875,0.9140625,-1.2890625,0.29492188,0.14257812,-0.30078125,0.23535156,-0.3046875,-0.546875,-0.32421875,0.49023438,-1.203125,0.546875,-0.5234375,-1.1015625,-0.90625,-0.51953125,-0.9921875,0.5703125,0.96875,-0.4296875,-0.80859375,-0.16503906,1.9296875,-0.18652344,0.93359375,-1.4609375,-0.16992188
+3,0,6,-0.31054688,0.45898438,-1.703125,-1.1796875,-0.609375,0.16015625,-0.45703125,2.109375,0.5625,-0.81640625,1.7421875,1.1171875,1.453125,-0.3671875,0.080078125,0.9609375,-0.07080078,-0.93359375,0.76171875,-0.53125,0.625,-0.010375977,0.8671875,0.96484375,-0.055419922,0.13574219,0.22558594,-0.30664062,-1.2890625,-1.7265625,-0.3125,1.6796875,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.24023438,0.52734375,0.75,-0.203125,-0.45117188,0.16503906,-1.5078125,0.27539062,0.84375,0.23828125,0.5546875,0.39648438,-0.30273438,-2.21875,0.796875,-0.9140625,0.35546875,-0.8671875,-1.015625,-0.24023438,-0.6171875,0.28320312,-0.047607422,-0.11816406,0.87109375,0.6875,1.5390625,1.4140625,0.08886719,-1.28125,0.54296875,-0.765625,0.54296875,-0.40429688,0.123535156,0.48632812,0.095703125,-0.625,2.421875,-0.609375,0.017333984,-0.14453125,-0.03112793,-0.88671875,0.092285156,1.0,-0.8359375,0.88671875,0.7578125,-0.61328125,-0.41210938,1.0859375,-0.34570312,1.3984375,-0.24902344,-0.875,0.8359375,-0.115722656,-1.125,0.671875,-0.8359375,-1.390625,-1.21875,-1.4140625,0.68359375,3.140625,-0.81640625,-1.2421875,-1.171875,-0.67578125,0.30664062,-0.15039062,0.9140625,-0.55078125,-0.47070312,0.87890625,-1.25,1.765625,-0.30273438,-0.5546875,-0.359375,0.011352539,-0.64453125,-0.0015106201,-0.828125,0.45898438,-0.43945312,-1.03125,1.28125,-0.88671875,-1.65625,1.6015625,0.53125,-0.93359375,-0.78125,0.37304688,-0.98046875,-1.734375,1.4921875,0.69140625,0.33398438,-0.43164062,-1.9765625,2.5,-1.375,-1.0078125,-0.064941406,-0.1796875,-0.26367188,0.328125,-1.59375,-1.1953125,0.24023438,-0.07080078,1.59375,1.421875,-2.109375,0.10449219,-0.18457031,0.4140625,-1.03125,1.6875,-0.578125,0.421875,-0.27539062,0.89453125,-0.71484375,-0.017578125,2.25,-0.9453125,-0.11328125,1.5859375,-0.2578125,-1.625,-1.1015625,-2.078125,-1.5546875,-0.734375,0.484375,-0.032470703,0.0038757324,-1.328125,-1.265625,-0.13183594,-0.48632812,0.7421875,-0.029785156,0.73828125,-1.0546875,-0.45703125,0.15429688,-1.1015625,1.078125,1.734375,-0.21484375,0.6875,-0.46484375,-0.94921875,0.75390625,1.90625,1.6015625,0.65625,1.4296875
+3,0,7,0.578125,0.421875,-0.9765625,-0.27929688,-1.546875,-1.1796875,0.515625,1.3671875,-0.12890625,-0.9140625,1.6796875,1.3984375,0.7578125,0.19824219,0.4296875,0.94921875,-0.83203125,-0.28125,1.0625,-0.7265625,0.80078125,-0.29101562,2.296875,1.9921875,-0.28320312,0.5546875,-0.17480469,-1.1640625,-0.578125,-0.32421875,0.35742188,0.64453125,1.0703125,2.125,0.00074768066,-0.4765625,-0.20507812,0.5,0.60546875,-1.2734375,-1.0703125,-1.84375,1.0,-0.48242188,0.66796875,0.20019531,0.3515625,-0.19433594,-2.375,-0.13085938,-0.34960938,0.14648438,-1.65625,0.95703125,1.6328125,-0.11035156,-0.53515625,-0.021728516,-0.53125,0.028198242,0.24609375,0.26367188,-1.40625,-1.1796875,1.1328125,1.5234375,-1.0703125,-0.23925781,1.0234375,0.09277344,-1.4375,0.88671875,-0.890625,0.49609375,0.9375,0.66015625,1.0703125,0.61328125,-0.67578125,0.39257812,-0.77734375,0.546875,1.7578125,0.03173828,1.0390625,0.059570312,-0.39257812,1.2578125,0.06689453,0.82421875,1.2265625,-0.90625,-1.3125,0.52734375,-0.42578125,-0.7578125,1.5625,-0.31640625,-0.51171875,-0.51171875,-0.9140625,-1.0078125,2.421875,-0.26367188,-1.953125,-1.6875,0.71484375,-1.515625,0.859375,0.640625,-0.26171875,-1.609375,0.92578125,-1.234375,0.2734375,-0.53515625,0.028686523,0.42578125,0.3515625,-1.359375,1.3671875,0.47265625,-0.31445312,-2.140625,0.45117188,2.03125,-0.73828125,-0.119628906,1.796875,-1.5859375,0.44921875,1.25,1.203125,-2.609375,-2.28125,-0.28320312,-1.0078125,0.5625,-0.028320312,-1.4765625,0.89453125,-0.099121094,-0.2890625,0.29296875,-0.40234375,0.54296875,-0.9921875,-0.30078125,-0.80078125,0.64453125,-0.43945312,0.29882812,-0.6953125,-0.15332031,-0.29101562,-2.109375,-0.7734375,-0.55859375,0.53515625,1.46875,-0.80859375,-1.375,0.076171875,-0.8515625,0.8984375,-0.04248047,-2.0,-1.5234375,2.046875,-0.60546875,-0.31054688,0.54296875,-1.5234375,-0.47265625,-0.22363281,-0.092285156,1.546875,0.21875,-0.3515625,0.11328125,0.47460938,1.765625,0.4765625,-0.25585938,0.63671875,-0.7109375,-0.37695312,0.31445312,0.52734375,1.5390625,0.828125,0.60546875,-0.21679688,-0.5078125,0.4140625,-1.546875,0.765625,0.1328125,1.8046875,0.20996094
+4,0,5,2.046875,0.008178711,-1.2265625,1.3046875,1.1015625,-1.0078125,-0.7109375,1.5546875,-0.13476562,-0.9296875,1.03125,1.0625,1.0546875,0.23925781,-1.75,0.84765625,-0.36914062,-0.08300781,0.37304688,0.011230469,0.87890625,0.49609375,0.6328125,0.36523438,0.40820312,0.088378906,-1.7734375,0.33203125,-1.6640625,0.45898438,0.1640625,0.79296875,1.34375,1.7578125,0.71484375,0.41015625,0.24414062,-1.2421875,0.73046875,0.9140625,-0.80078125,0.20507812,-1.0546875,-0.73046875,0.40429688,-0.106933594,1.296875,-0.20605469,-0.8203125,-0.78125,-0.44726562,-1.1953125,-2.203125,-1.546875,-1.6796875,-0.82421875,0.28125,0.29492188,0.3828125,-0.6953125,0.10058594,0.23242188,0.04711914,0.7578125,-0.21777344,-1.2578125,-0.051513672,0.20996094,0.13378906,0.921875,1.1953125,-2.828125,0.040527344,-0.037597656,0.61328125,2.03125,0.3984375,-0.65234375,0.15039062,-1.9375,-0.48046875,0.29882812,0.11279297,-1.6328125,1.125,0.64453125,0.25195312,-0.36914062,1.2734375,-0.06982422,1.953125,0.9296875,0.002456665,0.4765625,0.49804688,-1.546875,-0.18945312,-0.6328125,-0.67578125,-0.06738281,-0.90625,1.3203125,0.9453125,-0.29296875,-0.62890625,-1.859375,-1.25,-0.18066406,-0.118652344,1.390625,-0.37695312,0.51171875,0.38867188,-0.3203125,1.0,1.296875,-0.77734375,-0.8671875,0.078125,-0.265625,-0.12597656,-0.72265625,-1.0859375,0.83984375,-1.3671875,1.515625,-0.90625,-1.09375,0.14941406,0.29492188,0.6171875,-0.79296875,-0.40234375,-0.6171875,-0.84375,0.703125,0.13574219,-1.171875,0.484375,-2.28125,1.9765625,-1.984375,-1.09375,0.44726562,0.76171875,0.3125,-1.328125,0.6640625,1.421875,2.46875,-0.030517578,0.84765625,0.4609375,0.29101562,1.5234375,1.0859375,-0.953125,0.26953125,0.78515625,-0.97265625,-0.18847656,1.5,1.015625,1.15625,-0.67578125,0.88671875,-0.26367188,-0.15527344,1.1171875,-1.1328125,-1.2109375,-1.3828125,0.096191406,-1.828125,-1.6328125,-0.82421875,0.3515625,-1.8515625,0.5390625,-0.09326172,-0.48828125,-0.8828125,0.52734375,-2.0,1.5,0.39648438,0.58984375,-0.50390625,-0.053466797,-2.09375,1.125,-0.056396484,1.15625,-0.46875,-0.16992188,2.65625,0.609375,0.014770508,0.29296875,1.0234375
+4,0,6,0.38671875,1.015625,-0.84375,-1.046875,1.4921875,0.36132812,-1.2421875,1.28125,0.82421875,-0.13574219,0.16894531,0.70703125,-0.6484375,0.14550781,-0.390625,-0.62890625,-1.2109375,0.21386719,1.8671875,1.171875,0.6640625,-0.39453125,0.94921875,0.6796875,-0.84375,-0.78515625,-2.0625,-1.0625,-2.046875,-0.029174805,-0.38867188,0.07470703,0.22265625,1.046875,-0.5703125,0.096191406,0.38085938,-0.23925781,0.54296875,0.875,-1.671875,0.46875,-0.040527344,0.08105469,0.26953125,0.38085938,0.78125,-0.875,0.70703125,0.65234375,0.98046875,-1.2734375,0.103027344,0.28515625,-2.0,-0.6484375,-0.34179688,-1.0078125,0.6171875,-1.390625,0.28710938,0.76171875,2.0625,-0.71875,0.47851562,-1.0625,-0.04663086,-1.265625,-0.171875,1.484375,0.50390625,1.390625,0.04638672,-1.0390625,-0.040771484,0.83984375,-0.37890625,0.3046875,-0.71484375,0.063964844,0.51953125,-0.9140625,0.46484375,-0.32617188,-0.375,-0.26367188,-0.50390625,-1.34375,1.5,0.3828125,1.3125,1.7421875,-0.98046875,1.4375,-0.53515625,-0.73828125,-0.4140625,-0.024902344,-0.97265625,-0.095214844,-0.62890625,1.2265625,1.6640625,-1.609375,-1.1015625,0.5859375,-0.9296875,0.110839844,0.31445312,0.020263672,0.3515625,0.40039062,0.106933594,0.390625,0.99609375,-0.53125,-0.5703125,-0.31835938,0.6796875,0.07714844,0.25195312,-0.2265625,-1.0,0.22167969,-1.1640625,2.09375,-0.18554688,-2.1875,1.359375,2.828125,0.61328125,-1.625,-2.03125,0.48242188,-1.6875,-0.6953125,-1.1640625,-1.4765625,0.40820312,-1.6875,2.140625,-2.296875,-0.984375,1.8671875,0.796875,-0.7578125,0.78125,0.2265625,-0.37304688,1.75,0.37304688,-0.22460938,1.171875,-0.38085938,0.05444336,0.83984375,1.6796875,0.35351562,0.7578125,-2.046875,0.8828125,1.203125,0.97265625,2.21875,-1.0390625,1.0390625,-1.6171875,-1.4140625,2.03125,-0.81640625,0.390625,-1.5078125,0.11621094,-0.18847656,-0.71875,-1.0390625,-0.49023438,-1.375,0.23046875,0.0035247803,-0.8203125,-0.609375,0.83203125,0.60546875,-0.35546875,1.21875,-0.064941406,-0.56640625,1.0703125,-0.9765625,-1.109375,0.036865234,0.45898438,-0.5859375,-1.03125,1.9453125,0.7890625,0.671875,-0.421875,0.66015625
+4,0,7,1.2578125,-0.80078125,-0.62890625,0.19628906,-0.296875,-0.51953125,-0.765625,0.88671875,-0.6796875,-0.62890625,0.32421875,1.1953125,1.0546875,0.52734375,-0.6640625,1.5078125,-0.31054688,-0.44921875,2.265625,-0.46289062,1.6953125,0.14453125,-0.20214844,0.046142578,2.1875,-0.54296875,-0.984375,-1.3125,-0.546875,0.49023438,-0.30273438,0.5390625,1.0390625,-0.013305664,-1.984375,-0.13085938,0.53515625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.86328125,-1.0625,-0.23925781,-0.09716797,0.91796875,0.1953125,-0.68359375,-1.2578125,-0.10107422,-0.44335938,-1.0078125,-1.5859375,-0.59375,-0.5,1.953125,0.48632812,0.014770508,0.23535156,-0.71875,1.3203125,-0.578125,0.52734375,1.53125,0.5859375,0.1875,-0.48828125,-0.45898438,2.265625,-0.023071289,0.18359375,0.029052734,-0.37890625,0.28320312,1.0859375,2.375,-0.25195312,0.08203125,-1.6484375,-0.34375,0.88671875,-0.51953125,-0.45703125,0.111816406,0.99609375,-0.15625,0.3203125,1.5859375,0.5625,1.6015625,1.5859375,-0.40429688,0.2890625,0.97265625,-1.7265625,1.5546875,-0.90625,-0.030151367,0.38085938,-0.22167969,-0.83984375,0.19140625,-0.21777344,0.28710938,-0.91796875,-0.29882812,-0.4921875,-2.6875,-0.64453125,0.328125,0.48046875,-0.953125,0.6484375,-0.6796875,-0.609375,-0.10107422,-0.79296875,0.97265625,-0.24609375,0.21289062,0.78515625,1.1640625,-0.25195312,-1.7109375,-1.2109375,0.8984375,-1.421875,-0.9453125,0.024658203,0.23242188,0.96875,-0.29882812,-0.29296875,-1.296875,-0.61328125,-0.20605469,1.8671875,-0.62890625,-1.6875,-2.625,2.171875,0.64453125,-1.078125,-0.12792969,-0.84375,-0.20703125,0.6796875,-0.875,0.29882812,1.9140625,-0.36523438,-0.11035156,1.3203125,-2.09375,-0.30664062,-1.5859375,1.625,1.2734375,1.3984375,-0.83984375,-0.48046875,-0.8046875,0.36914062,-0.28515625,-1.1328125,0.18652344,0.09814453,-1.0859375,1.828125,-0.390625,0.30859375,-0.28125,-1.2890625,0.45703125,-0.51171875,1.6015625,0.21191406,-0.12695312,0.9765625,0.091308594,-1.328125,0.22558594,-0.020385742,0.5,0.019897461,-0.5546875,1.0,-0.44140625,1.5546875,-0.90234375,-0.16210938,-0.5390625,1.2421875,-0.67578125,-0.81640625,0.48046875,0.7109375,2.796875,2.171875,0.546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
index c8ece4d5..88760c11 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,5,1.890625,1.671875,-0.37304688,-0.3125,-0.74609375,-1.3984375,-0.87890625,0.22949219,0.5625,-1.1875,-0.19140625,-0.40039062,-2.421875,1.6015625,-0.6328125,1.2578125,0.546875,0.16992188,-1.4296875,0.5859375,1.7890625,0.5,-0.42773438,0.004333496,1.15625,-0.22167969,-1.609375,0.095703125,-2.140625,0.640625,-0.8828125,0.19726562,0.578125,0.75390625,0.29882812,0.24023438,1.046875,1.0078125,-0.62890625,0.55078125,-0.20605469,0.104003906,-0.33398438,0.8359375,-0.82421875,-0.03515625,1.0546875,-0.37304688,-0.46289062,-0.47460938,-0.33203125,-1.015625,-0.6328125,-0.390625,-0.5546875,2.375,1.6953125,-0.14257812,-0.15429688,-1.71875,0.56640625,0.7578125,0.38671875,0.37695312,0.98046875,0.4140625,-0.859375,1.0,1.640625,-0.6796875,-1.46875,-1.453125,-2.015625,-0.26757812,-0.19726562,0.47070312,1.0625,1.09375,0.107910156,-0.34765625,1.0234375,-1.3515625,0.9765625,-0.921875,0.45117188,-0.95703125,0.96875,0.42578125,-0.67578125,-0.66015625,2.84375,-0.4140625,1.5546875,1.1953125,-1.328125,0.72265625,-1.0234375,0.58203125,0.10253906,1.046875,-0.38671875,0.13378906,1.0703125,-1.6328125,0.546875,-0.578125,0.41601562,-0.44140625,-0.76171875,0.22265625,0.734375,-0.068847656,1.1953125,-0.05908203,0.53515625,-0.47460938,-0.17773438,0.07910156,0.18847656,-1.1015625,0.890625,-0.9921875,-0.4453125,-1.4140625,-1.3984375,2.484375,-0.40820312,-1.359375,-1.5078125,-1.203125,1.515625,-0.5546875,-1.5234375,0.28710938,-2.265625,1.5625,0.72265625,-0.4765625,-1.5390625,-2.3125,1.1640625,-2.234375,0.2109375,0.30273438,0.8203125,0.125,-1.0390625,0.34375,1.140625,1.828125,-1.2109375,-0.6953125,0.734375,-0.58203125,0.5625,0.15429688,0.41015625,0.06298828,0.21386719,0.49804688,-0.734375,-0.2734375,-0.765625,0.265625,-0.6796875,2.25,-0.107910156,0.19238281,1.515625,-0.57421875,-0.9609375,-1.125,-0.8515625,-1.3046875,0.33007812,1.2734375,1.15625,-0.33984375,-0.36328125,-1.6171875,0.69140625,1.34375,-0.5703125,-1.015625,0.234375,-1.109375,-0.06640625,1.0546875,0.6484375,-1.1171875,-0.1640625,0.34179688,0.10253906,-0.29882812,-0.52734375,1.203125,1.421875,0.23730469,0.73046875,1.1015625
-5,0,6,0.79296875,-0.24023438,-0.79296875,-1.0390625,0.19824219,-1.375,-1.296875,1.3125,-0.17578125,0.6328125,-0.9453125,-0.5078125,0.8359375,0.32421875,-0.8125,1.6328125,-0.18066406,-1.2109375,0.20214844,0.4140625,2.40625,-0.7109375,-1.265625,-0.02355957,0.77734375,-0.07763672,-1.640625,-1.1328125,-1.6640625,1.75,-0.28320312,1.4140625,-2.046875,1.1015625,-0.62890625,-0.13867188,-0.013183594,0.03491211,-0.3359375,-0.578125,0.5546875,0.03515625,-0.79296875,0.31835938,-0.33398438,1.5390625,-0.46484375,-0.71484375,1.0859375,-0.515625,-0.8671875,-3.375,-1.328125,0.2265625,-1.46875,2.03125,1.796875,-0.7421875,-0.65625,-2.03125,-1.0703125,-0.47851562,0.5,1.2734375,1.375,1.5859375,-1.3828125,1.0625,0.5546875,0.41015625,0.671875,-1.8515625,-1.5234375,-0.49023438,0.10595703,0.73046875,0.640625,0.08544922,0.59765625,-0.111328125,1.078125,0.11376953,0.103515625,0.67578125,0.22167969,-0.28320312,-0.23339844,-0.46875,0.060546875,-0.10595703,0.5390625,-0.47265625,2.765625,-0.26757812,-0.98046875,-0.28125,-1.625,1.0703125,-0.007293701,0.4140625,0.15820312,0.1640625,0.05029297,-0.104003906,0.8515625,0.38476562,-0.87109375,-1.6875,-1.1875,-0.91796875,0.60546875,1.0859375,-0.4453125,0.024414062,1.0078125,-2.609375,0.14648438,0.4921875,0.41992188,1.296875,0.24121094,-0.62109375,0.0070495605,-1.359375,-1.6875,0.9375,-1.5078125,-0.625,-0.78515625,0.75,2.328125,-1.546875,-0.76953125,0.5703125,-0.734375,0.1875,0.6484375,0.36132812,0.56640625,-2.015625,-0.12109375,2.46875,0.26367188,1.453125,0.6875,0.32421875,-1.1796875,1.3984375,0.7265625,1.40625,-0.122558594,-2.21875,0.68359375,-0.89453125,-0.38476562,0.40625,1.5234375,0.25390625,-0.07519531,0.59375,0.06347656,1.265625,0.15820312,0.65625,-0.11328125,1.296875,0.43945312,0.075683594,0.015197754,0.44140625,0.22851562,0.65625,-0.015258789,-0.34765625,-0.48632812,1.703125,-0.06982422,-1.2578125,-0.61328125,-0.75390625,-0.546875,0.83203125,-0.58984375,-0.5859375,-0.5390625,0.18554688,0.6953125,-0.15136719,0.14648438,-0.90625,-1.03125,-0.65625,0.53515625,-0.52734375,-0.048339844,1.171875,0.40820312,-0.19628906,1.8984375,1.5
-5,0,7,1.640625,0.83984375,0.90625,0.87109375,-1.1484375,-0.37109375,-1.3671875,-0.29492188,2.5625,-1.546875,1.96875,-0.04345703,-0.18457031,1.3125,0.07519531,-0.390625,-0.2734375,-0.016601562,2.171875,1.21875,0.044921875,-0.58984375,-0.4140625,0.37695312,-0.34765625,-0.1640625,1.078125,-0.48632812,0.4921875,0.013183594,-0.29882812,0.49023438,0.171875,-0.703125,0.9140625,0.42382812,1.296875,0.47070312,-0.104003906,1.34375,-1.7734375,-1.2109375,-1.4765625,1.015625,-0.49609375,-0.984375,0.59765625,-0.53125,0.8203125,-0.36914062,0.029541016,-0.33789062,-0.17578125,0.859375,0.8125,0.625,0.31054688,-0.66015625,-0.796875,0.92578125,1.2421875,-0.20507812,0.47070312,0.06640625,0.40429688,-0.06689453,0.31835938,-1.7890625,-0.75,0.16894531,-0.296875,1.40625,0.57421875,-0.8125,-0.53515625,0.6875,-0.34570312,0.7109375,-1.296875,0.65625,-2.53125,-0.71484375,0.984375,-1.8984375,-0.29492188,-0.22851562,0.17675781,-1.890625,0.37109375,0.16210938,0.69921875,1.015625,-0.31640625,1.0625,-0.53125,0.03930664,-1.6875,1.140625,-0.6015625,0.068359375,-1.265625,0.765625,0.6796875,-1.4921875,0.22070312,1.859375,-0.119140625,-0.064453125,-0.40429688,0.42382812,-0.22851562,0.5234375,1.734375,-0.31445312,0.89453125,0.34179688,-0.6953125,0.033691406,-0.71875,0.12988281,0.609375,-0.042236328,0.46679688,-1.3515625,-2.34375,0.921875,-0.072753906,0.13476562,-0.55859375,1.3203125,1.3203125,0.44335938,-0.8046875,0.22753906,-1.1484375,1.359375,0.26367188,-0.66015625,-0.27734375,-1.515625,1.75,-0.0234375,-1.1484375,1.625,-0.2265625,-0.11376953,1.1171875,-0.07910156,-0.375,0.26367188,-0.90625,-0.75390625,2.609375,-0.85546875,-1.296875,0.51171875,1.359375,-0.86328125,0.36914062,0.49023438,-0.796875,-0.14355469,1.28125,1.4765625,0.16503906,-0.59765625,-1.1015625,-0.7578125,0.50390625,-1.4921875,-1.703125,0.10205078,-3.078125,-0.75,-0.42578125,2.28125,0.875,-0.35351562,-0.62109375,-0.052001953,-0.21875,-0.1796875,-0.8203125,-1.359375,-1.703125,0.15527344,1.1015625,0.14453125,-0.37109375,-0.85546875,-1.5,-0.4453125,-0.18066406,-1.0390625,-2.609375,1.8984375,2.234375,0.2578125,0.69921875,1.2109375
+5,0,5,0.58984375,0.99609375,-0.34179688,-1.34375,-0.484375,-0.9140625,-0.88671875,0.4609375,0.94921875,-1.2421875,-0.19433594,-0.5859375,-1.09375,1.5390625,-1.671875,0.66796875,-0.3359375,0.453125,-0.8125,0.87890625,1.6640625,0.703125,0.06201172,0.34765625,0.40625,-0.58984375,-1.8828125,-0.15136719,-0.99609375,1.21875,-0.9296875,-0.2265625,0.83984375,0.2109375,-0.23046875,0.90625,0.40039062,-0.05908203,-0.58984375,0.62109375,-1.7890625,0.38085938,-0.12158203,0.41015625,-0.0019302368,1.3359375,0.70703125,0.13671875,-0.40234375,1.2578125,-0.41796875,-0.6796875,-1.5234375,-0.018432617,-0.041503906,1.8125,0.013305664,-0.23535156,0.40429688,-1.546875,0.29101562,0.39648438,1.2890625,0.24316406,0.96875,1.1796875,-1.5703125,1.46875,0.012756348,-0.22949219,-0.796875,-1.203125,-0.76953125,-0.07128906,-0.58984375,0.015563965,0.49414062,0.5546875,0.30859375,0.6484375,1.078125,-1.046875,1.015625,0.42382812,-0.53125,0.49609375,0.3515625,0.640625,-0.58984375,-0.50390625,3.59375,0.41992188,0.79296875,1.265625,-0.51953125,1.0390625,-0.98828125,0.31835938,-0.58203125,0.34570312,-0.828125,0.17089844,1.3828125,-1.140625,0.045410156,0.17773438,-0.390625,-1.8671875,-0.4453125,0.83203125,-0.15332031,-0.40625,0.4296875,-0.80859375,-0.625,-0.609375,-0.34960938,0.69921875,0.42773438,-0.12597656,0.734375,-1.0703125,-0.62890625,-1.8515625,-0.828125,1.65625,-0.35742188,-1.625,-1.1015625,-0.96875,1.3203125,0.2578125,-2.375,-0.5859375,-1.5703125,2.015625,0.14355469,-0.45507812,-1.34375,-2.5625,1.1015625,-2.21875,0.58203125,-0.09277344,0.6484375,0.35546875,-0.20605469,-0.76953125,-0.088378906,1.5625,-1.5859375,0.59765625,0.60546875,-0.09814453,1.296875,-0.064453125,0.4140625,0.9375,0.78515625,0.41601562,-0.87890625,-0.6640625,-0.37109375,0.09033203,-2.6875,2.390625,0.22949219,-0.0134887695,1.4140625,-0.9140625,-1.8984375,-0.86328125,-1.1171875,-0.9296875,-0.44921875,1.5078125,1.1796875,-0.99609375,-0.18261719,-0.91015625,0.44921875,1.3046875,0.50390625,-0.040771484,1.0078125,-1.109375,-0.13085938,1.609375,0.578125,-1.3125,-0.23730469,1.359375,1.2265625,0.38085938,-0.5546875,0.43554688,3.265625,-0.09863281,-0.18554688,0.45117188
+5,0,6,-0.44140625,-0.69921875,-0.38476562,-1.4375,0.17578125,-1.203125,-0.9609375,1.640625,0.057373047,0.8203125,-0.8671875,-0.53125,1.4140625,0.875,-2.15625,0.93359375,-0.51953125,-0.859375,0.10546875,0.5078125,2.46875,-0.18359375,-0.8515625,0.24316406,0.48046875,-0.38671875,-2.078125,-1.125,-0.42578125,1.75,-0.26757812,0.50390625,-0.98046875,0.5546875,-1.25,0.49804688,-0.37890625,-0.49804688,-0.27929688,-0.6015625,-0.9453125,0.328125,-0.6953125,-0.28710938,0.53515625,2.484375,-0.107910156,-0.3125,1.390625,1.0703125,-1.0078125,-2.421875,-2.34375,0.43945312,-0.609375,1.6640625,-0.28125,-0.80859375,-0.26757812,-1.5546875,-0.87890625,-0.80859375,1.578125,0.9609375,1.4375,2.015625,-1.6640625,1.625,-0.95703125,0.6171875,1.3046875,-1.2890625,-0.6953125,-0.265625,-0.234375,0.25,0.265625,-0.33984375,0.73046875,0.73046875,1.578125,-0.42382812,0.28320312,1.734375,-0.84375,0.64453125,-0.8203125,0.05126953,0.13476562,-0.11230469,1.4765625,-0.3671875,1.78125,-0.09765625,-0.15429688,0.027709961,-1.84375,0.78125,-0.54296875,0.375,-0.2890625,-0.26757812,0.36328125,0.17675781,0.3671875,1.3125,-1.671875,-2.59375,-1.0703125,-0.609375,-0.20019531,0.78125,-1.0546875,-0.53125,-0.09716797,-2.359375,-0.09716797,0.81640625,0.33007812,1.5859375,0.11279297,-0.3828125,-0.15917969,-1.6484375,-1.2578125,0.6328125,-1.359375,-0.77734375,-0.7890625,0.48046875,2.0625,-0.7578125,-1.7109375,-0.546875,-0.28710938,0.8359375,-0.13867188,0.07861328,-0.09716797,-2.046875,-0.32226562,1.609375,0.58984375,0.95703125,0.52734375,0.15917969,-0.51953125,0.20410156,-0.59765625,1.0234375,0.03515625,-0.90625,0.54296875,-0.52734375,0.43945312,-0.17382812,1.4140625,1.0078125,0.55078125,0.35742188,-0.19433594,0.875,0.4140625,0.50390625,-1.796875,1.734375,0.73828125,0.20800781,0.46289062,0.03930664,-0.49804688,0.63671875,-0.0390625,0.18066406,-1.1171875,1.984375,0.18164062,-1.671875,-0.57421875,-0.099609375,-0.6640625,0.875,0.91796875,0.3828125,0.28125,-0.28710938,0.28125,0.39453125,0.3203125,-0.796875,-1.453125,0.7578125,1.5078125,0.14746094,-0.55078125,0.2734375,2.375,-0.3984375,1.0078125,1.125
+5,0,7,0.91015625,0.5625,0.828125,0.203125,-1.140625,-0.08105469,-1.1953125,-0.031982422,2.609375,-1.46875,1.828125,-0.040771484,0.35351562,1.3671875,-0.62890625,-0.42578125,-0.4609375,0.18457031,2.265625,1.1484375,0.2421875,-0.5078125,-0.15722656,0.58203125,-0.48632812,-0.53515625,0.65234375,-0.5234375,0.90625,0.5078125,-0.40429688,-0.0074157715,0.78125,-0.66796875,0.37695312,0.91015625,0.94921875,-0.18847656,-0.20898438,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24707031,-0.23242188,0.515625,-0.29296875,0.87890625,0.6796875,-0.15429688,-0.265625,-0.74609375,0.95703125,0.9296875,0.40820312,-0.65625,-0.49804688,-0.578125,0.77734375,1.1875,-0.37890625,1.2578125,0.09765625,0.640625,0.51171875,-0.20703125,-1.25,-1.453125,0.41210938,-0.11230469,1.3828125,0.78515625,-0.640625,-0.65234375,0.47851562,-0.52734375,0.4765625,-1.09375,1.2578125,-1.8984375,-0.71484375,0.9296875,-1.1640625,-0.703125,0.51171875,-0.36523438,-1.5,0.47851562,0.28515625,1.5390625,1.1953125,-0.47070312,0.875,-0.45117188,0.16308594,-1.78125,0.73046875,-0.63671875,-0.3671875,-1.6875,0.609375,0.9140625,-1.2890625,-0.34960938,2.171875,-0.640625,-0.7890625,-0.375,0.7890625,-0.4765625,0.26171875,1.203125,-0.578125,0.26367188,0.34960938,-0.625,0.453125,-0.41015625,0.5546875,0.6875,-0.0013961792,0.4453125,-1.6875,-1.90625,0.5625,-0.16992188,-0.203125,-0.34570312,1.140625,1.078125,0.61328125,-1.5078125,-0.265625,-1.1015625,1.578125,0.16503906,-0.73046875,-0.41796875,-1.7578125,1.6796875,-0.26953125,-0.625,1.5078125,-0.28320312,-0.07519531,1.4609375,-0.734375,-1.1796875,0.22070312,-1.078125,-0.16503906,2.140625,-0.57421875,-0.60546875,0.19042969,1.40625,-0.39648438,0.65625,0.38867188,-0.75,-0.30664062,1.390625,1.2421875,-1.1796875,-0.14648438,-0.6640625,-0.6171875,0.64453125,-1.703125,-2.109375,-0.25585938,-3.0,-0.40039062,-0.578125,2.484375,0.80078125,-1.046875,-0.48046875,0.36328125,-0.33398438,-0.09277344,-0.06982422,-0.6640625,-1.1640625,0.20800781,1.125,0.58984375,-0.390625,-0.8203125,-1.5234375,0.296875,0.546875,-0.86328125,-2.796875,1.2578125,3.171875,0.19921875,0.34570312,0.859375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
index 7fb3dcc2..184fba8e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,5,0.94921875,0.62890625,0.4921875,-1.28125,1.1796875,0.37304688,0.73828125,0.33007812,-0.4921875,-1.1328125,0.36328125,-0.44921875,0.33398438,0.9609375,-0.61328125,1.015625,-0.71484375,-2.4375,0.81640625,-1.0703125,-0.09716797,-0.78515625,0.71484375,0.34960938,-1.1015625,-0.48632812,-0.69140625,-0.8828125,-2.875,-0.049804688,-0.91015625,0.3828125,-0.671875,-0.04321289,-0.022094727,0.59375,1.21875,0.42578125,-1.2421875,2.390625,-0.48046875,-0.004211426,0.515625,0.58203125,0.18847656,-0.39453125,0.73046875,-0.83984375,0.5234375,0.55078125,-1.3671875,-0.89453125,-0.27734375,1.2734375,-2.90625,0.40039062,0.17871094,-0.84375,1.2890625,-2.046875,1.71875,0.19726562,-0.62109375,0.73828125,-0.87109375,-0.5078125,-0.87890625,-0.22851562,1.421875,1.4453125,-0.33789062,-0.62109375,0.01940918,-0.30859375,-0.56640625,2.203125,1.0546875,1.1796875,-1.796875,-0.6796875,0.81640625,0.75,-0.11621094,-0.7421875,0.08642578,-0.75390625,0.42382812,1.125,-0.8671875,0.6328125,0.51953125,-0.055419922,-1.015625,0.50390625,-1.6875,-0.12695312,-0.29882812,-0.041259766,0.77734375,-0.7734375,-1.6328125,-0.34375,2.109375,0.94921875,-0.15917969,-0.73046875,0.765625,-0.58203125,0.8359375,0.82421875,1.0234375,-0.13574219,1.0703125,-0.64453125,1.1796875,0.21777344,-0.4375,-1.0625,-0.64453125,-1.984375,0.42382812,-1.0703125,0.875,-1.0546875,-0.03881836,1.0234375,-1.640625,-2.359375,-0.41992188,0.33007812,-0.15332031,0.3125,0.49804688,-0.45507812,-1.078125,-0.111328125,-0.73046875,0.6875,0.73046875,-1.421875,3.28125,-0.60546875,-0.022705078,0.74609375,-1.2734375,1.15625,-0.9140625,0.421875,-1.7109375,1.0234375,-0.609375,-0.859375,1.21875,-0.9453125,-1.328125,-0.71875,0.140625,0.70703125,0.5390625,1.6953125,-0.12890625,-0.030151367,1.890625,1.734375,-0.51953125,1.453125,-1.4296875,-0.984375,1.6953125,0.15722656,0.9296875,-0.3203125,0.37695312,0.51953125,1.578125,-0.0546875,-0.80859375,-0.104003906,0.8984375,-1.03125,0.16113281,-1.2265625,-0.82421875,0.15917969,-0.33398438,0.5859375,-0.05419922,-0.64453125,1.484375,-2.15625,0.87109375,0.6796875,0.29882812,0.4375,1.140625,0.06933594,0.5,0.39453125,1.421875,0.18554688
-6,0,6,1.75,1.09375,-0.24414062,0.59765625,1.21875,0.36914062,-0.49414062,0.010375977,0.13769531,-1.6015625,1.46875,-0.21191406,-0.020385742,0.033447266,-0.34960938,-0.29492188,-0.95703125,-0.6875,1.0859375,1.3203125,0.06591797,-2.1875,-0.092285156,-0.032714844,-1.0078125,-0.74609375,-0.90234375,-0.36523438,-0.38476562,-0.59375,-0.40625,0.27539062,0.8828125,0.41015625,0.12109375,-0.57421875,0.5078125,0.40234375,-0.06542969,1.2890625,0.42773438,0.62109375,0.1875,1.859375,-0.640625,-0.19628906,-0.7890625,-1.34375,0.08886719,0.5703125,1.0625,-1.3125,0.13085938,-0.5625,-1.109375,-1.9609375,0.93359375,1.28125,-0.31445312,-1.0703125,1.6484375,1.734375,1.140625,-0.18066406,-1.2265625,-1.0546875,0.6171875,-2.5,0.12451172,0.40820312,-0.103027344,0.106933594,1.234375,-1.6484375,0.43359375,0.65234375,0.33984375,-0.41992188,-1.578125,0.3125,-0.8203125,0.58203125,0.13769531,-0.11425781,-0.20117188,0.55859375,-2.4375,0.048095703,1.1875,1.0859375,1.03125,0.8203125,1.25,0.44335938,-0.6640625,-1.0078125,-0.66796875,1.828125,-0.096191406,0.390625,0.061767578,-0.41992188,0.578125,0.02331543,-0.59375,0.30664062,-0.40625,0.21777344,-0.040039062,-0.43359375,0.49023438,1.8671875,0.30859375,-0.39648438,-0.13085938,1.0546875,-0.44726562,-0.29882812,-0.64453125,0.49609375,-0.016357422,-1.3125,-1.1171875,-0.12207031,-1.2421875,1.5859375,-0.75,-0.3046875,-0.6015625,0.76171875,0.010803223,-2.359375,-0.24902344,1.5859375,-1.2578125,2.703125,-0.47070312,0.55859375,2.25,-1.2890625,1.3671875,-0.625,-0.7734375,0.890625,0.71484375,-0.11230469,-0.28710938,0.5078125,0.31445312,-0.010009766,-0.01159668,-0.9765625,-1.640625,0.8359375,-1.296875,1.75,0.6875,-0.953125,-0.7109375,1.3515625,0.65625,1.0703125,0.87890625,3.484375,-1.84375,1.34375,-2.171875,0.57421875,-0.25195312,-0.022705078,-1.71875,-1.6953125,0.8203125,-0.78515625,0.92578125,0.20703125,-0.37890625,-1.546875,1.4453125,-0.515625,0.4296875,0.35546875,0.053710938,-0.7890625,-1.453125,1.6796875,-1.0078125,-0.73828125,0.6796875,-0.45117188,1.0390625,0.77734375,-1.1328125,-0.875,-0.6015625,-0.30078125,-1.5625,-0.29882812,0.23144531,0.37304688
-6,0,7,-0.34765625,0.27929688,0.10546875,0.8515625,1.1875,-0.13476562,-0.44726562,0.58203125,0.78515625,-0.734375,0.44726562,1.59375,-0.9296875,-0.20605469,-0.65625,0.22070312,0.03540039,-1.1171875,1.4765625,-0.43554688,-0.265625,-2.015625,-1.1328125,0.2421875,-0.052246094,-0.041748047,1.0546875,0.5703125,-0.17285156,-0.21191406,0.46289062,0.7734375,0.5,-0.29882812,2.796875,0.453125,1.2265625,-0.068359375,-0.9140625,0.9921875,0.44921875,-1.0625,-1.3125,-1.5390625,-0.97265625,0.38476562,-0.35742188,-1.046875,0.62890625,0.6875,1.140625,-0.27539062,-0.69921875,-0.0021362305,-0.18847656,-1.453125,0.8984375,1.34375,0.32226562,0.31054688,0.62890625,0.8515625,-0.9921875,1.578125,-1.0390625,0.08935547,-0.81640625,-2.203125,0.97265625,-0.49804688,0.36132812,1.4609375,1.640625,-1.6953125,0.78515625,0.9140625,0.080078125,-0.5078125,-0.08544922,0.80859375,-1.8203125,0.34375,-0.31054688,-1.4296875,0.55078125,-1.65625,-0.10498047,-0.5546875,1.0546875,0.67578125,0.78125,-0.56640625,0.1015625,1.03125,-0.25585938,-0.27929688,-0.75390625,-0.48828125,-0.828125,-0.003540039,0.80859375,0.36914062,0.75,1.2734375,-0.5625,0.75,1.25,0.53125,0.07128906,0.734375,1.2265625,0.80078125,2.65625,-1.4765625,1.1328125,-0.28320312,-1.65625,-0.1328125,-1.046875,-1.015625,1.1484375,0.58984375,-0.69140625,-0.44921875,-1.265625,2.046875,-0.033935547,1.25,0.6953125,-1.2421875,2.046875,-2.1875,0.52734375,0.8359375,-0.1640625,0.82421875,0.53125,0.14941406,0.4921875,-0.64453125,1.5234375,0.30859375,-1.984375,1.4140625,-0.47070312,-1.125,0.31054688,0.3671875,0.6015625,-1.40625,0.35742188,-0.27734375,2.796875,-0.765625,-0.859375,0.6640625,0.27734375,-2.046875,0.78125,0.296875,0.5234375,0.44335938,0.8984375,0.375,0.8671875,0.765625,0.006866455,0.68359375,-0.828125,0.026367188,-1.125,-1.1640625,-2.3125,-1.484375,-0.051513672,0.97265625,-0.18359375,-1.1015625,-1.671875,-1.0859375,1.1640625,0.33398438,-0.5,-1.765625,-0.94140625,-1.203125,-0.640625,-0.17285156,-1.0234375,1.046875,-0.92578125,-0.75390625,-1.6328125,-1.1640625,-1.296875,0.9609375,-0.984375,0.1640625,0.19726562,1.6484375
-7,0,5,1.5546875,1.359375,-0.57421875,-0.25390625,-0.94140625,-0.37890625,-0.95703125,0.921875,-0.9375,-0.21875,2.265625,-0.26757812,-0.33007812,-0.77734375,-0.72265625,1.9453125,0.22851562,-1.7734375,0.234375,-0.014099121,1.8984375,0.40429688,-0.16796875,0.6796875,0.875,-0.32226562,-0.33789062,-1.75,-2.203125,-0.8125,-1.0078125,2.953125,-0.625,2.265625,0.5078125,-0.14160156,0.41015625,0.140625,0.7734375,0.41015625,1.0703125,-0.119628906,0.6875,0.08886719,0.50390625,-0.14257812,0.04345703,-0.15820312,0.44335938,-2.25,-0.83984375,-0.30078125,-0.9921875,-1.4609375,-1.265625,1.0703125,1.1953125,-0.58984375,0.08642578,-0.3046875,0.014770508,-0.057861328,0.8828125,1.4609375,1.421875,0.19726562,-0.15820312,-0.6484375,0.81640625,-0.28320312,-0.4609375,-0.71875,-1.3203125,-0.48046875,0.16015625,0.6796875,0.20605469,0.20703125,-0.02368164,-0.20996094,-0.6015625,-0.40234375,1.0234375,-0.359375,1.59375,0.19238281,-0.16796875,0.37695312,0.91796875,-0.54296875,-0.040771484,-0.14648438,1.1796875,1.0625,0.06982422,-1.2421875,0.71484375,0.90234375,-1.109375,-0.10205078,0.024658203,0.74609375,1.734375,-1.7265625,-0.69140625,-2.140625,-1.71875,-0.53125,-0.76171875,0.37109375,0.63671875,-0.44921875,0.27539062,0.703125,1.421875,-0.13378906,-0.026855469,-1.2734375,-0.36523438,-0.10058594,-0.328125,-0.09765625,-0.6875,-1.6953125,-1.9609375,1.109375,-2.078125,0.119140625,0.41796875,0.94921875,1.1796875,-0.5859375,-0.09765625,0.024047852,-1.71875,0.53515625,1.796875,-0.008544922,-1.0625,-2.3125,1.046875,0.06591797,-0.6015625,0.68359375,-0.106933594,0.56640625,-1.125,-0.20996094,0.8359375,1.171875,-0.2890625,-0.81640625,1.2265625,-1.9296875,-0.08544922,0.234375,-0.080078125,-0.93359375,0.25585938,-0.3515625,1.015625,-0.28125,0.3125,-0.72265625,0.33789062,3.15625,-0.94140625,-0.41992188,0.41210938,-0.375,-0.012329102,-0.67578125,-0.83984375,-1.3125,0.20410156,1.3828125,0.24804688,-0.44921875,0.88671875,-1.3125,0.84765625,-0.12158203,-0.296875,-1.7890625,0.095214844,0.092285156,0.16210938,-0.032470703,0.48828125,-0.14453125,1.9921875,-1.7265625,-0.78515625,-2.0,0.45507812,1.59375,0.72265625,2.515625,1.328125,0.796875
-7,0,6,1.4296875,1.3984375,-0.46289062,-0.3515625,-1.6015625,-1.1328125,0.66015625,0.61328125,-0.21777344,0.21679688,0.2265625,0.37695312,-0.390625,-0.859375,0.28515625,1.875,-1.1875,-0.76953125,0.265625,0.82421875,1.0,-1.0546875,0.82421875,-0.076171875,1.4140625,0.0068969727,-1.4296875,-0.80078125,0.25585938,-0.5078125,-0.7109375,0.16503906,0.89453125,-0.63671875,-0.07714844,-0.5078125,0.5703125,-0.70703125,-0.76953125,-0.38476562,0.45703125,-0.084472656,0.20605469,-0.10205078,-1.2109375,-0.31640625,-0.056396484,0.34765625,-0.59375,-0.79296875,0.7890625,0.17382812,-3.15625,-0.18359375,1.1953125,2.140625,0.93359375,0.34570312,0.34179688,-0.025878906,-0.92578125,-0.328125,-0.07470703,0.5390625,1.390625,-1.078125,-0.052001953,-0.73046875,0.96484375,0.022583008,-0.7890625,1.2578125,-0.50390625,-1.0234375,2.65625,1.140625,1.2109375,0.1328125,-1.4296875,0.041503906,0.08496094,0.37109375,1.203125,-0.5390625,1.0546875,-1.0,-0.609375,1.0859375,1.203125,0.111328125,-0.27539062,-0.83984375,1.5859375,0.63671875,-1.640625,0.69921875,-0.9296875,1.7421875,-0.94140625,0.9140625,1.078125,1.0234375,-0.8828125,0.65625,-0.19921875,-0.58984375,-1.3515625,-1.421875,-0.38476562,-0.0058288574,-0.75390625,-2.0625,0.97265625,-0.13867188,-0.43164062,0.87890625,-0.45507812,0.7421875,-1.3828125,-0.9765625,-0.095214844,0.24414062,-0.93359375,-1.4453125,0.31640625,1.015625,-1.453125,0.5703125,-0.515625,-0.984375,1.3515625,-0.78125,0.5703125,-1.6171875,-0.39453125,-0.97265625,2.53125,0.09082031,-0.90625,-2.40625,1.140625,0.41796875,-1.1484375,-1.0546875,-0.4296875,-0.51171875,0.20605469,1.2421875,2.734375,1.90625,-0.625,-2.125,-0.07470703,-0.73046875,-1.0078125,-0.765625,1.0,-0.15136719,0.65234375,0.828125,-0.5390625,-0.16699219,-0.171875,-0.43164062,-0.3046875,0.33789062,-0.40820312,-0.8125,0.29492188,1.15625,1.5,0.31445312,-0.40625,0.76171875,-1.53125,1.078125,2.890625,-0.53515625,1.3515625,1.140625,-0.46484375,0.49804688,-0.13574219,-1.5390625,-0.890625,-1.0,0.58203125,-0.4375,1.0625,-0.35742188,-0.2734375,-0.04736328,-0.43554688,-0.36914062,-1.15625,0.52734375,-0.6484375,2.21875,1.7890625,0.079589844
-7,0,7,1.65625,0.68359375,-0.87890625,-1.0234375,0.41015625,-0.9375,-0.44335938,1.4765625,0.6015625,0.35351562,-1.234375,1.53125,-0.46679688,0.9296875,0.6171875,0.7890625,0.609375,-0.5546875,0.87890625,1.640625,0.36914062,-0.31054688,-0.6953125,-0.55859375,-0.8671875,1.75,-0.14941406,-0.041015625,0.3359375,-2.265625,0.7890625,2.140625,-0.20117188,0.049072266,-0.009765625,-0.81640625,0.21972656,1.9609375,0.009460449,0.118652344,-0.38476562,-0.18554688,-0.14355469,0.98046875,-0.10058594,-0.40625,-0.53125,-0.65625,-0.44140625,-0.34570312,0.11328125,-1.4375,0.17578125,-0.953125,-0.115722656,0.6640625,-0.94140625,0.13964844,0.19238281,-0.62109375,-1.0390625,-0.20605469,0.62890625,-0.45703125,0.33789062,0.095214844,0.48828125,-0.38085938,0.4140625,0.953125,-0.31054688,2.125,-0.875,-0.65625,0.734375,1.6796875,0.50390625,-0.5859375,-2.5,0.33203125,0.87890625,0.7421875,1.640625,1.609375,0.043945312,0.06542969,-0.04296875,-0.8984375,-0.859375,1.5234375,0.4921875,0.58984375,0.016113281,1.5078125,-0.0390625,0.000541687,-1.25,0.9765625,0.24023438,0.87109375,1.21875,-0.24609375,-0.4765625,-0.27929688,-0.33398438,0.609375,0.08642578,-0.58984375,-1.4296875,-0.921875,0.33789062,-0.20410156,2.09375,-0.796875,0.64453125,-0.91796875,0.16210938,-0.15625,0.31835938,0.40234375,1.125,0.025390625,-0.96875,-0.6015625,-0.44335938,0.70703125,0.040771484,0.16308594,0.99609375,-0.14160156,0.98046875,-1.9453125,0.95703125,-0.234375,-1.203125,0.55859375,-1.34375,0.3984375,0.58984375,-1.96875,0.053710938,0.16601562,-1.359375,1.5078125,-0.43164062,0.12890625,-0.83984375,2.265625,0.8125,-1.90625,1.2734375,-2.71875,0.19824219,0.09765625,-0.12597656,1.0859375,1.1015625,0.6328125,-1.0078125,0.5546875,-1.7421875,1.875,-1.0390625,0.83203125,1.890625,-0.953125,-0.40039062,-2.625,-0.027954102,1.3359375,-0.31445312,0.7265625,-0.99609375,1.1953125,0.2890625,1.2890625,0.18261719,-1.34375,-0.8828125,0.39257812,-0.31054688,0.43945312,-1.0625,0.14453125,-2.125,0.29492188,-1.6484375,-2.40625,1.34375,-1.2421875,-1.390625,-1.6640625,-2.125,0.39453125,0.059326172,0.625,-0.66015625,-0.23730469,-0.70703125,1.0546875
+6,0,5,0.13769531,0.10595703,0.27148438,-0.8828125,1.3984375,0.6171875,0.48632812,1.03125,-0.49414062,-1.0390625,0.12792969,-0.26171875,0.58203125,0.35351562,-1.2578125,0.484375,-0.546875,-1.9140625,1.1328125,-0.5234375,-0.08544922,-0.6640625,1.2734375,0.52734375,-1.234375,-1.5078125,-0.9609375,-0.8671875,-1.7890625,0.43945312,-0.63671875,-0.26171875,0.013671875,-0.4609375,-0.52734375,0.953125,0.91796875,-0.73828125,-1.4609375,1.5703125,-1.7890625,0.03149414,-0.07714844,0.09423828,0.38085938,0.5234375,0.63671875,-0.69921875,0.37304688,1.859375,-1.1953125,-1.6015625,-0.97265625,1.1015625,-2.421875,0.092285156,-0.21484375,-0.8046875,1.328125,-1.71875,1.75,0.18554688,0.2890625,0.29882812,-0.515625,-0.067871094,-1.078125,0.16308594,0.546875,1.8515625,0.06298828,-0.31640625,0.57421875,-0.13769531,-0.671875,1.8671875,0.765625,1.1796875,-1.765625,-0.5234375,1.2890625,0.89453125,-0.12207031,0.12060547,-0.640625,-0.20117188,-0.46484375,1.4609375,-0.359375,0.99609375,1.3828125,0.21289062,-0.921875,0.36914062,-1.421875,0.22265625,-0.58984375,-0.640625,0.62109375,-0.890625,-2.359375,-0.17871094,2.5,0.78125,-0.79296875,0.0063171387,0.06640625,-1.0703125,0.48632812,0.984375,0.703125,0.03564453,0.5703125,-0.79296875,0.4921875,-0.3203125,-0.47460938,-0.3515625,-0.484375,-1.1953125,0.7109375,-0.7890625,0.60546875,-1.5625,-0.2265625,1.2578125,-1.3203125,-2.421875,0.014038086,0.17285156,-0.18652344,0.03955078,-0.021972656,-1.3671875,-1.0390625,-0.13769531,-0.66015625,0.25195312,0.20800781,-1.7265625,3.375,-0.40625,1.3203125,1.1015625,-1.2265625,1.078125,-0.6015625,-0.66796875,-2.515625,1.1015625,-0.30859375,-0.03540039,1.0625,-0.48828125,-1.1015625,-1.1953125,0.3828125,1.3515625,1.015625,1.421875,-0.26367188,-0.010314941,1.828125,1.25,-1.78125,1.65625,-1.671875,-0.24707031,2.109375,-0.20117188,0.3671875,-0.40039062,0.1015625,0.6171875,0.69140625,0.040283203,-0.8671875,-0.84375,0.859375,-0.609375,-0.30664062,-0.89453125,0.140625,0.85546875,0.14648438,0.5390625,0.59375,0.038085938,0.8671875,-1.828125,0.099609375,1.375,1.2578125,0.28515625,0.43945312,-0.18261719,1.6796875,0.15820312,1.4140625,0.5234375
+6,0,6,0.82421875,0.82421875,-0.3984375,0.84375,1.40625,0.62890625,-0.3515625,0.39453125,-0.10058594,-1.4375,1.1015625,-0.19433594,0.53515625,-0.54296875,-0.99609375,-0.375,-0.8203125,-0.40039062,1.109375,1.2421875,0.20117188,-1.890625,0.36132812,0.35742188,-0.90234375,-1.625,-1.3828125,-0.35351562,0.22753906,0.05859375,-0.23828125,-0.32421875,1.7109375,0.20800781,-0.328125,-0.17089844,0.31640625,-0.6015625,-0.26367188,0.78125,-0.6015625,0.71484375,-0.26953125,1.2734375,-0.27734375,0.25585938,-0.734375,-1.0234375,0.43945312,1.71875,1.140625,-1.765625,-0.4921875,-0.26757812,-1.0859375,-2.140625,0.38671875,1.3046875,-0.111328125,-1.015625,1.6171875,1.59375,1.734375,-0.34960938,-1.03125,-0.7109375,0.050048828,-2.078125,-0.3046875,0.85546875,0.3828125,0.16992188,1.3359375,-1.4921875,0.4609375,0.51953125,0.18554688,-0.20605469,-1.7109375,0.39453125,-0.2734375,0.71484375,-0.018432617,0.31835938,-0.7578125,0.7421875,-2.828125,0.453125,1.5859375,1.2109375,1.421875,1.1953125,1.125,0.22460938,-0.6328125,-0.93359375,-0.8671875,1.2421875,-0.23046875,0.12402344,-0.54296875,-0.39453125,0.89453125,-0.114746094,-1.0390625,0.77734375,-0.98828125,-0.18164062,-0.34960938,-0.33203125,0.18164062,1.96875,0.13378906,-0.63671875,-0.57421875,0.671875,-0.4140625,0.33203125,-0.46484375,0.84375,0.05078125,-1.1015625,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.6015625,-0.24121094,0.76171875,-0.04345703,-2.515625,-0.9375,0.7421875,-1.328125,2.546875,-0.31835938,0.01373291,1.8125,-1.3203125,1.421875,-0.75390625,0.12109375,1.171875,0.6640625,-0.20117188,-0.22460938,-0.27929688,-0.5625,0.20019531,0.43359375,-0.32226562,-1.3515625,1.015625,-1.1328125,1.234375,0.6953125,-0.36523438,-0.390625,1.046875,0.78515625,1.15625,1.0078125,3.046875,-2.671875,1.609375,-2.234375,1.1875,0.16796875,-0.296875,-1.7734375,-1.8359375,0.734375,-0.65234375,0.65625,0.27539062,-0.66796875,-1.96875,1.40625,-0.35742188,0.09814453,0.15039062,0.84375,-0.11621094,-1.0234375,1.578125,-0.4921875,-0.26953125,0.51171875,-0.36523438,0.66015625,1.546875,-0.31445312,-1.0390625,-0.82421875,-0.46679688,-0.46875,-0.19824219,0.19824219,0.51953125
+6,0,7,-0.91015625,0.03955078,-0.34375,1.078125,1.359375,-0.03100586,-0.43945312,1.0546875,0.59765625,-0.66015625,0.27539062,1.84375,-0.29296875,-0.6640625,-0.97265625,0.22949219,0.0019836426,-1.046875,1.9765625,-0.4765625,-0.123535156,-2.0625,-0.8203125,0.54296875,-0.12060547,-0.62890625,0.7109375,0.73828125,0.69140625,0.41015625,0.43554688,0.2734375,1.3203125,-0.52734375,2.140625,0.88671875,1.171875,-0.828125,-1.015625,0.6640625,-0.40625,-0.80859375,-1.859375,-1.875,-0.82421875,0.89453125,-0.3984375,-0.94140625,0.8515625,1.7578125,1.125,-0.5703125,-1.15625,0.31640625,-0.123046875,-2.03125,0.35742188,1.515625,0.43359375,0.15625,0.78125,0.83203125,-0.3515625,1.2890625,-1.3125,0.5546875,-1.3359375,-1.90625,0.47460938,-0.0546875,0.81640625,1.4453125,1.78125,-1.7734375,0.8125,0.921875,-0.08251953,-0.6953125,-0.28515625,0.9296875,-1.0390625,0.57421875,-0.29101562,-0.9140625,0.04296875,-1.3671875,-0.43554688,-0.19140625,1.5625,0.859375,1.4453125,-0.13769531,-0.20214844,1.0859375,-0.31835938,-0.27539062,-0.8203125,-0.9609375,-1.03125,-0.21777344,0.08300781,0.40625,1.390625,1.1796875,-1.0703125,1.3671875,0.61328125,0.19921875,-0.049316406,0.515625,0.75390625,0.83984375,2.5,-1.59375,0.76953125,-0.35351562,-1.5234375,0.671875,-0.83984375,-0.7890625,1.3125,0.46289062,-0.67578125,-0.64453125,-1.296875,2.03125,-0.06591797,0.76953125,0.8125,-1.2734375,1.4765625,-2.3125,0.15332031,0.19628906,-0.106933594,0.90234375,0.7265625,-0.12792969,0.41015625,-0.99609375,1.5625,0.265625,-1.453125,1.3828125,-0.36523438,-1.0234375,0.49414062,-0.31640625,-0.18652344,-1.2578125,0.78515625,0.22851562,2.671875,-0.93359375,-0.8203125,0.038330078,0.13867188,-1.578125,1.1875,0.12060547,0.63671875,0.6015625,1.125,0.48046875,0.0546875,0.7265625,0.023925781,0.984375,-0.33203125,-0.0034179688,-1.1875,-1.3984375,-2.359375,-1.28125,-0.4140625,0.9921875,-0.5,-1.5234375,-1.6328125,-0.953125,0.703125,0.16015625,0.25585938,-1.234375,-0.83203125,-1.21875,-0.25,0.19238281,-1.2578125,1.078125,-1.2265625,-0.03100586,-0.87109375,-1.1953125,-1.6171875,0.58203125,-0.18945312,0.21679688,0.20117188,1.609375
+7,0,5,0.8671875,1.2109375,-0.828125,-0.984375,-1.2890625,0.32421875,-0.5,1.515625,-0.7890625,-0.70703125,1.625,0.0040893555,-0.059326172,-0.28125,-1.15625,1.6328125,-0.28710938,-1.5234375,0.15136719,-0.66015625,1.3515625,0.84765625,0.48046875,1.125,0.8984375,-0.859375,-0.93359375,-1.9375,-1.6171875,-0.63671875,-1.2578125,2.453125,1.015625,2.515625,-0.171875,0.14648438,-0.40234375,-0.59765625,0.8671875,0.5625,-0.625,0.28125,0.63671875,0.20117188,1.3046875,0.36328125,0.578125,0.2734375,0.65625,-1.1796875,-0.84375,-0.51171875,-1.8984375,-1.609375,-1.03125,0.59765625,0.08300781,-0.33984375,0.3125,-0.75390625,-0.19140625,-0.515625,1.8515625,1.0,1.1796875,0.70703125,-0.88671875,-0.13867188,0.22558594,0.24902344,-0.036865234,-0.36132812,-0.671875,-0.1875,-0.4765625,-0.15820312,-0.23339844,-0.13183594,-0.01550293,0.45898438,-0.48242188,-0.546875,1.1328125,0.42382812,1.1171875,1.2890625,-0.97265625,0.52734375,1.375,-1.0234375,1.3359375,0.41015625,0.546875,1.21875,0.21972656,-1.3203125,0.76953125,0.99609375,-1.703125,-0.296875,-0.6484375,0.32421875,2.21875,-1.7734375,-1.21875,-1.4765625,-2.015625,-1.7109375,-0.47851562,0.9296875,0.20996094,-1.0078125,0.088378906,0.5703125,0.53125,0.20117188,0.0040283203,-0.98046875,-0.26757812,0.22851562,-0.36914062,-0.3125,-1.4375,-0.99609375,-1.4609375,1.0625,-1.75,-0.55078125,1.0390625,0.984375,0.9140625,-0.484375,-1.140625,-0.6484375,-1.765625,0.8203125,1.109375,-0.1875,-1.6640625,-2.4375,1.484375,-0.55078125,-0.640625,0.59375,0.4765625,0.18652344,-0.25,-0.5234375,0.72265625,1.5390625,-0.75390625,0.2734375,1.2734375,-1.21875,0.6796875,0.30859375,0.59375,-0.859375,0.7734375,-1.1484375,0.80078125,-0.40625,0.64453125,-0.64453125,-1.1328125,3.453125,-0.76171875,-1.1796875,0.328125,-1.015625,-1.0,-0.47070312,-1.2890625,-1.1171875,-0.38085938,1.1796875,0.7890625,-0.72265625,1.046875,-0.49804688,0.390625,0.34960938,0.73046875,-1.2890625,1.1484375,-0.053955078,0.08984375,0.28125,0.24609375,0.38085938,2.03125,-0.14257812,-0.033447266,-0.7265625,-0.0119018555,0.8359375,2.0625,1.796875,0.8515625,0.66796875
+7,0,6,0.5859375,1.3984375,-0.61328125,-0.51953125,-1.9140625,-0.7578125,1.0234375,0.9765625,0.18945312,0.011169434,0.115234375,0.43945312,-0.020263672,-0.828125,-0.51171875,1.8046875,-1.2578125,-0.82421875,0.34960938,0.27734375,0.45703125,-0.60546875,1.046875,0.3671875,1.28125,-0.51171875,-1.984375,-0.71875,0.5,-0.62890625,-0.73828125,-0.008300781,2.21875,-0.43554688,-0.625,-0.076660156,0.25976562,-1.28125,-0.51953125,-0.6796875,-0.4296875,0.18847656,0.29882812,-0.09814453,-0.73046875,0.15429688,0.37890625,0.66015625,-0.28125,-0.20117188,0.89453125,0.546875,-3.75,-0.15429688,1.15625,1.71875,-0.14941406,0.34179688,0.47265625,-0.095703125,-1.046875,-1.1171875,1.0390625,0.33007812,1.6015625,-0.375,-0.234375,-0.34765625,0.040771484,0.35351562,-0.32421875,1.125,-0.04248047,-0.98828125,2.453125,0.83203125,0.88671875,-0.12988281,-1.4375,0.34375,0.59375,-0.28125,1.53125,0.125,0.37304688,0.0019226074,-1.203125,1.1328125,1.6015625,-0.24707031,0.875,-0.3515625,0.99609375,0.5234375,-1.2734375,0.57421875,-0.83984375,1.828125,-1.3984375,0.35742188,0.33984375,0.97265625,-0.5625,0.71875,-0.62109375,-0.2109375,-1.6875,-1.75,-0.29101562,0.33398438,-1.46875,-2.296875,0.3046875,-0.20117188,-1.4140625,1.15625,-0.671875,1.03125,-1.140625,-0.546875,-0.12988281,0.41210938,-1.203125,-1.3359375,0.49023438,0.76171875,-1.625,-0.09667969,-0.18652344,-1.125,1.140625,-0.23144531,-0.11035156,-2.171875,-0.76171875,-0.66796875,1.6484375,0.048339844,-1.4296875,-2.796875,1.4453125,0.15136719,-1.390625,-1.21875,-0.328125,-0.44921875,0.40039062,0.77734375,2.25,1.96875,-0.5390625,-1.0703125,0.17382812,-0.31835938,-0.23925781,-1.2734375,1.234375,0.084472656,0.9921875,0.099121094,-0.27539062,-0.071777344,0.22949219,-0.40625,-1.15625,0.625,-0.515625,-1.375,0.57421875,0.46679688,0.84765625,0.61328125,-0.80078125,0.80859375,-1.59375,1.1328125,2.75,-0.35351562,1.6796875,1.8984375,-0.25195312,0.6484375,0.62109375,-1.0234375,0.045898438,-1.046875,0.4921875,-0.08691406,0.83984375,-0.041259766,0.18652344,0.96875,-0.012329102,0.3125,-1.09375,-0.20605469,0.014526367,2.0,1.0703125,0.028442383
+7,0,7,0.96484375,0.67578125,-1.0078125,-0.94921875,0.052734375,-0.69921875,-0.15136719,1.9921875,0.83203125,0.625,-1.328125,1.8359375,-0.36328125,1.1484375,0.020385742,0.7734375,0.62109375,-0.359375,0.87109375,1.421875,0.23535156,-0.052490234,-0.75390625,-0.328125,-0.6171875,1.3828125,-0.6484375,0.028320312,0.8984375,-2.203125,0.5625,1.7890625,1.1796875,0.16015625,-0.59765625,-0.41601562,0.22558594,1.4609375,0.09033203,-0.10644531,-0.984375,-0.07714844,-0.53515625,0.49414062,0.265625,0.049804688,-0.23828125,-0.33398438,0.06640625,0.17578125,0.099609375,-1.3828125,-0.359375,-0.7421875,0.068847656,0.50390625,-1.6484375,0.15234375,0.28125,-0.63671875,-1.1484375,-0.68359375,1.6171875,-0.34179688,0.45117188,0.55859375,0.25585938,-0.18066406,-0.29882812,1.0390625,0.21972656,2.109375,-0.87109375,-0.7421875,0.8671875,1.5859375,0.16894531,-0.81640625,-2.859375,0.7109375,1.609375,0.578125,1.84375,2.0625,-0.66796875,0.8046875,-0.6796875,-0.65234375,-0.5625,1.375,1.1484375,1.0,-0.16894531,1.2890625,0.021972656,0.111328125,-1.6328125,1.03125,-0.13085938,0.453125,0.66796875,-0.50390625,-0.48828125,-0.34570312,-0.6015625,0.94140625,-0.34570312,-0.91015625,-1.71875,-0.87109375,-0.17871094,-0.47070312,1.5859375,-0.83203125,-0.15234375,-1.0859375,0.21191406,0.19140625,0.8984375,0.80859375,1.2265625,0.06738281,-1.0703125,-0.77734375,-0.47070312,0.734375,-0.17871094,-0.234375,1.2265625,-0.5390625,0.8984375,-1.859375,0.29101562,-1.1640625,-1.359375,0.7734375,-1.578125,0.16699219,0.22167969,-2.0625,0.072265625,-0.13183594,-1.359375,1.2421875,-0.24023438,0.06738281,-0.53515625,1.71875,0.22167969,-1.875,1.5703125,-2.09375,0.34765625,0.20410156,0.1484375,0.46875,1.4140625,0.67578125,-0.54296875,0.23144531,-1.5703125,2.25,-0.83203125,0.796875,1.1640625,-0.78515625,-0.484375,-2.859375,0.23730469,1.2421875,-0.55859375,0.7421875,-1.1171875,1.3203125,0.09326172,1.703125,0.22558594,-1.4921875,-0.765625,0.8203125,-0.609375,0.51953125,-0.41992188,0.47460938,-1.7109375,0.29882812,-1.6953125,-2.046875,1.484375,-0.98046875,-1.5703125,-0.92578125,-1.5390625,0.48046875,0.072265625,0.06640625,0.24023438,-0.37890625,-0.85546875,1.265625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
index 28d6a225..3da2498e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,5,1.7890625,0.8515625,0.26367188,-0.18652344,-0.059570312,-0.56640625,-1.6015625,0.59375,-0.08984375,0.57421875,0.33789062,0.31445312,-0.14648438,-0.13574219,-0.07714844,-0.35351562,-1.2109375,-2.171875,0.78515625,0.74609375,1.9453125,-1.28125,-1.265625,-0.16308594,1.609375,-0.068359375,-0.010253906,-0.359375,-0.515625,-0.36328125,-1.2109375,0.22558594,0.65234375,-0.60546875,0.52734375,-1.5234375,0.31640625,-0.52734375,-2.140625,-0.005493164,1.4609375,-0.8359375,-0.59375,0.43164062,-1.7734375,-0.20800781,-0.359375,0.34179688,1.953125,0.328125,0.14746094,-0.33984375,-1.484375,1.1171875,-0.78515625,0.8125,0.44726562,1.390625,0.7890625,-0.640625,-1.125,-1.0390625,-0.087402344,1.1875,1.2109375,0.4140625,-0.41992188,-2.0625,1.6796875,0.765625,0.28125,1.328125,-0.79296875,-1.109375,2.734375,1.625,-0.16113281,0.22949219,-0.35351562,-0.067871094,-1.0390625,0.3984375,-0.012390137,1.078125,0.27734375,-1.5390625,0.049804688,2.453125,0.109375,1.2734375,-0.296875,0.21386719,1.7578125,-0.41601562,-0.9921875,-0.30664062,0.70703125,1.46875,0.578125,-0.19824219,1.4609375,1.046875,-0.91015625,0.15527344,0.27734375,-0.08642578,-0.8125,-0.6796875,0.14453125,0.4609375,-0.24023438,0.10253906,0.008056641,-1.703125,-0.92578125,0.51171875,-0.875,0.58203125,-1.5625,0.203125,0.21582031,0.38476562,-2.171875,-0.984375,-1.1875,0.625,-0.36914062,1.0859375,0.42773438,-0.19824219,1.984375,-0.98828125,0.58984375,-0.546875,-0.65234375,0.23535156,1.015625,0.18164062,-0.21875,-1.953125,1.3515625,2.703125,-0.33007812,-0.671875,-1.484375,0.8203125,0.21777344,0.86328125,1.4921875,0.96484375,0.40039062,-0.24414062,0.95703125,-0.69921875,-1.7890625,-0.38085938,-0.045410156,-0.7890625,0.5703125,-0.20214844,0.37890625,0.33789062,-0.41210938,-1.2265625,1.7265625,0.119140625,0.29101562,-0.18261719,-1.828125,1.734375,0.32226562,0.8359375,-0.703125,-0.9140625,-1.2578125,1.8203125,0.25,-0.578125,0.35742188,-0.8828125,1.515625,0.734375,-0.19628906,-1.765625,-1.671875,-0.96484375,-0.21289062,-0.48046875,0.7890625,-0.14941406,-0.61328125,-1.421875,-0.49023438,-1.2890625,-0.16992188,-0.65625,-1.7109375,1.203125,1.2421875,0.9375
-8,0,6,1.46875,1.421875,0.018798828,0.24511719,0.045898438,-0.7265625,0.734375,0.81640625,1.0546875,-0.8203125,1.5703125,1.234375,-0.8359375,-0.89453125,0.21972656,0.265625,-0.671875,-0.91796875,0.71875,0.515625,-0.05517578,-0.74609375,-0.20214844,1.4765625,-0.029174805,0.58203125,-2.265625,-0.14355469,-0.2578125,0.51953125,-0.84375,-0.21972656,0.4375,-0.796875,1.2734375,0.09326172,0.6328125,-0.38867188,0.025390625,1.40625,-0.45117188,1.0859375,-0.17675781,-1.90625,-1.234375,-1.03125,0.1953125,-0.53515625,1.390625,0.65234375,0.734375,-0.7421875,-0.703125,0.61328125,0.3828125,0.48632812,0.7890625,-1.046875,-1.1796875,0.064453125,1.4453125,0.6484375,-0.18847656,0.96484375,0.44140625,-1.3125,0.51953125,0.21972656,-0.8203125,0.36914062,-0.5703125,0.6484375,0.94140625,-0.62109375,1.703125,0.703125,1.6171875,1.3125,-0.125,0.20117188,-0.4921875,-1.03125,0.7109375,-2.234375,0.31445312,0.020019531,0.46679688,-0.37890625,1.25,0.21484375,0.625,-0.76171875,-0.80078125,1.171875,-1.2109375,-0.65625,-2.71875,0.93359375,-0.73046875,-0.016357422,-0.8125,1.4375,0.13476562,0.091308594,-0.27929688,1.3828125,-0.49804688,0.46289062,-1.1328125,0.32617188,-0.20117188,-0.19433594,1.390625,-0.111816406,1.078125,2.03125,-0.5546875,-0.09765625,-0.46289062,-0.765625,0.328125,-0.78125,-0.35546875,-0.87109375,-1.6796875,2.5,-1.0859375,-0.68359375,0.33007812,-0.890625,1.5234375,-1.546875,-0.16894531,0.029541016,-1.53125,0.45507812,1.4140625,-0.29296875,0.22167969,-2.078125,1.96875,-0.765625,-1.6875,1.625,0.6875,-0.296875,0.83984375,-0.28515625,-0.25585938,1.078125,-1.5078125,-1.546875,0.92578125,-1.421875,0.5625,1.171875,1.4375,-1.3203125,0.7578125,0.092285156,-0.20605469,1.546875,2.0625,1.71875,-0.74609375,0.17480469,-0.56640625,0.34179688,0.31640625,0.23730469,-0.8203125,-0.0859375,-1.890625,-0.3203125,-1.3203125,0.94140625,1.3125,-0.96875,0.44726562,-0.14941406,-1.1328125,-1.4296875,-0.33203125,-1.859375,-0.8046875,0.33007812,-0.22753906,0.20117188,-1.4296875,-2.015625,-0.44335938,1.484375,-1.2890625,-1.1953125,-1.3046875,2.15625,0.875,0.9453125,-0.375,-0.4375
-8,0,7,1.640625,1.9765625,0.36132812,0.13574219,-1.796875,-0.22851562,-1.4921875,1.5546875,1.421875,-2.71875,2.03125,-0.84765625,-0.49609375,1.3125,-0.22363281,0.5234375,0.36523438,0.37304688,1.71875,0.08935547,-0.72265625,-0.111328125,-0.54296875,-0.234375,-0.703125,-0.5859375,-0.35742188,-0.31835938,-0.19824219,-0.6171875,-0.4765625,0.46679688,0.5703125,0.75,0.80078125,-0.14648438,1.7265625,0.68359375,-0.62109375,1.8515625,-1.578125,-0.22265625,0.19042969,2.328125,0.033447266,0.62890625,-0.71875,0.453125,0.052734375,-0.037841797,-0.56640625,-0.14257812,-0.22851562,-0.359375,-0.53125,-0.16796875,0.7265625,0.5546875,-1.1953125,0.33789062,0.38867188,0.875,1.1484375,-0.33984375,1.0703125,-0.11035156,0.3203125,-1.640625,-0.60546875,-0.15039062,-0.44921875,-0.27148438,0.38476562,-0.14746094,-0.94921875,0.734375,0.14453125,0.35546875,-1.6484375,1.2265625,-1.125,-0.75390625,1.078125,-1.8125,0.23242188,-0.12207031,0.48632812,-1.546875,-0.28515625,0.47265625,1.34375,1.53125,-0.12988281,1.0,-0.89453125,-0.23535156,-0.63671875,0.31835938,-0.04248047,-0.16503906,-1.875,0.83984375,1.5,-2.171875,0.53515625,1.3828125,-1.0390625,-0.12695312,-0.107910156,-0.0014190674,0.2265625,-0.15820312,1.6640625,-0.42578125,0.53515625,0.20214844,-0.51171875,-0.17089844,-0.48632812,0.65234375,1.3359375,-0.828125,-0.0040893555,-1.59375,-0.9765625,0.48632812,0.5390625,-1.328125,-0.421875,1.6953125,-0.67578125,1.0234375,-0.9921875,1.0859375,-2.21875,1.390625,0.98828125,-0.4609375,-0.66796875,-2.046875,1.5703125,-1.125,-0.9453125,0.09326172,-0.16699219,0.031982422,0.22851562,0.0035095215,0.0031433105,1.28125,-1.1875,-0.33789062,0.98046875,-0.375,-0.9453125,0.5625,1.0546875,-0.052734375,0.3359375,0.20703125,-0.71875,0.15820312,0.14355469,0.33007812,-0.41015625,0.8125,-0.7421875,0.7265625,0.5390625,-1.9609375,-1.984375,-1.0859375,-2.328125,-0.75,0.5625,2.296875,0.578125,-0.23632812,0.032470703,-0.76171875,0.69140625,-0.19628906,-0.8671875,-0.4140625,-0.734375,0.22460938,0.66796875,1.2109375,-0.23046875,-0.14550781,-0.18359375,-0.53515625,0.22753906,-1.546875,-3.125,2.109375,3.0625,0.58984375,-0.25195312,0.73046875
+8,0,5,1.0703125,0.25,-0.16601562,-0.29296875,0.10986328,-0.42382812,-1.28125,0.7890625,0.23828125,0.46875,0.1328125,0.921875,1.2421875,-0.13574219,-0.56640625,-0.12792969,-1.078125,-2.234375,0.93359375,0.21386719,1.78125,-1.03125,-1.2890625,-0.20898438,1.4140625,-0.30078125,-0.71484375,-0.010803223,0.25585938,-0.14355469,-0.84375,-0.61328125,1.953125,-0.94140625,-0.9296875,-0.41601562,0.09277344,-1.140625,-2.59375,-0.46484375,0.4453125,-0.87890625,-0.70703125,-0.29882812,-1.2265625,0.5234375,-0.31835938,0.50390625,1.7578125,1.4375,-0.27148438,0.359375,-1.6640625,1.2578125,-0.103515625,0.38085938,-0.5,1.8828125,0.7109375,-0.82421875,-0.69140625,-1.9453125,0.76953125,0.80078125,1.328125,1.46875,-1.4609375,-1.859375,0.41601562,0.94140625,1.421875,1.21875,-0.41601562,-0.94921875,2.90625,1.78125,-0.35546875,-0.111816406,-0.49804688,0.27148438,-0.34179688,0.26953125,-0.0005874634,2.359375,-0.48046875,-0.58984375,-0.4140625,2.5,1.0546875,0.92578125,0.8125,0.54296875,0.81640625,0.06689453,-0.84375,-0.13574219,0.5546875,1.0390625,0.51171875,-0.94140625,0.19335938,1.0390625,-0.46875,0.36914062,-0.16503906,0.76953125,-1.171875,-1.140625,0.37695312,0.33203125,-1.015625,-0.15625,-0.55859375,-1.9453125,-1.78125,0.103515625,-0.625,1.2109375,-1.265625,0.4453125,0.61328125,0.41210938,-2.21875,-1.359375,-0.8046875,0.18652344,-0.5625,0.43359375,0.6640625,-0.6328125,1.7734375,-0.25,0.14355469,-1.5,-0.65625,0.5703125,0.5546875,0.19335938,-0.74609375,-2.5,1.078125,2.578125,-0.71484375,-1.21875,-1.265625,1.21875,0.32421875,-0.15722656,0.32226562,0.60546875,0.55859375,0.98046875,0.6328125,-0.6171875,-0.78125,-1.453125,0.28710938,-0.05053711,1.171875,-0.65625,0.36132812,0.57421875,0.075683594,-1.0625,0.5859375,-0.28515625,0.578125,-0.74609375,-1.28125,1.265625,0.14160156,0.65234375,-0.90234375,-0.53515625,-1.7109375,1.84375,-0.17773438,-0.5859375,0.079589844,-0.484375,1.25,0.40429688,0.8203125,-0.6796875,-1.0703125,-1.5625,0.18847656,0.36328125,0.69140625,0.625,-0.29882812,0.3671875,0.80078125,-0.93359375,-0.640625,-1.8515625,-0.6953125,1.2578125,1.0234375,0.7265625
+8,0,6,0.87109375,0.9921875,-0.1796875,0.12597656,0.08154297,-0.609375,0.7578125,1.0703125,1.0703125,-0.796875,1.328125,1.546875,0.051757812,-1.03125,-0.14355469,0.296875,-0.6796875,-1.0078125,0.83984375,0.33398438,0.026855469,-0.36523438,-0.24609375,1.625,-0.18261719,0.14550781,-2.78125,0.21289062,0.3125,0.7890625,-0.5078125,-0.8984375,1.3671875,-0.7734375,0.578125,0.59375,0.53125,-0.94140625,-0.22167969,1.078125,-1.203125,1.1171875,-0.33984375,-2.40625,-0.8671875,-0.78515625,0.26171875,-0.22851562,1.296875,1.6015625,0.55859375,-0.47460938,-0.92578125,0.67578125,0.47460938,0.265625,0.19921875,-0.515625,-1.0625,-0.033935547,1.46875,0.2109375,0.6328125,0.859375,0.50390625,-0.43359375,-0.14160156,0.5625,-1.359375,0.67578125,-0.099121094,0.54296875,0.84765625,-0.42578125,1.8359375,0.8046875,1.390625,1.0546875,-0.35351562,0.37304688,-0.083984375,-1.1328125,0.83984375,-1.4453125,-0.45507812,0.36523438,-0.03149414,-0.036132812,1.7578125,0.19042969,1.3515625,-0.5390625,-1.03125,1.25,-1.1015625,-0.59375,-2.734375,0.49414062,-0.59765625,-0.3671875,-1.4609375,1.1640625,0.37890625,0.19628906,-0.59765625,1.8359375,-0.75390625,0.16796875,-1.15625,0.30859375,-0.56640625,-0.42773438,0.91796875,-0.25976562,0.36328125,1.765625,-0.5078125,0.5546875,-0.16015625,-0.44335938,0.578125,-0.625,-0.49023438,-0.9375,-1.328125,2.109375,-1.265625,-1.046875,0.71484375,-1.28125,1.3203125,-1.21875,-0.6875,-0.76953125,-1.6015625,0.6484375,1.09375,-0.52734375,0.05908203,-2.46875,1.703125,-0.94921875,-1.765625,1.375,0.7109375,-0.04321289,0.8671875,-0.89453125,-1.1640625,0.9140625,-1.2890625,-0.703125,0.68359375,-1.015625,1.1171875,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.06933594,1.65625,2.453125,1.453125,-1.5234375,0.099121094,-0.54296875,0.08496094,0.859375,-0.15527344,-1.015625,-0.29492188,-1.640625,-0.14941406,-1.5390625,1.0390625,0.90625,-1.1640625,0.29492188,0.049804688,-1.2734375,-1.40625,0.53515625,-1.0,-0.45507812,0.24609375,0.021606445,0.56640625,-1.6171875,-1.6328125,-0.23535156,2.46875,-0.36914062,-0.97265625,-1.5625,1.03125,1.4921875,1.09375,-0.41992188,-0.59375
+8,0,7,1.2578125,1.5546875,0.1015625,-0.2578125,-1.5859375,-0.019165039,-1.328125,1.4375,1.4140625,-2.65625,1.7578125,-0.69140625,0.03100586,1.1171875,-0.4921875,0.46875,0.2890625,0.40625,1.859375,-0.015991211,-0.8046875,-0.036132812,-0.390625,-0.25,-0.86328125,-1.0625,-0.71484375,-0.22070312,0.099121094,-0.20605469,-0.25195312,0.016601562,1.0546875,0.71484375,0.09716797,0.29101562,1.40625,0.14648438,-0.88671875,1.6953125,-2.3125,-0.22265625,0.114746094,2.140625,0.25976562,0.91796875,-0.87109375,0.7109375,-0.061767578,0.88671875,-0.64453125,-0.08984375,-0.3515625,-0.24902344,-0.41601562,-0.45507812,0.19433594,0.80859375,-1.1328125,0.16601562,0.546875,0.65625,1.65625,-0.40820312,1.3671875,0.43164062,-0.36328125,-1.265625,-0.98046875,0.079589844,-0.27148438,-0.23730469,0.6171875,0.037841797,-0.91015625,0.734375,0.14160156,0.34375,-1.53125,1.3203125,-0.81640625,-0.66015625,0.953125,-1.2265625,-0.18066406,0.33398438,0.24511719,-1.3046875,0.103515625,0.51171875,1.9453125,1.5859375,-0.35546875,1.0546875,-0.89453125,-0.14648438,-0.68359375,0.004760742,0.19335938,-0.703125,-2.390625,0.87890625,1.7265625,-1.7421875,0.03515625,1.546875,-1.1171875,-0.46289062,0.0546875,0.20410156,-0.08935547,-0.3125,1.3359375,-0.73046875,0.06542969,0.080566406,-0.3828125,0.24902344,-0.25390625,0.859375,1.6484375,-0.56640625,-0.15429688,-1.796875,-0.515625,0.20117188,0.39257812,-1.5625,-0.11816406,1.390625,-0.890625,1.109375,-1.3984375,0.6484375,-2.078125,1.515625,0.90625,-0.59765625,-0.7109375,-2.328125,1.5859375,-1.1640625,-0.76171875,0.033691406,-0.19140625,0.20507812,0.48046875,-0.59375,-0.7578125,1.234375,-1.3984375,0.296875,0.58203125,-0.119628906,-0.31835938,0.09667969,1.1953125,0.34570312,0.52734375,0.037353516,-0.67578125,0.16699219,0.4375,0.13085938,-1.2578125,0.83984375,-0.6015625,0.6796875,0.66796875,-2.015625,-2.109375,-1.3359375,-2.171875,-0.49023438,0.34765625,2.296875,0.36914062,-0.52734375,0.037597656,-0.4453125,0.64453125,-0.27929688,-0.2578125,0.20898438,-0.38476562,0.14160156,0.91796875,1.4375,-0.47265625,0.045410156,0.080566406,0.25976562,0.78125,-1.4140625,-3.125,1.2578125,3.359375,0.64453125,-0.44726562,0.49609375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
index 4477b4bf..5c971edd 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,5,1.7421875,0.86328125,0.08642578,0.15820312,-0.040771484,-0.63671875,0.12792969,1.234375,0.028320312,-2.0,1.1875,0.027832031,-1.6875,-0.94921875,0.26953125,0.6015625,-1.03125,0.66796875,-0.765625,0.12792969,-0.055419922,0.2890625,-0.49804688,-0.484375,-0.30664062,-0.3125,-1.6875,0.110839844,-1.2265625,-0.2265625,-1.5390625,-0.38867188,-1.2421875,0.58984375,-0.061035156,-0.70703125,0.6015625,-0.100097656,0.34570312,0.9609375,0.8046875,0.328125,1.0859375,2.4375,-0.20410156,-0.33007812,-0.09033203,0.026245117,-2.28125,-0.984375,0.48242188,-1.203125,-1.53125,-0.26171875,-1.1171875,0.032226562,1.4609375,1.3046875,-0.7578125,-1.5546875,-0.27734375,1.453125,-0.9140625,0.33984375,0.984375,-0.46875,0.66796875,-0.63671875,0.74609375,0.21972656,-0.40429688,-1.7578125,-0.018676758,-1.15625,-0.053222656,1.4453125,2.109375,-0.07470703,0.27148438,-0.050048828,-0.24707031,0.68359375,-0.078125,-1.703125,0.9765625,0.640625,0.6015625,-2.0,-0.07421875,0.84765625,0.12060547,0.94921875,1.0546875,0.953125,-0.890625,-0.5390625,-1.046875,0.35546875,0.30664062,0.86328125,2.109375,0.41992188,-1.0703125,-0.53515625,0.9765625,-1.0859375,-0.484375,-0.36523438,0.73046875,-0.671875,-0.44140625,1.96875,0.41992188,1.0234375,0.63671875,0.47265625,-0.068359375,0.16503906,-0.51953125,0.875,0.92578125,-1.6953125,-0.74609375,-0.43554688,0.20117188,0.28320312,0.29492188,-0.71484375,-1.515625,-0.009338379,0.76953125,-1.0390625,-0.1640625,1.4453125,-0.64453125,1.0546875,0.12597656,-0.40039062,0.9375,-2.34375,1.6328125,-0.73046875,-1.2890625,0.15332031,1.59375,0.39453125,-1.0234375,0.76171875,2.90625,3.0,-0.89453125,-1.6015625,-2.421875,1.65625,-0.5078125,1.171875,0.0107421875,0.5625,-0.8359375,0.36132812,-0.94140625,0.828125,0.23046875,1.5625,-0.9609375,0.25195312,-0.82421875,0.37304688,0.56640625,-0.98046875,0.23339844,-1.265625,1.640625,-0.86328125,0.026611328,-0.091308594,0.55078125,-0.16699219,0.890625,-0.6875,-0.24023438,0.390625,0.58203125,-0.43945312,-0.34375,1.0703125,0.578125,0.36328125,1.5078125,-2.296875,-0.3203125,-0.671875,-0.24804688,-1.7578125,-1.265625,2.09375,0.09326172,-0.671875,-0.21484375,-0.4140625
-9,0,6,1.1328125,1.78125,0.05908203,-0.7734375,0.39453125,-0.114746094,0.078125,-1.1640625,2.09375,-1.5390625,2.453125,-1.359375,-1.0234375,0.3359375,0.5703125,-0.26171875,-0.88671875,-0.73046875,0.6015625,0.29101562,0.28320312,-0.859375,0.9296875,-0.28710938,-1.1015625,-1.28125,0.2578125,0.033447266,-0.8203125,0.43554688,0.43945312,0.041992188,0.087890625,1.015625,0.65625,-0.37890625,0.9453125,1.9140625,-0.4296875,0.35546875,0.65234375,0.07861328,1.015625,1.28125,-1.3046875,0.17480469,0.1875,-1.4140625,-1.515625,0.6328125,0.40234375,-0.49414062,-0.05517578,0.49609375,0.20214844,-0.32421875,0.84375,1.0078125,0.08691406,-0.32226562,0.83203125,2.09375,-0.84765625,0.87109375,1.2421875,-0.15332031,-0.27539062,-1.125,0.84375,0.875,-1.484375,0.5703125,0.052490234,-0.73046875,-0.44921875,-0.08935547,-0.16601562,-0.67578125,-0.2578125,1.0,0.053955078,0.23535156,0.57421875,1.625,1.0625,-0.6640625,-1.7109375,-0.9609375,0.8828125,-0.20996094,1.6875,-0.62890625,1.5859375,-0.111328125,-0.36914062,-0.19140625,-1.0625,0.47265625,-0.86328125,-1.078125,0.13476562,-0.70703125,1.46875,-0.5859375,0.10498047,0.15039062,-1.5390625,0.55859375,-0.0859375,-0.45507812,0.19628906,2.296875,-0.46484375,-2.0,0.14941406,0.8359375,-1.265625,-0.17773438,-0.36328125,-0.38085938,0.075683594,-1.3125,-0.26757812,-1.578125,-1.71875,2.421875,-1.5546875,-0.27539062,-0.66796875,-0.22851562,1.1796875,-0.032958984,-0.32421875,0.60546875,-1.40625,2.03125,-0.65234375,1.109375,1.046875,-0.8671875,1.359375,0.30859375,-0.55078125,2.234375,0.15234375,-1.5703125,-0.3125,0.029663086,-0.3359375,0.22753906,-0.6796875,-1.5390625,-0.1796875,0.0045166016,-1.140625,0.66015625,0.58984375,-0.8203125,-1.09375,2.15625,-0.28515625,-0.546875,1.3203125,1.6875,-0.94140625,1.890625,-1.0546875,1.4140625,-0.4921875,-0.076171875,-1.3203125,0.107910156,-0.26757812,-1.390625,1.0703125,0.1484375,0.14648438,-2.40625,-0.66015625,-2.03125,1.5234375,0.55078125,-0.48632812,0.1953125,-1.4765625,1.9140625,-1.734375,-1.28125,-0.296875,0.68359375,0.84765625,0.265625,-0.62890625,0.30664062,-0.022216797,-1.59375,-0.73828125,-0.08886719,1.5234375,0.953125
-9,0,7,0.546875,0.875,0.71484375,-2.0625,-0.20117188,0.33203125,1.6484375,-0.9140625,1.4296875,-0.85546875,1.375,-1.3515625,0.859375,-0.30664062,0.2578125,0.890625,-0.08251953,-1.1015625,2.09375,-1.171875,-0.91015625,-1.328125,0.96484375,-0.3828125,-2.203125,-1.1953125,0.83203125,-1.0234375,-1.078125,1.2578125,0.37695312,-0.57421875,-0.53125,0.34570312,1.6796875,0.55078125,0.21582031,0.54296875,-1.7890625,1.4765625,-0.9453125,0.18945312,1.0,1.34375,0.6796875,0.70703125,-0.83984375,-0.88671875,-0.75,0.37695312,-1.8984375,0.703125,-0.8828125,1.8046875,-0.6875,0.6171875,0.90625,0.17480469,-0.27929688,-0.41796875,0.31054688,0.20214844,-1.671875,0.875,0.51171875,1.0234375,-0.29882812,-1.40625,-0.35351562,2.125,-1.40625,-0.7890625,-0.29492188,0.6875,-0.4609375,1.109375,1.7421875,0.0859375,0.03100586,-0.052490234,0.40625,0.8359375,0.28320312,0.7109375,0.59375,-1.40625,-0.20703125,-1.5234375,-0.46875,-0.5625,0.41796875,-0.24121094,-0.78125,-0.58203125,-1.765625,0.40039062,-0.546875,-0.34765625,-0.50390625,-0.30859375,-0.796875,-0.029785156,2.40625,1.171875,0.94921875,1.0390625,-0.609375,-0.022583008,0.4609375,-0.80859375,0.890625,1.2734375,1.0859375,-2.203125,1.15625,-0.27734375,0.12792969,-0.48046875,-0.73828125,0.011169434,1.421875,-0.41015625,1.234375,-1.703125,0.47460938,0.33007812,-2.4375,-1.578125,-0.67578125,0.703125,0.8046875,0.40820312,0.8125,-1.2578125,0.734375,-0.08300781,-0.40820312,0.9375,0.15332031,-0.671875,2.453125,1.96875,-0.984375,1.6015625,-0.83203125,-0.75,0.1015625,0.6953125,-1.7890625,0.609375,-0.53125,-1.015625,0.22851562,-1.25,-1.796875,-0.34179688,0.68359375,-0.67578125,0.029663086,2.375,-1.8125,-0.24316406,1.296875,0.52734375,-0.57421875,-0.16992188,-0.359375,0.98046875,-0.3046875,0.65625,0.36523438,0.65625,-1.21875,-0.66015625,0.49414062,-0.22460938,1.0078125,-1.1015625,-0.25976562,-0.69921875,-0.40234375,-1.6015625,0.3203125,0.28320312,-1.109375,-0.018920898,0.09863281,-0.890625,-0.76171875,-0.020874023,-0.34570312,0.765625,0.13183594,0.0146484375,0.26953125,-0.15625,1.046875,-0.85546875,2.546875,0.60546875
+9,0,5,0.78515625,1.0,-0.059570312,0.65234375,0.36132812,-0.49804688,0.3359375,1.0859375,-0.15917969,-1.0625,0.72265625,-0.25,-1.125,-0.859375,-1.0546875,0.1484375,-1.0859375,1.375,-0.43554688,0.30078125,-0.025146484,0.79296875,-0.33203125,-0.13964844,-0.609375,-1.4765625,-2.765625,0.3125,-0.63671875,0.51171875,-1.046875,-1.1171875,-0.38476562,0.21972656,-0.63671875,-0.8984375,0.515625,-1.2578125,0.63671875,-0.103515625,-0.7578125,0.43945312,0.40039062,1.59375,0.29882812,0.69921875,0.05517578,0.29492188,-2.140625,0.8125,1.28125,-1.2890625,-2.296875,-0.14355469,-0.92578125,-0.44726562,0.078125,1.0078125,-0.2734375,-0.8125,-0.07763672,1.265625,-0.07373047,-0.17089844,1.3046875,-0.5859375,0.33789062,-0.0017471313,0.22070312,0.94140625,0.18164062,-2.015625,0.8984375,-1.1171875,-0.1328125,1.265625,1.765625,-0.13085938,0.22851562,-0.3671875,0.90234375,0.47265625,-0.17480469,-1.1015625,-0.4609375,1.0078125,0.07128906,-1.3203125,0.14355469,0.60546875,0.62109375,1.6796875,0.8828125,0.59765625,-0.27734375,-0.20898438,-1.09375,0.80078125,0.12207031,0.60546875,1.6796875,0.67578125,-0.90625,-0.21289062,0.50390625,-0.35546875,-0.94921875,-0.52734375,0.42578125,-0.1640625,-1.59375,2.015625,0.07763672,0.5,-0.36328125,0.16796875,-0.47851562,1.578125,-0.30078125,1.4453125,1.4609375,-1.296875,-0.77734375,-1.03125,0.02319336,0.20605469,0.33789062,-1.0546875,-2.203125,0.18457031,0.22265625,-0.76953125,-0.85546875,0.21289062,0.2421875,1.0,-0.44335938,-1.109375,1.2421875,-1.90625,1.3515625,-0.5390625,-0.43945312,-0.1171875,1.1796875,0.33789062,-0.97265625,-0.119140625,2.203125,3.46875,-0.27929688,-0.75,-2.375,2.21875,-0.48828125,0.2421875,0.203125,1.265625,-1.4921875,0.44335938,-0.95703125,1.0625,0.66015625,1.7421875,-1.9296875,0.23339844,-1.265625,0.828125,1.2109375,-1.1484375,-0.06542969,-1.0703125,1.8359375,-0.80078125,-0.47851562,0.0008468628,0.040039062,-0.734375,1.21875,0.17675781,-0.5546875,0.31445312,1.546875,0.30664062,-0.068847656,0.44140625,1.296875,0.50390625,0.94921875,-2.96875,-0.12402344,-0.10107422,0.007446289,-1.2421875,-1.40625,1.6640625,1.234375,-0.5390625,-0.890625,-0.51171875
+9,0,6,-0.16699219,1.359375,-0.18457031,-0.94140625,0.55859375,0.32421875,0.57421875,-0.81640625,1.5859375,-1.4609375,2.109375,-1.296875,-0.46484375,-0.18066406,-0.43164062,-0.47851562,-1.0546875,0.03173828,0.9296875,0.640625,0.30664062,-0.98046875,1.2578125,0.07373047,-1.609375,-2.046875,-0.65234375,-0.01574707,-0.09667969,1.328125,0.62890625,-0.27734375,0.703125,0.83984375,0.111816406,-0.17480469,0.59375,0.64453125,-0.3046875,-0.0014801025,-0.45117188,0.5703125,1.046875,0.90625,-0.7265625,0.99609375,0.07324219,-1.6484375,-0.70703125,2.25,0.578125,-1.109375,-0.609375,0.84765625,0.34765625,-0.51171875,-0.010070801,1.09375,0.546875,-0.171875,0.96875,1.6875,0.20800781,0.5,1.0,0.24511719,-0.85546875,-0.8984375,0.24414062,0.9765625,-1.1015625,0.5,0.38867188,-0.53515625,-0.53515625,-0.34179688,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.091796875,-0.038085938,2.1875,0.28320312,0.020141602,-2.21875,-0.55078125,0.66796875,-0.13378906,2.203125,0.044921875,1.3203125,-0.72265625,-0.11621094,-0.049072266,-1.2109375,0.21386719,-0.81640625,-1.5390625,-0.41210938,-0.625,1.5078125,-0.45117188,-0.54296875,1.1015625,-2.03125,0.33007812,-0.21191406,-0.34179688,-0.49414062,2.109375,-0.43164062,-2.59375,-0.5859375,0.65625,-0.98046875,0.84765625,-0.24902344,0.13085938,0.5625,-1.1796875,0.025756836,-2.21875,-1.2734375,2.28125,-1.4609375,-0.70703125,-0.41601562,-0.38085938,1.0859375,0.07714844,-1.0546875,-0.0390625,-1.078125,2.34375,-0.5390625,0.62890625,0.72265625,-0.89453125,1.4296875,0.22265625,0.16894531,2.203125,0.0625,-1.0703125,-0.052246094,-0.75390625,-1.265625,0.6484375,-0.5,-0.65234375,-0.68359375,0.053466797,-0.87109375,-0.05078125,0.20214844,-0.27148438,-1.2265625,2.296875,-0.69921875,-0.6484375,1.5,1.140625,-2.4375,1.921875,-0.78515625,2.03125,-0.028686523,-0.14941406,-1.2421875,-0.026245117,-0.484375,-1.1171875,0.6015625,0.33203125,-0.111816406,-2.6875,-0.49414062,-0.94140625,1.2890625,0.57421875,0.5078125,0.8046875,-1.0078125,1.7734375,-1.2734375,-0.25585938,-0.72265625,0.125,0.48828125,0.6640625,0.36328125,0.41796875,-0.12695312,-1.71875,0.2109375,-0.036132812,1.0078125,0.7265625
+9,0,7,-0.31445312,0.34179688,0.5546875,-2.0625,-0.029174805,0.671875,1.8046875,-0.69921875,1.0390625,-0.98046875,1.125,-1.3671875,1.3671875,-0.6484375,-0.54296875,0.6953125,-0.043701172,-0.62890625,2.109375,-0.734375,-0.78125,-1.40625,1.25,-0.15429688,-2.46875,-1.6171875,0.18457031,-0.79296875,-0.515625,1.65625,0.53125,-0.546875,-0.122558594,0.099121094,1.109375,0.7734375,-0.20898438,-0.14160156,-1.8046875,0.9609375,-1.53125,0.51953125,0.95703125,0.96875,0.59765625,1.3671875,-0.98828125,-1.1484375,-0.2265625,1.609375,-1.78125,0.25390625,-1.0,1.890625,-0.25195312,0.62109375,0.20898438,0.28125,-0.047851562,-0.34375,0.60546875,-0.043945312,-0.70703125,0.52734375,0.6328125,1.421875,-0.7890625,-1.1015625,-0.98828125,2.109375,-1.109375,-0.57421875,0.080078125,0.859375,-0.4140625,0.76953125,1.5703125,0.1953125,-0.21386719,0.2265625,1.125,0.4609375,-0.16308594,1.2578125,0.16894531,-0.82421875,-0.65625,-1.28125,-0.5703125,-0.32421875,0.875,0.037109375,-0.83984375,-1.0078125,-1.5390625,0.484375,-0.5703125,-0.60546875,-0.25195312,-0.62109375,-1.375,0.06225586,2.3125,1.1796875,0.17480469,1.7734375,-0.9296875,-0.10449219,0.53515625,-0.79296875,0.40039062,1.015625,0.90234375,-2.59375,0.54296875,-0.34179688,0.28515625,0.18945312,-0.76953125,0.53515625,1.75,-0.38671875,1.234375,-2.0625,0.83203125,0.15625,-2.140625,-1.6171875,-0.19238281,0.24707031,0.71875,0.68359375,0.32617188,-1.515625,0.578125,0.40234375,-0.25585938,0.6015625,-0.328125,-0.62890625,2.546875,1.671875,-0.25585938,1.7109375,-0.87109375,-0.35546875,0.18945312,-0.15332031,-2.421875,0.66015625,-0.5390625,-0.13867188,-0.41210938,-1.2578125,-1.234375,-0.93359375,0.3359375,-0.07519531,-0.05444336,2.09375,-2.0625,-0.67578125,1.171875,-0.0077209473,-1.6875,-0.109375,-0.17285156,1.3515625,0.024414062,0.46679688,0.3515625,0.49804688,-1.375,-0.29296875,0.03564453,-0.14941406,0.765625,-1.2890625,-0.25390625,0.11279297,-0.27539062,-1.140625,1.0390625,0.79296875,-0.63671875,-0.17382812,0.3671875,0.021606445,-1.1484375,-0.17089844,-0.54296875,1.203125,1.2578125,0.13671875,-0.030883789,-0.45703125,1.34375,-0.6875,2.0625,0.5859375
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
index 3841e07f..fb1845ae 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
-1,8,124
-1,9,124
-1,10,124
-1,11,124
-1,12,124
+1,8,113
+1,9,113
+1,10,113
+1,11,113
+1,12,113
 1,13,113
-1,14,124
-1,15,124
-1,16,124
-1,17,124
-1,18,124
-1,19,124
-1,20,124
-1,21,124
-1,22,124
-1,23,124
-1,24,124
-1,25,124
-1,26,124
-1,27,124
+1,14,113
+1,15,113
+1,16,113
+1,17,113
+1,18,113
+1,19,113
+1,20,113
+1,21,113
+1,22,113
+1,23,113
+1,24,113
+1,25,113
+1,26,113
+1,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
index 58d912fa..b59ddd38 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
 2,8,103
-2,9,124
-2,10,103
-2,11,124
-2,12,103
-2,13,124
-2,14,124
-2,15,124
-2,16,124
-2,17,124
-2,18,124
-2,19,124
-2,20,124
-2,21,124
-2,22,124
-2,23,124
-2,24,124
-2,25,124
-2,26,124
-2,27,124
+2,9,122
+2,10,122
+2,11,122
+2,12,122
+2,13,122
+2,14,122
+2,15,122
+2,16,122
+2,17,122
+2,18,122
+2,19,122
+2,20,122
+2,21,122
+2,22,122
+2,23,122
+2,24,122
+2,25,122
+2,26,122
+2,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
index 081f39f9..ec95ae85 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
@@ -1,5 +1,5 @@
 sequenceId,itemPosition,itemId
-3,8,124
+3,8,103
 3,9,124
 3,10,124
 3,11,124
@@ -19,23 +19,23 @@ sequenceId,itemPosition,itemId
 3,25,124
 3,26,124
 3,27,124
-4,8,124
-4,9,124
-4,10,113
-4,11,113
-4,12,113
-4,13,113
-4,14,113
-4,15,113
-4,16,113
-4,17,113
-4,18,113
-4,19,113
-4,20,113
-4,21,113
-4,22,113
-4,23,113
-4,24,113
-4,25,113
-4,26,113
-4,27,113
+4,8,122
+4,9,122
+4,10,122
+4,11,122
+4,12,122
+4,13,122
+4,14,122
+4,15,122
+4,16,122
+4,17,122
+4,18,122
+4,19,122
+4,20,122
+4,21,122
+4,22,122
+4,23,122
+4,24,122
+4,25,122
+4,26,122
+4,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
index da076443..e689f7f2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
-5,8,124
-5,9,103
-5,10,124
-5,11,124
-5,12,124
-5,13,124
-5,14,113
-5,15,124
-5,16,124
-5,17,124
-5,18,124
-5,19,124
-5,20,124
-5,21,124
-5,22,124
-5,23,124
-5,24,124
-5,25,124
-5,26,124
-5,27,124
+5,8,103
+5,9,122
+5,10,122
+5,11,122
+5,12,122
+5,13,122
+5,14,122
+5,15,122
+5,16,122
+5,17,122
+5,18,122
+5,19,122
+5,20,122
+5,21,122
+5,22,122
+5,23,122
+5,24,122
+5,25,122
+5,26,122
+5,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
index 541bd2be..0acf1d86 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
@@ -1,41 +1,41 @@
 sequenceId,itemPosition,itemId
-6,8,124
-6,9,124
-6,10,124
-6,11,124
-6,12,124
-6,13,124
-6,14,124
-6,15,124
-6,16,124
-6,17,124
-6,18,124
-6,19,124
-6,20,124
-6,21,124
-6,22,124
-6,23,124
-6,24,124
-6,25,124
-6,26,124
-6,27,124
-7,8,124
-7,9,124
-7,10,124
-7,11,124
-7,12,124
-7,13,124
+6,8,113
+6,9,113
+6,10,113
+6,11,113
+6,12,113
+6,13,113
+6,14,113
+6,15,113
+6,16,113
+6,17,113
+6,18,113
+6,19,113
+6,20,113
+6,21,113
+6,22,113
+6,23,113
+6,24,113
+6,25,113
+6,26,113
+6,27,113
+7,8,113
+7,9,113
+7,10,113
+7,11,113
+7,12,113
+7,13,113
 7,14,113
-7,15,124
-7,16,124
-7,17,124
-7,18,124
-7,19,124
-7,20,124
-7,21,124
-7,22,124
-7,23,124
-7,24,124
-7,25,124
-7,26,124
-7,27,124
+7,15,113
+7,16,113
+7,17,113
+7,18,113
+7,19,113
+7,20,113
+7,21,113
+7,22,113
+7,23,113
+7,24,113
+7,25,113
+7,26,113
+7,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
index 79dc2ec9..a074f362 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
-8,8,103
-8,9,124
-8,10,103
+8,8,122
+8,9,122
+8,10,122
 8,11,122
 8,12,122
 8,13,122
-8,14,122
+8,14,113
 8,15,122
 8,16,122
 8,17,122
 8,18,122
 8,19,122
 8,20,122
-8,21,122
-8,22,122
-8,23,122
-8,24,122
-8,25,122
-8,26,122
-8,27,122
+8,21,113
+8,22,113
+8,23,113
+8,24,113
+8,25,113
+8,26,113
+8,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
index 31db896c..c97879eb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId
-9,8,103
-9,9,124
+9,8,113
+9,9,113
 9,10,113
 9,11,113
 9,12,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
index c936861c..c3d900b3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,124
+1,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
index d279d22f..90765ba3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,124
-4,8,124
+3,8,103
+4,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
index 81899090..60a3b97e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,124
+5,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
index 25ad0933..953b3bbc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,124
-7,8,124
+6,8,113
+7,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
index bc91b117..f3ae7b5b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,103
+8,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
index b949bc17..ba964099 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,103
+9,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
index 8892a7bd..5158426c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
 1,6,122
-1,7,111
-1,8,117
+1,7,124
+1,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
index 8812ffc8..613c7045 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-2,6,122
+2,6,121
 2,7,102
 2,8,126
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
index 60c4449f..a4583793 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId
-3,6,122
+3,6,121
 3,7,109
-3,8,124
+3,8,102
 4,6,127
 4,7,119
 4,8,120
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
index 6475d519..79312602 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
 8,6,104
-8,7,122
+8,7,121
 8,8,112
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
index d566f05c..2ee41e48 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
 9,6,118
 9,7,118
-9,8,103
+9,8,118
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
index c3d900b3..1213b475 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,113
+1,8,127
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
index 52479655..90765ba3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,113
-4,8,124
+3,8,103
+4,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
index 4e9004da..60a3b97e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,113
+5,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
index 953b3bbc..132c76ec 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,113
-7,8,113
+6,8,122
+7,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
index 346e8242..bc91b117 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,113
+8,8,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
index ba964099..824e36f4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,113
+9,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
index cbe044a8..84336320 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-1,6,122,6,1
+1,6,122,0,1
 1,7,121,9,1
 1,8,108,9,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
index c9b5b93b..f2206ac1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
 2,6,121,6,1
-2,7,102,6,5
-2,8,112,3,0
+2,7,127,7,5
+2,8,112,4,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
index c39abddc..bba6a834 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
 3,6,121,6,1
 3,7,109,3,1
-3,8,110,2,6
-4,6,102,6,1
-4,7,119,1,1
-4,8,102,0,6
+3,8,110,0,2
+4,6,121,6,1
+4,7,121,3,1
+4,8,102,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
index f6da44a0..a3007f68 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-5,6,126,6,1
-5,7,113,6,0
+5,6,113,6,1
+5,7,113,4,0
 5,8,121,4,5
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
index 71855096..3fe6eafb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-6,6,102,5,5
-6,7,118,5,1
-6,8,109,5,6
+6,6,102,4,5
+6,7,118,9,1
+6,8,121,4,6
 7,6,109,3,1
-7,7,121,2,6
-7,8,119,5,7
+7,7,121,0,2
+7,8,108,5,7
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
index 03e17fbc..e453627c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-8,6,102,8,6
-8,7,121,6,1
+8,6,102,8,8
+8,7,121,4,1
 8,8,121,3,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
index 59439ca4..62834e55 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-9,6,122,6,1
-9,7,118,4,4
+9,6,108,9,1
+9,7,118,4,8
 9,8,103,4,8
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
index 7e8daf90..8302df2f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,113
+2,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
index 18e88ed1..3ee3129e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,113
+3,8,122
 4,8,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
index 953b3bbc..c71ccfa4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 6,8,113
-7,8,113
+7,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
index 5aa808f6..7240bc4f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,113
+0,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
index c3d900b3..dea53fa9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,113
+1,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
index 7e8daf90..1616d64b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,113
+2,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
index 18e88ed1..a467ff6d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,113
-4,8,113
+3,8,122
+4,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
index 4e9004da..81899090 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,113
+5,8,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
index 953b3bbc..3e2fb799 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 6,8,113
-7,8,113
+7,8,128
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
index ba964099..56680af1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,113
+9,8,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
index 9576f383..f27b432d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,115
+0,8,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
index 598df3ba..53195e3d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,110
+2,8,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
index 04c6c551..b08ec867 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 3,8,115
-4,8,115
+4,8,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
index 3e9b1443..a620b023 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,115
+5,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
index f97711ff..ebfed34f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 6,8,115
-7,8,115
+7,8,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
index 0d49e5bc..ae1ba6f4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,115
+8,8,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
index 41144ce1..56680af1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,115
+9,8,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
index 9576f383..f27b432d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,115
+0,8,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
index 598df3ba..8302df2f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,110
+2,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
index 04c6c551..b08ec867 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 3,8,115
-4,8,115
+4,8,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
index 3e9b1443..a620b023 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,115
+5,8,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
index f97711ff..ebfed34f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
 6,8,115
-7,8,115
+7,8,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
index 0d49e5bc..a2894aea 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,115
+8,8,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
index 41144ce1..56680af1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,115
+9,8,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
index a7c99704..31f237e5 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,8,109,6,0.035670765
+0,8,109,[other],0.00014168024
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
index d440d67e..5875e518 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,8,102,1,-0.094851345
+1,8,102,[unknown],-0.09726027
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
index 6c270219..7521792f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,[mask],9,-0.12064132
+2,8,[mask],[unknown],-0.12791729
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
index fc5d931d..9bf8759e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,8,103,4,0.08533255
-4,8,110,4,0.072973825
+3,8,103,4,0.043316968
+4,8,110,4,0.032937437
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
index 7d6c2949..c93e224e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,8,110,5,-0.04847882
+5,8,110,9,-0.075533785
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
index 9c8622d1..ddd61771 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,8,122,7,-0.012625184
-7,8,113,4,0.0941027
+6,8,122,7,-0.04820269
+7,8,113,[other],0.057539806
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
index d71ac4b8..2911647f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,110,7,-0.03440542
+8,8,110,[mask],-0.095947556
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
index 9a89327f..9ad6394e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,8,114,6,0.0070434585
+9,8,114,6,-0.0014844015
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
index b87c3693..09619471 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,8,102,6,0.035670765
+0,8,102,4,0.016973361
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
index d672e857..99102984 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,8,100,1,-0.094851345
+1,8,100,[unknown],-0.09465362
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
index cff9a2a5..b2187677 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,100,9,-0.12064132
+2,8,100,[unknown],-0.12798095
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
index e3a13c6f..2fbedb5c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,8,111,4,0.08533255
-4,8,109,4,0.072973825
+3,8,111,4,0.058097668
+4,8,128,4,0.047720857
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
index dcae774d..5063e0ad 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,8,102,5,-0.04847882
+5,8,102,6,-0.06665557
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
index 0ab95c0d..2740828e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,8,110,7,-0.012625184
-7,8,102,4,0.0941027
+6,8,110,7,-0.029965512
+7,8,102,4,0.083941795
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
index edf48cd7..a3fb49ac 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,[mask],7,-0.03440542
+8,8,128,0,-0.07864175
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
index b1aaa3c1..5d07e206 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,8,111,6,0.0070434585
+9,8,125,6,0.00074065477
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index 314cd804..9216a636 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,0.045526013
-0,22,0.20925649
-0,23,0.28213653
-0,24,0.33005765
-0,25,0.3220708
-0,26,0.13737482
-0,27,0.20725977
-0,28,0.09843891
-0,29,-0.060299788
-0,30,-0.003611853
-0,31,0.045026835
-0,32,0.029302716
-0,33,-0.05430965
-0,34,-0.06728829
-0,35,0.009834763
-0,36,0.23521377
-0,37,0.08845534
-0,38,0.040534228
-0,39,0.050268207
-0,40,-0.2969103
-1,19,0.25617927
-1,20,0.21324992
-1,21,0.38596562
-1,22,0.37797877
-1,23,0.07298081
-1,24,-0.05381047
-1,25,0.03778875
-1,26,0.098938085
-1,27,-0.020615114
-1,28,-0.037088
-1,29,0.020941481
-1,30,-0.024234157
-1,31,0.1423666
-1,32,0.09594302
-1,33,0.07198246
-1,34,-0.112713516
-1,35,0.042281352
-1,36,-0.053560883
-1,37,-0.032595392
-1,38,0.102931514
-2,17,0.38596562
-2,18,-0.12868722
-2,19,0.10043562
-2,20,0.014514559
-2,21,0.15434688
-2,22,0.08995288
-2,23,0.14536166
-2,24,0.06599232
-2,25,0.15235017
-2,26,0.0074636657
-2,27,0.047772314
-2,28,0.24220227
-2,29,-0.00005520758
-2,30,0.07148328
-2,31,-0.028352378
-2,32,-0.10422748
-2,33,-0.070782535
-2,34,0.0664915
-2,35,-0.12369544
-2,36,-0.27095303
-3,17,0.13038632
-3,18,0.14036989
-3,19,0.37398535
-3,20,-0.042578958
-3,21,0.34603137
-3,22,0.26216942
-3,23,-0.1177053
-3,24,0.25617927
-3,25,-0.029849913
-3,26,-0.0600502
-3,27,0.10343069
-3,28,0.07048492
-3,29,-0.07228007
-3,30,-0.015623331
-3,31,-0.0010243154
-3,32,0.12589371
-3,33,0.19228444
-3,34,0.086957805
-3,35,0.003563835
-3,36,-0.004672607
-4,14,0.13038632
-4,15,0.04877067
-4,16,0.0894537
-4,17,0.054261632
-4,18,0.18928936
-4,19,0.053263277
-4,20,-0.10322913
-4,21,-0.14116667
-4,22,-0.25897273
-4,23,-0.039583888
-4,24,-0.068785824
-4,25,-0.112713516
-4,26,0.09943727
-4,27,-0.0018198809
-4,28,0.14036989
-4,29,-0.15813874
-4,30,0.09494466
-4,31,-0.07976775
-4,32,0.24719404
-4,33,0.0709841
-5,14,0.23321705
-5,15,0.37198862
-5,16,0.21524663
-5,17,0.26416612
-5,18,0.40193933
-5,19,0.07697424
-5,20,0.18729265
-5,21,0.22722691
-5,22,-0.043577317
-5,23,0.0550104
-5,24,-0.08925214
-5,25,0.00047516936
-5,26,-0.09324556
-5,27,-0.14216504
-5,28,-0.09973488
-5,29,-0.09374474
-5,30,-0.043826904
-5,31,-0.14216504
-5,32,0.04278053
-5,33,-0.17211573
-6,18,0.4438703
-6,19,0.022938194
-6,20,0.23022199
-6,21,0.4119229
-6,22,0.19927293
-6,23,0.12539454
-6,24,-0.11371187
-6,25,-0.021613471
-6,26,0.16033702
-6,27,0.23022199
-6,28,-0.1501519
-6,29,-0.058802254
-6,30,-0.028976351
-6,31,-0.1591371
-6,32,-0.106224194
-6,33,-0.022861416
-6,34,-0.03384334
-6,35,0.09344713
-6,36,-0.021987854
-6,37,-0.14016832
-7,15,0.12339783
-7,16,0.3041004
-7,17,-0.0067005185
-7,18,0.00015538326
-7,19,-0.09374474
-7,20,0.07198246
-7,21,0.034544088
-7,22,0.23321705
-7,23,-0.06778747
-7,24,0.0054045552
-7,25,0.020067917
-7,26,-0.032221008
-7,27,0.033545732
-7,28,-0.04806992
-7,29,-0.034342516
-7,30,-0.13417818
-7,31,-0.094743095
-7,32,-0.118204474
-7,33,0.16233373
-7,34,-0.09524228
-8,20,0.18329921
-8,21,0.116409324
-8,22,0.2921201
-8,23,0.0100843515
-8,24,-0.052812114
-8,25,0.07198246
-8,26,0.08595945
-8,27,-0.26096946
-8,28,-0.00040619232
-8,29,0.09993644
-8,30,-0.312884
-8,31,-0.12070037
-8,32,-0.13417818
-8,33,-0.18709108
-8,34,0.06898739
-8,35,0.010333941
-8,36,-0.00027359813
-8,37,0.011956271
-8,38,-0.12519297
-8,39,0.086957805
-9,16,0.25817597
-9,17,0.3939525
-9,18,0.12489536
-9,19,0.20126964
-9,20,0.40393606
-9,21,0.27215296
-9,22,0.14835674
-9,23,-0.006326135
-9,24,-0.14715682
-9,25,-0.058053486
-9,26,-0.058552664
-9,27,0.22023842
-9,28,-0.00056998525
-9,29,0.024810111
-9,30,0.10992001
-9,31,0.010833119
-9,32,0.038287926
-9,33,0.020567097
-9,34,0.12739125
-9,35,-0.18709108
+0,21,-0.11176976
+0,22,0.34597677
+0,23,0.26810494
+0,24,0.27209836
+0,25,0.26411152
+0,26,0.20620683
+0,27,0.10986541
+0,28,0.16327749
+0,29,0.084906496
+0,30,0.23715588
+0,31,-0.2280783
+0,32,0.08041389
+0,33,-0.15819333
+0,34,0.06743526
+0,35,0.0055371495
+0,36,-0.011310118
+0,37,-0.0062559377
+0,38,-0.18614732
+0,39,0.11385884
+0,40,0.20321175
+1,19,0.100880206
+1,20,0.36195046
+1,21,0.14430872
+1,22,0.29406223
+1,23,0.17126435
+1,24,0.056453336
+1,25,0.11335966
+1,26,-0.08231824
+1,27,0.08191143
+1,28,-0.31693202
+1,29,0.09788513
+1,30,0.084407315
+1,31,-0.12325086
+1,32,0.047468126
+1,33,-0.04812452
+1,34,-0.14221963
+1,35,-0.08581249
+1,36,0.013212016
+1,37,-0.18015718
+1,38,-0.096794404
+2,17,0.18623969
+2,18,0.056702927
+2,19,-0.09579605
+2,20,0.06793443
+2,21,0.10287692
+2,22,0.20520847
+2,23,0.023881953
+2,24,-0.027907796
+2,25,0.0060675265
+2,26,-0.19513254
+2,27,-0.10478126
+2,28,0.07941554
+2,29,0.07442375
+2,30,0.21419367
+2,31,-0.16917527
+2,32,-0.07033796
+2,33,-0.11326729
+2,34,-0.122751676
+2,35,-0.120754965
+2,36,-0.13822621
+3,17,0.2740951
+3,18,0.2092019
+3,19,0.3060425
+3,20,0.06444018
+3,21,0.19821997
+3,22,0.13033172
+3,23,0.066936076
+3,24,-0.109273866
+3,25,-0.049871642
+3,26,0.07592129
+3,27,-0.0027070923
+3,28,-0.042134378
+3,29,-0.09779277
+3,30,-0.011247721
+3,31,-0.043631915
+3,32,-0.061851922
+3,33,-0.027783003
+3,34,-0.23007502
+3,35,0.0060675265
+3,36,-0.015740326
+4,14,0.090896636
+4,15,0.16227913
+4,16,0.15728734
+4,17,0.10487363
+4,18,-0.042134378
+4,19,0.08340896
+4,20,0.13732022
+4,21,0.09788513
+4,22,-0.10178619
+4,23,-0.15519828
+4,24,-0.18315226
+4,25,-0.008658233
+4,26,0.00049856835
+4,27,-0.011122926
+4,28,0.07592129
+4,29,0.001980504
+4,30,0.027500995
+4,31,-0.05411466
+4,32,0.028374556
+4,33,0.01265044
+5,14,0.27010167
+5,15,0.22916903
+5,16,0.19622326
+5,17,0.1493005
+5,18,0.042975523
+5,19,0.15129721
+5,20,0.04372429
+5,21,-0.10428208
+5,22,-0.020295328
+5,23,-0.05661055
+5,24,-0.018049026
+5,25,-0.122751676
+5,26,-0.071336314
+5,27,-0.025911083
+5,28,0.096387595
+5,29,-0.039888076
+5,30,-0.042134378
+5,31,-0.018049026
+5,32,-0.12325086
+5,33,-0.029405331
+6,18,0.100880206
+6,19,0.21319532
+6,20,0.32001948
+6,21,0.3539636
+6,22,0.38591102
+6,23,0.03823333
+6,24,-0.08281741
+6,25,-0.18315226
+6,26,-0.03489629
+6,27,-0.036144238
+6,28,0.05919882
+6,29,0.03848292
+6,30,-0.089306735
+6,31,-0.06784207
+6,32,-0.1267451
+6,33,0.054456625
+6,34,-0.06634453
+6,35,-0.004025235
+6,36,-0.10478126
+6,37,0.09189499
+7,15,0.20121504
+7,16,0.26610824
+7,17,0.118351445
+7,18,0.14630543
+7,19,-0.09330016
+7,20,0.12833501
+7,21,-0.01792423
+7,22,-0.05960562
+7,23,-0.021293685
+7,24,0.12284405
+7,25,0.15529063
+7,26,0.03012168
+7,27,-0.09379934
+7,28,0.04946484
+7,29,-0.06784207
+7,30,-0.087310016
+7,31,-0.04537904
+7,32,0.059947584
+7,33,0.33798993
+7,34,-0.15619662
+8,20,0.2391526
+8,21,0.010279343
+8,22,0.12134651
+8,23,0.29805565
+8,24,-0.25203884
+8,25,-0.03177643
+8,26,-0.053116303
+8,27,0.030371271
+8,28,-0.055861782
+8,29,-0.040137667
+8,30,-0.07183549
+8,31,0.14630543
+8,32,-0.10028865
+8,33,-0.12824264
+8,34,0.021760445
+8,35,0.005661944
+8,36,0.12833501
+8,37,-0.21509966
+8,38,-0.13423277
+8,39,-0.07433138
+9,16,0.27209836
+9,17,-0.05860726
+9,18,0.17326106
+9,19,0.41985515
+9,20,0.11735309
+9,21,0.35196692
+9,22,0.13532351
+9,23,-0.008096658
+9,24,-0.049871642
+9,25,-0.04812452
+9,26,-0.042633556
+9,27,-0.1591917
+9,28,-0.2280783
+9,29,-0.12624593
+9,30,0.08989828
+9,31,-0.023415193
+9,32,-0.15519828
+9,33,-0.14022292
+9,34,0.10337609
+9,35,-0.071336314
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
index 8a20bbbe..44ec0f7b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,0.33205438
-0,9,-0.029600324
-0,10,0.42190647
-0,11,0.29611352
-0,12,-0.30689386
-1,8,0.3180774
-1,9,0.34603137
-1,10,0.37797877
-2,8,0.045026835
-2,9,0.14336495
-3,8,0.40992618
-3,9,0.3360478
-4,8,-0.25897273
-5,8,0.16433044
-6,8,0.38596562
-6,9,0.32606423
-6,10,0.36000836
-7,8,0.216245
-8,8,0.19228444
-8,9,-0.25398096
-8,10,0.473821
-8,11,0.46583417
-9,8,0.36999193
-9,9,0.31608066
+0,8,0.2900688
+0,9,-0.17616376
+0,10,0.3799209
+0,11,0.26610824
+0,12,-0.2989616
+1,8,0.28208193
+1,9,-0.003034678
+1,10,0.33798993
+2,8,-0.14820977
+2,9,0.11785226
+3,8,0.36993733
+3,9,0.29605892
+4,8,-0.13523114
+5,8,0.1403153
+6,8,0.3359932
+6,9,0.26810494
+6,10,0.31402937
+7,8,-0.15320155
+8,8,0.16826928
+8,9,-0.25703064
+8,10,0.41785845
+8,11,0.40987158
+9,8,0.31602606
+9,9,0.25812137
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
index a06dd55d..dcfed5d1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.0067936005
-0,9,0.047275458
-0,10,0.06949562
-0,11,0.34379116
-0,12,0.53735346
-1,8,0.17491794
-1,9,0.031227563
-1,10,-0.15887825
-2,8,0.0593731
-2,9,-0.139127
-3,8,0.09344401
-3,9,0.183806
-4,8,0.33984092
-5,8,0.015025363
-6,8,0.35564193
-6,9,0.31218916
-6,10,0.0023722164
-7,8,0.6124082
-8,8,0.2608359
-8,9,0.30428866
-8,10,0.116651736
-8,11,0.1709677
-9,8,0.007306172
-9,9,-0.15789069
+0,8,-0.19187672
+0,9,-0.078307025
+0,10,0.01989374
+0,11,0.13599409
+0,12,0.30980512
+1,8,0.08809729
+1,9,-0.026706878
+1,10,-0.23631705
+2,8,0.046866544
+2,9,-0.18792649
+3,8,0.08710972
+3,9,0.17352147
+4,8,0.3789345
+5,8,-0.18002598
+6,8,0.2999295
+6,9,0.33548176
+6,10,0.007902182
+7,8,0.66335255
+8,8,0.19821054
+8,9,0.19031003
+8,10,-0.03189158
+8,11,-0.01386856
+9,8,0.001099186
+9,9,-0.2531056
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
index fae6087b..b3488afb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.08213164
-0,9,-0.36485112
-0,10,-0.04796779
-0,11,-0.23358852
-0,12,-0.23358852
-1,8,-0.24276772
-1,9,-0.048351455
-1,10,0.042246565
-2,8,-0.377702
-2,9,-0.29325333
-3,8,-0.115176775
-3,9,0.27402145
-4,8,0.16019934
-5,8,-0.36485112
-6,8,0.08217611
-6,9,-0.06515011
-6,10,0.18131152
-7,8,0.20058784
-8,8,-0.22624514
-8,9,-0.115176775
-8,10,-0.29508916
-8,11,-0.29508916
-9,8,-0.14087854
-9,9,0.024806082
+0,8,-0.075993046
+0,9,-0.34528795
+0,10,-0.06383061
+0,11,-0.22871205
+0,12,-0.23330164
+1,8,-0.2268762
+1,9,-0.10984136
+1,10,-0.017590366
+2,8,-0.3618105
+2,9,-0.29939193
+3,8,-0.13783793
+3,9,0.23850942
+4,8,0.10357512
+5,8,-0.3471238
+6,8,0.09990344
+6,9,-0.1116772
+6,10,0.15314281
+7,8,0.12560523
+8,8,-0.1773085
+8,9,-0.111218244
+8,10,-0.29939193
+8,11,-0.29204854
+9,8,-0.14334546
+9,9,-0.018737767
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
index 953319ec..8f72a904 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.061894365
-0,9,-0.20701185
-0,10,-0.11131864
-0,11,-0.020215044
-0,12,-0.03547547
-1,8,-0.035131253
-1,9,0.03623706
-1,10,-0.06484892
-2,8,-0.3391924
-2,9,-0.2226165
-3,8,0.095901884
-3,9,0.5291603
-4,8,0.41166648
-5,8,-0.13311926
-6,8,0.38780054
-6,9,0.3217103
-6,10,0.47775677
-7,8,0.41350234
-8,8,-0.054174513
-8,9,-0.03547547
-8,10,-0.03593443
-8,11,-0.1601979
-9,8,0.055054426
-9,9,0.23542577
+0,8,-0.081780255
+0,9,-0.22130415
+0,10,-0.060438603
+0,11,-0.065946124
+0,12,-0.09807333
+1,8,-0.055734262
+1,9,-0.02922931
+1,10,-0.12492251
+2,8,-0.25710306
+2,9,-0.24058047
+3,8,0.01597827
+3,9,0.36891866
+4,8,0.3579036
+5,8,-0.12675835
+6,8,0.3395452
+6,9,0.2771266
+6,10,0.3175151
+7,8,0.35606775
+8,8,-0.03835114
+8,9,-0.016148943
+8,10,-0.05389842
+8,11,-0.12675835
+9,8,-0.001691699
+9,9,0.18900627
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
index 923104fa..2a8d05f6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,121,4,9,8,1
-0,9,121,4,3,8,0
-0,10,121,4,3,6,1
-0,11,121,6,3,6,1
-0,12,121,6,3,6,1
-0,13,121,6,3,6,1
-0,14,121,4,3,6,1
-0,15,121,4,3,6,1
-0,16,121,4,3,6,1
-0,17,121,4,3,6,1
-0,18,121,4,3,8,1
-0,19,121,4,5,6,1
-0,20,121,4,3,8,1
-0,21,121,4,3,6,1
-0,22,121,4,3,6,0
-0,23,121,4,3,8,0
-0,24,121,4,3,6,0
-0,25,121,4,3,6,0
-0,26,121,4,3,6,0
-0,27,121,4,3,6,0
-0,28,121,4,3,6,0
-0,29,121,4,3,6,0
-0,30,121,4,3,6,0
-0,31,121,4,3,6,0
-0,32,121,4,3,6,1
-0,33,121,4,3,6,0
-0,34,121,4,3,6,1
-0,35,121,4,3,6,0
-0,36,121,4,3,6,1
-0,37,121,4,3,6,1
+0,8,113,6,2,0,1
+0,9,113,6,2,0,1
+0,10,113,6,2,0,1
+0,11,113,6,2,0,1
+0,12,113,6,2,0,1
+0,13,113,6,2,0,2
+0,14,113,6,2,0,2
+0,15,113,6,2,0,2
+0,16,113,6,2,0,2
+0,17,113,6,2,0,2
+0,18,113,6,2,0,2
+0,19,113,6,2,0,2
+0,20,113,6,2,0,2
+0,21,113,6,2,0,2
+0,22,113,6,2,0,2
+0,23,113,6,2,0,2
+0,24,113,6,2,0,2
+0,25,113,6,2,0,2
+0,26,113,6,2,0,2
+0,27,113,6,2,0,2
+0,28,113,6,2,0,2
+0,29,113,6,2,0,2
+0,30,113,6,2,0,2
+0,31,113,6,2,0,2
+0,32,113,6,2,0,2
+0,33,113,6,2,0,2
+0,34,113,6,2,0,2
+0,35,113,6,2,0,2
+0,36,113,6,2,0,2
+0,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
index aa4c0f4a..cd416b33 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,103,4,9,8,1
-1,9,121,4,3,6,1
-1,10,121,4,3,6,1
-1,11,121,4,3,6,1
-1,12,121,6,3,6,1
-1,13,121,4,3,6,0
-1,14,121,4,3,6,0
-1,15,121,4,3,6,0
-1,16,121,4,3,6,0
-1,17,121,4,3,6,1
-1,18,121,4,3,6,1
-1,19,121,4,3,6,1
-1,20,121,4,3,6,1
-1,21,121,4,3,6,1
-1,22,121,4,3,6,1
-1,23,121,4,3,6,1
-1,24,121,4,3,6,1
-1,25,121,4,3,8,1
-1,26,121,4,3,6,1
-1,27,121,4,3,6,1
-1,28,121,4,3,6,1
-1,29,121,4,3,6,1
-1,30,121,4,3,6,1
-1,31,121,4,3,6,0
-1,32,121,4,3,8,1
-1,33,121,4,3,6,0
-1,34,121,4,3,8,1
-1,35,121,4,3,6,1
-1,36,121,4,3,6,0
-1,37,121,4,3,6,0
+1,8,113,6,0,9,0
+1,9,100,6,0,8,0
+1,10,121,6,0,8,0
+1,11,120,4,0,8,0
+1,12,120,4,0,8,0
+1,13,120,0,0,5,0
+1,14,109,4,8,8,0
+1,15,109,4,9,5,0
+1,16,109,4,9,5,0
+1,17,109,4,9,5,0
+1,18,109,4,0,5,0
+1,19,109,4,9,5,0
+1,20,109,4,0,9,0
+1,21,109,4,0,5,0
+1,22,127,4,0,9,0
+1,23,120,4,0,9,0
+1,24,120,4,0,9,0
+1,25,109,4,0,5,0
+1,26,109,4,9,5,0
+1,27,127,4,0,9,0
+1,28,109,4,9,5,0
+1,29,109,4,9,5,0
+1,30,109,4,3,5,0
+1,31,109,4,3,5,0
+1,32,109,4,3,5,0
+1,33,109,4,3,5,0
+1,34,127,6,3,5,0
+1,35,109,6,3,5,0
+1,36,112,6,3,5,0
+1,37,103,6,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
index a00ae412..cedb1297 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,103,4,9,8,1
-2,9,103,4,9,5,1
-2,10,103,4,3,6,0
-2,11,121,4,3,6,0
-2,12,121,4,3,6,0
-2,13,121,6,3,6,0
-2,14,121,4,3,6,0
-2,15,121,4,3,6,0
-2,16,121,4,3,6,0
-2,17,121,4,3,6,0
-2,18,121,4,3,6,0
-2,19,121,4,3,6,1
-2,20,121,4,3,6,1
-2,21,121,4,3,6,1
-2,22,121,4,3,6,0
-2,23,121,4,3,6,1
-2,24,121,4,3,6,1
-2,25,121,4,3,6,1
-2,26,121,4,3,6,1
-2,27,121,4,3,6,1
-2,28,121,4,3,6,1
-2,29,121,4,3,6,1
-2,30,121,4,3,6,1
-2,31,121,4,3,8,1
-2,32,121,4,3,6,1
-2,33,121,4,3,6,1
-2,34,121,4,3,6,1
-2,35,121,4,3,6,1
-2,36,121,4,3,6,1
-2,37,121,4,3,6,0
+2,8,113,4,2,0,1
+2,9,113,6,2,0,0
+2,10,113,6,2,0,0
+2,11,113,6,2,0,1
+2,12,113,6,2,0,1
+2,13,113,6,2,0,1
+2,14,113,6,2,0,2
+2,15,113,6,2,0,2
+2,16,113,6,2,0,2
+2,17,113,6,2,0,2
+2,18,113,6,2,0,2
+2,19,113,6,2,0,2
+2,20,113,6,2,0,2
+2,21,113,6,2,0,2
+2,22,113,6,2,0,2
+2,23,113,6,2,0,2
+2,24,113,6,2,0,2
+2,25,113,6,2,0,2
+2,26,113,6,2,0,2
+2,27,113,6,2,0,2
+2,28,113,6,2,0,2
+2,29,113,6,2,0,2
+2,30,113,6,2,0,2
+2,31,113,6,2,0,2
+2,32,113,6,2,0,2
+2,33,113,6,2,0,2
+2,34,113,6,2,0,2
+2,35,113,6,2,0,2
+2,36,113,6,2,0,2
+2,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
index 48a9f496..477f3560 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,103,4,9,8,1
-3,9,103,4,9,8,1
-3,10,121,4,3,6,1
-3,11,121,4,3,6,0
-3,12,121,4,3,6,1
-3,13,121,6,3,6,1
-3,14,121,4,3,6,1
-3,15,121,6,3,6,0
-3,16,121,4,3,6,0
-3,17,121,4,3,6,0
-3,18,121,4,3,6,1
-3,19,121,4,3,6,1
-3,20,121,4,3,6,1
-3,21,121,4,3,6,1
-3,22,121,4,3,6,1
-3,23,121,4,3,6,1
-3,24,121,4,3,6,1
-3,25,121,4,3,6,1
-3,26,121,4,3,8,1
-3,27,121,4,3,6,1
-3,28,121,4,3,6,1
-3,29,121,4,3,6,1
-3,30,121,4,3,6,1
-3,31,121,4,3,6,1
-3,32,121,4,3,6,0
-3,33,121,4,3,8,1
-3,34,121,4,3,6,0
-3,35,121,4,3,8,1
-3,36,121,4,3,6,1
-3,37,121,4,3,6,0
-4,8,103,4,9,8,1
-4,9,103,4,9,8,1
-4,10,121,4,3,6,1
-4,11,121,4,3,6,0
-4,12,121,4,3,6,1
-4,13,121,6,3,6,0
-4,14,121,4,3,6,1
-4,15,121,6,3,6,0
-4,16,121,4,3,6,0
-4,17,121,4,3,6,0
-4,18,121,4,3,6,1
-4,19,121,4,3,6,1
-4,20,121,4,3,6,1
-4,21,121,4,3,6,1
-4,22,121,4,3,6,1
-4,23,121,4,3,6,1
-4,24,121,4,3,6,1
-4,25,121,4,3,6,1
-4,26,121,4,3,8,1
-4,27,121,4,3,6,1
-4,28,121,4,3,6,1
-4,29,121,4,3,6,1
-4,30,121,4,3,6,1
-4,31,121,4,3,6,1
-4,32,121,4,3,6,0
-4,33,121,4,3,8,1
-4,34,121,4,3,6,0
-4,35,121,4,3,8,1
-4,36,121,4,3,6,1
-4,37,121,4,3,6,0
+3,8,121,4,2,8,1
+3,9,121,4,2,8,1
+3,10,121,4,9,8,0
+3,11,109,4,9,5,0
+3,12,109,6,9,5,0
+3,13,121,6,0,5,0
+3,14,109,4,9,5,0
+3,15,109,4,0,5,0
+3,16,120,8,0,9,0
+3,17,120,4,0,5,0
+3,18,127,4,0,9,0
+3,19,120,4,0,9,0
+3,20,109,4,0,8,0
+3,21,109,4,9,5,0
+3,22,109,4,0,5,0
+3,23,109,4,9,5,0
+3,24,109,4,9,5,0
+3,25,109,4,3,5,0
+3,26,109,4,9,5,0
+3,27,127,4,0,9,0
+3,28,127,4,0,5,0
+3,29,127,4,0,9,0
+3,30,109,4,0,5,0
+3,31,109,4,0,9,0
+3,32,109,4,0,5,0
+3,33,109,4,9,5,0
+3,34,109,4,9,5,0
+3,35,109,4,9,5,0
+3,36,109,4,3,5,0
+3,37,109,4,3,5,0
+4,8,113,7,2,8,1
+4,9,113,6,2,8,0
+4,10,121,6,2,8,0
+4,11,121,4,2,8,0
+4,12,109,4,9,5,0
+4,13,109,6,9,5,0
+4,14,121,6,0,9,0
+4,15,120,4,0,5,0
+4,16,120,9,0,9,0
+4,17,120,0,0,5,0
+4,18,109,4,0,5,0
+4,19,109,4,0,5,0
+4,20,127,4,9,9,0
+4,21,127,4,9,5,0
+4,22,109,4,9,5,0
+4,23,109,4,9,5,0
+4,24,109,4,9,5,0
+4,25,109,6,3,9,0
+4,26,109,4,0,5,0
+4,27,127,6,0,9,0
+4,28,127,9,0,5,0
+4,29,127,4,0,9,0
+4,30,120,4,0,5,0
+4,31,109,4,3,9,0
+4,32,109,4,9,5,0
+4,33,109,4,9,5,0
+4,34,109,4,9,5,0
+4,35,109,4,9,5,0
+4,36,109,4,3,5,0
+4,37,109,4,3,5,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
index 551fdd40..e21fb885 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,103,3,9,8,1
-5,9,103,4,3,8,1
-5,10,121,4,3,6,1
-5,11,121,4,3,6,0
-5,12,121,4,3,6,1
-5,13,121,6,3,6,1
-5,14,121,6,3,6,1
-5,15,121,4,3,6,1
-5,16,121,4,3,6,0
-5,17,121,4,3,6,1
-5,18,121,4,3,6,1
-5,19,121,4,3,6,1
-5,20,121,4,3,6,1
-5,21,121,4,3,6,1
-5,22,121,4,3,6,1
-5,23,121,4,3,6,1
-5,24,121,4,3,6,1
-5,25,121,4,3,8,1
-5,26,121,4,3,6,1
-5,27,121,4,3,6,1
-5,28,121,4,3,6,1
-5,29,121,4,3,6,1
-5,30,121,4,3,6,1
-5,31,121,4,3,6,0
-5,32,121,4,3,8,1
-5,33,121,4,3,6,0
-5,34,121,4,3,8,1
-5,35,121,4,3,6,1
-5,36,121,4,3,6,0
-5,37,121,4,3,6,0
+5,8,113,6,2,1,1
+5,9,113,6,2,0,2
+5,10,113,6,2,0,2
+5,11,113,6,2,0,2
+5,12,113,6,2,0,2
+5,13,113,6,2,0,2
+5,14,113,6,2,0,2
+5,15,113,6,2,0,2
+5,16,113,6,2,0,2
+5,17,113,6,2,0,2
+5,18,113,6,2,0,2
+5,19,113,6,2,0,2
+5,20,113,6,2,0,2
+5,21,113,6,2,0,2
+5,22,113,6,2,0,2
+5,23,113,6,2,0,2
+5,24,113,6,2,0,2
+5,25,113,6,2,0,2
+5,26,113,6,2,0,2
+5,27,113,6,2,0,2
+5,28,113,6,2,0,2
+5,29,113,6,2,0,2
+5,30,113,6,2,0,2
+5,31,113,6,2,0,2
+5,32,113,6,2,0,2
+5,33,113,6,2,0,2
+5,34,113,6,2,0,2
+5,35,113,6,2,0,2
+5,36,113,6,2,0,2
+5,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
index d875f86b..908877f2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,121,4,9,8,1
-6,9,121,4,9,8,1
-6,10,121,4,3,6,1
-6,11,121,6,3,6,1
-6,12,121,6,3,6,1
-6,13,121,4,3,6,1
-6,14,121,4,3,6,1
-6,15,121,4,3,6,1
-6,16,121,4,3,6,0
-6,17,121,4,3,8,1
-6,18,121,4,3,6,1
-6,19,121,4,3,6,1
-6,20,121,4,3,6,1
-6,21,121,4,3,6,0
-6,22,121,4,3,6,1
-6,23,121,4,3,6,0
-6,24,121,4,3,6,0
-6,25,121,4,3,6,0
-6,26,121,4,3,6,1
-6,27,121,4,3,6,1
-6,28,121,4,3,6,1
-6,29,121,4,3,6,1
-6,30,121,4,3,6,1
-6,31,121,4,3,6,1
-6,32,121,4,3,6,1
-6,33,121,4,3,6,1
-6,34,121,4,3,8,1
-6,35,121,4,3,6,1
-6,36,121,4,3,6,1
-6,37,121,4,3,6,1
-7,8,121,4,9,8,1
-7,9,103,4,9,5,1
-7,10,121,4,3,6,1
-7,11,121,4,3,6,1
-7,12,121,6,3,6,1
-7,13,121,6,3,6,1
-7,14,121,4,3,6,1
-7,15,121,4,3,6,1
-7,16,121,4,3,6,0
-7,17,121,4,3,6,1
-7,18,121,4,3,6,1
-7,19,121,4,3,6,1
-7,20,121,4,3,6,1
-7,21,121,4,3,6,1
-7,22,121,4,3,6,1
-7,23,121,4,3,6,1
-7,24,121,4,3,6,1
-7,25,121,4,3,8,1
-7,26,121,4,3,6,1
-7,27,121,4,3,6,1
-7,28,121,4,3,6,1
-7,29,121,4,3,6,1
-7,30,121,4,3,6,1
-7,31,121,4,3,6,0
-7,32,121,4,3,8,1
-7,33,121,4,3,6,0
-7,34,121,4,3,8,1
-7,35,121,4,3,6,1
-7,36,121,4,3,6,0
-7,37,121,4,3,6,0
+6,8,113,6,2,8,0
+6,9,113,6,2,8,0
+6,10,113,6,2,8,0
+6,11,121,6,2,8,0
+6,12,121,6,0,8,0
+6,13,109,6,0,8,0
+6,14,109,4,0,8,0
+6,15,109,4,0,5,0
+6,16,109,4,0,5,0
+6,17,109,4,9,5,0
+6,18,109,4,0,5,0
+6,19,109,4,9,5,0
+6,20,109,4,9,5,0
+6,21,109,4,3,9,0
+6,22,109,4,0,5,0
+6,23,109,4,3,9,0
+6,24,109,4,0,5,0
+6,25,127,4,0,9,0
+6,26,109,4,0,5,0
+6,27,127,4,0,9,0
+6,28,109,4,0,5,0
+6,29,109,4,9,5,0
+6,30,109,4,9,5,0
+6,31,109,4,9,5,0
+6,32,109,4,3,5,0
+6,33,109,4,3,5,0
+6,34,109,4,3,5,0
+6,35,109,4,3,5,0
+6,36,109,6,3,5,0
+6,37,127,6,3,9,0
+7,8,109,4,9,6,0
+7,9,121,4,9,6,0
+7,10,121,4,2,8,1
+7,11,109,4,9,8,0
+7,12,109,4,9,5,0
+7,13,109,4,9,5,0
+7,14,109,6,3,5,0
+7,15,121,6,0,5,0
+7,16,109,6,0,8,0
+7,17,109,4,0,5,0
+7,18,127,4,0,9,0
+7,19,120,4,0,5,0
+7,20,127,4,3,9,0
+7,21,109,4,9,5,0
+7,22,109,4,3,0,0
+7,23,109,4,9,5,0
+7,24,109,4,9,5,0
+7,25,109,4,3,5,0
+7,26,109,4,3,5,0
+7,27,109,6,3,5,0
+7,28,109,6,3,5,0
+7,29,112,6,3,9,0
+7,30,121,6,3,5,0
+7,31,112,6,3,9,0
+7,32,121,6,3,5,0
+7,33,103,6,3,0,0
+7,34,103,6,3,0,0
+7,35,103,6,2,0,0
+7,36,103,6,2,0,0
+7,37,103,6,2,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
index c203f13c..d1cabc88 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,103,4,9,8,1
-8,9,121,4,3,6,1
-8,10,121,4,3,6,1
-8,11,121,6,3,6,1
-8,12,121,6,3,6,1
-8,13,121,6,3,6,1
-8,14,121,4,3,6,1
-8,15,121,4,3,6,1
-8,16,121,4,3,6,1
-8,17,121,4,3,8,1
-8,18,121,4,5,8,1
-8,19,121,4,9,6,1
-8,20,121,4,3,6,1
-8,21,121,4,3,6,1
-8,22,121,4,3,6,0
-8,23,121,4,3,6,0
-8,24,121,4,3,6,0
-8,25,121,4,3,6,0
-8,26,121,4,3,6,0
-8,27,121,4,3,6,1
-8,28,121,4,3,6,1
-8,29,121,4,3,6,1
-8,30,121,4,3,6,0
-8,31,121,4,3,6,1
-8,32,121,4,3,6,1
-8,33,121,4,3,6,1
-8,34,121,4,3,6,1
-8,35,121,4,3,6,1
-8,36,121,4,3,6,1
-8,37,121,4,3,6,1
+8,8,113,6,2,8,0
+8,9,113,6,2,8,0
+8,10,113,6,2,8,0
+8,11,113,6,2,0,0
+8,12,113,6,2,0,0
+8,13,113,6,8,0,0
+8,14,113,6,8,0,0
+8,15,113,9,7,0,0
+8,16,113,6,2,0,1
+8,17,113,6,2,0,1
+8,18,113,6,2,0,1
+8,19,113,6,2,0,1
+8,20,113,6,2,0,2
+8,21,113,6,2,0,2
+8,22,113,6,2,0,2
+8,23,113,6,2,0,2
+8,24,113,6,2,0,2
+8,25,113,6,2,0,2
+8,26,113,6,2,0,2
+8,27,113,6,2,0,2
+8,28,113,6,2,0,2
+8,29,113,6,2,0,2
+8,30,113,6,2,0,2
+8,31,113,6,2,0,2
+8,32,113,6,2,0,2
+8,33,113,6,2,0,2
+8,34,113,6,2,0,2
+8,35,113,6,2,0,2
+8,36,113,6,2,0,2
+8,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
index 629630a6..f9969b00 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,103,4,9,8,1
-9,9,103,4,3,8,0
-9,10,121,4,3,7,0
-9,11,121,4,3,7,0
-9,12,121,4,3,6,0
-9,13,121,4,3,6,0
-9,14,121,4,3,6,0
-9,15,121,4,3,6,0
-9,16,121,4,3,6,0
-9,17,121,4,3,6,0
-9,18,121,4,3,6,1
-9,19,121,4,3,6,1
-9,20,121,4,3,6,0
-9,21,121,4,3,6,1
-9,22,121,4,3,6,1
-9,23,121,4,3,6,1
-9,24,121,4,3,6,1
-9,25,121,4,3,6,1
-9,26,121,4,3,6,1
-9,27,121,4,3,6,1
-9,28,121,4,3,6,1
-9,29,121,4,3,8,1
-9,30,121,4,3,6,1
-9,31,121,4,3,6,1
-9,32,121,4,3,6,1
-9,33,121,4,3,6,1
-9,34,121,4,3,6,1
-9,35,121,4,3,6,0
-9,36,121,4,3,8,1
-9,37,121,4,3,6,0
+9,8,100,6,0,9,0
+9,9,100,9,8,9,0
+9,10,109,4,9,5,1
+9,11,113,6,0,9,0
+9,12,100,9,0,9,0
+9,13,109,4,9,5,1
+9,14,112,6,3,9,0
+9,15,121,6,0,5,0
+9,16,109,4,0,8,0
+9,17,109,4,0,5,0
+9,18,127,4,0,9,0
+9,19,109,4,9,5,0
+9,20,127,4,0,5,0
+9,21,127,4,0,9,0
+9,22,109,4,9,5,0
+9,23,109,4,9,5,0
+9,24,109,4,3,5,0
+9,25,109,4,3,5,0
+9,26,109,4,3,5,0
+9,27,109,6,3,5,0
+9,28,121,6,3,0,0
+9,29,109,4,3,7,0
+9,30,112,6,3,5,0
+9,31,103,6,3,0,0
+9,32,103,6,2,0,0
+9,33,103,6,2,0,0
+9,34,113,6,2,0,0
+9,35,113,6,2,0,0
+9,36,113,6,2,0,0
+9,37,113,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index 2a8d05f6..9f00e61d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,113,6,2,0,1
-0,9,113,6,2,0,1
-0,10,113,6,2,0,1
-0,11,113,6,2,0,1
-0,12,113,6,2,0,1
-0,13,113,6,2,0,2
-0,14,113,6,2,0,2
-0,15,113,6,2,0,2
-0,16,113,6,2,0,2
-0,17,113,6,2,0,2
-0,18,113,6,2,0,2
-0,19,113,6,2,0,2
-0,20,113,6,2,0,2
-0,21,113,6,2,0,2
-0,22,113,6,2,0,2
-0,23,113,6,2,0,2
-0,24,113,6,2,0,2
-0,25,113,6,2,0,2
-0,26,113,6,2,0,2
-0,27,113,6,2,0,2
-0,28,113,6,2,0,2
-0,29,113,6,2,0,2
-0,30,113,6,2,0,2
-0,31,113,6,2,0,2
-0,32,113,6,2,0,2
-0,33,113,6,2,0,2
-0,34,113,6,2,0,2
-0,35,113,6,2,0,2
-0,36,113,6,2,0,2
-0,37,113,6,2,0,2
+0,8,102,6,3,8,1
+0,9,121,6,3,8,1
+0,10,121,6,3,8,1
+0,11,121,6,3,8,1
+0,12,121,6,3,8,1
+0,13,104,6,3,8,0
+0,14,104,3,3,0,0
+0,15,104,3,3,0,0
+0,16,104,3,3,0,0
+0,17,104,3,3,0,0
+0,18,104,3,3,0,0
+0,19,104,3,3,0,0
+0,20,104,3,3,0,1
+0,21,103,6,3,0,1
+0,22,113,6,3,8,1
+0,23,113,6,3,8,1
+0,24,113,6,3,8,1
+0,25,104,6,3,8,1
+0,26,104,6,3,8,1
+0,27,104,6,3,8,1
+0,28,104,6,3,8,1
+0,29,104,6,3,8,1
+0,30,104,9,3,0,0
+0,31,104,3,3,0,1
+0,32,104,9,3,0,1
+0,33,104,9,3,0,1
+0,34,104,3,3,8,1
+0,35,104,6,3,8,0
+0,36,121,6,3,8,0
+0,37,121,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index fed4cb3b..81ebca01 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,4,2,6,0
-1,9,109,4,9,6,1
-1,10,121,6,2,0,0
-1,11,121,6,2,8,0
-1,12,121,6,0,8,0
-1,13,109,6,0,8,0
-1,14,121,6,0,8,0
-1,15,109,4,9,8,0
-1,16,109,4,9,5,0
-1,17,120,6,0,9,0
-1,18,109,4,0,5,0
-1,19,109,4,0,5,0
-1,20,109,4,9,5,0
-1,21,127,4,0,9,0
-1,22,120,4,0,5,0
-1,23,127,4,0,9,0
-1,24,127,4,0,5,0
-1,25,109,4,9,5,0
-1,26,109,4,3,5,0
-1,27,109,4,3,5,0
-1,28,109,4,3,5,0
-1,29,109,4,9,6,0
-1,30,109,4,3,8,0
-1,31,109,4,9,8,0
-1,32,109,4,0,8,0
-1,33,109,4,0,8,0
-1,34,109,4,9,8,0
-1,35,109,4,0,5,0
-1,36,109,4,0,5,0
-1,37,109,4,0,5,0
+1,8,121,6,2,8,1
+1,9,121,6,3,8,1
+1,10,121,6,3,8,0
+1,11,121,6,3,8,0
+1,12,121,6,3,8,0
+1,13,121,3,3,8,0
+1,14,104,3,3,8,0
+1,15,104,3,3,0,0
+1,16,104,3,3,0,0
+1,17,104,3,3,0,0
+1,18,104,3,3,0,0
+1,19,104,3,3,0,0
+1,20,104,3,3,0,0
+1,21,104,3,3,0,0
+1,22,121,6,3,0,1
+1,23,113,6,3,8,1
+1,24,121,6,3,8,1
+1,25,104,6,3,8,1
+1,26,104,6,3,8,0
+1,27,104,6,3,0,0
+1,28,104,9,3,0,1
+1,29,104,3,3,0,1
+1,30,104,9,3,0,1
+1,31,104,3,3,0,1
+1,32,104,3,3,0,1
+1,33,103,6,3,8,1
+1,34,113,6,3,8,1
+1,35,121,6,3,8,1
+1,36,121,6,3,8,1
+1,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index c4fe8a8e..c3ecaba4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,103,4,2,6,1
-2,9,113,6,2,0,1
-2,10,113,6,2,0,1
-2,11,113,6,2,0,1
-2,12,113,6,2,0,1
-2,13,113,6,2,0,2
-2,14,113,6,2,0,2
-2,15,113,6,2,0,2
-2,16,113,6,2,0,2
-2,17,113,6,2,0,2
-2,18,113,6,2,0,2
-2,19,113,6,2,0,2
-2,20,113,6,2,0,2
-2,21,113,6,2,0,2
-2,22,113,6,2,0,2
-2,23,113,6,2,0,2
-2,24,113,6,2,0,2
-2,25,113,6,2,0,2
-2,26,113,6,2,0,2
-2,27,113,6,2,0,2
-2,28,113,6,2,0,2
-2,29,113,6,2,0,2
-2,30,113,6,2,0,2
-2,31,113,6,2,0,2
-2,32,113,6,2,0,2
-2,33,113,6,2,0,2
-2,34,113,6,2,0,2
-2,35,113,6,2,0,2
-2,36,113,6,2,0,2
-2,37,113,6,2,0,2
+2,8,102,6,3,8,1
+2,9,121,6,3,8,1
+2,10,104,6,3,8,1
+2,11,104,6,3,8,1
+2,12,104,6,3,8,1
+2,13,104,6,3,8,1
+2,14,104,6,3,8,0
+2,15,104,9,3,0,0
+2,16,104,3,3,0,0
+2,17,104,3,3,0,0
+2,18,104,3,3,0,1
+2,19,104,3,3,0,1
+2,20,104,3,3,0,1
+2,21,103,6,3,0,1
+2,22,113,6,3,8,1
+2,23,113,6,3,8,1
+2,24,113,6,3,8,1
+2,25,104,6,3,8,1
+2,26,104,6,3,8,1
+2,27,104,6,3,8,1
+2,28,104,6,3,8,1
+2,29,104,6,3,8,1
+2,30,104,9,3,0,0
+2,31,104,3,3,0,1
+2,32,104,9,3,0,1
+2,33,104,9,3,0,1
+2,34,104,3,3,8,1
+2,35,104,6,3,8,0
+2,36,121,6,3,8,0
+2,37,121,6,3,8,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index 6052a5e2..cfc7190a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,109,4,2,6,1
-3,9,113,4,2,0,1
-3,10,113,6,2,0,0
-3,11,113,6,2,0,0
-3,12,113,6,2,0,0
-3,13,113,6,2,0,0
-3,14,113,6,2,0,0
-3,15,113,6,2,0,0
-3,16,113,6,2,0,1
-3,17,113,6,2,0,1
-3,18,113,6,2,0,1
-3,19,113,6,2,0,2
-3,20,113,6,2,0,2
-3,21,113,6,2,0,2
-3,22,113,6,2,0,2
-3,23,113,6,2,0,2
-3,24,113,6,2,0,2
-3,25,113,6,2,0,2
-3,26,113,6,2,0,2
-3,27,113,6,2,0,2
-3,28,113,6,2,0,2
-3,29,113,6,2,0,2
-3,30,113,6,2,0,2
-3,31,113,6,2,0,2
-3,32,113,6,2,0,2
-3,33,113,6,2,0,2
-3,34,113,6,2,0,2
-3,35,113,6,2,0,2
-3,36,113,6,2,0,2
-3,37,113,6,2,0,2
-4,8,121,4,9,6,1
-4,9,113,4,2,6,1
-4,10,113,6,2,0,0
-4,11,113,6,2,0,0
-4,12,113,6,2,0,0
-4,13,113,6,2,0,1
-4,14,113,6,2,0,1
-4,15,113,6,2,0,1
-4,16,113,6,2,0,2
-4,17,113,6,2,0,2
-4,18,113,6,2,0,2
-4,19,113,6,2,0,2
-4,20,113,6,2,0,2
-4,21,113,6,2,0,2
-4,22,113,6,2,0,2
-4,23,113,6,2,0,2
-4,24,113,6,2,0,2
-4,25,113,6,2,0,2
-4,26,113,6,2,0,2
-4,27,113,6,2,0,2
-4,28,113,6,2,0,2
-4,29,113,6,2,0,2
-4,30,113,6,2,0,2
-4,31,113,6,2,0,2
-4,32,113,6,2,0,2
-4,33,113,6,2,0,2
-4,34,113,6,2,0,2
-4,35,113,6,2,0,2
-4,36,113,6,2,0,2
-4,37,113,6,2,0,2
+3,8,121,6,3,8,1
+3,9,121,6,3,8,0
+3,10,121,3,3,8,0
+3,11,121,3,3,8,0
+3,12,121,3,3,8,0
+3,13,121,3,3,8,0
+3,14,104,3,3,0,0
+3,15,104,3,3,0,0
+3,16,104,3,3,0,0
+3,17,104,3,3,0,0
+3,18,104,3,3,0,0
+3,19,104,3,3,0,0
+3,20,104,3,3,0,0
+3,21,121,3,3,0,1
+3,22,113,6,3,8,1
+3,23,121,6,3,8,1
+3,24,104,6,3,8,1
+3,25,104,6,3,8,0
+3,26,104,6,3,0,0
+3,27,104,3,3,0,1
+3,28,104,9,3,0,1
+3,29,104,3,3,0,1
+3,30,104,9,3,0,1
+3,31,104,3,3,0,1
+3,32,103,6,3,8,1
+3,33,113,6,3,8,1
+3,34,113,6,3,8,1
+3,35,121,6,3,8,1
+3,36,121,6,3,8,1
+3,37,104,6,3,8,1
+4,8,102,6,2,8,1
+4,9,121,6,3,8,1
+4,10,121,6,3,8,1
+4,11,121,6,3,8,1
+4,12,121,6,3,8,1
+4,13,121,6,3,8,0
+4,14,104,6,3,8,0
+4,15,104,3,3,0,0
+4,16,104,3,3,0,0
+4,17,104,3,3,0,0
+4,18,104,3,3,0,0
+4,19,104,3,3,0,0
+4,20,104,3,3,0,0
+4,21,104,3,3,0,0
+4,22,121,6,3,0,1
+4,23,113,6,3,8,1
+4,24,121,6,3,8,1
+4,25,104,6,3,8,1
+4,26,104,6,3,8,0
+4,27,104,6,3,0,0
+4,28,104,9,3,0,1
+4,29,104,3,3,0,1
+4,30,104,9,3,0,1
+4,31,104,3,3,0,1
+4,32,104,3,3,0,1
+4,33,103,6,3,8,1
+4,34,113,6,3,8,1
+4,35,121,6,3,8,1
+4,36,121,6,3,8,1
+4,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index 77123cc4..69802abf 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,103,4,2,0,1
-5,9,113,6,2,0,1
-5,10,113,6,2,0,1
-5,11,113,6,2,0,1
-5,12,113,6,2,0,2
-5,13,113,6,2,0,2
-5,14,113,6,2,0,2
-5,15,113,6,2,0,2
-5,16,113,6,2,0,2
-5,17,113,6,2,0,2
-5,18,113,6,2,0,2
-5,19,113,6,2,0,2
-5,20,113,6,2,0,2
-5,21,113,6,2,0,2
-5,22,113,6,2,0,2
-5,23,113,6,2,0,2
-5,24,113,6,2,0,2
-5,25,113,6,2,0,2
-5,26,113,6,2,0,2
-5,27,113,6,2,0,2
-5,28,113,6,2,0,2
-5,29,113,6,2,0,2
-5,30,113,6,2,0,2
-5,31,113,6,2,0,2
-5,32,113,6,2,0,2
-5,33,113,6,2,0,2
-5,34,113,6,2,0,2
-5,35,113,6,2,0,2
-5,36,113,6,2,0,2
-5,37,113,6,2,0,2
+5,8,113,6,3,8,1
+5,9,121,6,3,8,0
+5,10,104,6,3,8,0
+5,11,104,3,3,0,0
+5,12,104,3,3,0,0
+5,13,104,3,3,0,0
+5,14,121,6,3,8,1
+5,15,104,3,3,8,0
+5,16,104,3,3,0,0
+5,17,104,3,3,0,0
+5,18,104,3,3,0,0
+5,19,121,3,3,0,1
+5,20,104,4,3,8,1
+5,21,104,3,3,0,0
+5,22,104,3,3,0,0
+5,23,104,3,3,0,1
+5,24,103,6,3,8,1
+5,25,113,6,3,8,1
+5,26,104,6,3,8,1
+5,27,104,9,3,8,1
+5,28,121,6,3,8,1
+5,29,104,6,3,8,1
+5,30,104,6,3,8,0
+5,31,104,6,3,0,0
+5,32,104,9,3,0,0
+5,33,104,3,3,0,1
+5,34,104,3,3,0,0
+5,35,104,3,3,0,1
+5,36,104,3,3,0,1
+5,37,103,6,3,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index f7596c81..4ff6731c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,121,4,2,6,0
-6,9,121,4,2,6,1
-6,10,113,4,2,0,1
-6,11,113,6,2,0,0
-6,12,113,6,2,0,0
-6,13,113,6,2,0,0
-6,14,113,6,2,0,0
-6,15,113,6,2,0,0
-6,16,113,6,2,0,0
-6,17,113,6,2,0,1
-6,18,113,6,2,0,1
-6,19,113,6,2,0,1
-6,20,113,6,2,0,2
-6,21,113,6,2,0,2
-6,22,113,6,2,0,2
-6,23,113,6,2,0,2
-6,24,113,6,2,0,2
-6,25,113,6,2,0,2
-6,26,113,6,2,0,2
-6,27,113,6,2,0,2
-6,28,113,6,2,0,2
-6,29,113,6,2,0,2
-6,30,113,6,2,0,2
-6,31,113,6,2,0,2
-6,32,113,6,2,0,2
-6,33,113,6,2,0,2
-6,34,113,6,2,0,2
-6,35,113,6,2,0,2
-6,36,113,6,2,0,2
-6,37,113,6,2,0,2
-7,8,109,4,9,6,0
-7,9,109,4,9,6,0
-7,10,121,4,9,6,0
-7,11,121,4,2,8,1
-7,12,109,4,9,8,0
-7,13,109,4,9,8,0
-7,14,109,4,0,5,0
-7,15,109,6,0,5,0
-7,16,109,6,0,8,0
-7,17,109,4,0,5,0
-7,18,109,4,0,5,0
-7,19,127,4,0,5,0
-7,20,127,4,0,5,0
-7,21,109,4,9,5,0
-7,22,109,4,3,0,0
-7,23,109,4,3,0,0
-7,24,109,4,9,5,0
-7,25,109,4,3,0,0
-7,26,109,4,3,0,0
-7,27,121,4,3,0,0
-7,28,109,4,9,8,0
-7,29,109,4,0,8,0
-7,30,109,4,9,8,0
-7,31,109,4,9,8,0
-7,32,109,4,0,5,0
-7,33,109,4,9,5,0
-7,34,109,4,0,5,0
-7,35,109,4,0,5,0
-7,36,127,4,0,5,0
-7,37,109,4,0,5,0
+6,8,102,6,2,8,1
+6,9,113,6,3,8,1
+6,10,121,6,3,8,1
+6,11,121,6,3,8,1
+6,12,121,6,3,8,1
+6,13,121,6,3,8,0
+6,14,121,6,3,8,0
+6,15,104,3,3,8,0
+6,16,104,3,3,0,0
+6,17,104,3,3,0,0
+6,18,104,3,3,0,0
+6,19,104,3,3,0,0
+6,20,104,3,3,0,0
+6,21,104,3,3,0,0
+6,22,104,3,3,0,0
+6,23,121,6,3,0,1
+6,24,113,6,3,8,1
+6,25,121,6,3,8,1
+6,26,104,6,3,8,1
+6,27,104,6,3,8,0
+6,28,104,6,3,0,0
+6,29,104,9,3,0,1
+6,30,104,3,3,0,1
+6,31,104,9,3,0,1
+6,32,104,3,3,0,1
+6,33,104,3,3,0,1
+6,34,103,6,3,8,1
+6,35,113,6,3,8,1
+6,36,121,6,3,8,1
+6,37,121,6,3,8,1
+7,8,102,6,2,8,1
+7,9,121,6,3,8,1
+7,10,121,6,3,8,1
+7,11,104,6,3,8,0
+7,12,121,6,3,8,0
+7,13,104,6,3,8,0
+7,14,104,6,3,0,0
+7,15,104,3,3,0,0
+7,16,104,3,3,0,0
+7,17,104,3,3,0,0
+7,18,104,3,3,0,0
+7,19,104,3,3,0,0
+7,20,104,3,3,0,1
+7,21,103,6,3,8,1
+7,22,113,6,3,8,1
+7,23,113,6,3,8,1
+7,24,104,6,3,8,1
+7,25,104,6,3,8,1
+7,26,104,6,3,8,1
+7,27,104,6,3,8,1
+7,28,104,6,3,8,0
+7,29,104,9,3,0,0
+7,30,104,3,3,0,1
+7,31,104,9,3,0,1
+7,32,104,3,3,0,1
+7,33,104,9,3,0,1
+7,34,103,3,3,8,1
+7,35,104,6,3,8,1
+7,36,121,6,3,8,1
+7,37,121,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index 49dff757..7d0ff1de 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,103,4,2,0,0
-8,9,113,6,2,0,0
-8,10,113,6,2,0,0
-8,11,113,6,2,0,0
-8,12,113,6,2,0,0
-8,13,113,6,2,0,0
-8,14,113,6,7,0,0
-8,15,113,6,7,0,0
-8,16,113,6,2,0,1
-8,17,113,6,2,0,1
-8,18,113,6,2,0,1
-8,19,113,6,2,0,2
-8,20,113,6,2,0,2
-8,21,113,6,2,0,2
-8,22,113,6,2,0,2
-8,23,113,6,2,0,2
-8,24,113,6,2,0,2
-8,25,113,6,2,0,2
-8,26,113,6,2,0,2
-8,27,113,6,2,0,2
-8,28,113,6,2,0,2
-8,29,113,6,2,0,2
-8,30,113,6,2,0,2
-8,31,113,6,2,0,2
-8,32,113,6,2,0,2
-8,33,113,6,2,0,2
-8,34,113,6,2,0,2
-8,35,113,6,2,0,2
-8,36,113,6,2,0,2
-8,37,113,6,2,0,2
+8,8,113,6,3,8,1
+8,9,121,6,3,8,1
+8,10,121,6,3,8,1
+8,11,104,6,3,8,1
+8,12,104,6,3,8,0
+8,13,121,6,3,8,0
+8,14,104,6,3,8,0
+8,15,104,3,3,0,0
+8,16,104,3,3,0,0
+8,17,104,3,3,0,0
+8,18,104,3,3,0,0
+8,19,104,3,3,0,0
+8,20,104,3,3,0,0
+8,21,104,3,3,0,0
+8,22,121,6,3,0,1
+8,23,113,6,3,8,1
+8,24,121,6,3,8,1
+8,25,104,6,3,8,1
+8,26,104,6,3,8,0
+8,27,104,6,3,0,0
+8,28,104,9,3,0,1
+8,29,104,3,3,0,1
+8,30,104,9,3,0,1
+8,31,104,3,3,0,1
+8,32,104,3,3,0,1
+8,33,103,6,3,8,1
+8,34,113,6,3,8,1
+8,35,121,6,3,8,1
+8,36,121,6,3,8,1
+8,37,104,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index 4a0c0385..057d4027 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,103,4,3,0,0
-9,9,103,4,2,0,0
-9,10,121,6,2,0,0
-9,11,121,6,2,0,0
-9,12,121,6,2,0,1
-9,13,113,6,2,8,0
-9,14,121,6,2,8,0
-9,15,121,6,2,8,0
-9,16,121,4,2,8,0
-9,17,109,4,9,8,0
-9,18,109,4,0,5,0
-9,19,109,4,0,5,0
-9,20,120,4,0,5,0
-9,21,109,4,0,5,0
-9,22,109,4,9,5,0
-9,23,109,4,0,5,0
-9,24,109,4,9,5,0
-9,25,127,4,0,9,0
-9,26,109,4,9,5,0
-9,27,109,4,3,9,0
-9,28,109,4,0,5,0
-9,29,109,4,3,9,0
-9,30,120,4,0,5,0
-9,31,109,4,3,9,0
-9,32,120,4,0,5,0
-9,33,109,4,9,5,0
-9,34,127,4,0,9,0
-9,35,109,4,9,5,0
-9,36,109,4,9,5,0
-9,37,109,4,0,5,0
+9,8,113,6,3,8,1
+9,9,121,6,3,8,1
+9,10,104,6,3,8,1
+9,11,104,6,3,8,1
+9,12,104,6,3,8,0
+9,13,104,6,3,8,0
+9,14,104,3,3,0,0
+9,15,104,3,3,0,0
+9,16,104,3,3,0,0
+9,17,104,3,3,0,0
+9,18,104,3,3,0,1
+9,19,103,4,3,0,1
+9,20,103,6,3,8,1
+9,21,113,6,3,8,1
+9,22,113,6,3,8,1
+9,23,113,6,3,8,1
+9,24,104,6,3,8,1
+9,25,104,6,3,8,1
+9,26,104,6,3,8,1
+9,27,104,6,3,8,1
+9,28,104,9,3,8,0
+9,29,104,3,3,0,0
+9,30,104,3,3,0,0
+9,31,104,3,3,0,1
+9,32,104,9,3,0,1
+9,33,104,3,3,0,1
+9,34,104,3,3,0,1
+9,35,103,6,3,8,1
+9,36,121,6,3,8,1
+9,37,113,6,3,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index 68b4bc7d..9e3ea8c3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -1,15 +1,15 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,104,4,2,8,1
-0,9,102,6,2,8,1
-0,10,102,6,2,8,0
+0,8,102,6,2,0,1
+0,9,102,6,2,0,1
+0,10,102,6,2,0,1
 0,11,102,6,2,8,1
 0,12,102,6,2,8,1
 0,13,102,6,2,8,1
 0,14,102,6,2,6,1
 0,15,102,6,2,8,1
-0,16,102,6,2,6,1
-0,17,102,6,2,8,1
-0,18,102,6,2,6,1
+0,16,102,6,2,0,1
+0,17,102,6,2,0,1
+0,18,102,6,2,0,1
 0,19,102,6,2,0,1
 0,20,102,6,2,0,1
 0,21,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index 875d7e7d..2a623d17 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,15 +1,15 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,104,4,7,6,0
-1,9,103,4,9,8,1
-1,10,102,6,2,8,0
+1,8,102,6,2,8,1
+1,9,102,6,2,8,1
+1,10,102,6,2,8,1
 1,11,102,6,2,8,1
 1,12,102,6,2,8,1
 1,13,102,6,2,8,1
 1,14,102,6,2,8,1
 1,15,102,6,2,8,1
-1,16,102,6,2,8,1
-1,17,102,6,2,6,1
-1,18,102,6,2,8,1
+1,16,102,6,2,6,1
+1,17,102,6,2,0,1
+1,18,102,6,2,0,1
 1,19,102,6,2,0,1
 1,20,102,6,2,0,1
 1,21,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index aaa36998..a8977415 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,16 +1,16 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,110,4,2,6,0
-2,9,110,4,2,8,1
-2,10,110,6,2,8,0
-2,11,102,6,2,8,0
+2,8,102,6,2,8,1
+2,9,102,6,2,8,1
+2,10,102,6,2,8,1
+2,11,102,6,2,8,1
 2,12,102,6,2,8,1
 2,13,102,6,2,8,1
-2,14,102,6,2,8,1
-2,15,102,6,2,8,1
-2,16,102,6,2,8,1
-2,17,102,6,2,6,1
-2,18,102,6,2,8,1
-2,19,102,6,2,6,1
+2,14,102,6,2,0,1
+2,15,102,6,2,0,1
+2,16,102,6,2,0,1
+2,17,102,6,2,0,1
+2,18,102,6,2,0,1
+2,19,102,6,2,0,1
 2,20,102,6,2,0,1
 2,21,102,6,2,0,1
 2,22,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index 9d5383b8..780b6ded 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -1,15 +1,15 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,104,4,2,6,0
-3,9,103,6,5,8,1
-3,10,102,4,2,6,0
+3,8,102,6,2,8,1
+3,9,102,6,2,8,1
+3,10,102,6,2,8,1
 3,11,102,6,2,8,1
 3,12,102,6,2,8,1
 3,13,102,6,2,8,1
 3,14,102,6,2,8,1
 3,15,102,6,2,8,1
-3,16,102,6,2,8,1
-3,17,102,6,2,8,1
-3,18,102,6,2,6,1
+3,16,102,6,2,6,1
+3,17,102,6,2,0,1
+3,18,102,6,2,0,1
 3,19,102,6,2,0,1
 3,20,102,6,2,0,1
 3,21,102,6,2,0,1
@@ -29,14 +29,14 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 3,35,102,6,2,0,1
 3,36,102,6,2,0,1
 3,37,102,6,2,0,1
-4,8,110,4,2,6,0
-4,9,110,4,2,8,1
-4,10,102,6,2,6,0
-4,11,102,6,2,8,1
-4,12,102,6,2,8,1
+4,8,113,5,2,0,1
+4,9,102,6,2,0,1
+4,10,102,6,2,0,1
+4,11,102,6,2,0,1
+4,12,102,6,2,0,1
 4,13,102,6,2,0,1
 4,14,102,6,2,8,1
-4,15,102,6,2,6,1
+4,15,102,6,2,8,1
 4,16,102,6,2,0,1
 4,17,102,6,2,0,1
 4,18,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index 956ddf9b..a6dfbfa4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -1,8 +1,8 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,110,4,9,6,0
-5,9,110,4,2,8,1
-5,10,110,4,2,6,0
-5,11,113,6,2,8,2
+5,8,113,7,2,4,2
+5,9,113,5,2,0,1
+5,10,113,6,2,0,1
+5,11,113,6,2,0,1
 5,12,113,6,2,0,1
 5,13,113,6,2,0,1
 5,14,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index b4ef4bd3..eeb5e8ef 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -1,14 +1,14 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,110,4,2,8,0
-6,9,102,6,2,8,0
+6,8,102,6,2,8,1
+6,9,102,6,2,8,1
 6,10,102,6,2,8,1
 6,11,102,6,2,8,1
 6,12,102,6,2,8,1
 6,13,102,6,2,8,1
 6,14,102,6,2,8,1
-6,15,102,6,3,6,1
-6,16,102,6,2,8,1
-6,17,102,6,2,6,1
+6,15,102,6,2,8,1
+6,16,102,6,2,6,1
+6,17,102,6,2,0,1
 6,18,102,6,2,0,1
 6,19,102,6,2,0,1
 6,20,102,6,2,0,1
@@ -29,17 +29,17 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,35,102,6,2,0,1
 6,36,102,6,2,0,1
 6,37,102,6,2,0,1
-7,8,104,4,7,6,0
-7,9,124,4,9,8,1
-7,10,110,6,2,8,0
-7,11,102,4,2,6,0
+7,8,102,6,2,8,1
+7,9,102,6,2,8,1
+7,10,102,6,2,8,1
+7,11,102,6,2,8,1
 7,12,102,6,2,8,1
 7,13,102,6,2,8,1
-7,14,102,6,2,6,1
+7,14,102,6,3,6,1
 7,15,102,6,2,8,1
-7,16,102,6,2,8,1
-7,17,102,6,2,6,1
-7,18,102,6,2,8,1
+7,16,102,6,2,0,1
+7,17,102,6,2,0,1
+7,18,102,6,2,0,1
 7,19,102,6,2,0,1
 7,20,102,6,2,0,1
 7,21,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index 11f63b7e..811522ec 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -1,15 +1,15 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,110,4,2,6,0
-8,9,102,4,2,8,1
-8,10,102,6,2,8,0
+8,8,102,6,2,8,0
+8,9,102,6,2,8,1
+8,10,102,6,2,8,1
 8,11,102,6,2,8,1
 8,12,102,6,2,8,1
 8,13,102,6,2,8,1
-8,14,102,6,2,8,1
-8,15,102,6,2,6,1
-8,16,102,6,2,8,1
-8,17,102,6,2,6,1
-8,18,102,6,2,8,1
+8,14,102,6,2,6,1
+8,15,102,6,3,6,1
+8,16,102,6,2,6,1
+8,17,102,6,2,0,1
+8,18,102,6,2,0,1
 8,19,102,6,2,0,1
 8,20,102,6,2,0,1
 8,21,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index c00f4102..12621cb2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -1,17 +1,17 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,110,4,8,6,0
-9,9,110,4,2,8,0
-9,10,102,4,2,8,0
-9,11,102,6,2,8,1
-9,12,102,6,2,8,0
+9,8,113,6,2,6,1
+9,9,102,6,2,0,1
+9,10,102,6,2,0,1
+9,11,102,6,2,0,1
+9,12,102,6,2,0,1
 9,13,102,6,2,0,1
 9,14,102,6,2,0,1
 9,15,102,6,2,0,1
-9,16,102,6,2,8,1
-9,17,102,6,2,8,1
-9,18,102,6,2,6,1
-9,19,102,6,2,8,1
-9,20,102,6,2,6,1
+9,16,102,6,2,0,1
+9,17,102,6,2,0,1
+9,18,102,6,2,0,1
+9,19,102,6,2,0,1
+9,20,102,6,2,0,1
 9,21,102,6,2,0,1
 9,22,102,6,2,0,1
 9,23,102,6,2,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
index 0920b6c1..2a8d05f6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,121,4,9,2,1
-0,9,113,9,7,8,1
-0,10,113,3,1,6,1
-0,11,113,9,3,2,1
-0,12,127,6,1,7,1
-0,13,121,6,1,7,1
-0,14,121,9,1,7,1
-0,15,121,6,0,7,1
-0,16,121,6,0,7,1
-0,17,121,6,7,7,1
-0,18,113,6,7,7,1
-0,19,121,6,7,2,1
-0,20,113,6,7,2,0
-0,21,113,4,7,2,0
-0,22,121,6,1,2,0
-0,23,127,4,0,8,1
-0,24,127,6,0,2,0
-0,25,127,4,0,8,1
-0,26,127,4,1,8,1
-0,27,127,4,1,8,0
-0,28,127,4,1,8,1
-0,29,127,4,1,7,1
-0,30,127,0,1,0,1
-0,31,113,1,4,0,1
-0,32,113,1,9,7,1
-0,33,113,1,7,7,1
-0,34,113,6,9,0,1
-0,35,113,1,7,7,1
-0,36,113,6,7,7,1
-0,37,113,6,7,2,1
+0,8,113,6,2,0,1
+0,9,113,6,2,0,1
+0,10,113,6,2,0,1
+0,11,113,6,2,0,1
+0,12,113,6,2,0,1
+0,13,113,6,2,0,2
+0,14,113,6,2,0,2
+0,15,113,6,2,0,2
+0,16,113,6,2,0,2
+0,17,113,6,2,0,2
+0,18,113,6,2,0,2
+0,19,113,6,2,0,2
+0,20,113,6,2,0,2
+0,21,113,6,2,0,2
+0,22,113,6,2,0,2
+0,23,113,6,2,0,2
+0,24,113,6,2,0,2
+0,25,113,6,2,0,2
+0,26,113,6,2,0,2
+0,27,113,6,2,0,2
+0,28,113,6,2,0,2
+0,29,113,6,2,0,2
+0,30,113,6,2,0,2
+0,31,113,6,2,0,2
+0,32,113,6,2,0,2
+0,33,113,6,2,0,2
+0,34,113,6,2,0,2
+0,35,113,6,2,0,2
+0,36,113,6,2,0,2
+0,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
index a73055c6..cd416b33 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,0,7,6,0
-1,9,121,6,3,8,0
-1,10,121,9,0,8,0
-1,11,121,4,0,8,0
-1,12,127,6,0,7,0
-1,13,121,6,0,7,1
-1,14,127,6,7,7,1
-1,15,127,6,7,7,1
-1,16,127,4,1,7,1
-1,17,127,4,1,2,1
-1,18,127,4,1,2,1
-1,19,127,6,7,2,1
-1,20,113,6,7,0,1
-1,21,113,1,7,0,1
-1,22,113,1,7,0,1
-1,23,113,1,7,0,1
-1,24,113,1,7,0,1
-1,25,113,1,7,0,1
-1,26,113,6,7,2,1
-1,27,113,6,1,2,1
-1,28,113,6,0,2,1
-1,29,113,6,0,2,1
-1,30,113,6,0,2,1
-1,31,113,6,0,2,1
-1,32,113,1,8,2,1
-1,33,113,1,1,8,1
-1,34,113,1,0,8,1
-1,35,113,1,8,8,1
-1,36,113,1,0,7,1
-1,37,113,1,8,7,1
+1,8,113,6,0,9,0
+1,9,100,6,0,8,0
+1,10,121,6,0,8,0
+1,11,120,4,0,8,0
+1,12,120,4,0,8,0
+1,13,120,0,0,5,0
+1,14,109,4,8,8,0
+1,15,109,4,9,5,0
+1,16,109,4,9,5,0
+1,17,109,4,9,5,0
+1,18,109,4,0,5,0
+1,19,109,4,9,5,0
+1,20,109,4,0,9,0
+1,21,109,4,0,5,0
+1,22,127,4,0,9,0
+1,23,120,4,0,9,0
+1,24,120,4,0,9,0
+1,25,109,4,0,5,0
+1,26,109,4,9,5,0
+1,27,127,4,0,9,0
+1,28,109,4,9,5,0
+1,29,109,4,9,5,0
+1,30,109,4,3,5,0
+1,31,109,4,3,5,0
+1,32,109,4,3,5,0
+1,33,109,4,3,5,0
+1,34,127,6,3,5,0
+1,35,109,6,3,5,0
+1,36,112,6,3,5,0
+1,37,103,6,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
index 087f11a3..cedb1297 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,121,4,7,8,0
-2,9,121,0,3,7,0
-2,10,121,0,7,6,1
-2,11,113,0,7,6,0
-2,12,121,4,1,6,1
-2,13,127,0,7,7,1
-2,14,127,6,7,0,1
-2,15,127,6,1,0,1
-2,16,127,6,7,2,1
-2,17,127,6,7,0,1
-2,18,127,6,1,0,1
-2,19,127,6,1,0,1
-2,20,113,6,7,0,1
-2,21,113,1,7,0,1
-2,22,113,1,7,0,1
-2,23,113,1,7,0,1
-2,24,113,1,7,0,1
-2,25,113,6,7,2,1
-2,26,113,6,1,2,1
-2,27,113,6,0,2,1
-2,28,113,6,0,2,1
-2,29,113,6,0,2,1
-2,30,113,6,0,2,1
-2,31,113,1,8,2,1
-2,32,113,1,1,8,1
-2,33,113,1,0,8,1
-2,34,113,1,8,8,1
-2,35,113,1,0,7,1
-2,36,113,1,8,7,1
-2,37,121,6,7,7,1
+2,8,113,4,2,0,1
+2,9,113,6,2,0,0
+2,10,113,6,2,0,0
+2,11,113,6,2,0,1
+2,12,113,6,2,0,1
+2,13,113,6,2,0,1
+2,14,113,6,2,0,2
+2,15,113,6,2,0,2
+2,16,113,6,2,0,2
+2,17,113,6,2,0,2
+2,18,113,6,2,0,2
+2,19,113,6,2,0,2
+2,20,113,6,2,0,2
+2,21,113,6,2,0,2
+2,22,113,6,2,0,2
+2,23,113,6,2,0,2
+2,24,113,6,2,0,2
+2,25,113,6,2,0,2
+2,26,113,6,2,0,2
+2,27,113,6,2,0,2
+2,28,113,6,2,0,2
+2,29,113,6,2,0,2
+2,30,113,6,2,0,2
+2,31,113,6,2,0,2
+2,32,113,6,2,0,2
+2,33,113,6,2,0,2
+2,34,113,6,2,0,2
+2,35,113,6,2,0,2
+2,36,113,6,2,0,2
+2,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
index c6ba0686..477f3560 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,4,9,6,0
-3,9,121,0,3,8,1
-3,10,121,0,7,8,0
-3,11,121,4,0,8,0
-3,12,127,4,0,8,0
-3,13,127,4,0,7,0
-3,14,127,6,0,7,1
-3,15,127,6,7,7,1
-3,16,127,6,7,7,1
-3,17,127,6,1,7,1
-3,18,127,6,1,7,1
-3,19,124,6,7,2,1
-3,20,113,6,7,2,1
-3,21,113,6,7,2,1
-3,22,113,6,7,2,1
-3,23,113,6,1,2,1
-3,24,113,1,2,8,1
-3,25,113,1,1,8,1
-3,26,113,1,0,8,1
-3,27,113,1,0,8,1
-3,28,113,1,0,7,1
-3,29,121,1,8,7,1
-3,30,121,6,0,7,1
-3,31,121,6,0,7,1
-3,32,121,6,7,7,1
-3,33,113,6,7,7,1
-3,34,121,6,7,7,1
-3,35,113,6,7,7,1
-3,36,113,6,7,2,1
-3,37,113,6,7,2,1
-4,8,113,4,9,8,1
-4,9,121,1,7,2,0
-4,10,113,3,7,8,0
-4,11,121,6,9,6,0
-4,12,121,9,1,7,1
-4,13,127,6,9,7,1
-4,14,127,6,1,7,1
-4,15,127,6,1,7,1
-4,16,127,9,1,7,1
-4,17,127,4,1,2,1
-4,18,127,6,1,2,1
-4,19,127,4,1,2,1
-4,20,113,6,2,2,1
-4,21,113,1,3,2,1
-4,22,113,6,7,8,1
-4,23,113,1,1,0,1
-4,24,113,1,2,8,1
-4,25,113,1,1,7,1
-4,26,113,1,8,2,1
-4,27,113,6,7,7,1
-4,28,121,6,0,2,1
-4,29,113,6,0,7,1
-4,30,121,6,8,2,1
-4,31,113,6,7,7,1
-4,32,121,6,7,2,1
-4,33,113,6,7,7,1
-4,34,121,6,7,2,1
-4,35,113,6,7,2,1
-4,36,113,4,1,2,1
-4,37,113,6,0,2,0
+3,8,121,4,2,8,1
+3,9,121,4,2,8,1
+3,10,121,4,9,8,0
+3,11,109,4,9,5,0
+3,12,109,6,9,5,0
+3,13,121,6,0,5,0
+3,14,109,4,9,5,0
+3,15,109,4,0,5,0
+3,16,120,8,0,9,0
+3,17,120,4,0,5,0
+3,18,127,4,0,9,0
+3,19,120,4,0,9,0
+3,20,109,4,0,8,0
+3,21,109,4,9,5,0
+3,22,109,4,0,5,0
+3,23,109,4,9,5,0
+3,24,109,4,9,5,0
+3,25,109,4,3,5,0
+3,26,109,4,9,5,0
+3,27,127,4,0,9,0
+3,28,127,4,0,5,0
+3,29,127,4,0,9,0
+3,30,109,4,0,5,0
+3,31,109,4,0,9,0
+3,32,109,4,0,5,0
+3,33,109,4,9,5,0
+3,34,109,4,9,5,0
+3,35,109,4,9,5,0
+3,36,109,4,3,5,0
+3,37,109,4,3,5,0
+4,8,113,7,2,8,1
+4,9,113,6,2,8,0
+4,10,121,6,2,8,0
+4,11,121,4,2,8,0
+4,12,109,4,9,5,0
+4,13,109,6,9,5,0
+4,14,121,6,0,9,0
+4,15,120,4,0,5,0
+4,16,120,9,0,9,0
+4,17,120,0,0,5,0
+4,18,109,4,0,5,0
+4,19,109,4,0,5,0
+4,20,127,4,9,9,0
+4,21,127,4,9,5,0
+4,22,109,4,9,5,0
+4,23,109,4,9,5,0
+4,24,109,4,9,5,0
+4,25,109,6,3,9,0
+4,26,109,4,0,5,0
+4,27,127,6,0,9,0
+4,28,127,9,0,5,0
+4,29,127,4,0,9,0
+4,30,120,4,0,5,0
+4,31,109,4,3,9,0
+4,32,109,4,9,5,0
+4,33,109,4,9,5,0
+4,34,109,4,9,5,0
+4,35,109,4,9,5,0
+4,36,109,4,3,5,0
+4,37,109,4,3,5,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
index 67fdd593..e21fb885 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,0,7,4,1
-5,9,113,1,9,6,1
-5,10,113,1,9,2,1
-5,11,113,1,7,2,1
-5,12,113,6,9,7,1
-5,13,121,1,3,2,1
-5,14,113,6,7,7,1
-5,15,121,6,3,2,1
-5,16,113,6,0,7,1
-5,17,121,1,8,2,1
-5,18,113,6,7,7,1
-5,19,121,6,1,2,1
-5,20,113,6,0,2,1
-5,21,113,4,7,2,1
-5,22,113,6,1,2,1
-5,23,113,1,0,2,1
-5,24,113,1,7,2,1
-5,25,113,1,7,7,1
-5,26,113,1,7,2,1
-5,27,113,1,1,7,1
-5,28,121,6,9,2,1
-5,29,113,6,0,7,1
-5,30,121,1,8,2,1
-5,31,113,6,7,7,1
-5,32,121,6,7,7,1
-5,33,113,6,7,7,1
-5,34,121,6,7,2,1
-5,35,113,6,7,7,1
-5,36,121,6,1,2,1
-5,37,113,6,7,8,1
+5,8,113,6,2,1,1
+5,9,113,6,2,0,2
+5,10,113,6,2,0,2
+5,11,113,6,2,0,2
+5,12,113,6,2,0,2
+5,13,113,6,2,0,2
+5,14,113,6,2,0,2
+5,15,113,6,2,0,2
+5,16,113,6,2,0,2
+5,17,113,6,2,0,2
+5,18,113,6,2,0,2
+5,19,113,6,2,0,2
+5,20,113,6,2,0,2
+5,21,113,6,2,0,2
+5,22,113,6,2,0,2
+5,23,113,6,2,0,2
+5,24,113,6,2,0,2
+5,25,113,6,2,0,2
+5,26,113,6,2,0,2
+5,27,113,6,2,0,2
+5,28,113,6,2,0,2
+5,29,113,6,2,0,2
+5,30,113,6,2,0,2
+5,31,113,6,2,0,2
+5,32,113,6,2,0,2
+5,33,113,6,2,0,2
+5,34,113,6,2,0,2
+5,35,113,6,2,0,2
+5,36,113,6,2,0,2
+5,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
index f223e485..908877f2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,121,4,7,8,0
-6,9,121,6,7,8,0
-6,10,121,6,7,7,0
-6,11,121,6,7,7,1
-6,12,127,6,7,7,1
-6,13,127,4,7,0,1
-6,14,127,4,1,0,1
-6,15,127,4,1,0,1
-6,16,127,4,1,0,1
-6,17,127,4,1,2,1
-6,18,127,6,7,8,1
-6,19,113,6,1,0,1
-6,20,113,1,7,0,1
-6,21,113,6,7,0,1
-6,22,113,1,7,0,1
-6,23,113,1,7,0,1
-6,24,113,1,7,0,1
-6,25,113,6,7,0,1
-6,26,113,6,1,2,1
-6,27,113,6,0,2,1
-6,28,113,6,0,2,1
-6,29,113,6,0,2,1
-6,30,113,6,0,2,1
-6,31,113,6,8,2,1
-6,32,113,1,1,2,1
-6,33,113,1,8,8,1
-6,34,113,1,0,8,1
-6,35,113,1,8,7,1
-6,36,121,1,8,2,1
-6,37,113,6,7,7,1
-7,8,121,4,9,6,0
-7,9,121,0,3,8,0
-7,10,121,0,3,6,0
-7,11,121,0,3,7,0
-7,12,121,0,9,7,0
-7,13,127,0,9,7,1
-7,14,127,6,7,7,1
-7,15,127,6,1,7,1
-7,16,127,6,7,2,1
-7,17,127,4,1,8,1
-7,18,127,6,1,8,1
-7,19,127,6,1,8,1
-7,20,127,6,1,0,1
-7,21,113,1,7,0,1
-7,22,113,1,7,0,1
-7,23,113,1,7,0,1
-7,24,113,1,1,0,1
-7,25,113,1,4,0,1
-7,26,113,1,7,7,1
-7,27,113,6,7,2,1
-7,28,113,6,7,2,1
-7,29,113,6,7,7,1
-7,30,121,6,7,2,1
-7,31,113,6,7,7,1
-7,32,121,6,7,2,1
-7,33,113,6,7,7,1
-7,34,121,6,1,2,1
-7,35,127,6,0,8,1
-7,36,127,6,0,2,1
-7,37,127,4,0,2,1
+6,8,113,6,2,8,0
+6,9,113,6,2,8,0
+6,10,113,6,2,8,0
+6,11,121,6,2,8,0
+6,12,121,6,0,8,0
+6,13,109,6,0,8,0
+6,14,109,4,0,8,0
+6,15,109,4,0,5,0
+6,16,109,4,0,5,0
+6,17,109,4,9,5,0
+6,18,109,4,0,5,0
+6,19,109,4,9,5,0
+6,20,109,4,9,5,0
+6,21,109,4,3,9,0
+6,22,109,4,0,5,0
+6,23,109,4,3,9,0
+6,24,109,4,0,5,0
+6,25,127,4,0,9,0
+6,26,109,4,0,5,0
+6,27,127,4,0,9,0
+6,28,109,4,0,5,0
+6,29,109,4,9,5,0
+6,30,109,4,9,5,0
+6,31,109,4,9,5,0
+6,32,109,4,3,5,0
+6,33,109,4,3,5,0
+6,34,109,4,3,5,0
+6,35,109,4,3,5,0
+6,36,109,6,3,5,0
+6,37,127,6,3,9,0
+7,8,109,4,9,6,0
+7,9,121,4,9,6,0
+7,10,121,4,2,8,1
+7,11,109,4,9,8,0
+7,12,109,4,9,5,0
+7,13,109,4,9,5,0
+7,14,109,6,3,5,0
+7,15,121,6,0,5,0
+7,16,109,6,0,8,0
+7,17,109,4,0,5,0
+7,18,127,4,0,9,0
+7,19,120,4,0,5,0
+7,20,127,4,3,9,0
+7,21,109,4,9,5,0
+7,22,109,4,3,0,0
+7,23,109,4,9,5,0
+7,24,109,4,9,5,0
+7,25,109,4,3,5,0
+7,26,109,4,3,5,0
+7,27,109,6,3,5,0
+7,28,109,6,3,5,0
+7,29,112,6,3,9,0
+7,30,121,6,3,5,0
+7,31,112,6,3,9,0
+7,32,121,6,3,5,0
+7,33,103,6,3,0,0
+7,34,103,6,3,0,0
+7,35,103,6,2,0,0
+7,36,103,6,2,0,0
+7,37,103,6,2,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
index c09630f3..d1cabc88 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,4,3,8,0
-8,9,121,1,3,8,0
-8,10,121,9,0,7,0
-8,11,121,4,9,7,1
-8,12,127,6,7,7,1
-8,13,127,6,7,7,1
-8,14,127,6,7,0,1
-8,15,127,6,1,0,1
-8,16,127,4,1,2,1
-8,17,127,4,1,2,1
-8,18,127,6,7,2,1
-8,19,113,6,7,0,1
-8,20,113,1,1,0,1
-8,21,113,1,2,8,1
-8,22,113,1,1,0,1
-8,23,113,1,2,8,1
-8,24,113,1,9,7,1
-8,25,113,6,7,7,1
-8,26,121,6,1,2,1
-8,27,113,6,0,7,1
-8,28,121,6,7,2,1
-8,29,113,6,0,7,1
-8,30,121,6,7,2,1
-8,31,113,6,0,7,1
-8,32,121,1,7,2,1
-8,33,113,6,7,8,1
-8,34,113,1,1,2,1
-8,35,113,6,7,8,1
-8,36,102,1,1,2,1
-8,37,113,6,7,8,0
+8,8,113,6,2,8,0
+8,9,113,6,2,8,0
+8,10,113,6,2,8,0
+8,11,113,6,2,0,0
+8,12,113,6,2,0,0
+8,13,113,6,8,0,0
+8,14,113,6,8,0,0
+8,15,113,9,7,0,0
+8,16,113,6,2,0,1
+8,17,113,6,2,0,1
+8,18,113,6,2,0,1
+8,19,113,6,2,0,1
+8,20,113,6,2,0,2
+8,21,113,6,2,0,2
+8,22,113,6,2,0,2
+8,23,113,6,2,0,2
+8,24,113,6,2,0,2
+8,25,113,6,2,0,2
+8,26,113,6,2,0,2
+8,27,113,6,2,0,2
+8,28,113,6,2,0,2
+8,29,113,6,2,0,2
+8,30,113,6,2,0,2
+8,31,113,6,2,0,2
+8,32,113,6,2,0,2
+8,33,113,6,2,0,2
+8,34,113,6,2,0,2
+8,35,113,6,2,0,2
+8,36,113,6,2,0,2
+8,37,113,6,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
index ac971d29..f9969b00 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,121,0,9,2,0
-9,9,121,3,3,8,0
-9,10,121,3,0,8,0
-9,11,121,3,0,8,0
-9,12,121,3,0,7,0
-9,13,121,6,8,7,1
-9,14,127,0,7,7,1
-9,15,127,0,7,7,1
-9,16,127,4,7,7,1
-9,17,127,6,7,7,1
-9,18,127,6,7,7,1
-9,19,127,6,7,2,1
-9,20,113,6,7,2,1
-9,21,113,6,7,2,1
-9,22,113,6,7,2,1
-9,23,113,6,1,2,1
-9,24,113,1,2,8,1
-9,25,113,1,1,8,1
-9,26,113,1,0,8,1
-9,27,113,1,0,8,1
-9,28,113,1,0,7,1
-9,29,121,1,8,7,1
-9,30,121,6,0,7,1
-9,31,121,6,0,7,1
-9,32,121,6,7,7,1
-9,33,113,6,7,7,1
-9,34,121,6,7,7,1
-9,35,113,6,7,7,1
-9,36,113,6,7,2,1
-9,37,113,6,7,2,1
+9,8,100,6,0,9,0
+9,9,100,9,8,9,0
+9,10,109,4,9,5,1
+9,11,113,6,0,9,0
+9,12,100,9,0,9,0
+9,13,109,4,9,5,1
+9,14,112,6,3,9,0
+9,15,121,6,0,5,0
+9,16,109,4,0,8,0
+9,17,109,4,0,5,0
+9,18,127,4,0,9,0
+9,19,109,4,9,5,0
+9,20,127,4,0,5,0
+9,21,127,4,0,9,0
+9,22,109,4,9,5,0
+9,23,109,4,9,5,0
+9,24,109,4,3,5,0
+9,25,109,4,3,5,0
+9,26,109,4,3,5,0
+9,27,109,6,3,5,0
+9,28,121,6,3,0,0
+9,29,109,4,3,7,0
+9,30,112,6,3,5,0
+9,31,103,6,3,0,0
+9,32,103,6,2,0,0
+9,33,103,6,2,0,0
+9,34,113,6,2,0,0
+9,35,113,6,2,0,0
+9,36,113,6,2,0,0
+9,37,113,6,2,0,1
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
index feb89b42..0cf0e0be 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0091290055,0.0077892784,0.007239608,0.020136122,0.006804843,0.065107465,0.09113526,0.03846739,0.046789855,0.011856737,0.010647341,0.0430088,0.01229634,0.024512613,0.043958448,0.015431894,0.10510967,0.008920799,0.012439459,0.013178185,0.0115483245,0.042840227,0.020539861,0.013582593,0.07060103,0.024588121,0.021217003,0.10432629,0.008402498,0.025339192,0.025642872,0.0106556,0.026757332
+0.009402489,0.007112399,0.006294906,0.025800984,0.005330445,0.058306154,0.09866103,0.031214684,0.053123873,0.008108479,0.006941985,0.03378066,0.015496237,0.024613176,0.034951206,0.013683666,0.15753025,0.0058497526,0.010813744,0.012419266,0.00899315,0.0433608,0.024071196,0.0070383553,0.061834656,0.040662296,0.013590249,0.08462712,0.007689061,0.02625725,0.025855735,0.011850585,0.024734205
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
index d7938c94..7a95c833 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008205366,0.008331864,0.0074459263,0.021188624,0.0073211472,0.061644685,0.07545811,0.034722853,0.043902382,0.010114443,0.010213885,0.042470828,0.012390661,0.022267908,0.032764185,0.01806368,0.12040593,0.009458063,0.014117386,0.011872736,0.013330289,0.041174386,0.020099496,0.013925595,0.072014526,0.024408026,0.018947104,0.13367255,0.00828295,0.024368336,0.02391859,0.011060069,0.022437433
+0.008010792,0.007172111,0.0058797314,0.027109819,0.005602905,0.055375658,0.075244375,0.029313534,0.04922832,0.006365298,0.006468321,0.034970127,0.01725487,0.02158429,0.023749217,0.016210178,0.18300672,0.006796729,0.011571762,0.008914565,0.0113315275,0.037273988,0.022148373,0.007266352,0.05995564,0.03540441,0.009834666,0.13455547,0.0071732774,0.021706866,0.02286324,0.011315751,0.01934116
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
index 06ba23c9..c2b44898 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012586267,0.009460248,0.01056256,0.024452873,0.0086881975,0.052435704,0.10992195,0.0706047,0.03934652,0.017093472,0.012070613,0.038141258,0.016003482,0.024356524,0.05110281,0.022539876,0.057360444,0.0130562065,0.014306477,0.011672565,0.013660459,0.019611526,0.033125043,0.01840493,0.062889695,0.025482355,0.023796566,0.06978374,0.011446344,0.029403456,0.027195802,0.014777219,0.034660116
+0.012989526,0.008451789,0.008343688,0.025282253,0.0077611837,0.049804118,0.12995926,0.073096894,0.043786965,0.012898271,0.010859507,0.029631373,0.019566044,0.019631185,0.046036806,0.028859269,0.060753997,0.012488813,0.016959455,0.0091642905,0.0140438145,0.013560966,0.04981232,0.0114713255,0.055660956,0.039123084,0.016607726,0.044842925,0.010157406,0.035077438,0.033839993,0.018593594,0.030883798
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
index d6044728..45b924b2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008913519,0.007365497,0.007884704,0.019221375,0.008245324,0.061635274,0.10626261,0.04613362,0.039076947,0.013314558,0.01237192,0.04430893,0.014824947,0.02413848,0.04149791,0.018489841,0.07907649,0.010472187,0.011848482,0.012690502,0.011933226,0.034617245,0.024047974,0.015420446,0.064966045,0.021884238,0.020587962,0.12284686,0.008948034,0.029838102,0.020556517,0.010585136,0.025995124
-0.0094871465,0.011149254,0.009741706,0.023295518,0.012869805,0.05367749,0.048896357,0.031161776,0.05537468,0.012018335,0.013214679,0.028068008,0.016052,0.03157924,0.03311481,0.021816775,0.052957285,0.012617017,0.022129932,0.015422226,0.02169244,0.031480502,0.015010904,0.02252592,0.07512524,0.07089812,0.022424813,0.11010506,0.011601552,0.021417523,0.044051338,0.016005058,0.02301758
+0.0084459195,0.005478003,0.005521863,0.022377402,0.0057007815,0.048822228,0.13396627,0.047674347,0.044100903,0.008718914,0.0076107793,0.030250654,0.022718882,0.023471467,0.03372159,0.020928407,0.121317275,0.007938454,0.010315681,0.009893803,0.009422824,0.028209142,0.035702728,0.00701184,0.057384826,0.034776427,0.009679542,0.101385765,0.0068074698,0.034072928,0.021294992,0.011075436,0.02420245
+0.01495256,0.011281415,0.011318162,0.04048125,0.013168076,0.043056514,0.0334818,0.020567056,0.03463795,0.00803167,0.010289714,0.013072755,0.027047135,0.024557816,0.022758493,0.027457278,0.042199492,0.008637161,0.033859514,0.014415746,0.032140847,0.015577597,0.018920667,0.013239309,0.03609407,0.24212311,0.0075512324,0.030155562,0.013689231,0.021117214,0.06719196,0.031909913,0.015017748
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
index d211d4e4..a80d23ab 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.00905651,0.008449699,0.007513583,0.01789081,0.009958172,0.059586458,0.09060577,0.05163472,0.05080496,0.015027517,0.0131017715,0.046035837,0.0129659865,0.021924116,0.043815948,0.022697417,0.07993372,0.0115594845,0.015695909,0.011922975,0.012971503,0.034338184,0.025293143,0.016621767,0.06711588,0.022368787,0.02659395,0.0939604,0.010572614,0.026793035,0.024672363,0.012390149,0.026126813
+0.010768626,0.0071453005,0.0061826305,0.020636555,0.007476879,0.053042114,0.121259026,0.055434912,0.06060058,0.011440382,0.008968602,0.034554403,0.018097715,0.020440187,0.040972438,0.030373393,0.08881224,0.010173473,0.015603304,0.009588735,0.01265485,0.026746185,0.037209485,0.0092568,0.06076117,0.03545333,0.017796088,0.065680444,0.0107882945,0.024638936,0.028522931,0.013651196,0.025268821
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
index 8e0cfcaa..9e3adcef 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008020919,0.0066487696,0.0072186273,0.023239477,0.009595765,0.059273914,0.062329326,0.027587652,0.0486198,0.008902664,0.009877493,0.03435876,0.017895322,0.022803104,0.0360042,0.016689297,0.112537004,0.008584821,0.014853362,0.013560702,0.015122532,0.043928403,0.0197575,0.01412689,0.07697556,0.037536968,0.014428177,0.14790235,0.008241942,0.016693683,0.024688615,0.013138984,0.018857364
-0.0086295465,0.007917115,0.007522475,0.023743981,0.0073922593,0.06037666,0.08741754,0.043700412,0.04167852,0.012650199,0.01019312,0.043437473,0.013766954,0.021962289,0.04165744,0.018287549,0.10325626,0.009215165,0.01214539,0.011779306,0.012235232,0.045917273,0.018463248,0.014126799,0.068317495,0.021482795,0.016241094,0.12552734,0.008585244,0.025119003,0.02262137,0.01049176,0.02414174
+0.008686484,0.005960353,0.008075502,0.039276745,0.0097908955,0.05850304,0.05206207,0.01889337,0.048300233,0.00518526,0.006793675,0.025432974,0.031197455,0.021319006,0.02276724,0.014915394,0.15458107,0.0053153243,0.016606817,0.011906255,0.0143988915,0.038660195,0.019179536,0.007658729,0.052357905,0.0989535,0.005735836,0.11204717,0.008564646,0.016890625,0.025278362,0.020940961,0.013764397
+0.008197727,0.007059469,0.0055742837,0.023094872,0.0057917624,0.051699355,0.11453689,0.0425388,0.043388393,0.00957253,0.0075618536,0.034410797,0.019690048,0.022121822,0.03347103,0.017320948,0.1331481,0.007196628,0.0096377,0.010214877,0.009963169,0.045684543,0.02452887,0.007794731,0.060744867,0.027903598,0.010524433,0.115211345,0.0076923347,0.03072128,0.020051872,0.0100175515,0.022933505
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
index e2701732..f309bc23 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.00940564,0.0063802544,0.0065206103,0.027256832,0.0066316505,0.05138203,0.10203898,0.048445128,0.049608555,0.00873864,0.010182529,0.026134778,0.014573924,0.02411956,0.038605724,0.036600724,0.07013269,0.011340922,0.020680869,0.011217838,0.01796068,0.016984288,0.026266541,0.013434889,0.07714,0.08002693,0.012619701,0.06227914,0.007597093,0.02409692,0.043419037,0.014213739,0.0239632
+0.011856947,0.007223844,0.0070571597,0.032826528,0.008080109,0.040391415,0.09255855,0.039391916,0.052117735,0.007856196,0.009969194,0.01840535,0.020338742,0.020724263,0.03398534,0.041261356,0.058055624,0.012493181,0.027867151,0.0117967855,0.0220204,0.011738093,0.035083402,0.009734925,0.05840189,0.1281792,0.009151511,0.030461226,0.008224331,0.028107058,0.059158992,0.020592097,0.024889482
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
index 99ee0e82..d6b436bb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.009661113,0.007831191,0.009208629,0.018906532,0.008831174,0.065691166,0.10545667,0.038684368,0.039912082,0.01415859,0.013239993,0.03936315,0.0131430235,0.025695173,0.04284285,0.015051846,0.08823259,0.009918131,0.012693808,0.015639171,0.01375921,0.053180333,0.021613518,0.016188124,0.064060025,0.024109922,0.022105852,0.08915296,0.01004957,0.026042204,0.028148724,0.010506249,0.026922056
+0.010474039,0.006521543,0.0074160905,0.023902066,0.0066056135,0.057496417,0.12374015,0.03752914,0.047728445,0.009753353,0.0080655515,0.027190102,0.014648493,0.026415588,0.031456366,0.0145098185,0.13790837,0.006406579,0.010360853,0.01592334,0.009660522,0.059283886,0.031192504,0.007285326,0.06007168,0.037452072,0.011793553,0.05825485,0.008734745,0.027273804,0.028270483,0.011573875,0.02510079
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
index a2fc7cd2..4cab3548 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0042463774,0.005140092,0.0060796365,0.016366256,0.0067424467,0.03495936,0.06623093,0.020122634,0.050006244,0.00649412,0.0041204235,0.02871717,0.018642396,0.028757636,0.033106435,0.01599839,0.23275402,0.007276528,0.008622721,0.017800169,0.010982657,0.028370103,0.02974473,0.0034585185,0.06776658,0.046286162,0.0064784237,0.120085396,0.0039499593,0.021950865,0.015566807,0.018041762,0.015134025
+0.0018376797,0.0023137012,0.0024238364,0.009155929,0.0033391858,0.037196394,0.11448541,0.03619819,0.020103797,0.0032478096,0.0022368645,0.042428207,0.030106904,0.017886922,0.041678857,0.017872667,0.17781384,0.002194447,0.005525137,0.01080326,0.0041624387,0.026048908,0.038844123,0.0014777568,0.105031826,0.010042691,0.004704572,0.116676375,0.0022287897,0.06710731,0.0103819845,0.011351856,0.023092275
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
index 609b1f0f..4526ff20 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0041903867,0.0054771546,0.0058246264,0.015669782,0.0071306545,0.030696195,0.06473787,0.018447235,0.05242684,0.0068101883,0.004801957,0.029178623,0.022633398,0.02907579,0.033386257,0.018674878,0.19909294,0.006572294,0.009357265,0.01596702,0.013374633,0.026702495,0.032056045,0.003591021,0.07047866,0.053875104,0.0050279247,0.13526632,0.0042218207,0.021093741,0.021248914,0.019235754,0.013676274
+0.0048656277,0.007918645,0.0054325806,0.0121447295,0.007898342,0.017814836,0.075257584,0.011748432,0.021347238,0.00770395,0.0054659205,0.023766177,0.042818088,0.014586231,0.021619631,0.062280655,0.065641224,0.0032016544,0.038712315,0.010045757,0.03476402,0.009708087,0.06701612,0.015924746,0.021149337,0.11031023,0.00664596,0.060393848,0.006036425,0.027078018,0.14114997,0.010524234,0.02902941
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
index a7ac31d6..a6528ce7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0054918397,0.0063162544,0.0061463765,0.016365495,0.0076878374,0.034990918,0.0610308,0.020188866,0.0626216,0.008143705,0.0056989286,0.028557653,0.01842186,0.026081735,0.02933535,0.018272739,0.24774212,0.008434298,0.009743969,0.017181823,0.013515602,0.028113525,0.035016388,0.004349286,0.05375894,0.048733603,0.0074889436,0.08502142,0.0040941467,0.018917236,0.024357485,0.024139129,0.014040137
+0.0023688173,0.0032759958,0.0024291654,0.0072109625,0.0038664865,0.03368376,0.037551492,0.020473365,0.020303426,0.00320672,0.002927277,0.019001897,0.0069159083,0.008086997,0.020394612,0.011339872,0.576207,0.0024872613,0.0064319503,0.010710103,0.005667849,0.009287372,0.02400277,0.002055522,0.026548946,0.011598315,0.0042506577,0.04309646,0.002323081,0.030272912,0.01817002,0.013025035,0.010827967
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
index 942890e0..29abb8a1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004381987,0.006287049,0.006620055,0.019419348,0.006867307,0.03929928,0.07064888,0.017149057,0.060524255,0.006789356,0.0050974856,0.027337514,0.024331838,0.03136498,0.029390393,0.020899236,0.15376608,0.0074551413,0.009935729,0.015628591,0.013709809,0.03817771,0.03494091,0.0051154993,0.06018743,0.075591475,0.0067811315,0.11117232,0.0047863433,0.015857743,0.028808644,0.024059782,0.017617658
-0.004469706,0.006364652,0.0068032146,0.020612448,0.007166283,0.039386127,0.08379896,0.017597469,0.06735702,0.007070649,0.005259082,0.027966166,0.024157904,0.03855127,0.026747128,0.019213978,0.11608029,0.0063566845,0.010583058,0.01697354,0.012141066,0.043337125,0.031579476,0.004105956,0.061916213,0.07502441,0.0060634622,0.11869589,0.005418634,0.019128878,0.027781855,0.025369624,0.01692181
+0.0066867806,0.009659095,0.006985347,0.018342886,0.008343686,0.02827222,0.13270424,0.009605769,0.023481674,0.0065346137,0.008042574,0.01586629,0.06682908,0.029205475,0.01841571,0.035887375,0.0029870616,0.00378063,0.021729896,0.010461546,0.025363758,0.0364591,0.043761298,0.02097016,0.028998114,0.09809047,0.012051953,0.032568526,0.00945276,0.014552994,0.11844721,0.013847485,0.08161431
+0.0062105553,0.008430051,0.0066421693,0.034530688,0.009992443,0.030213395,0.02225522,0.0048418874,0.034004305,0.006981886,0.0063810414,0.009619785,0.016620198,0.025754781,0.010917337,0.020015128,0.0059674657,0.004531538,0.036805414,0.011499066,0.03323312,0.025179174,0.01618241,0.030543782,0.015805239,0.36270088,0.008696667,0.013364138,0.008154805,0.006193264,0.11358851,0.015613565,0.03853005
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
index 7f7b97c7..b868c8da 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005170072,0.0052365907,0.00694314,0.017285356,0.008234528,0.037991844,0.07294639,0.022531565,0.05857518,0.0065963003,0.005045354,0.026125679,0.020520791,0.035622887,0.0324971,0.01749203,0.16650967,0.007892656,0.009556492,0.017534178,0.011642951,0.03663207,0.033574145,0.0035568238,0.07068761,0.048016,0.0072974605,0.12339532,0.0045464407,0.022555495,0.018934185,0.020668464,0.01818533
+0.0040303203,0.0029316142,0.0041374513,0.010778411,0.0056618657,0.032110527,0.1331622,0.03412711,0.018952146,0.00459283,0.0029058945,0.023700856,0.08413434,0.023126962,0.058894433,0.033277348,0.027406437,0.0037933602,0.005311759,0.011036976,0.007056581,0.04888226,0.041817285,0.0038434942,0.10182688,0.010030801,0.0061456566,0.10456378,0.0049827173,0.06323838,0.008559678,0.016620507,0.058359172
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
index 220a5080..c5d59b6b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0050350777,0.005766118,0.006457709,0.019448206,0.007035608,0.036554355,0.06908976,0.016720336,0.073623985,0.007380201,0.0050186585,0.027139133,0.019853646,0.03107149,0.027048368,0.018990275,0.18321688,0.0086597875,0.00932425,0.017520232,0.013038053,0.03883165,0.028723942,0.0043549896,0.060425516,0.060644113,0.0076746633,0.10625886,0.0049037784,0.016128076,0.024719888,0.024399208,0.014943133
-0.0042376784,0.005924756,0.0069834692,0.018832779,0.0078081987,0.04281036,0.07998875,0.020728823,0.057859186,0.006507186,0.005319783,0.035538692,0.020944735,0.03357084,0.033523228,0.017411817,0.14398022,0.0064492146,0.010395881,0.018382689,0.010860411,0.03802423,0.034201827,0.003840388,0.06658509,0.059049625,0.0067133494,0.10930205,0.00455577,0.024278667,0.023504738,0.026207874,0.015677737
+0.0028121886,0.004768915,0.0026038273,0.024106294,0.00428334,0.04001045,0.042205393,0.0055848244,0.051259898,0.0029990096,0.0024799826,0.018063348,0.013305797,0.016817851,0.013413405,0.02079274,0.06953616,0.002443421,0.023023874,0.012457441,0.017435335,0.022667246,0.025499817,0.009044353,0.021868205,0.2366301,0.006028528,0.06351859,0.0038850654,0.008795736,0.16167547,0.015960602,0.03402277
+0.0049432945,0.008161541,0.005641948,0.0334701,0.006687145,0.020943755,0.048687987,0.0067313584,0.026358342,0.0061188038,0.00532264,0.014945954,0.022759516,0.02219019,0.0143439155,0.022385858,0.0037059456,0.0030287185,0.029329462,0.010064615,0.025855048,0.019193592,0.026798103,0.025868736,0.018704478,0.27705717,0.008246222,0.02213319,0.0063955006,0.008136024,0.18003964,0.013626517,0.05212471
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
index 83108e29..420af8d1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005213573,0.005612019,0.005980635,0.014423169,0.008148407,0.035215456,0.08324456,0.021167181,0.05202259,0.006690955,0.0051461984,0.03389285,0.022059022,0.028906152,0.04061394,0.018913923,0.16657288,0.005920637,0.008955144,0.018136969,0.010939499,0.027240243,0.03896024,0.0037672839,0.06677831,0.05034332,0.006739948,0.11629482,0.004037235,0.02923933,0.021340776,0.020757355,0.016725339
+0.008235056,0.006630053,0.008237326,0.0065348125,0.013050176,0.030923633,0.14479923,0.043974973,0.015504926,0.009599815,0.00887134,0.027437832,0.024815973,0.021933725,0.078563236,0.020613033,0.04735088,0.005203651,0.006880082,0.015798736,0.006727233,0.021745805,0.099310055,0.006908331,0.07205278,0.00884878,0.012542185,0.016972749,0.0075840293,0.11026282,0.012869484,0.040418718,0.038798563
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
index d3b4ccb7..0010dfae 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004618134,0.006132982,0.006672614,0.015618401,0.0074373814,0.03397381,0.07677221,0.019593101,0.05670967,0.0069386126,0.0054081944,0.030513486,0.021875117,0.02945475,0.029283365,0.017094098,0.20585725,0.007897137,0.009288681,0.016525654,0.011750388,0.03215573,0.031738274,0.0035157073,0.06421899,0.043070234,0.00695176,0.11497121,0.004500923,0.020835267,0.023099473,0.021389142,0.014138295
+0.0014310861,0.002243976,0.0017065166,0.016686594,0.0020667855,0.053174898,0.095879674,0.014162464,0.032807916,0.0018454238,0.0015201237,0.026181472,0.039513864,0.022075271,0.021645933,0.0206909,0.15128985,0.0018012328,0.008061768,0.0071523706,0.007267346,0.053755138,0.028755592,0.00226818,0.050142813,0.034830853,0.0034951526,0.19666047,0.0019271312,0.020879522,0.041771542,0.010048885,0.02625917
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 7e5bfc90..068b8b9f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.016444147,0.017504703,0.010868917,0.024397992,0.02378631,0.05476737,0.018488651,0.01870659,0.03153483,0.01417574,0.018273251,0.04281981,0.028775947,0.030504966,0.017300766,0.020425139,0.019642664,0.01798995,0.01798995,0.06529241,0.026927112,0.06377991,0.09389312,0.034339547,0.03991245,0.021912975,0.010130945,0.04549251,0.012461275,0.0647843,0.021114701,0.01657312,0.03898788
-0.018649058,0.011534323,0.009525009,0.04894217,0.016076498,0.027988719,0.09323943,0.032794107,0.024086585,0.014523906,0.023028513,0.028638465,0.056332096,0.02776413,0.029872134,0.031101944,0.02357462,0.010298991,0.0192035,0.01811063,0.018979773,0.07148915,0.060672056,0.015280495,0.074919984,0.019054057,0.016522154,0.039403025,0.012767332,0.009750889,0.041294016,0.035528205,0.019054057
-0.009243722,0.015240321,0.012585413,0.012245942,0.011020156,0.044706393,0.04126598,0.022088086,0.022789238,0.016350478,0.017748276,0.052165084,0.025299398,0.037463184,0.07560385,0.03431107,0.060041644,0.014372992,0.013344871,0.014485721,0.019722441,0.03816479,0.025200764,0.012536347,0.06316938,0.05865078,0.021873433,0.038314164,0.036381554,0.024521017,0.025949989,0.020953465,0.062190026
+0.013403586,0.013561583,0.010157157,0.02523759,0.02000361,0.05363709,0.021128021,0.025016747,0.02920473,0.01225187,0.0151882535,0.046147842,0.030095028,0.030087681,0.019925622,0.02262283,0.03150853,0.014046838,0.017826365,0.06419519,0.02504119,0.059139382,0.08537767,0.030591354,0.04970318,0.023754872,0.006926579,0.045255262,0.009768042,0.06675244,0.026345653,0.018249106,0.037849132
+0.016586032,0.009750425,0.008877863,0.044213485,0.015101754,0.030926332,0.10564914,0.038188923,0.021930052,0.012667373,0.019466935,0.031108072,0.05049339,0.026259657,0.033276457,0.033570215,0.030092146,0.008471316,0.018575478,0.017898783,0.018978879,0.06433024,0.053123724,0.01424231,0.08489147,0.020401172,0.012519794,0.03767036,0.009599258,0.0104606785,0.04588504,0.03737721,0.017415995
+0.007843759,0.0123399645,0.010932602,0.012196199,0.010071585,0.041178703,0.048805505,0.02791397,0.021404997,0.01442688,0.014203211,0.060977157,0.024207711,0.034439713,0.07799098,0.035463613,0.07618432,0.0120071145,0.012583348,0.013552799,0.017436162,0.038533,0.023809142,0.010679349,0.07326574,0.052771334,0.016932746,0.037566938,0.030193252,0.02263027,0.02893749,0.023646941,0.054873504
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index c1bad420..06bccafb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.018813502,0.0097793965,0.019639514,0.023574388,0.030255334,0.08317135,0.027166266,0.015935654,0.042812943,0.011081507,0.016963415,0.03456955,0.039287366,0.025708733,0.03456955,0.021297682,0.02741194,0.017777506,0.02591037,0.016123498,0.03373579,0.01538515,0.036263976,0.033180345,0.039828185,0.11821033,0.015935654,0.05716278,0.01116842,0.03830239,0.035701755,0.008766194,0.014509578
-0.013969102,0.009713985,0.01286894,0.013969102,0.015045315,0.06638303,0.04526929,0.020847589,0.039484575,0.014639493,0.013645508,0.046162147,0.04162265,0.021467391,0.05503346,0.02815642,0.026431143,0.014356339,0.012231754,0.020206176,0.039484575,0.038720876,0.0321792,0.0149866585,0.092163704,0.023808694,0.018834226,0.04049995,0.016524045,0.019508144,0.04734924,0.046797603,0.03763967
-0.017729616,0.014929833,0.020768544,0.022854377,0.015403757,0.04781904,0.025558222,0.03895261,0.058589116,0.019701704,0.013646972,0.08070991,0.04781904,0.021261055,0.05070465,0.018726205,0.047077674,0.024115471,0.022086421,0.015646331,0.028528795,0.032546125,0.01435788,0.021136845,0.0401891,0.06310296,0.017626034,0.050903104,0.02488098,0.021302622,0.018873077,0.016655432,0.025796438
+0.019326234,0.008987534,0.01819081,0.021602122,0.03681829,0.06312124,0.03617675,0.017055342,0.04511071,0.0098323915,0.015110195,0.03783893,0.03835981,0.024985593,0.032971486,0.02502222,0.035477027,0.018369326,0.027649779,0.01682375,0.034184624,0.014531332,0.031430904,0.032019537,0.049933005,0.109490566,0.01257582,0.058377594,0.009717841,0.035477027,0.039423183,0.011185166,0.0128238555
+0.013355161,0.00794364,0.011924816,0.012399847,0.015252072,0.053862467,0.053862467,0.02285139,0.041217204,0.0128435325,0.011512868,0.051597085,0.037019096,0.019892447,0.055139776,0.030555336,0.03286122,0.013459907,0.0118320165,0.019393725,0.037602063,0.035810128,0.028006654,0.01410586,0.10880518,0.02236568,0.01507438,0.038644433,0.01388717,0.016685817,0.053027406,0.057112765,0.030096311
+0.015875405,0.0123637775,0.019149387,0.021762764,0.015875405,0.039832875,0.029979475,0.041419636,0.06085663,0.01757247,0.011569412,0.094256595,0.044872865,0.018487863,0.052461695,0.019603502,0.055845182,0.022963623,0.021300191,0.0145680895,0.028087579,0.031850714,0.013212684,0.019527076,0.04851914,0.05921513,0.014739814,0.050255228,0.023121139,0.019951142,0.019951142,0.019527076,0.021425363
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 4bb17616..b9936480 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.014180237,0.016257722,0.02075337,0.015035935,0.023724249,0.09210627,0.033686046,0.016321354,0.029331319,0.01988051,0.02091614,0.022904685,0.020075608,0.043507922,0.036637343,0.019610556,0.02790456,0.0151538635,0.01619434,0.025217365,0.032018133,0.042417135,0.02589112,0.017960545,0.12688175,0.066083014,0.015881114,0.024741694,0.022048742,0.031521738,0.026837192,0.013960393,0.02435811
-0.010987974,0.009621391,0.008102024,0.010004664,0.01264708,0.09947699,0.039184675,0.038127735,0.026918061,0.008229612,0.013890101,0.029495217,0.02808619,0.054296385,0.039108217,0.046989605,0.02911958,0.009696852,0.010649909,0.018258192,0.032708064,0.031547327,0.04019234,0.021597609,0.04019234,0.045543887,0.023859728,0.069717936,0.013890101,0.030031558,0.07900077,0.007976414,0.02085151
-0.016185524,0.016569352,0.011432509,0.008799916,0.01716223,0.0705819,0.029802466,0.025813933,0.0328878,0.008075258,0.016122423,0.023230024,0.018376462,0.034974713,0.0608454,0.10472156,0.020986577,0.009367462,0.015996957,0.017637983,0.0325523,0.024179006,0.022515312,0.025738416,0.0861416,0.026348786,0.020620897,0.04803876,0.019523472,0.043314718,0.05390604,0.013629578,0.02392066
+0.012297478,0.013665343,0.018972449,0.011826368,0.024950994,0.078946404,0.037073772,0.020235552,0.029241972,0.015067184,0.017892724,0.023132304,0.015667394,0.04334363,0.0390813,0.022333188,0.037584122,0.00976618,0.017546648,0.0258028,0.034895677,0.030525848,0.022267856,0.016710458,0.13534562,0.093021594,0.011826368,0.024006857,0.01638725,0.025976619,0.04160183,0.016841521,0.01616473
+0.011090748,0.008306586,0.0069948467,0.009486419,0.01348292,0.08195016,0.0491262,0.049899824,0.026644321,0.007049708,0.012324387,0.03201389,0.02492027,0.049318474,0.041046478,0.048176013,0.046149794,0.00805102,0.012421048,0.019088384,0.0321235,0.02361709,0.03528076,0.020478847,0.047244202,0.053118087,0.01741415,0.06137783,0.010177442,0.028612677,0.08655661,0.010257265,0.016200101
+0.015800804,0.01356804,0.009696641,0.008967927,0.01755835,0.065492526,0.037830207,0.038803037,0.030613404,0.0072624483,0.013835645,0.025572212,0.017904656,0.035158474,0.062493403,0.09418089,0.030650796,0.0078525795,0.018222168,0.018400995,0.032899566,0.02009147,0.021096846,0.026204217,0.08847476,0.029853182,0.01636618,0.04572115,0.013781705,0.039645717,0.059865013,0.016238818,0.01989622
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index b9714301..48bf280b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.013940462,0.019203799,0.019620843,0.012943281,0.026376959,0.08539569,0.034201056,0.016171219,0.036371298,0.017553693,0.018832363,0.023096437,0.022320347,0.02289433,0.035982683,0.026197264,0.0309963,0.018396111,0.017349185,0.024706373,0.039096992,0.03601784,0.021845924,0.017417088,0.139698,0.046610545,0.013511559,0.03353955,0.018649349,0.01886918,0.03457043,0.012447432,0.04517649
-0.018725712,0.011787114,0.01107297,0.033447456,0.019894524,0.08055031,0.03819819,0.035431203,0.03186907,0.018291932,0.019207258,0.02245556,0.13702068,0.054503065,0.023191012,0.02562002,0.01916978,0.013834468,0.016687555,0.02840035,0.030843345,0.021892576,0.026087089,0.01579946,0.04885623,0.023418596,0.015923375,0.042116705,0.01549387,0.026875785,0.03695072,0.019433666,0.016950345
-0.020445274,0.016654389,0.013200464,0.028482636,0.033642784,0.083429486,0.028041055,0.019777574,0.025619086,0.021657871,0.030393722,0.014611581,0.022830635,0.123300776,0.01745365,0.037973646,0.017728506,0.027734673,0.01872503,0.024196517,0.019738983,0.021280492,0.021657871,0.019509016,0.022542626,0.05344693,0.025725637,0.07596341,0.020049827,0.038271476,0.015432905,0.013780033,0.026701493
-0.016903484,0.014345801,0.01919161,0.02922801,0.017169677,0.073139146,0.028308017,0.035229895,0.035195507,0.019079488,0.01275942,0.023376677,0.053509746,0.026592921,0.03432988,0.015330795,0.024594432,0.016969644,0.017645637,0.023376677,0.03652613,0.047130045,0.024142297,0.023931038,0.06951778,0.054352403,0.024307897,0.06207239,0.022835156,0.02310433,0.034129318,0.010332893,0.03134177
-0.012535897,0.012634218,0.007906292,0.027009122,0.012584961,0.057965416,0.026101558,0.024281856,0.028694961,0.0115486095,0.016803129,0.074138895,0.03044326,0.030756325,0.042741016,0.019302549,0.026152587,0.010392624,0.011822477,0.047309995,0.01809768,0.06803367,0.085000664,0.020993682,0.06697891,0.02416358,0.010151879,0.041750923,0.010474134,0.058648694,0.01960652,0.01673762,0.028236296
-0.020117404,0.02431366,0.01404421,0.041117303,0.020796577,0.08436723,0.027712893,0.025282206,0.0328136,0.018245693,0.022224486,0.02491456,0.034054168,0.03721901,0.018496858,0.032352228,0.03961943,0.016101766,0.03537617,0.02481743,0.02498766,0.016743189,0.031564236,0.05642065,0.06519394,0.03134155,0.0072011556,0.046229426,0.02007815,0.036965452,0.027564444,0.01680872,0.02491456
+0.014360831,0.018260473,0.018224843,0.011053926,0.03076024,0.07321574,0.037321977,0.01873005,0.036528688,0.015227427,0.015527761,0.02097666,0.020291606,0.022482706,0.037103932,0.02936596,0.033982046,0.013024708,0.019821553,0.028408663,0.050124213,0.025789138,0.018803358,0.019362386,0.14334947,0.056798175,0.010143653,0.031520464,0.013229817,0.017977372,0.050320394,0.014586981,0.033324774
+0.01811298,0.010022493,0.010627294,0.029121654,0.02044467,0.07640756,0.046524912,0.04837825,0.029228497,0.015523141,0.01671908,0.021955362,0.12305657,0.05190229,0.027959907,0.030796167,0.022829963,0.0101011,0.016204689,0.027190795,0.0332174,0.017317316,0.02210596,0.016141513,0.05655991,0.03148033,0.012669618,0.03716541,0.011312696,0.025932973,0.047257572,0.02219248,0.013539525
+0.0188605,0.014545823,0.011733642,0.027108917,0.03445344,0.08535639,0.038024917,0.02668049,0.023018274,0.0188605,0.02539817,0.013825609,0.023982013,0.11485958,0.021226147,0.043937705,0.021122757,0.019535357,0.01927009,0.024431145,0.02162367,0.017409071,0.021350885,0.01799681,0.02659432,0.069665864,0.020835934,0.06621647,0.014320311,0.03824837,0.020714207,0.016482577,0.022310078
+0.016216978,0.013081708,0.014939779,0.024982793,0.017741427,0.057494164,0.03569894,0.043145325,0.033766083,0.014881535,0.010308138,0.02417871,0.0764655,0.026581064,0.04486404,0.018091345,0.031891134,0.012629794,0.014939779,0.023734218,0.031922292,0.033275068,0.015534913,0.023116592,0.07858521,0.06695543,0.014311432,0.06540441,0.016028045,0.02678954,0.036832146,0.011410069,0.024202334
+0.011195347,0.0092090415,0.0072849514,0.026208512,0.011416156,0.057524852,0.029870834,0.03149794,0.026517449,0.009281268,0.01366335,0.078934886,0.03681583,0.030525122,0.050370444,0.020671988,0.037873372,0.00787691,0.01097881,0.042916153,0.016871985,0.061715156,0.057300583,0.019230815,0.08402575,0.0281726,0.0070058694,0.04515177,0.0079386905,0.060521476,0.021453522,0.017407559,0.022571096
+0.01740354,0.01922637,0.01123657,0.03960435,0.018671269,0.0817404,0.03641405,0.032466494,0.028904509,0.014204355,0.018381797,0.028214265,0.03581451,0.036342997,0.021490498,0.035640057,0.050952412,0.011684184,0.033546194,0.023464803,0.024068216,0.015661491,0.026706208,0.0505559,0.078608975,0.03567488,0.005650102,0.044790044,0.014204355,0.04054354,0.028234936,0.018238753,0.02165905
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index c35131db..5bf42858 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.015050883,0.01172164,0.012093725,0.022309033,0.010756382,0.094016306,0.022882726,0.03638838,0.029932264,0.010672675,0.015347736,0.027256964,0.022027602,0.037543472,0.08695087,0.023957483,0.07613672,0.010714447,0.015407805,0.012674115,0.016659811,0.021770975,0.033035073,0.019553527,0.0648693,0.03820925,0.009755615,0.032081287,0.012924089,0.10570537,0.0215594,0.0116304215,0.018404752
-0.01221852,0.020105632,0.011523162,0.04298132,0.016506167,0.056940995,0.022168143,0.083172925,0.026531791,0.014566641,0.0132113695,0.06253745,0.035355467,0.036727987,0.024466112,0.017230876,0.11821256,0.01090991,0.019639885,0.026805244,0.014284897,0.02955142,0.036549088,0.017708533,0.03152267,0.038228378,0.0096279625,0.02415747,0.01929767,0.024778698,0.03875462,0.014008602,0.029717823
-0.0134996595,0.008414936,0.008818777,0.03387192,0.012195926,0.014370312,0.033657596,0.01964189,0.031905293,0.0134996595,0.019757316,0.038307022,0.07952743,0.028943907,0.039215446,0.05639339,0.046569496,0.01007136,0.015844474,0.011457012,0.019000426,0.048709184,0.071847044,0.01907479,0.08498784,0.01888942,0.009314485,0.045937136,0.01634743,0.01713196,0.024065744,0.06390209,0.024829673
+0.010741137,0.008597097,0.008732483,0.021761348,0.010615999,0.06205288,0.03822967,0.056279518,0.025933884,0.009895199,0.010699261,0.033606037,0.021071235,0.029112015,0.10072176,0.022342766,0.12244653,0.008139568,0.015147431,0.010699261,0.015325984,0.020106312,0.028507149,0.015266233,0.0655409,0.02857683,0.007127244,0.026108587,0.010492319,0.10311031,0.022039376,0.016799457,0.014174217
+0.009333175,0.014287093,0.009333175,0.03983506,0.0151492385,0.035707913,0.03427296,0.105981335,0.024172984,0.012510196,0.010052225,0.06503865,0.027972776,0.027532374,0.029922616,0.0156913,0.18455555,0.009046024,0.017920028,0.020828241,0.013008546,0.023856388,0.02995185,0.012364448,0.035987977,0.02517892,0.00704505,0.020148035,0.016284535,0.030334523,0.033939894,0.018131264,0.024625693
+0.010948875,0.0068784477,0.007794302,0.033332005,0.011385028,0.014280026,0.04296665,0.029207861,0.02900534,0.012024987,0.01593052,0.041807696,0.07056385,0.026035152,0.04887815,0.056921564,0.06475303,0.008232424,0.014906994,0.010086576,0.017841335,0.04450406,0.061787773,0.01656512,0.09203261,0.017360097,0.0076137474,0.04060063,0.013258555,0.019253442,0.023659023,0.06759603,0.02198812
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 6fdccfca..15dcf397 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.013031122,0.011321634,0.010389316,0.012531906,0.014766195,0.104928374,0.035284165,0.027871214,0.020864384,0.010389316,0.017811432,0.024512332,0.020182995,0.08139971,0.029885026,0.043062516,0.016154267,0.010552924,0.013817477,0.024947023,0.036833324,0.051740803,0.04085048,0.033833183,0.039362326,0.033146404,0.022692353,0.050938636,0.017263431,0.053175114,0.046470832,0.012385906,0.017603923
-0.013867004,0.0117917815,0.01430719,0.011837933,0.013572258,0.038663756,0.02143049,0.018033074,0.02613567,0.012749956,0.014475838,0.04330135,0.019403422,0.029014427,0.033526108,0.025405882,0.021388676,0.013921278,0.016086007,0.024153845,0.020156674,0.32372767,0.016957058,0.01336184,0.04830615,0.016403275,0.014877124,0.025381085,0.016825097,0.016661588,0.018156769,0.017359182,0.028760536
-0.02343217,0.018681755,0.01668093,0.015979351,0.02779923,0.038860146,0.042846564,0.01560919,0.039012242,0.023295274,0.038633116,0.022228505,0.060541473,0.026786525,0.037848905,0.03317399,0.03386134,0.021544605,0.028099464,0.034529194,0.031596933,0.040725086,0.03131662,0.013299196,0.04855152,0.04276296,0.021798568,0.028569853,0.029247366,0.014549365,0.03781196,0.029795108,0.03053148
-0.015153667,0.017171355,0.015880907,0.04367764,0.013012349,0.09067712,0.021834187,0.05186848,0.02257132,0.019533848,0.019572038,0.028627018,0.09138831,0.06738551,0.024596829,0.029257633,0.03824542,0.014573139,0.02051132,0.022926766,0.016708186,0.023174377,0.017306032,0.017855383,0.055864554,0.033130992,0.013425405,0.041758977,0.02423915,0.028627018,0.018748865,0.0150945885,0.025601646
-0.020325784,0.03276724,0.017183015,0.036877196,0.0409791,0.060563177,0.054928415,0.013097705,0.02611157,0.020930074,0.021573389,0.027016183,0.022258202,0.041219912,0.019700425,0.035533786,0.029733112,0.020991484,0.033840418,0.018255536,0.018219914,0.018835029,0.020807795,0.015075367,0.14931092,0.027404804,0.014328933,0.0464354,0.020167608,0.028804252,0.016817788,0.013252097,0.01665435
-0.014279534,0.026315438,0.013732494,0.018192599,0.034430902,0.022774111,0.018624028,0.013572506,0.040204696,0.016825402,0.018770097,0.065004244,0.019747883,0.029950626,0.025983466,0.04906778,0.041724697,0.02481782,0.01630774,0.040957645,0.027930688,0.04398407,0.07167262,0.020374749,0.04398407,0.026008852,0.0155002875,0.031533923,0.028012637,0.030362919,0.0260597,0.021021511,0.062270254
+0.011449223,0.008812782,0.008951563,0.014304557,0.012673135,0.088313274,0.04466766,0.040118113,0.019976614,0.008214416,0.014027881,0.034923363,0.0200548,0.07524385,0.032968048,0.04163484,0.031914745,0.008343774,0.014360543,0.024238,0.029458722,0.049249835,0.036670923,0.027613169,0.055049576,0.03465159,0.015466914,0.052528672,0.012428015,0.048962105,0.049927797,0.016272627,0.016528884
+0.01242395,0.008741341,0.012918863,0.0126689905,0.012136149,0.03917598,0.026045907,0.02556709,0.024926098,0.010878774,0.011535247,0.05655713,0.019660482,0.029255617,0.03887111,0.02514614,0.0321624,0.010836362,0.015983855,0.023936106,0.017657993,0.2894735,0.014244129,0.0121836485,0.06459027,0.01704799,0.01135641,0.026951538,0.012918863,0.01786614,0.021508634,0.019737432,0.025035877
+0.019095432,0.015463947,0.0148135545,0.016017271,0.02569571,0.03876264,0.05160351,0.019207647,0.036166057,0.019740103,0.031347863,0.026176937,0.057680424,0.025998604,0.03910483,0.03476358,0.042780813,0.01607996,0.028070044,0.03522492,0.029742124,0.04114191,0.026955184,0.01176435,0.06334955,0.042365067,0.016655328,0.029546713,0.022676386,0.014080084,0.04580756,0.034661878,0.027459998
+0.012427352,0.012427352,0.011812032,0.044231243,0.012922402,0.08660057,0.032709766,0.07465436,0.020609653,0.0156484,0.016271763,0.031952046,0.09075662,0.0546183,0.03434656,0.031395294,0.053145062,0.010713048,0.016208325,0.019398829,0.014700311,0.02252502,0.012722059,0.013384786,0.06958552,0.03223411,0.009985658,0.03335491,0.01854668,0.042703256,0.015285905,0.013863714,0.018259136
+0.0175087,0.024107471,0.015511848,0.03606612,0.04308127,0.057521038,0.0672489,0.017136548,0.023082271,0.018029237,0.017170051,0.03007557,0.023931548,0.03831727,0.02473345,0.03831727,0.036348987,0.017338548,0.031151721,0.015941853,0.018348955,0.019839952,0.018894475,0.0121516315,0.16644265,0.023990046,0.012784643,0.04135001,0.018931413,0.033748895,0.0147726275,0.013139047,0.012985972
+0.011374115,0.021458177,0.012639273,0.019884368,0.034073096,0.020920198,0.021291187,0.017754741,0.037733898,0.014950992,0.014661813,0.07216797,0.019923242,0.028580561,0.030113481,0.052593373,0.055985354,0.019652708,0.01661401,0.03874207,0.025682075,0.045249857,0.06596687,0.016484719,0.052593373,0.02253194,0.01215507,0.030408999,0.026111996,0.032960337,0.027338427,0.022269435,0.05913232
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 74de4f73..891c35c6 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.01910463,0.02436371,0.021312755,0.02668002,0.029352391,0.070567615,0.022909949,0.013841413,0.03352967,0.023545036,0.025808707,0.034057684,0.031127442,0.056261405,0.02002148,0.035972085,0.04840573,0.025758348,0.038366944,0.02652415,0.030842822,0.029618738,0.034543376,0.017982226,0.04520731,0.032403,0.018123262,0.04100125,0.022118514,0.023962572,0.035972085,0.020257486,0.020456282
-0.013497088,0.009385752,0.010718865,0.022892157,0.017195737,0.047109466,0.042228654,0.023229958,0.034365196,0.015841454,0.01488167,0.039592747,0.037339516,0.025289863,0.04841539,0.025228195,0.036049604,0.009608328,0.015474488,0.012289238,0.024380473,0.03366762,0.052554406,0.010148417,0.2136187,0.02663978,0.014536936,0.043314595,0.011321378,0.008347861,0.026646286,0.017845849,0.016344316
-0.009544797,0.0067682746,0.008827494,0.02225743,0.00889673,0.03658895,0.043281052,0.023995617,0.03151086,0.010858004,0.017283395,0.049428515,0.06729738,0.028021205,0.06756077,0.07025208,0.047720984,0.008758798,0.011423627,0.010320387,0.020766476,0.04203145,0.031113345,0.019470233,0.094904505,0.027490968,0.008489318,0.034540378,0.013618987,0.026697252,0.028558182,0.05139752,0.02032508
+0.016557448,0.020405825,0.017625311,0.025519764,0.03680612,0.06548598,0.03477934,0.021892255,0.029666752,0.022543117,0.019700896,0.034173325,0.027245352,0.046527296,0.022107093,0.042529393,0.057565864,0.019357618,0.046165217,0.027687918,0.035638895,0.025644677,0.032544788,0.017625311,0.044396657,0.027566511,0.015253436,0.038048714,0.016557448,0.025694814,0.04571658,0.022499131,0.018471168
+0.012519716,0.008405341,0.010099248,0.021738015,0.018539116,0.04669845,0.04799298,0.030121356,0.03165953,0.0144665055,0.012816613,0.040493153,0.034098655,0.024898428,0.05459602,0.026919158,0.040335283,0.007896087,0.016138554,0.012967691,0.025155049,0.032252446,0.047619496,0.0095991995,0.2101071,0.025044747,0.012917135,0.04153446,0.0099039115,0.009017614,0.028981574,0.019657847,0.014809568
+0.009015803,0.0061482433,0.008469563,0.021126784,0.009229606,0.037883893,0.05087888,0.031684104,0.028091125,0.010216243,0.015396422,0.050680522,0.056981638,0.026749171,0.074900955,0.0725965,0.05147862,0.007532985,0.011897433,0.010136739,0.022011328,0.039779596,0.028429195,0.019273764,0.093946844,0.02702198,0.007894501,0.032245975,0.011531387,0.026202993,0.028933344,0.053738806,0.01789508
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index b406953d..44487bad 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.015481911,0.021790396,0.019840382,0.03492312,0.020390378,0.059813716,0.027535558,0.028744513,0.04444978,0.020833192,0.014430728,0.048249777,0.029595483,0.019685984,0.057973444,0.018601947,0.035025585,0.018277813,0.016224904,0.025851479,0.029047865,0.061471812,0.022220174,0.02191845,0.041189697,0.06392057,0.01972447,0.050961923,0.024214173,0.033323877,0.018894885,0.01597336,0.019418672
-0.013983444,0.010351101,0.01371298,0.024782542,0.011960674,0.073268,0.036626223,0.02346937,0.02346364,0.022808464,0.020010805,0.03494899,0.019855078,0.04996443,0.029229501,0.023452187,0.026665796,0.013343096,0.015357809,0.02130139,0.01657335,0.20548517,0.04207427,0.009028391,0.04006921,0.015752748,0.028000148,0.020666188,0.018927356,0.024217322,0.020828273,0.027378248,0.02244389
-0.012367314,0.014458855,0.009707215,0.026529063,0.018349245,0.059930302,0.13718182,0.03231419,0.018601837,0.015572799,0.02426137,0.023676222,0.026946833,0.040948816,0.021202514,0.05785998,0.014515445,0.011172936,0.013424592,0.019570857,0.021961171,0.045237765,0.03869391,0.013635999,0.041593663,0.022746976,0.036456212,0.028863892,0.020113382,0.012033726,0.06530834,0.032015786,0.022746976
+0.012167323,0.016182173,0.019519432,0.031962857,0.019672524,0.042634357,0.030798472,0.034057528,0.04888053,0.015380938,0.010864196,0.07139911,0.027352633,0.018229676,0.069473244,0.016793964,0.05759534,0.015024639,0.014849597,0.022335472,0.023510464,0.06300949,0.016533598,0.019904418,0.051830195,0.050432164,0.013415492,0.05670241,0.024244925,0.03049917,0.019481344,0.02006053,0.015201745
+0.0109462105,0.008198333,0.012648323,0.023094064,0.010776506,0.070822805,0.049057525,0.03391485,0.023337783,0.018674767,0.014758534,0.046446733,0.017254474,0.044146996,0.03610217,0.023224108,0.042788733,0.0104449475,0.0142208915,0.019532802,0.0133592915,0.18805672,0.035611942,0.007946097,0.052837033,0.015406512,0.020093884,0.019705232,0.015497048,0.023872638,0.02404923,0.03451628,0.018656539
+0.01147871,0.0120766675,0.00944213,0.024731357,0.017097488,0.060144126,0.1535834,0.041579332,0.018271415,0.013791987,0.019640686,0.02893507,0.023210282,0.035703883,0.024359824,0.055624224,0.022061061,0.0103297215,0.013631306,0.018271415,0.019640686,0.040378857,0.03815534,0.013057991,0.04629459,0.02357579,0.027908226,0.02653935,0.01587448,0.012411445,0.06579788,0.036837246,0.019564113
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index c5305b93..bb8ba9c4 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.021211134,0.028320579,0.023478584,0.058909986,0.13095303,0.09981858,0.10119266,0.05229339,0.18758154,0.118306205,0.10178733,0.036364343,0.039782714
-0.016987843,0.024144594,0.015347238,0.05002766,0.08610305,0.046630908,0.095680416,0.25555444,0.118380524,0.08460274,0.08410848,0.03582306,0.08660904
-0.03188173,0.04349213,0.038607225,0.03515228,0.05744916,0.060088728,0.064544536,0.08375145,0.061064795,0.27327284,0.060530446,0.032006513,0.15815818
+0.015025156,0.019444002,0.016373487,0.078113556,0.1282854,0.08043589,0.11321145,0.07098442,0.1760316,0.1118925,0.11634979,0.035484683,0.038368087
+0.012999591,0.01708764,0.012599636,0.06817954,0.07748421,0.039885454,0.0934638,0.2801235,0.10305166,0.07075717,0.101851076,0.03280888,0.089707874
+0.025136897,0.034358066,0.031900384,0.042926658,0.059018586,0.05193135,0.06720416,0.11265576,0.05965599,0.23942567,0.0725941,0.02938801,0.1738044
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 5dc72934..8a8365f8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.036800325,0.042522695,0.031600185,0.123525076,0.04186344,0.10062151,0.12113588,0.06976658,0.029225392,0.18399003,0.06321378,0.057444494,0.098290615
-0.023137681,0.02156669,0.017740281,0.07512827,0.09349864,0.04173359,0.115230136,0.07210918,0.043226883,0.15873621,0.05558578,0.03740975,0.24489695
-0.03727073,0.06427239,0.031631254,0.054867815,0.053701587,0.09143873,0.115476556,0.026950166,0.027481709,0.1778087,0.052355234,0.033020034,0.23372512
+0.034291983,0.032466996,0.029909998,0.16106153,0.04377453,0.09321527,0.11443302,0.09523961,0.023753362,0.10792062,0.08903354,0.067189634,0.10771004
+0.020611567,0.017356684,0.01605231,0.09023478,0.0966189,0.037395693,0.10224941,0.094196565,0.03358682,0.112299,0.067714855,0.04229219,0.2693913
+0.03479445,0.054101624,0.031929184,0.06602827,0.057086926,0.08603304,0.10489644,0.034523677,0.022729797,0.13338077,0.05976819,0.033461496,0.2812662
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index dc7b23c0..27982a02 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.020409178,0.02520197,0.032233812,0.12212525,0.036383294,0.0602501,0.0626177,0.04644414,0.08024828,0.33719864,0.04397243,0.08328219,0.04963303
-0.018363692,0.022676133,0.023579445,0.082380764,0.027245997,0.11293152,0.13254708,0.07497192,0.07754087,0.18692133,0.097163096,0.057175048,0.08650309
-0.015567869,0.014854965,0.02337014,0.12197339,0.0134203425,0.0694305,0.18382038,0.11868336,0.12584524,0.1414916,0.04014407,0.09117559,0.04022255
+0.017545573,0.016227003,0.028927766,0.17582504,0.035030115,0.05548813,0.057663318,0.06000796,0.07401406,0.27232316,0.06309156,0.09522173,0.0486345
+0.014570021,0.015269251,0.02170198,0.118239954,0.02797496,0.09802437,0.12247076,0.116862416,0.06581848,0.12247076,0.1319062,0.06262835,0.08206254
+0.012689575,0.010939098,0.021585695,0.15763971,0.014379173,0.062643535,0.1632803,0.1765481,0.10177984,0.09413096,0.057163052,0.084543414,0.042677484
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index b1f3057f..556d2b62 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.021790942,0.02875576,0.023932666,0.11462324,0.033882566,0.074223466,0.087884694,0.03663579,0.07831908,0.31525108,0.041190725,0.09521178,0.048298124
-0.019185683,0.01774385,0.017605767,0.11346654,0.04976387,0.07859597,0.28414276,0.10032964,0.06620061,0.09333503,0.042648513,0.07882658,0.038155284
-0.049818195,0.04616438,0.07234362,0.14387244,0.037678268,0.15077703,0.11031084,0.13836077,0.054394927,0.069640145,0.04336742,0.054607823,0.028664112
-0.022936853,0.028434087,0.022405522,0.04789797,0.036652986,0.1330277,0.16882095,0.031969305,0.054701194,0.24659441,0.06426521,0.067086786,0.07520705
-0.01597938,0.02019983,0.017826287,0.037008327,0.15703289,0.073062755,0.15339524,0.051783957,0.12617949,0.14410149,0.08930007,0.02306893,0.09106136
-0.028527373,0.050263055,0.0540826,0.15242085,0.032834806,0.13770077,0.070279166,0.104757205,0.073994376,0.10133605,0.056182116,0.12104672,0.016574996
+0.019172488,0.01932286,0.023036273,0.20292799,0.03028061,0.07068008,0.0837709,0.042287517,0.07884933,0.21601573,0.054938298,0.11836687,0.040351033
+0.013898927,0.011343973,0.015028323,0.16800593,0.04053331,0.06949026,0.26021266,0.13033302,0.052788023,0.07113817,0.049856827,0.079515524,0.037855063
+0.033207193,0.02873843,0.053064942,0.20104882,0.02988324,0.13392739,0.11727074,0.17263922,0.043395016,0.05972081,0.047289208,0.053953256,0.025861789
+0.01586071,0.017016059,0.019585362,0.06508605,0.037604593,0.13971801,0.16591924,0.055089667,0.04819104,0.16919172,0.10221989,0.07776479,0.086752884
+0.011261298,0.013583719,0.013583719,0.044976395,0.14044411,0.06680432,0.15245065,0.07024145,0.10684425,0.12988958,0.116887964,0.023654649,0.10937799
+0.018765636,0.029638039,0.04139015,0.19101273,0.034113172,0.11185279,0.07816212,0.1505144,0.065538496,0.0822338,0.07793347,0.09948393,0.01936132
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 7a6d6645..bc78ed18 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.021908142,0.024061386,0.026947476,0.05529263,0.035699606,0.055726297,0.103704534,0.12558115,0.045839198,0.3591476,0.039055485,0.076020196,0.031016354
-0.031836722,0.03900717,0.033890016,0.036358677,0.06507003,0.07140441,0.04242466,0.11265075,0.088420585,0.23159625,0.14606592,0.054314982,0.04695982
-0.0118632335,0.019789722,0.010469266,0.04605744,0.101288736,0.03852003,0.21526752,0.2169559,0.096273586,0.13005732,0.03852003,0.046919998,0.02801721
+0.017322686,0.016020864,0.0268299,0.08280347,0.045728423,0.046267454,0.111206844,0.21773246,0.03754178,0.22116125,0.05969938,0.06348763,0.05419789
+0.02507379,0.024976037,0.03481151,0.056836713,0.074419186,0.054873265,0.049670782,0.18239927,0.070286855,0.1535956,0.16607644,0.048425484,0.058555067
+0.008939866,0.014064383,0.009591076,0.05823833,0.09724553,0.03360716,0.19875696,0.2737993,0.077986516,0.10722134,0.04580116,0.039870534,0.03487773
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index f5d1ffa1..466b8da1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.019913744,0.02274221,0.032832213,0.12439244,0.026381426,0.079455666,0.09334777,0.11459569,0.13293332,0.11731324,0.09190055,0.060269855,0.08392191
-0.017521279,0.033905335,0.016459718,0.045982454,0.0826155,0.044134606,0.0711495,0.08046563,0.19895993,0.1259736,0.060649738,0.052003276,0.17017946
-0.025699943,0.03512766,0.01985938,0.048484996,0.0704764,0.038580187,0.07725216,0.15866601,0.17562728,0.14334281,0.07823914,0.10631546,0.0223285
-0.021763925,0.03864713,0.032927766,0.06163743,0.033446305,0.11904154,0.23581937,0.08023351,0.070469536,0.1138123,0.043452136,0.103829876,0.0449191
-0.035014756,0.06755865,0.06371388,0.10998025,0.038008157,0.13370198,0.07059374,0.07311969,0.049186207,0.12958841,0.064717226,0.113471396,0.05134574
-0.044106077,0.03992442,0.04333755,0.04750411,0.14632298,0.071626194,0.04958979,0.079012446,0.19158807,0.16007784,0.05232613,0.05032155,0.024262724
+0.013754733,0.015586155,0.022677721,0.12847804,0.03051621,0.06587699,0.08986748,0.15864919,0.103033565,0.09547677,0.12501256,0.049436044,0.10163449
+0.013943452,0.025645932,0.013620452,0.05140305,0.07930422,0.041708842,0.07060354,0.10202771,0.17053087,0.10506101,0.07977025,0.044659745,0.2017209
+0.018764257,0.02466515,0.015314939,0.057096727,0.06734212,0.033450905,0.075309195,0.19099866,0.17664489,0.12380064,0.10006119,0.09039775,0.026153553
+0.014145564,0.020581674,0.029252458,0.070930876,0.034199588,0.10051818,0.2977565,0.12706688,0.046110548,0.08801666,0.046020575,0.077826284,0.047574252
+0.026296157,0.044383164,0.062163893,0.15014632,0.03826068,0.13198708,0.09405097,0.09818031,0.03994053,0.08800825,0.07123646,0.09924066,0.05610558
+0.034638647,0.031049881,0.038869288,0.061015755,0.14413515,0.06952924,0.05955869,0.10260714,0.18291703,0.14987685,0.054070164,0.04865788,0.023074314
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 81f95111..a0e4ca20 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.044508155,0.08047588,0.044508155,0.05337343,0.06501244,0.10866292,0.057261333,0.064758986,0.11499529,0.079421885,0.09061411,0.14939743,0.047009982
-0.012853926,0.02102762,0.016504768,0.088702664,0.063192055,0.041492984,0.12681279,0.11774168,0.07601974,0.21071875,0.06582174,0.07075447,0.088356845
-0.011193676,0.014945519,0.013086734,0.07242391,0.087019324,0.05896654,0.2649157,0.11393881,0.05971991,0.20471099,0.019190425,0.03407726,0.045811202
+0.034407627,0.052876856,0.04070077,0.09763607,0.057229374,0.09744556,0.06497616,0.10934721,0.109133855,0.04522797,0.11106918,0.13876884,0.041180536
+0.0108207315,0.016371245,0.015992004,0.12677342,0.058041103,0.041157305,0.12335392,0.1565443,0.07547841,0.14253521,0.076667026,0.06862328,0.08764204
+0.010323195,0.013255244,0.013783273,0.09987551,0.08344912,0.057697766,0.24241307,0.14760642,0.05647656,0.16022526,0.02466751,0.033980932,0.056246158
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 55254927..3724cf36 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.03258783,0.04750771,0.034689564,0.0454208,0.056416802,0.106019855,0.053991236,0.023748815,0.043766156,0.28041837,0.04988499,0.026701512,0.1988464
-0.02455211,0.0374375,0.024744675,0.051875398,0.11622004,0.027604679,0.038625892,0.22227663,0.1474909,0.11622004,0.06943205,0.04445813,0.079061925
-0.027025884,0.03122835,0.031596463,0.08698541,0.055997707,0.056547236,0.067414336,0.21009848,0.09963192,0.10380335,0.10002187,0.03098533,0.09866369
+0.027727641,0.038949795,0.032799017,0.049429864,0.06950336,0.087946326,0.038047526,0.030811826,0.03009807,0.18509464,0.07083959,0.02142626,0.31732604
+0.020138782,0.02752645,0.021105262,0.05810302,0.11679962,0.026471928,0.03851643,0.2511587,0.10718136,0.10187445,0.08782126,0.035069607,0.10823318
+0.023077397,0.02427956,0.028720284,0.102021314,0.054821804,0.052516073,0.06696441,0.23307347,0.076410435,0.08334873,0.11628466,0.028385684,0.110096104
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 6a547fcd..801421f5 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.017289048,0.016052337,0.023816664,0.022026813,0.09165594,0.034517944,0.036103908,0.02730594,0.11362295,0.14877227,0.39813527,0.0136234425,0.05707744
-0.019879099,0.019879099,0.028475493,0.0552105,0.080095805,0.04910524,0.08279988,0.06850958,0.17240135,0.12712078,0.11001386,0.107885994,0.07862336
-0.028068636,0.036751684,0.030827366,0.19407536,0.14006053,0.1583997,0.07188607,0.0728045,0.052336797,0.053473387,0.031681933,0.09230354,0.037330437
+0.011098017,0.012674333,0.016274165,0.025602855,0.10047324,0.047448613,0.043594476,0.029584043,0.11252469,0.14505006,0.39121857,0.01773458,0.046722345
+0.013920534,0.0164024,0.022595258,0.06077886,0.0823278,0.060823392,0.09757592,0.069157906,0.16761106,0.104888335,0.100870125,0.14114267,0.061905783
+0.019095892,0.028002275,0.021808194,0.20934938,0.12697682,0.17906602,0.07730733,0.07637869,0.052111182,0.04527498,0.026305703,0.106716566,0.031607028
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 97f72b85..64bbe388 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.028672202,0.035683133,0.04295809,0.05304616,0.20614591,0.06991537,0.099297024,0.038885273,0.108631276,0.13206205,0.030881178,0.055105437,0.09871691
-0.01782257,0.03189652,0.03448836,0.090687424,0.24411833,0.09002563,0.051571313,0.08114297,0.06220688,0.11419271,0.09202563,0.054790262,0.03503147
-0.038860124,0.03483399,0.03840739,0.11969761,0.24464603,0.101586044,0.07508794,0.03708059,0.077282585,0.058908433,0.050584264,0.046418726,0.07660632
+0.028366636,0.033554886,0.04283361,0.06522581,0.20635305,0.08057263,0.12913246,0.036709156,0.08909872,0.09540245,0.027493887,0.064884335,0.100372225
+0.015950827,0.027133264,0.032474265,0.11224481,0.24183615,0.102800645,0.06656782,0.08485103,0.0548107,0.08240122,0.07644393,0.067781396,0.034703974
+0.0346278,0.030558925,0.03476333,0.13378303,0.2403646,0.1126566,0.09158917,0.03893308,0.0698818,0.043517902,0.041769095,0.05421182,0.07334285
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index be8b8095..c6da86b1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.020406444,0.035674796,0.030395307,0.051704157,0.2998705,0.12500462,0.033382714,0.056234058,0.076600574,0.12598504,0.067698866,0.03051427,0.046528704
-0.015827617,0.024323527,0.020968191,0.17016296,0.15676147,0.063647375,0.036371585,0.03074782,0.1898305,0.057951596,0.07677339,0.11029547,0.0463386
-0.017605642,0.022783393,0.029254457,0.20951222,0.16063859,0.06966567,0.046384685,0.0307784,0.1224463,0.08047888,0.09829233,0.08552304,0.026636485
+0.015611125,0.028825635,0.024946507,0.05949394,0.25792566,0.17865989,0.061143175,0.05961025,0.08167672,0.07912997,0.05406415,0.05513047,0.043782484
+0.013057961,0.019449774,0.017848115,0.16736537,0.14540955,0.07191268,0.05906717,0.032956194,0.17133433,0.038080808,0.052356172,0.16933823,0.041823585
+0.015269818,0.018708972,0.024979755,0.21079336,0.16036314,0.06840117,0.06590937,0.03468096,0.12343589,0.05596387,0.06995491,0.1263631,0.025175674
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 0c828e54..b138c563 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.019000068,0.031819124,0.030243652,0.07024996,0.21407309,0.13396353,0.028858693,0.026585873,0.06696755,0.16285823,0.15180056,0.022563139,0.041016474
-0.017298587,0.02785987,0.026171926,0.022916906,0.29258338,0.047437537,0.034001548,0.03626522,0.08464918,0.20266658,0.11324245,0.06333742,0.03156937
-0.0686808,0.047946926,0.05949635,0.06376796,0.06336447,0.15089422,0.07095185,0.051540107,0.055402566,0.20384508,0.026530573,0.08195969,0.055619407
-0.028476672,0.036995783,0.024548426,0.05917686,0.32845432,0.06365823,0.01987992,0.022703582,0.13638632,0.07755896,0.11989127,0.025526324,0.056743342
-0.016698256,0.012903412,0.01798473,0.037851244,0.29368037,0.023183238,0.032629814,0.026476089,0.12052618,0.07571856,0.26531813,0.021524908,0.055505183
-0.045183532,0.035257764,0.031236647,0.11381345,0.10505485,0.08457898,0.07918349,0.04026563,0.08683875,0.17422402,0.07487796,0.07225598,0.057228997
+0.01688767,0.02476413,0.02698642,0.07116921,0.2106128,0.1646674,0.04896163,0.025952587,0.05929015,0.12724508,0.16085288,0.03118275,0.031427316
+0.014174571,0.020224977,0.022126192,0.0230076,0.35156232,0.052318905,0.053994514,0.03936724,0.0823695,0.12882835,0.09388537,0.09206946,0.026071027
+0.051626947,0.03439096,0.048215687,0.060181446,0.07837651,0.17194399,0.10227216,0.056314826,0.054316334,0.15594643,0.024674308,0.116798475,0.04494196
+0.019099979,0.023038972,0.014644466,0.044235647,0.47743943,0.06330279,0.023493377,0.01940076,0.11745949,0.038357608,0.09147754,0.02800827,0.040041707
+0.010872557,0.00997722,0.011483708,0.037251316,0.3685879,0.028645372,0.035406873,0.025929531,0.12250072,0.057527177,0.22182012,0.023840794,0.046156727
+0.03117595,0.026252784,0.023167998,0.11226891,0.13280286,0.08968357,0.100637205,0.042946685,0.09678186,0.12972648,0.068729214,0.096971065,0.048855443
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 15afec14..8748c33c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.022081608,0.018594578,0.018740417,0.103307895,0.47212556,0.04692973,0.030657277,0.045264293,0.059411697,0.078746155,0.02350575,0.05239225,0.028242808
-0.027631663,0.025856346,0.021603787,0.16926447,0.12096786,0.07964316,0.043301236,0.1349494,0.09815427,0.05714117,0.099311285,0.0701477,0.052027624
-0.026684716,0.020863418,0.016504321,0.07736575,0.10974434,0.03814951,0.06553134,0.049032796,0.21570995,0.21236569,0.061913315,0.043144707,0.062990114
+0.016345652,0.013981177,0.014882885,0.11213945,0.46299294,0.050342005,0.03998455,0.06528273,0.05401568,0.042521786,0.016603058,0.08975528,0.021152845
+0.021561448,0.02073544,0.017325114,0.17226464,0.13950415,0.09796191,0.052949805,0.16631368,0.06838009,0.038587876,0.062359657,0.103468396,0.038587876
+0.020325927,0.01738569,0.014526288,0.09583984,0.1389025,0.047259193,0.0788357,0.06192797,0.20210195,0.15195979,0.05210722,0.065693036,0.05313494
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index bf39a092..0bb283aa 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.016777907,0.029446134,0.027233222,0.09903301,0.15368521,0.065074414,0.049991827,0.071156874,0.1554968,0.06006656,0.10398861,0.11969016,0.048359197
-0.023548124,0.02840446,0.027638298,0.0627111,0.20662825,0.07437815,0.030832749,0.047290612,0.15904744,0.048412077,0.11411094,0.045656934,0.13134089
-0.020972665,0.028554477,0.028778432,0.029231627,0.05453193,0.057009615,0.037976593,0.024233868,0.18987098,0.32806766,0.0779229,0.04914536,0.073703915
-0.02953646,0.036472656,0.021108754,0.053275105,0.39211807,0.06938212,0.028074007,0.041734517,0.065625556,0.09798792,0.073174864,0.06043496,0.031075094
-0.054990366,0.03739047,0.034580532,0.07439615,0.102434985,0.16691916,0.07954307,0.051911525,0.051206626,0.2151673,0.025698068,0.05061005,0.05515171
-0.035944857,0.03580472,0.038714133,0.05932095,0.06886392,0.0958184,0.08983742,0.06345591,0.0639536,0.14272109,0.23808104,0.028769901,0.038714133
+0.011731772,0.022969693,0.018313037,0.12035259,0.14745906,0.07250908,0.05957885,0.068290815,0.15881966,0.04703879,0.07210307,0.15333316,0.0475004
+0.01647793,0.022347387,0.019721499,0.07884245,0.21705416,0.08566629,0.034075696,0.04938655,0.16707255,0.04240789,0.09426995,0.057626065,0.115051664
+0.014698837,0.023034362,0.022065567,0.03471401,0.057541963,0.07060523,0.0458093,0.026512388,0.21431695,0.279522,0.07709165,0.06710933,0.06697838
+0.019589733,0.019976106,0.014787107,0.039495137,0.57698613,0.047154225,0.023537608,0.0401759,0.043186717,0.049152493,0.045644898,0.058029015,0.022284957
+0.04979851,0.031715743,0.031346247,0.074355975,0.15856877,0.17689624,0.089297146,0.055445857,0.042346075,0.1642426,0.021044873,0.059688386,0.04525359
+0.027977334,0.027116563,0.0300153,0.06489063,0.08352489,0.118712805,0.09957648,0.06747558,0.053875085,0.14319499,0.2226528,0.02989828,0.031089291
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 4e43aa2e..f8f40bba 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.04036501,0.0526459,0.035483122,0.057146583,0.056979407,0.10676387,0.06257964,0.06425163,0.12384826,0.17465405,0.050137047,0.100884765,0.07426074
-0.0159961,0.026167873,0.02203556,0.084553465,0.22782892,0.074536234,0.040371586,0.03837265,0.13551247,0.18814044,0.07248592,0.03346922,0.040529598
-0.018406145,0.016499165,0.01400266,0.088499516,0.33661324,0.033329222,0.051798053,0.042049877,0.14087047,0.096441306,0.039967842,0.07855951,0.042963065
+0.0342464,0.038354147,0.0311817,0.06441956,0.055262636,0.10589919,0.09309136,0.071602516,0.12141403,0.13439308,0.04130904,0.15468548,0.054140862
+0.0143472245,0.022048492,0.020232834,0.09900767,0.2333686,0.08585151,0.05457054,0.04249958,0.13348985,0.1477598,0.06448847,0.047783565,0.034551915
+0.015537152,0.014036647,0.01318621,0.09705379,0.33219895,0.034673247,0.06452561,0.04620453,0.1406619,0.06882144,0.033279873,0.10210958,0.037711035
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index d35903cd..8f216025 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.04502345,0.037545096,0.032875586,0.1290136,0.23822105,0.09531464,0.04755424,0.04746145,0.110351175,0.038585886,0.04271068,0.039655525,0.09568769
-0.024061177,0.0339317,0.034197826,0.027478727,0.10804553,0.092235915,0.03138169,0.15936935,0.15446608,0.06424972,0.027264884,0.13900442,0.104313046
-0.01398243,0.020826768,0.033673387,0.059273835,0.023971463,0.071147464,0.054311454,0.12708203,0.15815614,0.043747153,0.033022083,0.31823865,0.04256715
+0.03765263,0.032458596,0.023470555,0.16874741,0.20756285,0.13245134,0.058517206,0.05027286,0.09747293,0.02659562,0.028421698,0.04208667,0.09428967
+0.01482675,0.025518453,0.02444518,0.04046103,0.08994222,0.11892158,0.036410987,0.1750695,0.14627604,0.042735364,0.020108057,0.18062681,0.08465808
+0.010264666,0.017460784,0.026728721,0.06693403,0.021815348,0.07467031,0.06196434,0.12652327,0.14004847,0.03016953,0.022773158,0.36467987,0.03596743
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
index a0bbf05d..8cd9eea8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0051731854,0.006115724,0.004452979,0.02984497,0.0035049068,0.023042496,0.04243705,0.017752096,0.043330517,0.005744317,0.005712806,0.02443036,0.03076382,0.016793936,0.020566238,0.014915136,0.23716973,0.006960561,0.016510267,0.008906468,0.016893843,0.032925818,0.022828635,0.0059587103,0.08213512,0.101522684,0.009693471,0.08036983,0.0065862564,0.016128413,0.025845902,0.011681713,0.02330209
+0.0053320643,0.005789764,0.0048760655,0.03720358,0.0035652393,0.025075478,0.04443915,0.016188094,0.035687704,0.0061186426,0.005490237,0.021655092,0.025955843,0.01783606,0.021378342,0.014357306,0.21706626,0.0067667514,0.016861673,0.011250843,0.018886115,0.02848178,0.022863349,0.006047647,0.06405485,0.14882801,0.010209717,0.07615009,0.0072470545,0.015511747,0.021830576,0.013150852,0.023844091
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
index 68d35a3b..fafc83a6 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0051706634,0.006392545,0.004375827,0.030993352,0.0034347752,0.021954894,0.041624684,0.017873196,0.04168018,0.00586722,0.005894553,0.022671,0.0292719,0.01654101,0.019867422,0.013532966,0.2661518,0.007395097,0.01631813,0.008761997,0.016077364,0.033230375,0.020873664,0.0064080334,0.07242437,0.097736195,0.008801166,0.08053392,0.0065339063,0.015795996,0.02301082,0.012271813,0.02052921
+0.005414255,0.0058297557,0.0046986197,0.036962785,0.0031884725,0.025659159,0.04501603,0.017804626,0.042272698,0.0064299456,0.0062360885,0.021977006,0.028345384,0.016593426,0.020529402,0.01365076,0.22053148,0.007261174,0.018225364,0.011267549,0.016855016,0.032483425,0.02040384,0.006038296,0.061356742,0.13856283,0.010517521,0.07868338,0.006724646,0.016816795,0.021524182,0.011967941,0.020171447
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
index 6ae5b497..a0bb7db1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0064157196,0.006527345,0.0056665037,0.027466085,0.0044602095,0.026517421,0.049759857,0.01652433,0.041261666,0.006874849,0.0067948974,0.023469295,0.029153131,0.016635327,0.019607816,0.016634526,0.17611296,0.007703006,0.023564544,0.010035094,0.022443883,0.03051713,0.025578223,0.0072342893,0.08143584,0.14751312,0.013761597,0.055317592,0.0074805734,0.01573484,0.03309507,0.01149615,0.027207157
+0.010200325,0.008255294,0.008725317,0.028328693,0.0071364073,0.031695973,0.050800975,0.014434751,0.036991633,0.010709722,0.009840103,0.023735259,0.027150556,0.015559098,0.017729346,0.021140106,0.07350094,0.010444301,0.03560199,0.015230986,0.03764622,0.023684604,0.0285523,0.010491406,0.051373497,0.22636293,0.02416806,0.02768691,0.010378675,0.017089728,0.040571064,0.013346023,0.031436794
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
index 99c55a45..c4c85200 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0055406922,0.006386817,0.004625329,0.02794532,0.0038382933,0.023122994,0.047041234,0.017717674,0.043066762,0.0060848948,0.006028181,0.024094954,0.031411633,0.017373681,0.020232791,0.016327478,0.21710728,0.0073115835,0.017294787,0.008706568,0.019171692,0.034296166,0.022187935,0.0064854,0.08433365,0.10816069,0.01044296,0.07764751,0.0066119726,0.016789336,0.028201265,0.012045506,0.022366948
-0.005273,0.006350921,0.0045870915,0.029855195,0.003674753,0.022280036,0.04056523,0.018330282,0.042522132,0.0058698533,0.005789554,0.023899432,0.030028421,0.016782813,0.018927349,0.014220322,0.25088266,0.0070974855,0.017512802,0.008408013,0.01705966,0.032813244,0.022579813,0.0065027904,0.08215564,0.09919083,0.009322651,0.072932474,0.0066912686,0.015727002,0.02790005,0.011463137,0.022804117
+0.006770478,0.0070975553,0.0061097112,0.032388054,0.00469689,0.029014563,0.054458927,0.015219773,0.036924854,0.0077053085,0.0074958084,0.02382878,0.028791584,0.017738653,0.01888046,0.020983137,0.11063638,0.008250365,0.025185334,0.012733429,0.02757118,0.027456699,0.025177583,0.008029096,0.0590341,0.21493421,0.017014707,0.05102823,0.007797203,0.017581401,0.030098991,0.014413131,0.024953334
+0.006460396,0.007043518,0.0061552622,0.043098416,0.0042703887,0.024615532,0.040515337,0.016898174,0.034789506,0.007203162,0.006736328,0.023457414,0.025878392,0.016775291,0.016562976,0.015162665,0.20098755,0.008355655,0.024063585,0.011879982,0.021613166,0.027997697,0.02484585,0.007972056,0.05440866,0.17081757,0.011742848,0.048810504,0.00834056,0.015540331,0.02875625,0.013147283,0.02509769
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
index ac00fb4a..f67dd196 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005243044,0.0068628322,0.004119111,0.027635053,0.0037787466,0.021621196,0.03984544,0.020376597,0.036851183,0.005732252,0.0054553016,0.02128327,0.026677908,0.019110823,0.01986349,0.013093736,0.30414927,0.006873597,0.014688124,0.007742951,0.014003141,0.031778734,0.020403402,0.006862341,0.07797771,0.064194806,0.0077092745,0.088538475,0.0069644405,0.016042879,0.022522885,0.012781177,0.019216869
+0.0055265864,0.0069941594,0.0047999513,0.03426367,0.0041857036,0.022670772,0.03401926,0.020106943,0.023731828,0.0055064764,0.0053447303,0.014821653,0.01768129,0.021764157,0.019465126,0.010148312,0.37970722,0.0066347406,0.0131839905,0.0095445,0.013646816,0.025851715,0.020261735,0.008503725,0.052374292,0.05979142,0.0068904874,0.08166511,0.008144954,0.014931847,0.014926138,0.01588961,0.017021054
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
index 545f567e..3aa6ada2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005483725,0.0066012214,0.005198228,0.033509348,0.0042043817,0.02317797,0.03937999,0.017307758,0.037636165,0.005854237,0.0058676037,0.02215927,0.024298837,0.016519763,0.01830959,0.013106523,0.27282035,0.0074146762,0.019438406,0.0094330795,0.016409066,0.027353697,0.023160962,0.0076946104,0.074612305,0.11091272,0.0100175105,0.06297743,0.007235825,0.013821495,0.024065388,0.011177126,0.022840766
-0.005132123,0.006258132,0.004523277,0.028863266,0.0038300497,0.02196964,0.043445766,0.018415095,0.051318515,0.0059415693,0.0059717335,0.024573747,0.03365483,0.017032608,0.022367323,0.015106722,0.21536613,0.0069446526,0.016772965,0.008130857,0.016136188,0.037611797,0.023627339,0.005972406,0.0856897,0.09184303,0.009807569,0.08422559,0.0065450855,0.01732655,0.02972546,0.012266841,0.023603454
+0.0079358015,0.008644697,0.008528162,0.050991315,0.006794538,0.027761,0.0358331,0.017671429,0.028613957,0.008022484,0.00793928,0.02051466,0.018362291,0.019954775,0.018463802,0.013146331,0.21864693,0.010478101,0.024880765,0.014835908,0.021379687,0.021088064,0.027753515,0.012069517,0.05148979,0.15231076,0.013349637,0.044053532,0.01088003,0.014509736,0.024076153,0.014572965,0.024447259
+0.005569799,0.005660792,0.005039816,0.03242026,0.0038034345,0.024715062,0.04848758,0.019317342,0.07592577,0.0073058056,0.0071936473,0.028503245,0.039357413,0.016763123,0.02464078,0.017066706,0.107044116,0.006419049,0.021850476,0.009070911,0.016291386,0.042864792,0.028132914,0.005143986,0.0739408,0.13126431,0.012961439,0.079355635,0.007179912,0.021646269,0.03454433,0.011274988,0.029244097
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
index 981d5e28..e7303aa2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0055701155,0.0063301595,0.0042756563,0.029668182,0.003665929,0.023308653,0.041472793,0.01839364,0.03860736,0.0060253735,0.0056214742,0.020402266,0.027661264,0.016783247,0.01859294,0.013065602,0.28502858,0.006916266,0.01649864,0.008324767,0.015853647,0.026072072,0.02126262,0.007229569,0.07821424,0.10320244,0.009021552,0.0650905,0.006879877,0.015151082,0.022952529,0.012675275,0.020181743
+0.006583762,0.0068928897,0.0048378515,0.03660268,0.004346066,0.02784286,0.042631052,0.017884584,0.033521336,0.0069473158,0.00634369,0.018111208,0.024766268,0.018511454,0.018494321,0.013050359,0.25470728,0.007486415,0.018900618,0.010490363,0.018111968,0.02067791,0.02412442,0.009464144,0.0647224,0.13972172,0.010750387,0.05095635,0.008137707,0.015822683,0.021603521,0.016543148,0.020411285
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
index af6571aa..27cebee6 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005370935,0.0065400284,0.004214182,0.028507866,0.004052519,0.023884999,0.0431287,0.01866619,0.038549002,0.0058642467,0.005612742,0.022007838,0.028794203,0.018062135,0.018920997,0.014034251,0.28207308,0.00724066,0.01554973,0.008246547,0.015637344,0.029362565,0.020395592,0.0072190757,0.08529724,0.08158563,0.008495348,0.07213983,0.0073112454,0.015740085,0.024467451,0.01239826,0.020629534
+0.008046482,0.010288932,0.0069967234,0.03990826,0.007162291,0.03375451,0.03726217,0.021291917,0.027013969,0.007882263,0.0078681735,0.01970519,0.02320241,0.024027871,0.017250976,0.014517644,0.26696384,0.011102694,0.021171335,0.014237669,0.020665448,0.017218802,0.024644138,0.013858891,0.057388205,0.10421708,0.01121666,0.042277344,0.01241735,0.016171634,0.01957094,0.01778471,0.02291343
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
index 9391aa2a..46d86b6f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.003818576,0.0031525614,0.0051886565,0.027662326,0.006760011,0.041391376,0.062559254,0.016534436,0.041413,0.006456904,0.0038409191,0.031356983,0.025992455,0.02640143,0.038692255,0.022995137,0.17438106,0.00587134,0.01310105,0.012426717,0.017008489,0.037873317,0.023261664,0.0044478937,0.07316778,0.08910169,0.00722368,0.09950632,0.0060560848,0.021839175,0.01798219,0.013436582,0.01909868
+0.0012847829,0.0013862588,0.0013743622,0.0049335216,0.002629261,0.04421479,0.07391029,0.04052809,0.04858283,0.0037130984,0.001400662,0.06348066,0.056958914,0.019132532,0.07674131,0.02009592,0.03749449,0.0011936828,0.0070741074,0.0039621834,0.0048973863,0.08357048,0.02983897,0.0016369526,0.09871048,0.0017693046,0.013229287,0.16916683,0.0018953648,0.019495614,0.012248143,0.0025117958,0.050937597
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
index 79957644..9d87985f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0038112525,0.0032467076,0.005103138,0.028848037,0.006253271,0.040058028,0.06397161,0.015401709,0.039240718,0.007229387,0.0042264964,0.031109253,0.022743622,0.026395943,0.03867059,0.0224054,0.16184604,0.006153928,0.014256049,0.012480883,0.018035192,0.033755783,0.022141878,0.0048214206,0.06857576,0.11654172,0.0072527346,0.091533765,0.0067206784,0.023300141,0.021024033,0.014561617,0.018283268
+0.015949078,0.013059553,0.009781349,0.048254672,0.01312685,0.01593082,0.028322889,0.007367935,0.007820241,0.0077832667,0.014817788,0.002786293,0.010103129,0.011100873,0.013910568,0.012986693,0.032847404,0.017707605,0.018416008,0.024353962,0.032842595,0.002778024,0.03274647,0.05513109,0.009160907,0.3112136,0.0045494093,0.0028305363,0.019623127,0.03260526,0.07252648,0.09204435,0.0055211657
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
index fc35a03f..110272db 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0038847511,0.0034304704,0.0054487633,0.030906213,0.0065313987,0.04176309,0.0651858,0.015840625,0.039366536,0.00712093,0.0041720322,0.030338436,0.021227272,0.027650725,0.03322719,0.02358945,0.17575502,0.0062058195,0.015167547,0.013033766,0.018510986,0.03583581,0.021288542,0.0049735317,0.066214904,0.10693726,0.0075933044,0.08376561,0.007044815,0.021833733,0.02074671,0.015919521,0.019489419
+0.0006185576,0.0006328632,0.00043876236,0.0064425473,0.0018244528,0.065627284,0.1284394,0.011077401,0.04086932,0.0013103667,0.00057274237,0.05383014,0.0138934385,0.03555001,0.035452865,0.016880834,0.100283615,0.0006685078,0.013122532,0.008209185,0.014611625,0.022872807,0.024365555,0.0038333896,0.035221875,0.08512628,0.003992684,0.14880891,0.00076322147,0.02621189,0.05397747,0.01884453,0.025624936
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
index dfae95aa..d92c0440 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0038600573,0.0032016537,0.0050423536,0.030087678,0.0062640845,0.042983707,0.06558176,0.016441714,0.037939113,0.0068796654,0.0041524945,0.033378396,0.02355689,0.02613971,0.03884025,0.022489678,0.16319337,0.005882346,0.0136809675,0.012107426,0.018530237,0.035350807,0.02193522,0.0048262407,0.069117285,0.106256664,0.007452489,0.08964618,0.0068337126,0.02311564,0.021159057,0.014839026,0.019234167
-0.0038384346,0.0032803454,0.005382596,0.028020917,0.0065062335,0.040472195,0.059231486,0.01510768,0.044317365,0.00677788,0.004063074,0.030269165,0.023738174,0.028254144,0.035682805,0.02283463,0.17437491,0.0062798257,0.014255069,0.013157966,0.018651582,0.03610014,0.021770932,0.0047950475,0.06709478,0.10704565,0.007234076,0.09272612,0.0065334365,0.020849966,0.019035984,0.013944705,0.018372733
+0.008535063,0.010116155,0.004999788,0.027722772,0.007546305,0.050576966,0.040861025,0.009750304,0.0048526525,0.0060036117,0.007846482,0.01775664,0.0069292323,0.034353506,0.021897852,0.011217101,0.012552527,0.009145636,0.011646283,0.026980927,0.0408808,0.010849433,0.05207903,0.02948556,0.016560948,0.22770509,0.00622852,0.007239846,0.012705731,0.037399787,0.10941857,0.108829156,0.009326774
+0.0008166164,0.0005132967,0.00059320376,0.006916493,0.002581364,0.021923427,0.050397992,0.009789066,0.11758413,0.001491984,0.0006522989,0.013462131,0.023066303,0.012532,0.02107708,0.016029773,0.19077104,0.00079144875,0.01853323,0.005924662,0.009348377,0.015089126,0.008362901,0.0038307437,0.022829419,0.035089612,0.0039825533,0.32077447,0.0008275365,0.014444679,0.024167405,0.0045341994,0.021271454
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
index 4800fc39..135277b4 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004071277,0.0036563596,0.005473332,0.026431864,0.0070418348,0.040103372,0.058523994,0.014499132,0.048744615,0.0068927244,0.003979709,0.024812536,0.02332055,0.028595718,0.034942232,0.022658326,0.19808845,0.0063588833,0.014898518,0.014481609,0.017180959,0.036216,0.021954877,0.004474714,0.07242044,0.09361984,0.0072108386,0.08443967,0.0062276307,0.019533861,0.01759654,0.013579895,0.01796969
+0.0006395509,0.0005395344,0.0006403406,0.0073080775,0.0016307676,0.04628504,0.114420965,0.033811003,0.04748827,0.0018689088,0.0006519,0.033128954,0.04977551,0.015585043,0.048301417,0.019319415,0.11359712,0.0006475441,0.0067579355,0.0037142283,0.0035758512,0.040649626,0.021625642,0.0012714841,0.08745198,0.003365021,0.0064466624,0.21271876,0.0010225932,0.022955855,0.010015839,0.0032300227,0.03955918
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
index bda5f04e..e5581ce7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0042564487,0.0033090666,0.0054760864,0.029484103,0.0063959234,0.032995798,0.058816995,0.013607776,0.048283372,0.0057379017,0.0038245237,0.021225223,0.023702899,0.026598837,0.030306822,0.02262228,0.23025683,0.0063128057,0.014794539,0.013137006,0.019704955,0.031384137,0.019589942,0.004760695,0.06250796,0.106131054,0.006412059,0.074764796,0.005834489,0.019974858,0.01842272,0.012555495,0.016811622
-0.0038192424,0.0033686042,0.0054370016,0.02769945,0.006779793,0.04220823,0.05973473,0.015520086,0.04562838,0.0071383878,0.00422071,0.03299786,0.025394814,0.0281485,0.038514096,0.023156043,0.16337366,0.0063498737,0.014283524,0.012739055,0.01869462,0.036976255,0.023111258,0.004908983,0.07237423,0.09736417,0.0072591524,0.08847638,0.006704482,0.022873966,0.021077238,0.0138563765,0.019810876
+0.0017209036,0.0015223114,0.0009925718,0.017091438,0.0052904864,0.029449824,0.05233395,0.005048244,0.11915878,0.0020151085,0.0016928521,0.011555765,0.01666688,0.016796239,0.025025416,0.015222794,0.24491838,0.002148038,0.034079727,0.010994179,0.035693437,0.0076755397,0.015851151,0.01280274,0.021139963,0.12886791,0.003564564,0.062393457,0.001786263,0.017634643,0.0507273,0.016218634,0.011920487
+0.027149918,0.03692235,0.0297103,0.07504655,0.025415707,0.06079237,0.025890756,0.02272309,0.008084381,0.018789234,0.039250992,0.011783172,0.017395535,0.029108455,0.016487945,0.0140047455,0.021839352,0.052460495,0.012594856,0.029682744,0.036679644,0.009503126,0.053610098,0.036722347,0.033857927,0.029855149,0.0132688945,0.0007574541,0.05655436,0.031545028,0.02689056,0.07513626,0.020486278
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
index 0c133e09..fd8cc3b9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0038691815,0.0031717422,0.005064685,0.029039107,0.0061659142,0.037487313,0.06612843,0.01513539,0.03811665,0.0066099614,0.0039828895,0.026459182,0.02207245,0.029161863,0.032705896,0.02051133,0.18691592,0.0063597304,0.0139315445,0.013663895,0.017307293,0.030300051,0.019515662,0.004841311,0.059204914,0.1291775,0.0065799546,0.08632184,0.006249234,0.022190582,0.020693045,0.014265866,0.016799755
+0.00085969456,0.0007951353,0.00066271354,0.018881613,0.0021610884,0.061503798,0.16510883,0.011513276,0.036499035,0.0013508685,0.0008861966,0.012494069,0.020483125,0.02126216,0.037900373,0.016543375,0.20081232,0.0012051113,0.011987639,0.0088223675,0.012496148,0.008076034,0.025651107,0.005890034,0.034473073,0.09267589,0.00274839,0.04009039,0.0014759087,0.044728093,0.048980862,0.031187622,0.019793583
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
index 1ffc4204..fc6463da 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004423313,0.004286555,0.00660057,0.023772126,0.008373217,0.040369432,0.054351408,0.01465567,0.05245067,0.008132533,0.0042896164,0.029554067,0.024817849,0.03035214,0.039045297,0.025620734,0.17074145,0.0072791767,0.015768386,0.016019871,0.01698007,0.041483015,0.025339559,0.00460742,0.08207413,0.08653425,0.008505664,0.07673737,0.00643838,0.020246439,0.016483322,0.014442152,0.01922423
+0.0068634474,0.009162191,0.009985762,0.015698336,0.0108448705,0.0904677,0.042359415,0.06670113,0.03318652,0.011744499,0.008607542,0.061853275,0.05889687,0.030315023,0.05424857,0.02326838,0.035711277,0.007862204,0.008442241,0.014744367,0.0083070565,0.06985575,0.029260479,0.0033493678,0.15207788,0.0012586035,0.028393278,0.01334909,0.010879728,0.022950096,0.0053134672,0.0070398482,0.047001738
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index 35ddcd10..11f16da0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.029515171,0.029384783,0.02892698,0.031414278,0.030240053,0.03790003,0.031263754,0.030703781,0.02985053,0.03231135,0.030141735,0.028332692,0.033920184,0.027992733,0.031994548,0.032312855,0.032003276,0.029009493,0.025961198,0.028668242,0.031744353,0.028004343,0.0351939,0.028472656,0.033312727,0.029293254,0.025253193,0.03191485,0.031521086,0.030837052,0.027421221,0.028424587,0.026759079
+0.031412095,0.029059455,0.028553972,0.032031063,0.03308489,0.036242105,0.030798635,0.031423856,0.029171485,0.033814613,0.03109685,0.027981894,0.031494338,0.028169379,0.03019742,0.03245603,0.031989317,0.03025197,0.027243363,0.028524006,0.030921329,0.026887607,0.033563793,0.028870368,0.032369044,0.030635152,0.025346385,0.0325448,0.03363609,0.029415186,0.025767326,0.027363604,0.027682606
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index b475e5bf..76fb6eb8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02660504,0.030707078,0.032001417,0.035039,0.029511122,0.034290012,0.03181159,0.0294397,0.0321122,0.03097334,0.0332632,0.027464185,0.030202633,0.028321788,0.025869891,0.03017357,0.031728413,0.033907395,0.02845779,0.031313535,0.028817244,0.02832817,0.033035055,0.026313312,0.030338703,0.028560834,0.03160956,0.031486206,0.030627396,0.029206958,0.030443136,0.027497737,0.030542677
+0.02718314,0.030715112,0.031837214,0.035137605,0.030595766,0.03317532,0.031649627,0.029798042,0.03140365,0.031467617,0.03369167,0.027727356,0.028963719,0.027858147,0.025119768,0.030851895,0.03187139,0.03497891,0.02884058,0.031090252,0.028769664,0.028061062,0.032640804,0.026473435,0.030201206,0.02919215,0.031749595,0.031676207,0.03188455,0.02838132,0.029804083,0.0268488,0.030360317
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index ca893131..f277d42b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027332451,0.031302933,0.030766888,0.034912553,0.029892711,0.03336584,0.031323344,0.029848998,0.03191401,0.031425472,0.033242542,0.027521798,0.029320594,0.027383283,0.025944008,0.03078899,0.032123405,0.03406133,0.028520554,0.031947363,0.0294212,0.028340716,0.032867085,0.02619513,0.02996483,0.028789517,0.03081592,0.031394005,0.03228332,0.029442273,0.02991017,0.026885256,0.03075153
+0.027903663,0.031013355,0.030396728,0.034959204,0.031085547,0.03232213,0.031048927,0.030432688,0.031227158,0.03177365,0.03364062,0.027822366,0.027917664,0.027073784,0.025336333,0.031431273,0.032738667,0.034493793,0.029069608,0.031952385,0.029592805,0.028263152,0.032590456,0.026255373,0.0297492,0.029384159,0.0307498,0.031711757,0.033399194,0.028632063,0.028912678,0.02644551,0.030674366
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index 736eedb1..1fd43647 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03185726,0.028593035,0.029586729,0.02717517,0.029036317,0.03239331,0.030127635,0.029756432,0.028826356,0.03008631,0.027377693,0.03182702,0.03495427,0.032569427,0.03616851,0.031138953,0.02905974,0.026557622,0.028117852,0.02744148,0.032749053,0.030639345,0.031042855,0.03220263,0.033083834,0.030625483,0.027325926,0.030158056,0.028103651,0.03071186,0.029626483,0.033320524,0.027759206
-0.031184139,0.0276112,0.032012727,0.027526684,0.029155925,0.032260306,0.030398509,0.029150296,0.029186625,0.029938763,0.02798222,0.03192229,0.034919117,0.034535985,0.03453919,0.0301626,0.02843686,0.026707487,0.02883699,0.027758451,0.03170132,0.030358283,0.030214984,0.032069325,0.033027902,0.029928787,0.02964113,0.029945217,0.026400918,0.029637858,0.030964117,0.033538274,0.028345458
+0.032669287,0.02876315,0.030224655,0.027241936,0.029947326,0.031176692,0.029444885,0.029328013,0.028524838,0.03066203,0.027863681,0.03166694,0.034933478,0.032277443,0.03640189,0.029949818,0.028301675,0.026075223,0.029140284,0.027568853,0.0323754,0.03053253,0.029847682,0.033827864,0.032350555,0.0302727,0.027539514,0.029550968,0.028082011,0.03102456,0.030212617,0.03395407,0.028267471
+0.03143606,0.028259205,0.03223914,0.027423099,0.029374318,0.030516151,0.029467668,0.028255668,0.028395286,0.030080462,0.028519074,0.031538118,0.034919344,0.03331209,0.034968793,0.029315263,0.027845407,0.02649866,0.030031862,0.027775807,0.032202866,0.030588228,0.028885216,0.033773314,0.03254831,0.030060904,0.03000054,0.02907594,0.026533103,0.030174984,0.03166629,0.035144202,0.029174669
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index 136902ac..9c79cfda 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027090317,0.031356793,0.031175932,0.034752592,0.028963773,0.035918295,0.032366708,0.029567668,0.031858996,0.031686503,0.031844135,0.027539104,0.03215706,0.02786867,0.028114032,0.030691825,0.03100698,0.032229245,0.026652839,0.030048568,0.02923739,0.02832744,0.035018757,0.0267673,0.031383105,0.028060874,0.028454317,0.03174636,0.030990787,0.030642902,0.030124784,0.02802195,0.028333979
+0.027990196,0.031357266,0.032083645,0.035231497,0.030950967,0.03400702,0.031629156,0.029572189,0.031043753,0.03288654,0.03334564,0.027693558,0.030571442,0.027530046,0.027211161,0.030589903,0.030670388,0.03259379,0.02737541,0.03036181,0.0295517,0.027888844,0.034106847,0.02740788,0.030691696,0.027825894,0.02871136,0.031667218,0.032172386,0.02943597,0.029995803,0.02745374,0.028395284
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index 7b1ad181..b0bcf281 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.031245146,0.026523279,0.03429213,0.028590597,0.030950105,0.030549498,0.030602774,0.029434195,0.02976708,0.030371303,0.02923278,0.03193279,0.032057486,0.036185876,0.031233333,0.028832216,0.028311577,0.027465537,0.031915396,0.028459525,0.029657632,0.029486306,0.02885231,0.031842597,0.032040454,0.029984664,0.032361843,0.029286683,0.025559865,0.028563855,0.03226416,0.032237228,0.029909858
-0.031189607,0.027768781,0.03171535,0.0286404,0.02989582,0.03686473,0.030307468,0.029793784,0.029219873,0.03136719,0.029169736,0.030565001,0.035063583,0.033113644,0.033308994,0.03176959,0.030947847,0.027378758,0.026388302,0.028127484,0.03166254,0.02842469,0.033268318,0.029070573,0.033865992,0.028430037,0.027634714,0.031596266,0.02863526,0.028753258,0.02869571,0.030206194,0.027160522
+0.0324019,0.026557397,0.03567093,0.028395623,0.03214945,0.030262461,0.029736726,0.02918344,0.02924375,0.031181242,0.030299323,0.031787664,0.031068454,0.03598506,0.03070766,0.028389841,0.028155206,0.027619353,0.033341076,0.028269062,0.029442511,0.028486561,0.028317692,0.032146245,0.031487968,0.02932897,0.03275529,0.028782543,0.025840137,0.02867312,0.032408092,0.031715825,0.030209484
+0.03502903,0.026921587,0.029170362,0.02918356,0.03458919,0.03605911,0.029871225,0.03156951,0.027554981,0.033587333,0.03044571,0.029222019,0.030610982,0.032198284,0.030960076,0.033182647,0.031899124,0.029759252,0.028811408,0.026992684,0.030438298,0.025892453,0.03120597,0.0292835,0.03345653,0.031768657,0.026362149,0.032247383,0.03288019,0.027643709,0.025063304,0.027542502,0.028597245
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index 5bd5387d..cb27b45f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03022107,0.026185002,0.035542563,0.030422375,0.030111654,0.032666244,0.03029986,0.028772637,0.029862503,0.031565245,0.030188533,0.031473093,0.03380565,0.034408633,0.031200787,0.029270852,0.029954033,0.026347158,0.029255822,0.028777694,0.031132765,0.028987736,0.03171347,0.029104978,0.032444056,0.027695423,0.03135624,0.030321337,0.025290275,0.027976021,0.03161053,0.03318183,0.028853968
+0.02995065,0.026482854,0.036322676,0.030786272,0.030653637,0.030291405,0.02871537,0.027630482,0.028931543,0.031819902,0.03167783,0.030811392,0.033273786,0.03227268,0.031228758,0.028594691,0.030216627,0.026278855,0.031208765,0.02872123,0.031650018,0.029130049,0.030444073,0.030609695,0.03161625,0.028118087,0.032231912,0.029491147,0.025260435,0.02785982,0.03239316,0.0353287,0.029997202
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index 211a7d66..006a90b1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.030450879,0.031356543,0.026286202,0.030208621,0.029305283,0.033798173,0.029997792,0.031071521,0.029736698,0.031199083,0.029278152,0.029431423,0.03262524,0.026415547,0.03385641,0.032870974,0.031922586,0.029316304,0.026414368,0.029405521,0.03292572,0.03013779,0.033768725,0.029333936,0.031706,0.030111287,0.025230156,0.03123331,0.033786975,0.032116424,0.027622495,0.029445931,0.027633844
+0.031105757,0.031319205,0.025952792,0.03074406,0.03020676,0.03298403,0.029882036,0.031156622,0.029681066,0.031637415,0.029420717,0.029296836,0.032249913,0.026572065,0.0334537,0.03261439,0.03169759,0.029466227,0.026537174,0.029645033,0.032349445,0.030258182,0.032920852,0.029934868,0.031194655,0.030517844,0.025205344,0.031641852,0.03456159,0.0313981,0.027084364,0.029406648,0.027902825
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index 488a103e..a462c215 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.069280244,0.08278568,0.072002776,0.068465054,0.073495515,0.069683656,0.08337737,0.08428103,0.079714045,0.08516558,0.07238791,0.07722001,0.08214111
+0.07371652,0.08747254,0.07478085,0.066930644,0.07320184,0.068280794,0.08230257,0.08293058,0.07530409,0.0836406,0.07084773,0.0773642,0.083227016
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index 287c3c11..1ffe1910 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.08120824,0.0770729,0.080374785,0.07006595,0.08169089,0.06985054,0.077308096,0.06746512,0.08158482,0.078970395,0.07628951,0.077936,0.08018279
+0.0846475,0.07865937,0.0813897,0.06975984,0.08057894,0.07018437,0.076035455,0.068347126,0.07994533,0.07798379,0.07339249,0.07953821,0.079537906
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index 6cd732d3..b9913b9f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.081267715,0.07711225,0.080673225,0.07033195,0.08024122,0.06911172,0.07555922,0.06642563,0.0804719,0.08202005,0.07461091,0.079976365,0.08219792
+0.083180524,0.07846569,0.0812608,0.07024677,0.07927988,0.06894813,0.075323164,0.06711979,0.07884185,0.08134555,0.07181041,0.081516474,0.08266096
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index 0caabb64..0ef1760e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.06782719,0.078334294,0.07045661,0.0793704,0.073115356,0.08177368,0.08440978,0.09057371,0.076156095,0.07454853,0.078214675,0.071796,0.07342377
-0.07023328,0.0774746,0.07357778,0.07948175,0.07633355,0.08361607,0.08418787,0.08768755,0.07620843,0.06998042,0.0815577,0.06901032,0.0706507
+0.06989463,0.07989096,0.07249861,0.08051619,0.07235672,0.081912465,0.08144418,0.08826559,0.074629635,0.07509785,0.079737626,0.071991235,0.07176424
+0.07401593,0.07802295,0.075928785,0.082235724,0.0748795,0.0842309,0.08068106,0.08477475,0.07419375,0.07135396,0.082080215,0.06978505,0.067817405
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index 75244a2c..30d5cd15 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.075564295,0.077890314,0.07621962,0.06734453,0.07971224,0.06832646,0.07854212,0.07214813,0.08339084,0.082671195,0.07736374,0.0789834,0.08184321
+0.07915403,0.08103098,0.07856794,0.06742558,0.07797891,0.068986446,0.07662418,0.0694121,0.0813124,0.08137604,0.07515498,0.081329875,0.08164642
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index e8c6c744..6911d301 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07543901,0.07629552,0.079985805,0.07938394,0.08105448,0.08392259,0.08113814,0.08250335,0.07510063,0.06616155,0.08535827,0.06484661,0.068810105
-0.0684685,0.08285618,0.07265628,0.07221835,0.074672826,0.07720729,0.087464236,0.08804019,0.07745041,0.072739474,0.07616589,0.07260381,0.07745654
+0.07836008,0.07883829,0.08306367,0.078179605,0.080974005,0.08502495,0.078466825,0.080645144,0.07336398,0.06484835,0.08606815,0.06423272,0.06793416
+0.0754353,0.091055885,0.07524671,0.07041598,0.07320034,0.07423061,0.087269105,0.0889919,0.069685824,0.073581785,0.071758285,0.069817245,0.07931107
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index 0a26093e..aec1ea17 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07194322,0.077352665,0.08031595,0.07728969,0.08099176,0.080330044,0.08255197,0.079121925,0.07681379,0.069612056,0.08324984,0.06879602,0.07163119
+0.07717769,0.07774059,0.083538964,0.08017507,0.07791889,0.08045961,0.07922038,0.07712727,0.07463028,0.07106932,0.08220443,0.07093723,0.06780032
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index fdcbac81..6558c6eb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.068312116,0.08032282,0.06863922,0.07357371,0.06984851,0.07086958,0.0803648,0.0826013,0.079164565,0.08935935,0.06972866,0.0837885,0.083426945
+0.06911243,0.081835695,0.06873813,0.0740041,0.06911324,0.06964999,0.080704,0.081733346,0.0777465,0.08935496,0.06924195,0.084877074,0.08388858
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index e296979b..abae3f67 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -1,3 +1,5 @@
+import copy
+
 import pytest
 import torch
 
@@ -69,6 +71,20 @@ def model(model_config):
     return TransformerModel(model_config)
 
 
+@pytest.fixture
+def causal_model(model_config):
+    config = copy.deepcopy(model_config)
+    config.training_spec.training_objective = "causal"
+    return TransformerModel(config)
+
+
+@pytest.fixture
+def bert_model(model_config):
+    config = copy.deepcopy(model_config)
+    config.training_spec.training_objective = "bert"
+    return TransformerModel(config)
+
+
 def test_transformer_model_initialization(model, model_config):
     """Tests that the model initializes with the correct layers."""
     # Check if encoder dicts were created
@@ -184,3 +200,95 @@ def test_calculate_loss(model, model_config):
     assert total_loss.item() > 0  # Valid loss value
     assert "cat_col" in component_losses
     assert "real_col" in component_losses
+
+
+def test_padding_keys_are_masked(bert_model):
+    seq_len = bert_model.seq_length
+
+    valid_mask = torch.ones(
+        2,
+        seq_len,
+        dtype=torch.bool,
+        device=bert_model.src_mask.device,
+    )
+
+    valid_mask[0, :2] = False
+    valid_mask[1, :1] = False
+
+    attn_mask = bert_model._build_attention_mask(valid_mask, dtype=torch.float32)
+
+    assert attn_mask.shape == (2, 1, seq_len, seq_len)
+
+    # Batch 0: keys 0 and 1 are padding.
+    assert torch.all(attn_mask[0, :, :, 0] < -1e20)
+    assert torch.all(attn_mask[0, :, :, 1] < -1e20)
+
+    # Batch 0: keys 2 onward are not padding-masked.
+    assert torch.all(attn_mask[0, :, :, 2:] > -1e20)
+
+    # Batch 1: key 0 is padding.
+    assert torch.all(attn_mask[1, :, :, 0] < -1e20)
+    assert torch.all(attn_mask[1, :, :, 1:] > -1e20)
+
+
+def test_causal_and_padding_masks_are_combined(causal_model):
+    seq_len = causal_model.seq_length
+
+    valid_mask = torch.ones(
+        1,
+        seq_len,
+        dtype=torch.bool,
+        device=causal_model.src_mask.device,
+    )
+
+    valid_mask[0, 0] = False
+
+    attn_mask = causal_model._build_attention_mask(
+        valid_mask,
+        dtype=torch.float32,
+    )
+
+    assert attn_mask.shape == (1, 1, seq_len, seq_len)
+
+    # Padding key 0 is masked for every query.
+    assert torch.all(attn_mask[0, :, :, 0] < -1e20)
+
+    # Causal future positions are masked.
+    assert attn_mask[0, 0, 1, 2] < -1e20
+    assert attn_mask[0, 0, 1, seq_len - 1] < -1e20
+
+    # A past valid key should not be masked by causal masking.
+    assert attn_mask[0, 0, 1, 1] > -1e20
+    assert attn_mask[0, 0, 2, 1] > -1e20
+    assert attn_mask[0, 0, seq_len - 1, seq_len - 1] > -1e20
+
+
+@pytest.fixture
+def batch(model):
+    seq_len = model.seq_length
+
+    return {
+        "cat_col": torch.tensor(
+            [
+                [0, 0] + [1] * (seq_len - 2),
+                [0] + [2] * (seq_len - 1),
+            ],
+            dtype=torch.long,
+            device=model.src_mask.device,
+        ),
+        "real_col": torch.tensor(
+            [
+                [0.0, 0.0] + [0.5] * (seq_len - 2),
+                [0.0] + [1.5] * (seq_len - 1),
+            ],
+            dtype=torch.float32,
+            device=model.src_mask.device,
+        ),
+    }
+
+
+def test_forward_no_nan_with_padding(model, batch):
+    out = model.forward_train(batch)
+
+    for tensor in out.values():
+        assert torch.isfinite(tensor).all()

From f61f621b2a43e89676947c5f127e6a1507b1fca4 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 5 Jun 2026 18:51:12 +0200
Subject: [PATCH 09/81] Store pad length explicitly

---
 src/sequifier/helpers.py                      | 139 +++++++++++++++---
 src/sequifier/infer.py                        |  12 +-
 .../io/sequifier_dataset_from_file.py         |   8 +
 .../sequifier_dataset_from_folder_parquet.py  |  30 +++-
 ...uifier_dataset_from_folder_parquet_lazy.py |  18 ++-
 .../io/sequifier_dataset_from_folder_pt.py    |  37 ++++-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  28 +++-
 src/sequifier/preprocess.py                   |  41 ++++--
 src/sequifier/train.py                        |  65 +++++---
 .../hyperparameter-search-custom-eval.yaml    |   4 +-
 tests/integration/test_preprocessing.py       |  66 +++++++--
 ...uifier_dataset_from_folder_parquet_lazy.py |  64 ++++++++
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  49 ++++++
 tests/unit/test_helpers.py                    | 116 ++++++++++++++-
 tests/unit/test_preprocess.py                 |  94 +++++++++++-
 tests/unit/test_train.py                      |  43 ++++++
 16 files changed, 737 insertions(+), 77 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index fb6c2541..5fa7340b 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -3,6 +3,7 @@
 import random
 import re
 import sys
+import warnings
 from datetime import datetime
 from typing import Any, Dict, Optional, Union
 
@@ -39,6 +40,12 @@
     "uint8": torch.int16,
 }
 
+EXPLICIT_PADDING_MASK_FALLBACK_WARNING = (
+    "Explicit padding mask not found. Falling back to value-based padding "
+    "inference for real-valued data; leading 0.0 values may be treated as "
+    "padding. Re-run preprocessing to generate explicit masks."
+)
+
 
 # Check an environment variable to see if we are in a testing context
 IS_TESTING = os.environ.get("SEQUIFIER_TESTING", "0") == "1"
@@ -313,9 +320,99 @@ def numpy_to_pytorch(
         )
         unified_tensors[f"{col_name}_target"] = target_tensor
 
+    left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
+    if left_pad_lengths is not None:
+        full_length = seq_length + 1
+        unified_tensors["_attention_valid_mask"] = build_valid_mask(
+            left_pad_lengths, full_length, data_offset, seq_length
+        )
+        unified_tensors["_target_valid_mask"] = build_valid_mask(
+            left_pad_lengths, full_length, target_offset, seq_length
+        )
+
     return unified_tensors
 
 
+@beartype
+def build_valid_mask(
+    left_pad_lengths: Tensor,
+    full_length: int,
+    offset: int,
+    seq_length: int,
+) -> Tensor:
+    """Builds a boolean validity mask from explicit left-padding metadata."""
+
+    full_positions = torch.arange(
+        full_length, device=left_pad_lengths.device, dtype=left_pad_lengths.dtype
+    )
+    full_valid = full_positions[None, :] >= left_pad_lengths[:, None]
+
+    return full_valid[:, -(seq_length + offset) : (-offset if offset > 0 else None)]
+
+
+@beartype
+def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Optional[Tensor]:
+    """Extracts one leftPadLength value per long-format subsequence."""
+    if "leftPadLength" not in data.columns:
+        return None
+
+    assert {"sequenceId", "subsequenceId"}.issubset(data.columns)
+
+    lengths = (
+        data.group_by(["sequenceId", "subsequenceId"], maintain_order=True)
+        .agg(pl.col("leftPadLength").first().alias("leftPadLength"))
+        .sort(["sequenceId", "subsequenceId"])
+        .get_column("leftPadLength")
+    )
+
+    return torch.tensor(lengths.to_numpy(), dtype=torch.int64)
+
+
+@beartype
+def attach_padding_masks(
+    data_batch: dict[str, Tensor],
+    targets_batch: dict[str, Tensor],
+    left_pad_lengths: Tensor,
+    seq_length: int,
+    data_offset: int,
+    target_offset: int,
+) -> None:
+    """Attaches explicit attention and target masks to batch dictionaries."""
+    full_length = seq_length + 1
+    data_batch["_attention_valid_mask"] = build_valid_mask(
+        left_pad_lengths, full_length, data_offset, seq_length
+    )
+    targets_batch["_target_valid_mask"] = build_valid_mask(
+        left_pad_lengths, full_length, target_offset, seq_length
+    )
+
+
+@beartype
+def unpack_preprocessed_pt_tuple(
+    data: tuple[Any, ...],
+) -> tuple[Any, Any, Any, Any, Optional[Any]]:
+    """Unpacks both v1 four-item and v2 five-item preprocessed PT tuples."""
+    if len(data) == 4:
+        sequences_dict, sequence_ids, subsequence_ids, start_positions = data
+        return sequences_dict, sequence_ids, subsequence_ids, start_positions, None
+    if len(data) == 5:
+        (
+            sequences_dict,
+            sequence_ids,
+            subsequence_ids,
+            start_positions,
+            left_pad_lengths,
+        ) = data
+        return (
+            sequences_dict,
+            sequence_ids,
+            subsequence_ids,
+            start_positions,
+            left_pad_lengths,
+        )
+    raise ValueError(f"Expected a 4- or 5-item preprocessed PT tuple, got {len(data)}")
+
+
 @beartype
 def normalize_path(path: str, project_root: str) -> str:
     """Normalizes a path to be relative to a project path, then joins them.
@@ -519,6 +616,16 @@ def get_last_training_batch_timedelta(
     return (t2 - t1).total_seconds()
 
 
+def infer_valid_mask_from_data(data_batch: Dict[str, torch.Tensor], config: Any):
+    if len(config.categorical_columns):
+        ref_col = config.categorical_columns[0]
+        return data_batch[ref_col] != 0
+    else:
+        warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
+        ref_col = list(data_batch.keys())[0]
+        return (data_batch[ref_col] != 0.0).long().cumsum(dim=1) > 0
+
+
 def apply_bert_masking(
     data_batch: Dict[str, torch.Tensor],
     targets_batch: Dict[str, torch.Tensor],
@@ -530,16 +637,14 @@ def apply_bert_masking(
     """
     data_batch = {k: tensor.clone() for k, tensor in data_batch.items()}
     targets_batch = {k: tensor.clone().detach() for k, tensor in targets_batch.items()}
-    batch_size, seq_len = config.training_spec.batch_size, config.seq_length
 
     # 1. Identify valid tokens (Renamed from padding_mask to valid_mask for clarity)
-    if len(config.categorical_columns):
-        ref_col = config.categorical_columns[0]
-        valid_mask = data_batch[ref_col] != 0
+    if "_attention_valid_mask" in data_batch:
+        valid_mask = data_batch["_attention_valid_mask"].bool()
     else:
-        ref_col = list(data_batch.keys())[0]
-        valid_mask = (data_batch[ref_col] != 0.0).long().cumsum(dim=1) > 0
+        valid_mask = infer_valid_mask_from_data(data_batch, config)
 
+    batch_size, seq_len = valid_mask.shape
     device = valid_mask.device
 
     # Calculate exact number of tokens to mask per sequence based on valid length
@@ -560,7 +665,8 @@ def apply_bert_masking(
     # 3. Span Masking Loop
     for i in range(batch_size):
         budget = budgets[i].item()
-        valid_len = valid_mask[i].sum().item()
+        valid_positions = valid_mask[i].nonzero(as_tuple=True)[0]
+        valid_len = len(valid_positions)
 
         if budget < 1 or valid_len < 1:
             continue
@@ -578,18 +684,17 @@ def apply_bert_masking(
             start_idx = sampled_starts[span_idx]
 
             end_idx = min(start_idx + span_len, valid_len)
+            span_positions = valid_positions[start_idx:end_idx]
+            if len(span_positions) == 0:
+                span_idx += 1
+                continue
 
-            span_view = bert_mask[i, start_idx:end_idx]
-            newly_masked = (~span_view).sum().item()
-
-            # Truncate the span if adding it exceeds our exact masking budget
-            if current_masked + newly_masked > budget:
-                allowance = budget - current_masked
-                unmasked_indices = (~span_view).nonzero(as_tuple=True)[0]
-                if len(unmasked_indices) > allowance:
-                    end_idx = start_idx + unmasked_indices[allowance - 1].item() + 1
+            unmasked_positions = span_positions[~bert_mask[i, span_positions]]
+            allowance = budget - current_masked
+            if len(unmasked_positions) > allowance:
+                unmasked_positions = unmasked_positions[:allowance]
 
-            bert_mask[i, start_idx:end_idx] = True
+            bert_mask[i, unmasked_positions] = True
             current_masked = bert_mask[i].sum().item()
             span_idx += 1
 
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index ba94086a..5b625952 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -20,6 +20,7 @@
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
+    unpack_preprocessed_pt_tuple,
     write_data,
 )
 from sequifier.train import (
@@ -373,7 +374,8 @@ def infer_embedding(
                 sequence_ids_tensor,
                 subsequence_ids_tensor,
                 start_positions_tensor,
-            ) = data
+                _,
+            ) = unpack_preprocessed_pt_tuple(data)
             embeddings = get_embeddings_pt(config, inferer, sequences_dict)
 
             sequence_ids_for_preds = sequence_ids_tensor.numpy()
@@ -580,7 +582,13 @@ def infer_generative(
                     )
                 )
         elif config.read_format == "pt":
-            sequences_dict, sequence_ids_tensor, _, start_positions_tensor = data
+            (
+                sequences_dict,
+                sequence_ids_tensor,
+                _,
+                start_positions_tensor,
+                _,
+            ) = unpack_preprocessed_pt_tuple(data)
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index c2f0916c..d12d672e 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -61,6 +61,14 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.target_tensors = {
             key: all_tensors[f"{key}_target"] for key in self.config.target_columns
         }
+        if "_attention_valid_mask" in all_tensors:
+            self.sequence_tensors["_attention_valid_mask"] = all_tensors[
+                "_attention_valid_mask"
+            ]
+        if "_target_valid_mask" in all_tensors:
+            self.target_tensors["_target_valid_mask"] = all_tensors[
+                "_target_valid_mask"
+            ]
         del all_tensors
 
         if config.training_spec.device.startswith("cuda"):
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index d37e11ab..34d806a9 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -10,7 +10,12 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import PANDAS_TO_TORCH_TYPES, normalize_path
+from sequifier.helpers import (
+    PANDAS_TO_TORCH_TYPES,
+    attach_padding_masks,
+    get_left_pad_lengths_from_preprocessed_data,
+    normalize_path,
+)
 
 
 class SequifierDatasetFromFolderParquet(IterableDataset):
@@ -76,12 +81,17 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         all_targets: Dict[str, list[torch.Tensor]] = {
             col: [] for col in config.target_columns
         }
+        all_left_pad_lengths: list[torch.Tensor] = []
 
         # Step 1: Eager I/O reduction pass over all chunk allocations
         for file_info in metadata["batch_files"]:
             file_path = os.path.join(self.data_dir, file_info["path"])
             df = pl.read_parquet(file_path)
 
+            left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(df)
+            if left_pad_lengths is not None:
+                all_left_pad_lengths.append(left_pad_lengths)
+
             for col in all_sequences.keys():
                 feature_df = df.filter(pl.col("inputCol") == col)
                 if not feature_df.is_empty():
@@ -112,12 +122,20 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col, tensors in all_targets.items()
             if tensors
         }
+        self.left_pad_lengths = (
+            torch.cat(all_left_pad_lengths)
+            if all_left_pad_lengths
+            and len(all_left_pad_lengths) == len(metadata["batch_files"])
+            else None
+        )
 
         # Step 3: Prevent serialization duplications across worker forks via shared memory flags
         for tensor in self.sequences.values():
             tensor.share_memory_()
         for tensor in self.targets.values():
             tensor.share_memory_()
+        if self.left_pad_lengths is not None:
+            self.left_pad_lengths.share_memory_()
 
         self.target_samples = self._get_target_samples()
         self.total_batches = self._calculate_total_batches(self.target_samples)
@@ -210,4 +228,14 @@ def __iter__(
                 for key, tensor in self.targets.items()
             }
 
+            if self.left_pad_lengths is not None:
+                attach_padding_masks(
+                    data_batch,
+                    targets_batch,
+                    self.left_pad_lengths[batch_indices],
+                    train_seq_len,
+                    self.config.training_spec.data_offset,
+                    self.config.training_spec.target_offset,
+                )
+
             yield data_batch, targets_batch, None, None, None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index adda1a6e..d1b368de 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -11,7 +11,12 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import PANDAS_TO_TORCH_TYPES, normalize_path
+from sequifier.helpers import (
+    PANDAS_TO_TORCH_TYPES,
+    attach_padding_masks,
+    get_left_pad_lengths_from_preprocessed_data,
+    normalize_path,
+)
 
 
 class SequifierDatasetFromFolderParquetLazy(IterableDataset):
@@ -247,6 +252,7 @@ def __iter__(
             # This file overlaps with our worker's assigned boundary. Load it.
             file_path = os.path.join(self.data_dir, self.batch_files_info[f_id]["path"])
             df = pl.read_parquet(file_path)
+            left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(df)
 
             # Generate indices for the whole file using torch (matching pt_lazy)
             indices = torch.arange(file_samples)
@@ -309,6 +315,16 @@ def __iter__(
                         f"Missing required column {col_name} in Parquet partition"
                     )
 
+            if left_pad_lengths is not None:
+                attach_padding_masks(
+                    new_seq,
+                    new_tgt,
+                    left_pad_lengths[worker_indices],
+                    train_seq_len,
+                    self.config.training_spec.data_offset,
+                    self.config.training_spec.target_offset,
+                )
+
             del df
 
             # Append the new slice to the cross-file buffer
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 9c29d8b5..eceda4b8 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -9,7 +9,11 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import normalize_path
+from sequifier.helpers import (
+    attach_padding_masks,
+    normalize_path,
+    unpack_preprocessed_pt_tuple,
+)
 
 
 class SequifierDatasetFromFolderPt(IterableDataset):
@@ -49,22 +53,39 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         all_sequences: Dict[str, list[torch.Tensor]] = {
             col: [] for col in set(config.input_columns + config.target_columns)
         }
+        all_left_pad_lengths: list[torch.Tensor] = []
 
         # Load all data files into RAM
         for file_info in metadata["batch_files"]:
             file_path = os.path.join(self.data_dir, file_info["path"])
-            (sequences_batch, _, _, _) = torch.load(
-                file_path, map_location="cpu", weights_only=False
+            (
+                sequences_batch,
+                _,
+                _,
+                _,
+                left_pad_lengths_batch,
+            ) = unpack_preprocessed_pt_tuple(
+                torch.load(file_path, map_location="cpu", weights_only=False)
             )
             for col in all_sequences.keys():
                 if col in sequences_batch:
                     all_sequences[col].append(sequences_batch[col])
+            if left_pad_lengths_batch is not None:
+                all_left_pad_lengths.append(left_pad_lengths_batch)
 
         self.sequences: Dict[str, torch.Tensor] = {
             col: torch.cat(tensors) for col, tensors in all_sequences.items() if tensors
         }
+        self.left_pad_lengths = (
+            torch.cat(all_left_pad_lengths)
+            if all_left_pad_lengths
+            and len(all_left_pad_lengths) == len(metadata["batch_files"])
+            else None
+        )
         for tensor in self.sequences.values():
             tensor.share_memory_()
+        if self.left_pad_lengths is not None:
+            self.left_pad_lengths.share_memory_()
 
         self.target_samples = self._get_target_samples()
         self.total_batches = self._calculate_total_batches(self.target_samples)
@@ -172,4 +193,14 @@ def __iter__(
                 if key in self.config.target_columns
             }
 
+            if self.left_pad_lengths is not None:
+                attach_padding_masks(
+                    data_batch,
+                    targets_batch,
+                    self.left_pad_lengths[batch_indices],
+                    train_seq_len,
+                    data_offset,
+                    target_offset,
+                )
+
             yield data_batch, targets_batch, None, None, None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index e07924e0..eaac2694 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -10,7 +10,11 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import normalize_path
+from sequifier.helpers import (
+    attach_padding_masks,
+    normalize_path,
+    unpack_preprocessed_pt_tuple,
+)
 
 
 class SequifierDatasetFromFolderPtLazy(IterableDataset):
@@ -223,8 +227,14 @@ def __iter__(
 
             # This file overlaps with our worker's assigned boundary. Load it.
             file_path = os.path.join(self.data_dir, self.batch_files_info[f_id]["path"])
-            (sequences_batch, _, _, _) = torch.load(
-                file_path, map_location="cpu", weights_only=False
+            (
+                sequences_batch,
+                _,
+                _,
+                _,
+                left_pad_lengths_batch,
+            ) = unpack_preprocessed_pt_tuple(
+                torch.load(file_path, map_location="cpu", weights_only=False)
             )
 
             # Generate indices for the whole file
@@ -269,8 +279,18 @@ def __iter__(
                 if k in self.config.target_columns
             }
 
+            if left_pad_lengths_batch is not None:
+                attach_padding_masks(
+                    new_seq,
+                    new_tgt,
+                    left_pad_lengths_batch[worker_indices],
+                    train_seq_len,
+                    data_offset,
+                    target_offset,
+                )
+
             # Free the large file immediately to keep RAM down
-            del sequences_batch
+            del sequences_batch, left_pad_lengths_batch
 
             # Append the new slice to the cross-file buffer
             if buffer_len == 0:
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 54ca8f14..ae658b89 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -15,7 +15,12 @@
 from loguru import logger
 
 from sequifier.config.preprocess_config import load_preprocessor_config
-from sequifier.helpers import PANDAS_TO_TORCH_TYPES, read_data, write_data
+from sequifier.helpers import (
+    PANDAS_TO_TORCH_TYPES,
+    read_data,
+    unpack_preprocessed_pt_tuple,
+    write_data,
+)
 
 
 @beartype
@@ -340,6 +345,7 @@ def _create_schema(
             "sequenceId": pl.Int64,
             "subsequenceId": pl.Int64,
             "startItemPosition": pl.Int64,
+            "leftPadLength": pl.Int64,
             "inputCol": pl.String,
         }
 
@@ -835,7 +841,9 @@ def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> No
         for file_path in files:
             try:
                 if write_format == "pt":
-                    sequences_dict, _, _, _ = torch.load(file_path, weights_only=False)
+                    sequences_dict, _, _, _, _ = unpack_preprocessed_pt_tuple(
+                        torch.load(file_path, weights_only=False)
+                    )
                     if sequences_dict:
                         n_samples = sequences_dict[
                             list(sequences_dict.keys())[0]
@@ -1591,10 +1599,13 @@ def process_and_write_data_pt(
     aggs = [
         pl.concat_list(sequence_cols)
         .filter(pl.col("inputCol") == col_name)
-        .flatten()
+        .list.explode(keep_nulls=False, empty_as_null=False)  # flatten
         .alias(f"seq_{col_name}")
         for col_name in all_feature_cols
-    ] + [pl.col("startItemPosition").first().alias("startItemPosition")]
+    ] + [
+        pl.col("startItemPosition").first().alias("startItemPosition"),
+        pl.col("leftPadLength").first().alias("leftPadLength"),
+    ]
 
     aggregated_data = (
         data.group_by(["sequenceId", "subsequenceId"])
@@ -1614,6 +1625,9 @@ def process_and_write_data_pt(
     start_item_positions_tensor = torch.tensor(
         aggregated_data.get_column("startItemPosition").to_numpy(), dtype=torch.int64
     )
+    left_pad_lengths_tensor = torch.tensor(
+        aggregated_data.get_column("leftPadLength").to_numpy(), dtype=torch.int64
+    )
     sequences_dict = {}
 
     for col_name in all_feature_cols:
@@ -1634,6 +1648,7 @@ def process_and_write_data_pt(
         sequence_ids_tensor,
         subsequence_ids_tensor,
         start_item_positions_tensor,
+        left_pad_lengths_tensor,
     )
     torch.save(data_to_save, path)
 
@@ -1869,7 +1884,7 @@ def extract_sequences(
     for in_row in raw_sequences.iter_rows(named=True):
         in_seq_lists_only = {col: in_row[col] for col in columns}
 
-        subsequences = extract_subsequences(
+        subsequences, left_pad_lengths = extract_subsequences(
             in_seq_lists_only,
             seq_length,
             stride_for_split,
@@ -1883,11 +1898,12 @@ def extract_sequences(
                     in_row["sequenceId"],
                     subsequence_id,
                     subsequence_id * stride_for_split,
+                    left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
-                if len(row) != (seq_length + 5):
+                if len(row) != (seq_length + 6):
                     raise RuntimeError(
-                        f"Row length mismatch. Expected {seq_length + 5}, got {len(row)}. Row: {row}"
+                        f"Row length mismatch. Expected {seq_length + 6}, got {len(row)}. Row: {row}"
                     )
                 rows.append(row)
 
@@ -1956,7 +1972,7 @@ def extract_subsequences(
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
-) -> dict[str, list[list[Union[float, int]]]]:
+) -> tuple[dict[str, list[list[Union[float, int]]]], list[int]]:
     """Extracts subsequences from a dictionary of sequence lists.
 
     This function takes a dictionary `in_seq` where keys are column
@@ -1978,10 +1994,8 @@ def extract_subsequences(
         A dictionary mapping column names to a *list of lists*, where
         each inner list is a subsequence.
     """
-    if not in_seq[columns[0]]:
-        return {col: [] for col in columns}
-
     in_seq_len = len(in_seq[columns[0]])
+    pad_len = 0
     if in_seq_len < (seq_length + 1):
         pad_len = (seq_length + 1) - in_seq_len
         in_seq = {col: ([0] * pad_len) + in_seq[col] for col in columns}
@@ -1996,10 +2010,13 @@ def extract_subsequences(
             f"Diff of {subsequence_starts = }, {subsequence_starts_diff = } larger than {stride_for_split = }"
         )
 
-    return {
+    result = {
         col: [list(in_seq[col][i : i + seq_length + 1]) for i in subsequence_starts]
         for col in columns
     }
+    left_pad_lengths = [pad_len] * len(subsequence_starts)
+
+    return result, left_pad_lengths
 
 
 @beartype
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index bd1086cf..a38a96f1 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -52,9 +52,13 @@
 
 torch._dynamo.config.suppress_errors = True
 
+DATA_BATCH_METADATA_KEYS = {"_attention_valid_mask"}
+TARGET_BATCH_METADATA_KEYS = {"_target_valid_mask", "_bert_mask"}
+
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
 from sequifier.helpers import (  # noqa: E402
+    EXPLICIT_PADDING_MASK_FALLBACK_WARNING,
     apply_bert_masking,
     conditional_beartype,
     configure_determinism,
@@ -991,6 +995,7 @@ def _infer_attention_valid_mask(self, src: dict[str, Tensor]) -> Tensor:
 
         # Fallback for real-only data.
         # This matches the existing heuristic used in BERT masking.
+        warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
         ref_col = next(col for col in self.input_columns if col in src)
         return (src[ref_col] != 0.0).long().cumsum(dim=1) > 0
 
@@ -1108,6 +1113,12 @@ def forward_inner(self, src: dict[str, Tensor]) -> Tensor:
             src2 = self.joint_embedding_layer(src2)
 
         valid_mask = self._infer_attention_valid_mask(src).to(device=src2.device)
+        if valid_mask.shape != src2.shape[:2]:
+            raise ValueError(
+                f"Invalid attention mask shape: got {tuple(valid_mask.shape)}, "
+                f"expected {tuple(src2.shape[:2])} = (batch_size, seq_length). "
+                "Check _attention_valid_mask / leftPadLength construction."
+            )
         src2 = self._zero_padding_positions(src2, valid_mask)
 
         mask = self._build_attention_mask(valid_mask, dtype=src2.dtype)
@@ -1505,12 +1516,12 @@ def _train_epoch(
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
-                    if k in self.input_columns
+                    if k in self.input_columns or k in DATA_BATCH_METADATA_KEYS
                 }
                 targets = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in targets.items()
-                    if k in self.target_column_types
+                    if k in self.target_column_types or k in TARGET_BATCH_METADATA_KEYS
                 }
                 if self.hparams.training_spec.training_objective == "bert":
                     data, targets = apply_bert_masking(data, targets, self.hparams)
@@ -1681,27 +1692,37 @@ def _calculate_loss(
             - A dictionary of individual (unweighted) loss Tensors for each
               target column.
         """
-        mask_col = next(
-            (
-                col
-                for col in targets.keys()
-                if self.target_column_types[col] == "categorical"
-            ),
-            list(targets.keys())[0],
-        )
+        target_names = [
+            k for k in targets.keys() if k not in TARGET_BATCH_METADATA_KEYS
+        ]
+        if not target_names:
+            raise RuntimeError("Loss calculation failed; no target columns were found.")
 
-        if self.target_column_types[mask_col] == "real":
-            seq_mask_2d = (targets[mask_col] != 0.0).long().cumsum(dim=1) > 0
+        if "_target_valid_mask" in targets:
+            seq_mask_2d = targets["_target_valid_mask"].bool()
         else:
-            seq_mask_2d = targets[mask_col] != 0
+            mask_col = next(
+                (
+                    col
+                    for col in target_names
+                    if self.target_column_types[col] == "categorical"
+                ),
+                target_names[0],
+            )
+
+            if self.target_column_types[mask_col] == "real":
+                warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
+                seq_mask_2d = (targets[mask_col] != 0.0).long().cumsum(dim=1) > 0
+            else:
+                seq_mask_2d = targets[mask_col] != 0
 
         if "_bert_mask" in targets:
-            seq_mask_2d = seq_mask_2d & targets["_bert_mask"]
+            seq_mask_2d = seq_mask_2d & targets["_bert_mask"].bool()
 
         mask = seq_mask_2d.T.contiguous().reshape(-1)
 
         losses = {}
-        for target_column in [k for k in targets.keys() if k != "_bert_mask"]:
+        for target_column in target_names:
             target_column_type = self.target_column_types[target_column]
             if target_column_type == "categorical":
                 output[target_column] = (
@@ -1730,7 +1751,7 @@ def _calculate_loss(
             )
 
         loss = None
-        for target_column in targets.keys():
+        for target_column in target_names:
             losses[target_column] = losses[target_column] * (
                 self.loss_weights[target_column]
                 if self.loss_weights is not None
@@ -1833,12 +1854,12 @@ def _evaluate(
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
-                    if k in self.input_columns
+                    if k in self.input_columns or k in DATA_BATCH_METADATA_KEYS
                 }
                 targets = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in targets.items()
-                    if k in self.target_column_types
+                    if k in self.target_column_types or k in TARGET_BATCH_METADATA_KEYS
                 }
                 if self.hparams.training_spec.training_objective == "bert":
                     data, targets = apply_bert_masking(data, targets, self.hparams)
@@ -1924,12 +1945,12 @@ def _evaluate(
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
-                    if k in self.input_columns
+                    if k in self.input_columns or k in DATA_BATCH_METADATA_KEYS
                 }
                 targets = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in targets.items()
-                    if k in self.target_column_types
+                    if k in self.target_column_types or k in TARGET_BATCH_METADATA_KEYS
                 }
 
                 pseudo_output = {}
@@ -1940,6 +1961,10 @@ def _evaluate(
                             col, data[col].transpose(0, 1)
                         )
                         targets_for_baseline[col] = targets[col]
+                if "_target_valid_mask" in targets:
+                    targets_for_baseline["_target_valid_mask"] = targets[
+                        "_target_valid_mask"
+                    ]
 
                 if len(pseudo_output) > 0:
                     loss, losses = self._calculate_loss(
diff --git a/tests/configs/hyperparameter-search-custom-eval.yaml b/tests/configs/hyperparameter-search-custom-eval.yaml
index 88a16baf..be3403d4 100644
--- a/tests/configs/hyperparameter-search-custom-eval.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval.yaml
@@ -56,7 +56,7 @@ training_hyperparameter_sampling:
   epochs: [1, 1]
   save_interval_epochs: 10
   batch_size: [5, 10]
-  learning_rate: [0.001, 0.01]
+  learning_rate: [0.001, 0.0001]
   criterion:
     itemId: CrossEntropyLoss
     supCat1: CrossEntropyLoss
@@ -64,7 +64,7 @@ training_hyperparameter_sampling:
     supCat3: CrossEntropyLoss
     supCat4: CrossEntropyLoss
   accumulation_steps: [1]
-  dropout: [0.0]
+  dropout: [0.5]
   optimizer:
     - name: Adam
   scheduler:
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 79e13b1f..cdd42c52 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -75,11 +75,25 @@ def load_parquet_folder_outputs(path):
     other_cols = [
         col
         for col in data.columns
-        if col not in ["sequenceId", "subsequenceId", "startItemPosition", "inputCol"]
+        if col
+        not in [
+            "sequenceId",
+            "subsequenceId",
+            "startItemPosition",
+            "leftPadLength",
+            "inputCol",
+        ]
     ]
 
     return data[
-        ["sequenceId", "subsequenceId", "startItemPosition", "inputCol"] + other_cols
+        [
+            "sequenceId",
+            "subsequenceId",
+            "startItemPosition",
+            "leftPadLength",
+            "inputCol",
+        ]
+        + other_cols
     ].sort(["sequenceId", "subsequenceId", "startItemPosition", "inputCol"])
 
 
@@ -93,7 +107,13 @@ def load_pt_outputs(path):
                     sequence_id,
                     subsequence_id,
                     start_item_position,
+                    *maybe_left_pad_length,
                 ) = torch.load(os.path.join(root, file))
+                left_pad_length = (
+                    maybe_left_pad_length[0]
+                    if maybe_left_pad_length
+                    else torch.zeros_like(sequence_id)
+                )
                 sequences2 = {}
                 for col, vals in sequences.items():
                     vals2 = vals.numpy()
@@ -128,6 +148,12 @@ def load_pt_outputs(path):
                             start_item_position,
                         ]
                     )
+                    sequences2["leftPadLength"] = np.concatenate(
+                        [
+                            sequences2.get("leftPadLength", []),
+                            left_pad_length,
+                        ]
+                    )
 
                 content = pl.DataFrame(sequences2)
                 contents.append(content)
@@ -137,10 +163,24 @@ def load_pt_outputs(path):
     other_cols = [
         col
         for col in data.columns
-        if col not in ["sequenceId", "subsequenceId", "startItemPosition", "inputCol"]
+        if col
+        not in [
+            "sequenceId",
+            "subsequenceId",
+            "startItemPosition",
+            "leftPadLength",
+            "inputCol",
+        ]
     ]
     return data[
-        ["sequenceId", "subsequenceId", "startItemPosition", "inputCol"] + other_cols
+        [
+            "sequenceId",
+            "subsequenceId",
+            "startItemPosition",
+            "leftPadLength",
+            "inputCol",
+        ]
+        + other_cols
     ].sort(["sequenceId", "subsequenceId", "startItemPosition", "inputCol"])
 
 
@@ -178,7 +218,7 @@ def test_preprocessed_data_real(data_splits):
         assert len(data_splits[name]) == 2
 
         for i, data in enumerate(data_splits[name]):
-            number_expected_columns = 13
+            number_expected_columns = 14
             assert data.shape[1] == (
                 number_expected_columns
             ), f"{name = } - {i = }: {data.shape = } - {data.columns = }"
@@ -195,7 +235,7 @@ def test_preprocessed_data_categorical(data_splits):
         assert len(data_splits[name]) == 3
 
         for i, data in enumerate(data_splits[name]):
-            number_expected_columns = 13
+            number_expected_columns = 14
             assert data.shape[1] == (
                 number_expected_columns
             ), f"{name = } - {i = }: {data.shape = } - {data.columns = }"
@@ -268,9 +308,9 @@ def test_preprocessed_data_exact(run_preprocessing):
     pt_output = load_pt_outputs(pt_out_path)
 
     assert np.all(
-        parquet_output.to_numpy()[:, [0, 1, 2, 4, 5, 6, 7, 8, 9]]
-        == pt_output.to_numpy()[:, [0, 1, 2, 4, 5, 6, 7, 8, 9]].astype(int)
-    ), f"{np.sum(parquet_output.to_numpy()[:,[0,1,2,4,5,6,7,8,9]] == pt_output.to_numpy()[:,[0,1,2,4,5,6,7,8,9]].astype(int)) = }"
+        parquet_output.to_numpy()[:, [0, 1, 2, 3, 5, 6, 7, 8, 9]]
+        == pt_output.to_numpy()[:, [0, 1, 2, 3, 5, 6, 7, 8, 9]].astype(int)
+    ), f"{np.sum(parquet_output.to_numpy()[:,[0,1,2,3,5,6,7,8,9]] == pt_output.to_numpy()[:,[0,1,2,3,5,6,7,8,9]].astype(int)) = }"
 
     assert np.all(
         parquet_output["sequenceId"].to_numpy() == np.repeat(np.arange(10), 9)
@@ -381,7 +421,13 @@ def test_preprocessing_from_precomputed_stats(
         ].to_numpy()
     )
 
-    metadata_cols = ["sequenceId", "subsequenceId", "startItemPosition", "inputCol"]
+    metadata_cols = [
+        "sequenceId",
+        "subsequenceId",
+        "startItemPosition",
+        "leftPadLength",
+        "inputCol",
+    ]
 
     assert (
         np.mean(
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index a47bbfd4..e32fe430 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -92,6 +92,70 @@ def test_iteration_yields_correct_batches(mock_config, dataset_path):
     assert tgt_batch["item"].shape == (5, 2)
 
 
+def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
+    data_dir = tmp_path / "parquet_masks"
+    data_dir.mkdir()
+
+    schema = {
+        "sequenceId": pl.Int64,
+        "subsequenceId": pl.Int64,
+        "startItemPosition": pl.Int64,
+        "leftPadLength": pl.Int64,
+        "inputCol": pl.String,
+        "2": pl.Float64,
+        "1": pl.Float64,
+        "0": pl.Float64,
+    }
+    rows = [
+        (0, 0, 0, 0, "item", 0.0, 1.0, 2.0),
+        (1, 0, 0, 1, "item", 0.0, 0.0, 2.0),
+        (2, 0, 0, 2, "item", 0.0, 0.0, 2.0),
+        (3, 0, 0, 3, "item", 0.0, 0.0, 2.0),
+        (4, 0, 0, 0, "item", 0.0, 1.0, 2.0),
+    ]
+    pl.DataFrame(rows, schema=schema, orient="row").write_parquet(
+        data_dir / "file.parquet"
+    )
+    with open(data_dir / "metadata.json", "w") as f:
+        json.dump(
+            {
+                "total_samples": 5,
+                "batch_files": [{"path": "file.parquet", "samples": 5}],
+            },
+            f,
+        )
+
+    dataset = SequifierDatasetFromFolderParquetLazy(
+        str(data_dir), mock_config, shuffle=False
+    )
+    seq_batch, tgt_batch, _, _, _ = next(iter(dataset))
+
+    assert torch.equal(
+        seq_batch["_attention_valid_mask"],
+        torch.tensor(
+            [
+                [True, True],
+                [False, True],
+                [False, False],
+                [False, False],
+                [True, True],
+            ]
+        ),
+    )
+    assert torch.equal(
+        tgt_batch["_target_valid_mask"],
+        torch.tensor(
+            [
+                [True, True],
+                [True, True],
+                [False, True],
+                [False, False],
+                [True, True],
+            ]
+        ),
+    )
+
+
 @patch("torch.distributed.is_initialized", return_value=True)
 @patch("torch.distributed.get_rank", return_value=0)
 @patch("torch.distributed.get_world_size", return_value=2)
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 29cd1af6..733b3301 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -108,6 +108,55 @@ def test_iteration_yields_correct_batches(mock_config, dataset_path, mock_torch_
     assert tgt_dict["tgt1"].shape == (5, 5)
 
 
+def test_iteration_attaches_explicit_padding_masks(mock_config, dataset_path):
+    with patch("torch.load") as mock_load:
+
+        def side_effect(path, map_location, weights_only):
+            dummy_seq = {
+                "col1": torch.ones((10, 6)),
+                "tgt1": torch.ones((10, 6)),
+            }
+            left_pad_lengths = torch.tensor([0, 1, 2, 3, 4, 5, 0, 0, 0, 0])
+            return (
+                dummy_seq,
+                torch.arange(10),
+                torch.zeros(10, dtype=torch.int64),
+                torch.zeros(10, dtype=torch.int64),
+                left_pad_lengths,
+            )
+
+        mock_load.side_effect = side_effect
+        dataset = SequifierDatasetFromFolderPtLazy(
+            dataset_path, mock_config, shuffle=False
+        )
+        seq_dict, tgt_dict, _, _, _ = next(iter(dataset))
+
+    assert torch.equal(
+        seq_dict["_attention_valid_mask"],
+        torch.tensor(
+            [
+                [True, True, True, True, True],
+                [False, True, True, True, True],
+                [False, False, True, True, True],
+                [False, False, False, True, True],
+                [False, False, False, False, True],
+            ]
+        ),
+    )
+    assert torch.equal(
+        tgt_dict["_target_valid_mask"],
+        torch.tensor(
+            [
+                [True, True, True, True, True],
+                [True, True, True, True, True],
+                [False, True, True, True, True],
+                [False, False, True, True, True],
+                [False, False, False, True, True],
+            ]
+        ),
+    )
+
+
 @patch("torch.distributed.is_initialized", return_value=True)
 @patch("torch.distributed.get_world_size", return_value=2)
 @patch("torch.distributed.get_rank", return_value=0)
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index cae5c87f..bd06c425 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -1,7 +1,14 @@
+from types import SimpleNamespace
+
 import polars as pl
 import torch
 
-from sequifier.helpers import construct_index_maps, numpy_to_pytorch
+from sequifier.helpers import (
+    apply_bert_masking,
+    build_valid_mask,
+    construct_index_maps,
+    numpy_to_pytorch,
+)
 
 # ==========================================
 # 1. Test Index Map Construction
@@ -132,3 +139,110 @@ def test_numpy_to_pytorch_dtypes():
         target_offset=0,
     )
     assert tensors_float["float_col"].dtype == torch.float32
+
+
+def test_build_valid_mask_from_left_pad_lengths():
+    left_pad_lengths = torch.tensor([0, 2, 5], dtype=torch.int64)
+
+    input_mask = build_valid_mask(
+        left_pad_lengths, full_length=6, offset=1, seq_length=5
+    )
+    target_mask = build_valid_mask(
+        left_pad_lengths, full_length=6, offset=0, seq_length=5
+    )
+
+    assert torch.equal(
+        input_mask,
+        torch.tensor(
+            [
+                [True, True, True, True, True],
+                [False, False, True, True, True],
+                [False, False, False, False, False],
+            ]
+        ),
+    )
+    assert torch.equal(
+        target_mask,
+        torch.tensor(
+            [
+                [True, True, True, True, True],
+                [False, True, True, True, True],
+                [False, False, False, False, True],
+            ]
+        ),
+    )
+
+
+def test_numpy_to_pytorch_includes_explicit_padding_masks():
+    data = pl.DataFrame(
+        {
+            "sequenceId": [1, 2],
+            "subsequenceId": [0, 0],
+            "startItemPosition": [0, 0],
+            "leftPadLength": [0, 2],
+            "inputCol": ["A", "A"],
+            "3": [0.0, 0.0],
+            "2": [0.0, 0.0],
+            "1": [1.0, 1.0],
+            "0": [2.0, 2.0],
+        }
+    )
+
+    tensors = numpy_to_pytorch(
+        data,
+        {"A": torch.float32},
+        ["A"],
+        seq_length=3,
+        data_offset=1,
+        target_offset=0,
+    )
+
+    assert torch.equal(
+        tensors["_attention_valid_mask"],
+        torch.tensor([[True, True, True], [False, False, True]]),
+    )
+    assert torch.equal(
+        tensors["_target_valid_mask"],
+        torch.tensor([[True, True, True], [False, True, True]]),
+    )
+
+
+class _OnesSpanMasking:
+    def sample(self, shape, device):
+        return torch.ones(shape, dtype=torch.long, device=device)
+
+
+def test_apply_bert_masking_uses_explicit_valid_mask_for_zero_values():
+    config = SimpleNamespace(
+        categorical_columns=[],
+        real_columns=["real_col"],
+        n_classes={},
+        seq_length=4,
+        training_spec=SimpleNamespace(
+            batch_size=1,
+            bert_spec=SimpleNamespace(
+                masking_probability=1.0,
+                span_masking=_OnesSpanMasking(),
+                replacement_distribution=SimpleNamespace(
+                    masked=1.0,
+                    random=0.0,
+                    identical=0.0,
+                ),
+            ),
+        ),
+    )
+    data_batch = {
+        "real_col": torch.tensor([[99.0, 0.0, 1.0, 2.0]]),
+        "_attention_valid_mask": torch.tensor([[False, True, True, True]]),
+    }
+    targets_batch = {
+        "real_col": torch.tensor([[99.0, 0.0, 1.0, 2.0]]),
+        "_target_valid_mask": torch.tensor([[False, True, True, True]]),
+    }
+
+    _, masked_targets = apply_bert_masking(data_batch, targets_batch, config)
+
+    assert torch.equal(
+        masked_targets["_bert_mask"],
+        torch.tensor([[False, True, True, True]]),
+    )
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 359919df..95b7e54c 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -1,13 +1,16 @@
 import numpy as np
 import polars as pl
 import pytest
+import torch
 
 from sequifier.preprocess import (
     _get_column_statistics,
     create_id_map,
+    extract_sequences,
     extract_subsequences,
     get_batch_limits,
     get_combined_statistics,
+    process_and_write_data_pt,
 )
 
 # ==========================================
@@ -25,7 +28,7 @@ def test_extract_subsequences_basic():
     # Expected behavior: Window size is seq_length + 1 (history + target)
     # Windows: [10,11,12,13], [11,12,13,14], [12,13,14,15]
 
-    result = extract_subsequences(
+    result, left_pad_lengths = extract_subsequences(
         input_data, seq_length, stride, columns, subsequence_start_mode="distribute"
     )
 
@@ -43,7 +46,7 @@ def test_extract_subsequences_padding():
 
     # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
 
-    result = extract_subsequences(
+    result, left_pad_lengths = extract_subsequences(
         input_data, seq_length, stride, columns, subsequence_start_mode="distribute"
     )
 
@@ -51,6 +54,85 @@ def test_extract_subsequences_padding():
     assert result["col1"][0] == [0, 0, 0, 1, 2]
 
 
+def test_extract_subsequences_returns_left_pad_lengths_when_requested():
+    input_data = {"col1": [0.0, 1.5]}
+    result, left_pad_lengths = extract_subsequences(
+        input_data,
+        seq_length=4,
+        stride_for_split=1,
+        columns=["col1"],
+        subsequence_start_mode="distribute",
+    )
+
+    assert result["col1"][0] == [0, 0, 0, 0.0, 1.5]
+    assert left_pad_lengths == [3]
+
+
+def test_extract_sequences_persists_left_pad_length_metadata():
+    schema = {
+        "sequenceId": pl.Int64,
+        "subsequenceId": pl.Int64,
+        "startItemPosition": pl.Int64,
+        "leftPadLength": pl.Int64,
+        "inputCol": pl.String,
+        "4": pl.Float64,
+        "3": pl.Float64,
+        "2": pl.Float64,
+        "1": pl.Float64,
+        "0": pl.Float64,
+    }
+    data = pl.DataFrame(
+        {
+            "sequenceId": [10, 10],
+            "itemPosition": [0, 1],
+            "col1": [0.0, 1.5],
+        }
+    )
+
+    sequences = extract_sequences(
+        data,
+        schema,
+        seq_length=4,
+        stride_for_split=1,
+        columns=["col1"],
+        subsequence_start_mode="distribute",
+    )
+
+    assert sequences.get_column("leftPadLength").to_list() == [3]
+
+
+def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
+    data = pl.DataFrame(
+        {
+            "sequenceId": [1, 2],
+            "subsequenceId": [0, 0],
+            "startItemPosition": [0, 0],
+            "leftPadLength": [0, 2],
+            "inputCol": ["col1", "col1"],
+            "3": [1.0, 0.0],
+            "2": [2.0, 0.0],
+            "1": [3.0, 1.0],
+            "0": [4.0, 2.0],
+        }
+    )
+    out_path = tmp_path / "batch.pt"
+
+    process_and_write_data_pt(
+        data,
+        seq_length=3,
+        path=str(out_path),
+        column_types={"col1": "Float64"},
+    )
+
+    sequences, _, _, _, left_pad_lengths = torch.load(out_path, weights_only=False)
+
+    assert torch.equal(left_pad_lengths, torch.tensor([0, 2]))
+    assert torch.equal(
+        sequences["col1"],
+        torch.tensor([[1.0, 2.0, 3.0, 4.0], [0.0, 0.0, 1.0, 2.0]]),
+    )
+
+
 @pytest.mark.parametrize("mode", ["distribute", "exact"])
 def test_extract_subsequences_modes(mode):
     """Tests logic for 'distribute' vs 'exact' modes."""
@@ -64,7 +146,9 @@ def test_extract_subsequences_modes(mode):
     if mode == "distribute":
         stride = 4
         # distribute might adjust indices to maximize coverage
-        result = extract_subsequences(input_data, seq_length, stride, columns, mode)
+        result, left_pad_lengths = extract_subsequences(
+            input_data, seq_length, stride, columns, mode
+        )
         assert len(result["col1"]) > 0
 
     elif mode == "exact":
@@ -78,7 +162,9 @@ def test_extract_subsequences_modes(mode):
         # Testing a passing exact case
         # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
         input_data_exact = {"col1": list(range(11))}
-        result = extract_subsequences(input_data_exact, seq_length, 4, columns, mode)
+        result, left_pad_lengths = extract_subsequences(
+            input_data_exact, seq_length, 4, columns, mode
+        )
         assert len(result["col1"]) > 0
 
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index abae3f67..0b0bdd0b 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -202,6 +202,49 @@ def test_calculate_loss(model, model_config):
     assert "real_col" in component_losses
 
 
+def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_column_types = {"real_col": "real"}
+    model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
+    model.loss_weights = None
+
+    outputs = {
+        "real_col": torch.tensor(
+            [
+                [[1.0]],
+                [[2.0]],
+                [[1.0]],
+            ]
+        )
+    }
+    targets = {
+        "real_col": torch.tensor([[0.0, 0.0, 2.0]]),
+        "_target_valid_mask": torch.tensor([[True, True, True]]),
+    }
+
+    total_loss, component_losses = TransformerModel._calculate_loss(
+        model, outputs, targets
+    )
+
+    assert torch.isclose(total_loss, torch.tensor(2.0))
+    assert torch.isclose(component_losses["real_col"], torch.tensor(2.0))
+
+
+def test_infer_attention_valid_mask_prefers_explicit_mask_for_real_zero_inputs():
+    model = TransformerModel.__new__(TransformerModel)
+    model.categorical_columns = []
+    model.input_columns = ["real_col"]
+
+    src = {
+        "real_col": torch.tensor([[0.0, 0.0, 1.0]]),
+        "_attention_valid_mask": torch.tensor([[True, True, True]]),
+    }
+
+    mask = TransformerModel._infer_attention_valid_mask(model, src)
+
+    assert torch.equal(mask, torch.tensor([[True, True, True]]))
+
+
 def test_padding_keys_are_masked(bert_model):
     seq_len = bert_model.seq_length
 

From 9624e8c58f4045ecc526f9d84bd8a95a10a34488 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 5 Jun 2026 19:37:57 +0200
Subject: [PATCH 10/81] Move to metadata dict

---
 src/sequifier/helpers.py                      |  67 +++----
 src/sequifier/infer.py                        | 120 +++++++++++--
 .../io/sequifier_dataset_from_file.py         |  31 ++--
 .../sequifier_dataset_from_folder_parquet.py  |  17 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  36 +++-
 .../io/sequifier_dataset_from_folder_pt.py    |  17 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  36 +++-
 src/sequifier/train.py                        | 164 +++++++++++-------
 ...uifier_dataset_from_folder_parquet_lazy.py |   6 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |   6 +-
 tests/unit/test_helpers.py                    |  25 +--
 tests/unit/test_train.py                      |  12 +-
 12 files changed, 367 insertions(+), 170 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 5fa7340b..af839d0e 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -257,7 +257,7 @@ def numpy_to_pytorch(
     seq_length: int,
     data_offset: int,
     target_offset: int,
-) -> dict[str, Tensor]:
+) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
     """Converts a long-format Polars DataFrame to a dict of sequence tensors.
 
     This function assumes the input DataFrame `data` is in a long format
@@ -286,9 +286,11 @@ def numpy_to_pytorch(
             column names for time steps (e.g., "0" to "L").
 
     Returns:
-        A dictionary mapping feature names to their corresponding PyTorch
-        tensors. Target tensors are stored with a `_target` suffix
-        (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
+        A tuple of:
+        - a dictionary mapping feature names to their corresponding PyTorch
+          tensors. Target tensors are stored with a `_target` suffix
+          (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
+        - a metadata dictionary containing any explicit masks.
     """
     input_seq_cols = [
         str(c) for c in range(seq_length - 1 + data_offset, (-1 + data_offset), -1)
@@ -322,15 +324,13 @@ def numpy_to_pytorch(
 
     left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
     if left_pad_lengths is not None:
-        full_length = seq_length + 1
-        unified_tensors["_attention_valid_mask"] = build_valid_mask(
-            left_pad_lengths, full_length, data_offset, seq_length
-        )
-        unified_tensors["_target_valid_mask"] = build_valid_mask(
-            left_pad_lengths, full_length, target_offset, seq_length
+        metadata = generate_padding_masks(
+            left_pad_lengths, seq_length, data_offset, target_offset
         )
+    else:
+        metadata = {}
 
-    return unified_tensors
+    return unified_tensors, metadata
 
 
 @beartype
@@ -369,22 +369,22 @@ def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Optional[
 
 
 @beartype
-def attach_padding_masks(
-    data_batch: dict[str, Tensor],
-    targets_batch: dict[str, Tensor],
+def generate_padding_masks(
     left_pad_lengths: Tensor,
     seq_length: int,
     data_offset: int,
     target_offset: int,
-) -> None:
-    """Attaches explicit attention and target masks to batch dictionaries."""
+) -> dict[str, Tensor]:
+    """Generates explicit attention and target masks as a metadata dictionary."""
     full_length = seq_length + 1
-    data_batch["_attention_valid_mask"] = build_valid_mask(
-        left_pad_lengths, full_length, data_offset, seq_length
-    )
-    targets_batch["_target_valid_mask"] = build_valid_mask(
-        left_pad_lengths, full_length, target_offset, seq_length
-    )
+    return {
+        "attention_valid_mask": build_valid_mask(
+            left_pad_lengths, full_length, data_offset, seq_length
+        ),
+        "target_valid_mask": build_valid_mask(
+            left_pad_lengths, full_length, target_offset, seq_length
+        ),
+    }
 
 
 @beartype
@@ -629,18 +629,24 @@ def infer_valid_mask_from_data(data_batch: Dict[str, torch.Tensor], config: Any)
 def apply_bert_masking(
     data_batch: Dict[str, torch.Tensor],
     targets_batch: Dict[str, torch.Tensor],
+    metadata_batch: Optional[Dict[str, torch.Tensor]],
     config: Any,  # TrainConfig
-) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
+) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
     """
     Applies BERT-style span corruption to the input data using custom distributions.
-    Explicitly passes the boolean prediction mask via the targets dictionary.
+    Explicitly passes the boolean prediction mask via the metadata dictionary.
     """
     data_batch = {k: tensor.clone() for k, tensor in data_batch.items()}
     targets_batch = {k: tensor.clone().detach() for k, tensor in targets_batch.items()}
+    metadata_batch = (
+        {k: tensor.clone().detach() for k, tensor in metadata_batch.items()}
+        if metadata_batch
+        else {}
+    )
 
     # 1. Identify valid tokens (Renamed from padding_mask to valid_mask for clarity)
-    if "_attention_valid_mask" in data_batch:
-        valid_mask = data_batch["_attention_valid_mask"].bool()
+    if "attention_valid_mask" in metadata_batch:
+        valid_mask = metadata_batch["attention_valid_mask"].bool()
     else:
         valid_mask = infer_valid_mask_from_data(data_batch, config)
 
@@ -744,9 +750,8 @@ def apply_bert_masking(
             tensor[mask_token_mask] = mask_val
             tensor[random_token_mask] = random_noise[random_token_mask]
 
-    # 6. Append the explicit prediction mask
-    targets_batch["_bert_mask"] = bert_mask
-
-    data_batch["_attention_valid_mask"] = valid_mask.detach()
+    # 6. Append explicit prediction and attention masks to metadata.
+    metadata_batch["bert_mask"] = bert_mask
+    metadata_batch["attention_valid_mask"] = valid_mask.detach()
 
-    return data_batch, targets_batch
+    return data_batch, targets_batch, metadata_batch
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 5b625952..1777cc4a 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -17,6 +17,7 @@
     PANDAS_TO_TORCH_TYPES,
     configure_determinism,
     construct_index_maps,
+    generate_padding_masks,
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
@@ -374,9 +375,20 @@ def infer_embedding(
                 sequence_ids_tensor,
                 subsequence_ids_tensor,
                 start_positions_tensor,
-                _,
+                left_pad_lengths_tensor,
             ) = unpack_preprocessed_pt_tuple(data)
-            embeddings = get_embeddings_pt(config, inferer, sequences_dict)
+            metadata = {}
+            if left_pad_lengths_tensor is not None:
+                target_offset = 0 if config.training_objective == "causal" else 1
+                metadata = generate_padding_masks(
+                    left_pad_lengths_tensor,
+                    config.seq_length,
+                    data_offset=1,
+                    target_offset=target_offset,
+                )
+            embeddings = get_embeddings_pt(
+                config, inferer, sequences_dict, metadata=metadata
+            )
 
             sequence_ids_for_preds = sequence_ids_tensor.numpy()
             subsequence_ids_for_preds = subsequence_ids_tensor.numpy()
@@ -587,7 +599,7 @@ def infer_generative(
                 sequence_ids_tensor,
                 _,
                 start_positions_tensor,
-                _,
+                left_pad_lengths_tensor,
             ) = unpack_preprocessed_pt_tuple(data)
             total_steps = (
                 1
@@ -601,8 +613,18 @@ def infer_generative(
                 if key in config.input_columns
             }
 
+            metadata = {}
+            if left_pad_lengths_tensor is not None:
+                target_offset = 0 if config.training_objective == "causal" else 1
+                metadata = generate_padding_masks(
+                    left_pad_lengths_tensor,
+                    config.seq_length,
+                    data_offset=1,
+                    target_offset=target_offset,
+                )
+
             probs, preds = get_probs_preds_from_dict(
-                config, inferer, sequences_dict, total_steps
+                config, inferer, sequences_dict, total_steps, metadata=metadata
             )
 
             prediction_length = inferer.prediction_length  # Get prediction_length
@@ -740,6 +762,7 @@ def get_embeddings_pt(
     config: Any,
     inferer: "Inferer",
     data: dict[str, torch.Tensor],
+    metadata: Optional[dict[str, torch.Tensor]] = None,
 ) -> np.ndarray:
     """Generates embeddings from a batch of PyTorch tensor data.
 
@@ -764,7 +787,10 @@ def get_embeddings_pt(
         for key, val in data.items()
         if key in config.input_columns
     }
-    embeddings = inferer.infer_embedding(X)
+    metadata_np = (
+        {key: val.numpy() for key, val in metadata.items()} if metadata else None
+    )
+    embeddings = inferer.infer_embedding(X, metadata=metadata_np)
     return embeddings
 
 
@@ -774,6 +800,7 @@ def get_probs_preds_from_dict(
     inferer: "Inferer",
     data: dict[str, torch.Tensor],
     total_steps: int = 1,
+    metadata: Optional[dict[str, torch.Tensor]] = None,
 ) -> tuple[Optional[dict[str, np.ndarray]], dict[str, np.ndarray]]:
     """Generates predictions from PyTorch tensor data, supporting autoregression.
 
@@ -816,18 +843,26 @@ def get_probs_preds_from_dict(
         for key, tensor in data.items()
         if key in config.input_columns
     }
+    metadata_np = (
+        {key: tensor.numpy() for key, tensor in metadata.items()} if metadata else None
+    )
     all_probs_list = {col: [] for col in target_cols}
     all_preds_list = {col: [] for col in target_cols}
 
     # 3. Autoregressive loop
+    metadata_for_step = metadata_np
     for i in range(total_steps):
         if config.output_probabilities:
-            probs_for_step = inferer.infer_generative(X, return_probs=True)
+            probs_for_step = inferer.infer_generative(
+                X, return_probs=True, metadata=metadata_for_step
+            )
             preds_for_step = inferer.infer_generative(None, probs_for_step)
             for col in target_cols:
                 all_probs_list[col].append(probs_for_step[col])
         else:
-            preds_for_step = inferer.infer_generative(X, return_probs=False)
+            preds_for_step = inferer.infer_generative(
+                X, return_probs=False, metadata=metadata_for_step
+            )
 
         for col in target_cols:
             all_preds_list[col].append(preds_for_step[col])
@@ -844,6 +879,17 @@ def get_probs_preds_from_dict(
             X_next[col] = np.concatenate([shifted_input, new_value], axis=1)
 
         X = X_next
+        if (
+            metadata_for_step is not None
+            and "attention_valid_mask" in metadata_for_step
+        ):
+            shifted_mask = metadata_for_step["attention_valid_mask"][:, 1:]
+            appended_valid = np.ones(
+                (shifted_mask.shape[0], 1), dtype=shifted_mask.dtype
+            )
+            metadata_for_step["attention_valid_mask"] = np.concatenate(
+                [shifted_mask, appended_valid], axis=1
+            )
 
     final_preds = {
         col: np.array(preds_list).T.reshape(-1, 1).flatten()
@@ -888,13 +934,18 @@ def get_embeddings(
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
     target_offset = 0 if config.training_objective == "causal" else 1
-    X = numpy_to_pytorch(
+    X, metadata = numpy_to_pytorch(
         data, column_types, all_columns, config.seq_length, 1, target_offset
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
+    metadata_np = (
+        {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
+        if metadata
+        else None
+    )
     del data
 
-    embeddings = inferer.infer_embedding(X)
+    embeddings = inferer.infer_embedding(X, metadata=metadata_np)
 
     return embeddings
 
@@ -932,18 +983,23 @@ def get_probs_preds_from_df(
 
     target_offset = 0 if config.training_objective == "causal" else 1
 
-    X = numpy_to_pytorch(
+    X, metadata = numpy_to_pytorch(
         data, column_types, all_columns, config.seq_length, 1, target_offset
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
+    metadata_np = (
+        {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
+        if metadata
+        else None
+    )
     del data
 
     if config.output_probabilities:
-        probs = inferer.infer_generative(X, return_probs=True)
+        probs = inferer.infer_generative(X, return_probs=True, metadata=metadata_np)
         preds = inferer.infer_generative(None, probs)
     else:
         probs = None
-        preds = inferer.infer_generative(X)
+        preds = inferer.infer_generative(X, metadata=metadata_np)
 
     return (probs, preds)
 
@@ -1051,7 +1107,7 @@ def get_probs_preds_autoregression(
         + seq_length
     )
 
-    head_data = numpy_to_pytorch(
+    head_data, metadata = numpy_to_pytorch(
         head_data_df,
         column_types,
         config.input_columns,
@@ -1062,7 +1118,11 @@ def get_probs_preds_autoregression(
 
     # Run the autoregressive PyTorch inference
     probs, preds = get_probs_preds_from_dict(
-        config, inferer, head_data, total_steps=config.autoregression_total_steps
+        config,
+        inferer,
+        head_data,
+        total_steps=config.autoregression_total_steps,
+        metadata=metadata,
     )
 
     # 4. Generate the final output arrays using the perfectly aligned bases
@@ -1223,6 +1283,7 @@ def invert_normalization(
     def infer_embedding(
         self,
         x: dict[str, np.ndarray],
+        metadata: Optional[dict[str, np.ndarray]] = None,
     ) -> np.ndarray:
         """Performs inference with an embedding model.
 
@@ -1239,7 +1300,7 @@ def infer_embedding(
         """
         assert x is not None
         size = x[list(x.keys())[0]].shape[0]
-        embedding = self.adjust_and_infer_embedding(x, size)
+        embedding = self.adjust_and_infer_embedding(x, size, metadata=metadata)
 
         return embedding
 
@@ -1249,6 +1310,7 @@ def infer_generative(
         x: Optional[dict[str, np.ndarray]],
         probs: Optional[dict[str, np.ndarray]] = None,
         return_probs: bool = False,
+        metadata: Optional[dict[str, np.ndarray]] = None,
     ) -> dict[str, np.ndarray]:
         """Performs generative inference, returning probabilities or predictions.
 
@@ -1294,7 +1356,7 @@ def infer_generative(
                     f"Not all keys in x are in probs - {x.keys() = } != {probs.keys() = }. Full inference is executed."
                 )
 
-            outs = self.adjust_and_infer_generative(x, size)
+            outs = self.adjust_and_infer_generative(x, size, metadata=metadata)
 
             for target_column, target_outs in outs.items():
                 if np.any(target_outs == np.inf):
@@ -1332,7 +1394,12 @@ def infer_generative(
         return outs
 
     @beartype
-    def adjust_and_infer_embedding(self, x: dict[str, np.ndarray], size: int):
+    def adjust_and_infer_embedding(
+        self,
+        x: dict[str, np.ndarray],
+        size: int,
+        metadata: Optional[dict[str, np.ndarray]] = None,
+    ):
         """Handles batching and backend-specific calls for embedding inference.
 
         This function prepares the input data `x` into batches using
@@ -1356,19 +1423,30 @@ def adjust_and_infer_embedding(self, x: dict[str, np.ndarray], size: int):
             embeddings = np.concatenate(inference_batch_embeddings, axis=0)[:size]
         elif self.inference_model_type == "pt":
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=False)
+            metadata_adjusted = (
+                self.prepare_inference_batches(metadata, pad_to_batch_size=False)
+                if metadata
+                else None
+            )
             embeddings = infer_with_embedding_model(
                 self.inference_model,
                 x_adjusted,
                 self.device,
                 size,
                 self.target_columns,
+                metadata=metadata_adjusted,
             )
         else:
             assert False, "not possible"
         return embeddings
 
     @beartype
-    def adjust_and_infer_generative(self, x: dict[str, np.ndarray], size: int):
+    def adjust_and_infer_generative(
+        self,
+        x: dict[str, np.ndarray],
+        size: int,
+        metadata: Optional[dict[str, np.ndarray]] = None,
+    ):
         """Handles batching and backend-specific calls for generative inference.
 
         This function prepares the input data `x` into batches using
@@ -1401,12 +1479,18 @@ def adjust_and_infer_generative(self, x: dict[str, np.ndarray], size: int):
         elif self.inference_model_type == "pt":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=False)
+            metadata_adjusted = (
+                self.prepare_inference_batches(metadata, pad_to_batch_size=False)
+                if metadata
+                else None
+            )
             outs = infer_with_generative_model(
                 self.inference_model,
                 x_adjusted,
                 self.device,
                 size * self.prediction_length,
                 self.target_columns,
+                metadata=metadata_adjusted,
             )
         else:
             assert False
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index d12d672e..ffc7d2aa 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -43,7 +43,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         target_offset = 0 if config.training_spec.training_objective == "causal" else 1
 
         # self.all_tensors now holds both inputs and targets
-        all_tensors = numpy_to_pytorch(
+        all_tensors, metadata_tensors = numpy_to_pytorch(
             data=data_df,
             column_types=column_types,
             all_columns=all_columns,
@@ -61,14 +61,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.target_tensors = {
             key: all_tensors[f"{key}_target"] for key in self.config.target_columns
         }
-        if "_attention_valid_mask" in all_tensors:
-            self.sequence_tensors["_attention_valid_mask"] = all_tensors[
-                "_attention_valid_mask"
-            ]
-        if "_target_valid_mask" in all_tensors:
-            self.target_tensors["_target_valid_mask"] = all_tensors[
-                "_target_valid_mask"
-            ]
+        self.metadata_tensors = metadata_tensors
         del all_tensors
 
         if config.training_spec.device.startswith("cuda"):
@@ -76,6 +69,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
                 self.sequence_tensors[key] = self.sequence_tensors[key].pin_memory()
             for key in self.target_tensors:
                 self.target_tensors[key] = self.target_tensors[key].pin_memory()
+            for key in self.metadata_tensors:
+                self.metadata_tensors[key] = self.metadata_tensors[key].pin_memory()
 
         logger.info(f"[INFO] Dataset loaded with {self.n_samples} samples.")
 
@@ -90,7 +85,13 @@ def __len__(self) -> int:
     def __iter__(
         self,
     ) -> Iterator[
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]
+        Tuple[
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            None,
+            None,
+        ]
     ]:
         """Yields batches of data.
 
@@ -98,13 +99,13 @@ def __iter__(
         rank and worker ID.
 
         Yields:
-            Iterator[Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]]:
+            Iterator[Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None]]:
             An iterator where each item is a tuple containing:
                 - data_batch (dict): Dictionary of feature tensors for the batch.
                 - targets_batch (dict): Dictionary of target tensors for the batch.
+                - metadata_batch (dict): Dictionary of metadata tensors for the batch.
                 - None: Placeholder for sequence_id (not used in this dataset type).
                 - None: Placeholder for subsequence_id (not used in this dataset type).
-                - None: Placeholder for start_position (not used in this dataset type).
         """
         worker_info = torch.utils.data.get_worker_info()
         world_size = dist.get_world_size() if dist.is_initialized() else 1
@@ -144,5 +145,9 @@ def __iter__(
                 key: tensor[batch_indices]
                 for key, tensor in self.target_tensors.items()
             }
+            metadata_batch = {
+                key: tensor[batch_indices]
+                for key, tensor in self.metadata_tensors.items()
+            }
 
-            yield data_batch, targets_batch, None, None, None
+            yield data_batch, targets_batch, metadata_batch, None, None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 34d806a9..99e58efc 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -12,7 +12,7 @@
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
-    attach_padding_masks,
+    generate_padding_masks,
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
 )
@@ -183,7 +183,13 @@ def __len__(self) -> int:
     def __iter__(
         self,
     ) -> Iterator[
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]
+        Tuple[
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            None,
+            None,
+        ]
     ]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
@@ -228,14 +234,13 @@ def __iter__(
                 for key, tensor in self.targets.items()
             }
 
+            metadata_batch = {}
             if self.left_pad_lengths is not None:
-                attach_padding_masks(
-                    data_batch,
-                    targets_batch,
+                metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
                     self.config.training_spec.data_offset,
                     self.config.training_spec.target_offset,
                 )
 
-            yield data_batch, targets_batch, None, None, None
+            yield data_batch, targets_batch, metadata_batch, None, None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index d1b368de..f4120cc3 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -13,7 +13,7 @@
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
-    attach_padding_masks,
+    generate_padding_masks,
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
 )
@@ -34,9 +34,9 @@ class SequifierDatasetFromFolderParquetLazy(IterableDataset):
             sample indices per epoch. Defaults to True.
 
     Yields:
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]:
+        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None]:
             A batch tuple containing sequence dictionaries, target dictionaries,
-            and three `None` placeholders (for API compatibility).
+            metadata dictionaries, and two `None` placeholders (for API compatibility).
 
     Raises:
         FileNotFoundError: If `metadata.json` is missing.
@@ -156,7 +156,13 @@ def __len__(self) -> int:
     def __iter__(
         self,
     ) -> Iterator[
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]
+        Tuple[
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            None,
+            None,
+        ]
     ]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
@@ -234,6 +240,7 @@ def __iter__(
         # Initialize cross-file buffers
         seq_buffer: Dict[str, torch.Tensor] = {}
         tgt_buffer: Dict[str, torch.Tensor] = {}
+        meta_buffer: Dict[str, torch.Tensor] = {}
         buffer_len = 0
 
         for f_id in extended_files:
@@ -315,10 +322,9 @@ def __iter__(
                         f"Missing required column {col_name} in Parquet partition"
                     )
 
+            new_meta = {}
             if left_pad_lengths is not None:
-                attach_padding_masks(
-                    new_seq,
-                    new_tgt,
+                new_meta = generate_padding_masks(
                     left_pad_lengths[worker_indices],
                     train_seq_len,
                     self.config.training_spec.data_offset,
@@ -331,6 +337,7 @@ def __iter__(
             if buffer_len == 0:
                 seq_buffer = new_seq
                 tgt_buffer = new_tgt
+                meta_buffer = new_meta
             else:
                 seq_buffer = {
                     k: torch.cat([seq_buffer[k], new_seq[k]], dim=0) for k in seq_buffer
@@ -338,6 +345,14 @@ def __iter__(
                 tgt_buffer = {
                     k: torch.cat([tgt_buffer[k], new_tgt[k]], dim=0) for k in tgt_buffer
                 }
+                if set(meta_buffer) != set(new_meta):
+                    raise RuntimeError(
+                        "Inconsistent leftPadLength metadata across Parquet chunks."
+                    )
+                meta_buffer = {
+                    k: torch.cat([meta_buffer[k], new_meta[k]], dim=0)
+                    for k in meta_buffer
+                }
 
             buffer_len += num_new_samples
 
@@ -349,13 +364,15 @@ def __iter__(
                 # Slice out a perfect batch from the top of the buffer
                 batch_seq = {k: v[: self.batch_size] for k, v in seq_buffer.items()}
                 batch_tgt = {k: v[: self.batch_size] for k, v in tgt_buffer.items()}
+                batch_meta = {k: v[: self.batch_size] for k, v in meta_buffer.items()}
 
-                yield batch_seq, batch_tgt, None, None, None
+                yield batch_seq, batch_tgt, batch_meta, None, None
                 yielded_samples += self.batch_size
 
                 # Keep the remainder in the buffer for the next loop/file
                 seq_buffer = {k: v[self.batch_size :] for k, v in seq_buffer.items()}
                 tgt_buffer = {k: v[self.batch_size :] for k, v in tgt_buffer.items()}
+                meta_buffer = {k: v[self.batch_size :] for k, v in meta_buffer.items()}
                 buffer_len -= self.batch_size
 
         # 6. Yield the final partial batch from the buffer if any remains
@@ -365,5 +382,6 @@ def __iter__(
 
             batch_seq = {k: v[:final_yield_size] for k, v in seq_buffer.items()}
             batch_tgt = {k: v[:final_yield_size] for k, v in tgt_buffer.items()}
+            batch_meta = {k: v[:final_yield_size] for k, v in meta_buffer.items()}
 
-            yield batch_seq, batch_tgt, None, None, None
+            yield batch_seq, batch_tgt, batch_meta, None, None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index eceda4b8..85a5d258 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -10,7 +10,7 @@
 
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
-    attach_padding_masks,
+    generate_padding_masks,
     normalize_path,
     unpack_preprocessed_pt_tuple,
 )
@@ -133,7 +133,13 @@ def __len__(self) -> int:
     def __iter__(
         self,
     ) -> Iterator[
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]
+        Tuple[
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            None,
+            None,
+        ]
     ]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
@@ -193,14 +199,13 @@ def __iter__(
                 if key in self.config.target_columns
             }
 
+            metadata_batch = {}
             if self.left_pad_lengths is not None:
-                attach_padding_masks(
-                    data_batch,
-                    targets_batch,
+                metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
                     data_offset,
                     target_offset,
                 )
 
-            yield data_batch, targets_batch, None, None, None
+            yield data_batch, targets_batch, metadata_batch, None, None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index eaac2694..41d5f115 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -11,7 +11,7 @@
 
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
-    attach_padding_masks,
+    generate_padding_masks,
     normalize_path,
     unpack_preprocessed_pt_tuple,
 )
@@ -32,9 +32,9 @@ class SequifierDatasetFromFolderPtLazy(IterableDataset):
             sample indices per epoch. Defaults to True.
 
     Yields:
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]:
+        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None]:
             A batch tuple containing sequence dictionaries, target dictionaries,
-            and three `None` placeholders (for API compatibility).
+            metadata dictionaries, and two `None` placeholders (for API compatibility).
 
     Raises:
         FileNotFoundError: If `metadata.json` is missing.
@@ -149,7 +149,13 @@ def __len__(self) -> int:
     def __iter__(
         self,
     ) -> Iterator[
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None, None]
+        Tuple[
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            Dict[str, torch.Tensor],
+            None,
+            None,
+        ]
     ]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
@@ -210,6 +216,7 @@ def __iter__(
         # Initialize cross-file buffers
         seq_buffer: Dict[str, torch.Tensor] = {}
         tgt_buffer: Dict[str, torch.Tensor] = {}
+        meta_buffer: Dict[str, torch.Tensor] = {}
         buffer_len = 0
 
         for f_id in extended_files:
@@ -279,10 +286,9 @@ def __iter__(
                 if k in self.config.target_columns
             }
 
+            new_meta = {}
             if left_pad_lengths_batch is not None:
-                attach_padding_masks(
-                    new_seq,
-                    new_tgt,
+                new_meta = generate_padding_masks(
                     left_pad_lengths_batch[worker_indices],
                     train_seq_len,
                     data_offset,
@@ -296,6 +302,7 @@ def __iter__(
             if buffer_len == 0:
                 seq_buffer = new_seq
                 tgt_buffer = new_tgt
+                meta_buffer = new_meta
             else:
                 seq_buffer = {
                     k: torch.cat([seq_buffer[k], new_seq[k]], dim=0) for k in seq_buffer
@@ -303,6 +310,14 @@ def __iter__(
                 tgt_buffer = {
                     k: torch.cat([tgt_buffer[k], new_tgt[k]], dim=0) for k in tgt_buffer
                 }
+                if set(meta_buffer) != set(new_meta):
+                    raise RuntimeError(
+                        "Inconsistent leftPadLength metadata across PT chunks."
+                    )
+                meta_buffer = {
+                    k: torch.cat([meta_buffer[k], new_meta[k]], dim=0)
+                    for k in meta_buffer
+                }
 
             buffer_len += num_new_samples
 
@@ -314,13 +329,15 @@ def __iter__(
                 # Slice out a perfect batch from the top of the buffer
                 batch_seq = {k: v[: self.batch_size] for k, v in seq_buffer.items()}
                 batch_tgt = {k: v[: self.batch_size] for k, v in tgt_buffer.items()}
+                batch_meta = {k: v[: self.batch_size] for k, v in meta_buffer.items()}
 
-                yield batch_seq, batch_tgt, None, None, None
+                yield batch_seq, batch_tgt, batch_meta, None, None
                 yielded_samples += self.batch_size
 
                 # Keep the remainder in the buffer for the next loop/file
                 seq_buffer = {k: v[self.batch_size :] for k, v in seq_buffer.items()}
                 tgt_buffer = {k: v[self.batch_size :] for k, v in tgt_buffer.items()}
+                meta_buffer = {k: v[self.batch_size :] for k, v in meta_buffer.items()}
                 buffer_len -= self.batch_size
 
         # 6. Yield the final partial batch from the buffer if any remains
@@ -330,5 +347,6 @@ def __iter__(
 
             batch_seq = {k: v[:final_yield_size] for k, v in seq_buffer.items()}
             batch_tgt = {k: v[:final_yield_size] for k, v in tgt_buffer.items()}
+            batch_meta = {k: v[:final_yield_size] for k, v in meta_buffer.items()}
 
-            yield batch_seq, batch_tgt, None, None, None
+            yield batch_seq, batch_tgt, batch_meta, None, None
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index a38a96f1..badba5cf 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -52,9 +52,6 @@
 
 torch._dynamo.config.suppress_errors = True
 
-DATA_BATCH_METADATA_KEYS = {"_attention_valid_mask"}
-TARGET_BATCH_METADATA_KEYS = {"_target_valid_mask", "_bert_mask"}
-
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
 from sequifier.helpers import (  # noqa: E402
@@ -522,7 +519,9 @@ def _copy_model(self):
         return model_copy
 
     @conditional_beartype
-    def forward(self, src: dict[str, Tensor]):
+    def forward(
+        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+    ):
         """Forward pass for the embedding model.
 
         Args:
@@ -531,7 +530,7 @@ def forward(self, src: dict[str, Tensor]):
         Returns:
             The embedded output.
         """
-        return self.transformer_model.forward_embed(src)
+        return self.transformer_model.forward_embed(src, metadata=metadata)
 
 
 class TransformerModel(nn.Module):
@@ -978,15 +977,17 @@ def _recursive_concat(self, srcs: list[Tensor]):
             return self._recursive_concat(srcs_inner)
 
     @conditional_beartype
-    def _infer_attention_valid_mask(self, src: dict[str, Tensor]) -> Tensor:
+    def _infer_attention_valid_mask(
+        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+    ) -> Tensor:
         """
         Returns a boolean mask of shape (batch_size, seq_length).
 
         True means valid/non-padding token.
         False means padding token.
         """
-        if "_attention_valid_mask" in src:
-            return src["_attention_valid_mask"].bool()
+        if metadata is not None and "attention_valid_mask" in metadata:
+            return metadata["attention_valid_mask"].bool()
 
         # Prefer categorical columns because padding is unambiguously encoded as 0.
         for col in self.categorical_columns:
@@ -1043,7 +1044,9 @@ def _zero_padding_positions(self, x: Tensor, valid_mask: Tensor) -> Tensor:
         return x * valid_mask[:, :, None].to(dtype=x.dtype)
 
     @conditional_beartype
-    def forward_inner(self, src: dict[str, Tensor]) -> Tensor:
+    def forward_inner(
+        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+    ) -> Tensor:
         """The inner forward pass of the model.
 
         This handles embedding lookup, positional encoding, and passing the
@@ -1112,12 +1115,14 @@ def forward_inner(self, src: dict[str, Tensor]) -> Tensor:
         if self.joint_embedding_layer is not None:
             src2 = self.joint_embedding_layer(src2)
 
-        valid_mask = self._infer_attention_valid_mask(src).to(device=src2.device)
+        valid_mask = self._infer_attention_valid_mask(src, metadata).to(
+            device=src2.device
+        )
         if valid_mask.shape != src2.shape[:2]:
             raise ValueError(
                 f"Invalid attention mask shape: got {tuple(valid_mask.shape)}, "
                 f"expected {tuple(src2.shape[:2])} = (batch_size, seq_length). "
-                "Check _attention_valid_mask / leftPadLength construction."
+                "Check attention_valid_mask / leftPadLength construction."
             )
         src2 = self._zero_padding_positions(src2, valid_mask)
 
@@ -1133,7 +1138,9 @@ def forward_inner(self, src: dict[str, Tensor]) -> Tensor:
         return src2.transpose(0, 1)
 
     @conditional_beartype
-    def forward_embed(self, src: dict[str, Tensor]) -> Tensor:
+    def forward_embed(
+        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+    ) -> Tensor:
         """Forward pass for the embedding model.
 
         This returns only the embedding from the *last* token in the sequence.
@@ -1146,10 +1153,12 @@ def forward_embed(self, src: dict[str, Tensor]) -> Tensor:
             The embedding tensor for the last token
             (batch_size, dim_model).
         """
-        return self.forward_inner(src)[-self.prediction_length :, :, :]
+        return self.forward_inner(src, metadata)[-self.prediction_length :, :, :]
 
     @conditional_beartype
-    def forward_train(self, src: dict[str, Tensor]) -> dict[str, Tensor]:
+    def forward_train(
+        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+    ) -> dict[str, Tensor]:
         """Forward pass for training.
 
         This runs the inner forward pass and then applies the appropriate
@@ -1163,7 +1172,7 @@ def forward_train(self, src: dict[str, Tensor]) -> dict[str, Tensor]:
             A dictionary mapping target column names to their raw output
             (logit) tensors (seq_length, batch_size, n_classes/1).
         """
-        output = self.forward_inner(src)
+        output = self.forward_inner(src, metadata)
         output = {
             target_column: self.decode(target_column, output)
             for target_column in self.target_columns
@@ -1213,7 +1222,10 @@ def apply_softmax(self, target_column: str, output: Tensor) -> Tensor:
 
     @conditional_beartype
     def forward(
-        self, src: dict[str, Tensor], return_logits: Union[bool, Tensor] = False
+        self,
+        src: dict[str, Tensor],
+        metadata: Optional[dict[str, Tensor]] = None,
+        return_logits: Union[bool, Tensor] = False,
     ) -> dict[str, Tensor]:
         """The main forward pass of the model.
 
@@ -1230,7 +1242,7 @@ def forward(
             output (LogSoftmax probabilities or real values) for the
             last token (batch_size, n_classes/1).
         """
-        output = self.forward_train(src)
+        output = self.forward_train(src, metadata)
         if return_logits:
             return output
         return {
@@ -1492,8 +1504,9 @@ def _train_epoch(
 
         Iterates through the training DataLoader, computes loss, performs
         backpropagation, and updates model parameters. The DataLoader is expected
-        to yield tuples of (sequences_dict, targets_dict, sequence_ids, subsequence_ids, start_positions).
-        The IDs and positions are currently unused in this training loop.
+        to yield tuples of
+        (sequences_dict, targets_dict, metadata_dict, sequence_ids, subsequence_ids).
+        The IDs are currently unused in this training loop.
 
         Args:
             train_loader: DataLoader for the training dataset.
@@ -1511,20 +1524,26 @@ def _train_epoch(
 
         model_to_call.train()
 
-        for batch_count, (data, targets, _, _, _) in enumerate(train_loader):
+        for batch_count, (data, targets, metadata, _, _) in enumerate(train_loader):
             if batch_count >= start_batch:
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
-                    if k in self.input_columns or k in DATA_BATCH_METADATA_KEYS
+                    if k in self.input_columns
                 }
                 targets = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in targets.items()
-                    if k in self.target_column_types or k in TARGET_BATCH_METADATA_KEYS
+                    if k in self.target_column_types
+                }
+                metadata = {
+                    k: v.to(self.device, non_blocking=True)
+                    for k, v in (metadata or {}).items()
                 }
                 if self.hparams.training_spec.training_objective == "bert":
-                    data, targets = apply_bert_masking(data, targets, self.hparams)
+                    data, targets, metadata = apply_bert_masking(
+                        data, targets, metadata, self.hparams
+                    )
 
                 # Only use standard torch.autocast if FSDP MixedPrecision is NOT handling it natively
                 if (
@@ -1541,11 +1560,13 @@ def _train_epoch(
                     with torch.autocast(
                         device_type=self.device.split(":")[0], dtype=amp_dtype
                     ):
-                        output = model_to_call(data, True)
-                        loss, losses = self._calculate_loss(output, targets)
+                        output = model_to_call(
+                            data, metadata=metadata, return_logits=True
+                        )
+                        loss, losses = self._calculate_loss(output, targets, metadata)
                 else:
-                    output = model_to_call(data, True)
-                    loss, losses = self._calculate_loss(output, targets)
+                    output = model_to_call(data, metadata=metadata, return_logits=True)
+                    loss, losses = self._calculate_loss(output, targets, metadata)
 
                 self.scaler.scale(loss).backward()
 
@@ -1672,7 +1693,10 @@ def _train_epoch(
 
     @beartype
     def _calculate_loss(
-        self, output: dict[str, Tensor], targets: dict[str, Tensor]
+        self,
+        output: dict[str, Tensor],
+        targets: dict[str, Tensor],
+        metadata: Optional[dict[str, Tensor]] = None,
     ) -> tuple[Tensor, dict[str, Tensor]]:
         """Calculates the loss for the given output and targets.
 
@@ -1692,14 +1716,12 @@ def _calculate_loss(
             - A dictionary of individual (unweighted) loss Tensors for each
               target column.
         """
-        target_names = [
-            k for k in targets.keys() if k not in TARGET_BATCH_METADATA_KEYS
-        ]
+        target_names = [k for k in targets.keys() if k in self.target_column_types]
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
-        if "_target_valid_mask" in targets:
-            seq_mask_2d = targets["_target_valid_mask"].bool()
+        if metadata is not None and "target_valid_mask" in metadata:
+            seq_mask_2d = metadata["target_valid_mask"].bool()
         else:
             mask_col = next(
                 (
@@ -1716,8 +1738,8 @@ def _calculate_loss(
             else:
                 seq_mask_2d = targets[mask_col] != 0
 
-        if "_bert_mask" in targets:
-            seq_mask_2d = seq_mask_2d & targets["_bert_mask"].bool()
+        if metadata is not None and "bert_mask" in metadata:
+            seq_mask_2d = seq_mask_2d & metadata["bert_mask"].bool()
 
         mask = seq_mask_2d.T.contiguous().reshape(-1)
 
@@ -1825,8 +1847,8 @@ def _evaluate(
         and aggregates results across all processes if in distributed mode.
         Also calculates a one-time baseline loss on the first call.
         The DataLoader is expected to yield tuples of
-        (sequences_dict, targets_dict, sequence_ids, subsequence_ids, start_positions).
-        The IDs and positions are currently unused during evaluation.
+        (sequences_dict, targets_dict, metadata_dict, sequence_ids, subsequence_ids).
+        The IDs are currently unused during evaluation.
 
         Args:
             valid_loader: DataLoader for the validation dataset.
@@ -1849,20 +1871,26 @@ def _evaluate(
         model_to_call.eval()
 
         with torch.no_grad():
-            for data, targets, _, _, _ in valid_loader:
+            for data, targets, metadata, _, _ in valid_loader:
                 # Move data to the current process's assigned GPU
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
-                    if k in self.input_columns or k in DATA_BATCH_METADATA_KEYS
+                    if k in self.input_columns
                 }
                 targets = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in targets.items()
-                    if k in self.target_column_types or k in TARGET_BATCH_METADATA_KEYS
+                    if k in self.target_column_types
+                }
+                metadata = {
+                    k: v.to(self.device, non_blocking=True)
+                    for k, v in (metadata or {}).items()
                 }
                 if self.hparams.training_spec.training_objective == "bert":
-                    data, targets = apply_bert_masking(data, targets, self.hparams)
+                    data, targets, metadata = apply_bert_masking(
+                        data, targets, metadata, self.hparams
+                    )
 
                 if (
                     self.hparams.training_spec.layer_autocast
@@ -1878,18 +1906,20 @@ def _evaluate(
                     with torch.autocast(
                         device_type=self.device.split(":")[0], dtype=amp_dtype
                     ):
-                        output = model_to_call(data, True)
-                        loss, losses = self._calculate_loss(output, targets)
+                        output = model_to_call(
+                            data, metadata=metadata, return_logits=True
+                        )
+                        loss, losses = self._calculate_loss(output, targets, metadata)
                 else:
-                    output = model_to_call(data, True)
-                    loss, losses = self._calculate_loss(output, targets)
+                    output = model_to_call(data, metadata=metadata, return_logits=True)
+                    loss, losses = self._calculate_loss(output, targets, metadata)
 
                 total_loss_collect.append(loss.item())
                 for col, loss in losses.items():
                     total_losses_collect[col].append(loss.item())
 
                 # Free up GPU memory
-                del data, targets, loss, losses
+                del data, targets, metadata, loss, losses
                 if self.device == "cuda":
                     torch.cuda.empty_cache()
 
@@ -1941,16 +1971,20 @@ def _evaluate(
             baseline_losses_local_collect = {col: [] for col in self.target_columns}
 
             # Iterate over the sharded validation loader
-            for data, targets, _, _, _ in valid_loader:
+            for data, targets, metadata, _, _ in valid_loader:
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
-                    if k in self.input_columns or k in DATA_BATCH_METADATA_KEYS
+                    if k in self.input_columns
                 }
                 targets = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in targets.items()
-                    if k in self.target_column_types or k in TARGET_BATCH_METADATA_KEYS
+                    if k in self.target_column_types
+                }
+                metadata = {
+                    k: v.to(self.device, non_blocking=True)
+                    for k, v in (metadata or {}).items()
                 }
 
                 pseudo_output = {}
@@ -1961,14 +1995,10 @@ def _evaluate(
                             col, data[col].transpose(0, 1)
                         )
                         targets_for_baseline[col] = targets[col]
-                if "_target_valid_mask" in targets:
-                    targets_for_baseline["_target_valid_mask"] = targets[
-                        "_target_valid_mask"
-                    ]
 
                 if len(pseudo_output) > 0:
                     loss, losses = self._calculate_loss(
-                        pseudo_output, targets_for_baseline
+                        pseudo_output, targets_for_baseline, metadata
                     )
                     baseline_loss_local_collect.append(loss.item())
                     for col, loss_ in losses.items():
@@ -2478,6 +2508,7 @@ def infer_with_embedding_model(
     device: str,
     size: int,
     target_columns: list[str],
+    metadata: Optional[list[dict[str, np.ndarray]]] = None,
 ) -> np.ndarray:
     """Performs inference with an embedding model.
 
@@ -2496,7 +2527,7 @@ def infer_with_embedding_model(
     categorical_cols = set(model.transformer_model.categorical_columns)
 
     with torch.no_grad():
-        for x_sub in x:
+        for batch_idx, x_sub in enumerate(x):
             layer_types = (
                 model.transformer_model.hparams.training_spec.layer_type_dtypes or {}
             )
@@ -2508,8 +2539,16 @@ def infer_with_embedding_model(
                     data_gpu[col] = torch.from_numpy(x_).to(device, dtype=torch.int64)
                 else:
                     data_gpu[col] = torch.from_numpy(x_).to(device, dtype=ref_dtype)
+            metadata_gpu = (
+                {
+                    col: torch.from_numpy(x_).to(device)
+                    for col, x_ in metadata[batch_idx].items()
+                }
+                if metadata
+                else {}
+            )
 
-            output_gpu = model.forward(data_gpu)
+            output_gpu = model.forward(data_gpu, metadata=metadata_gpu)
             output_cpu = output_gpu.cpu().detach().float().numpy()
             output_cpu = output_cpu.transpose(1, 0, 2).reshape(
                 output_cpu.shape[0] * output_cpu.shape[1], output_cpu.shape[2]
@@ -2529,6 +2568,7 @@ def infer_with_generative_model(
     device: str,
     size: int,
     target_columns: list[str],
+    metadata: Optional[list[dict[str, np.ndarray]]] = None,
 ) -> dict[str, np.ndarray]:
     """Performs inference with a generative model.
 
@@ -2548,7 +2588,7 @@ def infer_with_generative_model(
     categorical_cols = set(model.categorical_columns)
 
     with torch.no_grad():
-        for x_sub in x:
+        for batch_idx, x_sub in enumerate(x):
             layer_types = model.hparams.training_spec.layer_type_dtypes or {}
             dtype_str = layer_types.get("linear", "float32")
             ref_dtype = get_torch_dtype(dtype_str)
@@ -2558,8 +2598,16 @@ def infer_with_generative_model(
                     data_gpu[col] = torch.from_numpy(x_).to(device, dtype=torch.int64)
                 else:
                     data_gpu[col] = torch.from_numpy(x_).to(device, dtype=ref_dtype)
+            metadata_gpu = (
+                {
+                    col: torch.from_numpy(x_).to(device)
+                    for col, x_ in metadata[batch_idx].items()
+                }
+                if metadata
+                else {}
+            )
 
-            output_gpu = model.forward(data_gpu)
+            output_gpu = model.forward(data_gpu, metadata=metadata_gpu)
             output_cpu = {k: v.cpu().detach() for k, v in output_gpu.items()}
             outs0.append(output_cpu)
             if device == "cuda":
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index e32fe430..dff33ff3 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -128,10 +128,10 @@ def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
     dataset = SequifierDatasetFromFolderParquetLazy(
         str(data_dir), mock_config, shuffle=False
     )
-    seq_batch, tgt_batch, _, _, _ = next(iter(dataset))
+    seq_batch, tgt_batch, metadata_batch, _, _ = next(iter(dataset))
 
     assert torch.equal(
-        seq_batch["_attention_valid_mask"],
+        metadata_batch["attention_valid_mask"],
         torch.tensor(
             [
                 [True, True],
@@ -143,7 +143,7 @@ def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
         ),
     )
     assert torch.equal(
-        tgt_batch["_target_valid_mask"],
+        metadata_batch["target_valid_mask"],
         torch.tensor(
             [
                 [True, True],
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 733b3301..f3194651 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -129,10 +129,10 @@ def side_effect(path, map_location, weights_only):
         dataset = SequifierDatasetFromFolderPtLazy(
             dataset_path, mock_config, shuffle=False
         )
-        seq_dict, tgt_dict, _, _, _ = next(iter(dataset))
+        seq_dict, tgt_dict, metadata_dict, _, _ = next(iter(dataset))
 
     assert torch.equal(
-        seq_dict["_attention_valid_mask"],
+        metadata_dict["attention_valid_mask"],
         torch.tensor(
             [
                 [True, True, True, True, True],
@@ -144,7 +144,7 @@ def side_effect(path, map_location, weights_only):
         ),
     )
     assert torch.equal(
-        tgt_dict["_target_valid_mask"],
+        metadata_dict["target_valid_mask"],
         torch.tensor(
             [
                 [True, True, True, True, True],
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index bd06c425..f046e9e5 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -87,13 +87,14 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     all_columns = ["A"]
     seq_length = 3
 
-    tensors = numpy_to_pytorch(
+    tensors, metadata = numpy_to_pytorch(
         data, column_types, all_columns, seq_length, data_offset=1, target_offset=0
     )
 
     # 1. Check Keys
     assert "A" in tensors
     assert "A_target" in tensors
+    assert metadata == {}
 
     # 2. Check Input Tensor (Cols 3, 2, 1)
     # Row 0: [10, 20, 30]
@@ -118,7 +119,7 @@ def test_numpy_to_pytorch_dtypes():
 
     # Case 1: Integer
     data_int = pl.DataFrame({"inputCol": ["int_col"], "1": [10], "0": [20]})
-    tensors_int = numpy_to_pytorch(
+    tensors_int, _ = numpy_to_pytorch(
         data_int,
         {"int_col": torch.int64},
         ["int_col"],
@@ -130,7 +131,7 @@ def test_numpy_to_pytorch_dtypes():
 
     # Case 2: Float
     data_float = pl.DataFrame({"inputCol": ["float_col"], "1": [10.5], "0": [20.5]})
-    tensors_float = numpy_to_pytorch(
+    tensors_float, _ = numpy_to_pytorch(
         data_float,
         {"float_col": torch.float32},
         ["float_col"],
@@ -188,7 +189,7 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
         }
     )
 
-    tensors = numpy_to_pytorch(
+    _, metadata = numpy_to_pytorch(
         data,
         {"A": torch.float32},
         ["A"],
@@ -198,11 +199,11 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
     )
 
     assert torch.equal(
-        tensors["_attention_valid_mask"],
+        metadata["attention_valid_mask"],
         torch.tensor([[True, True, True], [False, False, True]]),
     )
     assert torch.equal(
-        tensors["_target_valid_mask"],
+        metadata["target_valid_mask"],
         torch.tensor([[True, True, True], [False, True, True]]),
     )
 
@@ -233,16 +234,20 @@ def test_apply_bert_masking_uses_explicit_valid_mask_for_zero_values():
     )
     data_batch = {
         "real_col": torch.tensor([[99.0, 0.0, 1.0, 2.0]]),
-        "_attention_valid_mask": torch.tensor([[False, True, True, True]]),
     }
     targets_batch = {
         "real_col": torch.tensor([[99.0, 0.0, 1.0, 2.0]]),
-        "_target_valid_mask": torch.tensor([[False, True, True, True]]),
+    }
+    metadata = {
+        "attention_valid_mask": torch.tensor([[False, True, True, True]]),
+        "target_valid_mask": torch.tensor([[False, True, True, True]]),
     }
 
-    _, masked_targets = apply_bert_masking(data_batch, targets_batch, config)
+    _, _, masked_metadata = apply_bert_masking(
+        data_batch, targets_batch, metadata, config
+    )
 
     assert torch.equal(
-        masked_targets["_bert_mask"],
+        masked_metadata["bert_mask"],
         torch.tensor([[False, True, True, True]]),
     )
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 0b0bdd0b..5a5bedc7 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -219,11 +219,13 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
     }
     targets = {
         "real_col": torch.tensor([[0.0, 0.0, 2.0]]),
-        "_target_valid_mask": torch.tensor([[True, True, True]]),
+    }
+    metadata = {
+        "target_valid_mask": torch.tensor([[True, True, True]]),
     }
 
     total_loss, component_losses = TransformerModel._calculate_loss(
-        model, outputs, targets
+        model, outputs, targets, metadata
     )
 
     assert torch.isclose(total_loss, torch.tensor(2.0))
@@ -237,10 +239,12 @@ def test_infer_attention_valid_mask_prefers_explicit_mask_for_real_zero_inputs()
 
     src = {
         "real_col": torch.tensor([[0.0, 0.0, 1.0]]),
-        "_attention_valid_mask": torch.tensor([[True, True, True]]),
+    }
+    metadata = {
+        "attention_valid_mask": torch.tensor([[True, True, True]]),
     }
 
-    mask = TransformerModel._infer_attention_valid_mask(model, src)
+    mask = TransformerModel._infer_attention_valid_mask(model, src, metadata)
 
     assert torch.equal(mask, torch.tensor([[True, True, True]]))
 

From 513cdf8a49d35ba3867e755d391ebcdcc442f925 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 5 Jun 2026 20:59:38 +0200
Subject: [PATCH 11/81] Deduplicate padding inference

---
 src/sequifier/helpers.py       | 18 ++++++++---
 src/sequifier/preprocess.py    | 18 ++++++++++-
 src/sequifier/train.py         | 56 ++++++----------------------------
 tests/integration-test-log.txt |  3 --
 tests/unit/test_preprocess.py  | 11 +++++--
 tests/unit/test_train.py       | 11 ++++---
 6 files changed, 56 insertions(+), 61 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index af839d0e..85f46d41 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -616,9 +616,17 @@ def get_last_training_batch_timedelta(
     return (t2 - t1).total_seconds()
 
 
-def infer_valid_mask_from_data(data_batch: Dict[str, torch.Tensor], config: Any):
-    if len(config.categorical_columns):
-        ref_col = config.categorical_columns[0]
+def infer_valid_mask_from_data(
+    data_batch: Dict[str, torch.Tensor],
+    categorical_columns: list[str],
+    default_key: str,
+    metadata: Optional[dict[str, Tensor]] = None,
+):
+    if metadata is not None and default_key in metadata:
+        return metadata[default_key].bool()
+
+    if len(categorical_columns):
+        ref_col = categorical_columns[0]
         return data_batch[ref_col] != 0
     else:
         warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
@@ -648,7 +656,9 @@ def apply_bert_masking(
     if "attention_valid_mask" in metadata_batch:
         valid_mask = metadata_batch["attention_valid_mask"].bool()
     else:
-        valid_mask = infer_valid_mask_from_data(data_batch, config)
+        valid_mask = infer_valid_mask_from_data(
+            data_batch, config.categorical_columns, "attention_valid_mask"
+        )
 
     batch_size, seq_len = valid_mask.shape
     device = valid_mask.device
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index ae658b89..5e4f3ed7 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -4,6 +4,7 @@
 import os
 import re
 import shutil
+import warnings
 from pathlib import Path
 from typing import Any, Optional, Union
 
@@ -1466,7 +1467,22 @@ def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]
     ids = sorted(
         [int(x) if not isinstance(x, str) else x for x in np.unique(data[column])]
     )  # type: ignore
-    id_map = {id_: i + 3 for i, id_ in enumerate(ids)}
+
+    if isinstance(ids[0], str):
+        if "[mask]" in ids:
+            raise ValueError(f"Found value '[mask]' in {column}, this is invalid")
+
+        for special_val in ["[unknown]", "[other]"]:
+            if special_val in ids:
+                warnings.warn(
+                    f"Found special value {special_val} in {column}, these will be combined with the sequifier-internal special value {special_val}"
+                )
+        ids = [id_ for id_ in ids if id_ not in ["[unknown]", "[other]"]]
+        id_map = {id_: i + 3 for i, id_ in enumerate(ids)}
+        id_map["[unknown]"] = 0
+        id_map["[other]"] = 1
+    else:
+        id_map = {id_: i + 3 for i, id_ in enumerate(ids)}
     return dict(id_map)
 
 
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index badba5cf..29e8121c 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -55,13 +55,13 @@
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
 from sequifier.helpers import (  # noqa: E402
-    EXPLICIT_PADDING_MASK_FALLBACK_WARNING,
     apply_bert_masking,
     conditional_beartype,
     configure_determinism,
     configure_logger,
     construct_index_maps,
     get_torch_dtype,
+    infer_valid_mask_from_data,
     normalize_path,
 )
 from sequifier.io.sequifier_dataset_from_file import (  # noqa: E402
@@ -976,30 +976,6 @@ def _recursive_concat(self, srcs: list[Tensor]):
                 srcs_inner.append(src)
             return self._recursive_concat(srcs_inner)
 
-    @conditional_beartype
-    def _infer_attention_valid_mask(
-        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
-    ) -> Tensor:
-        """
-        Returns a boolean mask of shape (batch_size, seq_length).
-
-        True means valid/non-padding token.
-        False means padding token.
-        """
-        if metadata is not None and "attention_valid_mask" in metadata:
-            return metadata["attention_valid_mask"].bool()
-
-        # Prefer categorical columns because padding is unambiguously encoded as 0.
-        for col in self.categorical_columns:
-            if col in src:
-                return src[col] != 0
-
-        # Fallback for real-only data.
-        # This matches the existing heuristic used in BERT masking.
-        warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
-        ref_col = next(col for col in self.input_columns if col in src)
-        return (src[ref_col] != 0.0).long().cumsum(dim=1) > 0
-
     @conditional_beartype
     def _build_attention_mask(self, valid_mask: Tensor, dtype: torch.dtype) -> Tensor:
         batch_size, seq_length = valid_mask.shape
@@ -1115,9 +1091,9 @@ def forward_inner(
         if self.joint_embedding_layer is not None:
             src2 = self.joint_embedding_layer(src2)
 
-        valid_mask = self._infer_attention_valid_mask(src, metadata).to(
-            device=src2.device
-        )
+        valid_mask = infer_valid_mask_from_data(
+            src, self.hparams.categorical_columns, "attention_valid_mask", metadata
+        ).to(device=src2.device)
         if valid_mask.shape != src2.shape[:2]:
             raise ValueError(
                 f"Invalid attention mask shape: got {tuple(valid_mask.shape)}, "
@@ -1720,28 +1696,14 @@ def _calculate_loss(
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
-        if metadata is not None and "target_valid_mask" in metadata:
-            seq_mask_2d = metadata["target_valid_mask"].bool()
-        else:
-            mask_col = next(
-                (
-                    col
-                    for col in target_names
-                    if self.target_column_types[col] == "categorical"
-                ),
-                target_names[0],
-            )
-
-            if self.target_column_types[mask_col] == "real":
-                warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
-                seq_mask_2d = (targets[mask_col] != 0.0).long().cumsum(dim=1) > 0
-            else:
-                seq_mask_2d = targets[mask_col] != 0
+        valid_mask = infer_valid_mask_from_data(
+            targets, self.categorical_columns, "target_valid_mask", metadata
+        )
 
         if metadata is not None and "bert_mask" in metadata:
-            seq_mask_2d = seq_mask_2d & metadata["bert_mask"].bool()
+            valid_mask = valid_mask & metadata["bert_mask"].bool()
 
-        mask = seq_mask_2d.T.contiguous().reshape(-1)
+        mask = valid_mask.T.contiguous().reshape(-1)
 
         losses = {}
         for target_column in target_names:
diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
index ef9fa309..8ba6ff1d 100644
--- a/tests/integration-test-log.txt
+++ b/tests/integration-test-log.txt
@@ -64,9 +64,6 @@ sequifier visualize-training model-categorical-distributed-lazy-parquet --projec
 sequifier visualize-training model-categorical-lazy --project-root tests/project_folder
 sequifier visualize-training model-categorical-multitarget-5 --project-root tests/project_folder
 sequifier visualize-training model-categorical-multitarget-5-eager --project-root tests/project_folder
-sequifier visualize-training model-real-1 --project-root tests/project_folder
-sequifier visualize-training model-real-1-from-epoch-checkpoint --project-root tests/project_folder
 sequifier visualize-training model-real-3 --project-root tests/project_folder
 sequifier visualize-training model-real-5 --project-root tests/project_folder
 sequifier visualize-training model-real-50 --project-root tests/project_folder
-sequifier visualize-training test-hp-search-grid-run-0,test-hp-search-grid-run-1,test-hp-search-grid-run-2,test-hp-search-grid-run-3 --project-root tests/project_folder --log-scale --bucket-training-batches 5
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 95b7e54c..82b56c35 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -267,8 +267,15 @@ def test_get_column_statistics_state_accumulation():
 
     # Validations
     # 1. Categorical: Should contain all unique keys a,b,c,d
-    assert len(id_maps["cat_col"]) == 4
-    assert set(id_maps["cat_col"].keys()) == {"a", "b", "c", "d"}
+    assert len(id_maps["cat_col"]) == 6
+    assert set(id_maps["cat_col"].keys()) == {
+        "[unknown]",
+        "[other]",
+        "a",
+        "b",
+        "c",
+        "d",
+    }
 
     # 2. Numerical: Should match full dataset calculation
     expected_mean = data_full["num_col"].mean()
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 5a5bedc7..7a4518b2 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -4,6 +4,7 @@
 import torch
 
 from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
+from sequifier.helpers import infer_valid_mask_from_data
 from sequifier.train import TransformerModel
 
 
@@ -207,7 +208,7 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
     model.target_column_types = {"real_col": "real"}
     model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
     model.loss_weights = None
-
+    model.categorical_columns = []
     outputs = {
         "real_col": torch.tensor(
             [
@@ -221,7 +222,7 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
         "real_col": torch.tensor([[0.0, 0.0, 2.0]]),
     }
     metadata = {
-        "target_valid_mask": torch.tensor([[True, True, True]]),
+        "attention_valid_mask": torch.tensor([[True, True, True]]),
     }
 
     total_loss, component_losses = TransformerModel._calculate_loss(
@@ -232,7 +233,7 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
     assert torch.isclose(component_losses["real_col"], torch.tensor(2.0))
 
 
-def test_infer_attention_valid_mask_prefers_explicit_mask_for_real_zero_inputs():
+def test_infer_valid_mask_from_data():
     model = TransformerModel.__new__(TransformerModel)
     model.categorical_columns = []
     model.input_columns = ["real_col"]
@@ -244,7 +245,9 @@ def test_infer_attention_valid_mask_prefers_explicit_mask_for_real_zero_inputs()
         "attention_valid_mask": torch.tensor([[True, True, True]]),
     }
 
-    mask = TransformerModel._infer_attention_valid_mask(model, src, metadata)
+    mask = infer_valid_mask_from_data(
+        src, model.categorical_columns, "attention_valid_mask", metadata
+    )
 
     assert torch.equal(mask, torch.tensor([[True, True, True]]))
 

From 0b99cbea235d383be33f393eca05820492442c7d Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 5 Jun 2026 22:24:15 +0200
Subject: [PATCH 12/81] Fix test

---
 tests/integration-test-log.txt | 3 +++
 tests/unit/test_train.py       | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
index 8ba6ff1d..ef9fa309 100644
--- a/tests/integration-test-log.txt
+++ b/tests/integration-test-log.txt
@@ -64,6 +64,9 @@ sequifier visualize-training model-categorical-distributed-lazy-parquet --projec
 sequifier visualize-training model-categorical-lazy --project-root tests/project_folder
 sequifier visualize-training model-categorical-multitarget-5 --project-root tests/project_folder
 sequifier visualize-training model-categorical-multitarget-5-eager --project-root tests/project_folder
+sequifier visualize-training model-real-1 --project-root tests/project_folder
+sequifier visualize-training model-real-1-from-epoch-checkpoint --project-root tests/project_folder
 sequifier visualize-training model-real-3 --project-root tests/project_folder
 sequifier visualize-training model-real-5 --project-root tests/project_folder
 sequifier visualize-training model-real-50 --project-root tests/project_folder
+sequifier visualize-training test-hp-search-grid-run-0,test-hp-search-grid-run-1,test-hp-search-grid-run-2,test-hp-search-grid-run-3 --project-root tests/project_folder --log-scale --bucket-training-batches 5
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 7a4518b2..f581b40b 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -222,7 +222,7 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
         "real_col": torch.tensor([[0.0, 0.0, 2.0]]),
     }
     metadata = {
-        "attention_valid_mask": torch.tensor([[True, True, True]]),
+        "target_valid_mask": torch.tensor([[True, True, True]]),
     }
 
     total_loss, component_losses = TransformerModel._calculate_loss(

From d412e826af8546957c59e307f87ccc5cf85ee9f4 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 8 Jun 2026 11:39:23 +0200
Subject: [PATCH 13/81] Update outputs

---
 ...orical-1-best-embedding-3-3-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-4-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-5-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-6-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-7-embeddings.csv |   2 +-
 ...inf-size-best-embedding-3-0-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-1-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-2-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-3-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-4-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-5-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-6-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-7-embeddings.csv |   6 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...uifier-model-real-1-best-3-predictions.csv |  20 +-
 ...uifier-model-real-3-best-3-predictions.csv |  36 +-
 ...uifier-model-real-5-best-3-predictions.csv |  28 +-
 ...ifier-model-real-50-best-3-predictions.csv |  38 +-
 ...custom-eval-run-0-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-7-predictions.csv |  60 +--
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 74 files changed, 1588 insertions(+), 1588 deletions(-)

diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
index 961e9b5e..ecec6901 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,7,0.38537663,-0.76236117,0.54453766,1.1489445,-0.93219095,-1.0841125,-0.279618,0.190426,0.066999875,0.60205054,0.7307121,0.975402,-2.318475,0.97368747,1.6034338,0.94340336,1.2762036,-0.30378067,-0.9700833,0.62717986,1.3900626,-0.94647413,0.48373678,-0.030570993,-0.42904085,-1.1276957,-0.7923079,0.563961,-0.40978292,-0.2334528,-0.27136517,-0.8576837,0.69629365,-1.6003033,0.11132377,-0.5791753,1.9399358,-0.9257662,0.5349056,-0.49527618,-1.3456327,-0.17603794,1.3494829,-1.1282389,0.20037875,-0.9559931,-0.8822725,-0.07519408,-0.77170074,-0.9350284,-0.5532877,-0.94019765,-0.10448791,1.1707094,0.6575927,1.3083346,-0.61123514,1.6713382,0.16979483,-2.0898654,0.009840045,-0.44183388,-1.1921828,-1.265742,1.6432867,-1.1305618,-0.46437404,-0.8430221,0.7453205,-0.65720564,0.35852164,1.6237742,-2.0383017,1.1427602,-0.8917945,0.2076105,-1.3489536,0.6207456,-0.44713888,-1.8621222,-0.5244267,-0.45581636,-0.3975574,-0.4345815,1.8005577,-2.0100255,0.3588848,-1.2393867,0.95991385,-0.65878433,0.7419609,0.43640068,-0.37811753,-1.3537853,1.0124421,0.12569557,0.45669934,-0.062139753,-0.68999034,-0.7770923,-1.26833,-0.42139238,-0.41232172,1.722412,0.38184437,-0.50226235,0.19698814,-2.7483191,-0.18802643,1.3159317,0.4502178,-0.26388723,0.5703611,-0.85277057,-0.46127146,0.2382723,-0.42989612,0.4115149,-2.6187692,0.5646321,1.1139393,-0.032928452,0.10219407,-0.6601092,-1.3054988,0.093954325,0.43201578,-1.4673975,0.08933542,-0.7154099,-1.6057762,0.40757596,-1.6277497,0.05748511,-0.1892006,0.10566295,-0.8918468,-0.17913234,-1.4436157,0.07747513,0.6109399,-0.8347807,-1.3056538,-0.8713027,0.28275943,0.87232286,0.8705629,-0.14861424,0.91572547,0.6586051,0.67494464,-1.7701943,0.16297512,0.4728508,1.0638494,1.7032028,-0.6542731,-0.017882321,2.1696196,-0.04748722,-0.77344686,-2.259672,-1.1815661,-0.58230156,-1.6384532,-1.8893389,0.917642,0.3281549,0.40393066,0.178792,-0.32735556,-1.1835171,-1.0626796,-0.12413608,-0.13082595,0.59983134,-1.7939893,0.5362146,1.2261001,-1.4384091,0.4796257,-0.45061558,-0.10099141,0.51506233,-1.2538948,1.3633755,-0.588958,-0.21253218,-1.3333261,0.5079261,-0.97876984,-1.5163478,-0.66665024,1.8272513,-1.481488,2.3199954,-0.635016,0.18672499,0.83364874,0.9136821
-4,0,7,1.832777,0.9652215,-0.2531473,0.2162134,1.9031242,0.70369786,-0.9306243,-0.88834244,0.55274016,0.5570371,1.0132977,-0.9646427,-0.5195061,-0.055798862,-0.37479177,0.4407284,0.16671644,0.5790386,0.41375777,0.80011296,-1.5359619,-0.22593135,1.7854866,-0.71542126,-1.0921632,-0.04137182,-0.21795602,-0.33691144,-1.2671196,-0.7473329,0.18677165,-1.8774186,2.0318227,0.273334,-0.9697613,-0.5860969,-0.36416003,0.2273136,0.47210863,-1.3540341,-0.25599045,0.6586222,1.5556432,-0.8154356,1.168724,-2.2140863,0.42116508,0.4787695,-0.64788985,0.4423342,0.20755024,-0.55014634,-0.595621,-0.6086162,-1.3033568,-0.92190075,1.9484235,-0.028180895,0.9555864,-0.610766,1.3185904,0.42626792,-1.1561954,-0.5161907,-0.83520865,0.22098021,-0.32627407,-1.1956958,1.73692,-1.1766446,0.07778312,-1.4228519,-2.1649313,0.6194682,-0.44674367,-0.010885017,-0.43634287,1.7373236,0.6609659,-0.20005386,-0.5650337,0.29500005,0.115850784,-0.054047436,1.7766252,-1.624335,0.4819138,1.0377665,-1.2305183,0.80048895,-1.9219421,1.6361128,0.13642035,-1.2401028,0.56233996,-0.15820792,1.5966988,-0.2590406,-0.5514963,0.43703985,0.40010962,-0.12492114,1.6866367,2.0976324,-0.49593025,0.8581704,0.21792285,-0.63466465,-1.7050067,-1.6924849,-0.08375951,-0.38175452,-0.3269363,-0.96400464,-0.90264994,-0.5941345,0.1949616,-0.8319494,-0.5528981,-0.80228716,1.6503075,1.2991755,-1.0835283,0.3950482,0.2928661,-0.13108586,-0.38219264,-0.5976724,-1.3788452,-0.2650713,0.34214625,1.046247,1.518125,-0.5940697,0.589522,2.7725646,-1.0286809,1.1897798,0.33696088,-1.0014184,-0.029821016,-0.5898694,-0.9996754,0.0104257725,1.4441028,-0.87508523,0.57642156,2.500753,-1.5390791,-1.6458074,-1.7197642,-0.74723464,1.16423,-0.3134553,-0.005579411,1.5905838,-0.9592407,-0.41740352,-0.6339657,0.8530593,1.299676,-0.11439452,0.39143115,0.46125954,-3.0907333,-0.27514967,0.20107317,-1.4204426,-1.7638286,-0.4874675,-1.2546072,-0.23415713,-0.10110022,-0.24766764,1.7387943,-0.75891596,0.75030476,0.5836463,0.10834095,0.7989047,-0.48462987,-0.62584233,0.5027705,1.1866,-0.2813258,-0.8015657,0.4489704,1.1347668,0.07279327,-0.6131345,0.88922334,0.71706647,0.65837896,0.33229914,1.1745048,-0.74355304,0.5451966,0.36238006,1.4138216,0.37747213
+3,0,7,0.3853767,-0.7623611,0.5445376,1.1489445,-0.93219095,-1.0841124,-0.27961808,0.19042599,0.06699983,0.6020505,0.7307122,0.9754021,-2.318475,0.97368747,1.6034338,0.94340336,1.2762036,-0.30378067,-0.97008324,0.62717986,1.3900626,-0.9464742,0.4837369,-0.030570975,-0.4290407,-1.1276956,-0.7923079,0.563961,-0.40978292,-0.23345281,-0.27136502,-0.8576837,0.69629353,-1.6003033,0.11132374,-0.5791753,1.9399358,-0.92576635,0.53490555,-0.49527624,-1.3456327,-0.176038,1.349483,-1.1282389,0.20037875,-0.955993,-0.8822724,-0.07519414,-0.77170074,-0.9350284,-0.5532876,-0.94019765,-0.10448791,1.1707093,0.6575927,1.3083346,-0.6112351,1.6713382,0.16979486,-2.0898654,0.00984007,-0.44183394,-1.1921827,-1.2657421,1.6432866,-1.130562,-0.46437404,-0.8430221,0.7453206,-0.65720564,0.35852155,1.6237744,-2.0383017,1.1427602,-0.8917945,0.20761068,-1.3489536,0.62074554,-0.44713894,-1.8621219,-0.5244267,-0.45581624,-0.39755735,-0.43458134,1.8005577,-2.0100255,0.3588848,-1.2393866,0.95991385,-0.65878445,0.741961,0.43640068,-0.37811747,-1.3537853,1.0124421,0.12569562,0.45669925,-0.062139705,-0.6899904,-0.7770923,-1.2683299,-0.42139253,-0.4123218,1.7224118,0.38184428,-0.5022623,0.1969878,-2.748319,-0.18802647,1.3159317,0.4502178,-0.26388726,0.5703611,-0.85277075,-0.46127146,0.23827219,-0.4298961,0.41151482,-2.6187694,0.5646321,1.1139392,-0.032928437,0.1021941,-0.6601091,-1.3054987,0.09395422,0.43201578,-1.4673975,0.08933542,-0.71540993,-1.6057761,0.4075759,-1.6277498,0.05748504,-0.18920064,0.1056629,-0.8918468,-0.17913246,-1.4436157,0.077475116,0.6109399,-0.83478075,-1.3056538,-0.87130284,0.28275952,0.87232274,0.8705629,-0.14861424,0.91572547,0.6586051,0.6749446,-1.7701945,0.1629751,0.4728509,1.0638493,1.7032028,-0.6542733,-0.017882407,2.1696196,-0.047487322,-0.77344674,-2.259672,-1.1815661,-0.5823015,-1.638453,-1.8893389,0.9176421,0.32815483,0.4039307,0.17879207,-0.32735565,-1.1835169,-1.0626798,-0.12413608,-0.13082603,0.5998312,-1.7939893,0.5362146,1.2261001,-1.4384091,0.4796257,-0.45061567,-0.100991376,0.51506233,-1.2538948,1.3633755,-0.58895797,-0.21253212,-1.3333261,0.50792605,-0.97876984,-1.5163478,-0.6666502,1.8272516,-1.4814882,2.3199954,-0.6350161,0.1867251,0.83364874,0.9136821
+4,0,7,1.8327771,0.9652215,-0.25314748,0.21621384,1.9031243,0.7036974,-0.930624,-0.8883423,0.55273986,0.5570372,1.0132973,-0.96464247,-0.51950616,-0.05579906,-0.37479147,0.44072837,0.16671647,0.5790387,0.41375738,0.80011296,-1.5359617,-0.22593148,1.7854865,-0.71542114,-1.0921632,-0.041372035,-0.21795599,-0.3369114,-1.2671196,-0.7473329,0.18677177,-1.8774186,2.0318227,0.2733341,-0.9697609,-0.58609706,-0.36415994,0.22731349,0.4721085,-1.3540337,-0.25599056,0.65862226,1.5556433,-0.81543595,1.168724,-2.2140863,0.42116505,0.47876942,-0.6478898,0.44233415,0.20755047,-0.5501464,-0.5956206,-0.60861623,-1.3033564,-0.9219007,1.9484233,-0.02818083,0.9555862,-0.6107661,1.3185904,0.4262676,-1.1561952,-0.5161904,-0.8352084,0.22098036,-0.32627368,-1.1956955,1.7369198,-1.1766452,0.07778335,-1.4228516,-2.164931,0.61946803,-0.4467436,-0.01088463,-0.43634295,1.737324,0.6609657,-0.20005353,-0.56503403,0.29499987,0.11585043,-0.05404717,1.7766248,-1.624335,0.4819137,1.0377663,-1.2305183,0.8004889,-1.9219416,1.6361125,0.13642022,-1.2401025,0.56233984,-0.15820779,1.5966985,-0.25904042,-0.5514964,0.43703973,0.4001095,-0.12492117,1.6866364,2.0976322,-0.49593046,0.8581702,0.2179227,-0.6346643,-1.7050065,-1.6924847,-0.08375938,-0.38175428,-0.32693657,-0.9640046,-0.9026502,-0.5941344,0.19496141,-0.8319495,-0.55289805,-0.80228716,1.650307,1.2991756,-1.0835283,0.39504835,0.29286614,-0.13108593,-0.38219267,-0.59767234,-1.3788447,-0.26507112,0.34214637,1.0462471,1.5181249,-0.59406966,0.58952206,2.7725644,-1.0286809,1.18978,0.3369609,-1.0014182,-0.02982089,-0.5898692,-0.99967504,0.01042568,1.4441023,-0.87508506,0.57642156,2.5007522,-1.5390786,-1.6458069,-1.7197639,-0.74723464,1.1642295,-0.31345525,-0.0055794707,1.5905838,-0.9592406,-0.4174035,-0.6339656,0.8530592,1.2996758,-0.11439483,0.3914311,0.46125937,-3.0907335,-0.27514952,0.20107323,-1.4204425,-1.7638282,-0.48746726,-1.2546072,-0.2341572,-0.10110028,-0.24766739,1.7387944,-0.7589158,0.75030494,0.5836463,0.10834081,0.7989048,-0.4846295,-0.62584215,0.5027704,1.1866003,-0.28132582,-0.80156577,0.44897035,1.1347668,0.07279313,-0.61313444,0.8892234,0.71706593,0.65837866,0.33229896,1.174505,-0.7435528,0.5451965,0.36238024,1.4138215,0.37747207
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
index 1ede8ed4..5278ebbd 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,7,0.8678485,-1.1492084,-0.29577827,0.9786946,-1.9542456,-0.07664234,-0.03973026,-0.09718513,-0.46550712,0.088307075,0.014937765,1.1039448,-1.9457144,0.7994327,1.8773035,1.5537525,0.39788535,-0.4953131,-1.4370357,1.0003624,0.841088,-0.99132866,0.15745488,0.4423288,-1.511852,-1.9748083,-1.3705088,0.117527954,0.0471839,0.72272456,0.15624413,-1.3321223,0.8274153,-0.50209063,-0.85796344,-0.82082087,1.3705729,-0.32510358,0.92136246,-0.4567902,-2.0167284,0.778191,1.3592294,-1.6879605,0.3511618,-1.6433352,-0.8796016,0.4710184,-0.49356312,-1.540319,-0.5620308,-0.7479297,0.11485024,0.5929443,1.0806792,0.8520335,-0.28449163,2.3426988,-0.7176778,-2.077216,0.44693425,-0.51118195,-1.8134435,-0.3199498,0.5275097,-0.7924806,-0.90600455,-0.5945148,0.62720704,-0.76678073,1.4406856,0.83508265,-1.3691335,0.43675628,-0.93423057,-0.587019,-0.5398221,1.3742958,-0.36840782,-1.3804276,-1.3538328,-1.2029364,-0.6361642,-0.20616663,1.1179503,-0.86969507,0.46374834,-0.9316529,0.36905083,0.18406925,0.57265496,-0.09701405,-0.3084512,-1.0851078,1.2294931,-0.43448195,0.4714254,0.2976226,-0.6675889,-0.24110621,-0.86843914,-0.15433894,0.3217168,1.7343627,0.64216596,-0.39611214,1.2057974,-3.0774295,0.8041049,1.0713865,0.31967443,0.8850117,0.5413772,-1.8904046,-0.64331496,-0.7131405,-0.8982892,0.49162266,-1.7892696,1.266022,1.2143359,0.16972244,-0.42226094,-0.920911,-1.287436,-0.30553937,0.78019506,-1.3130422,-0.54638034,-0.06269855,-1.2766076,0.3067186,-1.1979859,-0.5041536,0.24134868,0.9985403,-0.20058866,-0.19003277,-0.91562223,0.7388305,0.60628974,-0.39632004,-1.401826,0.266171,1.2859257,1.7502289,1.1291436,-0.2593596,1.0141768,-0.0024154119,-0.012608729,-1.6622946,0.84452707,0.4089601,0.24172992,0.8491719,-1.2459073,0.44094768,2.3070037,-0.20285222,-0.9055848,-1.9905902,-0.62915814,-0.7049501,-1.5236275,-1.6411532,0.28731275,0.23528811,-0.8580845,0.87269217,-0.8224613,-1.3814896,-0.8996288,0.0075937505,1.4490063,0.6224928,-1.3583354,-0.39259246,0.55505186,-1.1764414,1.2858897,-0.51201284,-0.36201963,0.028119631,-1.0583313,0.626196,-0.93444836,-0.6852238,-0.55748594,0.1277266,-1.1170981,-1.7487799,-0.62091154,1.1936655,-0.5524272,1.6852114,-1.5539126,0.93077433,0.54523623,-0.3293535
+5,0,7,0.8678487,-1.1492083,-0.2957783,0.9786945,-1.9542453,-0.07664241,-0.039730284,-0.09718514,-0.465507,0.088307,0.014937709,1.1039451,-1.9457144,0.7994327,1.8773037,1.5537523,0.39788526,-0.4953133,-1.4370357,1.0003625,0.84108806,-0.99132866,0.15745479,0.4423289,-1.5118519,-1.9748083,-1.3705087,0.11752797,0.04718379,0.7227247,0.15624402,-1.3321223,0.8274153,-0.5020909,-0.8579634,-0.8208209,1.370573,-0.32510355,0.9213625,-0.45679018,-2.0167282,0.77819085,1.3592292,-1.6879606,0.35116172,-1.6433353,-0.8796014,0.47101834,-0.49356297,-1.540319,-0.56203103,-0.74793,0.114850305,0.5929445,1.0806793,0.8520335,-0.28449142,2.3426988,-0.71767795,-2.077216,0.4469343,-0.51118195,-1.8134437,-0.31995004,0.5275096,-0.7924808,-0.90600455,-0.5945146,0.62720716,-0.76678073,1.4406855,0.8350828,-1.3691337,0.43675637,-0.93423057,-0.58701897,-0.53982204,1.3742961,-0.3684079,-1.3804277,-1.3538327,-1.2029366,-0.6361641,-0.2061667,1.1179498,-0.8696951,0.46374846,-0.9316529,0.36905092,0.18406926,0.5726551,-0.097014,-0.3084513,-1.0851082,1.2294934,-0.43448213,0.4714254,0.29762262,-0.66758883,-0.2411063,-0.8684391,-0.15433899,0.3217168,1.7343627,0.6421659,-0.39611214,1.2057977,-3.0774305,0.8041049,1.0713868,0.3196744,0.885012,0.5413774,-1.8904048,-0.6433149,-0.7131405,-0.89828914,0.49162254,-1.7892698,1.266022,1.2143357,0.16972247,-0.4222612,-0.9209112,-1.2874361,-0.30553934,0.78019506,-1.3130422,-0.54638046,-0.06269859,-1.2766076,0.3067186,-1.1979859,-0.5041537,0.24134871,0.9985403,-0.20058915,-0.19003288,-0.9156221,0.73883057,0.60628974,-0.39632013,-1.401826,0.26617125,1.2859261,1.7502289,1.1291437,-0.2593595,1.014177,-0.002415313,-0.0126089025,-1.6622946,0.84452736,0.40896028,0.24172997,0.849172,-1.2459072,0.4409478,2.3070042,-0.20285252,-0.9055849,-1.9905902,-0.6291581,-0.70495003,-1.5236276,-1.6411536,0.28731304,0.23528825,-0.85808456,0.8726922,-0.8224612,-1.3814898,-0.8996288,0.0075938655,1.4490063,0.62249285,-1.3583359,-0.39259228,0.55505204,-1.1764418,1.2858897,-0.5120127,-0.3620197,0.028119635,-1.0583314,0.6261959,-0.93444824,-0.68522364,-0.5574861,0.12772661,-1.1170981,-1.7487799,-0.62091166,1.1936655,-0.55242705,1.6852115,-1.5539129,0.93077433,0.54523623,-0.32935345
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
index 541113f3..257f845b 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,7,0.47420257,1.2638454,-0.29428783,0.6063462,0.3250649,-0.011651794,-1.383389,-1.234404,-0.45170534,1.0377884,1.8840407,-0.33119473,-1.0696625,0.9726884,0.80814546,0.9354395,0.71671236,0.428451,-0.015223929,0.81312406,-1.2162952,-1.1050556,0.6366257,-1.6387161,-0.9819109,-0.4953308,0.12158297,-0.67469645,-1.717451,-0.21087128,1.0931989,-1.6311557,1.6655924,-1.1295466,-0.8290883,-0.27834374,0.3571047,0.7679167,1.286279,-1.5655476,-0.47575822,0.3837134,0.82579845,-0.65413463,-0.060213037,-1.6886673,0.31571612,-1.3439902,-1.3430762,0.0015742916,-0.8718748,-1.2474332,-0.8262217,0.5390429,0.43257564,0.7165555,-0.013134941,1.0078024,1.2916026,-1.6230772,0.18352468,1.4117193,-0.99339473,-0.17660125,-0.8539951,0.17099175,1.0725634,-0.23263906,1.7940855,0.49845868,-0.58279014,0.9100562,-2.1336784,-0.08051058,-0.15371843,0.53641343,0.036809027,0.27225932,0.43129542,-0.51692873,0.86170065,0.3777755,-0.3340567,-0.5066735,2.3349266,-3.2664793,0.23907657,0.13636138,-0.13565235,-0.29753345,-0.56251806,1.5495608,-0.25081062,-2.0128417,1.4139545,-0.16621536,0.784448,0.5935694,-0.85539657,-0.5690536,-0.5415991,-0.8971023,1.8330036,2.7651486,1.039496,-0.23426558,-0.57880366,-1.1803632,-1.0954609,-0.30837312,0.2056301,-0.45784757,0.79928094,-0.58580524,0.8425376,0.11244702,0.72019076,-1.2291323,-1.666961,-0.4446293,2.2034054,1.0664692,0.81338817,-0.4559747,-1.2628947,-0.2225234,-1.3892788,-0.7460107,0.097924024,-0.91731095,-1.1128799,0.40236434,-0.7147696,-0.7736082,0.39257404,2.3405876,-0.27565396,0.40534735,-0.81300443,-0.097918786,0.61959285,-1.4504018,-1.5624175,-1.1680082,1.2857004,-1.2578357,0.41585726,1.0423983,-0.53722256,-0.38153195,-0.51949257,-1.5859617,1.3256052,-0.097611226,0.5610492,2.6576302,1.24299,-0.7566358,0.91004014,0.7399129,-0.1096572,-2.0155513,-0.009600659,-0.6858703,-1.7821821,-1.058989,-0.13436024,0.42209816,-1.1379089,0.4846043,-0.6162597,-0.33495048,0.5386396,0.18685773,0.26678535,-0.20175369,-0.72045624,0.21726054,1.1120163,0.79319596,-0.36998135,-0.67134833,0.7267623,-0.18856248,0.4676755,-1.0685669,0.32332313,-0.06708168,-1.2248515,-0.7845494,0.5460868,-0.6168076,1.5106831,1.7673893,-0.6841837,0.44412613,0.7936944,-0.8888349,0.039817397,1.6578218
-7,0,7,-0.08779775,-0.7206459,0.39195007,0.30277744,-1.0405563,-0.8779835,-0.7976403,0.071234435,-0.29967046,0.27734953,-0.105664976,0.77108765,-1.7671883,0.824929,1.5373951,1.5010271,1.4846263,-0.7896349,-0.8870087,0.69192314,0.6787518,-0.4330691,0.6862604,-0.2904696,-0.9619547,-0.3613372,-0.92799646,0.18862405,-0.13149612,0.33348697,0.35246965,-0.86275643,0.7647196,-1.5118189,-0.24293661,-0.62797713,1.9434764,-0.1523092,1.2640053,-0.53051686,-1.5717334,-0.23882772,1.2569783,-1.5155659,0.316097,-1.4623227,-1.5276753,-0.6301773,-0.15530248,-1.2590177,-1.1136596,-0.8283695,-0.3316041,1.6386757,0.7145627,1.4172399,-0.92403394,1.1554213,-0.2054311,-2.2690709,0.032772172,-0.60741836,-0.840842,-0.43909273,1.1674483,-1.5343083,0.39456716,-0.6319559,0.694979,-0.17052399,0.6638801,1.6770401,-1.6856104,0.8501758,-1.0100726,-0.06956186,-0.49753255,0.45815906,-0.8491857,-1.5388988,-0.21994215,-0.6159065,-0.43685117,-0.5483946,1.8440413,-1.9056607,0.239437,-0.8901972,1.0178717,-0.41829032,0.47147453,0.2824819,-0.4255798,-1.3693498,1.3788786,-0.8324858,0.2925801,0.038576633,-0.9931528,-0.88978964,-1.3541443,-0.3884408,-0.5666751,2.0802114,0.20410398,-0.5688041,-0.25536698,-2.8470824,0.22910744,1.3398811,0.41405335,0.37236086,1.0213393,-0.612114,0.20765163,0.22666231,-0.052260194,0.20032823,-2.3570495,1.430299,1.4743409,-0.2883624,0.44256932,-0.7141606,-0.87645125,-0.20459764,0.5313735,-1.1576004,0.2882263,-0.31856775,-2.041501,0.55573547,-1.3157725,0.2728312,0.18704545,-0.34522098,-0.7096739,-0.432244,-1.6015022,1.0732054,0.9189849,-0.6834102,-1.5872613,-0.5969934,0.90725124,1.1939424,0.6279301,-0.4908578,0.9603637,0.6682552,0.66679686,-1.698892,0.26019415,0.5773275,0.53784454,1.2200284,0.32803503,-0.15348214,2.141097,-0.0141839655,-1.0378202,-2.6077065,-0.3066732,-0.06714368,-1.3892244,-1.7354015,0.5355618,0.9041617,-0.3285599,0.12410717,-0.4072241,-1.3432459,-0.7320013,-0.25400367,0.17272092,1.1034756,-2.2850537,0.62561816,1.75507,-1.3452936,-0.017986165,-0.54053295,0.05288172,0.2870577,-0.98898715,0.76313764,0.15638581,-0.6830085,-1.3387133,0.6244851,-0.7181842,-1.9862365,0.07094988,1.840984,-1.5815738,1.6396283,-0.85405743,0.19265085,0.44211808,0.6777788
+6,0,7,0.47420236,1.2638448,-0.2942875,0.6063466,0.32506454,-0.011651637,-1.3833891,-1.2344038,-0.45170528,1.0377885,1.8840404,-0.33119497,-1.0696628,0.9726884,0.80814546,0.9354395,0.7167119,0.42845112,-0.015223959,0.81312424,-1.2162951,-1.1050558,0.6366257,-1.6387163,-0.9819105,-0.4953311,0.12158322,-0.67469686,-1.7174512,-0.2108714,1.0931984,-1.6311551,1.6655928,-1.1295462,-0.82908773,-0.27834365,0.35710472,0.7679168,1.2862792,-1.5655477,-0.47575817,0.38371342,0.8257985,-0.6541342,-0.06021277,-1.688667,0.31571636,-1.3439896,-1.3430765,0.0015745121,-0.87187445,-1.2474327,-0.82622105,0.53904265,0.4325758,0.7165552,-0.013134802,1.0078027,1.2916025,-1.6230769,0.1835245,1.4117191,-0.9933948,-0.17660141,-0.85399544,0.1709921,1.0725634,-0.23263937,1.7940853,0.49845877,-0.58278996,0.9100562,-2.1336787,-0.0805108,-0.15371813,0.5364138,0.036808733,0.27225935,0.43129516,-0.5169288,0.8617008,0.37777528,-0.3340564,-0.50667363,2.3349266,-3.2664785,0.23907632,0.13636121,-0.1356525,-0.2975334,-0.56251806,1.5495608,-0.25081053,-2.012842,1.4139545,-0.16621536,0.7844478,0.5935697,-0.85539633,-0.5690532,-0.5415988,-0.8971024,1.8330035,2.7651486,1.039496,-0.23426563,-0.57880294,-1.180363,-1.095461,-0.30837286,0.20563056,-0.45784765,0.799281,-0.585805,0.84253716,0.11244704,0.7201907,-1.2291319,-1.666961,-0.44462907,2.2034056,1.0664692,0.81338817,-0.4559743,-1.2628946,-0.22252318,-1.3892785,-0.74601096,0.09792413,-0.9173109,-1.1128796,0.4023642,-0.7147697,-0.77360827,0.39257395,2.3405876,-0.2756536,0.4053474,-0.8130045,-0.097918354,0.619593,-1.450402,-1.562418,-1.1680082,1.2857003,-1.2578353,0.41585723,1.0423979,-0.53722227,-0.38153183,-0.51949257,-1.5859616,1.3256049,-0.097610995,0.5610497,2.6576302,1.24299,-0.7566359,0.91003966,0.73991305,-0.10965724,-2.0155516,-0.009600935,-0.68587035,-1.7821821,-1.0589889,-0.13436046,0.4220983,-1.1379086,0.48460454,-0.61625963,-0.3349504,0.53863937,0.18685772,0.26678517,-0.20175351,-0.7204558,0.21726036,1.1120167,0.79319596,-0.36998117,-0.6713484,0.72676253,-0.18856229,0.46767554,-1.068567,0.32332298,-0.06708165,-1.2248515,-0.7845498,0.5460865,-0.61680746,1.510683,1.7673895,-0.68418354,0.44412625,0.7936944,-0.8888349,0.03981725,1.6578218
+7,0,7,-0.087797716,-0.72064584,0.3919503,0.30277744,-1.0405561,-0.8779837,-0.7976403,0.071234636,-0.29967022,0.2773496,-0.10566512,0.77108765,-1.7671878,0.824929,1.5373951,1.5010269,1.4846264,-0.7896349,-0.88700837,0.6919234,0.6787521,-0.4330691,0.6862605,-0.2904699,-0.9619546,-0.36133736,-0.9279964,0.18862396,-0.13149612,0.33348706,0.35246968,-0.86275643,0.7647198,-1.5118189,-0.24293631,-0.62797725,1.9434764,-0.15230915,1.2640054,-0.53051674,-1.571733,-0.23882791,1.2569784,-1.5155659,0.31609696,-1.4623228,-1.5276749,-0.63017756,-0.15530244,-1.2590177,-1.1136601,-0.8283698,-0.33160418,1.6386752,0.7145629,1.4172399,-0.9240341,1.155421,-0.20543078,-2.2690709,0.032772336,-0.6074186,-0.8408421,-0.43909267,1.1674483,-1.5343087,0.394567,-0.6319556,0.69497925,-0.17052399,0.66387993,1.6770401,-1.6856111,0.8501759,-1.0100725,-0.06956196,-0.49753228,0.45815924,-0.8491857,-1.5388991,-0.21994229,-0.61590695,-0.43685135,-0.54839456,1.844041,-1.9056602,0.23943688,-0.8901972,1.0178715,-0.4182901,0.47147426,0.28248206,-0.42557982,-1.3693498,1.378879,-0.832486,0.2925802,0.03857666,-0.99315274,-0.88979,-1.3541446,-0.38844082,-0.56667507,2.0802116,0.20410404,-0.5688037,-0.25536713,-2.8470826,0.22910759,1.339881,0.41405326,0.3723613,1.0213392,-0.61211425,0.20765173,0.22666255,-0.052260146,0.20032822,-2.3570492,1.4302993,1.4743407,-0.2883624,0.44256923,-0.7141606,-0.87645143,-0.20459794,0.5313736,-1.1576004,0.2882262,-0.3185675,-2.041501,0.55573565,-1.3157725,0.2728311,0.18704554,-0.34522107,-0.7096739,-0.43224367,-1.601502,1.0732054,0.91898465,-0.68341035,-1.5872612,-0.59699357,0.90725124,1.1939424,0.6279298,-0.49085775,0.9603635,0.668255,0.66679686,-1.698892,0.26019427,0.57732755,0.5378443,1.2200284,0.32803506,-0.15348187,2.141097,-0.014183752,-1.0378197,-2.6077065,-0.3066731,-0.06714382,-1.3892245,-1.7354015,0.53556174,0.9041617,-0.32855996,0.12410693,-0.40722385,-1.3432457,-0.73200125,-0.25400352,0.17272082,1.1034756,-2.2850537,0.6256181,1.7550697,-1.3452938,-0.017986335,-0.54053277,0.05288191,0.28705782,-0.98898697,0.7631376,0.15638618,-0.6830087,-1.3387133,0.62448514,-0.7181843,-1.9862368,0.070950225,1.8409837,-1.5815738,1.639628,-0.8540574,0.19265085,0.4421182,0.67777854
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
index 84ef00be..e1edc7ec 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,7,1.6896031,0.29085124,-0.23292182,1.3335884,0.45459226,0.34372637,-0.4165882,-0.42207715,0.60733074,1.2122438,0.8359352,0.5889716,-1.7851549,1.4177666,0.7491194,0.5095495,0.26941702,0.21018945,-0.88023686,0.56112635,0.4965653,-0.8069379,1.1553066,-0.012635032,-0.1019034,-1.7098439,-0.778535,0.08457131,-1.0872964,-0.46575657,0.1705186,-1.0981392,1.2801948,-0.2028321,-0.3777968,-1.0450279,0.83502,-0.4449086,0.3772813,-1.3267674,-1.3647468,0.4980426,1.0723178,-1.55848,0.6211332,-1.7105299,0.93637806,0.962023,-0.5543043,-0.2215327,0.51130736,-0.5170269,0.11910182,-0.16186877,0.3342436,-0.8031384,1.6559254,0.4785719,0.5240285,-0.85450006,1.1488028,-0.123829335,-2.314366,-1.7480177,0.34907198,0.33689144,-0.96712893,-1.1980494,1.7140707,-1.4368652,0.665594,-0.2777855,-1.740489,1.0585523,-0.34576628,-0.026606081,-1.963539,2.3253458,0.718786,-0.63990796,-2.111302,-0.76395285,-0.37198,-0.00446455,1.4715577,-1.4121685,0.046070743,0.04943243,-0.6720886,0.77220064,-0.78486633,1.6651013,-0.5048683,-1.0169151,0.19552472,0.60602754,1.0623702,0.15507461,-0.32467106,0.2723276,-0.9548157,-0.0605602,1.6082473,1.8419511,-0.6165263,0.5530546,0.5313712,-1.6408414,-0.0837646,-0.62233794,0.053751737,0.17250709,-0.5837834,-1.9890919,-2.1878119,-0.6350772,-1.3977064,-0.09614897,-0.99611235,-0.87689996,0.8914697,1.3250514,-1.3865021,0.07022134,0.14449094,-0.4635786,0.22937033,-1.0663302,-1.124088,0.17105857,0.38474694,1.002527,0.06262806,-1.439073,-0.079469405,2.1103234,-1.7293836,1.1202102,-0.19353467,-1.0497702,-0.18288192,-0.8191539,-0.91999054,-0.0011520431,1.4552093,0.0285639,1.5781827,1.5662957,-0.37651268,-0.7598728,-0.96915853,-1.9078715,1.1331103,-0.11489686,0.537274,2.1304922,-2.0829363,0.2657547,0.73497164,-0.30037352,0.74554676,-1.2468303,-0.9414173,-0.96966434,-2.6400287,-1.1223296,0.87630385,-0.9678277,-0.9243955,1.2501684,-1.4089605,-0.80661803,-0.8789456,0.509323,0.86616665,-1.0657727,-0.44514504,0.16784553,-0.82460415,-0.6202602,0.22096892,-0.5744518,-0.051139973,0.95496964,-0.73176867,0.21573815,-0.73088634,1.3564851,-0.56419003,-0.3550247,-1.0255396,-0.57220864,-0.90364045,0.36711964,-0.09569923,0.6653445,-0.5859866,1.2525154,1.7926838,0.031228097
+8,0,7,1.6896031,0.2908513,-0.23292185,1.3335884,0.4545921,0.34372616,-0.4165884,-0.42207703,0.60733074,1.2122438,0.8359352,0.5889715,-1.7851549,1.4177665,0.7491191,0.5095494,0.26941696,0.21018945,-0.880237,0.5611264,0.49656537,-0.8069379,1.1553068,-0.0126349665,-0.10190317,-1.7098439,-0.7785349,0.08457157,-1.0872965,-0.46575686,0.17051879,-1.0981392,1.2801948,-0.20283222,-0.37779662,-1.045028,0.8350199,-0.4449086,0.37728134,-1.3267673,-1.3647467,0.49804264,1.072318,-1.5584799,0.62113297,-1.7105299,0.936378,0.96202284,-0.5543043,-0.22153261,0.5113075,-0.5170268,0.11910198,-0.16186874,0.33424348,-0.8031384,1.6559254,0.47857162,0.5240287,-0.85450006,1.1488028,-0.12382944,-2.3143659,-1.7480177,0.34907168,0.3368916,-0.96712863,-1.1980493,1.7140709,-1.4368651,0.66559404,-0.27778557,-1.7404891,1.0585526,-0.3457661,-0.026605887,-1.9635391,2.3253458,0.7187862,-0.6399079,-2.111302,-0.7639529,-0.3719801,-0.0044646794,1.4715577,-1.4121687,0.046070594,0.049432546,-0.67208856,0.77220047,-0.78486633,1.6651013,-0.5048683,-1.0169152,0.19552465,0.6060275,1.0623702,0.15507454,-0.3246709,0.2723275,-0.9548158,-0.060560215,1.6082474,1.8419513,-0.6165263,0.5530547,0.53137124,-1.6408414,-0.083764516,-0.6223381,0.053751897,0.17250697,-0.5837834,-1.9890918,-2.1878116,-0.63507724,-1.3977062,-0.09614903,-0.99611247,-0.8769,0.8914697,1.3250514,-1.3865021,0.070221275,0.14449106,-0.4635786,0.22937016,-1.06633,-1.124088,0.17105861,0.38474694,1.002527,0.062628046,-1.4390727,-0.07946947,2.1103234,-1.7293836,1.1202102,-0.1935348,-1.0497704,-0.1828823,-0.8191538,-0.9199905,-0.0011521083,1.4552091,0.028564112,1.5781827,1.5662959,-0.37651274,-0.75987285,-0.96915853,-1.9078714,1.1331102,-0.114896916,0.53727424,2.130492,-2.0829363,0.2657546,0.7349715,-0.30037367,0.7455469,-1.2468303,-0.94141734,-0.96966463,-2.640029,-1.1223298,0.8763038,-0.96782756,-0.92439544,1.2501682,-1.4089605,-0.8066179,-0.87894547,0.5093231,0.8661666,-1.0657727,-0.44514516,0.16784558,-0.82460415,-0.62026,0.22096872,-0.5744515,-0.05113996,0.9549697,-0.7317686,0.2157382,-0.73088646,1.3564851,-0.56419,-0.3550247,-1.0255395,-0.57220864,-0.9036402,0.36711925,-0.09569928,0.6653445,-0.5859867,1.2525154,1.7926836,0.031228034
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
index 69bb2f81..7c1f56ff 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,7,-0.16720565,-0.95600957,0.28606528,0.6061069,-0.38458014,-0.20699821,-1.1377771,0.20950352,-0.13692231,-0.39809802,1.1865551,1.4303648,-1.0548229,0.47550696,0.9458503,2.1207294,2.1566584,-0.24435447,-0.64201504,1.7221577,0.931517,-0.7352646,0.34909317,0.7247565,-0.5341214,-0.38585183,-0.7807762,0.55336505,0.0011219969,1.0769404,0.17883597,-1.319804,1.3214653,-1.3146976,-0.71112573,-0.4876438,1.6924084,-0.5158393,1.3672885,-0.55709517,-0.8911706,-0.76211774,1.6342849,-1.2171471,0.21610762,-2.0004432,-0.90466094,-0.66080266,0.099175125,-1.5156087,-1.408444,-1.4207674,-1.6178052,1.158086,0.932027,1.5196209,-1.0533147,1.268581,0.0013256798,-1.9838752,-0.02037545,-0.6469344,-1.380692,-1.059281,0.5494309,-1.3961787,-0.15649295,-0.5929407,0.9686789,-0.40543148,1.0557822,0.8631479,-2.2666907,0.67865163,-0.9861262,-0.4633933,-0.5806146,1.3590292,-0.97610635,-1.6902065,0.015555903,-1.0601696,-0.6836848,-0.48810723,1.0633672,-2.1536193,0.27614263,-0.9428519,0.8829793,-0.13327233,-0.69135016,0.057943318,0.29546723,-1.8476412,1.6050954,-0.77256197,-0.34552777,-0.3999227,-0.91350764,-1.4197913,-1.4342732,0.054446727,0.38957366,1.6343316,-0.018683603,0.20149976,-0.51843685,-2.7988346,-0.23812824,0.3947416,0.7759951,0.26147768,0.7695459,-0.8783734,0.24983662,0.22944754,-0.50273734,0.17921257,-2.96453,0.84934795,0.8776484,0.17855342,-0.3165259,-0.49537578,-1.4034133,-0.5197445,0.65165275,-1.1936586,0.4376411,-0.3197974,-1.1554341,-0.09437472,-1.4360906,-0.666431,0.19690531,0.22096209,-0.77558017,-0.6480959,-1.0999116,1.1237246,0.32265925,-0.71968275,-2.2545595,-1.4951689,0.5562803,1.1043609,1.1391095,0.12772344,0.39196163,0.68036157,-0.48576477,-1.285276,0.81914747,0.30907387,-0.07296551,1.534054,0.30991098,-0.007137307,1.7125996,-0.8783491,-0.079967216,-2.1059916,-0.31304887,-0.31512117,-1.8248315,-1.5771934,0.5107034,0.7182178,-0.25824896,-0.45623496,-0.26445267,-0.9527643,-0.7283463,0.17178623,0.2119341,0.9493435,-1.3720604,0.15322246,0.70129305,-0.55172205,-0.15245205,0.06241885,-0.6320953,0.5529451,-0.4287735,1.0122429,-0.0774352,-0.28508043,-1.175873,0.21220754,-1.3312136,-1.4357239,0.21157016,0.5917681,-1.1742185,1.9612663,-1.2883048,0.06013697,1.073086,0.5143654
+9,0,7,-0.16720563,-0.95600957,0.28606522,0.6061068,-0.3845801,-0.20699826,-1.1377771,0.20950346,-0.13692231,-0.39809802,1.1865551,1.4303648,-1.0548229,0.47550696,0.9458502,2.1207294,2.1566582,-0.24435441,-0.64201504,1.7221576,0.93151706,-0.7352646,0.34909317,0.72475654,-0.53412133,-0.38585177,-0.7807762,0.55336505,0.0011219378,1.0769404,0.178836,-1.3198038,1.3214653,-1.3146976,-0.7111258,-0.48764378,1.6924084,-0.51583934,1.3672884,-0.5570951,-0.8911707,-0.7621178,1.6342849,-1.217147,0.21610765,-2.0004432,-0.9046609,-0.66080266,0.099175125,-1.5156087,-1.408444,-1.4207674,-1.6178051,1.1580858,0.932027,1.5196209,-1.0533147,1.2685812,0.0013257095,-1.983875,-0.02037545,-0.64693433,-1.3806919,-1.0592809,0.54943085,-1.3961786,-0.15649292,-0.5929407,0.9686789,-0.40543142,1.0557822,0.863148,-2.2666905,0.6786515,-0.98612607,-0.46339324,-0.5806146,1.359029,-0.97610635,-1.6902065,0.015555903,-1.0601696,-0.68368477,-0.4881073,1.0633672,-2.1536195,0.27614263,-0.94285184,0.8829792,-0.1332724,-0.69135016,0.057943374,0.29546717,-1.8476412,1.6050953,-0.77256185,-0.34552783,-0.39992267,-0.9135076,-1.4197912,-1.4342732,0.05444667,0.38957366,1.6343316,-0.018683603,0.20149979,-0.5184369,-2.7988343,-0.23812824,0.39474156,0.7759951,0.26147765,0.7695459,-0.8783734,0.24983662,0.22944757,-0.50273734,0.17921253,-2.96453,0.8493479,0.8776484,0.17855343,-0.3165259,-0.49537572,-1.4034132,-0.5197445,0.6516528,-1.1936586,0.4376411,-0.31979737,-1.1554341,-0.094374776,-1.4360906,-0.666431,0.19690529,0.22096215,-0.77558017,-0.6480959,-1.0999115,1.1237246,0.32265925,-0.7196827,-2.2545595,-1.4951689,0.5562803,1.1043608,1.1391095,0.12772338,0.39196157,0.68036157,-0.48576477,-1.285276,0.81914747,0.30907387,-0.07296551,1.534054,0.30991095,-0.007137307,1.7125996,-0.8783491,-0.07996723,-2.1059916,-0.3130489,-0.31512117,-1.8248315,-1.5771933,0.51070344,0.7182178,-0.2582489,-0.45623493,-0.2644527,-0.9527642,-0.7283463,0.17178611,0.21193407,0.9493434,-1.3720604,0.15322249,0.701293,-0.551722,-0.15245207,0.06241879,-0.6320952,0.5529451,-0.4287735,1.0122428,-0.07743517,-0.28508037,-1.175873,0.21220748,-1.3312136,-1.4357239,0.21157013,0.5917681,-1.1742185,1.9612663,-1.2883047,0.06013697,1.073086,0.5143654
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
index 9a165d17..d4e16b52 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,5,0.24902344,-0.003112793,-1.1171875,-0.2890625,1.21875,0.76953125,-0.56640625,1.03125,0.9609375,-0.14257812,0.66796875,2.828125,0.4453125,0.28320312,-0.91015625,-0.048583984,-0.67578125,0.5625,3.265625,1.6875,1.0078125,-0.90234375,0.63671875,0.2578125,0.028686523,0.7890625,-0.5703125,-0.44140625,-0.91796875,-0.875,0.095703125,0.05810547,0.16601562,1.1328125,-1.125,0.0028686523,0.2890625,0.22851562,0.19140625,0.015991211,-0.37695312,-0.42578125,-0.25195312,-1.078125,0.11767578,0.7578125,0.14355469,-1.484375,0.13183594,0.13085938,0.49804688,-1.546875,0.0703125,0.119140625,-1.3515625,0.052490234,-0.33789062,-0.44726562,1.2421875,-1.2578125,-0.06982422,-0.359375,2.21875,-0.40820312,-0.6953125,-1.15625,0.02746582,-1.4140625,-0.25976562,0.8125,0.044921875,1.40625,-0.53125,-0.78125,1.0859375,2.875,-0.91015625,-0.13671875,-2.328125,0.02319336,0.49609375,-0.06640625,0.546875,0.3125,0.044189453,-0.29492188,-1.15625,-1.34375,0.38671875,1.65625,1.421875,0.66796875,-1.625,1.640625,-0.014099121,-0.024658203,-0.32617188,0.546875,-0.76953125,-0.25195312,-0.16210938,0.24316406,1.0078125,-1.2421875,-1.4765625,0.27148438,-0.63671875,0.6953125,-0.45507812,-0.14746094,1.2109375,0.49804688,0.13867188,-0.24902344,1.4296875,-1.53125,-1.0,-0.4140625,0.81640625,0.18847656,-0.21484375,0.39648438,0.4765625,-0.12402344,-1.4765625,2.421875,-0.49804688,-1.265625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40820312,-1.6796875,0.546875,-1.3671875,-0.609375,-0.083496094,-0.7890625,1.4140625,-1.640625,-0.87890625,2.03125,-0.66796875,-0.37695312,-0.14160156,1.0,-0.76953125,0.390625,1.09375,-0.9375,0.54296875,-0.87890625,-1.0390625,-0.0859375,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.375,0.73046875,2.28125,0.064941406,0.07910156,-1.28125,-2.359375,1.7734375,0.033691406,0.37890625,-0.78515625,-0.3359375,0.7421875,-0.95703125,-0.1953125,-0.90234375,-1.109375,-0.69140625,0.6484375,-0.76953125,-0.053222656,0.41210938,1.109375,-1.4453125,0.9453125,-0.9609375,-1.2890625,1.796875,-0.93359375,-1.4296875,-0.5390625,0.43359375,-0.578125,-0.20507812,1.375,-0.421875,0.31054688,0.33398438,1.21875
-0,0,6,0.29882812,0.5,0.65625,-0.48046875,0.25976562,-0.025512695,0.6796875,-0.6171875,0.58984375,-0.6640625,0.68359375,0.095703125,1.6015625,-0.73046875,0.007873535,-0.3125,-0.19628906,-0.005218506,2.625,0.16503906,-0.15234375,-1.3984375,0.6640625,0.057373047,-0.703125,-1.0,0.29296875,-0.3125,-0.37890625,1.609375,0.64453125,-1.0546875,1.5078125,-1.4140625,0.19238281,0.20996094,0.06982422,-1.046875,-0.84375,0.8203125,-2.140625,0.29101562,-0.82421875,-0.6484375,-0.28125,-0.51953125,0.31445312,-0.984375,1.3515625,0.671875,-0.78515625,0.34179688,-0.9296875,1.5390625,0.20117188,-0.30273438,0.140625,-0.421875,-1.1484375,-0.83203125,1.3359375,0.50390625,-0.051757812,1.25,-0.19140625,0.4453125,-0.05908203,-1.796875,-1.265625,1.5390625,-0.6640625,0.578125,0.640625,0.064453125,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.012451172,-0.80078125,-0.66796875,0.5625,-0.18066406,-0.27734375,-0.609375,-1.2421875,-1.953125,1.484375,-0.5,1.1953125,0.953125,-0.9609375,0.328125,-1.109375,-0.74609375,-1.8984375,-0.75,-1.296875,0.04248047,-1.375,1.125,1.796875,-0.017578125,0.44335938,2.609375,-1.03125,0.58984375,0.28515625,-0.625,0.69921875,1.1484375,0.640625,-0.078125,0.46484375,0.30859375,-0.21484375,0.24707031,-0.06347656,1.296875,0.5390625,-0.6171875,1.1875,-0.90625,-1.1171875,1.1171875,-1.8203125,-1.40625,-0.18847656,-0.10107422,1.1171875,-0.609375,-0.54296875,-1.3515625,0.38671875,0.40234375,0.40234375,-0.40820312,0.21582031,-1.8828125,2.484375,0.57421875,-0.671875,2.34375,-0.24902344,-1.5234375,2.21875,-1.0703125,-1.34375,0.82421875,-0.9609375,0.20019531,1.0390625,-1.4140625,-0.32617188,-0.030151367,1.2578125,-0.8203125,1.0859375,0.32421875,-0.390625,-0.53515625,2.390625,1.7734375,-0.94921875,-0.7734375,-1.1484375,1.015625,1.1015625,-0.39453125,-0.30664062,-0.50390625,-1.734375,-0.9140625,-0.88671875,0.265625,0.98046875,-1.4453125,0.5078125,0.2578125,-1.9296875,-1.015625,1.171875,0.26953125,-1.5390625,-0.41992188,0.43359375,-0.20898438,-0.65625,-0.05883789,-1.7109375,1.7890625,1.1875,-0.53125,-1.0859375,0.8984375,1.796875,-0.34570312,1.359375,1.0703125
-0,0,7,0.66796875,1.21875,-1.203125,-1.3125,-0.51171875,-1.8984375,0.31445312,1.1875,-0.53125,-0.6796875,-0.05419922,-1.1796875,-1.828125,-0.33203125,-0.9375,1.0703125,-0.27929688,0.91796875,0.1640625,0.022216797,0.90234375,-0.23339844,-0.03466797,1.328125,-1.7265625,-0.96484375,-2.390625,0.6640625,1.984375,0.9765625,0.76953125,1.3984375,-1.5234375,-0.38671875,0.73828125,-0.020141602,0.9453125,1.3671875,-0.044921875,0.6171875,-2.15625,-0.27734375,0.072265625,1.75,0.7109375,2.03125,0.43359375,0.5859375,0.18847656,1.25,0.28515625,-0.23242188,-1.3515625,-0.80078125,-0.1328125,-0.43164062,-0.9765625,0.012329102,-0.064941406,-0.609375,-0.1640625,0.53515625,0.34765625,-0.3828125,1.3359375,0.23632812,0.1640625,0.06542969,-1.0625,0.59375,0.047851562,-0.16796875,0.390625,-0.828125,0.6796875,0.09082031,2.09375,-0.044677734,-0.42578125,0.03955078,1.2734375,0.609375,1.578125,0.9140625,-0.08691406,0.9765625,-1.328125,0.515625,-1.7421875,0.35351562,0.9609375,2.203125,1.7890625,-0.40820312,-0.068359375,-1.421875,-0.33203125,0.012512207,0.96484375,1.03125,0.87890625,-0.18457031,1.0234375,-0.17285156,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0068969727,-0.37890625,-0.22851562,2.546875,-0.88671875,-0.47851562,-0.890625,-0.35546875,-1.03125,0.99609375,0.6875,2.0625,-1.8828125,-0.86328125,-2.96875,-0.21582031,0.19042969,-0.98046875,1.2421875,0.17089844,-1.9765625,0.90234375,-0.37695312,0.67578125,-1.078125,-1.875,0.58984375,0.53515625,-0.546875,0.111816406,-2.1875,0.32226562,0.89453125,1.703125,1.21875,-0.62109375,-0.49414062,-1.625,1.078125,-0.37695312,0.12890625,-0.041015625,-1.359375,-0.7890625,1.4296875,0.19238281,-0.7109375,-0.6015625,1.0625,-0.37695312,1.90625,-2.0625,0.96875,-0.33007812,0.079589844,-1.296875,-0.1484375,-0.26953125,0.29296875,0.46679688,-0.58203125,-0.32421875,0.47265625,-1.109375,0.9921875,0.953125,0.25,0.984375,-1.546875,0.041015625,-0.103515625,1.0,0.53125,0.62109375,-0.87890625,-0.49804688,-0.32421875,-0.24511719,0.0625,-0.54296875,-1.046875,-1.2421875,0.31640625,-0.036132812,-0.86328125,1.2890625,0.119140625,1.921875,-0.23046875,0.09375,0.42382812
+0,0,5,0.25,-0.0029144287,-1.1171875,-0.29296875,1.234375,0.76953125,-0.5703125,1.0390625,0.96875,-0.13964844,0.67578125,2.8125,0.44726562,0.28320312,-0.90234375,-0.04638672,-0.67578125,0.5625,3.265625,1.6875,1.015625,-0.90625,0.640625,0.2578125,0.030883789,0.79296875,-0.56640625,-0.4375,-0.921875,-0.87890625,0.09814453,0.05859375,0.16503906,1.140625,-1.125,0.0069274902,0.2890625,0.23144531,0.19335938,0.017700195,-0.37890625,-0.42578125,-0.25585938,-1.0703125,0.119628906,0.7578125,0.14355469,-1.484375,0.13476562,0.1328125,0.5,-1.5390625,0.068847656,0.119628906,-1.3515625,0.05126953,-0.33789062,-0.44726562,1.2265625,-1.2734375,-0.068359375,-0.36132812,2.1875,-0.40820312,-0.6953125,-1.1640625,0.025756836,-1.3984375,-0.2578125,0.8203125,0.047607422,1.4140625,-0.53125,-0.78515625,1.09375,2.859375,-0.90234375,-0.13671875,-2.34375,0.025390625,0.49414062,-0.068359375,0.54296875,0.31054688,0.04711914,-0.296875,-1.15625,-1.3359375,0.390625,1.65625,1.421875,0.66796875,-1.625,1.640625,-0.013427734,-0.026367188,-0.32617188,0.5390625,-0.765625,-0.25195312,-0.16503906,0.24121094,1.0078125,-1.234375,-1.4609375,0.27148438,-0.640625,0.6953125,-0.45507812,-0.14746094,1.203125,0.5,0.14160156,-0.24804688,1.4375,-1.5390625,-1.0,-0.41992188,0.8203125,0.18652344,-0.21972656,0.39257812,0.47460938,-0.125,-1.484375,2.40625,-0.49804688,-1.2890625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40625,-1.6953125,0.546875,-1.3671875,-0.62109375,-0.08544922,-0.78515625,1.4140625,-1.640625,-0.875,2.03125,-0.671875,-0.375,-0.14160156,0.99609375,-0.76953125,0.38671875,1.1015625,-0.94140625,0.55078125,-0.875,-1.03125,-0.088378906,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.3984375,0.73046875,2.28125,0.068847656,0.079589844,-1.28125,-2.375,1.7734375,0.033935547,0.3828125,-0.79296875,-0.33398438,0.7421875,-0.94921875,-0.19433594,-0.90234375,-1.125,-0.6953125,0.64453125,-0.765625,-0.052490234,0.4140625,1.109375,-1.4453125,0.9375,-0.9609375,-1.2890625,1.796875,-0.9375,-1.4140625,-0.5390625,0.43554688,-0.58203125,-0.20507812,1.3828125,-0.4140625,0.30859375,0.33789062,1.2265625
+0,0,6,0.29882812,0.50390625,0.65625,-0.48828125,0.25976562,-0.027954102,0.671875,-0.6015625,0.59375,-0.66796875,0.68359375,0.09765625,1.609375,-0.73046875,0.0064086914,-0.31640625,-0.19628906,0.00049972534,2.625,0.16210938,-0.15332031,-1.3984375,0.66796875,0.05883789,-0.703125,-1.0078125,0.296875,-0.3125,-0.3828125,1.609375,0.64453125,-1.0703125,1.5078125,-1.40625,0.19335938,0.20996094,0.06640625,-1.046875,-0.84765625,0.8203125,-2.15625,0.29101562,-0.828125,-0.6484375,-0.27539062,-0.51953125,0.3125,-0.97265625,1.3359375,0.67578125,-0.78515625,0.34570312,-0.93359375,1.5390625,0.20117188,-0.30078125,0.13769531,-0.42382812,-1.140625,-0.83203125,1.3515625,0.5078125,-0.05419922,1.2578125,-0.18945312,0.4453125,-0.06201172,-1.8046875,-1.265625,1.546875,-0.66796875,0.5859375,0.64453125,0.061279297,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.015563965,-0.80859375,-0.6640625,0.5625,-0.1796875,-0.27734375,-0.609375,-1.234375,-1.9375,1.484375,-0.5078125,1.1953125,0.94921875,-0.96484375,0.328125,-1.1015625,-0.73828125,-1.921875,-0.75390625,-1.296875,0.042236328,-1.375,1.125,1.796875,-0.018066406,0.44335938,2.609375,-1.0390625,0.59375,0.28710938,-0.62109375,0.703125,1.1484375,0.64453125,-0.075683594,0.46484375,0.31054688,-0.21679688,0.24121094,-0.06591797,1.3046875,0.54296875,-0.61328125,1.1953125,-0.91015625,-1.1171875,1.1171875,-1.8125,-1.3984375,-0.19140625,-0.10058594,1.109375,-0.609375,-0.54296875,-1.359375,0.3828125,0.40234375,0.39453125,-0.40429688,0.21484375,-1.875,2.5,0.5703125,-0.67578125,2.328125,-0.25,-1.5234375,2.21875,-1.0859375,-1.34375,0.82421875,-0.9609375,0.203125,1.03125,-1.4140625,-0.32421875,-0.034423828,1.2578125,-0.82421875,1.09375,0.31835938,-0.39257812,-0.53515625,2.390625,1.78125,-0.953125,-0.7734375,-1.1484375,1.015625,1.1015625,-0.39453125,-0.30859375,-0.5,-1.7265625,-0.9140625,-0.890625,0.265625,0.98046875,-1.4375,0.50390625,0.26171875,-1.921875,-1.015625,1.171875,0.2734375,-1.5390625,-0.41796875,0.43164062,-0.20996094,-0.65234375,-0.05810547,-1.703125,1.78125,1.1796875,-0.53125,-1.0859375,0.90234375,1.796875,-0.3515625,1.3671875,1.0703125
+0,0,7,0.66796875,1.21875,-1.203125,-1.3125,-0.515625,-1.90625,0.31640625,1.1875,-0.52734375,-0.67578125,-0.053466797,-1.203125,-1.828125,-0.33203125,-0.9296875,1.0625,-0.27734375,0.91796875,0.1640625,0.020629883,0.90234375,-0.234375,-0.03466797,1.3359375,-1.71875,-0.96484375,-2.375,0.6640625,1.9921875,0.9765625,0.76953125,1.3984375,-1.5234375,-0.38671875,0.734375,-0.017211914,0.9375,1.3671875,-0.049804688,0.62109375,-2.15625,-0.27539062,0.072753906,1.7578125,0.71484375,2.015625,0.43164062,0.59765625,0.1875,1.25,0.28320312,-0.22949219,-1.3515625,-0.796875,-0.13183594,-0.43359375,-0.98046875,0.012390137,-0.06347656,-0.61328125,-0.1640625,0.53515625,0.34960938,-0.38085938,1.328125,0.23535156,0.16503906,0.06347656,-1.0703125,0.59765625,0.049316406,-0.171875,0.390625,-0.8359375,0.6796875,0.08544922,2.09375,-0.043945312,-0.42773438,0.03857422,1.2734375,0.609375,1.5703125,0.9140625,-0.08544922,0.98828125,-1.328125,0.515625,-1.7421875,0.3515625,0.95703125,2.21875,1.796875,-0.41015625,-0.068847656,-1.40625,-0.328125,0.013549805,0.96875,1.0390625,0.875,-0.18457031,1.03125,-0.17382812,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0077819824,-0.37890625,-0.22558594,2.546875,-0.8828125,-0.4765625,-0.88671875,-0.35351562,-1.0390625,0.98828125,0.68359375,2.046875,-1.875,-0.86328125,-2.953125,-0.21582031,0.18847656,-0.98046875,1.25,0.17382812,-1.9765625,0.90234375,-0.37695312,0.67578125,-1.0703125,-1.8828125,0.58984375,0.53515625,-0.55078125,0.111816406,-2.1875,0.32421875,0.89453125,1.7109375,1.2109375,-0.6171875,-0.49804688,-1.625,1.0703125,-0.37695312,0.13085938,-0.041992188,-1.3515625,-0.7890625,1.4375,0.19433594,-0.7109375,-0.6015625,1.0625,-0.37695312,1.90625,-2.0625,0.9609375,-0.32617188,0.078125,-1.3046875,-0.15039062,-0.26953125,0.29296875,0.46679688,-0.5859375,-0.32421875,0.46875,-1.1015625,0.984375,0.953125,0.25,0.99609375,-1.546875,0.041992188,-0.1015625,1.0,0.53515625,0.62109375,-0.8828125,-0.50390625,-0.32617188,-0.24804688,0.064453125,-0.5390625,-1.0546875,-1.234375,0.31640625,-0.037841797,-0.87890625,1.296875,0.1171875,1.8984375,-0.2265625,0.09667969,0.42578125
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
index f802b0fc..a0974d2a 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,5,1.0546875,0.89453125,-0.546875,0.82421875,-0.65234375,-0.37695312,0.7734375,1.6484375,0.55859375,-1.6171875,-0.53515625,0.546875,0.7578125,-0.09765625,-1.4453125,0.83203125,-0.057617188,0.5859375,0.33789062,-0.37304688,0.83984375,-0.93359375,-0.035888672,-0.9140625,0.16015625,-0.9453125,-1.5078125,-0.14550781,-1.1484375,-0.8828125,1.3203125,-1.328125,1.71875,0.7578125,0.22753906,-1.578125,-0.12597656,-1.7265625,-0.7265625,-1.1328125,-1.2421875,-1.3046875,-1.28125,0.015991211,-1.875,0.92578125,0.48046875,1.015625,1.125,1.546875,-1.3125,-0.12792969,-0.73046875,1.7578125,-0.671875,0.5078125,1.15625,0.66796875,-0.15429688,-0.22753906,0.05419922,0.51171875,0.08251953,-0.12109375,0.7265625,-0.578125,-0.8984375,-0.40625,-0.703125,0.25195312,1.59375,-1.8125,0.57421875,-1.046875,2.296875,2.59375,1.53125,0.82421875,-0.34375,-1.125,-0.38671875,-1.25,-0.42578125,-1.2734375,0.9375,-1.1875,1.28125,0.59765625,-0.30664062,1.4921875,2.234375,-1.34375,0.16992188,0.734375,-0.4609375,0.5703125,0.18554688,-0.6796875,-0.21484375,0.58203125,-1.5703125,1.15625,1.03125,-0.029785156,2.0,-0.55859375,0.40234375,0.57421875,0.8828125,1.5859375,-0.31835938,0.014892578,0.296875,-0.609375,1.0,-0.29101562,-1.6875,0.17382812,-0.15136719,-0.28515625,-0.66796875,-0.41015625,0.22363281,-0.024902344,-0.578125,0.58984375,0.57421875,-0.58984375,-0.5,-0.515625,-0.37890625,0.2109375,-0.25585938,-1.28125,-1.75,-0.46679688,-0.609375,0.37890625,-0.4921875,-2.34375,2.171875,-0.66796875,-0.51953125,-1.9375,-0.30664062,-0.19238281,-0.7265625,-1.09375,0.609375,3.640625,-0.44726562,1.0546875,-1.796875,0.828125,0.19726562,-0.984375,0.14550781,1.203125,0.29492188,-0.8515625,0.2578125,0.36914062,-0.029174805,0.12451172,-0.8359375,0.53125,0.1484375,0.83203125,1.96875,-2.34375,-0.33398438,0.29882812,-0.4140625,-0.25195312,0.27734375,1.15625,1.0078125,1.203125,-0.42773438,0.3828125,-0.828125,1.4375,0.87109375,0.13476562,0.32226562,-1.203125,1.4375,0.91796875,-1.3203125,-0.19238281,-0.40429688,-0.828125,0.8125,0.29296875,-1.6796875,1.109375,0.052001953,-0.037109375,-0.35351562,-0.3828125
-1,0,6,-0.5546875,1.2734375,-1.359375,-1.5078125,0.31054688,-1.9609375,0.34179688,1.6328125,1.59375,-0.58203125,0.21679688,1.03125,-0.953125,0.515625,-1.2109375,0.53515625,0.26953125,0.22167969,-0.37304688,-0.07128906,0.33984375,-0.390625,1.328125,-0.036621094,-0.73828125,-0.9296875,-2.421875,1.1015625,1.578125,-0.47851562,1.375,0.20800781,-1.125,0.1484375,1.5703125,-1.09375,0.6640625,0.87109375,-0.77734375,1.0546875,-1.5546875,0.43164062,-0.21777344,-0.265625,-0.3203125,2.609375,-0.060058594,1.8125,1.15625,1.6015625,-1.0234375,1.078125,-0.63671875,-0.09033203,-0.828125,-0.088378906,-0.51953125,0.484375,-1.109375,1.3984375,0.39453125,0.9453125,-0.15917969,0.609375,0.546875,-1.1953125,-0.12451172,0.71875,-1.6484375,1.2578125,0.53125,0.78125,2.0625,-0.5390625,0.5625,2.078125,0.5078125,0.703125,-0.8359375,-0.40429688,0.48242188,-0.36328125,0.765625,-0.13183594,-0.12011719,0.9296875,-1.421875,-0.26757812,-1.15625,-0.68359375,0.58203125,1.5390625,0.99609375,-0.25976562,0.78515625,-1.875,-0.5390625,-0.25976562,0.41015625,-0.5625,0.14648438,0.4921875,0.6953125,0.15722656,-0.765625,0.06347656,-1.046875,-0.40039062,0.044677734,0.5703125,-1.6484375,-0.067871094,1.6640625,-2.3125,0.64453125,-0.14941406,-1.578125,-0.31054688,0.921875,0.083496094,1.1796875,-2.359375,0.028686523,-1.3671875,0.37890625,0.97265625,-0.25976562,-1.359375,1.03125,-0.24902344,-1.3203125,-0.91015625,-0.5390625,0.5859375,-1.90625,1.5390625,1.1953125,-0.19335938,0.9140625,-2.03125,1.3046875,0.12695312,0.6484375,0.83203125,-1.234375,-1.3359375,-0.81640625,-1.3046875,-1.4609375,-0.6953125,0.60546875,-0.61328125,-0.48242188,0.87109375,-0.73828125,1.8203125,0.5625,0.39453125,0.4609375,0.29492188,-0.26757812,1.1171875,0.7265625,0.36523438,0.02722168,-0.7265625,-0.38671875,0.76171875,-0.49414062,-2.0,-1.03125,-0.060791016,-1.5859375,-0.6953125,0.8515625,-0.12402344,1.0703125,0.84375,-0.31445312,0.020874023,0.08251953,-1.1171875,0.30664062,0.16894531,0.3359375,0.06591797,-0.83203125,-0.84765625,-1.7734375,-0.99609375,0.29101562,0.8203125,-0.7265625,0.6796875,-0.81640625,2.625,2.03125,0.100097656,-0.8984375,-0.48632812
-1,0,7,1.2578125,0.5703125,-0.98828125,0.10546875,-1.359375,-0.1171875,0.61328125,1.0078125,-1.1640625,-1.4609375,-0.21484375,-0.38085938,-0.4296875,0.453125,-1.75,-0.8203125,-1.0625,0.2890625,-0.953125,-0.53125,0.59375,-0.22070312,-0.578125,-0.46289062,-0.10107422,-2.1875,-2.390625,-0.43554688,-0.014282227,0.19238281,-0.3828125,-1.4453125,0.28710938,0.87890625,-0.32617188,-0.546875,-0.09814453,-2.375,0.44921875,-0.083496094,0.41210938,-0.21386719,-0.86328125,1.2421875,0.94140625,0.34960938,-1.0234375,0.6015625,-0.019897461,1.171875,0.16308594,-0.20117188,-1.65625,0.70703125,-0.90625,-0.671875,-0.640625,1.265625,-1.0546875,-1.0234375,-0.28125,0.765625,1.2734375,0.7734375,1.921875,-0.3515625,-0.28125,-0.5078125,-1.9375,0.49804688,1.2109375,-0.36523438,1.65625,-0.69140625,0.890625,0.82421875,2.515625,1.2265625,0.60546875,0.10253906,-0.4140625,-0.38476562,0.375,-1.3359375,-0.28125,1.0,0.32421875,-0.60546875,0.23828125,0.984375,0.5390625,1.34375,0.22363281,0.578125,-0.9609375,-1.2890625,-0.7421875,-0.13378906,1.1484375,1.0859375,0.390625,1.25,-0.30078125,-1.2421875,1.2734375,-1.078125,-0.7109375,0.69140625,0.41210938,-0.33398438,-0.9609375,1.0078125,1.578125,0.26953125,0.118652344,0.47070312,-0.30859375,0.515625,0.12451172,1.921875,0.30664062,-2.09375,-1.1484375,-0.6015625,-0.055664062,-1.25,0.17285156,-1.046875,0.0017166138,-0.16894531,0.421875,-0.5703125,-0.19140625,-0.022949219,-1.046875,2.453125,-0.91015625,-1.703125,1.4140625,-2.3125,0.64453125,-0.671875,-0.28320312,-0.65234375,0.875,0.31054688,0.375,0.20019531,2.1875,3.578125,0.41992188,0.265625,-1.6171875,1.0234375,-0.008728027,0.29492188,0.66015625,1.140625,-1.359375,0.62890625,0.9765625,1.734375,0.8046875,-0.62109375,-0.5078125,-0.546875,-0.34765625,1.6796875,0.51953125,-1.7578125,-0.61328125,-0.93359375,0.23242188,-0.55078125,0.703125,1.390625,-0.3125,1.1171875,1.328125,-0.14257812,0.067871094,0.123535156,1.359375,-0.640625,0.48828125,-0.2890625,0.64453125,1.4140625,-0.0038909912,-1.5234375,0.21582031,-1.6015625,0.54296875,-1.7265625,-0.80859375,1.2109375,0.30664062,-1.0859375,-1.2421875,-0.35742188
+1,0,5,1.0546875,0.89453125,-0.55078125,0.83203125,-0.65234375,-0.37890625,0.77734375,1.6484375,0.5625,-1.6171875,-0.53515625,0.54296875,0.76953125,-0.1015625,-1.4453125,0.828125,-0.05859375,0.58203125,0.34179688,-0.37304688,0.83984375,-0.9375,-0.032226562,-0.9140625,0.15820312,-0.953125,-1.5078125,-0.14160156,-1.1484375,-0.87890625,1.3203125,-1.3359375,1.71875,0.76171875,0.22167969,-1.578125,-0.122558594,-1.7265625,-0.7265625,-1.125,-1.2421875,-1.296875,-1.28125,0.017822266,-1.8828125,0.9296875,0.48046875,1.0078125,1.125,1.546875,-1.3046875,-0.13378906,-0.73046875,1.7421875,-0.671875,0.5078125,1.15625,0.66796875,-0.15234375,-0.23046875,0.05883789,0.51171875,0.08251953,-0.12207031,0.72265625,-0.5859375,-0.90625,-0.40429688,-0.7109375,0.25195312,1.5859375,-1.8125,0.57421875,-1.0390625,2.296875,2.609375,1.5390625,0.82421875,-0.34179688,-1.1328125,-0.38085938,-1.25,-0.42578125,-1.2734375,0.9296875,-1.1875,1.28125,0.59765625,-0.3046875,1.4921875,2.25,-1.3359375,0.16992188,0.734375,-0.45703125,0.57421875,0.19042969,-0.67578125,-0.21484375,0.5859375,-1.5703125,1.15625,1.03125,-0.02734375,1.9921875,-0.55859375,0.40429688,0.57421875,0.87890625,1.59375,-0.31835938,0.0119018555,0.296875,-0.61328125,1.0,-0.2890625,-1.6953125,0.16992188,-0.15234375,-0.28125,-0.66796875,-0.40039062,0.22363281,-0.026977539,-0.57421875,0.5859375,0.5703125,-0.59375,-0.48828125,-0.515625,-0.38476562,0.2109375,-0.25195312,-1.28125,-1.7578125,-0.46484375,-0.609375,0.37695312,-0.49609375,-2.328125,2.15625,-0.66796875,-0.5234375,-1.9453125,-0.30859375,-0.19140625,-0.7265625,-1.09375,0.61328125,3.640625,-0.44921875,1.0625,-1.7890625,0.828125,0.19335938,-0.9921875,0.14355469,1.2109375,0.29492188,-0.8515625,0.2578125,0.36523438,-0.029418945,0.12792969,-0.83984375,0.53125,0.15625,0.8359375,1.9765625,-2.328125,-0.33203125,0.29882812,-0.41601562,-0.25195312,0.27539062,1.1328125,1.0078125,1.203125,-0.42773438,0.38476562,-0.83203125,1.4375,0.875,0.13183594,0.32617188,-1.203125,1.4296875,0.91015625,-1.328125,-0.1953125,-0.41015625,-0.8203125,0.8203125,0.29101562,-1.6875,1.109375,0.049072266,-0.037353516,-0.34960938,-0.3828125
+1,0,6,-0.5546875,1.265625,-1.3671875,-1.5078125,0.30664062,-1.9453125,0.34375,1.6328125,1.609375,-0.58984375,0.21875,1.03125,-0.9453125,0.515625,-1.21875,0.53515625,0.2734375,0.21972656,-0.37109375,-0.07373047,0.33984375,-0.39453125,1.3359375,-0.0390625,-0.7421875,-0.921875,-2.421875,1.1015625,1.578125,-0.4765625,1.3671875,0.21191406,-1.1171875,0.14941406,1.5703125,-1.0859375,0.6640625,0.86328125,-0.7734375,1.046875,-1.546875,0.43164062,-0.22070312,-0.26953125,-0.32421875,2.625,-0.06201172,1.8125,1.140625,1.609375,-1.03125,1.0859375,-0.640625,-0.09033203,-0.82421875,-0.08984375,-0.51953125,0.48828125,-1.109375,1.40625,0.3984375,0.94140625,-0.16015625,0.609375,0.54296875,-1.203125,-0.12695312,0.71875,-1.65625,1.265625,0.52734375,0.7890625,2.046875,-0.53515625,0.55859375,2.09375,0.51171875,0.703125,-0.8359375,-0.40039062,0.48242188,-0.36523438,0.76171875,-0.13378906,-0.119628906,0.94140625,-1.421875,-0.27148438,-1.1640625,-0.6796875,0.5859375,1.53125,0.98046875,-0.25976562,0.7890625,-1.875,-0.53515625,-0.26171875,0.40625,-0.55859375,0.14160156,0.49414062,0.6953125,0.15527344,-0.76953125,0.068847656,-1.046875,-0.40039062,0.041259766,0.56640625,-1.6484375,-0.068359375,1.6640625,-2.3125,0.6484375,-0.14941406,-1.5625,-0.3125,0.91796875,0.08203125,1.171875,-2.34375,0.026367188,-1.359375,0.37890625,0.96484375,-0.26171875,-1.359375,1.0390625,-0.24902344,-1.328125,-0.91796875,-0.5390625,0.578125,-1.8984375,1.5390625,1.1875,-0.1953125,0.91015625,-2.03125,1.296875,0.123535156,0.6484375,0.8359375,-1.2421875,-1.3359375,-0.81640625,-1.3203125,-1.453125,-0.703125,0.60546875,-0.61328125,-0.48242188,0.875,-0.73828125,1.828125,0.5625,0.39453125,0.45898438,0.29101562,-0.27148438,1.1171875,0.73046875,0.36914062,0.026733398,-0.72265625,-0.38671875,0.7578125,-0.49414062,-1.9921875,-1.03125,-0.05883789,-1.578125,-0.6953125,0.8515625,-0.13183594,1.0703125,0.84375,-0.31445312,0.021362305,0.08154297,-1.1171875,0.30664062,0.16894531,0.33203125,0.06640625,-0.82421875,-0.8515625,-1.7734375,-1.0,0.2890625,0.81640625,-0.7265625,0.68359375,-0.81640625,2.640625,2.03125,0.09814453,-0.8984375,-0.48242188
+1,0,7,1.2578125,0.5703125,-0.9921875,0.103515625,-1.359375,-0.11767578,0.609375,1.0078125,-1.1640625,-1.4453125,-0.21582031,-0.37890625,-0.42382812,0.453125,-1.7421875,-0.8203125,-1.0625,0.28320312,-0.953125,-0.53125,0.59375,-0.21972656,-0.57421875,-0.46679688,-0.107421875,-2.171875,-2.390625,-0.4375,-0.008850098,0.19433594,-0.38085938,-1.4453125,0.2890625,0.87890625,-0.328125,-0.546875,-0.09423828,-2.390625,0.45507812,-0.08203125,0.40820312,-0.20996094,-0.859375,1.2421875,0.9375,0.3515625,-1.0234375,0.60546875,-0.025146484,1.1640625,0.16015625,-0.19824219,-1.65625,0.70703125,-0.90625,-0.67578125,-0.63671875,1.25,-1.0625,-1.03125,-0.28125,0.765625,1.265625,0.7734375,1.921875,-0.34960938,-0.28125,-0.5078125,-1.9296875,0.5,1.21875,-0.36132812,1.6640625,-0.6953125,0.8828125,0.828125,2.53125,1.234375,0.609375,0.099609375,-0.41210938,-0.3828125,0.37109375,-1.3359375,-0.27734375,0.99609375,0.31835938,-0.60546875,0.24121094,0.984375,0.5390625,1.34375,0.21972656,0.58203125,-0.95703125,-1.2890625,-0.734375,-0.13476562,1.1484375,1.0859375,0.38085938,1.2578125,-0.296875,-1.2265625,1.2578125,-1.078125,-0.71484375,0.69140625,0.41015625,-0.33398438,-0.96875,1.015625,1.578125,0.26367188,0.119628906,0.46679688,-0.3046875,0.515625,0.122558594,1.921875,0.30664062,-2.109375,-1.1484375,-0.59765625,-0.056152344,-1.25,0.171875,-1.046875,0.001335144,-0.16796875,0.421875,-0.5703125,-0.18945312,-0.022338867,-1.046875,2.4375,-0.9140625,-1.6875,1.40625,-2.3125,0.64453125,-0.671875,-0.28320312,-0.65234375,0.8671875,0.30664062,0.375,0.20214844,2.203125,3.59375,0.41992188,0.26367188,-1.609375,1.015625,-0.0071411133,0.29492188,0.65625,1.1484375,-1.359375,0.62890625,0.97265625,1.75,0.8046875,-0.62109375,-0.51171875,-0.55078125,-0.3515625,1.6796875,0.5234375,-1.7578125,-0.62109375,-0.94140625,0.23242188,-0.55078125,0.703125,1.3984375,-0.3125,1.1171875,1.3125,-0.14257812,0.064941406,0.122558594,1.3671875,-0.64453125,0.49023438,-0.2890625,0.640625,1.4140625,-0.0033111572,-1.515625,0.21582031,-1.6015625,0.5390625,-1.7421875,-0.81640625,1.2109375,0.30273438,-1.09375,-1.234375,-0.36328125
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
index df728e04..6031742c 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,5,1.2109375,-0.096191406,-1.3359375,-0.6015625,0.9609375,-1.484375,-0.8203125,3.140625,-0.5234375,-1.6640625,0.0234375,1.5859375,-0.53515625,-0.75,-0.5703125,1.75,-0.27148438,-1.40625,0.059570312,0.48632812,-1.375,-0.6796875,-0.44921875,1.2421875,-0.095703125,0.33007812,-2.484375,-0.15527344,-0.890625,-0.06933594,-0.020141602,0.7890625,0.6953125,0.7734375,-0.07373047,-1.0546875,1.25,-0.171875,0.515625,0.6953125,-0.06201172,-0.008300781,0.11279297,-0.78125,-0.115234375,0.65625,-0.46484375,0.96484375,-0.43945312,2.21875,-0.609375,-2.234375,-1.3359375,-0.048339844,-0.15039062,-0.33398438,0.921875,0.29882812,0.40234375,-1.046875,-0.3671875,1.4765625,-0.12988281,0.53515625,-0.26953125,-0.030273438,-0.00074005127,1.4765625,0.16796875,-0.11328125,-0.4140625,-0.2890625,0.9140625,-0.67578125,1.6328125,2.140625,0.62890625,0.26953125,-0.24414062,-0.75390625,1.671875,0.7578125,0.53515625,-0.6328125,-0.44140625,-1.1484375,1.328125,1.84375,0.50390625,1.65625,1.734375,-0.7734375,-1.046875,1.375,-0.28320312,0.033691406,-1.7265625,-0.73828125,-1.0859375,-0.21875,-1.0703125,-0.24804688,1.4296875,0.94140625,-0.46875,0.25390625,-1.640625,-0.62109375,-0.28515625,0.016479492,-1.140625,-1.9453125,1.421875,-2.265625,0.84765625,0.27539062,-2.0625,1.2734375,-0.30273438,-1.21875,1.2578125,-0.47265625,-0.92578125,-0.36523438,0.4140625,2.71875,-0.43945312,-0.41210938,0.4453125,-1.8671875,-1.484375,-1.640625,0.3671875,-1.1015625,-0.8984375,-0.012451172,1.3203125,0.34960938,0.43359375,-1.3515625,1.453125,-1.8203125,-0.6328125,-0.15917969,0.16210938,0.47070312,-0.32226562,0.83984375,-0.73828125,0.828125,-0.6171875,-0.019165039,0.80859375,-0.453125,0.046142578,-0.76171875,0.31835938,0.014770508,1.6640625,1.015625,-1.1484375,1.859375,1.2890625,0.8203125,-0.57421875,0.77734375,0.17285156,1.3203125,1.2578125,0.7734375,-1.2890625,-0.16796875,0.06640625,1.09375,-1.28125,0.43359375,0.35351562,-0.6953125,-0.056884766,-1.171875,0.46679688,0.38476562,0.3046875,-1.0859375,0.22265625,-0.24023438,-1.40625,0.0134887695,-0.84375,-1.34375,0.21386719,1.4375,-0.11621094,-0.42773438,0.123046875,1.1953125,0.49804688,0.6015625,-1.0703125,0.45703125
-2,0,6,0.31445312,-0.33398438,-0.72265625,-0.44140625,0.40625,-0.546875,-0.005645752,1.578125,-0.13769531,-0.8359375,-0.5390625,-0.37695312,1.2265625,0.5,-1.6484375,0.4453125,-0.51953125,-1.984375,0.42773438,-0.6953125,0.5625,-0.328125,-0.27734375,-0.06689453,-1.328125,-1.8125,-1.625,-0.088378906,-2.0625,0.48828125,-0.68359375,-1.1171875,0.55859375,0.076660156,0.546875,1.671875,0.625,-1.453125,-1.8671875,1.1875,-1.4765625,0.34960938,-0.54296875,-0.04272461,0.23144531,1.0625,0.5546875,-0.55078125,0.2421875,1.734375,-2.109375,-1.3203125,-1.4140625,0.7265625,-1.96875,0.6171875,0.12792969,-0.671875,-0.19238281,-2.171875,1.1875,-0.16015625,0.42773438,-0.07324219,0.17675781,0.05517578,-1.0859375,0.18652344,-0.67578125,2.0,0.90625,-1.609375,-0.45898438,-0.5,0.045410156,1.921875,1.109375,1.1015625,-0.61328125,-0.8828125,1.5546875,0.32226562,0.27539062,-0.5078125,-0.671875,-0.028320312,0.28125,0.9609375,-0.6640625,0.6796875,2.109375,-0.17285156,0.1484375,0.41992188,-1.3984375,-0.19824219,-0.3046875,-0.53125,1.421875,-0.33007812,-2.25,0.140625,2.21875,1.15625,-0.46875,-0.22851562,1.4765625,-0.9296875,0.40429688,0.7109375,0.27539062,0.35742188,0.859375,-1.4375,0.30859375,-0.47070312,0.18945312,-0.26953125,-0.66015625,-0.65625,1.3359375,-0.55078125,0.4140625,-1.1875,-0.24414062,1.2734375,-0.44921875,-2.3125,-0.006286621,-0.04321289,0.6796875,-0.061035156,0.19140625,-1.3046875,-1.328125,-0.20019531,-0.35351562,-0.46875,-0.01928711,-1.8046875,3.09375,-0.765625,0.390625,-0.052246094,-0.067871094,1.8359375,-1.21875,-0.016967773,-1.9296875,1.921875,0.23535156,1.140625,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.43359375,1.625,1.4453125,0.66796875,-0.72265625,1.34375,1.3671875,0.49609375,-1.3359375,0.53515625,-1.1171875,-0.33789062,2.375,-0.25390625,0.22949219,-0.1484375,-0.66796875,-0.31445312,0.12011719,0.453125,-0.94140625,-0.91796875,0.9296875,-0.5859375,-0.17871094,-0.107421875,1.2734375,0.49023438,0.66015625,-0.30078125,1.3203125,0.8203125,-0.5390625,-1.046875,-0.23730469,0.53515625,2.15625,0.02746582,0.828125,-0.15429688,1.4140625,-0.2109375,1.2265625,0.26953125
-2,0,7,0.7734375,0.110839844,-0.36914062,-0.22363281,0.2109375,-0.76171875,-0.26953125,1.4140625,0.5546875,-1.09375,0.5859375,1.421875,0.44335938,0.55078125,-1.59375,0.796875,0.18847656,-1.203125,-0.11621094,0.1171875,0.9921875,-0.53515625,-0.3984375,0.57421875,-0.29492188,-0.7265625,-0.51171875,0.061035156,-1.0078125,-0.061035156,-0.68359375,-0.51171875,1.796875,-0.453125,1.28125,1.59375,0.828125,-0.15820312,-2.21875,1.7265625,-1.8359375,0.37695312,-0.59765625,-0.049316406,0.62890625,0.81640625,-0.3359375,-0.1328125,-0.4453125,1.90625,-1.3671875,-0.10986328,-0.6875,-0.08984375,-0.796875,0.71484375,0.6171875,0.60546875,-0.079589844,-1.25,1.609375,-0.21484375,1.8671875,-0.34375,-1.7421875,0.8515625,-1.171875,0.53125,-0.52734375,1.265625,-0.10205078,-1.171875,-0.099609375,0.609375,-0.15625,1.9375,1.390625,0.38867188,-0.99609375,-1.46875,1.390625,-0.20019531,0.59765625,-1.2890625,-1.375,-0.33789062,0.19042969,1.40625,-0.8125,0.359375,1.9609375,-0.03515625,-0.8203125,0.62109375,-1.5390625,0.94140625,-0.17578125,0.88671875,0.73828125,0.9921875,-2.46875,0.011230469,2.0,1.03125,-1.390625,0.35351562,0.90625,-1.6640625,-1.0703125,0.18066406,0.578125,-0.36328125,1.546875,-1.5078125,-0.36132812,-1.21875,0.953125,-0.04736328,-0.3046875,-1.296875,2.125,0.44140625,-0.6015625,-1.2578125,0.60546875,0.41992188,-0.73046875,-2.171875,0.26367188,-0.36523438,0.1796875,-1.171875,-0.54296875,-0.79296875,-1.8515625,1.6640625,0.30664062,-0.625,-0.5390625,-1.3515625,1.8125,-1.6953125,-0.6875,-0.3828125,-0.17773438,1.3125,-0.5859375,0.5234375,-2.6875,-0.1171875,0.2578125,1.1328125,0.69921875,-1.03125,-0.23046875,-0.77734375,0.8984375,0.4765625,1.1171875,0.55078125,-0.41015625,0.36523438,-0.26171875,0.91796875,-1.5625,1.2734375,-1.1015625,-0.578125,1.3984375,0.091308594,-1.6796875,-0.7890625,-0.640625,-0.34960938,0.16503906,0.36523438,0.80078125,-0.46875,0.30859375,-0.8359375,-0.49804688,-0.031982422,0.640625,-0.49023438,1.234375,-1.5625,0.46875,0.61328125,0.24804688,-0.86328125,1.09375,1.546875,0.859375,-0.74609375,-0.7109375,-0.66796875,2.21875,1.6484375,1.2109375,-0.24121094
+2,0,5,1.203125,-0.09814453,-1.3359375,-0.6015625,0.96484375,-1.4765625,-0.8203125,3.140625,-0.5234375,-1.671875,0.023925781,1.5859375,-0.53125,-0.74609375,-0.57421875,1.7421875,-0.26953125,-1.3984375,0.059326172,0.48828125,-1.3671875,-0.6796875,-0.44921875,1.234375,-0.09033203,0.32617188,-2.5,-0.15136719,-0.8828125,-0.06982422,-0.022216797,0.78515625,0.69140625,0.7734375,-0.07470703,-1.0546875,1.2578125,-0.17285156,0.515625,0.6875,-0.0625,-0.008483887,0.11669922,-0.7890625,-0.11816406,0.65625,-0.46875,0.96875,-0.43945312,2.21875,-0.60546875,-2.21875,-1.3359375,-0.04736328,-0.15039062,-0.33203125,0.91796875,0.296875,0.40039062,-1.046875,-0.36523438,1.484375,-0.1328125,0.53515625,-0.265625,-0.03125,-0.002380371,1.4921875,0.16699219,-0.11767578,-0.4140625,-0.29296875,0.8984375,-0.671875,1.625,2.140625,0.625,0.2734375,-0.2421875,-0.75,1.671875,0.75,0.53515625,-0.62890625,-0.44140625,-1.1640625,1.328125,1.828125,0.49414062,1.65625,1.7421875,-0.7734375,-1.046875,1.3671875,-0.28710938,0.03100586,-1.7265625,-0.734375,-1.0859375,-0.21777344,-1.0625,-0.24414062,1.4375,0.94140625,-0.47265625,0.25195312,-1.640625,-0.62109375,-0.29101562,0.0234375,-1.140625,-1.9296875,1.4140625,-2.265625,0.84765625,0.27929688,-2.046875,1.2734375,-0.30078125,-1.21875,1.25,-0.4765625,-0.92578125,-0.36523438,0.4140625,2.75,-0.43164062,-0.40625,0.44335938,-1.8671875,-1.4765625,-1.640625,0.36914062,-1.1015625,-0.890625,-0.012817383,1.328125,0.34570312,0.43554688,-1.3515625,1.46875,-1.8203125,-0.62890625,-0.15527344,0.16113281,0.46875,-0.32421875,0.83203125,-0.73828125,0.83203125,-0.6171875,-0.018188477,0.8125,-0.4453125,0.045898438,-0.7578125,0.3203125,0.015563965,1.65625,1.015625,-1.1484375,1.875,1.2890625,0.8203125,-0.57421875,0.77734375,0.16894531,1.3125,1.2578125,0.76953125,-1.296875,-0.17089844,0.060546875,1.0859375,-1.28125,0.43359375,0.3515625,-0.69140625,-0.059326172,-1.171875,0.45898438,0.38476562,0.30273438,-1.09375,0.22363281,-0.24414062,-1.3984375,0.0134887695,-0.83984375,-1.3515625,0.20996094,1.421875,-0.11621094,-0.43164062,0.12207031,1.1953125,0.49609375,0.6015625,-1.0703125,0.4609375
+2,0,6,0.31445312,-0.33398438,-0.72265625,-0.4453125,0.40429688,-0.546875,-0.006439209,1.5859375,-0.13769531,-0.828125,-0.53515625,-0.375,1.2265625,0.49804688,-1.6484375,0.44726562,-0.5234375,-1.984375,0.42578125,-0.6875,0.5625,-0.32421875,-0.27734375,-0.06640625,-1.328125,-1.8203125,-1.640625,-0.08984375,-2.078125,0.49023438,-0.6875,-1.1171875,0.5625,0.07519531,0.546875,1.6796875,0.62109375,-1.453125,-1.8671875,1.1953125,-1.4765625,0.3515625,-0.54296875,-0.043701172,0.23144531,1.0625,0.5625,-0.546875,0.24023438,1.71875,-2.109375,-1.328125,-1.40625,0.73046875,-1.9921875,0.62109375,0.125,-0.66796875,-0.19140625,-2.140625,1.1796875,-0.16113281,0.42578125,-0.072265625,0.17675781,0.055908203,-1.0859375,0.18847656,-0.671875,2.0,0.90625,-1.6015625,-0.4609375,-0.49609375,0.044189453,1.921875,1.1171875,1.109375,-0.61328125,-0.8828125,1.5546875,0.31835938,0.27734375,-0.5078125,-0.67578125,-0.030639648,0.28710938,0.9609375,-0.66796875,0.67578125,2.125,-0.17285156,0.14941406,0.41796875,-1.390625,-0.19921875,-0.30273438,-0.53125,1.4140625,-0.328125,-2.25,0.14160156,2.25,1.15625,-0.46679688,-0.22851562,1.484375,-0.9296875,0.40039062,0.71484375,0.2734375,0.36328125,0.8515625,-1.4296875,0.30664062,-0.46875,0.18554688,-0.27148438,-0.65234375,-0.65625,1.328125,-0.55078125,0.4140625,-1.1875,-0.24609375,1.2734375,-0.44921875,-2.296875,-0.005645752,-0.041015625,0.671875,-0.061767578,0.19238281,-1.3046875,-1.34375,-0.19921875,-0.35351562,-0.47070312,-0.019042969,-1.8046875,3.0625,-0.76953125,0.39257812,-0.053955078,-0.06933594,1.828125,-1.234375,-0.01977539,-1.9375,1.921875,0.23632812,1.125,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.4375,1.6171875,1.4296875,0.671875,-0.71484375,1.3515625,1.375,0.49804688,-1.3359375,0.53515625,-1.1171875,-0.33984375,2.375,-0.25195312,0.2265625,-0.14550781,-0.66796875,-0.31835938,0.118652344,0.453125,-0.9375,-0.91796875,0.93359375,-0.58984375,-0.18164062,-0.10888672,1.265625,0.48828125,0.6640625,-0.30273438,1.3203125,0.82421875,-0.5390625,-1.0390625,-0.23242188,0.53125,2.15625,0.024658203,0.83203125,-0.15625,1.40625,-0.20996094,1.21875,0.27148438
+2,0,7,0.76953125,0.111328125,-0.36523438,-0.22363281,0.20703125,-0.7578125,-0.2734375,1.4140625,0.5546875,-1.09375,0.5859375,1.4140625,0.4375,0.54296875,-1.6015625,0.8046875,0.19335938,-1.2109375,-0.11230469,0.119628906,0.98828125,-0.5390625,-0.39257812,0.578125,-0.29296875,-0.73046875,-0.51171875,0.06201172,-1.0078125,-0.059814453,-0.68359375,-0.51171875,1.78125,-0.45117188,1.296875,1.59375,0.83203125,-0.16015625,-2.21875,1.7265625,-1.8359375,0.375,-0.59765625,-0.05078125,0.6328125,0.81640625,-0.33984375,-0.1328125,-0.4453125,1.90625,-1.375,-0.11328125,-0.6875,-0.08935547,-0.796875,0.71484375,0.62109375,0.60546875,-0.07861328,-1.2421875,1.609375,-0.21484375,1.8671875,-0.34375,-1.7421875,0.8515625,-1.171875,0.53515625,-0.53125,1.265625,-0.10449219,-1.1640625,-0.09814453,0.6171875,-0.15527344,1.9296875,1.390625,0.38671875,-1.0,-1.46875,1.3828125,-0.20019531,0.59765625,-1.2890625,-1.3671875,-0.3359375,0.19433594,1.40625,-0.81640625,0.36132812,1.9609375,-0.03540039,-0.8203125,0.62109375,-1.546875,0.93359375,-0.17773438,0.890625,0.734375,0.9921875,-2.46875,0.013305664,2.0,1.03125,-1.3984375,0.35546875,0.91015625,-1.6640625,-1.0703125,0.18652344,0.578125,-0.36132812,1.546875,-1.5078125,-0.359375,-1.21875,0.95703125,-0.045410156,-0.3046875,-1.296875,2.125,0.44140625,-0.59765625,-1.2734375,0.60546875,0.41992188,-0.73046875,-2.1875,0.26367188,-0.36523438,0.18261719,-1.1640625,-0.54296875,-0.79296875,-1.84375,1.65625,0.30859375,-0.62109375,-0.53125,-1.34375,1.8203125,-1.6953125,-0.68359375,-0.3828125,-0.17871094,1.3125,-0.58203125,0.51953125,-2.6875,-0.1171875,0.25976562,1.125,0.69921875,-1.0390625,-0.22753906,-0.78515625,0.90234375,0.47851562,1.1171875,0.54296875,-0.41015625,0.3671875,-0.26171875,0.91796875,-1.5703125,1.28125,-1.109375,-0.58203125,1.3984375,0.08935547,-1.6796875,-0.79296875,-0.64453125,-0.34960938,0.16503906,0.36328125,0.80078125,-0.4609375,0.31054688,-0.828125,-0.49804688,-0.03466797,0.640625,-0.48828125,1.234375,-1.5546875,0.47070312,0.61328125,0.24902344,-0.85546875,1.09375,1.546875,0.86328125,-0.7421875,-0.7109375,-0.67578125,2.21875,1.6484375,1.1953125,-0.23925781
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
index f0aaa556..1f353e1c 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,5,1.09375,-0.07519531,-1.4921875,0.24121094,0.8515625,-1.703125,-0.57421875,2.953125,-0.059326172,-0.76171875,0.22753906,2.6875,0.265625,-0.36328125,-0.703125,1.734375,0.27929688,-1.328125,0.359375,0.921875,-0.24414062,-0.734375,-0.65234375,0.55078125,0.94140625,1.328125,-2.703125,0.38085938,-0.60546875,-1.0234375,0.34570312,-0.042236328,1.6953125,1.3515625,0.55078125,-0.62890625,0.99609375,-0.015625,-0.40429688,0.10839844,0.044433594,0.30859375,-0.18164062,-1.65625,-0.13867188,0.875,-1.5703125,0.44921875,0.072265625,0.7890625,-0.8515625,-1.421875,-1.1796875,-1.046875,-0.18945312,-0.20800781,0.74609375,0.29882812,-0.24804688,-1.0703125,0.21289062,0.71484375,1.3125,0.80859375,-1.265625,-0.32421875,0.3125,0.28320312,-0.578125,-0.40234375,0.4140625,-0.42578125,0.50390625,-0.46679688,2.265625,3.09375,1.0390625,0.024169922,-0.80859375,-0.84375,1.4453125,0.6875,0.82421875,-1.6171875,0.03930664,-0.3984375,0.47460938,1.3125,0.578125,2.203125,1.8671875,-0.765625,-1.0078125,1.359375,-0.6171875,-0.091308594,-1.390625,0.0067749023,-0.1484375,0.6796875,-0.51171875,0.16601562,0.056640625,1.265625,-1.0390625,-0.18261719,-1.109375,0.03173828,-1.3828125,-0.35742188,-1.1328125,-1.53125,1.765625,-1.8828125,0.61328125,-0.036865234,-1.4609375,0.6953125,-0.0051879883,-0.39257812,1.2734375,0.11425781,-0.8828125,-0.10644531,-0.045166016,2.421875,-0.7578125,-0.62890625,1.40625,-1.5390625,-1.1640625,-2.8125,0.10205078,-0.89453125,-1.203125,0.33984375,1.1640625,-0.27929688,0.18359375,-1.1875,0.92578125,-1.2734375,-1.3984375,-0.55078125,-0.20117188,0.53515625,-0.8203125,0.765625,-0.34960938,0.02746582,0.41796875,-0.54296875,0.21386719,-1.2109375,0.28125,-0.12988281,0.96875,0.044189453,1.328125,0.58984375,-0.18164062,2.46875,0.42578125,0.94921875,-0.064453125,0.3984375,0.18652344,0.48046875,1.2734375,1.2578125,-1.03125,-0.640625,-0.88671875,0.9140625,-1.2890625,0.29492188,0.14257812,-0.30078125,0.23535156,-0.3046875,-0.546875,-0.32421875,0.49023438,-1.203125,0.546875,-0.5234375,-1.1015625,-0.90625,-0.51953125,-0.9921875,0.5703125,0.96875,-0.4296875,-0.80859375,-0.16503906,1.9296875,-0.18652344,0.93359375,-1.4609375,-0.16992188
-3,0,6,-0.31054688,0.45898438,-1.703125,-1.1796875,-0.609375,0.16015625,-0.45703125,2.109375,0.5625,-0.81640625,1.7421875,1.1171875,1.453125,-0.3671875,0.080078125,0.9609375,-0.07080078,-0.93359375,0.76171875,-0.53125,0.625,-0.010375977,0.8671875,0.96484375,-0.055419922,0.13574219,0.22558594,-0.30664062,-1.2890625,-1.7265625,-0.3125,1.6796875,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.24023438,0.52734375,0.75,-0.203125,-0.45117188,0.16503906,-1.5078125,0.27539062,0.84375,0.23828125,0.5546875,0.39648438,-0.30273438,-2.21875,0.796875,-0.9140625,0.35546875,-0.8671875,-1.015625,-0.24023438,-0.6171875,0.28320312,-0.047607422,-0.11816406,0.87109375,0.6875,1.5390625,1.4140625,0.08886719,-1.28125,0.54296875,-0.765625,0.54296875,-0.40429688,0.123535156,0.48632812,0.095703125,-0.625,2.421875,-0.609375,0.017333984,-0.14453125,-0.03112793,-0.88671875,0.092285156,1.0,-0.8359375,0.88671875,0.7578125,-0.61328125,-0.41210938,1.0859375,-0.34570312,1.3984375,-0.24902344,-0.875,0.8359375,-0.115722656,-1.125,0.671875,-0.8359375,-1.390625,-1.21875,-1.4140625,0.68359375,3.140625,-0.81640625,-1.2421875,-1.171875,-0.67578125,0.30664062,-0.15039062,0.9140625,-0.55078125,-0.47070312,0.87890625,-1.25,1.765625,-0.30273438,-0.5546875,-0.359375,0.011352539,-0.64453125,-0.0015106201,-0.828125,0.45898438,-0.43945312,-1.03125,1.28125,-0.88671875,-1.65625,1.6015625,0.53125,-0.93359375,-0.78125,0.37304688,-0.98046875,-1.734375,1.4921875,0.69140625,0.33398438,-0.43164062,-1.9765625,2.5,-1.375,-1.0078125,-0.064941406,-0.1796875,-0.26367188,0.328125,-1.59375,-1.1953125,0.24023438,-0.07080078,1.59375,1.421875,-2.109375,0.10449219,-0.18457031,0.4140625,-1.03125,1.6875,-0.578125,0.421875,-0.27539062,0.89453125,-0.71484375,-0.017578125,2.25,-0.9453125,-0.11328125,1.5859375,-0.2578125,-1.625,-1.1015625,-2.078125,-1.5546875,-0.734375,0.484375,-0.032470703,0.0038757324,-1.328125,-1.265625,-0.13183594,-0.48632812,0.7421875,-0.029785156,0.73828125,-1.0546875,-0.45703125,0.15429688,-1.1015625,1.078125,1.734375,-0.21484375,0.6875,-0.46484375,-0.94921875,0.75390625,1.90625,1.6015625,0.65625,1.4296875
-3,0,7,0.578125,0.421875,-0.9765625,-0.27929688,-1.546875,-1.1796875,0.515625,1.3671875,-0.12890625,-0.9140625,1.6796875,1.3984375,0.7578125,0.19824219,0.4296875,0.94921875,-0.83203125,-0.28125,1.0625,-0.7265625,0.80078125,-0.29101562,2.296875,1.9921875,-0.28320312,0.5546875,-0.17480469,-1.1640625,-0.578125,-0.32421875,0.35742188,0.64453125,1.0703125,2.125,0.00074768066,-0.4765625,-0.20507812,0.5,0.60546875,-1.2734375,-1.0703125,-1.84375,1.0,-0.48242188,0.66796875,0.20019531,0.3515625,-0.19433594,-2.375,-0.13085938,-0.34960938,0.14648438,-1.65625,0.95703125,1.6328125,-0.11035156,-0.53515625,-0.021728516,-0.53125,0.028198242,0.24609375,0.26367188,-1.40625,-1.1796875,1.1328125,1.5234375,-1.0703125,-0.23925781,1.0234375,0.09277344,-1.4375,0.88671875,-0.890625,0.49609375,0.9375,0.66015625,1.0703125,0.61328125,-0.67578125,0.39257812,-0.77734375,0.546875,1.7578125,0.03173828,1.0390625,0.059570312,-0.39257812,1.2578125,0.06689453,0.82421875,1.2265625,-0.90625,-1.3125,0.52734375,-0.42578125,-0.7578125,1.5625,-0.31640625,-0.51171875,-0.51171875,-0.9140625,-1.0078125,2.421875,-0.26367188,-1.953125,-1.6875,0.71484375,-1.515625,0.859375,0.640625,-0.26171875,-1.609375,0.92578125,-1.234375,0.2734375,-0.53515625,0.028686523,0.42578125,0.3515625,-1.359375,1.3671875,0.47265625,-0.31445312,-2.140625,0.45117188,2.03125,-0.73828125,-0.119628906,1.796875,-1.5859375,0.44921875,1.25,1.203125,-2.609375,-2.28125,-0.28320312,-1.0078125,0.5625,-0.028320312,-1.4765625,0.89453125,-0.099121094,-0.2890625,0.29296875,-0.40234375,0.54296875,-0.9921875,-0.30078125,-0.80078125,0.64453125,-0.43945312,0.29882812,-0.6953125,-0.15332031,-0.29101562,-2.109375,-0.7734375,-0.55859375,0.53515625,1.46875,-0.80859375,-1.375,0.076171875,-0.8515625,0.8984375,-0.04248047,-2.0,-1.5234375,2.046875,-0.60546875,-0.31054688,0.54296875,-1.5234375,-0.47265625,-0.22363281,-0.092285156,1.546875,0.21875,-0.3515625,0.11328125,0.47460938,1.765625,0.4765625,-0.25585938,0.63671875,-0.7109375,-0.37695312,0.31445312,0.52734375,1.5390625,0.828125,0.60546875,-0.21679688,-0.5078125,0.4140625,-1.546875,0.765625,0.1328125,1.8046875,0.20996094
-4,0,5,2.046875,0.008178711,-1.2265625,1.3046875,1.1015625,-1.0078125,-0.7109375,1.5546875,-0.13476562,-0.9296875,1.03125,1.0625,1.0546875,0.23925781,-1.75,0.84765625,-0.36914062,-0.08300781,0.37304688,0.011230469,0.87890625,0.49609375,0.6328125,0.36523438,0.40820312,0.088378906,-1.7734375,0.33203125,-1.6640625,0.45898438,0.1640625,0.79296875,1.34375,1.7578125,0.71484375,0.41015625,0.24414062,-1.2421875,0.73046875,0.9140625,-0.80078125,0.20507812,-1.0546875,-0.73046875,0.40429688,-0.106933594,1.296875,-0.20605469,-0.8203125,-0.78125,-0.44726562,-1.1953125,-2.203125,-1.546875,-1.6796875,-0.82421875,0.28125,0.29492188,0.3828125,-0.6953125,0.10058594,0.23242188,0.04711914,0.7578125,-0.21777344,-1.2578125,-0.051513672,0.20996094,0.13378906,0.921875,1.1953125,-2.828125,0.040527344,-0.037597656,0.61328125,2.03125,0.3984375,-0.65234375,0.15039062,-1.9375,-0.48046875,0.29882812,0.11279297,-1.6328125,1.125,0.64453125,0.25195312,-0.36914062,1.2734375,-0.06982422,1.953125,0.9296875,0.002456665,0.4765625,0.49804688,-1.546875,-0.18945312,-0.6328125,-0.67578125,-0.06738281,-0.90625,1.3203125,0.9453125,-0.29296875,-0.62890625,-1.859375,-1.25,-0.18066406,-0.118652344,1.390625,-0.37695312,0.51171875,0.38867188,-0.3203125,1.0,1.296875,-0.77734375,-0.8671875,0.078125,-0.265625,-0.12597656,-0.72265625,-1.0859375,0.83984375,-1.3671875,1.515625,-0.90625,-1.09375,0.14941406,0.29492188,0.6171875,-0.79296875,-0.40234375,-0.6171875,-0.84375,0.703125,0.13574219,-1.171875,0.484375,-2.28125,1.9765625,-1.984375,-1.09375,0.44726562,0.76171875,0.3125,-1.328125,0.6640625,1.421875,2.46875,-0.030517578,0.84765625,0.4609375,0.29101562,1.5234375,1.0859375,-0.953125,0.26953125,0.78515625,-0.97265625,-0.18847656,1.5,1.015625,1.15625,-0.67578125,0.88671875,-0.26367188,-0.15527344,1.1171875,-1.1328125,-1.2109375,-1.3828125,0.096191406,-1.828125,-1.6328125,-0.82421875,0.3515625,-1.8515625,0.5390625,-0.09326172,-0.48828125,-0.8828125,0.52734375,-2.0,1.5,0.39648438,0.58984375,-0.50390625,-0.053466797,-2.09375,1.125,-0.056396484,1.15625,-0.46875,-0.16992188,2.65625,0.609375,0.014770508,0.29296875,1.0234375
-4,0,6,0.38671875,1.015625,-0.84375,-1.046875,1.4921875,0.36132812,-1.2421875,1.28125,0.82421875,-0.13574219,0.16894531,0.70703125,-0.6484375,0.14550781,-0.390625,-0.62890625,-1.2109375,0.21386719,1.8671875,1.171875,0.6640625,-0.39453125,0.94921875,0.6796875,-0.84375,-0.78515625,-2.0625,-1.0625,-2.046875,-0.029174805,-0.38867188,0.07470703,0.22265625,1.046875,-0.5703125,0.096191406,0.38085938,-0.23925781,0.54296875,0.875,-1.671875,0.46875,-0.040527344,0.08105469,0.26953125,0.38085938,0.78125,-0.875,0.70703125,0.65234375,0.98046875,-1.2734375,0.103027344,0.28515625,-2.0,-0.6484375,-0.34179688,-1.0078125,0.6171875,-1.390625,0.28710938,0.76171875,2.0625,-0.71875,0.47851562,-1.0625,-0.04663086,-1.265625,-0.171875,1.484375,0.50390625,1.390625,0.04638672,-1.0390625,-0.040771484,0.83984375,-0.37890625,0.3046875,-0.71484375,0.063964844,0.51953125,-0.9140625,0.46484375,-0.32617188,-0.375,-0.26367188,-0.50390625,-1.34375,1.5,0.3828125,1.3125,1.7421875,-0.98046875,1.4375,-0.53515625,-0.73828125,-0.4140625,-0.024902344,-0.97265625,-0.095214844,-0.62890625,1.2265625,1.6640625,-1.609375,-1.1015625,0.5859375,-0.9296875,0.110839844,0.31445312,0.020263672,0.3515625,0.40039062,0.106933594,0.390625,0.99609375,-0.53125,-0.5703125,-0.31835938,0.6796875,0.07714844,0.25195312,-0.2265625,-1.0,0.22167969,-1.1640625,2.09375,-0.18554688,-2.1875,1.359375,2.828125,0.61328125,-1.625,-2.03125,0.48242188,-1.6875,-0.6953125,-1.1640625,-1.4765625,0.40820312,-1.6875,2.140625,-2.296875,-0.984375,1.8671875,0.796875,-0.7578125,0.78125,0.2265625,-0.37304688,1.75,0.37304688,-0.22460938,1.171875,-0.38085938,0.05444336,0.83984375,1.6796875,0.35351562,0.7578125,-2.046875,0.8828125,1.203125,0.97265625,2.21875,-1.0390625,1.0390625,-1.6171875,-1.4140625,2.03125,-0.81640625,0.390625,-1.5078125,0.11621094,-0.18847656,-0.71875,-1.0390625,-0.49023438,-1.375,0.23046875,0.0035247803,-0.8203125,-0.609375,0.83203125,0.60546875,-0.35546875,1.21875,-0.064941406,-0.56640625,1.0703125,-0.9765625,-1.109375,0.036865234,0.45898438,-0.5859375,-1.03125,1.9453125,0.7890625,0.671875,-0.421875,0.66015625
-4,0,7,1.2578125,-0.80078125,-0.62890625,0.19628906,-0.296875,-0.51953125,-0.765625,0.88671875,-0.6796875,-0.62890625,0.32421875,1.1953125,1.0546875,0.52734375,-0.6640625,1.5078125,-0.31054688,-0.44921875,2.265625,-0.46289062,1.6953125,0.14453125,-0.20214844,0.046142578,2.1875,-0.54296875,-0.984375,-1.3125,-0.546875,0.49023438,-0.30273438,0.5390625,1.0390625,-0.013305664,-1.984375,-0.13085938,0.53515625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.86328125,-1.0625,-0.23925781,-0.09716797,0.91796875,0.1953125,-0.68359375,-1.2578125,-0.10107422,-0.44335938,-1.0078125,-1.5859375,-0.59375,-0.5,1.953125,0.48632812,0.014770508,0.23535156,-0.71875,1.3203125,-0.578125,0.52734375,1.53125,0.5859375,0.1875,-0.48828125,-0.45898438,2.265625,-0.023071289,0.18359375,0.029052734,-0.37890625,0.28320312,1.0859375,2.375,-0.25195312,0.08203125,-1.6484375,-0.34375,0.88671875,-0.51953125,-0.45703125,0.111816406,0.99609375,-0.15625,0.3203125,1.5859375,0.5625,1.6015625,1.5859375,-0.40429688,0.2890625,0.97265625,-1.7265625,1.5546875,-0.90625,-0.030151367,0.38085938,-0.22167969,-0.83984375,0.19140625,-0.21777344,0.28710938,-0.91796875,-0.29882812,-0.4921875,-2.6875,-0.64453125,0.328125,0.48046875,-0.953125,0.6484375,-0.6796875,-0.609375,-0.10107422,-0.79296875,0.97265625,-0.24609375,0.21289062,0.78515625,1.1640625,-0.25195312,-1.7109375,-1.2109375,0.8984375,-1.421875,-0.9453125,0.024658203,0.23242188,0.96875,-0.29882812,-0.29296875,-1.296875,-0.61328125,-0.20605469,1.8671875,-0.62890625,-1.6875,-2.625,2.171875,0.64453125,-1.078125,-0.12792969,-0.84375,-0.20703125,0.6796875,-0.875,0.29882812,1.9140625,-0.36523438,-0.11035156,1.3203125,-2.09375,-0.30664062,-1.5859375,1.625,1.2734375,1.3984375,-0.83984375,-0.48046875,-0.8046875,0.36914062,-0.28515625,-1.1328125,0.18652344,0.09814453,-1.0859375,1.828125,-0.390625,0.30859375,-0.28125,-1.2890625,0.45703125,-0.51171875,1.6015625,0.21191406,-0.12695312,0.9765625,0.091308594,-1.328125,0.22558594,-0.020385742,0.5,0.019897461,-0.5546875,1.0,-0.44140625,1.5546875,-0.90234375,-0.16210938,-0.5390625,1.2421875,-0.67578125,-0.81640625,0.48046875,0.7109375,2.796875,2.171875,0.546875
+3,0,5,1.0859375,-0.07324219,-1.484375,0.24511719,0.85546875,-1.703125,-0.57421875,2.953125,-0.0625,-0.76171875,0.23046875,2.65625,0.26757812,-0.36328125,-0.69921875,1.734375,0.2734375,-1.328125,0.359375,0.92578125,-0.24414062,-0.73828125,-0.65625,0.5390625,0.9375,1.3359375,-2.703125,0.38671875,-0.59375,-1.0234375,0.34375,-0.045166016,1.6953125,1.34375,0.54296875,-0.62890625,0.99609375,-0.021362305,-0.40820312,0.10644531,0.04736328,0.30859375,-0.17773438,-1.6484375,-0.13574219,0.8828125,-1.5625,0.44726562,0.07519531,0.79296875,-0.84765625,-1.421875,-1.171875,-1.046875,-0.19042969,-0.20800781,0.74609375,0.30078125,-0.24511719,-1.0625,0.21191406,0.71484375,1.3203125,0.8125,-1.265625,-0.32226562,0.30859375,0.27929688,-0.57421875,-0.40234375,0.41601562,-0.42382812,0.49609375,-0.47070312,2.28125,3.09375,1.046875,0.02355957,-0.81640625,-0.84765625,1.453125,0.6875,0.82421875,-1.6015625,0.04321289,-0.3984375,0.47070312,1.328125,0.578125,2.203125,1.875,-0.765625,-1.015625,1.359375,-0.61328125,-0.08691406,-1.3984375,0.0048828125,-0.14941406,0.6796875,-0.51171875,0.16015625,0.052734375,1.265625,-1.046875,-0.18261719,-1.125,0.028930664,-1.3828125,-0.35351562,-1.125,-1.53125,1.765625,-1.8984375,0.61328125,-0.03857422,-1.4609375,0.69140625,-0.0063476562,-0.39453125,1.2734375,0.11425781,-0.890625,-0.11328125,-0.04321289,2.421875,-0.75,-0.62109375,1.3828125,-1.5390625,-1.171875,-2.828125,0.103027344,-0.890625,-1.2109375,0.33984375,1.15625,-0.27539062,0.18066406,-1.171875,0.92578125,-1.2734375,-1.3984375,-0.54296875,-0.19824219,0.54296875,-0.82421875,0.765625,-0.34960938,0.026489258,0.41992188,-0.5390625,0.21484375,-1.2109375,0.28125,-0.12695312,0.96875,0.041503906,1.3359375,0.58984375,-0.18261719,2.46875,0.421875,0.94921875,-0.05883789,0.40039062,0.18945312,0.4765625,1.2734375,1.265625,-1.0234375,-0.640625,-0.88671875,0.91015625,-1.2890625,0.29492188,0.14355469,-0.30078125,0.23828125,-0.30078125,-0.5390625,-0.328125,0.49023438,-1.2109375,0.546875,-0.5234375,-1.1015625,-0.91015625,-0.515625,-0.97265625,0.57421875,0.96484375,-0.42773438,-0.8046875,-0.16699219,1.921875,-0.18457031,0.93359375,-1.453125,-0.171875
+3,0,6,-0.31054688,0.4609375,-1.6953125,-1.1796875,-0.61328125,0.16113281,-0.45703125,2.109375,0.5625,-0.8203125,1.734375,1.125,1.453125,-0.36328125,0.07763672,0.9609375,-0.07128906,-0.9375,0.76171875,-0.53515625,0.62109375,-0.012634277,0.8671875,0.96875,-0.055664062,0.13867188,0.2265625,-0.30859375,-1.2890625,-1.7265625,-0.31054688,1.6953125,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.2421875,0.53515625,0.75,-0.20214844,-0.45703125,0.16113281,-1.5,0.27539062,0.84375,0.23828125,0.55078125,0.3984375,-0.30273438,-2.21875,0.80078125,-0.9140625,0.35546875,-0.85546875,-1.015625,-0.24023438,-0.61328125,0.28125,-0.050048828,-0.12060547,0.86328125,0.69140625,1.5546875,1.40625,0.08886719,-1.28125,0.5390625,-0.765625,0.5390625,-0.40625,0.12109375,0.484375,0.09765625,-0.62109375,2.421875,-0.609375,0.017333984,-0.14648438,-0.03149414,-0.88671875,0.09033203,0.99609375,-0.8359375,0.88671875,0.76171875,-0.6171875,-0.4140625,1.0859375,-0.34570312,1.390625,-0.24902344,-0.875,0.8359375,-0.1171875,-1.125,0.6796875,-0.83203125,-1.3984375,-1.2109375,-1.4140625,0.6796875,3.15625,-0.81640625,-1.234375,-1.1640625,-0.6796875,0.30664062,-0.15429688,0.91796875,-0.546875,-0.47070312,0.8828125,-1.25,1.7734375,-0.30273438,-0.5546875,-0.36523438,0.011047363,-0.64453125,-0.0018310547,-0.828125,0.45507812,-0.44140625,-1.03125,1.28125,-0.88671875,-1.6640625,1.6015625,0.53125,-0.93359375,-0.78125,0.37304688,-0.984375,-1.734375,1.5,0.69140625,0.3359375,-0.43359375,-1.96875,2.484375,-1.375,-1.0078125,-0.061523438,-0.18261719,-0.26171875,0.32421875,-1.59375,-1.1953125,0.24121094,-0.068359375,1.59375,1.421875,-2.140625,0.106933594,-0.18359375,0.41210938,-1.0234375,1.6796875,-0.58203125,0.42382812,-0.27734375,0.90234375,-0.71484375,-0.013977051,2.234375,-0.953125,-0.110839844,1.5859375,-0.2578125,-1.6171875,-1.1015625,-2.0625,-1.5546875,-0.734375,0.47460938,-0.030151367,0.001625061,-1.328125,-1.2734375,-0.12890625,-0.48632812,0.7421875,-0.03100586,0.734375,-1.0546875,-0.45507812,0.15820312,-1.09375,1.0703125,1.734375,-0.21289062,0.6953125,-0.46679688,-0.95703125,0.75,1.90625,1.609375,0.66015625,1.421875
+3,0,7,0.578125,0.41992188,-0.9765625,-0.27929688,-1.546875,-1.1796875,0.515625,1.3671875,-0.12890625,-0.91015625,1.6640625,1.390625,0.76171875,0.19921875,0.42773438,0.95703125,-0.8359375,-0.28320312,1.0625,-0.7265625,0.80078125,-0.29101562,2.28125,2.0,-0.28125,0.5546875,-0.17382812,-1.1640625,-0.57421875,-0.32617188,0.35546875,0.6484375,1.078125,2.125,-0.0016326904,-0.47460938,-0.20605469,0.5,0.60546875,-1.2734375,-1.0546875,-1.84375,1.0078125,-0.4765625,0.66796875,0.19921875,0.3515625,-0.19433594,-2.375,-0.13085938,-0.34765625,0.14355469,-1.6640625,0.95703125,1.625,-0.106933594,-0.53515625,-0.020629883,-0.53125,0.026367188,0.2421875,0.25976562,-1.40625,-1.1796875,1.1484375,1.53125,-1.078125,-0.23925781,1.0234375,0.092285156,-1.4296875,0.88671875,-0.88671875,0.49609375,0.9453125,0.65625,1.0703125,0.61328125,-0.67578125,0.39257812,-0.76953125,0.546875,1.7734375,0.037597656,1.0390625,0.057128906,-0.39453125,1.265625,0.06640625,0.828125,1.2265625,-0.90625,-1.3046875,0.52734375,-0.42578125,-0.75390625,1.5546875,-0.31640625,-0.5078125,-0.5078125,-0.921875,-1.0234375,2.40625,-0.26367188,-1.9453125,-1.6953125,0.7109375,-1.5234375,0.86328125,0.64453125,-0.26171875,-1.6171875,0.921875,-1.234375,0.27539062,-0.53125,0.028320312,0.421875,0.3515625,-1.3515625,1.34375,0.4765625,-0.31445312,-2.140625,0.45507812,2.015625,-0.734375,-0.119628906,1.796875,-1.59375,0.453125,1.234375,1.2109375,-2.625,-2.296875,-0.28710938,-1.015625,0.5625,-0.030273438,-1.4765625,0.89453125,-0.09814453,-0.28710938,0.296875,-0.40234375,0.54296875,-0.9921875,-0.30273438,-0.8046875,0.640625,-0.43945312,0.29492188,-0.69140625,-0.15136719,-0.29101562,-2.109375,-0.78125,-0.55859375,0.53515625,1.4609375,-0.80859375,-1.3828125,0.078125,-0.8515625,0.90625,-0.040039062,-2.0,-1.515625,2.046875,-0.59765625,-0.30664062,0.5390625,-1.5234375,-0.47265625,-0.22363281,-0.09326172,1.5390625,0.22265625,-0.34960938,0.11425781,0.4765625,1.7578125,0.47851562,-0.25390625,0.63671875,-0.7109375,-0.37695312,0.31640625,0.52734375,1.546875,0.8359375,0.60546875,-0.21386719,-0.50390625,0.40820312,-1.53125,0.76953125,0.12988281,1.8046875,0.21191406
+4,0,5,2.046875,0.015991211,-1.2265625,1.3203125,1.1015625,-1.015625,-0.7109375,1.546875,-0.13476562,-0.93359375,1.03125,1.0546875,1.0390625,0.23828125,-1.75,0.83984375,-0.37109375,-0.0859375,0.375,0.013183594,0.875,0.49804688,0.625,0.36523438,0.40429688,0.08935547,-1.7578125,0.33007812,-1.6640625,0.45507812,0.16796875,0.79296875,1.34375,1.765625,0.7109375,0.41015625,0.24316406,-1.2421875,0.734375,0.91015625,-0.80078125,0.203125,-1.0546875,-0.734375,0.40429688,-0.106933594,1.2890625,-0.21386719,-0.81640625,-0.78125,-0.4453125,-1.1953125,-2.203125,-1.5390625,-1.6796875,-0.8203125,0.27929688,0.29296875,0.3828125,-0.6953125,0.09716797,0.23046875,0.049072266,0.76171875,-0.21875,-1.2578125,-0.049560547,0.20996094,0.13867188,0.92578125,1.1875,-2.8125,0.037597656,-0.033935547,0.62109375,2.046875,0.40039062,-0.6484375,0.1484375,-1.9375,-0.48046875,0.29492188,0.114746094,-1.640625,1.1328125,0.6484375,0.25195312,-0.3671875,1.2734375,-0.0703125,1.9609375,0.92578125,0.002456665,0.47460938,0.49804688,-1.546875,-0.19433594,-0.63671875,-0.67578125,-0.064941406,-0.90625,1.3125,0.94140625,-0.29296875,-0.6328125,-1.875,-1.2578125,-0.1796875,-0.12109375,1.3984375,-0.37695312,0.5078125,0.38476562,-0.3203125,1.0,1.296875,-0.78125,-0.859375,0.071777344,-0.265625,-0.12597656,-0.7265625,-1.0859375,0.8359375,-1.3828125,1.5234375,-0.8984375,-1.0859375,0.1484375,0.29101562,0.6171875,-0.7890625,-0.3984375,-0.62109375,-0.84375,0.6953125,0.13183594,-1.1640625,0.48632812,-2.28125,1.96875,-1.984375,-1.1015625,0.44726562,0.76953125,0.31054688,-1.3359375,0.671875,1.4296875,2.46875,-0.029663086,0.85546875,0.45703125,0.2890625,1.53125,1.0859375,-0.953125,0.27148438,0.78125,-0.98046875,-0.18945312,1.4921875,1.0,1.15625,-0.6640625,0.88671875,-0.26757812,-0.15527344,1.125,-1.1328125,-1.203125,-1.375,0.09814453,-1.828125,-1.6484375,-0.82421875,0.35546875,-1.8515625,0.5390625,-0.09033203,-0.49023438,-0.88671875,0.5234375,-1.9921875,1.5,0.39648438,0.58984375,-0.5,-0.052001953,-2.0625,1.1171875,-0.055664062,1.15625,-0.47265625,-0.171875,2.65625,0.6171875,0.014770508,0.3046875,1.015625
+4,0,6,0.38867188,1.015625,-0.84375,-1.046875,1.4921875,0.36523438,-1.2578125,1.28125,0.83203125,-0.13476562,0.16894531,0.7109375,-0.65234375,0.14550781,-0.390625,-0.63671875,-1.1953125,0.2109375,1.8671875,1.171875,0.6640625,-0.39257812,0.9375,0.6796875,-0.84375,-0.7734375,-2.078125,-1.0625,-2.015625,-0.031982422,-0.38671875,0.07421875,0.22460938,1.0546875,-0.57421875,0.09472656,0.38085938,-0.24316406,0.546875,0.86328125,-1.6796875,0.47265625,-0.044433594,0.08251953,0.26953125,0.37890625,0.7890625,-0.875,0.703125,0.65234375,0.984375,-1.2734375,0.107910156,0.28320312,-1.9921875,-0.6484375,-0.34570312,-1.0078125,0.62109375,-1.390625,0.2890625,0.76171875,2.03125,-0.73046875,0.47851562,-1.0546875,-0.05053711,-1.2734375,-0.16992188,1.484375,0.49804688,1.4140625,0.044189453,-1.046875,-0.04296875,0.83203125,-0.37695312,0.3046875,-0.71484375,0.064941406,0.51171875,-0.9140625,0.46484375,-0.32617188,-0.37890625,-0.26367188,-0.50390625,-1.3359375,1.5078125,0.38671875,1.3125,1.75,-1.0,1.453125,-0.53515625,-0.7421875,-0.4140625,-0.028686523,-0.98046875,-0.09326172,-0.62890625,1.234375,1.6640625,-1.609375,-1.1015625,0.5859375,-0.9296875,0.11279297,0.31054688,0.021728516,0.35351562,0.3984375,0.106933594,0.38867188,1.0,-0.53125,-0.5703125,-0.3203125,0.67578125,0.07324219,0.25195312,-0.22070312,-1.0,0.22167969,-1.15625,2.078125,-0.1875,-2.203125,1.359375,2.84375,0.62109375,-1.625,-2.03125,0.47851562,-1.6875,-0.69921875,-1.1640625,-1.4765625,0.41015625,-1.6875,2.140625,-2.296875,-0.98828125,1.875,0.796875,-0.75390625,0.7734375,0.22265625,-0.37695312,1.75,0.375,-0.22753906,1.1640625,-0.37695312,0.051513672,0.83203125,1.6875,0.35351562,0.7578125,-2.046875,0.875,1.203125,0.9765625,2.203125,-1.03125,1.03125,-1.6171875,-1.421875,2.03125,-0.81640625,0.39453125,-1.515625,0.11816406,-0.1875,-0.72265625,-1.0390625,-0.49023438,-1.3828125,0.22851562,0.0050354004,-0.8203125,-0.61328125,0.8359375,0.6015625,-0.35351562,1.2265625,-0.06347656,-0.56640625,1.0625,-0.9765625,-1.109375,0.034423828,0.45898438,-0.5859375,-1.03125,1.953125,0.79296875,0.66796875,-0.421875,0.6640625
+4,0,7,1.265625,-0.80859375,-0.62890625,0.19921875,-0.29101562,-0.51171875,-0.76953125,0.8828125,-0.671875,-0.6328125,0.32617188,1.1875,1.0546875,0.5234375,-0.66015625,1.515625,-0.31445312,-0.44921875,2.265625,-0.46289062,1.6953125,0.15039062,-0.20507812,0.05078125,2.1875,-0.5390625,-0.98046875,-1.3125,-0.5546875,0.49023438,-0.29882812,0.54296875,1.03125,-0.008605957,-1.984375,-0.13183594,0.5390625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.86328125,-1.0625,-0.23925781,-0.099609375,0.921875,0.19726562,-0.6796875,-1.2578125,-0.103515625,-0.44140625,-1.0078125,-1.59375,-0.59375,-0.50390625,1.9765625,0.48632812,0.014526367,0.23632812,-0.7109375,1.328125,-0.578125,0.5234375,1.5234375,0.58203125,0.18652344,-0.48632812,-0.4609375,2.265625,-0.023071289,0.17675781,0.029907227,-0.37890625,0.28515625,1.0859375,2.375,-0.25,0.083984375,-1.6484375,-0.34375,0.88671875,-0.51953125,-0.45703125,0.111328125,1.0,-0.15722656,0.31445312,1.578125,0.55859375,1.6015625,1.578125,-0.40429688,0.28710938,0.97265625,-1.7265625,1.5703125,-0.90625,-0.034179688,0.375,-0.22167969,-0.83984375,0.19335938,-0.22460938,0.28710938,-0.92578125,-0.29882812,-0.48828125,-2.6875,-0.64453125,0.328125,0.48046875,-0.953125,0.65234375,-0.6796875,-0.61328125,-0.100097656,-0.79296875,0.96875,-0.24511719,0.21386719,0.77734375,1.15625,-0.25195312,-1.7265625,-1.2109375,0.8984375,-1.421875,-0.94921875,0.022949219,0.23046875,0.97265625,-0.30273438,-0.2890625,-1.28125,-0.60546875,-0.20898438,1.8515625,-0.62890625,-1.6796875,-2.625,2.171875,0.64453125,-1.078125,-0.12597656,-0.83984375,-0.20507812,0.671875,-0.87109375,0.296875,1.9296875,-0.36523438,-0.111816406,1.3125,-2.09375,-0.30859375,-1.6015625,1.609375,1.265625,1.40625,-0.84765625,-0.48242188,-0.8046875,0.37890625,-0.28710938,-1.140625,0.18652344,0.096191406,-1.0859375,1.8359375,-0.390625,0.30859375,-0.28125,-1.2890625,0.4609375,-0.51171875,1.5859375,0.21289062,-0.12695312,0.96484375,0.09326172,-1.328125,0.22460938,-0.020507812,0.5,0.018676758,-0.546875,1.0,-0.44335938,1.5625,-0.89453125,-0.15917969,-0.53515625,1.25,-0.67578125,-0.81640625,0.48632812,0.7109375,2.765625,2.171875,0.546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
index 88760c11..f76bfe98 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,5,0.58984375,0.99609375,-0.34179688,-1.34375,-0.484375,-0.9140625,-0.88671875,0.4609375,0.94921875,-1.2421875,-0.19433594,-0.5859375,-1.09375,1.5390625,-1.671875,0.66796875,-0.3359375,0.453125,-0.8125,0.87890625,1.6640625,0.703125,0.06201172,0.34765625,0.40625,-0.58984375,-1.8828125,-0.15136719,-0.99609375,1.21875,-0.9296875,-0.2265625,0.83984375,0.2109375,-0.23046875,0.90625,0.40039062,-0.05908203,-0.58984375,0.62109375,-1.7890625,0.38085938,-0.12158203,0.41015625,-0.0019302368,1.3359375,0.70703125,0.13671875,-0.40234375,1.2578125,-0.41796875,-0.6796875,-1.5234375,-0.018432617,-0.041503906,1.8125,0.013305664,-0.23535156,0.40429688,-1.546875,0.29101562,0.39648438,1.2890625,0.24316406,0.96875,1.1796875,-1.5703125,1.46875,0.012756348,-0.22949219,-0.796875,-1.203125,-0.76953125,-0.07128906,-0.58984375,0.015563965,0.49414062,0.5546875,0.30859375,0.6484375,1.078125,-1.046875,1.015625,0.42382812,-0.53125,0.49609375,0.3515625,0.640625,-0.58984375,-0.50390625,3.59375,0.41992188,0.79296875,1.265625,-0.51953125,1.0390625,-0.98828125,0.31835938,-0.58203125,0.34570312,-0.828125,0.17089844,1.3828125,-1.140625,0.045410156,0.17773438,-0.390625,-1.8671875,-0.4453125,0.83203125,-0.15332031,-0.40625,0.4296875,-0.80859375,-0.625,-0.609375,-0.34960938,0.69921875,0.42773438,-0.12597656,0.734375,-1.0703125,-0.62890625,-1.8515625,-0.828125,1.65625,-0.35742188,-1.625,-1.1015625,-0.96875,1.3203125,0.2578125,-2.375,-0.5859375,-1.5703125,2.015625,0.14355469,-0.45507812,-1.34375,-2.5625,1.1015625,-2.21875,0.58203125,-0.09277344,0.6484375,0.35546875,-0.20605469,-0.76953125,-0.088378906,1.5625,-1.5859375,0.59765625,0.60546875,-0.09814453,1.296875,-0.064453125,0.4140625,0.9375,0.78515625,0.41601562,-0.87890625,-0.6640625,-0.37109375,0.09033203,-2.6875,2.390625,0.22949219,-0.0134887695,1.4140625,-0.9140625,-1.8984375,-0.86328125,-1.1171875,-0.9296875,-0.44921875,1.5078125,1.1796875,-0.99609375,-0.18261719,-0.91015625,0.44921875,1.3046875,0.50390625,-0.040771484,1.0078125,-1.109375,-0.13085938,1.609375,0.578125,-1.3125,-0.23730469,1.359375,1.2265625,0.38085938,-0.5546875,0.43554688,3.265625,-0.09863281,-0.18554688,0.45117188
-5,0,6,-0.44140625,-0.69921875,-0.38476562,-1.4375,0.17578125,-1.203125,-0.9609375,1.640625,0.057373047,0.8203125,-0.8671875,-0.53125,1.4140625,0.875,-2.15625,0.93359375,-0.51953125,-0.859375,0.10546875,0.5078125,2.46875,-0.18359375,-0.8515625,0.24316406,0.48046875,-0.38671875,-2.078125,-1.125,-0.42578125,1.75,-0.26757812,0.50390625,-0.98046875,0.5546875,-1.25,0.49804688,-0.37890625,-0.49804688,-0.27929688,-0.6015625,-0.9453125,0.328125,-0.6953125,-0.28710938,0.53515625,2.484375,-0.107910156,-0.3125,1.390625,1.0703125,-1.0078125,-2.421875,-2.34375,0.43945312,-0.609375,1.6640625,-0.28125,-0.80859375,-0.26757812,-1.5546875,-0.87890625,-0.80859375,1.578125,0.9609375,1.4375,2.015625,-1.6640625,1.625,-0.95703125,0.6171875,1.3046875,-1.2890625,-0.6953125,-0.265625,-0.234375,0.25,0.265625,-0.33984375,0.73046875,0.73046875,1.578125,-0.42382812,0.28320312,1.734375,-0.84375,0.64453125,-0.8203125,0.05126953,0.13476562,-0.11230469,1.4765625,-0.3671875,1.78125,-0.09765625,-0.15429688,0.027709961,-1.84375,0.78125,-0.54296875,0.375,-0.2890625,-0.26757812,0.36328125,0.17675781,0.3671875,1.3125,-1.671875,-2.59375,-1.0703125,-0.609375,-0.20019531,0.78125,-1.0546875,-0.53125,-0.09716797,-2.359375,-0.09716797,0.81640625,0.33007812,1.5859375,0.11279297,-0.3828125,-0.15917969,-1.6484375,-1.2578125,0.6328125,-1.359375,-0.77734375,-0.7890625,0.48046875,2.0625,-0.7578125,-1.7109375,-0.546875,-0.28710938,0.8359375,-0.13867188,0.07861328,-0.09716797,-2.046875,-0.32226562,1.609375,0.58984375,0.95703125,0.52734375,0.15917969,-0.51953125,0.20410156,-0.59765625,1.0234375,0.03515625,-0.90625,0.54296875,-0.52734375,0.43945312,-0.17382812,1.4140625,1.0078125,0.55078125,0.35742188,-0.19433594,0.875,0.4140625,0.50390625,-1.796875,1.734375,0.73828125,0.20800781,0.46289062,0.03930664,-0.49804688,0.63671875,-0.0390625,0.18066406,-1.1171875,1.984375,0.18164062,-1.671875,-0.57421875,-0.099609375,-0.6640625,0.875,0.91796875,0.3828125,0.28125,-0.28710938,0.28125,0.39453125,0.3203125,-0.796875,-1.453125,0.7578125,1.5078125,0.14746094,-0.55078125,0.2734375,2.375,-0.3984375,1.0078125,1.125
-5,0,7,0.91015625,0.5625,0.828125,0.203125,-1.140625,-0.08105469,-1.1953125,-0.031982422,2.609375,-1.46875,1.828125,-0.040771484,0.35351562,1.3671875,-0.62890625,-0.42578125,-0.4609375,0.18457031,2.265625,1.1484375,0.2421875,-0.5078125,-0.15722656,0.58203125,-0.48632812,-0.53515625,0.65234375,-0.5234375,0.90625,0.5078125,-0.40429688,-0.0074157715,0.78125,-0.66796875,0.37695312,0.91015625,0.94921875,-0.18847656,-0.20898438,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24707031,-0.23242188,0.515625,-0.29296875,0.87890625,0.6796875,-0.15429688,-0.265625,-0.74609375,0.95703125,0.9296875,0.40820312,-0.65625,-0.49804688,-0.578125,0.77734375,1.1875,-0.37890625,1.2578125,0.09765625,0.640625,0.51171875,-0.20703125,-1.25,-1.453125,0.41210938,-0.11230469,1.3828125,0.78515625,-0.640625,-0.65234375,0.47851562,-0.52734375,0.4765625,-1.09375,1.2578125,-1.8984375,-0.71484375,0.9296875,-1.1640625,-0.703125,0.51171875,-0.36523438,-1.5,0.47851562,0.28515625,1.5390625,1.1953125,-0.47070312,0.875,-0.45117188,0.16308594,-1.78125,0.73046875,-0.63671875,-0.3671875,-1.6875,0.609375,0.9140625,-1.2890625,-0.34960938,2.171875,-0.640625,-0.7890625,-0.375,0.7890625,-0.4765625,0.26171875,1.203125,-0.578125,0.26367188,0.34960938,-0.625,0.453125,-0.41015625,0.5546875,0.6875,-0.0013961792,0.4453125,-1.6875,-1.90625,0.5625,-0.16992188,-0.203125,-0.34570312,1.140625,1.078125,0.61328125,-1.5078125,-0.265625,-1.1015625,1.578125,0.16503906,-0.73046875,-0.41796875,-1.7578125,1.6796875,-0.26953125,-0.625,1.5078125,-0.28320312,-0.07519531,1.4609375,-0.734375,-1.1796875,0.22070312,-1.078125,-0.16503906,2.140625,-0.57421875,-0.60546875,0.19042969,1.40625,-0.39648438,0.65625,0.38867188,-0.75,-0.30664062,1.390625,1.2421875,-1.1796875,-0.14648438,-0.6640625,-0.6171875,0.64453125,-1.703125,-2.109375,-0.25585938,-3.0,-0.40039062,-0.578125,2.484375,0.80078125,-1.046875,-0.48046875,0.36328125,-0.33398438,-0.09277344,-0.06982422,-0.6640625,-1.1640625,0.20800781,1.125,0.58984375,-0.390625,-0.8203125,-1.5234375,0.296875,0.546875,-0.86328125,-2.796875,1.2578125,3.171875,0.19921875,0.34570312,0.859375
+5,0,5,0.58984375,0.99609375,-0.34570312,-1.3515625,-0.484375,-0.91015625,-0.88671875,0.4609375,0.94921875,-1.25,-0.1953125,-0.58203125,-1.09375,1.546875,-1.6640625,0.671875,-0.33007812,0.453125,-0.81640625,0.87109375,1.671875,0.70703125,0.06347656,0.34570312,0.41015625,-0.58984375,-1.8828125,-0.15136719,-1.0,1.2109375,-0.92578125,-0.22558594,0.83984375,0.20800781,-0.22753906,0.91015625,0.40039062,-0.060546875,-0.58984375,0.6171875,-1.8046875,0.38085938,-0.12109375,0.40625,-0.0045166016,1.34375,0.703125,0.13867188,-0.40039062,1.25,-0.41796875,-0.66796875,-1.515625,-0.021972656,-0.045654297,1.8203125,0.018310547,-0.23730469,0.40429688,-1.546875,0.29101562,0.39453125,1.2890625,0.2421875,0.9765625,1.1875,-1.5859375,1.4765625,0.013549805,-0.22753906,-0.80078125,-1.2109375,-0.765625,-0.07080078,-0.58203125,0.013427734,0.4921875,0.55859375,0.3125,0.64453125,1.078125,-1.046875,1.0,0.42382812,-0.53125,0.49804688,0.34375,0.62890625,-0.5859375,-0.50390625,3.578125,0.421875,0.79296875,1.2734375,-0.52734375,1.03125,-0.97265625,0.31640625,-0.578125,0.34179688,-0.82421875,0.17285156,1.375,-1.140625,0.04663086,0.17480469,-0.38671875,-1.859375,-0.4453125,0.828125,-0.15234375,-0.40429688,0.43554688,-0.8125,-0.6171875,-0.609375,-0.34570312,0.6953125,0.42773438,-0.125,0.73046875,-1.0703125,-0.62890625,-1.8359375,-0.82421875,1.6484375,-0.35742188,-1.6328125,-1.1015625,-0.96484375,1.3125,0.25585938,-2.375,-0.58984375,-1.5625,2.0,0.13769531,-0.45507812,-1.3515625,-2.5625,1.1015625,-2.25,0.5859375,-0.092285156,0.64453125,0.35351562,-0.20507812,-0.76171875,-0.087402344,1.5546875,-1.578125,0.59375,0.60546875,-0.10253906,1.2890625,-0.061279297,0.41601562,0.9375,0.796875,0.41992188,-0.87890625,-0.6640625,-0.37109375,0.09082031,-2.6875,2.40625,0.22851562,-0.012207031,1.421875,-0.91796875,-1.9140625,-0.8671875,-1.1171875,-0.92578125,-0.44921875,1.5078125,1.1875,-1.0078125,-0.18554688,-0.91015625,0.44921875,1.3203125,0.50390625,-0.041992188,1.0078125,-1.109375,-0.13476562,1.609375,0.58203125,-1.3125,-0.23535156,1.3671875,1.2265625,0.3828125,-0.55078125,0.44140625,3.265625,-0.100097656,-0.18652344,0.44726562
+5,0,6,-0.43945312,-0.69921875,-0.38476562,-1.4296875,0.18261719,-1.203125,-0.9453125,1.6484375,0.064453125,0.8125,-0.8671875,-0.53125,1.4140625,0.87890625,-2.140625,0.92578125,-0.515625,-0.86328125,0.10058594,0.5078125,2.453125,-0.18164062,-0.84765625,0.23925781,0.484375,-0.37890625,-2.0625,-1.1328125,-0.421875,1.7421875,-0.26367188,0.50390625,-0.98046875,0.546875,-1.2578125,0.49609375,-0.37304688,-0.49609375,-0.27734375,-0.6015625,-0.94921875,0.33203125,-0.6953125,-0.28515625,0.5390625,2.5,-0.10595703,-0.31640625,1.390625,1.0625,-1.0,-2.421875,-2.328125,0.4375,-0.609375,1.671875,-0.28125,-0.80859375,-0.26367188,-1.5546875,-0.87890625,-0.80859375,1.5859375,0.9609375,1.4296875,2.015625,-1.671875,1.6328125,-0.96875,0.62109375,1.296875,-1.296875,-0.69921875,-0.26367188,-0.23632812,0.24609375,0.26367188,-0.33789062,0.7265625,0.73046875,1.5703125,-0.43359375,0.27929688,1.75,-0.84375,0.64453125,-0.8203125,0.049316406,0.13769531,-0.11621094,1.4765625,-0.36132812,1.7734375,-0.103515625,-0.14453125,0.025878906,-1.84375,0.78125,-0.5390625,0.375,-0.28710938,-0.26953125,0.35742188,0.17578125,0.3671875,1.3125,-1.671875,-2.625,-1.0703125,-0.62109375,-0.19824219,0.7734375,-1.0546875,-0.53125,-0.100097656,-2.375,-0.09667969,0.81640625,0.33007812,1.5859375,0.11230469,-0.3828125,-0.15625,-1.65625,-1.265625,0.63671875,-1.359375,-0.77734375,-0.78515625,0.4765625,2.0625,-0.75390625,-1.71875,-0.5546875,-0.28515625,0.83984375,-0.13671875,0.079589844,-0.100097656,-2.046875,-0.328125,1.609375,0.58984375,0.95703125,0.52734375,0.15820312,-0.5234375,0.20507812,-0.59375,1.0390625,0.034423828,-0.90234375,0.54296875,-0.53515625,0.4375,-0.1796875,1.40625,1.0078125,0.55078125,0.359375,-0.19433594,0.875,0.41601562,0.49804688,-1.7890625,1.7265625,0.734375,0.20996094,0.45898438,0.037597656,-0.49609375,0.63671875,-0.039794922,0.18457031,-1.1171875,1.9921875,0.18261719,-1.6640625,-0.5703125,-0.09667969,-0.66015625,0.875,0.91796875,0.3828125,0.28515625,-0.28515625,0.27734375,0.3984375,0.32226562,-0.7890625,-1.4609375,0.7578125,1.5,0.15039062,-0.55078125,0.27734375,2.375,-0.39453125,1.015625,1.109375
+5,0,7,0.90234375,0.56640625,0.828125,0.19921875,-1.140625,-0.07861328,-1.1953125,-0.034179688,2.609375,-1.453125,1.8359375,-0.03930664,0.34960938,1.3671875,-0.625,-0.42578125,-0.46484375,0.18164062,2.265625,1.1484375,0.24023438,-0.51171875,-0.15625,0.578125,-0.48242188,-0.53515625,0.6484375,-0.52734375,0.90625,0.5078125,-0.40625,-0.008117676,0.78125,-0.6796875,0.38085938,0.91796875,0.953125,-0.19042969,-0.21191406,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24707031,-0.234375,0.50390625,-0.29101562,0.87890625,0.6796875,-0.16015625,-0.265625,-0.74609375,0.95703125,0.9375,0.41015625,-0.65234375,-0.49804688,-0.578125,0.7734375,1.1953125,-0.37890625,1.2578125,0.099609375,0.640625,0.51953125,-0.20703125,-1.2578125,-1.46875,0.41210938,-0.111328125,1.3828125,0.78515625,-0.640625,-0.65234375,0.47851562,-0.52734375,0.48046875,-1.09375,1.2578125,-1.8828125,-0.70703125,0.9375,-1.1640625,-0.6953125,0.5078125,-0.36132812,-1.515625,0.48242188,0.28710938,1.5390625,1.2109375,-0.46875,0.87109375,-0.453125,0.16308594,-1.7890625,0.73828125,-0.640625,-0.36914062,-1.6875,0.609375,0.9140625,-1.296875,-0.34960938,2.171875,-0.6328125,-0.7890625,-0.375,0.78515625,-0.47265625,0.25976562,1.2265625,-0.58203125,0.26757812,0.34960938,-0.625,0.453125,-0.41210938,0.55859375,0.6875,-0.0054626465,0.4453125,-1.6796875,-1.9140625,0.55859375,-0.16894531,-0.19921875,-0.34765625,1.140625,1.0703125,0.61328125,-1.5078125,-0.265625,-1.09375,1.578125,0.16503906,-0.734375,-0.41601562,-1.75,1.6875,-0.26953125,-0.62890625,1.5078125,-0.28710938,-0.075683594,1.4609375,-0.73046875,-1.1875,0.22167969,-1.0859375,-0.16503906,2.15625,-0.578125,-0.609375,0.19628906,1.40625,-0.39648438,0.65625,0.390625,-0.75,-0.3046875,1.390625,1.2421875,-1.171875,-0.14550781,-0.6640625,-0.6171875,0.6484375,-1.6953125,-2.109375,-0.25585938,-2.984375,-0.40234375,-0.578125,2.484375,0.80078125,-1.0546875,-0.4765625,0.36132812,-0.33007812,-0.092285156,-0.068359375,-0.65625,-1.15625,0.20800781,1.1171875,0.58984375,-0.38671875,-0.8203125,-1.5234375,0.296875,0.546875,-0.8671875,-2.8125,1.2578125,3.171875,0.20019531,0.34375,0.85546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
index 184fba8e..a0c6dbf9 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,5,0.13769531,0.10595703,0.27148438,-0.8828125,1.3984375,0.6171875,0.48632812,1.03125,-0.49414062,-1.0390625,0.12792969,-0.26171875,0.58203125,0.35351562,-1.2578125,0.484375,-0.546875,-1.9140625,1.1328125,-0.5234375,-0.08544922,-0.6640625,1.2734375,0.52734375,-1.234375,-1.5078125,-0.9609375,-0.8671875,-1.7890625,0.43945312,-0.63671875,-0.26171875,0.013671875,-0.4609375,-0.52734375,0.953125,0.91796875,-0.73828125,-1.4609375,1.5703125,-1.7890625,0.03149414,-0.07714844,0.09423828,0.38085938,0.5234375,0.63671875,-0.69921875,0.37304688,1.859375,-1.1953125,-1.6015625,-0.97265625,1.1015625,-2.421875,0.092285156,-0.21484375,-0.8046875,1.328125,-1.71875,1.75,0.18554688,0.2890625,0.29882812,-0.515625,-0.067871094,-1.078125,0.16308594,0.546875,1.8515625,0.06298828,-0.31640625,0.57421875,-0.13769531,-0.671875,1.8671875,0.765625,1.1796875,-1.765625,-0.5234375,1.2890625,0.89453125,-0.12207031,0.12060547,-0.640625,-0.20117188,-0.46484375,1.4609375,-0.359375,0.99609375,1.3828125,0.21289062,-0.921875,0.36914062,-1.421875,0.22265625,-0.58984375,-0.640625,0.62109375,-0.890625,-2.359375,-0.17871094,2.5,0.78125,-0.79296875,0.0063171387,0.06640625,-1.0703125,0.48632812,0.984375,0.703125,0.03564453,0.5703125,-0.79296875,0.4921875,-0.3203125,-0.47460938,-0.3515625,-0.484375,-1.1953125,0.7109375,-0.7890625,0.60546875,-1.5625,-0.2265625,1.2578125,-1.3203125,-2.421875,0.014038086,0.17285156,-0.18652344,0.03955078,-0.021972656,-1.3671875,-1.0390625,-0.13769531,-0.66015625,0.25195312,0.20800781,-1.7265625,3.375,-0.40625,1.3203125,1.1015625,-1.2265625,1.078125,-0.6015625,-0.66796875,-2.515625,1.1015625,-0.30859375,-0.03540039,1.0625,-0.48828125,-1.1015625,-1.1953125,0.3828125,1.3515625,1.015625,1.421875,-0.26367188,-0.010314941,1.828125,1.25,-1.78125,1.65625,-1.671875,-0.24707031,2.109375,-0.20117188,0.3671875,-0.40039062,0.1015625,0.6171875,0.69140625,0.040283203,-0.8671875,-0.84375,0.859375,-0.609375,-0.30664062,-0.89453125,0.140625,0.85546875,0.14648438,0.5390625,0.59375,0.038085938,0.8671875,-1.828125,0.099609375,1.375,1.2578125,0.28515625,0.43945312,-0.18261719,1.6796875,0.15820312,1.4140625,0.5234375
-6,0,6,0.82421875,0.82421875,-0.3984375,0.84375,1.40625,0.62890625,-0.3515625,0.39453125,-0.10058594,-1.4375,1.1015625,-0.19433594,0.53515625,-0.54296875,-0.99609375,-0.375,-0.8203125,-0.40039062,1.109375,1.2421875,0.20117188,-1.890625,0.36132812,0.35742188,-0.90234375,-1.625,-1.3828125,-0.35351562,0.22753906,0.05859375,-0.23828125,-0.32421875,1.7109375,0.20800781,-0.328125,-0.17089844,0.31640625,-0.6015625,-0.26367188,0.78125,-0.6015625,0.71484375,-0.26953125,1.2734375,-0.27734375,0.25585938,-0.734375,-1.0234375,0.43945312,1.71875,1.140625,-1.765625,-0.4921875,-0.26757812,-1.0859375,-2.140625,0.38671875,1.3046875,-0.111328125,-1.015625,1.6171875,1.59375,1.734375,-0.34960938,-1.03125,-0.7109375,0.050048828,-2.078125,-0.3046875,0.85546875,0.3828125,0.16992188,1.3359375,-1.4921875,0.4609375,0.51953125,0.18554688,-0.20605469,-1.7109375,0.39453125,-0.2734375,0.71484375,-0.018432617,0.31835938,-0.7578125,0.7421875,-2.828125,0.453125,1.5859375,1.2109375,1.421875,1.1953125,1.125,0.22460938,-0.6328125,-0.93359375,-0.8671875,1.2421875,-0.23046875,0.12402344,-0.54296875,-0.39453125,0.89453125,-0.114746094,-1.0390625,0.77734375,-0.98828125,-0.18164062,-0.34960938,-0.33203125,0.18164062,1.96875,0.13378906,-0.63671875,-0.57421875,0.671875,-0.4140625,0.33203125,-0.46484375,0.84375,0.05078125,-1.1015625,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.6015625,-0.24121094,0.76171875,-0.04345703,-2.515625,-0.9375,0.7421875,-1.328125,2.546875,-0.31835938,0.01373291,1.8125,-1.3203125,1.421875,-0.75390625,0.12109375,1.171875,0.6640625,-0.20117188,-0.22460938,-0.27929688,-0.5625,0.20019531,0.43359375,-0.32226562,-1.3515625,1.015625,-1.1328125,1.234375,0.6953125,-0.36523438,-0.390625,1.046875,0.78515625,1.15625,1.0078125,3.046875,-2.671875,1.609375,-2.234375,1.1875,0.16796875,-0.296875,-1.7734375,-1.8359375,0.734375,-0.65234375,0.65625,0.27539062,-0.66796875,-1.96875,1.40625,-0.35742188,0.09814453,0.15039062,0.84375,-0.11621094,-1.0234375,1.578125,-0.4921875,-0.26953125,0.51171875,-0.36523438,0.66015625,1.546875,-0.31445312,-1.0390625,-0.82421875,-0.46679688,-0.46875,-0.19824219,0.19824219,0.51953125
-6,0,7,-0.91015625,0.03955078,-0.34375,1.078125,1.359375,-0.03100586,-0.43945312,1.0546875,0.59765625,-0.66015625,0.27539062,1.84375,-0.29296875,-0.6640625,-0.97265625,0.22949219,0.0019836426,-1.046875,1.9765625,-0.4765625,-0.123535156,-2.0625,-0.8203125,0.54296875,-0.12060547,-0.62890625,0.7109375,0.73828125,0.69140625,0.41015625,0.43554688,0.2734375,1.3203125,-0.52734375,2.140625,0.88671875,1.171875,-0.828125,-1.015625,0.6640625,-0.40625,-0.80859375,-1.859375,-1.875,-0.82421875,0.89453125,-0.3984375,-0.94140625,0.8515625,1.7578125,1.125,-0.5703125,-1.15625,0.31640625,-0.123046875,-2.03125,0.35742188,1.515625,0.43359375,0.15625,0.78125,0.83203125,-0.3515625,1.2890625,-1.3125,0.5546875,-1.3359375,-1.90625,0.47460938,-0.0546875,0.81640625,1.4453125,1.78125,-1.7734375,0.8125,0.921875,-0.08251953,-0.6953125,-0.28515625,0.9296875,-1.0390625,0.57421875,-0.29101562,-0.9140625,0.04296875,-1.3671875,-0.43554688,-0.19140625,1.5625,0.859375,1.4453125,-0.13769531,-0.20214844,1.0859375,-0.31835938,-0.27539062,-0.8203125,-0.9609375,-1.03125,-0.21777344,0.08300781,0.40625,1.390625,1.1796875,-1.0703125,1.3671875,0.61328125,0.19921875,-0.049316406,0.515625,0.75390625,0.83984375,2.5,-1.59375,0.76953125,-0.35351562,-1.5234375,0.671875,-0.83984375,-0.7890625,1.3125,0.46289062,-0.67578125,-0.64453125,-1.296875,2.03125,-0.06591797,0.76953125,0.8125,-1.2734375,1.4765625,-2.3125,0.15332031,0.19628906,-0.106933594,0.90234375,0.7265625,-0.12792969,0.41015625,-0.99609375,1.5625,0.265625,-1.453125,1.3828125,-0.36523438,-1.0234375,0.49414062,-0.31640625,-0.18652344,-1.2578125,0.78515625,0.22851562,2.671875,-0.93359375,-0.8203125,0.038330078,0.13867188,-1.578125,1.1875,0.12060547,0.63671875,0.6015625,1.125,0.48046875,0.0546875,0.7265625,0.023925781,0.984375,-0.33203125,-0.0034179688,-1.1875,-1.3984375,-2.359375,-1.28125,-0.4140625,0.9921875,-0.5,-1.5234375,-1.6328125,-0.953125,0.703125,0.16015625,0.25585938,-1.234375,-0.83203125,-1.21875,-0.25,0.19238281,-1.2578125,1.078125,-1.2265625,-0.03100586,-0.87109375,-1.1953125,-1.6171875,0.58203125,-0.18945312,0.21679688,0.20117188,1.609375
-7,0,5,0.8671875,1.2109375,-0.828125,-0.984375,-1.2890625,0.32421875,-0.5,1.515625,-0.7890625,-0.70703125,1.625,0.0040893555,-0.059326172,-0.28125,-1.15625,1.6328125,-0.28710938,-1.5234375,0.15136719,-0.66015625,1.3515625,0.84765625,0.48046875,1.125,0.8984375,-0.859375,-0.93359375,-1.9375,-1.6171875,-0.63671875,-1.2578125,2.453125,1.015625,2.515625,-0.171875,0.14648438,-0.40234375,-0.59765625,0.8671875,0.5625,-0.625,0.28125,0.63671875,0.20117188,1.3046875,0.36328125,0.578125,0.2734375,0.65625,-1.1796875,-0.84375,-0.51171875,-1.8984375,-1.609375,-1.03125,0.59765625,0.08300781,-0.33984375,0.3125,-0.75390625,-0.19140625,-0.515625,1.8515625,1.0,1.1796875,0.70703125,-0.88671875,-0.13867188,0.22558594,0.24902344,-0.036865234,-0.36132812,-0.671875,-0.1875,-0.4765625,-0.15820312,-0.23339844,-0.13183594,-0.01550293,0.45898438,-0.48242188,-0.546875,1.1328125,0.42382812,1.1171875,1.2890625,-0.97265625,0.52734375,1.375,-1.0234375,1.3359375,0.41015625,0.546875,1.21875,0.21972656,-1.3203125,0.76953125,0.99609375,-1.703125,-0.296875,-0.6484375,0.32421875,2.21875,-1.7734375,-1.21875,-1.4765625,-2.015625,-1.7109375,-0.47851562,0.9296875,0.20996094,-1.0078125,0.088378906,0.5703125,0.53125,0.20117188,0.0040283203,-0.98046875,-0.26757812,0.22851562,-0.36914062,-0.3125,-1.4375,-0.99609375,-1.4609375,1.0625,-1.75,-0.55078125,1.0390625,0.984375,0.9140625,-0.484375,-1.140625,-0.6484375,-1.765625,0.8203125,1.109375,-0.1875,-1.6640625,-2.4375,1.484375,-0.55078125,-0.640625,0.59375,0.4765625,0.18652344,-0.25,-0.5234375,0.72265625,1.5390625,-0.75390625,0.2734375,1.2734375,-1.21875,0.6796875,0.30859375,0.59375,-0.859375,0.7734375,-1.1484375,0.80078125,-0.40625,0.64453125,-0.64453125,-1.1328125,3.453125,-0.76171875,-1.1796875,0.328125,-1.015625,-1.0,-0.47070312,-1.2890625,-1.1171875,-0.38085938,1.1796875,0.7890625,-0.72265625,1.046875,-0.49804688,0.390625,0.34960938,0.73046875,-1.2890625,1.1484375,-0.053955078,0.08984375,0.28125,0.24609375,0.38085938,2.03125,-0.14257812,-0.033447266,-0.7265625,-0.0119018555,0.8359375,2.0625,1.796875,0.8515625,0.66796875
-7,0,6,0.5859375,1.3984375,-0.61328125,-0.51953125,-1.9140625,-0.7578125,1.0234375,0.9765625,0.18945312,0.011169434,0.115234375,0.43945312,-0.020263672,-0.828125,-0.51171875,1.8046875,-1.2578125,-0.82421875,0.34960938,0.27734375,0.45703125,-0.60546875,1.046875,0.3671875,1.28125,-0.51171875,-1.984375,-0.71875,0.5,-0.62890625,-0.73828125,-0.008300781,2.21875,-0.43554688,-0.625,-0.076660156,0.25976562,-1.28125,-0.51953125,-0.6796875,-0.4296875,0.18847656,0.29882812,-0.09814453,-0.73046875,0.15429688,0.37890625,0.66015625,-0.28125,-0.20117188,0.89453125,0.546875,-3.75,-0.15429688,1.15625,1.71875,-0.14941406,0.34179688,0.47265625,-0.095703125,-1.046875,-1.1171875,1.0390625,0.33007812,1.6015625,-0.375,-0.234375,-0.34765625,0.040771484,0.35351562,-0.32421875,1.125,-0.04248047,-0.98828125,2.453125,0.83203125,0.88671875,-0.12988281,-1.4375,0.34375,0.59375,-0.28125,1.53125,0.125,0.37304688,0.0019226074,-1.203125,1.1328125,1.6015625,-0.24707031,0.875,-0.3515625,0.99609375,0.5234375,-1.2734375,0.57421875,-0.83984375,1.828125,-1.3984375,0.35742188,0.33984375,0.97265625,-0.5625,0.71875,-0.62109375,-0.2109375,-1.6875,-1.75,-0.29101562,0.33398438,-1.46875,-2.296875,0.3046875,-0.20117188,-1.4140625,1.15625,-0.671875,1.03125,-1.140625,-0.546875,-0.12988281,0.41210938,-1.203125,-1.3359375,0.49023438,0.76171875,-1.625,-0.09667969,-0.18652344,-1.125,1.140625,-0.23144531,-0.11035156,-2.171875,-0.76171875,-0.66796875,1.6484375,0.048339844,-1.4296875,-2.796875,1.4453125,0.15136719,-1.390625,-1.21875,-0.328125,-0.44921875,0.40039062,0.77734375,2.25,1.96875,-0.5390625,-1.0703125,0.17382812,-0.31835938,-0.23925781,-1.2734375,1.234375,0.084472656,0.9921875,0.099121094,-0.27539062,-0.071777344,0.22949219,-0.40625,-1.15625,0.625,-0.515625,-1.375,0.57421875,0.46679688,0.84765625,0.61328125,-0.80078125,0.80859375,-1.59375,1.1328125,2.75,-0.35351562,1.6796875,1.8984375,-0.25195312,0.6484375,0.62109375,-1.0234375,0.045898438,-1.046875,0.4921875,-0.08691406,0.83984375,-0.041259766,0.18652344,0.96875,-0.012329102,0.3125,-1.09375,-0.20605469,0.014526367,2.0,1.0703125,0.028442383
-7,0,7,0.96484375,0.67578125,-1.0078125,-0.94921875,0.052734375,-0.69921875,-0.15136719,1.9921875,0.83203125,0.625,-1.328125,1.8359375,-0.36328125,1.1484375,0.020385742,0.7734375,0.62109375,-0.359375,0.87109375,1.421875,0.23535156,-0.052490234,-0.75390625,-0.328125,-0.6171875,1.3828125,-0.6484375,0.028320312,0.8984375,-2.203125,0.5625,1.7890625,1.1796875,0.16015625,-0.59765625,-0.41601562,0.22558594,1.4609375,0.09033203,-0.10644531,-0.984375,-0.07714844,-0.53515625,0.49414062,0.265625,0.049804688,-0.23828125,-0.33398438,0.06640625,0.17578125,0.099609375,-1.3828125,-0.359375,-0.7421875,0.068847656,0.50390625,-1.6484375,0.15234375,0.28125,-0.63671875,-1.1484375,-0.68359375,1.6171875,-0.34179688,0.45117188,0.55859375,0.25585938,-0.18066406,-0.29882812,1.0390625,0.21972656,2.109375,-0.87109375,-0.7421875,0.8671875,1.5859375,0.16894531,-0.81640625,-2.859375,0.7109375,1.609375,0.578125,1.84375,2.0625,-0.66796875,0.8046875,-0.6796875,-0.65234375,-0.5625,1.375,1.1484375,1.0,-0.16894531,1.2890625,0.021972656,0.111328125,-1.6328125,1.03125,-0.13085938,0.453125,0.66796875,-0.50390625,-0.48828125,-0.34570312,-0.6015625,0.94140625,-0.34570312,-0.91015625,-1.71875,-0.87109375,-0.17871094,-0.47070312,1.5859375,-0.83203125,-0.15234375,-1.0859375,0.21191406,0.19140625,0.8984375,0.80859375,1.2265625,0.06738281,-1.0703125,-0.77734375,-0.47070312,0.734375,-0.17871094,-0.234375,1.2265625,-0.5390625,0.8984375,-1.859375,0.29101562,-1.1640625,-1.359375,0.7734375,-1.578125,0.16699219,0.22167969,-2.0625,0.072265625,-0.13183594,-1.359375,1.2421875,-0.24023438,0.06738281,-0.53515625,1.71875,0.22167969,-1.875,1.5703125,-2.09375,0.34765625,0.20410156,0.1484375,0.46875,1.4140625,0.67578125,-0.54296875,0.23144531,-1.5703125,2.25,-0.83203125,0.796875,1.1640625,-0.78515625,-0.484375,-2.859375,0.23730469,1.2421875,-0.55859375,0.7421875,-1.1171875,1.3203125,0.09326172,1.703125,0.22558594,-1.4921875,-0.765625,0.8203125,-0.609375,0.51953125,-0.41992188,0.47460938,-1.7109375,0.29882812,-1.6953125,-2.046875,1.484375,-0.98046875,-1.5703125,-0.92578125,-1.5390625,0.48046875,0.072265625,0.06640625,0.24023438,-0.37890625,-0.85546875,1.265625
+6,0,5,0.13964844,0.10888672,0.27539062,-0.88671875,1.390625,0.61328125,0.48828125,1.0234375,-0.4921875,-1.046875,0.12597656,-0.26757812,0.57421875,0.35351562,-1.25,0.48632812,-0.546875,-1.9140625,1.1328125,-0.53125,-0.087402344,-0.6640625,1.28125,0.5234375,-1.234375,-1.5078125,-0.953125,-0.8671875,-1.7890625,0.44335938,-0.64453125,-0.26171875,0.010559082,-0.4609375,-0.5234375,0.953125,0.91796875,-0.73828125,-1.4609375,1.578125,-1.8046875,0.030517578,-0.07519531,0.09814453,0.38671875,0.5234375,0.63671875,-0.69921875,0.38085938,1.8515625,-1.1953125,-1.6015625,-0.98046875,1.1015625,-2.421875,0.091796875,-0.21289062,-0.80859375,1.328125,-1.6953125,1.734375,0.18359375,0.29296875,0.30273438,-0.5078125,-0.06640625,-1.0703125,0.16210938,0.546875,1.84375,0.05517578,-0.31835938,0.57421875,-0.13574219,-0.66796875,1.8515625,0.76953125,1.1796875,-1.7734375,-0.5234375,1.28125,0.89453125,-0.11816406,0.12207031,-0.63671875,-0.20117188,-0.46484375,1.4453125,-0.36132812,0.98828125,1.3828125,0.21386719,-0.921875,0.36523438,-1.4296875,0.22363281,-0.59375,-0.640625,0.625,-0.890625,-2.375,-0.18066406,2.515625,0.78125,-0.79296875,0.0051574707,0.063964844,-1.0703125,0.49609375,0.98828125,0.703125,0.036376953,0.56640625,-0.796875,0.49414062,-0.31640625,-0.47851562,-0.34765625,-0.484375,-1.1796875,0.71484375,-0.78515625,0.60546875,-1.5703125,-0.22460938,1.265625,-1.3203125,-2.40625,0.013122559,0.17382812,-0.18652344,0.044921875,-0.025512695,-1.3671875,-1.03125,-0.13671875,-0.65234375,0.25390625,0.20507812,-1.7265625,3.375,-0.40429688,1.3125,1.109375,-1.234375,1.0703125,-0.6015625,-0.671875,-2.515625,1.1015625,-0.30859375,-0.03564453,1.0625,-0.484375,-1.078125,-1.2109375,0.38085938,1.359375,1.015625,1.4375,-0.26367188,-0.011962891,1.8203125,1.2421875,-1.78125,1.6640625,-1.671875,-0.24511719,2.125,-0.20117188,0.36914062,-0.40234375,0.099609375,0.62109375,0.68359375,0.041992188,-0.86328125,-0.84765625,0.859375,-0.61328125,-0.3046875,-0.890625,0.13964844,0.86328125,0.1484375,0.54296875,0.59375,0.037841797,0.87109375,-1.8046875,0.09863281,1.375,1.2578125,0.28710938,0.43554688,-0.17871094,1.6875,0.15625,1.40625,0.51953125
+6,0,6,0.828125,0.83203125,-0.3984375,0.84375,1.40625,0.62890625,-0.3515625,0.39257812,-0.10253906,-1.4375,1.109375,-0.19335938,0.53125,-0.54296875,-0.99609375,-0.37109375,-0.81640625,-0.39648438,1.1171875,1.25,0.20410156,-1.8984375,0.36132812,0.35742188,-0.90234375,-1.609375,-1.3828125,-0.34960938,0.22753906,0.057861328,-0.23632812,-0.328125,1.71875,0.20703125,-0.33007812,-0.16992188,0.31445312,-0.6015625,-0.26367188,0.78125,-0.6015625,0.71875,-0.26757812,1.2578125,-0.27929688,0.25390625,-0.734375,-1.0234375,0.43554688,1.7109375,1.140625,-1.7734375,-0.49414062,-0.26757812,-1.0859375,-2.140625,0.3828125,1.3046875,-0.11279297,-1.015625,1.609375,1.59375,1.734375,-0.34570312,-1.0390625,-0.71875,0.05029297,-2.078125,-0.30078125,0.85546875,0.38476562,0.17382812,1.3515625,-1.4921875,0.45898438,0.51953125,0.18359375,-0.20214844,-1.71875,0.39648438,-0.2734375,0.71875,-0.016235352,0.32226562,-0.76171875,0.734375,-2.828125,0.45507812,1.578125,1.2109375,1.4296875,1.203125,1.125,0.22558594,-0.6328125,-0.9375,-0.87109375,1.25,-0.22753906,0.123535156,-0.54296875,-0.40039062,0.890625,-0.11279297,-1.0390625,0.78125,-0.98828125,-0.18164062,-0.34765625,-0.328125,0.18261719,1.96875,0.13378906,-0.63671875,-0.57421875,0.6796875,-0.4140625,0.33398438,-0.46484375,0.83984375,0.052490234,-1.09375,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.59765625,-0.24511719,0.765625,-0.04272461,-2.5,-0.93359375,0.73828125,-1.3125,2.546875,-0.31445312,0.013977051,1.8203125,-1.3203125,1.4296875,-0.75,0.12158203,1.171875,0.6640625,-0.20117188,-0.22558594,-0.28320312,-0.56640625,0.203125,0.4296875,-0.32226562,-1.34375,1.0234375,-1.140625,1.2265625,0.69921875,-0.36523438,-0.39257812,1.0546875,0.78515625,1.1484375,1.0078125,3.03125,-2.671875,1.6171875,-2.25,1.1796875,0.16796875,-0.29492188,-1.7734375,-1.84375,0.7265625,-0.6484375,0.65625,0.27734375,-0.66796875,-1.96875,1.40625,-0.35351562,0.09667969,0.15332031,0.83984375,-0.11230469,-1.015625,1.578125,-0.49609375,-0.26757812,0.5078125,-0.36914062,0.65625,1.546875,-0.31835938,-1.046875,-0.8125,-0.46289062,-0.46875,-0.19726562,0.19726562,0.51953125
+6,0,7,-0.9140625,0.041748047,-0.34179688,1.09375,1.3671875,-0.032226562,-0.43945312,1.0625,0.59765625,-0.66015625,0.27539062,1.84375,-0.2890625,-0.65625,-0.97265625,0.22949219,-0.00076293945,-1.046875,1.96875,-0.48046875,-0.123046875,-2.0625,-0.8125,0.53515625,-0.11816406,-0.62890625,0.71484375,0.73828125,0.68359375,0.40820312,0.44140625,0.2734375,1.328125,-0.515625,2.125,0.88671875,1.171875,-0.82421875,-1.015625,0.6640625,-0.40429688,-0.80859375,-1.8671875,-1.875,-0.828125,0.890625,-0.3984375,-0.94140625,0.86328125,1.75,1.140625,-0.5703125,-1.1484375,0.31054688,-0.12109375,-2.03125,0.35546875,1.515625,0.43359375,0.15722656,0.78125,0.83203125,-0.3515625,1.2890625,-1.3203125,0.5546875,-1.3359375,-1.9140625,0.47851562,-0.049316406,0.8046875,1.4453125,1.765625,-1.7734375,0.8203125,0.921875,-0.08251953,-0.69140625,-0.2890625,0.92578125,-1.03125,0.5703125,-0.28710938,-0.91015625,0.04296875,-1.359375,-0.4375,-0.19238281,1.5625,0.859375,1.4453125,-0.13574219,-0.20117188,1.0859375,-0.3203125,-0.27734375,-0.8203125,-0.9609375,-1.0234375,-0.21289062,0.083984375,0.40625,1.390625,1.1875,-1.0703125,1.3671875,0.61328125,0.19824219,-0.049560547,0.5234375,0.74609375,0.84375,2.484375,-1.578125,0.77734375,-0.35546875,-1.53125,0.66796875,-0.83984375,-0.7890625,1.328125,0.4609375,-0.671875,-0.64453125,-1.296875,2.015625,-0.06689453,0.7578125,0.8203125,-1.28125,1.484375,-2.328125,0.15722656,0.19335938,-0.110839844,0.8984375,0.72265625,-0.12988281,0.41210938,-1.0,1.5703125,0.26367188,-1.4453125,1.3828125,-0.36523438,-1.015625,0.49804688,-0.31640625,-0.1875,-1.25,0.7890625,0.22851562,2.671875,-0.921875,-0.8125,0.0390625,0.140625,-1.59375,1.1953125,0.12207031,0.63671875,0.59765625,1.1328125,0.47851562,0.0546875,0.73046875,0.020141602,0.98046875,-0.33398438,-0.005218506,-1.1953125,-1.40625,-2.359375,-1.2890625,-0.41796875,0.99609375,-0.49804688,-1.53125,-1.640625,-0.953125,0.7109375,0.16601562,0.25585938,-1.234375,-0.83984375,-1.21875,-0.25,0.19042969,-1.2578125,1.078125,-1.2109375,-0.030517578,-0.8671875,-1.1875,-1.609375,0.578125,-0.1875,0.21386719,0.203125,1.609375
+7,0,5,0.8671875,1.2109375,-0.828125,-0.984375,-1.28125,0.32617188,-0.5,1.5234375,-0.7890625,-0.71484375,1.625,0.010192871,-0.060791016,-0.28125,-1.15625,1.640625,-0.28515625,-1.5234375,0.15039062,-0.66796875,1.34375,0.84375,0.47851562,1.125,0.88671875,-0.85546875,-0.9296875,-1.9453125,-1.6171875,-0.62890625,-1.2578125,2.453125,1.015625,2.515625,-0.16992188,0.14550781,-0.40234375,-0.59765625,0.86328125,0.5546875,-0.62890625,0.28125,0.640625,0.20019531,1.3046875,0.36328125,0.58203125,0.27539062,0.6484375,-1.1796875,-0.83984375,-0.5078125,-1.890625,-1.609375,-1.03125,0.59375,0.08642578,-0.33789062,0.31445312,-0.75,-0.19042969,-0.515625,1.8515625,1.0,1.1796875,0.7109375,-0.890625,-0.13964844,0.22460938,0.24707031,-0.03857422,-0.36132812,-0.6796875,-0.18847656,-0.47265625,-0.15332031,-0.23339844,-0.13085938,-0.014709473,0.45898438,-0.48046875,-0.5390625,1.1328125,0.4296875,1.109375,1.2890625,-0.97265625,0.53125,1.359375,-1.015625,1.3359375,0.40625,0.546875,1.2109375,0.21875,-1.3125,0.765625,1.0,-1.71875,-0.29882812,-0.64453125,0.328125,2.234375,-1.7734375,-1.2265625,-1.484375,-2.0,-1.7265625,-0.48242188,0.93359375,0.21289062,-1.0078125,0.09082031,0.57421875,0.52734375,0.203125,0.008850098,-0.984375,-0.265625,0.22753906,-0.36914062,-0.31054688,-1.4375,-1.0,-1.46875,1.0625,-1.7578125,-0.55078125,1.0390625,0.98828125,0.90625,-0.484375,-1.140625,-0.64453125,-1.7578125,0.83203125,1.109375,-0.18359375,-1.6640625,-2.421875,1.4765625,-0.55078125,-0.640625,0.59765625,0.4765625,0.18261719,-0.24902344,-0.52734375,0.71875,1.5390625,-0.75390625,0.27539062,1.2734375,-1.2109375,0.67578125,0.3125,0.59375,-0.86328125,0.77734375,-1.1328125,0.796875,-0.40234375,0.64453125,-0.6484375,-1.140625,3.46875,-0.7578125,-1.1953125,0.33007812,-1.0234375,-1.0,-0.47460938,-1.28125,-1.1171875,-0.38085938,1.171875,0.78515625,-0.72265625,1.0390625,-0.49804688,0.39453125,0.3515625,0.73046875,-1.2890625,1.1484375,-0.053710938,0.08935547,0.28515625,0.25,0.3828125,2.03125,-0.14257812,-0.033203125,-0.7265625,-0.011657715,0.84375,2.0625,1.796875,0.84765625,0.66796875
+7,0,6,0.58984375,1.3984375,-0.609375,-0.51171875,-1.9140625,-0.7421875,1.0234375,0.9765625,0.18652344,0.0099487305,0.11669922,0.44335938,-0.01940918,-0.82421875,-0.51953125,1.796875,-1.2578125,-0.82421875,0.34960938,0.27929688,0.45703125,-0.6015625,1.046875,0.3671875,1.2734375,-0.50390625,-1.984375,-0.71875,0.5,-0.6328125,-0.73828125,-0.008544922,2.21875,-0.43359375,-0.62109375,-0.07519531,0.25976562,-1.28125,-0.515625,-0.6875,-0.42773438,0.18847656,0.29882812,-0.09863281,-0.73046875,0.15234375,0.37890625,0.6640625,-0.27734375,-0.20019531,0.89453125,0.546875,-3.765625,-0.15625,1.15625,1.71875,-0.15332031,0.34375,0.47070312,-0.100097656,-1.0546875,-1.1171875,1.0390625,0.328125,1.6015625,-0.375,-0.23828125,-0.34765625,0.04638672,0.35546875,-0.32617188,1.125,-0.045654297,-0.9921875,2.4375,0.83203125,0.88671875,-0.12792969,-1.4453125,0.34179688,0.59765625,-0.27929688,1.5234375,0.123046875,0.37304688,0.003616333,-1.203125,1.125,1.59375,-0.24609375,0.87890625,-0.34570312,1.0,0.52734375,-1.2734375,0.57421875,-0.84375,1.8203125,-1.3984375,0.35742188,0.33789062,0.9765625,-0.5546875,0.71484375,-0.625,-0.21484375,-1.6875,-1.75,-0.29296875,0.33398438,-1.4609375,-2.3125,0.31445312,-0.20410156,-1.4140625,1.15625,-0.66796875,1.0390625,-1.140625,-0.54296875,-0.13085938,0.41210938,-1.203125,-1.328125,0.484375,0.765625,-1.6328125,-0.09716797,-0.18652344,-1.1328125,1.1484375,-0.23730469,-0.110839844,-2.1875,-0.765625,-0.6640625,1.65625,0.048095703,-1.421875,-2.78125,1.453125,0.15136719,-1.3984375,-1.203125,-0.328125,-0.44921875,0.40625,0.77734375,2.21875,1.9765625,-0.5390625,-1.0703125,0.17578125,-0.3203125,-0.23925781,-1.2734375,1.234375,0.08105469,0.99609375,0.09765625,-0.27148438,-0.072753906,0.22949219,-0.40234375,-1.15625,0.62890625,-0.515625,-1.375,0.58203125,0.46875,0.84375,0.61328125,-0.796875,0.80859375,-1.6015625,1.1328125,2.75,-0.35546875,1.6796875,1.90625,-0.25390625,0.65234375,0.625,-1.0234375,0.047607422,-1.03125,0.49023438,-0.087890625,0.84375,-0.04272461,0.18847656,0.96875,-0.0099487305,0.3125,-1.09375,-0.20214844,0.014343262,1.9921875,1.0703125,0.028198242
+7,0,7,0.9609375,0.671875,-1.0234375,-0.95703125,0.057128906,-0.6953125,-0.15332031,1.9921875,0.82421875,0.625,-1.3359375,1.84375,-0.36328125,1.15625,0.025634766,0.765625,0.62890625,-0.35546875,0.87890625,1.4140625,0.22949219,-0.049072266,-0.74609375,-0.33203125,-0.6171875,1.3828125,-0.64453125,0.028442383,0.8984375,-2.203125,0.5703125,1.796875,1.1796875,0.16113281,-0.59375,-0.4140625,0.22558594,1.4609375,0.080078125,-0.107421875,-0.9765625,-0.07910156,-0.53125,0.49023438,0.26171875,0.05053711,-0.24121094,-0.33203125,0.071777344,0.18066406,0.09472656,-1.390625,-0.35742188,-0.73828125,0.07324219,0.49804688,-1.6484375,0.15332031,0.28515625,-0.64453125,-1.1484375,-0.6875,1.6171875,-0.33789062,0.45117188,0.55859375,0.25195312,-0.18261719,-0.29492188,1.046875,0.22167969,2.078125,-0.859375,-0.7421875,0.86328125,1.5859375,0.16601562,-0.81640625,-2.84375,0.70703125,1.6015625,0.578125,1.859375,2.0625,-0.66796875,0.80078125,-0.6796875,-0.65625,-0.55859375,1.375,1.140625,1.0,-0.16992188,1.2890625,0.020141602,0.111816406,-1.640625,1.03125,-0.12890625,0.45507812,0.66796875,-0.50390625,-0.484375,-0.34375,-0.59765625,0.9375,-0.3359375,-0.9140625,-1.71875,-0.875,-0.17871094,-0.47265625,1.59375,-0.83203125,-0.15039062,-1.078125,0.21484375,0.19238281,0.8984375,0.8046875,1.234375,0.06225586,-1.0703125,-0.78515625,-0.46679688,0.73828125,-0.17675781,-0.23925781,1.2265625,-0.53125,0.8984375,-1.859375,0.29492188,-1.1640625,-1.359375,0.765625,-1.578125,0.16601562,0.22460938,-2.046875,0.071777344,-0.13378906,-1.375,1.234375,-0.23828125,0.06689453,-0.53515625,1.7109375,0.21679688,-1.875,1.5859375,-2.109375,0.34375,0.20410156,0.15039062,0.46679688,1.421875,0.68359375,-0.54296875,0.23535156,-1.578125,2.234375,-0.8359375,0.796875,1.1484375,-0.78515625,-0.484375,-2.875,0.23535156,1.25,-0.55859375,0.73828125,-1.1171875,1.3203125,0.091308594,1.703125,0.23144531,-1.4921875,-0.765625,0.82421875,-0.61328125,0.51953125,-0.41601562,0.47070312,-1.7109375,0.29882812,-1.6796875,-2.046875,1.484375,-0.9765625,-1.5703125,-0.92578125,-1.5234375,0.48242188,0.06982422,0.064941406,0.24121094,-0.37890625,-0.859375,1.265625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
index 3da2498e..0aac125f 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,5,1.0703125,0.25,-0.16601562,-0.29296875,0.10986328,-0.42382812,-1.28125,0.7890625,0.23828125,0.46875,0.1328125,0.921875,1.2421875,-0.13574219,-0.56640625,-0.12792969,-1.078125,-2.234375,0.93359375,0.21386719,1.78125,-1.03125,-1.2890625,-0.20898438,1.4140625,-0.30078125,-0.71484375,-0.010803223,0.25585938,-0.14355469,-0.84375,-0.61328125,1.953125,-0.94140625,-0.9296875,-0.41601562,0.09277344,-1.140625,-2.59375,-0.46484375,0.4453125,-0.87890625,-0.70703125,-0.29882812,-1.2265625,0.5234375,-0.31835938,0.50390625,1.7578125,1.4375,-0.27148438,0.359375,-1.6640625,1.2578125,-0.103515625,0.38085938,-0.5,1.8828125,0.7109375,-0.82421875,-0.69140625,-1.9453125,0.76953125,0.80078125,1.328125,1.46875,-1.4609375,-1.859375,0.41601562,0.94140625,1.421875,1.21875,-0.41601562,-0.94921875,2.90625,1.78125,-0.35546875,-0.111816406,-0.49804688,0.27148438,-0.34179688,0.26953125,-0.0005874634,2.359375,-0.48046875,-0.58984375,-0.4140625,2.5,1.0546875,0.92578125,0.8125,0.54296875,0.81640625,0.06689453,-0.84375,-0.13574219,0.5546875,1.0390625,0.51171875,-0.94140625,0.19335938,1.0390625,-0.46875,0.36914062,-0.16503906,0.76953125,-1.171875,-1.140625,0.37695312,0.33203125,-1.015625,-0.15625,-0.55859375,-1.9453125,-1.78125,0.103515625,-0.625,1.2109375,-1.265625,0.4453125,0.61328125,0.41210938,-2.21875,-1.359375,-0.8046875,0.18652344,-0.5625,0.43359375,0.6640625,-0.6328125,1.7734375,-0.25,0.14355469,-1.5,-0.65625,0.5703125,0.5546875,0.19335938,-0.74609375,-2.5,1.078125,2.578125,-0.71484375,-1.21875,-1.265625,1.21875,0.32421875,-0.15722656,0.32226562,0.60546875,0.55859375,0.98046875,0.6328125,-0.6171875,-0.78125,-1.453125,0.28710938,-0.05053711,1.171875,-0.65625,0.36132812,0.57421875,0.075683594,-1.0625,0.5859375,-0.28515625,0.578125,-0.74609375,-1.28125,1.265625,0.14160156,0.65234375,-0.90234375,-0.53515625,-1.7109375,1.84375,-0.17773438,-0.5859375,0.079589844,-0.484375,1.25,0.40429688,0.8203125,-0.6796875,-1.0703125,-1.5625,0.18847656,0.36328125,0.69140625,0.625,-0.29882812,0.3671875,0.80078125,-0.93359375,-0.640625,-1.8515625,-0.6953125,1.2578125,1.0234375,0.7265625
-8,0,6,0.87109375,0.9921875,-0.1796875,0.12597656,0.08154297,-0.609375,0.7578125,1.0703125,1.0703125,-0.796875,1.328125,1.546875,0.051757812,-1.03125,-0.14355469,0.296875,-0.6796875,-1.0078125,0.83984375,0.33398438,0.026855469,-0.36523438,-0.24609375,1.625,-0.18261719,0.14550781,-2.78125,0.21289062,0.3125,0.7890625,-0.5078125,-0.8984375,1.3671875,-0.7734375,0.578125,0.59375,0.53125,-0.94140625,-0.22167969,1.078125,-1.203125,1.1171875,-0.33984375,-2.40625,-0.8671875,-0.78515625,0.26171875,-0.22851562,1.296875,1.6015625,0.55859375,-0.47460938,-0.92578125,0.67578125,0.47460938,0.265625,0.19921875,-0.515625,-1.0625,-0.033935547,1.46875,0.2109375,0.6328125,0.859375,0.50390625,-0.43359375,-0.14160156,0.5625,-1.359375,0.67578125,-0.099121094,0.54296875,0.84765625,-0.42578125,1.8359375,0.8046875,1.390625,1.0546875,-0.35351562,0.37304688,-0.083984375,-1.1328125,0.83984375,-1.4453125,-0.45507812,0.36523438,-0.03149414,-0.036132812,1.7578125,0.19042969,1.3515625,-0.5390625,-1.03125,1.25,-1.1015625,-0.59375,-2.734375,0.49414062,-0.59765625,-0.3671875,-1.4609375,1.1640625,0.37890625,0.19628906,-0.59765625,1.8359375,-0.75390625,0.16796875,-1.15625,0.30859375,-0.56640625,-0.42773438,0.91796875,-0.25976562,0.36328125,1.765625,-0.5078125,0.5546875,-0.16015625,-0.44335938,0.578125,-0.625,-0.49023438,-0.9375,-1.328125,2.109375,-1.265625,-1.046875,0.71484375,-1.28125,1.3203125,-1.21875,-0.6875,-0.76953125,-1.6015625,0.6484375,1.09375,-0.52734375,0.05908203,-2.46875,1.703125,-0.94921875,-1.765625,1.375,0.7109375,-0.04321289,0.8671875,-0.89453125,-1.1640625,0.9140625,-1.2890625,-0.703125,0.68359375,-1.015625,1.1171875,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.06933594,1.65625,2.453125,1.453125,-1.5234375,0.099121094,-0.54296875,0.08496094,0.859375,-0.15527344,-1.015625,-0.29492188,-1.640625,-0.14941406,-1.5390625,1.0390625,0.90625,-1.1640625,0.29492188,0.049804688,-1.2734375,-1.40625,0.53515625,-1.0,-0.45507812,0.24609375,0.021606445,0.56640625,-1.6171875,-1.6328125,-0.23535156,2.46875,-0.36914062,-0.97265625,-1.5625,1.03125,1.4921875,1.09375,-0.41992188,-0.59375
-8,0,7,1.2578125,1.5546875,0.1015625,-0.2578125,-1.5859375,-0.019165039,-1.328125,1.4375,1.4140625,-2.65625,1.7578125,-0.69140625,0.03100586,1.1171875,-0.4921875,0.46875,0.2890625,0.40625,1.859375,-0.015991211,-0.8046875,-0.036132812,-0.390625,-0.25,-0.86328125,-1.0625,-0.71484375,-0.22070312,0.099121094,-0.20605469,-0.25195312,0.016601562,1.0546875,0.71484375,0.09716797,0.29101562,1.40625,0.14648438,-0.88671875,1.6953125,-2.3125,-0.22265625,0.114746094,2.140625,0.25976562,0.91796875,-0.87109375,0.7109375,-0.061767578,0.88671875,-0.64453125,-0.08984375,-0.3515625,-0.24902344,-0.41601562,-0.45507812,0.19433594,0.80859375,-1.1328125,0.16601562,0.546875,0.65625,1.65625,-0.40820312,1.3671875,0.43164062,-0.36328125,-1.265625,-0.98046875,0.079589844,-0.27148438,-0.23730469,0.6171875,0.037841797,-0.91015625,0.734375,0.14160156,0.34375,-1.53125,1.3203125,-0.81640625,-0.66015625,0.953125,-1.2265625,-0.18066406,0.33398438,0.24511719,-1.3046875,0.103515625,0.51171875,1.9453125,1.5859375,-0.35546875,1.0546875,-0.89453125,-0.14648438,-0.68359375,0.004760742,0.19335938,-0.703125,-2.390625,0.87890625,1.7265625,-1.7421875,0.03515625,1.546875,-1.1171875,-0.46289062,0.0546875,0.20410156,-0.08935547,-0.3125,1.3359375,-0.73046875,0.06542969,0.080566406,-0.3828125,0.24902344,-0.25390625,0.859375,1.6484375,-0.56640625,-0.15429688,-1.796875,-0.515625,0.20117188,0.39257812,-1.5625,-0.11816406,1.390625,-0.890625,1.109375,-1.3984375,0.6484375,-2.078125,1.515625,0.90625,-0.59765625,-0.7109375,-2.328125,1.5859375,-1.1640625,-0.76171875,0.033691406,-0.19140625,0.20507812,0.48046875,-0.59375,-0.7578125,1.234375,-1.3984375,0.296875,0.58203125,-0.119628906,-0.31835938,0.09667969,1.1953125,0.34570312,0.52734375,0.037353516,-0.67578125,0.16699219,0.4375,0.13085938,-1.2578125,0.83984375,-0.6015625,0.6796875,0.66796875,-2.015625,-2.109375,-1.3359375,-2.171875,-0.49023438,0.34765625,2.296875,0.36914062,-0.52734375,0.037597656,-0.4453125,0.64453125,-0.27929688,-0.2578125,0.20898438,-0.38476562,0.14160156,0.91796875,1.4375,-0.47265625,0.045410156,0.080566406,0.25976562,0.78125,-1.4140625,-3.125,1.2578125,3.359375,0.64453125,-0.44726562,0.49609375
+8,0,5,1.078125,0.25195312,-0.16601562,-0.296875,0.109375,-0.42578125,-1.28125,0.7890625,0.23632812,0.46484375,0.13574219,0.921875,1.2421875,-0.13476562,-0.55859375,-0.13183594,-1.078125,-2.25,0.9375,0.21386719,1.7890625,-1.0234375,-1.2890625,-0.2109375,1.40625,-0.30078125,-0.71484375,-0.013305664,0.2578125,-0.14648438,-0.84765625,-0.61328125,1.9453125,-0.94921875,-0.93359375,-0.41601562,0.09472656,-1.140625,-2.578125,-0.453125,0.44726562,-0.87890625,-0.71484375,-0.29882812,-1.21875,0.5234375,-0.3203125,0.5,1.765625,1.4296875,-0.27148438,0.36328125,-1.6640625,1.2578125,-0.10595703,0.38671875,-0.5,1.8828125,0.71484375,-0.82421875,-0.69140625,-1.9453125,0.76953125,0.80078125,1.328125,1.46875,-1.4609375,-1.875,0.41601562,0.9375,1.421875,1.2265625,-0.41992188,-0.94921875,2.90625,1.78125,-0.3515625,-0.11669922,-0.5,0.26953125,-0.34765625,0.26757812,0.0007362366,2.34375,-0.48046875,-0.58984375,-0.41601562,2.5,1.0546875,0.91015625,0.8125,0.5390625,0.8203125,0.06933594,-0.84765625,-0.13769531,0.5546875,1.0390625,0.5078125,-0.9453125,0.19824219,1.0390625,-0.46484375,0.37109375,-0.16796875,0.76953125,-1.171875,-1.15625,0.375,0.33398438,-1.015625,-0.15820312,-0.55859375,-1.9453125,-1.78125,0.10253906,-0.625,1.2265625,-1.2578125,0.44140625,0.61328125,0.41210938,-2.203125,-1.359375,-0.8046875,0.18847656,-0.5625,0.43945312,0.6640625,-0.62890625,1.7578125,-0.25390625,0.14550781,-1.5078125,-0.65234375,0.5703125,0.55859375,0.19726562,-0.75,-2.5,1.0703125,2.59375,-0.7109375,-1.21875,-1.265625,1.234375,0.31640625,-0.15332031,0.31640625,0.6015625,0.56640625,0.98046875,0.6328125,-0.6171875,-0.77734375,-1.4609375,0.29296875,-0.049316406,1.171875,-0.64453125,0.35546875,0.57421875,0.080078125,-1.0703125,0.5859375,-0.28515625,0.57421875,-0.74609375,-1.2890625,1.2890625,0.14550781,0.65234375,-0.8984375,-0.52734375,-1.6953125,1.84375,-0.17871094,-0.58203125,0.07861328,-0.48632812,1.25,0.40625,0.828125,-0.68359375,-1.0703125,-1.5546875,0.18652344,0.36328125,0.6953125,0.625,-0.30078125,0.36328125,0.796875,-0.93359375,-0.6328125,-1.8828125,-0.6953125,1.25,1.03125,0.72265625
+8,0,6,0.87109375,0.984375,-0.17871094,0.12597656,0.08203125,-0.60546875,0.765625,1.0703125,1.078125,-0.80859375,1.3203125,1.5546875,0.053710938,-1.03125,-0.14355469,0.30078125,-0.67578125,-1.0078125,0.8359375,0.3359375,0.026489258,-0.36328125,-0.24609375,1.640625,-0.18164062,0.14550781,-2.78125,0.21191406,0.3203125,0.7890625,-0.5078125,-0.90625,1.3671875,-0.78125,0.57421875,0.59375,0.53515625,-0.93359375,-0.21972656,1.0859375,-1.2109375,1.109375,-0.34179688,-2.421875,-0.8671875,-0.7890625,0.25976562,-0.22949219,1.3046875,1.6015625,0.5546875,-0.47070312,-0.92578125,0.6796875,0.46875,0.26757812,0.19921875,-0.5078125,-1.0703125,-0.036376953,1.4765625,0.20703125,0.6328125,0.859375,0.50390625,-0.43554688,-0.14160156,0.55859375,-1.3671875,0.66796875,-0.09863281,0.5390625,0.84765625,-0.42578125,1.8359375,0.8046875,1.3984375,1.046875,-0.35546875,0.36914062,-0.0859375,-1.125,0.83984375,-1.4375,-0.45507812,0.3671875,-0.03173828,-0.036132812,1.75,0.1875,1.34375,-0.5390625,-1.03125,1.2421875,-1.09375,-0.59765625,-2.75,0.49023438,-0.6015625,-0.3671875,-1.453125,1.171875,0.38085938,0.19140625,-0.58984375,1.84375,-0.75,0.16601562,-1.1640625,0.30859375,-0.56640625,-0.42578125,0.91796875,-0.26367188,0.36328125,1.765625,-0.5078125,0.5546875,-0.15917969,-0.44335938,0.57421875,-0.62890625,-0.4921875,-0.94140625,-1.328125,2.109375,-1.2578125,-1.0546875,0.71875,-1.265625,1.3203125,-1.2109375,-0.6875,-0.76171875,-1.59375,0.65234375,1.09375,-0.52734375,0.057373047,-2.46875,1.703125,-0.9453125,-1.7578125,1.3671875,0.71484375,-0.04272461,0.875,-0.89453125,-1.171875,0.9140625,-1.296875,-0.70703125,0.68359375,-1.0234375,1.109375,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.07128906,1.65625,2.453125,1.4609375,-1.53125,0.100097656,-0.5390625,0.084472656,0.859375,-0.16113281,-1.015625,-0.296875,-1.6328125,-0.14746094,-1.5390625,1.0390625,0.9140625,-1.1640625,0.29492188,0.046875,-1.2734375,-1.4140625,0.54296875,-1.0,-0.45703125,0.24902344,0.022705078,0.56640625,-1.6171875,-1.625,-0.234375,2.46875,-0.3671875,-0.9609375,-1.546875,1.03125,1.4921875,1.09375,-0.421875,-0.59375
+8,0,7,1.25,1.5546875,0.10205078,-0.26171875,-1.609375,-0.016479492,-1.328125,1.4375,1.40625,-2.640625,1.7578125,-0.69140625,0.03100586,1.1171875,-0.49609375,0.46679688,0.29101562,0.41210938,1.859375,-0.016845703,-0.8046875,-0.034179688,-0.39453125,-0.25195312,-0.8671875,-1.0546875,-0.71484375,-0.22363281,0.09765625,-0.20605469,-0.25195312,0.01928711,1.0546875,0.71875,0.099609375,0.28710938,1.421875,0.14550781,-0.8828125,1.6875,-2.3125,-0.22265625,0.114746094,2.140625,0.26367188,0.921875,-0.87109375,0.71484375,-0.0625,0.87890625,-0.6484375,-0.09033203,-0.34960938,-0.25390625,-0.4140625,-0.45703125,0.19628906,0.8125,-1.1328125,0.16796875,0.546875,0.65234375,1.65625,-0.41601562,1.359375,0.43359375,-0.36328125,-1.2578125,-0.98046875,0.076660156,-0.2734375,-0.234375,0.6171875,0.04272461,-0.9140625,0.734375,0.14160156,0.34375,-1.53125,1.3203125,-0.81640625,-0.66796875,0.95703125,-1.2421875,-0.17773438,0.3359375,0.24609375,-1.3046875,0.10058594,0.51171875,1.953125,1.5859375,-0.35546875,1.0546875,-0.89453125,-0.14648438,-0.6796875,0.0028839111,0.19433594,-0.703125,-2.390625,0.87890625,1.71875,-1.7421875,0.03149414,1.5390625,-1.109375,-0.46484375,0.053466797,0.20800781,-0.08691406,-0.3125,1.328125,-0.7265625,0.06738281,0.080078125,-0.37890625,0.24707031,-0.25390625,0.859375,1.640625,-0.56640625,-0.15332031,-1.796875,-0.51171875,0.20214844,0.39257812,-1.5546875,-0.11767578,1.3828125,-0.890625,1.1171875,-1.3984375,0.6484375,-2.09375,1.515625,0.9140625,-0.59765625,-0.7109375,-2.328125,1.5859375,-1.1640625,-0.765625,0.03930664,-0.18945312,0.203125,0.48046875,-0.5859375,-0.76171875,1.2421875,-1.40625,0.29492188,0.578125,-0.114746094,-0.31835938,0.09765625,1.1953125,0.34179688,0.53125,0.03564453,-0.66796875,0.16796875,0.44140625,0.13085938,-1.2578125,0.84375,-0.6015625,0.67578125,0.671875,-2.015625,-2.109375,-1.328125,-2.171875,-0.49414062,0.34765625,2.28125,0.36914062,-0.5234375,0.036865234,-0.44726562,0.64453125,-0.27734375,-0.25390625,0.21191406,-0.38085938,0.14160156,0.91796875,1.453125,-0.47460938,0.048583984,0.084472656,0.2578125,0.77734375,-1.421875,-3.140625,1.265625,3.359375,0.6484375,-0.44726562,0.4921875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
index 5c971edd..f8e891b1 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,5,0.78515625,1.0,-0.059570312,0.65234375,0.36132812,-0.49804688,0.3359375,1.0859375,-0.15917969,-1.0625,0.72265625,-0.25,-1.125,-0.859375,-1.0546875,0.1484375,-1.0859375,1.375,-0.43554688,0.30078125,-0.025146484,0.79296875,-0.33203125,-0.13964844,-0.609375,-1.4765625,-2.765625,0.3125,-0.63671875,0.51171875,-1.046875,-1.1171875,-0.38476562,0.21972656,-0.63671875,-0.8984375,0.515625,-1.2578125,0.63671875,-0.103515625,-0.7578125,0.43945312,0.40039062,1.59375,0.29882812,0.69921875,0.05517578,0.29492188,-2.140625,0.8125,1.28125,-1.2890625,-2.296875,-0.14355469,-0.92578125,-0.44726562,0.078125,1.0078125,-0.2734375,-0.8125,-0.07763672,1.265625,-0.07373047,-0.17089844,1.3046875,-0.5859375,0.33789062,-0.0017471313,0.22070312,0.94140625,0.18164062,-2.015625,0.8984375,-1.1171875,-0.1328125,1.265625,1.765625,-0.13085938,0.22851562,-0.3671875,0.90234375,0.47265625,-0.17480469,-1.1015625,-0.4609375,1.0078125,0.07128906,-1.3203125,0.14355469,0.60546875,0.62109375,1.6796875,0.8828125,0.59765625,-0.27734375,-0.20898438,-1.09375,0.80078125,0.12207031,0.60546875,1.6796875,0.67578125,-0.90625,-0.21289062,0.50390625,-0.35546875,-0.94921875,-0.52734375,0.42578125,-0.1640625,-1.59375,2.015625,0.07763672,0.5,-0.36328125,0.16796875,-0.47851562,1.578125,-0.30078125,1.4453125,1.4609375,-1.296875,-0.77734375,-1.03125,0.02319336,0.20605469,0.33789062,-1.0546875,-2.203125,0.18457031,0.22265625,-0.76953125,-0.85546875,0.21289062,0.2421875,1.0,-0.44335938,-1.109375,1.2421875,-1.90625,1.3515625,-0.5390625,-0.43945312,-0.1171875,1.1796875,0.33789062,-0.97265625,-0.119140625,2.203125,3.46875,-0.27929688,-0.75,-2.375,2.21875,-0.48828125,0.2421875,0.203125,1.265625,-1.4921875,0.44335938,-0.95703125,1.0625,0.66015625,1.7421875,-1.9296875,0.23339844,-1.265625,0.828125,1.2109375,-1.1484375,-0.06542969,-1.0703125,1.8359375,-0.80078125,-0.47851562,0.0008468628,0.040039062,-0.734375,1.21875,0.17675781,-0.5546875,0.31445312,1.546875,0.30664062,-0.068847656,0.44140625,1.296875,0.50390625,0.94921875,-2.96875,-0.12402344,-0.10107422,0.007446289,-1.2421875,-1.40625,1.6640625,1.234375,-0.5390625,-0.890625,-0.51171875
-9,0,6,-0.16699219,1.359375,-0.18457031,-0.94140625,0.55859375,0.32421875,0.57421875,-0.81640625,1.5859375,-1.4609375,2.109375,-1.296875,-0.46484375,-0.18066406,-0.43164062,-0.47851562,-1.0546875,0.03173828,0.9296875,0.640625,0.30664062,-0.98046875,1.2578125,0.07373047,-1.609375,-2.046875,-0.65234375,-0.01574707,-0.09667969,1.328125,0.62890625,-0.27734375,0.703125,0.83984375,0.111816406,-0.17480469,0.59375,0.64453125,-0.3046875,-0.0014801025,-0.45117188,0.5703125,1.046875,0.90625,-0.7265625,0.99609375,0.07324219,-1.6484375,-0.70703125,2.25,0.578125,-1.109375,-0.609375,0.84765625,0.34765625,-0.51171875,-0.010070801,1.09375,0.546875,-0.171875,0.96875,1.6875,0.20800781,0.5,1.0,0.24511719,-0.85546875,-0.8984375,0.24414062,0.9765625,-1.1015625,0.5,0.38867188,-0.53515625,-0.53515625,-0.34179688,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.091796875,-0.038085938,2.1875,0.28320312,0.020141602,-2.21875,-0.55078125,0.66796875,-0.13378906,2.203125,0.044921875,1.3203125,-0.72265625,-0.11621094,-0.049072266,-1.2109375,0.21386719,-0.81640625,-1.5390625,-0.41210938,-0.625,1.5078125,-0.45117188,-0.54296875,1.1015625,-2.03125,0.33007812,-0.21191406,-0.34179688,-0.49414062,2.109375,-0.43164062,-2.59375,-0.5859375,0.65625,-0.98046875,0.84765625,-0.24902344,0.13085938,0.5625,-1.1796875,0.025756836,-2.21875,-1.2734375,2.28125,-1.4609375,-0.70703125,-0.41601562,-0.38085938,1.0859375,0.07714844,-1.0546875,-0.0390625,-1.078125,2.34375,-0.5390625,0.62890625,0.72265625,-0.89453125,1.4296875,0.22265625,0.16894531,2.203125,0.0625,-1.0703125,-0.052246094,-0.75390625,-1.265625,0.6484375,-0.5,-0.65234375,-0.68359375,0.053466797,-0.87109375,-0.05078125,0.20214844,-0.27148438,-1.2265625,2.296875,-0.69921875,-0.6484375,1.5,1.140625,-2.4375,1.921875,-0.78515625,2.03125,-0.028686523,-0.14941406,-1.2421875,-0.026245117,-0.484375,-1.1171875,0.6015625,0.33203125,-0.111816406,-2.6875,-0.49414062,-0.94140625,1.2890625,0.57421875,0.5078125,0.8046875,-1.0078125,1.7734375,-1.2734375,-0.25585938,-0.72265625,0.125,0.48828125,0.6640625,0.36328125,0.41796875,-0.12695312,-1.71875,0.2109375,-0.036132812,1.0078125,0.7265625
-9,0,7,-0.31445312,0.34179688,0.5546875,-2.0625,-0.029174805,0.671875,1.8046875,-0.69921875,1.0390625,-0.98046875,1.125,-1.3671875,1.3671875,-0.6484375,-0.54296875,0.6953125,-0.043701172,-0.62890625,2.109375,-0.734375,-0.78125,-1.40625,1.25,-0.15429688,-2.46875,-1.6171875,0.18457031,-0.79296875,-0.515625,1.65625,0.53125,-0.546875,-0.122558594,0.099121094,1.109375,0.7734375,-0.20898438,-0.14160156,-1.8046875,0.9609375,-1.53125,0.51953125,0.95703125,0.96875,0.59765625,1.3671875,-0.98828125,-1.1484375,-0.2265625,1.609375,-1.78125,0.25390625,-1.0,1.890625,-0.25195312,0.62109375,0.20898438,0.28125,-0.047851562,-0.34375,0.60546875,-0.043945312,-0.70703125,0.52734375,0.6328125,1.421875,-0.7890625,-1.1015625,-0.98828125,2.109375,-1.109375,-0.57421875,0.080078125,0.859375,-0.4140625,0.76953125,1.5703125,0.1953125,-0.21386719,0.2265625,1.125,0.4609375,-0.16308594,1.2578125,0.16894531,-0.82421875,-0.65625,-1.28125,-0.5703125,-0.32421875,0.875,0.037109375,-0.83984375,-1.0078125,-1.5390625,0.484375,-0.5703125,-0.60546875,-0.25195312,-0.62109375,-1.375,0.06225586,2.3125,1.1796875,0.17480469,1.7734375,-0.9296875,-0.10449219,0.53515625,-0.79296875,0.40039062,1.015625,0.90234375,-2.59375,0.54296875,-0.34179688,0.28515625,0.18945312,-0.76953125,0.53515625,1.75,-0.38671875,1.234375,-2.0625,0.83203125,0.15625,-2.140625,-1.6171875,-0.19238281,0.24707031,0.71875,0.68359375,0.32617188,-1.515625,0.578125,0.40234375,-0.25585938,0.6015625,-0.328125,-0.62890625,2.546875,1.671875,-0.25585938,1.7109375,-0.87109375,-0.35546875,0.18945312,-0.15332031,-2.421875,0.66015625,-0.5390625,-0.13867188,-0.41210938,-1.2578125,-1.234375,-0.93359375,0.3359375,-0.07519531,-0.05444336,2.09375,-2.0625,-0.67578125,1.171875,-0.0077209473,-1.6875,-0.109375,-0.17285156,1.3515625,0.024414062,0.46679688,0.3515625,0.49804688,-1.375,-0.29296875,0.03564453,-0.14941406,0.765625,-1.2890625,-0.25390625,0.11279297,-0.27539062,-1.140625,1.0390625,0.79296875,-0.63671875,-0.17382812,0.3671875,0.021606445,-1.1484375,-0.17089844,-0.54296875,1.203125,1.2578125,0.13671875,-0.030883789,-0.45703125,1.34375,-0.6875,2.0625,0.5859375
+9,0,5,0.78515625,1.0,-0.059326172,0.65234375,0.359375,-0.50390625,0.3359375,1.0859375,-0.15917969,-1.0546875,0.72265625,-0.24804688,-1.1328125,-0.8515625,-1.046875,0.14648438,-1.0859375,1.3671875,-0.43164062,0.30273438,-0.024780273,0.80078125,-0.33203125,-0.13867188,-0.61328125,-1.4765625,-2.765625,0.3125,-0.6328125,0.5078125,-1.0390625,-1.109375,-0.38671875,0.21972656,-0.63671875,-0.8984375,0.51171875,-1.265625,0.63671875,-0.10107422,-0.7578125,0.43554688,0.40039062,1.5859375,0.29882812,0.6953125,0.059570312,0.296875,-2.15625,0.80859375,1.28125,-1.2890625,-2.3125,-0.14160156,-0.92578125,-0.4453125,0.07861328,1.0078125,-0.27734375,-0.8125,-0.07714844,1.265625,-0.076660156,-0.171875,1.3046875,-0.58984375,0.33789062,-0.0034332275,0.22070312,0.94140625,0.18554688,-2.0,0.8984375,-1.1015625,-0.13476562,1.2734375,1.7578125,-0.1328125,0.22949219,-0.36328125,0.8984375,0.47265625,-0.17578125,-1.0859375,-0.46289062,1.0078125,0.07128906,-1.3203125,0.14355469,0.60546875,0.625,1.6875,0.88671875,0.59765625,-0.28125,-0.20898438,-1.1015625,0.80078125,0.12158203,0.6015625,1.6640625,0.671875,-0.90625,-0.21191406,0.5,-0.34960938,-0.9453125,-0.53125,0.42382812,-0.16308594,-1.59375,2.015625,0.078125,0.50390625,-0.359375,0.16796875,-0.4765625,1.5625,-0.3046875,1.4453125,1.46875,-1.296875,-0.78515625,-1.03125,0.025268555,0.20703125,0.3359375,-1.0546875,-2.203125,0.18457031,0.22167969,-0.78515625,-0.8515625,0.21289062,0.24511719,1.0,-0.44140625,-1.1015625,1.2421875,-1.890625,1.3515625,-0.54296875,-0.4375,-0.12060547,1.1796875,0.33398438,-0.96875,-0.115722656,2.21875,3.484375,-0.27734375,-0.74609375,-2.359375,2.203125,-0.48828125,0.23925781,0.20214844,1.265625,-1.5,0.4453125,-0.95703125,1.0546875,0.671875,1.75,-1.9375,0.234375,-1.2578125,0.828125,1.21875,-1.140625,-0.06640625,-1.0703125,1.8359375,-0.80078125,-0.48046875,0.0024719238,0.038330078,-0.7265625,1.2265625,0.17480469,-0.5625,0.31054688,1.53125,0.30273438,-0.06640625,0.43554688,1.3046875,0.50390625,0.953125,-2.96875,-0.12597656,-0.103027344,0.011413574,-1.2421875,-1.4140625,1.6640625,1.234375,-0.5390625,-0.890625,-0.50390625
+9,0,6,-0.16210938,1.3515625,-0.18554688,-0.9375,0.55859375,0.32421875,0.58203125,-0.81640625,1.578125,-1.453125,2.109375,-1.2890625,-0.4609375,-0.17871094,-0.43554688,-0.48242188,-1.0625,0.029785156,0.9296875,0.63671875,0.30664062,-0.98046875,1.2578125,0.07373047,-1.6171875,-2.0625,-0.64453125,-0.014831543,-0.08935547,1.328125,0.62109375,-0.2734375,0.703125,0.83984375,0.114746094,-0.17382812,0.59375,0.65234375,-0.30859375,0.00042533875,-0.45117188,0.56640625,1.0546875,0.90625,-0.72265625,0.99609375,0.07324219,-1.6484375,-0.703125,2.234375,0.578125,-1.109375,-0.609375,0.84765625,0.34960938,-0.515625,-0.0063476562,1.09375,0.55078125,-0.171875,0.96875,1.671875,0.20996094,0.49804688,0.99609375,0.24511719,-0.8515625,-0.90234375,0.24316406,0.9765625,-1.1015625,0.49609375,0.38476562,-0.53515625,-0.53515625,-0.34375,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.09033203,-0.040039062,2.171875,0.28320312,0.023925781,-2.203125,-0.55078125,0.671875,-0.13183594,2.203125,0.045898438,1.328125,-0.72265625,-0.11621094,-0.049804688,-1.2109375,0.20996094,-0.80859375,-1.5390625,-0.41210938,-0.6171875,1.4921875,-0.45507812,-0.54296875,1.109375,-2.046875,0.33007812,-0.21484375,-0.34179688,-0.4921875,2.109375,-0.43164062,-2.609375,-0.58203125,0.65625,-0.984375,0.84765625,-0.25,0.13378906,0.5625,-1.171875,0.026733398,-2.21875,-1.265625,2.296875,-1.46875,-0.70703125,-0.4140625,-0.38085938,1.0859375,0.07324219,-1.0546875,-0.03540039,-1.078125,2.375,-0.54296875,0.625,0.72265625,-0.8984375,1.421875,0.22167969,0.17089844,2.203125,0.06591797,-1.0703125,-0.053222656,-0.75,-1.265625,0.6484375,-0.49804688,-0.65234375,-0.68359375,0.052490234,-0.86328125,-0.05078125,0.19726562,-0.27148438,-1.2265625,2.296875,-0.69921875,-0.6484375,1.5,1.125,-2.421875,1.90625,-0.78515625,2.03125,-0.029541016,-0.14746094,-1.2421875,-0.02709961,-0.48242188,-1.125,0.6015625,0.3359375,-0.11328125,-2.6875,-0.5,-0.94140625,1.28125,0.578125,0.5078125,0.796875,-0.98828125,1.7734375,-1.28125,-0.25390625,-0.71484375,0.12695312,0.48632812,0.6640625,0.36523438,0.41796875,-0.12597656,-1.734375,0.2109375,-0.034423828,1.0,0.7265625
+9,0,7,-0.31445312,0.34179688,0.55859375,-2.0625,-0.025634766,0.67578125,1.8046875,-0.69921875,1.046875,-0.98046875,1.1171875,-1.359375,1.3671875,-0.6484375,-0.54296875,0.6953125,-0.045166016,-0.62890625,2.09375,-0.7421875,-0.78515625,-1.4140625,1.25,-0.15234375,-2.484375,-1.6171875,0.18164062,-0.7890625,-0.51171875,1.65625,0.52734375,-0.546875,-0.118652344,0.099121094,1.1015625,0.78125,-0.20898438,-0.13964844,-1.8203125,0.953125,-1.546875,0.51953125,0.94921875,0.97265625,0.6015625,1.375,-0.984375,-1.140625,-0.2265625,1.609375,-1.78125,0.25390625,-1.0078125,1.8671875,-0.25,0.625,0.20898438,0.27734375,-0.045166016,-0.34375,0.60546875,-0.045654297,-0.703125,0.52734375,0.6328125,1.4140625,-0.78515625,-1.1015625,-0.984375,2.109375,-1.1015625,-0.5703125,0.083496094,0.87109375,-0.40625,0.76953125,1.5703125,0.19335938,-0.21386719,0.22753906,1.125,0.4609375,-0.16113281,1.25,0.17089844,-0.82421875,-0.66015625,-1.2734375,-0.5703125,-0.328125,0.87109375,0.03540039,-0.83984375,-1.0,-1.546875,0.48242188,-0.5703125,-0.60546875,-0.25585938,-0.625,-1.3828125,0.0625,2.3125,1.1875,0.17480469,1.7890625,-0.92578125,-0.10107422,0.53125,-0.78515625,0.40039062,1.015625,0.90625,-2.578125,0.546875,-0.34179688,0.28710938,0.19042969,-0.76953125,0.5390625,1.75,-0.38671875,1.2421875,-2.0625,0.83203125,0.15625,-2.140625,-1.6015625,-0.19628906,0.24804688,0.7109375,0.6875,0.32421875,-1.5078125,0.578125,0.40429688,-0.25195312,0.6015625,-0.328125,-0.62890625,2.546875,1.671875,-0.25390625,1.71875,-0.875,-0.36132812,0.19042969,-0.15136719,-2.421875,0.66015625,-0.5390625,-0.140625,-0.4140625,-1.2578125,-1.234375,-0.92578125,0.33789062,-0.07519531,-0.055908203,2.09375,-2.078125,-0.67578125,1.1640625,-0.014465332,-1.6875,-0.109375,-0.17285156,1.3515625,0.024536133,0.46679688,0.34960938,0.50390625,-1.375,-0.30078125,0.035888672,-0.14941406,0.765625,-1.2890625,-0.25390625,0.111328125,-0.26953125,-1.140625,1.0546875,0.79296875,-0.640625,-0.17480469,0.3671875,0.0234375,-1.1484375,-0.17480469,-0.54296875,1.203125,1.265625,0.1328125,-0.029907227,-0.45703125,1.3359375,-0.6953125,2.046875,0.58203125
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index 9216a636..78d03599 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.11176976
-0,22,0.34597677
-0,23,0.26810494
-0,24,0.27209836
-0,25,0.26411152
-0,26,0.20620683
-0,27,0.10986541
-0,28,0.16327749
-0,29,0.084906496
-0,30,0.23715588
-0,31,-0.2280783
-0,32,0.08041389
-0,33,-0.15819333
-0,34,0.06743526
-0,35,0.0055371495
-0,36,-0.011310118
-0,37,-0.0062559377
-0,38,-0.18614732
-0,39,0.11385884
-0,40,0.20321175
-1,19,0.100880206
-1,20,0.36195046
-1,21,0.14430872
-1,22,0.29406223
-1,23,0.17126435
-1,24,0.056453336
-1,25,0.11335966
-1,26,-0.08231824
-1,27,0.08191143
-1,28,-0.31693202
-1,29,0.09788513
-1,30,0.084407315
-1,31,-0.12325086
-1,32,0.047468126
-1,33,-0.04812452
-1,34,-0.14221963
-1,35,-0.08581249
-1,36,0.013212016
-1,37,-0.18015718
-1,38,-0.096794404
-2,17,0.18623969
-2,18,0.056702927
-2,19,-0.09579605
-2,20,0.06793443
-2,21,0.10287692
-2,22,0.20520847
-2,23,0.023881953
-2,24,-0.027907796
-2,25,0.0060675265
-2,26,-0.19513254
-2,27,-0.10478126
-2,28,0.07941554
-2,29,0.07442375
-2,30,0.21419367
-2,31,-0.16917527
-2,32,-0.07033796
-2,33,-0.11326729
-2,34,-0.122751676
-2,35,-0.120754965
-2,36,-0.13822621
-3,17,0.2740951
-3,18,0.2092019
-3,19,0.3060425
-3,20,0.06444018
-3,21,0.19821997
-3,22,0.13033172
-3,23,0.066936076
-3,24,-0.109273866
-3,25,-0.049871642
-3,26,0.07592129
-3,27,-0.0027070923
-3,28,-0.042134378
-3,29,-0.09779277
-3,30,-0.011247721
-3,31,-0.043631915
-3,32,-0.061851922
-3,33,-0.027783003
-3,34,-0.23007502
-3,35,0.0060675265
-3,36,-0.015740326
-4,14,0.090896636
-4,15,0.16227913
-4,16,0.15728734
-4,17,0.10487363
-4,18,-0.042134378
-4,19,0.08340896
-4,20,0.13732022
-4,21,0.09788513
-4,22,-0.10178619
-4,23,-0.15519828
-4,24,-0.18315226
-4,25,-0.008658233
-4,26,0.00049856835
-4,27,-0.011122926
-4,28,0.07592129
-4,29,0.001980504
-4,30,0.027500995
-4,31,-0.05411466
-4,32,0.028374556
-4,33,0.01265044
-5,14,0.27010167
-5,15,0.22916903
-5,16,0.19622326
-5,17,0.1493005
-5,18,0.042975523
-5,19,0.15129721
-5,20,0.04372429
-5,21,-0.10428208
-5,22,-0.020295328
-5,23,-0.05661055
-5,24,-0.018049026
-5,25,-0.122751676
-5,26,-0.071336314
-5,27,-0.025911083
-5,28,0.096387595
-5,29,-0.039888076
-5,30,-0.042134378
-5,31,-0.018049026
-5,32,-0.12325086
-5,33,-0.029405331
-6,18,0.100880206
-6,19,0.21319532
-6,20,0.32001948
-6,21,0.3539636
-6,22,0.38591102
-6,23,0.03823333
-6,24,-0.08281741
-6,25,-0.18315226
-6,26,-0.03489629
-6,27,-0.036144238
-6,28,0.05919882
-6,29,0.03848292
-6,30,-0.089306735
-6,31,-0.06784207
-6,32,-0.1267451
-6,33,0.054456625
-6,34,-0.06634453
-6,35,-0.004025235
-6,36,-0.10478126
-6,37,0.09189499
-7,15,0.20121504
-7,16,0.26610824
-7,17,0.118351445
-7,18,0.14630543
-7,19,-0.09330016
-7,20,0.12833501
-7,21,-0.01792423
-7,22,-0.05960562
-7,23,-0.021293685
-7,24,0.12284405
-7,25,0.15529063
-7,26,0.03012168
-7,27,-0.09379934
-7,28,0.04946484
-7,29,-0.06784207
-7,30,-0.087310016
-7,31,-0.04537904
-7,32,0.059947584
-7,33,0.33798993
-7,34,-0.15619662
-8,20,0.2391526
-8,21,0.010279343
-8,22,0.12134651
-8,23,0.29805565
-8,24,-0.25203884
-8,25,-0.03177643
-8,26,-0.053116303
-8,27,0.030371271
-8,28,-0.055861782
-8,29,-0.040137667
-8,30,-0.07183549
-8,31,0.14630543
-8,32,-0.10028865
-8,33,-0.12824264
-8,34,0.021760445
-8,35,0.005661944
-8,36,0.12833501
-8,37,-0.21509966
-8,38,-0.13423277
-8,39,-0.07433138
-9,16,0.27209836
-9,17,-0.05860726
-9,18,0.17326106
-9,19,0.41985515
-9,20,0.11735309
-9,21,0.35196692
-9,22,0.13532351
-9,23,-0.008096658
-9,24,-0.049871642
-9,25,-0.04812452
-9,26,-0.042633556
-9,27,-0.1591917
-9,28,-0.2280783
-9,29,-0.12624593
-9,30,0.08989828
-9,31,-0.023415193
-9,32,-0.15519828
-9,33,-0.14022292
-9,34,0.10337609
-9,35,-0.071336314
+0,21,-0.03839054
+0,22,0.3938979
+0,23,0.48175326
+0,24,0.19821997
+0,25,0.3539636
+0,26,0.25612468
+0,27,0.24314602
+0,28,0.44780913
+0,29,0.01864058
+0,30,0.039980453
+0,31,0.22218053
+0,32,-0.2989616
+0,33,-0.10028865
+0,34,-0.07033796
+0,35,-0.002902084
+0,36,-0.01792423
+0,37,-0.036144238
+0,38,0.11984898
+0,39,0.051461555
+0,40,0.11685391
+1,19,0.24713944
+1,20,0.2920655
+1,21,0.13532351
+1,22,0.22817066
+1,23,0.090397455
+1,24,-0.13223606
+1,25,0.19821997
+1,26,-0.03339876
+1,27,0.072926216
+1,28,-0.11127058
+1,29,-0.33290574
+1,30,0.034239903
+1,31,-0.07483056
+1,32,-0.14122127
+1,33,0.06344183
+1,34,-0.11426565
+1,35,-0.09978948
+1,36,0.05919882
+1,37,-0.035395473
+1,38,0.0037588265
+2,17,0.2920655
+2,18,0.15029885
+2,19,0.07043032
+2,20,-0.050121233
+2,21,0.12184569
+2,22,-0.060354386
+2,23,0.20021668
+2,24,0.054955803
+2,25,-0.04438068
+2,26,-0.07233467
+2,27,-0.05062041
+2,28,-0.1981276
+2,29,0.012400852
+2,30,-0.03127725
+2,31,-0.121254146
+2,32,-0.04138561
+2,33,-0.122751676
+2,34,-0.11426565
+2,35,-0.052117944
+2,36,0.066936076
+3,17,0.03473908
+3,18,0.18623969
+3,19,0.32600963
+3,20,0.030496065
+3,21,0.06843361
+3,22,0.3938979
+3,23,0.25013453
+3,24,0.19222984
+3,25,-0.10577962
+3,26,-0.114764825
+3,27,0.04971443
+3,28,0.102377735
+3,29,-0.055861782
+3,30,-0.0040096357
+3,31,0.12533994
+3,32,-0.21709637
+3,33,-0.007753473
+3,34,0.021136472
+3,35,-0.055362605
+3,36,0.17326106
+4,14,0.16926762
+4,15,-0.10028865
+4,16,-0.042883147
+4,17,0.17725448
+4,18,0.21119861
+4,19,-0.054863427
+4,20,-0.018797792
+4,21,-0.0050235917
+4,22,-0.13523114
+4,23,-0.2670142
+4,24,0.048466485
+4,25,-0.037392184
+4,26,0.061694708
+4,27,0.13432515
+4,28,-0.06634453
+4,29,-0.065346174
+4,30,-0.21609803
+4,31,-0.03639383
+4,32,-0.04188479
+4,33,0.19222984
+5,14,0.106371164
+5,15,0.44581243
+5,16,0.24414438
+5,17,0.14730379
+5,18,0.40188473
+5,19,0.17625612
+5,20,0.17026599
+5,21,0.1403153
+5,22,-0.07383221
+5,23,0.03149442
+5,24,-0.22308652
+5,25,-0.0021962146
+5,26,0.042226754
+5,27,-0.021917658
+5,28,0.016394278
+5,29,-0.02690944
+5,30,-0.07832481
+5,31,-0.06684371
+5,32,-0.20910953
+5,33,-0.095296875
+6,18,0.024505924
+6,19,0.13831857
+6,20,0.15329392
+6,21,0.16427584
+6,22,0.07192786
+6,23,0.044972237
+6,24,0.0412284
+6,25,-0.10078783
+6,26,-0.13822621
+6,27,-0.24604872
+6,28,-0.061103154
+6,29,-0.07782563
+6,30,-0.17915882
+6,31,-0.13323443
+6,32,0.044972237
+6,33,-0.03489629
+6,34,0.018016607
+6,35,-0.012994845
+6,36,0.05520539
+6,37,-0.006505527
+7,15,0.15229557
+7,16,0.4877434
+7,17,0.18424298
+7,18,0.029622503
+7,19,0.24015094
+7,20,0.32001948
+7,21,0.0098425625
+7,22,-0.109273866
+7,23,-0.11626236
+7,24,-0.03364835
+7,25,-0.20611446
+7,26,0.056952514
+7,27,-0.20711282
+7,28,0.009031397
+7,29,-0.006973507
+7,30,-0.12025579
+7,31,-0.049622055
+7,32,-0.16917527
+7,33,-0.0004529903
+7,34,0.02163565
+8,20,0.076919645
+8,21,-0.09978948
+8,22,-0.011310118
+8,23,0.10187856
+8,24,0.1722627
+8,25,0.09538924
+8,26,0.061694708
+8,27,0.16727091
+8,28,-0.23107336
+8,29,0.07592129
+8,30,-0.0009950667
+8,31,-0.018298615
+8,32,-0.013369229
+8,33,0.0527095
+8,34,0.04721854
+8,35,-0.019047381
+8,36,0.000061787316
+8,37,0.18524134
+8,38,-0.07882399
+8,39,0.007346671
+9,16,0.23016739
+9,17,0.25412795
+9,18,0.23815423
+9,19,0.16028242
+9,20,0.10736952
+9,21,0.20720518
+9,22,0.06394101
+9,23,-0.095296875
+9,24,0.05894923
+9,25,-0.04837411
+9,26,-0.057359315
+9,27,0.03224319
+9,28,-0.13523114
+9,29,-0.046127804
+9,30,0.3180228
+9,31,-0.15519828
+9,32,-0.14221963
+9,33,0.08640403
+9,34,0.08390814
+9,35,-0.017425053
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
index 44ec0f7b..744668f3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
@@ -2,24 +2,24 @@ sequenceId,itemPosition,itemValue
 0,8,0.2900688
 0,9,-0.17616376
 0,10,0.3799209
-0,11,0.26610824
-0,12,-0.2989616
+0,11,0.26411152
+0,12,-0.3009583
 1,8,0.28208193
-1,9,-0.003034678
+1,9,-0.0039628376
 1,10,0.33798993
 2,8,-0.14820977
-2,9,0.11785226
+2,9,0.11735309
 3,8,0.36993733
-3,9,0.29605892
+3,9,0.29805565
 4,8,-0.13523114
-5,8,0.1403153
+5,8,0.14131364
 6,8,0.3359932
 6,9,0.26810494
 6,10,0.31402937
-7,8,-0.15320155
+7,8,-0.15419991
 8,8,0.16826928
 8,9,-0.25703064
-8,10,0.41785845
-8,11,0.40987158
-9,8,0.31602606
+8,10,0.41586173
+8,11,0.40787488
+9,8,0.31402937
 9,9,0.25812137
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
index dcfed5d1..2d1901f7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.19187672
+0,8,-0.19088916
 0,9,-0.078307025
-0,10,0.01989374
-0,11,0.13599409
+0,10,0.020881303
+0,11,0.13500652
 0,12,0.30980512
-1,8,0.08809729
-1,9,-0.026706878
-1,10,-0.23631705
-2,8,0.046866544
-2,9,-0.18792649
-3,8,0.08710972
-3,9,0.17352147
-4,8,0.3789345
+1,8,0.0876035
+1,9,-0.026953768
+1,10,-0.2353295
+2,8,0.046619654
+2,9,-0.18891405
+3,8,0.08957863
+3,9,0.17450903
+4,8,0.37695938
 5,8,-0.18002598
 6,8,0.2999295
-6,9,0.33548176
-6,10,0.007902182
-7,8,0.66335255
-8,8,0.19821054
+6,9,0.33745688
+6,10,0.007867463
+7,8,0.66730285
+8,8,0.20117323
 8,9,0.19031003
-8,10,-0.03189158
-8,11,-0.01386856
-9,8,0.001099186
+8,10,-0.03213847
+8,11,-0.012880999
+9,8,0.0023336397
 9,9,-0.2531056
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
index b3488afb..27254750 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.075993046
+0,8,-0.07564883
 0,9,-0.34528795
-0,10,-0.06383061
+0,10,-0.06342901
 0,11,-0.22871205
-0,12,-0.23330164
+0,12,-0.23238373
 1,8,-0.2268762
 1,9,-0.10984136
-1,10,-0.017590366
+1,10,-0.016901925
 2,8,-0.3618105
 2,9,-0.29939193
-3,8,-0.13783793
+3,8,-0.1360021
 3,9,0.23850942
-4,8,0.10357512
+4,8,0.104493044
 5,8,-0.3471238
 6,8,0.09990344
-6,9,-0.1116772
-6,10,0.15314281
-7,8,0.12560523
-8,8,-0.1773085
-8,9,-0.111218244
-8,10,-0.29939193
+6,9,-0.11305408
+6,10,0.15406075
+7,8,0.12652314
+8,8,-0.17822644
+8,9,-0.11075929
+8,10,-0.29755607
 8,11,-0.29204854
-9,8,-0.14334546
-9,9,-0.018737767
+9,8,-0.14472234
+9,9,-0.019196726
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
index 8f72a904..e1bc7a5b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.081780255
-0,9,-0.22130415
-0,10,-0.060438603
-0,11,-0.065946124
-0,12,-0.09807333
-1,8,-0.055734262
-1,9,-0.02922931
-1,10,-0.12492251
+0,8,-0.08269817
+0,9,-0.22038624
+0,10,-0.060151752
+0,11,-0.066634566
+0,12,-0.0973849
+1,8,-0.055131875
+1,9,-0.02945879
+1,10,-0.12446355
 2,8,-0.25710306
 2,9,-0.24058047
-3,8,0.01597827
-3,9,0.36891866
+3,8,0.01506035
+3,9,0.37075448
 4,8,0.3579036
-5,8,-0.12675835
+5,8,-0.12813523
 6,8,0.3395452
 6,9,0.2771266
-6,10,0.3175151
-7,8,0.35606775
-8,8,-0.03835114
-8,9,-0.016148943
-8,10,-0.05389842
-8,11,-0.12675835
-9,8,-0.001691699
-9,9,0.18900627
+6,10,0.31567925
+7,8,0.35423192
+8,8,-0.039154325
+8,9,-0.015919466
+8,10,-0.053353403
+8,11,-0.12629938
+9,8,-0.0007737763
+9,9,0.19084209
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
index 2a8d05f6..aaee17e3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,113,6,2,0,1
-0,9,113,6,2,0,1
-0,10,113,6,2,0,1
-0,11,113,6,2,0,1
-0,12,113,6,2,0,1
-0,13,113,6,2,0,2
-0,14,113,6,2,0,2
-0,15,113,6,2,0,2
-0,16,113,6,2,0,2
-0,17,113,6,2,0,2
-0,18,113,6,2,0,2
-0,19,113,6,2,0,2
-0,20,113,6,2,0,2
-0,21,113,6,2,0,2
-0,22,113,6,2,0,2
-0,23,113,6,2,0,2
-0,24,113,6,2,0,2
-0,25,113,6,2,0,2
-0,26,113,6,2,0,2
-0,27,113,6,2,0,2
-0,28,113,6,2,0,2
-0,29,113,6,2,0,2
-0,30,113,6,2,0,2
-0,31,113,6,2,0,2
-0,32,113,6,2,0,2
-0,33,113,6,2,0,2
-0,34,113,6,2,0,2
-0,35,113,6,2,0,2
-0,36,113,6,2,0,2
-0,37,113,6,2,0,2
+0,8,127,1,1,2,1
+0,9,113,4,9,8,1
+0,10,113,1,1,8,1
+0,11,113,1,9,8,1
+0,12,113,1,9,8,1
+0,13,113,1,9,8,1
+0,14,113,1,8,8,1
+0,15,113,1,0,8,1
+0,16,113,1,9,7,1
+0,17,121,1,8,7,1
+0,18,105,1,8,7,1
+0,19,105,1,8,7,1
+0,20,105,1,8,7,1
+0,21,113,1,7,7,1
+0,22,113,6,7,7,1
+0,23,113,6,7,2,1
+0,24,113,1,7,8,1
+0,25,113,1,7,8,1
+0,26,113,1,7,8,1
+0,27,113,1,0,8,1
+0,28,113,1,0,8,1
+0,29,113,1,0,8,1
+0,30,113,1,0,8,1
+0,31,113,1,0,8,1
+0,32,113,1,0,7,1
+0,33,108,1,8,7,1
+0,34,108,1,0,7,1
+0,35,108,6,0,7,1
+0,36,108,1,0,7,1
+0,37,108,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
index cd416b33..7ae8e23a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,113,6,0,9,0
-1,9,100,6,0,8,0
-1,10,121,6,0,8,0
-1,11,120,4,0,8,0
-1,12,120,4,0,8,0
-1,13,120,0,0,5,0
-1,14,109,4,8,8,0
-1,15,109,4,9,5,0
-1,16,109,4,9,5,0
-1,17,109,4,9,5,0
-1,18,109,4,0,5,0
-1,19,109,4,9,5,0
-1,20,109,4,0,9,0
-1,21,109,4,0,5,0
-1,22,127,4,0,9,0
-1,23,120,4,0,9,0
-1,24,120,4,0,9,0
-1,25,109,4,0,5,0
-1,26,109,4,9,5,0
-1,27,127,4,0,9,0
-1,28,109,4,9,5,0
-1,29,109,4,9,5,0
-1,30,109,4,3,5,0
-1,31,109,4,3,5,0
-1,32,109,4,3,5,0
-1,33,109,4,3,5,0
-1,34,127,6,3,5,0
-1,35,109,6,3,5,0
-1,36,112,6,3,5,0
-1,37,103,6,3,0,0
+1,8,121,6,0,8,0
+1,9,121,1,0,8,0
+1,10,121,1,0,8,0
+1,11,121,1,0,8,0
+1,12,121,1,0,8,0
+1,13,126,1,0,8,0
+1,14,127,1,0,8,0
+1,15,127,0,0,8,0
+1,16,127,0,9,8,0
+1,17,127,0,7,7,0
+1,18,105,8,9,7,1
+1,19,127,0,7,2,1
+1,20,113,4,7,8,1
+1,21,113,1,9,8,1
+1,22,113,1,7,8,1
+1,23,113,1,9,8,1
+1,24,113,1,0,8,1
+1,25,113,1,0,8,1
+1,26,113,1,0,8,1
+1,27,113,1,0,8,1
+1,28,113,1,0,7,1
+1,29,108,1,8,7,1
+1,30,108,1,8,7,1
+1,31,105,6,8,7,1
+1,32,113,1,8,7,1
+1,33,113,6,7,7,1
+1,34,113,6,7,7,1
+1,35,113,6,7,2,1
+1,36,113,1,7,8,1
+1,37,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
index cedb1297..38fb6ff6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,113,4,2,0,1
-2,9,113,6,2,0,0
-2,10,113,6,2,0,0
-2,11,113,6,2,0,1
-2,12,113,6,2,0,1
-2,13,113,6,2,0,1
-2,14,113,6,2,0,2
-2,15,113,6,2,0,2
-2,16,113,6,2,0,2
-2,17,113,6,2,0,2
-2,18,113,6,2,0,2
-2,19,113,6,2,0,2
-2,20,113,6,2,0,2
-2,21,113,6,2,0,2
-2,22,113,6,2,0,2
-2,23,113,6,2,0,2
-2,24,113,6,2,0,2
-2,25,113,6,2,0,2
-2,26,113,6,2,0,2
-2,27,113,6,2,0,2
-2,28,113,6,2,0,2
-2,29,113,6,2,0,2
-2,30,113,6,2,0,2
-2,31,113,6,2,0,2
-2,32,113,6,2,0,2
-2,33,113,6,2,0,2
-2,34,113,6,2,0,2
-2,35,113,6,2,0,2
-2,36,113,6,2,0,2
-2,37,113,6,2,0,2
+2,8,121,1,8,0,0
+2,9,105,0,8,7,0
+2,10,121,1,8,2,0
+2,11,113,4,7,8,0
+2,12,121,1,9,8,0
+2,13,127,1,0,8,0
+2,14,127,4,0,8,1
+2,15,127,1,0,8,0
+2,16,127,4,7,8,1
+2,17,127,1,0,8,1
+2,18,103,1,0,8,1
+2,19,113,1,0,8,1
+2,20,103,1,9,7,1
+2,21,105,1,7,7,1
+2,22,113,1,9,7,1
+2,23,105,1,9,7,1
+2,24,113,1,7,7,1
+2,25,113,6,9,7,1
+2,26,113,1,7,7,1
+2,27,113,6,5,7,1
+2,28,113,1,8,8,1
+2,29,113,1,0,8,1
+2,30,113,1,0,8,1
+2,31,113,1,0,8,1
+2,32,113,1,0,8,1
+2,33,113,1,0,8,1
+2,34,113,1,0,7,1
+2,35,108,1,8,7,1
+2,36,105,1,8,7,1
+2,37,105,1,9,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
index 477f3560..932a6a31 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,4,2,8,1
-3,9,121,4,2,8,1
-3,10,121,4,9,8,0
-3,11,109,4,9,5,0
-3,12,109,6,9,5,0
-3,13,121,6,0,5,0
-3,14,109,4,9,5,0
-3,15,109,4,0,5,0
-3,16,120,8,0,9,0
-3,17,120,4,0,5,0
-3,18,127,4,0,9,0
-3,19,120,4,0,9,0
-3,20,109,4,0,8,0
-3,21,109,4,9,5,0
-3,22,109,4,0,5,0
-3,23,109,4,9,5,0
-3,24,109,4,9,5,0
-3,25,109,4,3,5,0
-3,26,109,4,9,5,0
-3,27,127,4,0,9,0
-3,28,127,4,0,5,0
-3,29,127,4,0,9,0
-3,30,109,4,0,5,0
-3,31,109,4,0,9,0
-3,32,109,4,0,5,0
-3,33,109,4,9,5,0
-3,34,109,4,9,5,0
-3,35,109,4,9,5,0
-3,36,109,4,3,5,0
-3,37,109,4,3,5,0
-4,8,113,7,2,8,1
-4,9,113,6,2,8,0
-4,10,121,6,2,8,0
-4,11,121,4,2,8,0
-4,12,109,4,9,5,0
-4,13,109,6,9,5,0
-4,14,121,6,0,9,0
-4,15,120,4,0,5,0
-4,16,120,9,0,9,0
-4,17,120,0,0,5,0
-4,18,109,4,0,5,0
-4,19,109,4,0,5,0
-4,20,127,4,9,9,0
-4,21,127,4,9,5,0
-4,22,109,4,9,5,0
-4,23,109,4,9,5,0
-4,24,109,4,9,5,0
-4,25,109,6,3,9,0
-4,26,109,4,0,5,0
-4,27,127,6,0,9,0
-4,28,127,9,0,5,0
-4,29,127,4,0,9,0
-4,30,120,4,0,5,0
-4,31,109,4,3,9,0
-4,32,109,4,9,5,0
-4,33,109,4,9,5,0
-4,34,109,4,9,5,0
-4,35,109,4,9,5,0
-4,36,109,4,3,5,0
-4,37,109,4,3,5,0
+3,8,127,4,0,2,0
+3,9,127,4,0,8,1
+3,10,127,4,0,8,1
+3,11,127,1,0,2,1
+3,12,113,4,9,8,1
+3,13,121,1,1,2,1
+3,14,113,0,9,8,1
+3,15,113,1,1,8,1
+3,16,113,1,9,7,1
+3,17,105,1,9,7,1
+3,18,105,1,8,7,1
+3,19,105,1,9,7,1
+3,20,113,1,8,7,1
+3,21,105,8,9,7,1
+3,22,113,1,8,8,1
+3,23,113,1,9,8,1
+3,24,113,1,7,8,1
+3,25,113,1,9,7,1
+3,26,113,1,8,7,1
+3,27,113,1,8,7,1
+3,28,113,1,8,7,1
+3,29,113,1,8,7,1
+3,30,113,1,5,7,1
+3,31,113,1,5,8,1
+3,32,113,1,0,8,1
+3,33,113,1,0,8,1
+3,34,113,1,0,8,1
+3,35,113,1,0,8,1
+3,36,113,1,0,7,1
+3,37,108,1,8,7,1
+4,8,127,4,0,8,0
+4,9,127,4,0,8,1
+4,10,127,1,0,2,1
+4,11,127,4,7,8,1
+4,12,127,1,1,8,1
+4,13,127,1,0,8,1
+4,14,113,1,0,8,1
+4,15,113,1,9,8,1
+4,16,113,1,9,7,1
+4,17,105,1,9,7,1
+4,18,105,1,9,7,1
+4,19,105,1,9,7,1
+4,20,105,1,9,7,1
+4,21,113,1,9,7,1
+4,22,113,1,9,7,1
+4,23,113,1,9,7,1
+4,24,113,1,9,8,1
+4,25,113,1,9,8,1
+4,26,113,1,9,8,1
+4,27,113,1,9,8,1
+4,28,113,1,9,7,1
+4,29,113,1,8,7,1
+4,30,105,1,8,7,1
+4,31,113,1,8,7,1
+4,32,113,1,8,7,1
+4,33,113,1,8,7,1
+4,34,113,1,9,7,1
+4,35,113,1,7,8,1
+4,36,113,1,7,8,1
+4,37,113,1,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
index e21fb885..0c148033 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,6,2,1,1
-5,9,113,6,2,0,2
-5,10,113,6,2,0,2
-5,11,113,6,2,0,2
-5,12,113,6,2,0,2
-5,13,113,6,2,0,2
-5,14,113,6,2,0,2
-5,15,113,6,2,0,2
-5,16,113,6,2,0,2
-5,17,113,6,2,0,2
-5,18,113,6,2,0,2
-5,19,113,6,2,0,2
-5,20,113,6,2,0,2
-5,21,113,6,2,0,2
-5,22,113,6,2,0,2
-5,23,113,6,2,0,2
-5,24,113,6,2,0,2
-5,25,113,6,2,0,2
-5,26,113,6,2,0,2
-5,27,113,6,2,0,2
-5,28,113,6,2,0,2
-5,29,113,6,2,0,2
-5,30,113,6,2,0,2
-5,31,113,6,2,0,2
-5,32,113,6,2,0,2
-5,33,113,6,2,0,2
-5,34,113,6,2,0,2
-5,35,113,6,2,0,2
-5,36,113,6,2,0,2
-5,37,113,6,2,0,2
+5,8,121,1,5,4,1
+5,9,127,1,0,8,1
+5,10,127,1,0,2,1
+5,11,127,1,7,2,1
+5,12,127,1,1,0,1
+5,13,105,1,9,2,1
+5,14,113,1,7,0,1
+5,15,113,1,9,7,1
+5,16,105,8,9,7,1
+5,17,105,1,8,2,1
+5,18,113,0,7,8,1
+5,19,113,1,9,8,1
+5,20,113,1,9,8,1
+5,21,113,1,9,8,1
+5,22,113,1,0,8,1
+5,23,113,1,0,8,1
+5,24,113,1,0,8,1
+5,25,113,1,0,7,1
+5,26,108,1,8,7,1
+5,27,105,1,8,7,1
+5,28,105,1,8,7,1
+5,29,105,1,9,7,1
+5,30,113,1,7,7,1
+5,31,113,6,7,7,1
+5,32,113,6,7,2,1
+5,33,113,1,7,8,1
+5,34,113,1,7,8,1
+5,35,113,1,9,8,1
+5,36,113,1,0,8,1
+5,37,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
index 908877f2..1b1ae93e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,113,6,2,8,0
-6,9,113,6,2,8,0
-6,10,113,6,2,8,0
-6,11,121,6,2,8,0
-6,12,121,6,0,8,0
-6,13,109,6,0,8,0
-6,14,109,4,0,8,0
-6,15,109,4,0,5,0
-6,16,109,4,0,5,0
-6,17,109,4,9,5,0
-6,18,109,4,0,5,0
-6,19,109,4,9,5,0
-6,20,109,4,9,5,0
-6,21,109,4,3,9,0
-6,22,109,4,0,5,0
-6,23,109,4,3,9,0
-6,24,109,4,0,5,0
-6,25,127,4,0,9,0
-6,26,109,4,0,5,0
-6,27,127,4,0,9,0
-6,28,109,4,0,5,0
-6,29,109,4,9,5,0
-6,30,109,4,9,5,0
-6,31,109,4,9,5,0
-6,32,109,4,3,5,0
-6,33,109,4,3,5,0
-6,34,109,4,3,5,0
-6,35,109,4,3,5,0
-6,36,109,6,3,5,0
-6,37,127,6,3,9,0
-7,8,109,4,9,6,0
-7,9,121,4,9,6,0
-7,10,121,4,2,8,1
-7,11,109,4,9,8,0
-7,12,109,4,9,5,0
-7,13,109,4,9,5,0
-7,14,109,6,3,5,0
-7,15,121,6,0,5,0
-7,16,109,6,0,8,0
-7,17,109,4,0,5,0
-7,18,127,4,0,9,0
-7,19,120,4,0,5,0
-7,20,127,4,3,9,0
-7,21,109,4,9,5,0
-7,22,109,4,3,0,0
-7,23,109,4,9,5,0
-7,24,109,4,9,5,0
-7,25,109,4,3,5,0
-7,26,109,4,3,5,0
-7,27,109,6,3,5,0
-7,28,109,6,3,5,0
-7,29,112,6,3,9,0
-7,30,121,6,3,5,0
-7,31,112,6,3,9,0
-7,32,121,6,3,5,0
-7,33,103,6,3,0,0
-7,34,103,6,3,0,0
-7,35,103,6,2,0,0
-7,36,103,6,2,0,0
-7,37,103,6,2,0,0
+6,8,121,4,0,8,0
+6,9,113,1,0,8,0
+6,10,113,1,0,8,1
+6,11,113,1,0,8,1
+6,12,113,1,0,8,1
+6,13,113,1,0,8,1
+6,14,113,1,0,8,1
+6,15,113,1,9,7,1
+6,16,105,8,9,7,1
+6,17,105,1,8,7,1
+6,18,105,1,9,7,1
+6,19,105,1,9,7,1
+6,20,113,1,7,7,1
+6,21,113,6,9,7,1
+6,22,113,1,3,2,1
+6,23,113,6,7,8,1
+6,24,113,1,9,8,1
+6,25,113,1,0,8,1
+6,26,113,1,0,8,1
+6,27,113,1,0,8,1
+6,28,113,1,0,8,1
+6,29,113,1,0,8,1
+6,30,113,1,0,7,1
+6,31,108,1,8,7,1
+6,32,105,1,8,7,1
+6,33,105,1,9,7,1
+6,34,105,1,7,7,1
+6,35,113,6,7,7,1
+6,36,113,6,7,7,1
+6,37,113,6,7,2,1
+7,8,105,4,9,8,0
+7,9,121,0,8,8,0
+7,10,113,0,8,8,0
+7,11,121,4,8,8,0
+7,12,127,0,0,8,0
+7,13,121,4,8,8,0
+7,14,127,4,0,8,0
+7,15,127,1,0,8,0
+7,16,127,4,0,8,1
+7,17,127,1,0,8,0
+7,18,127,1,0,8,1
+7,19,127,1,7,8,1
+7,20,113,1,7,8,1
+7,21,113,1,9,7,1
+7,22,103,1,7,7,1
+7,23,113,8,7,7,1
+7,24,113,1,9,7,1
+7,25,113,1,7,7,1
+7,26,113,6,5,7,1
+7,27,113,1,5,8,1
+7,28,113,1,0,8,1
+7,29,113,1,0,8,1
+7,30,113,1,0,8,1
+7,31,113,1,0,8,1
+7,32,113,1,0,7,1
+7,33,108,1,8,7,1
+7,34,108,1,8,7,1
+7,35,108,1,8,7,1
+7,36,122,6,7,7,1
+7,37,113,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
index d1cabc88..9dc6a76d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,6,2,8,0
-8,9,113,6,2,8,0
-8,10,113,6,2,8,0
-8,11,113,6,2,0,0
-8,12,113,6,2,0,0
-8,13,113,6,8,0,0
-8,14,113,6,8,0,0
-8,15,113,9,7,0,0
-8,16,113,6,2,0,1
-8,17,113,6,2,0,1
-8,18,113,6,2,0,1
-8,19,113,6,2,0,1
-8,20,113,6,2,0,2
-8,21,113,6,2,0,2
-8,22,113,6,2,0,2
-8,23,113,6,2,0,2
-8,24,113,6,2,0,2
-8,25,113,6,2,0,2
-8,26,113,6,2,0,2
-8,27,113,6,2,0,2
-8,28,113,6,2,0,2
-8,29,113,6,2,0,2
-8,30,113,6,2,0,2
-8,31,113,6,2,0,2
-8,32,113,6,2,0,2
-8,33,113,6,2,0,2
-8,34,113,6,2,0,2
-8,35,113,6,2,0,2
-8,36,113,6,2,0,2
-8,37,113,6,2,0,2
+8,8,113,1,0,8,0
+8,9,113,1,0,8,1
+8,10,113,1,0,2,1
+8,11,113,1,8,2,1
+8,12,113,1,0,8,1
+8,13,113,1,0,8,1
+8,14,113,1,0,2,1
+8,15,113,1,0,2,1
+8,16,113,1,8,7,1
+8,17,105,8,9,7,1
+8,18,105,1,8,2,1
+8,19,113,1,9,7,1
+8,20,105,1,9,7,1
+8,21,113,1,7,7,1
+8,22,113,1,9,7,1
+8,23,113,1,7,8,1
+8,24,113,1,9,8,1
+8,25,113,1,0,8,1
+8,26,113,1,9,8,1
+8,27,113,1,0,7,1
+8,28,113,1,8,7,1
+8,29,105,1,8,7,1
+8,30,113,1,8,7,1
+8,31,105,1,9,7,1
+8,32,113,1,8,7,1
+8,33,113,6,9,7,1
+8,34,113,1,8,2,1
+8,35,113,1,7,8,1
+8,36,113,1,7,8,1
+8,37,113,1,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
index f9969b00..06013440 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,100,6,0,9,0
-9,9,100,9,8,9,0
-9,10,109,4,9,5,1
-9,11,113,6,0,9,0
-9,12,100,9,0,9,0
-9,13,109,4,9,5,1
-9,14,112,6,3,9,0
-9,15,121,6,0,5,0
-9,16,109,4,0,8,0
-9,17,109,4,0,5,0
-9,18,127,4,0,9,0
-9,19,109,4,9,5,0
-9,20,127,4,0,5,0
-9,21,127,4,0,9,0
-9,22,109,4,9,5,0
-9,23,109,4,9,5,0
-9,24,109,4,3,5,0
-9,25,109,4,3,5,0
-9,26,109,4,3,5,0
-9,27,109,6,3,5,0
-9,28,121,6,3,0,0
-9,29,109,4,3,7,0
-9,30,112,6,3,5,0
-9,31,103,6,3,0,0
-9,32,103,6,2,0,0
-9,33,103,6,2,0,0
-9,34,113,6,2,0,0
-9,35,113,6,2,0,0
-9,36,113,6,2,0,0
-9,37,113,6,2,0,1
+9,8,121,1,8,8,1
+9,9,113,1,0,8,1
+9,10,113,1,8,8,1
+9,11,113,1,0,7,1
+9,12,105,1,8,7,1
+9,13,105,1,8,7,1
+9,14,105,1,7,7,1
+9,15,113,1,7,7,1
+9,16,113,1,7,7,1
+9,17,113,6,7,7,1
+9,18,113,6,5,2,1
+9,19,113,1,0,8,1
+9,20,113,1,0,8,1
+9,21,113,1,0,8,1
+9,22,113,1,0,8,1
+9,23,113,1,0,8,1
+9,24,113,1,0,8,1
+9,25,113,1,0,8,1
+9,26,113,1,0,7,1
+9,27,105,1,8,7,1
+9,28,105,1,8,7,1
+9,29,105,1,9,7,1
+9,30,105,1,7,7,1
+9,31,113,6,7,7,1
+9,32,113,6,7,2,1
+9,33,113,1,7,8,1
+9,34,113,1,7,8,1
+9,35,113,1,9,8,1
+9,36,113,1,0,8,1
+9,37,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index 9f00e61d..fd1bb4cf 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,102,6,3,8,1
-0,9,121,6,3,8,1
-0,10,121,6,3,8,1
-0,11,121,6,3,8,1
-0,12,121,6,3,8,1
-0,13,104,6,3,8,0
-0,14,104,3,3,0,0
-0,15,104,3,3,0,0
-0,16,104,3,3,0,0
-0,17,104,3,3,0,0
-0,18,104,3,3,0,0
-0,19,104,3,3,0,0
-0,20,104,3,3,0,1
-0,21,103,6,3,0,1
-0,22,113,6,3,8,1
-0,23,113,6,3,8,1
-0,24,113,6,3,8,1
-0,25,104,6,3,8,1
-0,26,104,6,3,8,1
-0,27,104,6,3,8,1
-0,28,104,6,3,8,1
-0,29,104,6,3,8,1
-0,30,104,9,3,0,0
-0,31,104,3,3,0,1
-0,32,104,9,3,0,1
-0,33,104,9,3,0,1
-0,34,104,3,3,8,1
-0,35,104,6,3,8,0
-0,36,121,6,3,8,0
-0,37,121,6,3,8,0
+0,8,104,7,[mask],[unknown],1
+0,9,104,7,[mask],[unknown],0
+0,10,104,7,1,[unknown],0
+0,11,101,7,1,[unknown],0
+0,12,106,7,1,[unknown],[other]
+0,13,112,9,1,[unknown],0
+0,14,129,9,1,5,[unknown]
+0,15,112,2,1,1,[other]
+0,16,112,7,1,[unknown],2
+0,17,129,2,1,[unknown],2
+0,18,129,7,1,5,2
+0,19,112,7,1,1,2
+0,20,129,7,1,[unknown],2
+0,21,129,7,1,1,[other]
+0,22,112,7,1,5,[mask]
+0,23,118,7,[mask],[unknown],[mask]
+0,24,129,9,1,5,[other]
+0,25,112,9,1,5,[mask]
+0,26,[other],7,[mask],[unknown],[mask]
+0,27,129,9,1,5,[other]
+0,28,112,9,1,5,[other]
+0,29,129,7,1,[unknown],[mask]
+0,30,104,0,1,[unknown],[other]
+0,31,112,7,1,[unknown],[other]
+0,32,104,7,1,[unknown],2
+0,33,129,7,1,[unknown],[other]
+0,34,112,9,1,5,[other]
+0,35,112,7,1,[unknown],2
+0,36,108,7,1,[unknown],2
+0,37,129,7,1,5,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index 81ebca01..d23a5aeb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,6,2,8,1
-1,9,121,6,3,8,1
-1,10,121,6,3,8,0
-1,11,121,6,3,8,0
-1,12,121,6,3,8,0
-1,13,121,3,3,8,0
-1,14,104,3,3,8,0
-1,15,104,3,3,0,0
-1,16,104,3,3,0,0
-1,17,104,3,3,0,0
-1,18,104,3,3,0,0
-1,19,104,3,3,0,0
-1,20,104,3,3,0,0
-1,21,104,3,3,0,0
-1,22,121,6,3,0,1
-1,23,113,6,3,8,1
-1,24,121,6,3,8,1
-1,25,104,6,3,8,1
-1,26,104,6,3,8,0
-1,27,104,6,3,0,0
-1,28,104,9,3,0,1
-1,29,104,3,3,0,1
-1,30,104,9,3,0,1
-1,31,104,3,3,0,1
-1,32,104,3,3,0,1
-1,33,103,6,3,8,1
-1,34,113,6,3,8,1
-1,35,121,6,3,8,1
-1,36,121,6,3,8,1
-1,37,104,6,3,8,1
+1,8,104,7,[mask],1,1
+1,9,104,7,[mask],[unknown],[mask]
+1,10,[other],9,1,8,[mask]
+1,11,[other],9,1,5,[other]
+1,12,104,9,1,1,[other]
+1,13,112,9,1,[unknown],[other]
+1,14,112,7,1,[unknown],[other]
+1,15,112,2,1,5,[other]
+1,16,112,2,1,[unknown],[mask]
+1,17,129,2,1,[unknown],[mask]
+1,18,129,7,1,5,2
+1,19,129,2,1,1,2
+1,20,129,7,1,1,2
+1,21,129,7,1,1,2
+1,22,129,7,1,1,2
+1,23,129,7,1,1,2
+1,24,129,2,1,5,[mask]
+1,25,129,2,1,5,[mask]
+1,26,129,2,1,5,[mask]
+1,27,129,2,1,5,[mask]
+1,28,129,2,1,5,[mask]
+1,29,112,2,1,5,[mask]
+1,30,107,2,1,5,[mask]
+1,31,107,2,1,5,1
+1,32,107,2,[mask],[unknown],1
+1,33,107,2,[mask],[unknown],1
+1,34,107,2,[mask],[unknown],1
+1,35,107,7,[mask],[unknown],1
+1,36,107,7,[mask],[unknown],1
+1,37,107,[mask],[mask],[unknown],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index c3ecaba4..cc5b8a64 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,102,6,3,8,1
-2,9,121,6,3,8,1
-2,10,104,6,3,8,1
-2,11,104,6,3,8,1
-2,12,104,6,3,8,1
-2,13,104,6,3,8,1
-2,14,104,6,3,8,0
-2,15,104,9,3,0,0
-2,16,104,3,3,0,0
-2,17,104,3,3,0,0
-2,18,104,3,3,0,1
-2,19,104,3,3,0,1
-2,20,104,3,3,0,1
-2,21,103,6,3,0,1
-2,22,113,6,3,8,1
-2,23,113,6,3,8,1
-2,24,113,6,3,8,1
-2,25,104,6,3,8,1
-2,26,104,6,3,8,1
-2,27,104,6,3,8,1
-2,28,104,6,3,8,1
-2,29,104,6,3,8,1
-2,30,104,9,3,0,0
-2,31,104,3,3,0,1
-2,32,104,9,3,0,1
-2,33,104,9,3,0,1
-2,34,104,3,3,8,1
-2,35,104,6,3,8,0
-2,36,121,6,3,8,0
-2,37,121,6,3,8,0
+2,8,112,2,[mask],5,[mask]
+2,9,104,7,[mask],[unknown],[mask]
+2,10,104,7,1,8,[mask]
+2,11,112,7,1,[unknown],[other]
+2,12,112,7,1,[unknown],[mask]
+2,13,107,7,[mask],[unknown],[mask]
+2,14,107,7,[mask],[unknown],[mask]
+2,15,[other],7,[mask],[unknown],2
+2,16,129,9,[other],1,[other]
+2,17,112,9,[other],1,[mask]
+2,18,104,1,[other],5,[mask]
+2,19,104,2,[other],[unknown],[other]
+2,20,110,9,[mask],[unknown],[other]
+2,21,110,2,[mask],[unknown],[mask]
+2,22,[other],2,[mask],[unknown],[other]
+2,23,110,2,[mask],5,[mask]
+2,24,[other],2,[other],7,[mask]
+2,25,104,2,[other],8,[mask]
+2,26,107,2,[other],1,[mask]
+2,27,107,2,[other],1,[mask]
+2,28,107,2,[other],8,[mask]
+2,29,107,2,1,5,[mask]
+2,30,107,2,1,5,[unknown]
+2,31,129,2,1,5,[mask]
+2,32,129,9,1,5,[other]
+2,33,107,2,1,5,[mask]
+2,34,107,2,1,5,[mask]
+2,35,107,2,1,[unknown],1
+2,36,107,2,[mask],[unknown],1
+2,37,107,7,[mask],[unknown],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index cfc7190a..2a4d06f3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,6,3,8,1
-3,9,121,6,3,8,0
-3,10,121,3,3,8,0
-3,11,121,3,3,8,0
-3,12,121,3,3,8,0
-3,13,121,3,3,8,0
-3,14,104,3,3,0,0
-3,15,104,3,3,0,0
-3,16,104,3,3,0,0
-3,17,104,3,3,0,0
-3,18,104,3,3,0,0
-3,19,104,3,3,0,0
-3,20,104,3,3,0,0
-3,21,121,3,3,0,1
-3,22,113,6,3,8,1
-3,23,121,6,3,8,1
-3,24,104,6,3,8,1
-3,25,104,6,3,8,0
-3,26,104,6,3,0,0
-3,27,104,3,3,0,1
-3,28,104,9,3,0,1
-3,29,104,3,3,0,1
-3,30,104,9,3,0,1
-3,31,104,3,3,0,1
-3,32,103,6,3,8,1
-3,33,113,6,3,8,1
-3,34,113,6,3,8,1
-3,35,121,6,3,8,1
-3,36,121,6,3,8,1
-3,37,104,6,3,8,1
-4,8,102,6,2,8,1
-4,9,121,6,3,8,1
-4,10,121,6,3,8,1
-4,11,121,6,3,8,1
-4,12,121,6,3,8,1
-4,13,121,6,3,8,0
-4,14,104,6,3,8,0
-4,15,104,3,3,0,0
-4,16,104,3,3,0,0
-4,17,104,3,3,0,0
-4,18,104,3,3,0,0
-4,19,104,3,3,0,0
-4,20,104,3,3,0,0
-4,21,104,3,3,0,0
-4,22,121,6,3,0,1
-4,23,113,6,3,8,1
-4,24,121,6,3,8,1
-4,25,104,6,3,8,1
-4,26,104,6,3,8,0
-4,27,104,6,3,0,0
-4,28,104,9,3,0,1
-4,29,104,3,3,0,1
-4,30,104,9,3,0,1
-4,31,104,3,3,0,1
-4,32,104,3,3,0,1
-4,33,103,6,3,8,1
-4,34,113,6,3,8,1
-4,35,121,6,3,8,1
-4,36,121,6,3,8,1
-4,37,104,6,3,8,1
+3,8,104,[mask],[mask],[unknown],[mask]
+3,9,104,0,[mask],8,[other]
+3,10,104,[mask],[mask],6,[mask]
+3,11,104,0,[mask],8,[mask]
+3,12,104,0,[mask],8,[mask]
+3,13,104,1,[mask],8,[mask]
+3,14,[other],2,[mask],[unknown],[mask]
+3,15,129,1,[other],6,[mask]
+3,16,107,2,[other],1,[mask]
+3,17,107,2,[other],6,[mask]
+3,18,129,9,[other],6,[mask]
+3,19,129,9,[other],5,[mask]
+3,20,129,9,[other],5,[mask]
+3,21,129,9,[other],5,[mask]
+3,22,129,9,[other],5,[other]
+3,23,110,9,[other],5,[mask]
+3,24,107,2,[other],8,[other]
+3,25,107,2,1,5,[other]
+3,26,110,9,1,[unknown],[other]
+3,27,[other],9,1,[unknown],[other]
+3,28,[other],2,1,[unknown],[other]
+3,29,[other],2,1,[unknown],[other]
+3,30,107,2,1,[unknown],[unknown]
+3,31,129,2,1,[unknown],[mask]
+3,32,129,2,1,[unknown],2
+3,33,129,2,1,5,2
+3,34,129,2,1,5,2
+3,35,129,2,1,5,2
+3,36,129,2,1,5,2
+3,37,129,2,1,5,2
+4,8,104,9,1,[unknown],[mask]
+4,9,[other],7,1,[unknown],[other]
+4,10,104,9,1,[unknown],[other]
+4,11,[other],9,1,[unknown],[other]
+4,12,[other],9,1,[unknown],[other]
+4,13,[other],9,1,[unknown],[other]
+4,14,129,2,1,5,[unknown]
+4,15,129,2,1,1,2
+4,16,129,2,1,1,2
+4,17,129,7,1,5,2
+4,18,129,2,1,1,2
+4,19,129,7,1,5,2
+4,20,129,7,1,1,2
+4,21,129,7,1,1,[mask]
+4,22,129,7,1,5,[mask]
+4,23,129,2,1,5,[mask]
+4,24,112,2,1,5,[mask]
+4,25,107,2,[mask],[unknown],[mask]
+4,26,129,2,[mask],5,[mask]
+4,27,129,9,1,5,[mask]
+4,28,129,9,1,5,[mask]
+4,29,129,9,1,5,[mask]
+4,30,129,9,1,5,[mask]
+4,31,129,9,1,5,[mask]
+4,32,129,9,1,5,[mask]
+4,33,129,9,1,5,[other]
+4,34,112,9,1,5,[mask]
+4,35,107,7,[mask],[unknown],[mask]
+4,36,107,7,[mask],[unknown],[mask]
+4,37,104,7,[mask],[unknown],[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index 69802abf..529df285 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,6,3,8,1
-5,9,121,6,3,8,0
-5,10,104,6,3,8,0
-5,11,104,3,3,0,0
-5,12,104,3,3,0,0
-5,13,104,3,3,0,0
-5,14,121,6,3,8,1
-5,15,104,3,3,8,0
-5,16,104,3,3,0,0
-5,17,104,3,3,0,0
-5,18,104,3,3,0,0
-5,19,121,3,3,0,1
-5,20,104,4,3,8,1
-5,21,104,3,3,0,0
-5,22,104,3,3,0,0
-5,23,104,3,3,0,1
-5,24,103,6,3,8,1
-5,25,113,6,3,8,1
-5,26,104,6,3,8,1
-5,27,104,9,3,8,1
-5,28,121,6,3,8,1
-5,29,104,6,3,8,1
-5,30,104,6,3,8,0
-5,31,104,6,3,0,0
-5,32,104,9,3,0,0
-5,33,104,3,3,0,1
-5,34,104,3,3,0,0
-5,35,104,3,3,0,1
-5,36,104,3,3,0,1
-5,37,103,6,3,0,1
+5,8,102,7,[mask],9,1
+5,9,107,[mask],[mask],[unknown],1
+5,10,107,[mask],[mask],[unknown],1
+5,11,107,[mask],[mask],[unknown],1
+5,12,104,[mask],[mask],[unknown],1
+5,13,104,[mask],[mask],[unknown],1
+5,14,104,[mask],[mask],[unknown],1
+5,15,104,[mask],[mask],3,1
+5,16,104,[mask],[mask],3,[other]
+5,17,112,[unknown],[mask],3,[other]
+5,18,111,[mask],[mask],3,2
+5,19,104,9,[mask],3,2
+5,20,104,5,[mask],3,2
+5,21,104,5,[mask],3,2
+5,22,104,5,[mask],3,2
+5,23,104,[unknown],[mask],3,2
+5,24,104,9,[mask],3,[unknown]
+5,25,104,9,[mask],3,[mask]
+5,26,112,[unknown],[mask],3,[mask]
+5,27,112,[unknown],[mask],3,[mask]
+5,28,112,[mask],[mask],3,[mask]
+5,29,104,[mask],[mask],7,[mask]
+5,30,107,9,[mask],3,[mask]
+5,31,107,2,[mask],3,[mask]
+5,32,107,2,[mask],6,[mask]
+5,33,104,2,[mask],6,[mask]
+5,34,104,9,[mask],6,[mask]
+5,35,[other],9,[mask],6,[mask]
+5,36,[other],2,[other],6,[mask]
+5,37,107,9,[other],6,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index 4ff6731c..819defac 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,102,6,2,8,1
-6,9,113,6,3,8,1
-6,10,121,6,3,8,1
-6,11,121,6,3,8,1
-6,12,121,6,3,8,1
-6,13,121,6,3,8,0
-6,14,121,6,3,8,0
-6,15,104,3,3,8,0
-6,16,104,3,3,0,0
-6,17,104,3,3,0,0
-6,18,104,3,3,0,0
-6,19,104,3,3,0,0
-6,20,104,3,3,0,0
-6,21,104,3,3,0,0
-6,22,104,3,3,0,0
-6,23,121,6,3,0,1
-6,24,113,6,3,8,1
-6,25,121,6,3,8,1
-6,26,104,6,3,8,1
-6,27,104,6,3,8,0
-6,28,104,6,3,0,0
-6,29,104,9,3,0,1
-6,30,104,3,3,0,1
-6,31,104,9,3,0,1
-6,32,104,3,3,0,1
-6,33,104,3,3,0,1
-6,34,103,6,3,8,1
-6,35,113,6,3,8,1
-6,36,121,6,3,8,1
-6,37,121,6,3,8,1
-7,8,102,6,2,8,1
-7,9,121,6,3,8,1
-7,10,121,6,3,8,1
-7,11,104,6,3,8,0
-7,12,121,6,3,8,0
-7,13,104,6,3,8,0
-7,14,104,6,3,0,0
-7,15,104,3,3,0,0
-7,16,104,3,3,0,0
-7,17,104,3,3,0,0
-7,18,104,3,3,0,0
-7,19,104,3,3,0,0
-7,20,104,3,3,0,1
-7,21,103,6,3,8,1
-7,22,113,6,3,8,1
-7,23,113,6,3,8,1
-7,24,104,6,3,8,1
-7,25,104,6,3,8,1
-7,26,104,6,3,8,1
-7,27,104,6,3,8,1
-7,28,104,6,3,8,0
-7,29,104,9,3,0,0
-7,30,104,3,3,0,1
-7,31,104,9,3,0,1
-7,32,104,3,3,0,1
-7,33,104,9,3,0,1
-7,34,103,3,3,8,1
-7,35,104,6,3,8,1
-7,36,121,6,3,8,1
-7,37,121,6,3,8,1
+6,8,129,9,1,5,[other]
+6,9,129,9,1,5,[mask]
+6,10,129,9,1,5,[other]
+6,11,129,9,1,5,[other]
+6,12,129,9,1,5,[other]
+6,13,129,2,1,5,[mask]
+6,14,129,2,1,8,[other]
+6,15,112,7,1,[unknown],[unknown]
+6,16,107,7,1,[unknown],2
+6,17,108,7,[mask],[unknown],2
+6,18,104,7,1,5,2
+6,19,108,7,1,[unknown],2
+6,20,104,7,1,5,2
+6,21,112,9,[mask],1,[mask]
+6,22,129,9,[other],5,[mask]
+6,23,104,9,[other],5,[mask]
+6,24,112,9,8,1,[mask]
+6,25,129,9,[other],5,[mask]
+6,26,104,9,[other],1,[mask]
+6,27,104,9,[other],1,[other]
+6,28,112,7,[other],1,[mask]
+6,29,[other],1,[other],8,[other]
+6,30,[other],2,[mask],[unknown],[other]
+6,31,[other],2,1,[unknown],[other]
+6,32,110,2,1,5,[other]
+6,33,[other],2,1,[unknown],[other]
+6,34,[other],2,1,5,[mask]
+6,35,129,2,1,8,[unknown]
+6,36,129,2,1,5,[unknown]
+6,37,129,2,1,5,2
+7,8,129,2,1,5,[mask]
+7,9,129,2,1,5,[mask]
+7,10,129,2,1,5,[mask]
+7,11,129,2,1,5,[mask]
+7,12,129,2,1,5,1
+7,13,107,2,1,5,[mask]
+7,14,107,2,1,[unknown],1
+7,15,107,2,[mask],[unknown],1
+7,16,107,2,[mask],[unknown],1
+7,17,107,7,[mask],[unknown],1
+7,18,107,7,[mask],[unknown],1
+7,19,107,7,[mask],[unknown],1
+7,20,107,7,[mask],[unknown],1
+7,21,107,[mask],[mask],[unknown],1
+7,22,104,[mask],[mask],[unknown],2
+7,23,104,9,[mask],1,[other]
+7,24,104,9,[mask],3,[mask]
+7,25,104,[unknown],[mask],3,[other]
+7,26,104,[mask],[mask],3,[mask]
+7,27,104,9,[mask],[unknown],[other]
+7,28,112,9,[mask],[unknown],[mask]
+7,29,104,2,[mask],[unknown],[mask]
+7,30,107,2,[other],[unknown],[mask]
+7,31,107,2,3,1,[mask]
+7,32,107,2,[other],[unknown],[mask]
+7,33,107,2,[other],[unknown],[mask]
+7,34,107,2,[other],[unknown],[mask]
+7,35,107,2,[other],8,[mask]
+7,36,[other],2,[other],8,[mask]
+7,37,[other],9,[other],5,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index 7d0ff1de..a98e349c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,6,3,8,1
-8,9,121,6,3,8,1
-8,10,121,6,3,8,1
-8,11,104,6,3,8,1
-8,12,104,6,3,8,0
-8,13,121,6,3,8,0
-8,14,104,6,3,8,0
-8,15,104,3,3,0,0
-8,16,104,3,3,0,0
-8,17,104,3,3,0,0
-8,18,104,3,3,0,0
-8,19,104,3,3,0,0
-8,20,104,3,3,0,0
-8,21,104,3,3,0,0
-8,22,121,6,3,0,1
-8,23,113,6,3,8,1
-8,24,121,6,3,8,1
-8,25,104,6,3,8,1
-8,26,104,6,3,8,0
-8,27,104,6,3,0,0
-8,28,104,9,3,0,1
-8,29,104,3,3,0,1
-8,30,104,9,3,0,1
-8,31,104,3,3,0,1
-8,32,104,3,3,0,1
-8,33,103,6,3,8,1
-8,34,113,6,3,8,1
-8,35,121,6,3,8,1
-8,36,121,6,3,8,1
-8,37,104,6,3,8,1
+8,8,104,2,1,5,[mask]
+8,9,129,9,1,5,[mask]
+8,10,129,9,1,5,[mask]
+8,11,129,9,1,5,[mask]
+8,12,129,9,1,5,[mask]
+8,13,129,9,1,5,[mask]
+8,14,112,9,1,5,[other]
+8,15,107,7,[mask],[unknown],[mask]
+8,16,107,7,[mask],[unknown],[mask]
+8,17,104,7,[mask],[unknown],[mask]
+8,18,104,0,1,[unknown],2
+8,19,104,7,[mask],[unknown],[mask]
+8,20,104,9,1,8,[mask]
+8,21,112,9,1,[unknown],[mask]
+8,22,104,9,1,1,[mask]
+8,23,129,9,[other],1,[other]
+8,24,112,2,[other],1,[mask]
+8,25,129,2,[other],5,[mask]
+8,26,129,2,[other],1,[mask]
+8,27,129,2,[other],1,[other]
+8,28,129,2,1,1,[mask]
+8,29,129,9,1,5,[other]
+8,30,129,2,1,5,[other]
+8,31,129,2,1,5,[other]
+8,32,129,2,1,5,[other]
+8,33,129,2,1,5,[unknown]
+8,34,129,2,1,5,2
+8,35,129,2,1,5,2
+8,36,129,2,1,5,2
+8,37,107,2,1,[unknown],2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index 057d4027..9cd75ca4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,113,6,3,8,1
-9,9,121,6,3,8,1
-9,10,104,6,3,8,1
-9,11,104,6,3,8,1
-9,12,104,6,3,8,0
-9,13,104,6,3,8,0
-9,14,104,3,3,0,0
-9,15,104,3,3,0,0
-9,16,104,3,3,0,0
-9,17,104,3,3,0,0
-9,18,104,3,3,0,1
-9,19,103,4,3,0,1
-9,20,103,6,3,8,1
-9,21,113,6,3,8,1
-9,22,113,6,3,8,1
-9,23,113,6,3,8,1
-9,24,104,6,3,8,1
-9,25,104,6,3,8,1
-9,26,104,6,3,8,1
-9,27,104,6,3,8,1
-9,28,104,9,3,8,0
-9,29,104,3,3,0,0
-9,30,104,3,3,0,0
-9,31,104,3,3,0,1
-9,32,104,9,3,0,1
-9,33,104,3,3,0,1
-9,34,104,3,3,0,1
-9,35,103,6,3,8,1
-9,36,121,6,3,8,1
-9,37,113,6,3,8,1
+9,8,104,7,[mask],[unknown],[other]
+9,9,104,7,[mask],[unknown],[other]
+9,10,104,7,[mask],[unknown],[other]
+9,11,104,0,[mask],[unknown],0
+9,12,104,7,[mask],[unknown],2
+9,13,104,9,[other],[unknown],2
+9,14,104,9,[other],[unknown],[mask]
+9,15,[other],9,[other],[unknown],[mask]
+9,16,107,2,[other],1,[mask]
+9,17,107,2,[other],1,[mask]
+9,18,107,2,[other],1,[mask]
+9,19,107,2,[other],[unknown],[mask]
+9,20,110,9,[other],[unknown],[other]
+9,21,110,2,[other],[unknown],[other]
+9,22,110,2,3,[unknown],[other]
+9,23,110,1,1,[unknown],[other]
+9,24,110,2,[mask],[unknown],[other]
+9,25,110,2,[mask],7,[mask]
+9,26,[other],2,[mask],[unknown],[mask]
+9,27,[other],2,[mask],5,[mask]
+9,28,129,2,[other],8,[mask]
+9,29,129,2,1,5,[mask]
+9,30,129,9,1,5,[mask]
+9,31,129,9,1,5,[mask]
+9,32,129,9,1,5,[unknown]
+9,33,129,9,1,5,[mask]
+9,34,129,9,1,5,2
+9,35,129,2,1,5,2
+9,36,112,2,1,5,2
+9,37,107,7,[mask],[unknown],[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index 9e3ea8c3..fd1bb4cf 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,102,6,2,0,1
-0,9,102,6,2,0,1
-0,10,102,6,2,0,1
-0,11,102,6,2,8,1
-0,12,102,6,2,8,1
-0,13,102,6,2,8,1
-0,14,102,6,2,6,1
-0,15,102,6,2,8,1
-0,16,102,6,2,0,1
-0,17,102,6,2,0,1
-0,18,102,6,2,0,1
-0,19,102,6,2,0,1
-0,20,102,6,2,0,1
-0,21,102,6,2,0,1
-0,22,102,6,2,0,1
-0,23,102,6,2,0,1
-0,24,102,6,2,0,1
-0,25,102,6,2,0,1
-0,26,102,6,2,0,1
-0,27,102,6,2,0,1
-0,28,102,6,2,0,1
-0,29,102,6,2,0,1
-0,30,102,6,2,0,1
-0,31,102,6,2,0,1
-0,32,102,6,2,0,1
-0,33,102,6,2,0,1
-0,34,102,6,2,0,1
-0,35,102,6,2,0,1
-0,36,102,6,2,0,1
-0,37,102,6,2,0,1
+0,8,104,7,[mask],[unknown],1
+0,9,104,7,[mask],[unknown],0
+0,10,104,7,1,[unknown],0
+0,11,101,7,1,[unknown],0
+0,12,106,7,1,[unknown],[other]
+0,13,112,9,1,[unknown],0
+0,14,129,9,1,5,[unknown]
+0,15,112,2,1,1,[other]
+0,16,112,7,1,[unknown],2
+0,17,129,2,1,[unknown],2
+0,18,129,7,1,5,2
+0,19,112,7,1,1,2
+0,20,129,7,1,[unknown],2
+0,21,129,7,1,1,[other]
+0,22,112,7,1,5,[mask]
+0,23,118,7,[mask],[unknown],[mask]
+0,24,129,9,1,5,[other]
+0,25,112,9,1,5,[mask]
+0,26,[other],7,[mask],[unknown],[mask]
+0,27,129,9,1,5,[other]
+0,28,112,9,1,5,[other]
+0,29,129,7,1,[unknown],[mask]
+0,30,104,0,1,[unknown],[other]
+0,31,112,7,1,[unknown],[other]
+0,32,104,7,1,[unknown],2
+0,33,129,7,1,[unknown],[other]
+0,34,112,9,1,5,[other]
+0,35,112,7,1,[unknown],2
+0,36,108,7,1,[unknown],2
+0,37,129,7,1,5,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index 2a623d17..d23a5aeb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,102,6,2,8,1
-1,9,102,6,2,8,1
-1,10,102,6,2,8,1
-1,11,102,6,2,8,1
-1,12,102,6,2,8,1
-1,13,102,6,2,8,1
-1,14,102,6,2,8,1
-1,15,102,6,2,8,1
-1,16,102,6,2,6,1
-1,17,102,6,2,0,1
-1,18,102,6,2,0,1
-1,19,102,6,2,0,1
-1,20,102,6,2,0,1
-1,21,102,6,2,0,1
-1,22,102,6,2,0,1
-1,23,102,6,2,0,1
-1,24,102,6,2,0,1
-1,25,102,6,2,0,1
-1,26,102,6,2,0,1
-1,27,102,6,2,0,1
-1,28,102,6,2,0,1
-1,29,102,6,2,0,1
-1,30,102,6,2,0,1
-1,31,102,6,2,0,1
-1,32,102,6,2,0,1
-1,33,102,6,2,0,1
-1,34,102,6,2,0,1
-1,35,102,6,2,0,1
-1,36,102,6,2,0,1
-1,37,102,6,2,0,1
+1,8,104,7,[mask],1,1
+1,9,104,7,[mask],[unknown],[mask]
+1,10,[other],9,1,8,[mask]
+1,11,[other],9,1,5,[other]
+1,12,104,9,1,1,[other]
+1,13,112,9,1,[unknown],[other]
+1,14,112,7,1,[unknown],[other]
+1,15,112,2,1,5,[other]
+1,16,112,2,1,[unknown],[mask]
+1,17,129,2,1,[unknown],[mask]
+1,18,129,7,1,5,2
+1,19,129,2,1,1,2
+1,20,129,7,1,1,2
+1,21,129,7,1,1,2
+1,22,129,7,1,1,2
+1,23,129,7,1,1,2
+1,24,129,2,1,5,[mask]
+1,25,129,2,1,5,[mask]
+1,26,129,2,1,5,[mask]
+1,27,129,2,1,5,[mask]
+1,28,129,2,1,5,[mask]
+1,29,112,2,1,5,[mask]
+1,30,107,2,1,5,[mask]
+1,31,107,2,1,5,1
+1,32,107,2,[mask],[unknown],1
+1,33,107,2,[mask],[unknown],1
+1,34,107,2,[mask],[unknown],1
+1,35,107,7,[mask],[unknown],1
+1,36,107,7,[mask],[unknown],1
+1,37,107,[mask],[mask],[unknown],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index a8977415..cc5b8a64 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,102,6,2,8,1
-2,9,102,6,2,8,1
-2,10,102,6,2,8,1
-2,11,102,6,2,8,1
-2,12,102,6,2,8,1
-2,13,102,6,2,8,1
-2,14,102,6,2,0,1
-2,15,102,6,2,0,1
-2,16,102,6,2,0,1
-2,17,102,6,2,0,1
-2,18,102,6,2,0,1
-2,19,102,6,2,0,1
-2,20,102,6,2,0,1
-2,21,102,6,2,0,1
-2,22,102,6,2,0,1
-2,23,102,6,2,0,1
-2,24,102,6,2,0,1
-2,25,102,6,2,0,1
-2,26,102,6,2,0,1
-2,27,102,6,2,0,1
-2,28,102,6,2,0,1
-2,29,102,6,2,0,1
-2,30,102,6,2,0,1
-2,31,102,6,2,0,1
-2,32,102,6,2,0,1
-2,33,102,6,2,0,1
-2,34,102,6,2,0,1
-2,35,102,6,2,0,1
-2,36,102,6,2,0,1
-2,37,102,6,2,0,1
+2,8,112,2,[mask],5,[mask]
+2,9,104,7,[mask],[unknown],[mask]
+2,10,104,7,1,8,[mask]
+2,11,112,7,1,[unknown],[other]
+2,12,112,7,1,[unknown],[mask]
+2,13,107,7,[mask],[unknown],[mask]
+2,14,107,7,[mask],[unknown],[mask]
+2,15,[other],7,[mask],[unknown],2
+2,16,129,9,[other],1,[other]
+2,17,112,9,[other],1,[mask]
+2,18,104,1,[other],5,[mask]
+2,19,104,2,[other],[unknown],[other]
+2,20,110,9,[mask],[unknown],[other]
+2,21,110,2,[mask],[unknown],[mask]
+2,22,[other],2,[mask],[unknown],[other]
+2,23,110,2,[mask],5,[mask]
+2,24,[other],2,[other],7,[mask]
+2,25,104,2,[other],8,[mask]
+2,26,107,2,[other],1,[mask]
+2,27,107,2,[other],1,[mask]
+2,28,107,2,[other],8,[mask]
+2,29,107,2,1,5,[mask]
+2,30,107,2,1,5,[unknown]
+2,31,129,2,1,5,[mask]
+2,32,129,9,1,5,[other]
+2,33,107,2,1,5,[mask]
+2,34,107,2,1,5,[mask]
+2,35,107,2,1,[unknown],1
+2,36,107,2,[mask],[unknown],1
+2,37,107,7,[mask],[unknown],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index 780b6ded..2a4d06f3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,102,6,2,8,1
-3,9,102,6,2,8,1
-3,10,102,6,2,8,1
-3,11,102,6,2,8,1
-3,12,102,6,2,8,1
-3,13,102,6,2,8,1
-3,14,102,6,2,8,1
-3,15,102,6,2,8,1
-3,16,102,6,2,6,1
-3,17,102,6,2,0,1
-3,18,102,6,2,0,1
-3,19,102,6,2,0,1
-3,20,102,6,2,0,1
-3,21,102,6,2,0,1
-3,22,102,6,2,0,1
-3,23,102,6,2,0,1
-3,24,102,6,2,0,1
-3,25,102,6,2,0,1
-3,26,102,6,2,0,1
-3,27,102,6,2,0,1
-3,28,102,6,2,0,1
-3,29,102,6,2,0,1
-3,30,102,6,2,0,1
-3,31,102,6,2,0,1
-3,32,102,6,2,0,1
-3,33,102,6,2,0,1
-3,34,102,6,2,0,1
-3,35,102,6,2,0,1
-3,36,102,6,2,0,1
-3,37,102,6,2,0,1
-4,8,113,5,2,0,1
-4,9,102,6,2,0,1
-4,10,102,6,2,0,1
-4,11,102,6,2,0,1
-4,12,102,6,2,0,1
-4,13,102,6,2,0,1
-4,14,102,6,2,8,1
-4,15,102,6,2,8,1
-4,16,102,6,2,0,1
-4,17,102,6,2,0,1
-4,18,102,6,2,0,1
-4,19,102,6,2,0,1
-4,20,102,6,2,0,1
-4,21,102,6,2,0,1
-4,22,102,6,2,0,1
-4,23,102,6,2,0,1
-4,24,102,6,2,0,1
-4,25,102,6,2,0,1
-4,26,102,6,2,0,1
-4,27,102,6,2,0,1
-4,28,102,6,2,0,1
-4,29,102,6,2,0,1
-4,30,102,6,2,0,1
-4,31,102,6,2,0,1
-4,32,102,6,2,0,1
-4,33,102,6,2,0,1
-4,34,102,6,2,0,1
-4,35,102,6,2,0,1
-4,36,102,6,2,0,1
-4,37,102,6,2,0,1
+3,8,104,[mask],[mask],[unknown],[mask]
+3,9,104,0,[mask],8,[other]
+3,10,104,[mask],[mask],6,[mask]
+3,11,104,0,[mask],8,[mask]
+3,12,104,0,[mask],8,[mask]
+3,13,104,1,[mask],8,[mask]
+3,14,[other],2,[mask],[unknown],[mask]
+3,15,129,1,[other],6,[mask]
+3,16,107,2,[other],1,[mask]
+3,17,107,2,[other],6,[mask]
+3,18,129,9,[other],6,[mask]
+3,19,129,9,[other],5,[mask]
+3,20,129,9,[other],5,[mask]
+3,21,129,9,[other],5,[mask]
+3,22,129,9,[other],5,[other]
+3,23,110,9,[other],5,[mask]
+3,24,107,2,[other],8,[other]
+3,25,107,2,1,5,[other]
+3,26,110,9,1,[unknown],[other]
+3,27,[other],9,1,[unknown],[other]
+3,28,[other],2,1,[unknown],[other]
+3,29,[other],2,1,[unknown],[other]
+3,30,107,2,1,[unknown],[unknown]
+3,31,129,2,1,[unknown],[mask]
+3,32,129,2,1,[unknown],2
+3,33,129,2,1,5,2
+3,34,129,2,1,5,2
+3,35,129,2,1,5,2
+3,36,129,2,1,5,2
+3,37,129,2,1,5,2
+4,8,104,9,1,[unknown],[mask]
+4,9,[other],7,1,[unknown],[other]
+4,10,104,9,1,[unknown],[other]
+4,11,[other],9,1,[unknown],[other]
+4,12,[other],9,1,[unknown],[other]
+4,13,[other],9,1,[unknown],[other]
+4,14,129,2,1,5,[unknown]
+4,15,129,2,1,1,2
+4,16,129,2,1,1,2
+4,17,129,7,1,5,2
+4,18,129,2,1,1,2
+4,19,129,7,1,5,2
+4,20,129,7,1,1,2
+4,21,129,7,1,1,[mask]
+4,22,129,7,1,5,[mask]
+4,23,129,2,1,5,[mask]
+4,24,112,2,1,5,[mask]
+4,25,107,2,[mask],[unknown],[mask]
+4,26,129,2,[mask],5,[mask]
+4,27,129,9,1,5,[mask]
+4,28,129,9,1,5,[mask]
+4,29,129,9,1,5,[mask]
+4,30,129,9,1,5,[mask]
+4,31,129,9,1,5,[mask]
+4,32,129,9,1,5,[mask]
+4,33,129,9,1,5,[other]
+4,34,112,9,1,5,[mask]
+4,35,107,7,[mask],[unknown],[mask]
+4,36,107,7,[mask],[unknown],[mask]
+4,37,104,7,[mask],[unknown],[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index a6dfbfa4..529df285 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,7,2,4,2
-5,9,113,5,2,0,1
-5,10,113,6,2,0,1
-5,11,113,6,2,0,1
-5,12,113,6,2,0,1
-5,13,113,6,2,0,1
-5,14,102,6,2,0,1
-5,15,102,6,2,0,1
-5,16,102,6,2,0,1
-5,17,102,6,2,0,1
-5,18,102,6,2,0,1
-5,19,102,6,2,0,1
-5,20,102,6,2,0,1
-5,21,102,6,2,0,1
-5,22,102,6,2,0,1
-5,23,102,6,2,0,1
-5,24,102,6,2,0,1
-5,25,102,6,2,0,1
-5,26,102,6,2,0,1
-5,27,102,6,2,0,1
-5,28,102,6,2,0,1
-5,29,102,6,2,0,1
-5,30,102,6,2,0,1
-5,31,102,6,2,0,1
-5,32,102,6,2,0,1
-5,33,102,6,2,0,1
-5,34,102,6,2,0,1
-5,35,102,6,2,0,1
-5,36,102,6,2,0,1
-5,37,102,6,2,0,1
+5,8,102,7,[mask],9,1
+5,9,107,[mask],[mask],[unknown],1
+5,10,107,[mask],[mask],[unknown],1
+5,11,107,[mask],[mask],[unknown],1
+5,12,104,[mask],[mask],[unknown],1
+5,13,104,[mask],[mask],[unknown],1
+5,14,104,[mask],[mask],[unknown],1
+5,15,104,[mask],[mask],3,1
+5,16,104,[mask],[mask],3,[other]
+5,17,112,[unknown],[mask],3,[other]
+5,18,111,[mask],[mask],3,2
+5,19,104,9,[mask],3,2
+5,20,104,5,[mask],3,2
+5,21,104,5,[mask],3,2
+5,22,104,5,[mask],3,2
+5,23,104,[unknown],[mask],3,2
+5,24,104,9,[mask],3,[unknown]
+5,25,104,9,[mask],3,[mask]
+5,26,112,[unknown],[mask],3,[mask]
+5,27,112,[unknown],[mask],3,[mask]
+5,28,112,[mask],[mask],3,[mask]
+5,29,104,[mask],[mask],7,[mask]
+5,30,107,9,[mask],3,[mask]
+5,31,107,2,[mask],3,[mask]
+5,32,107,2,[mask],6,[mask]
+5,33,104,2,[mask],6,[mask]
+5,34,104,9,[mask],6,[mask]
+5,35,[other],9,[mask],6,[mask]
+5,36,[other],2,[other],6,[mask]
+5,37,107,9,[other],6,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index eeb5e8ef..819defac 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,102,6,2,8,1
-6,9,102,6,2,8,1
-6,10,102,6,2,8,1
-6,11,102,6,2,8,1
-6,12,102,6,2,8,1
-6,13,102,6,2,8,1
-6,14,102,6,2,8,1
-6,15,102,6,2,8,1
-6,16,102,6,2,6,1
-6,17,102,6,2,0,1
-6,18,102,6,2,0,1
-6,19,102,6,2,0,1
-6,20,102,6,2,0,1
-6,21,102,6,2,0,1
-6,22,102,6,2,0,1
-6,23,102,6,2,0,1
-6,24,102,6,2,0,1
-6,25,102,6,2,0,1
-6,26,102,6,2,0,1
-6,27,102,6,2,0,1
-6,28,102,6,2,0,1
-6,29,102,6,2,0,1
-6,30,102,6,2,0,1
-6,31,102,6,2,0,1
-6,32,102,6,2,0,1
-6,33,102,6,2,0,1
-6,34,102,6,2,0,1
-6,35,102,6,2,0,1
-6,36,102,6,2,0,1
-6,37,102,6,2,0,1
-7,8,102,6,2,8,1
-7,9,102,6,2,8,1
-7,10,102,6,2,8,1
-7,11,102,6,2,8,1
-7,12,102,6,2,8,1
-7,13,102,6,2,8,1
-7,14,102,6,3,6,1
-7,15,102,6,2,8,1
-7,16,102,6,2,0,1
-7,17,102,6,2,0,1
-7,18,102,6,2,0,1
-7,19,102,6,2,0,1
-7,20,102,6,2,0,1
-7,21,102,6,2,0,1
-7,22,102,6,2,0,1
-7,23,102,6,2,0,1
-7,24,102,6,2,0,1
-7,25,102,6,2,0,1
-7,26,102,6,2,0,1
-7,27,102,6,2,0,1
-7,28,102,6,2,0,1
-7,29,102,6,2,0,1
-7,30,102,6,2,0,1
-7,31,102,6,2,0,1
-7,32,102,6,2,0,1
-7,33,102,6,2,0,1
-7,34,102,6,2,0,1
-7,35,102,6,2,0,1
-7,36,102,6,2,0,1
-7,37,102,6,2,0,1
+6,8,129,9,1,5,[other]
+6,9,129,9,1,5,[mask]
+6,10,129,9,1,5,[other]
+6,11,129,9,1,5,[other]
+6,12,129,9,1,5,[other]
+6,13,129,2,1,5,[mask]
+6,14,129,2,1,8,[other]
+6,15,112,7,1,[unknown],[unknown]
+6,16,107,7,1,[unknown],2
+6,17,108,7,[mask],[unknown],2
+6,18,104,7,1,5,2
+6,19,108,7,1,[unknown],2
+6,20,104,7,1,5,2
+6,21,112,9,[mask],1,[mask]
+6,22,129,9,[other],5,[mask]
+6,23,104,9,[other],5,[mask]
+6,24,112,9,8,1,[mask]
+6,25,129,9,[other],5,[mask]
+6,26,104,9,[other],1,[mask]
+6,27,104,9,[other],1,[other]
+6,28,112,7,[other],1,[mask]
+6,29,[other],1,[other],8,[other]
+6,30,[other],2,[mask],[unknown],[other]
+6,31,[other],2,1,[unknown],[other]
+6,32,110,2,1,5,[other]
+6,33,[other],2,1,[unknown],[other]
+6,34,[other],2,1,5,[mask]
+6,35,129,2,1,8,[unknown]
+6,36,129,2,1,5,[unknown]
+6,37,129,2,1,5,2
+7,8,129,2,1,5,[mask]
+7,9,129,2,1,5,[mask]
+7,10,129,2,1,5,[mask]
+7,11,129,2,1,5,[mask]
+7,12,129,2,1,5,1
+7,13,107,2,1,5,[mask]
+7,14,107,2,1,[unknown],1
+7,15,107,2,[mask],[unknown],1
+7,16,107,2,[mask],[unknown],1
+7,17,107,7,[mask],[unknown],1
+7,18,107,7,[mask],[unknown],1
+7,19,107,7,[mask],[unknown],1
+7,20,107,7,[mask],[unknown],1
+7,21,107,[mask],[mask],[unknown],1
+7,22,104,[mask],[mask],[unknown],2
+7,23,104,9,[mask],1,[other]
+7,24,104,9,[mask],3,[mask]
+7,25,104,[unknown],[mask],3,[other]
+7,26,104,[mask],[mask],3,[mask]
+7,27,104,9,[mask],[unknown],[other]
+7,28,112,9,[mask],[unknown],[mask]
+7,29,104,2,[mask],[unknown],[mask]
+7,30,107,2,[other],[unknown],[mask]
+7,31,107,2,3,1,[mask]
+7,32,107,2,[other],[unknown],[mask]
+7,33,107,2,[other],[unknown],[mask]
+7,34,107,2,[other],[unknown],[mask]
+7,35,107,2,[other],8,[mask]
+7,36,[other],2,[other],8,[mask]
+7,37,[other],9,[other],5,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index 811522ec..a98e349c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,102,6,2,8,0
-8,9,102,6,2,8,1
-8,10,102,6,2,8,1
-8,11,102,6,2,8,1
-8,12,102,6,2,8,1
-8,13,102,6,2,8,1
-8,14,102,6,2,6,1
-8,15,102,6,3,6,1
-8,16,102,6,2,6,1
-8,17,102,6,2,0,1
-8,18,102,6,2,0,1
-8,19,102,6,2,0,1
-8,20,102,6,2,0,1
-8,21,102,6,2,0,1
-8,22,102,6,2,0,1
-8,23,102,6,2,0,1
-8,24,102,6,2,0,1
-8,25,102,6,2,0,1
-8,26,102,6,2,0,1
-8,27,102,6,2,0,1
-8,28,102,6,2,0,1
-8,29,102,6,2,0,1
-8,30,102,6,2,0,1
-8,31,102,6,2,0,1
-8,32,102,6,2,0,1
-8,33,102,6,2,0,1
-8,34,102,6,2,0,1
-8,35,102,6,2,0,1
-8,36,102,6,2,0,1
-8,37,102,6,2,0,1
+8,8,104,2,1,5,[mask]
+8,9,129,9,1,5,[mask]
+8,10,129,9,1,5,[mask]
+8,11,129,9,1,5,[mask]
+8,12,129,9,1,5,[mask]
+8,13,129,9,1,5,[mask]
+8,14,112,9,1,5,[other]
+8,15,107,7,[mask],[unknown],[mask]
+8,16,107,7,[mask],[unknown],[mask]
+8,17,104,7,[mask],[unknown],[mask]
+8,18,104,0,1,[unknown],2
+8,19,104,7,[mask],[unknown],[mask]
+8,20,104,9,1,8,[mask]
+8,21,112,9,1,[unknown],[mask]
+8,22,104,9,1,1,[mask]
+8,23,129,9,[other],1,[other]
+8,24,112,2,[other],1,[mask]
+8,25,129,2,[other],5,[mask]
+8,26,129,2,[other],1,[mask]
+8,27,129,2,[other],1,[other]
+8,28,129,2,1,1,[mask]
+8,29,129,9,1,5,[other]
+8,30,129,2,1,5,[other]
+8,31,129,2,1,5,[other]
+8,32,129,2,1,5,[other]
+8,33,129,2,1,5,[unknown]
+8,34,129,2,1,5,2
+8,35,129,2,1,5,2
+8,36,129,2,1,5,2
+8,37,107,2,1,[unknown],2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index 12621cb2..9cd75ca4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,113,6,2,6,1
-9,9,102,6,2,0,1
-9,10,102,6,2,0,1
-9,11,102,6,2,0,1
-9,12,102,6,2,0,1
-9,13,102,6,2,0,1
-9,14,102,6,2,0,1
-9,15,102,6,2,0,1
-9,16,102,6,2,0,1
-9,17,102,6,2,0,1
-9,18,102,6,2,0,1
-9,19,102,6,2,0,1
-9,20,102,6,2,0,1
-9,21,102,6,2,0,1
-9,22,102,6,2,0,1
-9,23,102,6,2,0,1
-9,24,102,6,2,0,1
-9,25,102,6,2,0,1
-9,26,102,6,2,0,1
-9,27,102,6,2,0,1
-9,28,102,6,2,0,1
-9,29,102,6,2,0,1
-9,30,102,6,2,0,1
-9,31,102,6,2,0,1
-9,32,102,6,2,0,1
-9,33,102,6,2,0,1
-9,34,102,6,2,0,1
-9,35,102,6,2,0,1
-9,36,102,6,2,0,1
-9,37,102,6,2,0,1
+9,8,104,7,[mask],[unknown],[other]
+9,9,104,7,[mask],[unknown],[other]
+9,10,104,7,[mask],[unknown],[other]
+9,11,104,0,[mask],[unknown],0
+9,12,104,7,[mask],[unknown],2
+9,13,104,9,[other],[unknown],2
+9,14,104,9,[other],[unknown],[mask]
+9,15,[other],9,[other],[unknown],[mask]
+9,16,107,2,[other],1,[mask]
+9,17,107,2,[other],1,[mask]
+9,18,107,2,[other],1,[mask]
+9,19,107,2,[other],[unknown],[mask]
+9,20,110,9,[other],[unknown],[other]
+9,21,110,2,[other],[unknown],[other]
+9,22,110,2,3,[unknown],[other]
+9,23,110,1,1,[unknown],[other]
+9,24,110,2,[mask],[unknown],[other]
+9,25,110,2,[mask],7,[mask]
+9,26,[other],2,[mask],[unknown],[mask]
+9,27,[other],2,[mask],5,[mask]
+9,28,129,2,[other],8,[mask]
+9,29,129,2,1,5,[mask]
+9,30,129,9,1,5,[mask]
+9,31,129,9,1,5,[mask]
+9,32,129,9,1,5,[unknown]
+9,33,129,9,1,5,[mask]
+9,34,129,9,1,5,2
+9,35,129,2,1,5,2
+9,36,112,2,1,5,2
+9,37,107,7,[mask],[unknown],[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
index 2a8d05f6..8cab47cb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,113,6,2,0,1
-0,9,113,6,2,0,1
-0,10,113,6,2,0,1
-0,11,113,6,2,0,1
-0,12,113,6,2,0,1
-0,13,113,6,2,0,2
-0,14,113,6,2,0,2
-0,15,113,6,2,0,2
-0,16,113,6,2,0,2
-0,17,113,6,2,0,2
-0,18,113,6,2,0,2
-0,19,113,6,2,0,2
-0,20,113,6,2,0,2
-0,21,113,6,2,0,2
-0,22,113,6,2,0,2
-0,23,113,6,2,0,2
-0,24,113,6,2,0,2
-0,25,113,6,2,0,2
-0,26,113,6,2,0,2
-0,27,113,6,2,0,2
-0,28,113,6,2,0,2
-0,29,113,6,2,0,2
-0,30,113,6,2,0,2
-0,31,113,6,2,0,2
-0,32,113,6,2,0,2
-0,33,113,6,2,0,2
-0,34,113,6,2,0,2
-0,35,113,6,2,0,2
-0,36,113,6,2,0,2
-0,37,113,6,2,0,2
+0,8,102,3,9,5,1
+0,9,102,4,3,0,1
+0,10,102,6,9,2,1
+0,11,102,4,3,0,1
+0,12,112,6,9,2,1
+0,13,102,4,3,0,1
+0,14,112,6,9,0,0
+0,15,102,4,3,0,1
+0,16,112,6,9,0,0
+0,17,102,4,3,0,2
+0,18,112,6,3,0,0
+0,19,102,4,9,0,0
+0,20,112,4,3,0,0
+0,21,112,4,3,0,0
+0,22,112,4,3,0,0
+0,23,112,4,3,0,0
+0,24,112,4,3,0,0
+0,25,112,4,3,0,0
+0,26,112,4,3,0,0
+0,27,112,4,3,0,0
+0,28,112,4,3,0,0
+0,29,112,4,3,0,0
+0,30,112,4,3,0,0
+0,31,112,4,3,0,0
+0,32,112,4,3,0,0
+0,33,112,4,3,0,0
+0,34,112,4,3,0,0
+0,35,112,4,3,0,0
+0,36,112,4,3,0,0
+0,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
index cd416b33..5429f965 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,113,6,0,9,0
-1,9,100,6,0,8,0
-1,10,121,6,0,8,0
-1,11,120,4,0,8,0
-1,12,120,4,0,8,0
-1,13,120,0,0,5,0
-1,14,109,4,8,8,0
-1,15,109,4,9,5,0
-1,16,109,4,9,5,0
-1,17,109,4,9,5,0
-1,18,109,4,0,5,0
-1,19,109,4,9,5,0
-1,20,109,4,0,9,0
-1,21,109,4,0,5,0
-1,22,127,4,0,9,0
-1,23,120,4,0,9,0
-1,24,120,4,0,9,0
-1,25,109,4,0,5,0
-1,26,109,4,9,5,0
-1,27,127,4,0,9,0
-1,28,109,4,9,5,0
-1,29,109,4,9,5,0
-1,30,109,4,3,5,0
-1,31,109,4,3,5,0
-1,32,109,4,3,5,0
-1,33,109,4,3,5,0
-1,34,127,6,3,5,0
-1,35,109,6,3,5,0
-1,36,112,6,3,5,0
-1,37,103,6,3,0,0
+1,8,102,4,0,5,0
+1,9,102,6,0,2,1
+1,10,111,4,2,2,1
+1,11,112,4,9,5,1
+1,12,112,4,9,8,1
+1,13,112,4,9,0,0
+1,14,112,7,9,0,1
+1,15,112,4,3,0,0
+1,16,112,4,9,0,0
+1,17,112,4,9,0,1
+1,18,112,4,9,0,0
+1,19,112,4,3,0,0
+1,20,112,4,9,0,0
+1,21,112,4,3,0,0
+1,22,112,4,9,0,0
+1,23,112,4,3,0,0
+1,24,112,4,3,0,0
+1,25,112,4,3,0,0
+1,26,112,4,3,0,0
+1,27,112,4,3,0,0
+1,28,112,4,3,0,0
+1,29,112,4,3,0,0
+1,30,112,4,3,0,0
+1,31,112,4,3,0,0
+1,32,112,4,3,0,0
+1,33,112,4,3,0,0
+1,34,112,4,3,0,0
+1,35,112,4,3,0,0
+1,36,112,4,3,0,0
+1,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
index cedb1297..d61975e1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,113,4,2,0,1
-2,9,113,6,2,0,0
-2,10,113,6,2,0,0
-2,11,113,6,2,0,1
-2,12,113,6,2,0,1
-2,13,113,6,2,0,1
-2,14,113,6,2,0,2
-2,15,113,6,2,0,2
-2,16,113,6,2,0,2
-2,17,113,6,2,0,2
-2,18,113,6,2,0,2
-2,19,113,6,2,0,2
-2,20,113,6,2,0,2
-2,21,113,6,2,0,2
-2,22,113,6,2,0,2
-2,23,113,6,2,0,2
-2,24,113,6,2,0,2
-2,25,113,6,2,0,2
-2,26,113,6,2,0,2
-2,27,113,6,2,0,2
-2,28,113,6,2,0,2
-2,29,113,6,2,0,2
-2,30,113,6,2,0,2
-2,31,113,6,2,0,2
-2,32,113,6,2,0,2
-2,33,113,6,2,0,2
-2,34,113,6,2,0,2
-2,35,113,6,2,0,2
-2,36,113,6,2,0,2
-2,37,113,6,2,0,2
+2,8,112,4,0,5,1
+2,9,112,4,3,0,0
+2,10,112,4,9,0,1
+2,11,112,4,9,0,0
+2,12,112,4,3,0,0
+2,13,112,4,9,0,0
+2,14,112,4,3,0,0
+2,15,112,4,9,0,0
+2,16,112,4,3,0,0
+2,17,112,4,3,0,0
+2,18,112,4,3,0,0
+2,19,112,4,3,0,0
+2,20,112,4,3,0,0
+2,21,112,4,3,0,0
+2,22,112,4,3,0,0
+2,23,112,4,3,0,0
+2,24,112,4,3,0,0
+2,25,112,4,3,0,0
+2,26,112,4,3,0,0
+2,27,112,4,3,0,0
+2,28,112,4,3,0,0
+2,29,112,4,3,0,0
+2,30,112,4,3,0,0
+2,31,112,4,3,0,0
+2,32,112,4,3,0,0
+2,33,112,4,3,0,0
+2,34,112,4,3,0,0
+2,35,112,4,3,0,0
+2,36,112,4,3,0,0
+2,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
index 477f3560..dae0b6c2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,121,4,2,8,1
-3,9,121,4,2,8,1
-3,10,121,4,9,8,0
-3,11,109,4,9,5,0
-3,12,109,6,9,5,0
-3,13,121,6,0,5,0
-3,14,109,4,9,5,0
-3,15,109,4,0,5,0
-3,16,120,8,0,9,0
-3,17,120,4,0,5,0
-3,18,127,4,0,9,0
-3,19,120,4,0,9,0
-3,20,109,4,0,8,0
-3,21,109,4,9,5,0
-3,22,109,4,0,5,0
-3,23,109,4,9,5,0
-3,24,109,4,9,5,0
-3,25,109,4,3,5,0
-3,26,109,4,9,5,0
-3,27,127,4,0,9,0
-3,28,127,4,0,5,0
-3,29,127,4,0,9,0
-3,30,109,4,0,5,0
-3,31,109,4,0,9,0
-3,32,109,4,0,5,0
-3,33,109,4,9,5,0
-3,34,109,4,9,5,0
-3,35,109,4,9,5,0
-3,36,109,4,3,5,0
-3,37,109,4,3,5,0
-4,8,113,7,2,8,1
-4,9,113,6,2,8,0
-4,10,121,6,2,8,0
-4,11,121,4,2,8,0
-4,12,109,4,9,5,0
-4,13,109,6,9,5,0
-4,14,121,6,0,9,0
-4,15,120,4,0,5,0
-4,16,120,9,0,9,0
-4,17,120,0,0,5,0
-4,18,109,4,0,5,0
-4,19,109,4,0,5,0
-4,20,127,4,9,9,0
-4,21,127,4,9,5,0
-4,22,109,4,9,5,0
-4,23,109,4,9,5,0
-4,24,109,4,9,5,0
-4,25,109,6,3,9,0
-4,26,109,4,0,5,0
-4,27,127,6,0,9,0
-4,28,127,9,0,5,0
-4,29,127,4,0,9,0
-4,30,120,4,0,5,0
-4,31,109,4,3,9,0
-4,32,109,4,9,5,0
-4,33,109,4,9,5,0
-4,34,109,4,9,5,0
-4,35,109,4,9,5,0
-4,36,109,4,3,5,0
-4,37,109,4,3,5,0
+3,8,112,3,3,5,0
+3,9,102,4,3,5,0
+3,10,112,3,3,0,0
+3,11,112,4,3,5,0
+3,12,112,4,3,0,0
+3,13,112,4,3,0,0
+3,14,112,4,9,0,0
+3,15,112,4,3,0,0
+3,16,112,4,9,0,0
+3,17,112,4,3,0,0
+3,18,112,4,9,0,0
+3,19,112,4,3,0,0
+3,20,112,4,3,0,0
+3,21,112,4,3,0,0
+3,22,112,4,3,0,0
+3,23,112,4,3,0,0
+3,24,112,4,3,0,0
+3,25,112,4,3,0,0
+3,26,112,4,3,0,0
+3,27,112,4,3,0,0
+3,28,112,4,3,0,0
+3,29,112,4,3,0,0
+3,30,112,4,3,0,0
+3,31,112,4,3,0,0
+3,32,112,4,3,0,0
+3,33,112,4,3,0,0
+3,34,112,4,3,0,0
+3,35,112,4,3,0,0
+3,36,112,4,3,0,0
+3,37,112,4,3,0,0
+4,8,113,4,9,5,1
+4,9,112,4,9,8,1
+4,10,118,4,9,8,1
+4,11,102,4,9,0,1
+4,12,112,4,9,0,1
+4,13,112,4,9,0,1
+4,14,112,4,9,0,1
+4,15,112,4,9,0,1
+4,16,112,4,3,0,0
+4,17,112,4,9,0,1
+4,18,112,4,3,0,0
+4,19,112,4,3,0,1
+4,20,112,4,9,0,0
+4,21,112,4,3,0,0
+4,22,112,4,9,0,0
+4,23,112,4,3,0,0
+4,24,112,4,9,0,0
+4,25,112,4,3,0,0
+4,26,112,4,3,0,0
+4,27,112,4,9,0,0
+4,28,112,4,3,0,0
+4,29,112,4,3,0,0
+4,30,112,4,3,0,0
+4,31,112,4,3,0,0
+4,32,112,4,3,0,0
+4,33,112,4,3,0,0
+4,34,112,4,3,0,0
+4,35,112,4,3,0,0
+4,36,112,4,3,0,0
+4,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
index e21fb885..83cec41e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,113,6,2,1,1
-5,9,113,6,2,0,2
-5,10,113,6,2,0,2
-5,11,113,6,2,0,2
-5,12,113,6,2,0,2
-5,13,113,6,2,0,2
-5,14,113,6,2,0,2
-5,15,113,6,2,0,2
-5,16,113,6,2,0,2
-5,17,113,6,2,0,2
-5,18,113,6,2,0,2
-5,19,113,6,2,0,2
-5,20,113,6,2,0,2
-5,21,113,6,2,0,2
-5,22,113,6,2,0,2
-5,23,113,6,2,0,2
-5,24,113,6,2,0,2
-5,25,113,6,2,0,2
-5,26,113,6,2,0,2
-5,27,113,6,2,0,2
-5,28,113,6,2,0,2
-5,29,113,6,2,0,2
-5,30,113,6,2,0,2
-5,31,113,6,2,0,2
-5,32,113,6,2,0,2
-5,33,113,6,2,0,2
-5,34,113,6,2,0,2
-5,35,113,6,2,0,2
-5,36,113,6,2,0,2
-5,37,113,6,2,0,2
+5,8,108,4,9,8,1
+5,9,104,4,9,8,1
+5,10,119,4,9,8,1
+5,11,104,4,9,8,1
+5,12,119,4,9,8,1
+5,13,104,4,9,8,0
+5,14,102,4,9,5,1
+5,15,119,4,3,0,0
+5,16,112,4,3,6,0
+5,17,112,4,3,5,0
+5,18,112,4,3,5,0
+5,19,112,4,3,0,0
+5,20,112,4,3,0,0
+5,21,112,4,3,6,0
+5,22,112,4,3,0,0
+5,23,112,4,9,0,0
+5,24,112,4,3,6,0
+5,25,112,4,3,0,0
+5,26,112,4,9,0,0
+5,27,112,4,3,0,0
+5,28,112,4,3,0,0
+5,29,112,4,3,6,0
+5,30,112,4,3,0,0
+5,31,112,4,3,0,0
+5,32,112,4,3,6,0
+5,33,112,4,3,0,0
+5,34,112,4,3,0,0
+5,35,112,4,3,0,0
+5,36,112,4,3,6,0
+5,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
index 908877f2..2ce84b09 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,113,6,2,8,0
-6,9,113,6,2,8,0
-6,10,113,6,2,8,0
-6,11,121,6,2,8,0
-6,12,121,6,0,8,0
-6,13,109,6,0,8,0
-6,14,109,4,0,8,0
-6,15,109,4,0,5,0
-6,16,109,4,0,5,0
-6,17,109,4,9,5,0
-6,18,109,4,0,5,0
-6,19,109,4,9,5,0
-6,20,109,4,9,5,0
-6,21,109,4,3,9,0
-6,22,109,4,0,5,0
-6,23,109,4,3,9,0
-6,24,109,4,0,5,0
-6,25,127,4,0,9,0
-6,26,109,4,0,5,0
-6,27,127,4,0,9,0
-6,28,109,4,0,5,0
-6,29,109,4,9,5,0
-6,30,109,4,9,5,0
-6,31,109,4,9,5,0
-6,32,109,4,3,5,0
-6,33,109,4,3,5,0
-6,34,109,4,3,5,0
-6,35,109,4,3,5,0
-6,36,109,6,3,5,0
-6,37,127,6,3,9,0
-7,8,109,4,9,6,0
-7,9,121,4,9,6,0
-7,10,121,4,2,8,1
-7,11,109,4,9,8,0
-7,12,109,4,9,5,0
-7,13,109,4,9,5,0
-7,14,109,6,3,5,0
-7,15,121,6,0,5,0
-7,16,109,6,0,8,0
-7,17,109,4,0,5,0
-7,18,127,4,0,9,0
-7,19,120,4,0,5,0
-7,20,127,4,3,9,0
-7,21,109,4,9,5,0
-7,22,109,4,3,0,0
-7,23,109,4,9,5,0
-7,24,109,4,9,5,0
-7,25,109,4,3,5,0
-7,26,109,4,3,5,0
-7,27,109,6,3,5,0
-7,28,109,6,3,5,0
-7,29,112,6,3,9,0
-7,30,121,6,3,5,0
-7,31,112,6,3,9,0
-7,32,121,6,3,5,0
-7,33,103,6,3,0,0
-7,34,103,6,3,0,0
-7,35,103,6,2,0,0
-7,36,103,6,2,0,0
-7,37,103,6,2,0,0
+6,8,102,4,9,5,1
+6,9,112,4,3,0,1
+6,10,112,4,9,5,1
+6,11,112,4,3,0,1
+6,12,112,4,9,0,1
+6,13,112,4,3,0,0
+6,14,112,4,9,0,1
+6,15,112,4,3,0,0
+6,16,112,4,9,0,0
+6,17,112,4,3,0,0
+6,18,112,4,9,0,0
+6,19,112,4,3,0,0
+6,20,112,4,9,0,0
+6,21,112,4,3,0,0
+6,22,112,4,9,0,0
+6,23,112,4,3,0,0
+6,24,112,4,3,0,0
+6,25,112,4,3,0,0
+6,26,112,4,3,0,0
+6,27,112,4,3,0,0
+6,28,112,4,3,0,0
+6,29,112,4,3,0,0
+6,30,112,4,3,0,0
+6,31,112,4,3,0,0
+6,32,112,4,3,0,0
+6,33,112,4,3,0,0
+6,34,112,4,3,0,0
+6,35,112,4,3,0,0
+6,36,112,4,3,0,0
+6,37,112,4,3,0,0
+7,8,102,4,0,5,1
+7,9,112,4,0,8,1
+7,10,118,4,9,8,1
+7,11,112,4,9,0,1
+7,12,112,4,9,0,1
+7,13,112,4,9,0,1
+7,14,112,4,9,0,1
+7,15,112,4,9,0,0
+7,16,112,4,3,0,0
+7,17,112,4,9,0,1
+7,18,112,4,9,0,0
+7,19,112,4,3,0,0
+7,20,112,4,9,0,0
+7,21,112,4,3,0,0
+7,22,112,4,9,0,0
+7,23,112,4,3,0,0
+7,24,112,4,3,0,0
+7,25,112,4,3,0,0
+7,26,112,4,3,0,0
+7,27,112,4,3,0,0
+7,28,112,4,3,0,0
+7,29,112,4,3,0,0
+7,30,112,4,3,0,0
+7,31,112,4,3,0,0
+7,32,112,4,3,0,0
+7,33,112,4,3,0,0
+7,34,112,4,3,0,0
+7,35,112,4,3,0,0
+7,36,112,4,3,0,0
+7,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
index d1cabc88..c373d0f8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,6,2,8,0
-8,9,113,6,2,8,0
-8,10,113,6,2,8,0
-8,11,113,6,2,0,0
-8,12,113,6,2,0,0
-8,13,113,6,8,0,0
-8,14,113,6,8,0,0
-8,15,113,9,7,0,0
-8,16,113,6,2,0,1
-8,17,113,6,2,0,1
-8,18,113,6,2,0,1
-8,19,113,6,2,0,1
-8,20,113,6,2,0,2
-8,21,113,6,2,0,2
-8,22,113,6,2,0,2
-8,23,113,6,2,0,2
-8,24,113,6,2,0,2
-8,25,113,6,2,0,2
-8,26,113,6,2,0,2
-8,27,113,6,2,0,2
-8,28,113,6,2,0,2
-8,29,113,6,2,0,2
-8,30,113,6,2,0,2
-8,31,113,6,2,0,2
-8,32,113,6,2,0,2
-8,33,113,6,2,0,2
-8,34,113,6,2,0,2
-8,35,113,6,2,0,2
-8,36,113,6,2,0,2
-8,37,113,6,2,0,2
+8,8,112,4,9,5,0
+8,9,112,4,0,5,0
+8,10,112,7,8,5,0
+8,11,112,4,9,5,0
+8,12,112,4,9,5,0
+8,13,112,4,9,5,1
+8,14,112,4,9,0,0
+8,15,112,4,9,0,0
+8,16,112,4,3,0,0
+8,17,112,4,9,0,0
+8,18,112,4,3,0,0
+8,19,112,4,3,0,0
+8,20,112,4,3,0,0
+8,21,112,4,9,0,0
+8,22,112,4,3,0,0
+8,23,112,4,3,0,0
+8,24,112,4,3,0,0
+8,25,112,4,3,0,0
+8,26,112,4,3,0,0
+8,27,112,4,3,0,0
+8,28,112,4,3,0,0
+8,29,112,4,3,0,0
+8,30,112,4,3,0,0
+8,31,112,4,3,0,0
+8,32,112,4,3,0,0
+8,33,112,4,3,0,0
+8,34,112,4,3,0,0
+8,35,112,4,3,0,0
+8,36,112,4,3,0,0
+8,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
index f9969b00..3613e328 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,100,6,0,9,0
-9,9,100,9,8,9,0
-9,10,109,4,9,5,1
-9,11,113,6,0,9,0
-9,12,100,9,0,9,0
-9,13,109,4,9,5,1
-9,14,112,6,3,9,0
-9,15,121,6,0,5,0
-9,16,109,4,0,8,0
-9,17,109,4,0,5,0
-9,18,127,4,0,9,0
-9,19,109,4,9,5,0
-9,20,127,4,0,5,0
-9,21,127,4,0,9,0
-9,22,109,4,9,5,0
-9,23,109,4,9,5,0
-9,24,109,4,3,5,0
-9,25,109,4,3,5,0
-9,26,109,4,3,5,0
-9,27,109,6,3,5,0
-9,28,121,6,3,0,0
-9,29,109,4,3,7,0
-9,30,112,6,3,5,0
-9,31,103,6,3,0,0
-9,32,103,6,2,0,0
-9,33,103,6,2,0,0
-9,34,113,6,2,0,0
-9,35,113,6,2,0,0
-9,36,113,6,2,0,0
-9,37,113,6,2,0,1
+9,8,119,4,9,8,1
+9,9,102,4,9,8,1
+9,10,102,4,9,8,1
+9,11,102,4,9,0,1
+9,12,102,4,9,0,1
+9,13,102,6,9,0,1
+9,14,112,4,9,0,1
+9,15,112,4,9,0,0
+9,16,112,4,3,0,0
+9,17,112,4,3,0,0
+9,18,112,4,3,0,0
+9,19,112,4,3,0,0
+9,20,112,4,3,0,0
+9,21,112,4,9,0,0
+9,22,112,4,3,0,0
+9,23,112,4,3,0,0
+9,24,112,4,3,0,0
+9,25,112,4,3,0,0
+9,26,112,4,3,0,0
+9,27,112,4,3,0,0
+9,28,112,4,3,0,0
+9,29,112,4,3,0,0
+9,30,112,4,3,0,0
+9,31,112,4,3,0,0
+9,32,112,4,3,0,0
+9,33,112,4,3,0,0
+9,34,112,4,3,0,0
+9,35,112,4,3,0,0
+9,36,112,4,3,0,0
+9,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 068b8b9f..e820f648 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.013403586,0.013561583,0.010157157,0.02523759,0.02000361,0.05363709,0.021128021,0.025016747,0.02920473,0.01225187,0.0151882535,0.046147842,0.030095028,0.030087681,0.019925622,0.02262283,0.03150853,0.014046838,0.017826365,0.06419519,0.02504119,0.059139382,0.08537767,0.030591354,0.04970318,0.023754872,0.006926579,0.045255262,0.009768042,0.06675244,0.026345653,0.018249106,0.037849132
-0.016586032,0.009750425,0.008877863,0.044213485,0.015101754,0.030926332,0.10564914,0.038188923,0.021930052,0.012667373,0.019466935,0.031108072,0.05049339,0.026259657,0.033276457,0.033570215,0.030092146,0.008471316,0.018575478,0.017898783,0.018978879,0.06433024,0.053123724,0.01424231,0.08489147,0.020401172,0.012519794,0.03767036,0.009599258,0.0104606785,0.04588504,0.03737721,0.017415995
-0.007843759,0.0123399645,0.010932602,0.012196199,0.010071585,0.041178703,0.048805505,0.02791397,0.021404997,0.01442688,0.014203211,0.060977157,0.024207711,0.034439713,0.07799098,0.035463613,0.07618432,0.0120071145,0.012583348,0.013552799,0.017436162,0.038533,0.023809142,0.010679349,0.07326574,0.052771334,0.016932746,0.037566938,0.030193252,0.02263027,0.02893749,0.023646941,0.054873504
+0.013453301,0.013558816,0.010155084,0.025183218,0.019921565,0.05383606,0.02112372,0.025011655,0.029356034,0.012249369,0.015066982,0.046228647,0.030096248,0.030052193,0.019999536,0.022618225,0.031463686,0.013989219,0.017822728,0.064182125,0.025011655,0.059127342,0.0850275,0.030607535,0.049790215,0.023750037,0.0069251657,0.04524605,0.009766049,0.06673885,0.026443383,0.018245382,0.03795245
+0.016586587,0.009750751,0.00887816,0.044301387,0.015043382,0.030987818,0.10565262,0.038190182,0.021930778,0.0126184095,0.019467577,0.031093914,0.05049505,0.026299018,0.033277556,0.033669822,0.030034423,0.0084716,0.018576091,0.017899383,0.01901661,0.06408156,0.05312548,0.014242788,0.08489428,0.020441733,0.012520213,0.037671603,0.0095995795,0.01046103,0.04588656,0.037451517,0.017382594
+0.007844531,0.012341179,0.0109336795,0.01224514,0.009994192,0.041102406,0.04862002,0.027862249,0.021365335,0.014428301,0.014204611,0.06122185,0.02423375,0.034409486,0.07799866,0.035467107,0.07589478,0.012055296,0.012633842,0.013554134,0.017437879,0.038687624,0.023811487,0.010680402,0.07327296,0.05277653,0.017000694,0.03757064,0.030218352,0.0226325,0.028972154,0.02364927,0.05487891
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 06bccafb..8cfb24ae 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.019326234,0.008987534,0.01819081,0.021602122,0.03681829,0.06312124,0.03617675,0.017055342,0.04511071,0.0098323915,0.015110195,0.03783893,0.03835981,0.024985593,0.032971486,0.02502222,0.035477027,0.018369326,0.027649779,0.01682375,0.034184624,0.014531332,0.031430904,0.032019537,0.049933005,0.109490566,0.01257582,0.058377594,0.009717841,0.035477027,0.039423183,0.011185166,0.0128238555
-0.013355161,0.00794364,0.011924816,0.012399847,0.015252072,0.053862467,0.053862467,0.02285139,0.041217204,0.0128435325,0.011512868,0.051597085,0.037019096,0.019892447,0.055139776,0.030555336,0.03286122,0.013459907,0.0118320165,0.019393725,0.037602063,0.035810128,0.028006654,0.01410586,0.10880518,0.02236568,0.01507438,0.038644433,0.01388717,0.016685817,0.053027406,0.057112765,0.030096311
-0.015875405,0.0123637775,0.019149387,0.021762764,0.015875405,0.039832875,0.029979475,0.041419636,0.06085663,0.01757247,0.011569412,0.094256595,0.044872865,0.018487863,0.052461695,0.019603502,0.055845182,0.022963623,0.021300191,0.0145680895,0.028087579,0.031850714,0.013212684,0.019527076,0.04851914,0.05921513,0.014739814,0.050255228,0.023121139,0.019951142,0.019951142,0.019527076,0.021425363
+0.019366598,0.008988731,0.01826444,0.021583911,0.036751345,0.063376725,0.036252305,0.017057614,0.04511672,0.009872189,0.015112207,0.037917957,0.03821535,0.024958435,0.03297588,0.025031663,0.03541252,0.018371774,0.027673723,0.016891846,0.03425602,0.014533268,0.03134313,0.03208641,0.050135113,0.10865299,0.012577496,0.058385372,0.009719135,0.03555112,0.03950552,0.011186656,0.012875763
+0.013347994,0.007877592,0.011918416,0.012393192,0.015243887,0.053833574,0.053833574,0.022816842,0.04135633,0.01283664,0.01146183,0.051569402,0.037071574,0.019804265,0.054895345,0.030598652,0.032779507,0.013452683,0.011825667,0.01938332,0.03758189,0.035721082,0.027969431,0.01409829,0.10959972,0.022375524,0.01506629,0.0386237,0.013825605,0.016676864,0.05299896,0.057082128,0.030080168
+0.015875753,0.01241244,0.019149808,0.021741997,0.01581386,0.039911628,0.030156313,0.041339725,0.060857967,0.017572856,0.011569666,0.094258666,0.045049477,0.01848827,0.052462846,0.019565681,0.055846408,0.022941712,0.021279868,0.01456841,0.028019704,0.031789266,0.013212974,0.019489402,0.0485202,0.059216425,0.014740137,0.050256327,0.02307653,0.019912649,0.020029668,0.019489402,0.021384027
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index b9936480..1f2ed7bf 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012297478,0.013665343,0.018972449,0.011826368,0.024950994,0.078946404,0.037073772,0.020235552,0.029241972,0.015067184,0.017892724,0.023132304,0.015667394,0.04334363,0.0390813,0.022333188,0.037584122,0.00976618,0.017546648,0.0258028,0.034895677,0.030525848,0.022267856,0.016710458,0.13534562,0.093021594,0.011826368,0.024006857,0.01638725,0.025976619,0.04160183,0.016841521,0.01616473
-0.011090748,0.008306586,0.0069948467,0.009486419,0.01348292,0.08195016,0.0491262,0.049899824,0.026644321,0.007049708,0.012324387,0.03201389,0.02492027,0.049318474,0.041046478,0.048176013,0.046149794,0.00805102,0.012421048,0.019088384,0.0321235,0.02361709,0.03528076,0.020478847,0.047244202,0.053118087,0.01741415,0.06137783,0.010177442,0.028612677,0.08655661,0.010257265,0.016200101
-0.015800804,0.01356804,0.009696641,0.008967927,0.01755835,0.065492526,0.037830207,0.038803037,0.030613404,0.0072624483,0.013835645,0.025572212,0.017904656,0.035158474,0.062493403,0.09418089,0.030650796,0.0078525795,0.018222168,0.018400995,0.032899566,0.02009147,0.021096846,0.026204217,0.08847476,0.029853182,0.01636618,0.04572115,0.013781705,0.039645717,0.059865013,0.016238818,0.01989622
+0.012260135,0.01367717,0.018951818,0.011836603,0.024954304,0.079014726,0.037033457,0.020253064,0.029310184,0.015080225,0.017873267,0.023197588,0.015742326,0.043296497,0.039191596,0.022287127,0.037763875,0.009736525,0.017561834,0.025771594,0.034925878,0.030522447,0.022330698,0.016692288,0.13546278,0.09237757,0.011836603,0.024109898,0.01633749,0.026002275,0.041637834,0.016823206,0.016147152
+0.011046612,0.008305911,0.0069942786,0.009485649,0.013481825,0.08194348,0.048930686,0.050091043,0.026681205,0.0070491354,0.012323385,0.032058205,0.024869617,0.049314454,0.040963046,0.048172083,0.04605599,0.008050365,0.012420039,0.019124145,0.031980034,0.023615165,0.035277884,0.020437222,0.04724035,0.053113755,0.017480887,0.061613034,0.010176616,0.028699104,0.086549565,0.010256432,0.016198784
+0.015744025,0.013519284,0.00962413,0.008970675,0.017563729,0.06551258,0.0379899,0.038890805,0.030600358,0.007264674,0.013839886,0.02553013,0.017910143,0.035237998,0.062268827,0.09420972,0.030697633,0.007854985,0.018227752,0.01840663,0.032877516,0.020058408,0.021103306,0.026186656,0.08850185,0.0299061,0.016371196,0.045735154,0.0137859285,0.039580476,0.059883345,0.016243795,0.019902313
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 48bf280b..ab6b8364 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.014360831,0.018260473,0.018224843,0.011053926,0.03076024,0.07321574,0.037321977,0.01873005,0.036528688,0.015227427,0.015527761,0.02097666,0.020291606,0.022482706,0.037103932,0.02936596,0.033982046,0.013024708,0.019821553,0.028408663,0.050124213,0.025789138,0.018803358,0.019362386,0.14334947,0.056798175,0.010143653,0.031520464,0.013229817,0.017977372,0.050320394,0.014586981,0.033324774
-0.01811298,0.010022493,0.010627294,0.029121654,0.02044467,0.07640756,0.046524912,0.04837825,0.029228497,0.015523141,0.01671908,0.021955362,0.12305657,0.05190229,0.027959907,0.030796167,0.022829963,0.0101011,0.016204689,0.027190795,0.0332174,0.017317316,0.02210596,0.016141513,0.05655991,0.03148033,0.012669618,0.03716541,0.011312696,0.025932973,0.047257572,0.02219248,0.013539525
-0.0188605,0.014545823,0.011733642,0.027108917,0.03445344,0.08535639,0.038024917,0.02668049,0.023018274,0.0188605,0.02539817,0.013825609,0.023982013,0.11485958,0.021226147,0.043937705,0.021122757,0.019535357,0.01927009,0.024431145,0.02162367,0.017409071,0.021350885,0.01799681,0.02659432,0.069665864,0.020835934,0.06621647,0.014320311,0.03824837,0.020714207,0.016482577,0.022310078
-0.016216978,0.013081708,0.014939779,0.024982793,0.017741427,0.057494164,0.03569894,0.043145325,0.033766083,0.014881535,0.010308138,0.02417871,0.0764655,0.026581064,0.04486404,0.018091345,0.031891134,0.012629794,0.014939779,0.023734218,0.031922292,0.033275068,0.015534913,0.023116592,0.07858521,0.06695543,0.014311432,0.06540441,0.016028045,0.02678954,0.036832146,0.011410069,0.024202334
-0.011195347,0.0092090415,0.0072849514,0.026208512,0.011416156,0.057524852,0.029870834,0.03149794,0.026517449,0.009281268,0.01366335,0.078934886,0.03681583,0.030525122,0.050370444,0.020671988,0.037873372,0.00787691,0.01097881,0.042916153,0.016871985,0.061715156,0.057300583,0.019230815,0.08402575,0.0281726,0.0070058694,0.04515177,0.0079386905,0.060521476,0.021453522,0.017407559,0.022571096
-0.01740354,0.01922637,0.01123657,0.03960435,0.018671269,0.0817404,0.03641405,0.032466494,0.028904509,0.014204355,0.018381797,0.028214265,0.03581451,0.036342997,0.021490498,0.035640057,0.050952412,0.011684184,0.033546194,0.023464803,0.024068216,0.015661491,0.026706208,0.0505559,0.078608975,0.03567488,0.005650102,0.044790044,0.014204355,0.04054354,0.028234936,0.018238753,0.02165905
+0.01434968,0.018246295,0.018246295,0.011062379,0.030698866,0.073096424,0.03732033,0.018661758,0.03649588,0.015284481,0.015576431,0.020900974,0.020248648,0.022526698,0.037143078,0.029393356,0.03404698,0.013014595,0.01981463,0.028412612,0.050005887,0.025869979,0.0187521,0.019310784,0.14371982,0.056729846,0.010097493,0.031495996,0.013216318,0.018000763,0.05034006,0.014573876,0.0333467
+0.01807688,0.009905313,0.01062685,0.029172484,0.020458804,0.07633914,0.046557073,0.04852078,0.029202323,0.015531968,0.016718382,0.021956462,0.1231266,0.051788602,0.027899088,0.03077986,0.022778912,0.0100612985,0.016242629,0.027246153,0.033163358,0.017325047,0.022083469,0.016141823,0.05652651,0.031496324,0.012720227,0.037182026,0.011353725,0.025973134,0.047310814,0.022188853,0.013545157
+0.018902939,0.0145501075,0.01169134,0.027147124,0.034455173,0.08530861,0.038063984,0.026669418,0.023067255,0.01891448,0.025368463,0.013871527,0.023945192,0.11464649,0.021188384,0.043950647,0.02112382,0.019579314,0.019284002,0.02452501,0.021680264,0.017422704,0.021336326,0.018038407,0.026665552,0.06965661,0.020803943,0.06600996,0.014377083,0.038264383,0.020744508,0.01648542,0.022261553
+0.016216656,0.01308145,0.014939484,0.025020843,0.017754074,0.057443947,0.035689533,0.043020573,0.033728357,0.014948606,0.010307934,0.024133276,0.07651071,0.026627235,0.044945393,0.018055689,0.031882733,0.012678976,0.014916706,0.023783064,0.031886622,0.03319329,0.015595406,0.023117555,0.07884793,0.06692554,0.014312897,0.06517998,0.01608653,0.026792346,0.0368025,0.01140845,0.024165705
+0.011208817,0.009184176,0.007265281,0.026254857,0.011393673,0.05732051,0.029912245,0.031399224,0.026623994,0.00926186,0.013679789,0.07865209,0.036882624,0.030619256,0.050721206,0.02077786,0.037872672,0.0077945087,0.010975259,0.04301501,0.0168655,0.06145841,0.05736951,0.019180054,0.083752826,0.028208211,0.006933426,0.045228165,0.007977403,0.060601793,0.02146246,0.017494582,0.022652796
+0.017393991,0.019215824,0.0112304045,0.03964368,0.018711211,0.08226599,0.036385193,0.03245042,0.02884637,0.014205229,0.018300092,0.028105145,0.035746835,0.036245,0.021434186,0.035585742,0.0507631,0.011632246,0.033574894,0.023546595,0.024087338,0.01569116,0.026756804,0.050531257,0.07852272,0.03570976,0.005625673,0.044787344,0.014193096,0.04060556,0.028245514,0.018262153,0.021699423
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 5bf42858..934e0e29 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010741137,0.008597097,0.008732483,0.021761348,0.010615999,0.06205288,0.03822967,0.056279518,0.025933884,0.009895199,0.010699261,0.033606037,0.021071235,0.029112015,0.10072176,0.022342766,0.12244653,0.008139568,0.015147431,0.010699261,0.015325984,0.020106312,0.028507149,0.015266233,0.0655409,0.02857683,0.007127244,0.026108587,0.010492319,0.10311031,0.022039376,0.016799457,0.014174217
-0.009333175,0.014287093,0.009333175,0.03983506,0.0151492385,0.035707913,0.03427296,0.105981335,0.024172984,0.012510196,0.010052225,0.06503865,0.027972776,0.027532374,0.029922616,0.0156913,0.18455555,0.009046024,0.017920028,0.020828241,0.013008546,0.023856388,0.02995185,0.012364448,0.035987977,0.02517892,0.00704505,0.020148035,0.016284535,0.030334523,0.033939894,0.018131264,0.024625693
-0.010948875,0.0068784477,0.007794302,0.033332005,0.011385028,0.014280026,0.04296665,0.029207861,0.02900534,0.012024987,0.01593052,0.041807696,0.07056385,0.026035152,0.04887815,0.056921564,0.06475303,0.008232424,0.014906994,0.010086576,0.017841335,0.04450406,0.061787773,0.01656512,0.09203261,0.017360097,0.0076137474,0.04060063,0.013258555,0.019253442,0.023659023,0.06759603,0.02198812
+0.010731453,0.008589347,0.008724609,0.021720512,0.01064794,0.062239602,0.038269885,0.05600958,0.025929496,0.009886278,0.010731453,0.033510234,0.021114009,0.029029025,0.101420246,0.022333529,0.12233617,0.0081322305,0.015133775,0.010731453,0.015312167,0.020029424,0.028453656,0.015312167,0.065481834,0.028537137,0.0071766684,0.026116919,0.010441991,0.10301738,0.022019511,0.016718876,0.01416144
+0.009332455,0.014341904,0.009259829,0.039909847,0.015148069,0.035705153,0.03433731,0.10597313,0.02412395,0.01255819,0.01005145,0.06503362,0.027943308,0.02745641,0.02977456,0.01569009,0.18454129,0.009045325,0.017988779,0.02082663,0.013007542,0.023807997,0.029949531,0.012363493,0.03598519,0.025091063,0.0070445063,0.020107165,0.01634701,0.030421168,0.033937268,0.018200824,0.024696033
+0.0109061785,0.0068784403,0.0077942936,0.033202026,0.011385017,0.01428001,0.043134782,0.029179327,0.029015938,0.011978094,0.015992854,0.041807663,0.070288695,0.025984332,0.04887811,0.056921516,0.06500641,0.008232416,0.014906977,0.010086566,0.017806504,0.044417188,0.06178772,0.016565101,0.09203254,0.017326204,0.007613739,0.040600598,0.01325854,0.019178364,0.023705257,0.06786054,0.021988103
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 15dcf397..1593701e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.011449223,0.008812782,0.008951563,0.014304557,0.012673135,0.088313274,0.04466766,0.040118113,0.019976614,0.008214416,0.014027881,0.034923363,0.0200548,0.07524385,0.032968048,0.04163484,0.031914745,0.008343774,0.014360543,0.024238,0.029458722,0.049249835,0.036670923,0.027613169,0.055049576,0.03465159,0.015466914,0.052528672,0.012428015,0.048962105,0.049927797,0.016272627,0.016528884
-0.01242395,0.008741341,0.012918863,0.0126689905,0.012136149,0.03917598,0.026045907,0.02556709,0.024926098,0.010878774,0.011535247,0.05655713,0.019660482,0.029255617,0.03887111,0.02514614,0.0321624,0.010836362,0.015983855,0.023936106,0.017657993,0.2894735,0.014244129,0.0121836485,0.06459027,0.01704799,0.01135641,0.026951538,0.012918863,0.01786614,0.021508634,0.019737432,0.025035877
-0.019095432,0.015463947,0.0148135545,0.016017271,0.02569571,0.03876264,0.05160351,0.019207647,0.036166057,0.019740103,0.031347863,0.026176937,0.057680424,0.025998604,0.03910483,0.03476358,0.042780813,0.01607996,0.028070044,0.03522492,0.029742124,0.04114191,0.026955184,0.01176435,0.06334955,0.042365067,0.016655328,0.029546713,0.022676386,0.014080084,0.04580756,0.034661878,0.027459998
-0.012427352,0.012427352,0.011812032,0.044231243,0.012922402,0.08660057,0.032709766,0.07465436,0.020609653,0.0156484,0.016271763,0.031952046,0.09075662,0.0546183,0.03434656,0.031395294,0.053145062,0.010713048,0.016208325,0.019398829,0.014700311,0.02252502,0.012722059,0.013384786,0.06958552,0.03223411,0.009985658,0.03335491,0.01854668,0.042703256,0.015285905,0.013863714,0.018259136
-0.0175087,0.024107471,0.015511848,0.03606612,0.04308127,0.057521038,0.0672489,0.017136548,0.023082271,0.018029237,0.017170051,0.03007557,0.023931548,0.03831727,0.02473345,0.03831727,0.036348987,0.017338548,0.031151721,0.015941853,0.018348955,0.019839952,0.018894475,0.0121516315,0.16644265,0.023990046,0.012784643,0.04135001,0.018931413,0.033748895,0.0147726275,0.013139047,0.012985972
-0.011374115,0.021458177,0.012639273,0.019884368,0.034073096,0.020920198,0.021291187,0.017754741,0.037733898,0.014950992,0.014661813,0.07216797,0.019923242,0.028580561,0.030113481,0.052593373,0.055985354,0.019652708,0.01661401,0.03874207,0.025682075,0.045249857,0.06596687,0.016484719,0.052593373,0.02253194,0.01215507,0.030408999,0.026111996,0.032960337,0.027338427,0.022269435,0.05913232
+0.011400348,0.008775162,0.008843986,0.014321295,0.012677706,0.08820507,0.044946373,0.040066205,0.019908346,0.008184344,0.014077552,0.03491359,0.020098807,0.075348005,0.033000086,0.041619375,0.031996496,0.008308156,0.014333318,0.024208302,0.029419037,0.049255587,0.036585778,0.027604597,0.05521417,0.034725506,0.015523573,0.052534807,0.012420362,0.048949987,0.04967562,0.016328253,0.016530307
+0.012443908,0.008755383,0.012939616,0.012758654,0.012164551,0.0392054,0.02605592,0.025547082,0.024999686,0.010902902,0.011553777,0.05659787,0.019704089,0.029325426,0.03885284,0.025211142,0.03223767,0.010853769,0.016016373,0.023959927,0.01762386,0.28838542,0.014211388,0.012203965,0.06491157,0.017034778,0.0113316905,0.026981654,0.012987089,0.01789706,0.021515755,0.019766726,0.025063088
+0.01905886,0.015464511,0.014814094,0.01604256,0.025715468,0.038806655,0.051542424,0.019265728,0.03616295,0.019714328,0.03132796,0.026078215,0.05771773,0.025969015,0.039216194,0.0347988,0.042813707,0.016017856,0.028096773,0.035178926,0.029812263,0.0413246,0.026982497,0.011719628,0.06307021,0.0423485,0.016723165,0.029562214,0.02267167,0.01413746,0.04577323,0.034625072,0.027446749
+0.01243457,0.01243457,0.011818892,0.044152383,0.012939381,0.08657686,0.03272077,0.07462879,0.02063925,0.0156059675,0.016217737,0.03191114,0.09086477,0.054746,0.03436231,0.031505693,0.05321489,0.010677479,0.01619301,0.019355675,0.0146282725,0.0225271,0.012729448,0.013341161,0.069316395,0.03227055,0.009992678,0.033357985,0.018552922,0.042900607,0.015252949,0.013815998,0.01831384
+0.017431313,0.024062537,0.015443287,0.035962105,0.04309029,0.05766668,0.06726299,0.017111776,0.023123773,0.01796051,0.017127581,0.0300626,0.02393364,0.038289644,0.024756758,0.03822251,0.036356602,0.01729566,0.031027302,0.015982209,0.018346079,0.019839263,0.018921515,0.012098662,0.166915,0.024002396,0.012779513,0.041348573,0.01895388,0.033669606,0.0148399975,0.013130568,0.01298512
+0.011333878,0.02142407,0.01259456,0.019844584,0.034227207,0.020909905,0.02135618,0.01781424,0.037632555,0.015024115,0.014667127,0.07213027,0.01998162,0.028571296,0.03012072,0.052715294,0.05604668,0.019621471,0.016529996,0.038723007,0.025725903,0.045199998,0.065990776,0.016491702,0.05237854,0.02253048,0.012113549,0.030353254,0.026191704,0.032936145,0.027407028,0.022231326,0.059180837
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 891c35c6..99f876c0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.016557448,0.020405825,0.017625311,0.025519764,0.03680612,0.06548598,0.03477934,0.021892255,0.029666752,0.022543117,0.019700896,0.034173325,0.027245352,0.046527296,0.022107093,0.042529393,0.057565864,0.019357618,0.046165217,0.027687918,0.035638895,0.025644677,0.032544788,0.017625311,0.044396657,0.027566511,0.015253436,0.038048714,0.016557448,0.025694814,0.04571658,0.022499131,0.018471168
-0.012519716,0.008405341,0.010099248,0.021738015,0.018539116,0.04669845,0.04799298,0.030121356,0.03165953,0.0144665055,0.012816613,0.040493153,0.034098655,0.024898428,0.05459602,0.026919158,0.040335283,0.007896087,0.016138554,0.012967691,0.025155049,0.032252446,0.047619496,0.0095991995,0.2101071,0.025044747,0.012917135,0.04153446,0.0099039115,0.009017614,0.028981574,0.019657847,0.014809568
-0.009015803,0.0061482433,0.008469563,0.021126784,0.009229606,0.037883893,0.05087888,0.031684104,0.028091125,0.010216243,0.015396422,0.050680522,0.056981638,0.026749171,0.074900955,0.0725965,0.05147862,0.007532985,0.011897433,0.010136739,0.022011328,0.039779596,0.028429195,0.019273764,0.093946844,0.02702198,0.007894501,0.032245975,0.011531387,0.026202993,0.028933344,0.053738806,0.01789508
+0.016554631,0.020482207,0.017691284,0.025540352,0.036871806,0.06521957,0.034739483,0.021931322,0.029562298,0.022627497,0.019697545,0.034100845,0.027227417,0.046701454,0.022060204,0.042522155,0.05755607,0.019354325,0.0462476,0.027696729,0.03559805,0.025690442,0.032507494,0.01755361,0.04430249,0.027561821,0.015310531,0.037968013,0.016554631,0.025690442,0.045798164,0.022539282,0.018540308
+0.012518779,0.008404711,0.010098492,0.021736387,0.01857397,0.046512906,0.047989387,0.030192723,0.031626258,0.014409027,0.012815653,0.04049012,0.034029573,0.024860121,0.054591935,0.026916321,0.040332265,0.007895496,0.016200505,0.01296672,0.025196189,0.03225003,0.04780229,0.0095984815,0.21009137,0.025030646,0.012916167,0.041450314,0.009903169,0.0090169385,0.029021885,0.019694805,0.014866418
+0.00902418,0.0061539556,0.008477432,0.02116707,0.009238182,0.037993215,0.050727595,0.031682577,0.028151562,0.010265757,0.015410727,0.050727595,0.057034567,0.026796907,0.07497053,0.072663926,0.05152644,0.007539984,0.0119084865,0.010146158,0.022053301,0.039661318,0.028497316,0.019291667,0.09330233,0.027031815,0.007901835,0.03230746,0.011542101,0.026227333,0.028946083,0.05378872,0.017841876
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 44487bad..3381365f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012167323,0.016182173,0.019519432,0.031962857,0.019672524,0.042634357,0.030798472,0.034057528,0.04888053,0.015380938,0.010864196,0.07139911,0.027352633,0.018229676,0.069473244,0.016793964,0.05759534,0.015024639,0.014849597,0.022335472,0.023510464,0.06300949,0.016533598,0.019904418,0.051830195,0.050432164,0.013415492,0.05670241,0.024244925,0.03049917,0.019481344,0.02006053,0.015201745
-0.0109462105,0.008198333,0.012648323,0.023094064,0.010776506,0.070822805,0.049057525,0.03391485,0.023337783,0.018674767,0.014758534,0.046446733,0.017254474,0.044146996,0.03610217,0.023224108,0.042788733,0.0104449475,0.0142208915,0.019532802,0.0133592915,0.18805672,0.035611942,0.007946097,0.052837033,0.015406512,0.020093884,0.019705232,0.015497048,0.023872638,0.02404923,0.03451628,0.018656539
-0.01147871,0.0120766675,0.00944213,0.024731357,0.017097488,0.060144126,0.1535834,0.041579332,0.018271415,0.013791987,0.019640686,0.02893507,0.023210282,0.035703883,0.024359824,0.055624224,0.022061061,0.0103297215,0.013631306,0.018271415,0.019640686,0.040378857,0.03815534,0.013057991,0.04629459,0.02357579,0.027908226,0.02653935,0.01587448,0.012411445,0.06579788,0.036837246,0.019564113
+0.012163671,0.016177313,0.019475494,0.03201573,0.019628244,0.042621553,0.03087956,0.0340473,0.048865855,0.015436501,0.010860935,0.07137767,0.027397878,0.018224202,0.06918161,0.016821744,0.05757805,0.015020127,0.0148451375,0.02235058,0.02353786,0.062990576,0.016560948,0.019937342,0.05201743,0.050417025,0.013411464,0.056685384,0.024237646,0.030490013,0.01951357,0.020093713,0.015137932
+0.010944551,0.00819709,0.012646405,0.023028629,0.010774871,0.070812054,0.049050074,0.0339097,0.023305774,0.018763326,0.014756296,0.046439677,0.017285585,0.04431305,0.03609669,0.0231781,0.042782236,0.010443363,0.0142187355,0.019548915,0.013357266,0.18802817,0.035606537,0.007944892,0.05282901,0.015404175,0.020090831,0.019663798,0.015524992,0.023804635,0.024000125,0.03457851,0.018671932
+0.011467976,0.012065373,0.009433299,0.024672067,0.0170815,0.060087897,0.15464325,0.041540455,0.018254327,0.013779089,0.019622322,0.028830487,0.023233917,0.035531435,0.024444234,0.05557222,0.022083525,0.010320061,0.013618558,0.018254327,0.019622322,0.040341105,0.03819419,0.012994919,0.046251304,0.023599796,0.027953701,0.026456343,0.015859634,0.012399838,0.065480076,0.036802802,0.019507684
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index bb8ba9c4..b0c6cbe3 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.015025156,0.019444002,0.016373487,0.078113556,0.1282854,0.08043589,0.11321145,0.07098442,0.1760316,0.1118925,0.11634979,0.035484683,0.038368087
-0.012999591,0.01708764,0.012599636,0.06817954,0.07748421,0.039885454,0.0934638,0.2801235,0.10305166,0.07075717,0.101851076,0.03280888,0.089707874
-0.025136897,0.034358066,0.031900384,0.042926658,0.059018586,0.05193135,0.06720416,0.11265576,0.05965599,0.23942567,0.0725941,0.02938801,0.1738044
+0.015029022,0.019297646,0.0163777,0.07805737,0.12882058,0.08037804,0.113240555,0.07114149,0.17607687,0.111921266,0.11592598,0.035355426,0.03837795
+0.0130002005,0.017088441,0.012600226,0.06811618,0.077639334,0.03980949,0.0932858,0.28013662,0.10305648,0.07076048,0.101855844,0.032938834,0.089712076
+0.0252677,0.03453685,0.031941365,0.04289794,0.05886402,0.051947314,0.0674056,0.11280049,0.05979099,0.23786764,0.072545536,0.029425764,0.17470881
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 8a8365f8..4f7e8351 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034291983,0.032466996,0.029909998,0.16106153,0.04377453,0.09321527,0.11443302,0.09523961,0.023753362,0.10792062,0.08903354,0.067189634,0.10771004
-0.020611567,0.017356684,0.01605231,0.09023478,0.0966189,0.037395693,0.10224941,0.094196565,0.03358682,0.112299,0.067714855,0.04229219,0.2693913
-0.03479445,0.054101624,0.031929184,0.06602827,0.057086926,0.08603304,0.10489644,0.034523677,0.022729797,0.13338077,0.05976819,0.033461496,0.2812662
+0.034147155,0.032329876,0.029900245,0.1616392,0.043674875,0.09300306,0.11439571,0.095208555,0.023745617,0.10788543,0.08917852,0.06721694,0.10767492
+0.020598466,0.017345646,0.016042102,0.09035372,0.096557476,0.037299003,0.10258434,0.094136685,0.03356547,0.11222761,0.06780411,0.042265307,0.26922008
+0.034789838,0.054094452,0.03192495,0.06606789,0.05707936,0.08610568,0.10488253,0.034519102,0.022726784,0.13336308,0.059760265,0.033457063,0.281229
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 27982a02..8aaf1681 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.017545573,0.016227003,0.028927766,0.17582504,0.035030115,0.05548813,0.057663318,0.06000796,0.07401406,0.27232316,0.06309156,0.09522173,0.0486345
-0.014570021,0.015269251,0.02170198,0.118239954,0.02797496,0.09802437,0.12247076,0.116862416,0.06581848,0.12247076,0.1319062,0.06262835,0.08206254
-0.012689575,0.010939098,0.021585695,0.15763971,0.014379173,0.062643535,0.1632803,0.1765481,0.10177984,0.09413096,0.057163052,0.084543414,0.042677484
+0.017511223,0.016195234,0.028871126,0.17548077,0.035098363,0.055460665,0.057403073,0.059726194,0.073725,0.2739216,0.06293729,0.095035285,0.048634168
+0.014574757,0.015274214,0.021879302,0.11781726,0.028093578,0.09824792,0.12251057,0.1169004,0.0659042,0.12251057,0.13143465,0.06260283,0.08224971
+0.012698688,0.010946953,0.021601195,0.15775292,0.014389498,0.06268852,0.16339757,0.17667489,0.10145584,0.09401476,0.057232037,0.08443905,0.04270813
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 556d2b62..55d4c245 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.019172488,0.01932286,0.023036273,0.20292799,0.03028061,0.07068008,0.0837709,0.042287517,0.07884933,0.21601573,0.054938298,0.11836687,0.040351033
-0.013898927,0.011343973,0.015028323,0.16800593,0.04053331,0.06949026,0.26021266,0.13033302,0.052788023,0.07113817,0.049856827,0.079515524,0.037855063
-0.033207193,0.02873843,0.053064942,0.20104882,0.02988324,0.13392739,0.11727074,0.17263922,0.043395016,0.05972081,0.047289208,0.053953256,0.025861789
-0.01586071,0.017016059,0.019585362,0.06508605,0.037604593,0.13971801,0.16591924,0.055089667,0.04819104,0.16919172,0.10221989,0.07776479,0.086752884
-0.011261298,0.013583719,0.013583719,0.044976395,0.14044411,0.06680432,0.15245065,0.07024145,0.10684425,0.12988958,0.116887964,0.023654649,0.10937799
-0.018765636,0.029638039,0.04139015,0.19101273,0.034113172,0.11185279,0.07816212,0.1505144,0.065538496,0.0822338,0.07793347,0.09948393,0.01936132
+0.019247921,0.019242048,0.023122676,0.2023879,0.030257503,0.070795946,0.083553046,0.04229355,0.07910523,0.21612567,0.054946132,0.118629664,0.040292792
+0.013859086,0.011220005,0.014865906,0.16772893,0.04046441,0.06960686,0.26081634,0.1302294,0.052909687,0.071038246,0.049805023,0.07961735,0.03783882
+0.03323299,0.028751979,0.053304255,0.20066538,0.029766515,0.13464227,0.11751235,0.17178494,0.043314233,0.059854817,0.047320165,0.053955637,0.025894517
+0.01579723,0.016942782,0.019503407,0.06499989,0.03757122,0.13979305,0.16676494,0.055090666,0.048058998,0.16925676,0.101997375,0.077775694,0.086447895
+0.011209881,0.013517572,0.013519222,0.04491335,0.14133994,0.06666387,0.15254389,0.070202954,0.10669937,0.12986594,0.11656749,0.023598434,0.10935808
+0.01869931,0.029639825,0.041397702,0.19057001,0.0340989,0.11240363,0.078290984,0.15029393,0.0654575,0.08262617,0.078181155,0.09896303,0.01937786
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index bc78ed18..f0ffa839 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.017322686,0.016020864,0.0268299,0.08280347,0.045728423,0.046267454,0.111206844,0.21773246,0.03754178,0.22116125,0.05969938,0.06348763,0.05419789
-0.02507379,0.024976037,0.03481151,0.056836713,0.074419186,0.054873265,0.049670782,0.18239927,0.070286855,0.1535956,0.16607644,0.048425484,0.058555067
-0.008939866,0.014064383,0.009591076,0.05823833,0.09724553,0.03360716,0.19875696,0.2737993,0.077986516,0.10722134,0.04580116,0.039870534,0.03487773
+0.0173287,0.016026428,0.026839208,0.0828322,0.04565503,0.046283506,0.11124542,0.21780796,0.037408393,0.22123794,0.0596618,0.06350966,0.05416377
+0.025201809,0.025005685,0.034852836,0.05695978,0.0744348,0.05504581,0.049729746,0.18261582,0.07040466,0.15317841,0.16562536,0.048577756,0.058367517
+0.008935493,0.014057503,0.0095863845,0.058039587,0.09719801,0.03352519,0.19865984,0.2736655,0.07825349,0.107588395,0.04577878,0.03985105,0.034860685
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 466b8da1..35299c71 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.013754733,0.015586155,0.022677721,0.12847804,0.03051621,0.06587699,0.08986748,0.15864919,0.103033565,0.09547677,0.12501256,0.049436044,0.10163449
-0.013943452,0.025645932,0.013620452,0.05140305,0.07930422,0.041708842,0.07060354,0.10202771,0.17053087,0.10506101,0.07977025,0.044659745,0.2017209
-0.018764257,0.02466515,0.015314939,0.057096727,0.06734212,0.033450905,0.075309195,0.19099866,0.17664489,0.12380064,0.10006119,0.09039775,0.026153553
-0.014145564,0.020581674,0.029252458,0.070930876,0.034199588,0.10051818,0.2977565,0.12706688,0.046110548,0.08801666,0.046020575,0.077826284,0.047574252
-0.026296157,0.044383164,0.062163893,0.15014632,0.03826068,0.13198708,0.09405097,0.09818031,0.03994053,0.08800825,0.07123646,0.09924066,0.05610558
-0.034638647,0.031049881,0.038869288,0.061015755,0.14413515,0.06952924,0.05955869,0.10260714,0.18291703,0.14987685,0.054070164,0.04865788,0.023074314
+0.013696977,0.015395226,0.022755446,0.12909949,0.0304831,0.06601206,0.0899556,0.15893084,0.10253051,0.095681295,0.12471578,0.049288597,0.10145512
+0.013873818,0.025609914,0.01354995,0.05140924,0.07923226,0.041689683,0.07039259,0.102127224,0.17122677,0.10499674,0.0796728,0.044621523,0.2015975
+0.018696813,0.024568997,0.015257101,0.05712812,0.06721138,0.033482578,0.07528165,0.19145387,0.17692275,0.123415925,0.09988427,0.09062429,0.026072279
+0.0140934745,0.020499628,0.02925345,0.07079055,0.03404695,0.10040815,0.29820317,0.12686196,0.046179004,0.08807876,0.046114925,0.07786219,0.04760782
+0.026012044,0.044062022,0.06208434,0.15045841,0.038114253,0.13218476,0.09425221,0.09837236,0.040025376,0.08812661,0.07108019,0.09915601,0.05607141
+0.034625676,0.03115023,0.03877182,0.061037596,0.14396825,0.069717765,0.05949643,0.10258124,0.18308024,0.14989388,0.05404332,0.048556603,0.023076938
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index a0e4ca20..2446af67 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034407627,0.052876856,0.04070077,0.09763607,0.057229374,0.09744556,0.06497616,0.10934721,0.109133855,0.04522797,0.11106918,0.13876884,0.041180536
-0.0108207315,0.016371245,0.015992004,0.12677342,0.058041103,0.041157305,0.12335392,0.1565443,0.07547841,0.14253521,0.076667026,0.06862328,0.08764204
-0.010323195,0.013255244,0.013783273,0.09987551,0.08344912,0.057697766,0.24241307,0.14760642,0.05647656,0.16022526,0.02466751,0.033980932,0.056246158
+0.034523316,0.053054646,0.040678408,0.0977732,0.05697494,0.09720199,0.06494046,0.109287135,0.109287135,0.045203123,0.11122518,0.13869257,0.041157912
+0.010823801,0.016375888,0.015996542,0.12631503,0.058085933,0.04108866,0.12338894,0.15658872,0.07549984,0.14257565,0.0769138,0.068508826,0.087838314
+0.010226722,0.013234361,0.013869491,0.09971816,0.08315507,0.05762093,0.24392943,0.14737387,0.056230776,0.15997283,0.024628649,0.033927396,0.056112282
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 3724cf36..99292a49 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.027727641,0.038949795,0.032799017,0.049429864,0.06950336,0.087946326,0.038047526,0.030811826,0.03009807,0.18509464,0.07083959,0.02142626,0.31732604
-0.020138782,0.02752645,0.021105262,0.05810302,0.11679962,0.026471928,0.03851643,0.2511587,0.10718136,0.10187445,0.08782126,0.035069607,0.10823318
-0.023077397,0.02427956,0.028720284,0.102021314,0.054821804,0.052516073,0.06696441,0.23307347,0.076410435,0.08334873,0.11628466,0.028385684,0.110096104
+0.027722225,0.038942188,0.03279261,0.04932378,0.06952372,0.08810105,0.038040094,0.030926377,0.03009219,0.18505847,0.07079118,0.021422075,0.31726402
+0.020294143,0.02752294,0.02110257,0.05809561,0.11678473,0.026572147,0.038662247,0.25112665,0.10716769,0.101861455,0.08781006,0.035202377,0.10779747
+0.023249542,0.02427032,0.028709352,0.10198248,0.054908074,0.052547373,0.06692258,0.23298474,0.07608357,0.08356145,0.1162404,0.028485935,0.110054195
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index 801421f5..a90a0f9c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.011098017,0.012674333,0.016274165,0.025602855,0.10047324,0.047448613,0.043594476,0.029584043,0.11252469,0.14505006,0.39121857,0.01773458,0.046722345
-0.013920534,0.0164024,0.022595258,0.06077886,0.0823278,0.060823392,0.09757592,0.069157906,0.16761106,0.104888335,0.100870125,0.14114267,0.061905783
-0.019095892,0.028002275,0.021808194,0.20934938,0.12697682,0.17906602,0.07730733,0.07637869,0.052111182,0.04527498,0.026305703,0.106716566,0.031607028
+0.011096026,0.012672059,0.016271247,0.02559827,0.100848414,0.047379345,0.043607958,0.02952103,0.112504534,0.14502408,0.3911485,0.017731398,0.046597134
+0.013814418,0.016405025,0.022598883,0.060714453,0.082501985,0.0610787,0.097782366,0.06913524,0.16763797,0.10449618,0.10068946,0.14116533,0.061979994
+0.019121636,0.028040025,0.021837592,0.20881435,0.12739657,0.17860837,0.07736431,0.07641633,0.052181434,0.045159265,0.026341166,0.10706935,0.031649638
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 64bbe388..c5539a6d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.028366636,0.033554886,0.04283361,0.06522581,0.20635305,0.08057263,0.12913246,0.036709156,0.08909872,0.09540245,0.027493887,0.064884335,0.100372225
-0.015950827,0.027133264,0.032474265,0.11224481,0.24183615,0.102800645,0.06656782,0.08485103,0.0548107,0.08240122,0.07644393,0.067781396,0.034703974
-0.0346278,0.030558925,0.03476333,0.13378303,0.2403646,0.1126566,0.09158917,0.03893308,0.0698818,0.043517902,0.041769095,0.05421182,0.07334285
+0.028286915,0.033591546,0.04288041,0.0652413,0.20657852,0.08050327,0.12876955,0.036749262,0.08919606,0.09513434,0.02763165,0.06495523,0.10048189
+0.015949441,0.027130906,0.03247144,0.11179749,0.24181513,0.10279171,0.06656204,0.08480224,0.05502044,0.082635805,0.07668217,0.0677755,0.034565672
+0.034620997,0.030552922,0.034756497,0.13375674,0.24031734,0.11263447,0.0917502,0.03892543,0.06997049,0.043424457,0.04176089,0.05420117,0.07332844
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index c6da86b1..263dcbe9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.015611125,0.028825635,0.024946507,0.05949394,0.25792566,0.17865989,0.061143175,0.05961025,0.08167672,0.07912997,0.05406415,0.05513047,0.043782484
-0.013057961,0.019449774,0.017848115,0.16736537,0.14540955,0.07191268,0.05906717,0.032956194,0.17133433,0.038080808,0.052356172,0.16933823,0.041823585
-0.015269818,0.018708972,0.024979755,0.21079336,0.16036314,0.06840117,0.06590937,0.03468096,0.12343589,0.05596387,0.06995491,0.1263631,0.025175674
+0.01560967,0.028822947,0.024944182,0.059488393,0.2579016,0.17794675,0.061197206,0.059604697,0.08186873,0.07943712,0.054059114,0.055341084,0.0437784
+0.013061907,0.019455653,0.01785351,0.16741595,0.14602278,0.07193442,0.058912177,0.032966156,0.17071792,0.03809232,0.052423164,0.1693894,0.041754596
+0.015262611,0.018700138,0.024967961,0.2106938,0.1602874,0.068318814,0.065813944,0.034664582,0.123377606,0.056046806,0.06990481,0.12679777,0.025163786
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index b138c563..682787e3 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.01688767,0.02476413,0.02698642,0.07116921,0.2106128,0.1646674,0.04896163,0.025952587,0.05929015,0.12724508,0.16085288,0.03118275,0.031427316
-0.014174571,0.020224977,0.022126192,0.0230076,0.35156232,0.052318905,0.053994514,0.03936724,0.0823695,0.12882835,0.09388537,0.09206946,0.026071027
-0.051626947,0.03439096,0.048215687,0.060181446,0.07837651,0.17194399,0.10227216,0.056314826,0.054316334,0.15594643,0.024674308,0.116798475,0.04494196
-0.019099979,0.023038972,0.014644466,0.044235647,0.47743943,0.06330279,0.023493377,0.01940076,0.11745949,0.038357608,0.09147754,0.02800827,0.040041707
-0.010872557,0.00997722,0.011483708,0.037251316,0.3685879,0.028645372,0.035406873,0.025929531,0.12250072,0.057527177,0.22182012,0.023840794,0.046156727
-0.03117595,0.026252784,0.023167998,0.11226891,0.13280286,0.08968357,0.100637205,0.042946685,0.09678186,0.12972648,0.068729214,0.096971065,0.048855443
+0.016827505,0.024672894,0.026888637,0.071175836,0.2112891,0.1647487,0.048896004,0.025926603,0.05938984,0.12714048,0.16040213,0.031259973,0.031382322
+0.013997956,0.020205943,0.022106718,0.022882363,0.35228354,0.052252322,0.05392724,0.039436106,0.08225181,0.12857363,0.0942475,0.09183136,0.026003601
+0.05171333,0.03437709,0.04829341,0.060120467,0.07857956,0.1719225,0.10227148,0.05633357,0.054162037,0.15633135,0.024398591,0.116559125,0.044937547
+0.019069683,0.023180008,0.014620337,0.04426263,0.4761877,0.06358374,0.023508424,0.019419778,0.11789036,0.038552396,0.09140214,0.028023642,0.040299125
+0.01085677,0.009961516,0.011466335,0.03731547,0.36767113,0.02872022,0.035538275,0.026009187,0.1229666,0.05771425,0.22166713,0.023857085,0.046255976
+0.031226596,0.02629222,0.023113754,0.112368956,0.13236259,0.08975561,0.1010255,0.043042876,0.09706933,0.12928063,0.06875058,0.09676764,0.048943765
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 8748c33c..b2377fc1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.016345652,0.013981177,0.014882885,0.11213945,0.46299294,0.050342005,0.03998455,0.06528273,0.05401568,0.042521786,0.016603058,0.08975528,0.021152845
-0.021561448,0.02073544,0.017325114,0.17226464,0.13950415,0.09796191,0.052949805,0.16631368,0.06838009,0.038587876,0.062359657,0.103468396,0.038587876
-0.020325927,0.01738569,0.014526288,0.09583984,0.1389025,0.047259193,0.0788357,0.06192797,0.20210195,0.15195979,0.05210722,0.065693036,0.05313494
+0.016353136,0.013987578,0.015006481,0.11175343,0.4632051,0.05027907,0.040120237,0.0651852,0.05404043,0.042499743,0.01661066,0.0897964,0.021162536
+0.021549432,0.020723887,0.017451258,0.17284249,0.13888285,0.09771629,0.052868646,0.16687156,0.06822529,0.038566373,0.06232491,0.10341074,0.038566373
+0.020322079,0.01738239,0.014523531,0.0958217,0.1388762,0.04725025,0.07889779,0.061897356,0.20206368,0.15193103,0.052148256,0.06576083,0.053124882
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 0bb283aa..5af54919 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.011731772,0.022969693,0.018313037,0.12035259,0.14745906,0.07250908,0.05957885,0.068290815,0.15881966,0.04703879,0.07210307,0.15333316,0.0475004
-0.01647793,0.022347387,0.019721499,0.07884245,0.21705416,0.08566629,0.034075696,0.04938655,0.16707255,0.04240789,0.09426995,0.057626065,0.115051664
-0.014698837,0.023034362,0.022065567,0.03471401,0.057541963,0.07060523,0.0458093,0.026512388,0.21431695,0.279522,0.07709165,0.06710933,0.06697838
-0.019589733,0.019976106,0.014787107,0.039495137,0.57698613,0.047154225,0.023537608,0.0401759,0.043186717,0.049152493,0.045644898,0.058029015,0.022284957
-0.04979851,0.031715743,0.031346247,0.074355975,0.15856877,0.17689624,0.089297146,0.055445857,0.042346075,0.1642426,0.021044873,0.059688386,0.04525359
-0.027977334,0.027116563,0.0300153,0.06489063,0.08352489,0.118712805,0.09957648,0.06747558,0.053875085,0.14319499,0.2226528,0.02989828,0.031089291
+0.011690648,0.022708276,0.018247731,0.12031194,0.14736429,0.072557904,0.059560545,0.06837643,0.15909584,0.047002815,0.072019346,0.15372172,0.047342557
+0.016566617,0.022464922,0.019672144,0.07886144,0.21629961,0.08572125,0.0340683,0.04920004,0.16754173,0.04258672,0.09411154,0.057664577,0.115241036
+0.0146416435,0.02312187,0.021978369,0.03468886,0.057679538,0.07068581,0.045911815,0.026580743,0.21460655,0.27920017,0.077151075,0.066860445,0.0668931
+0.019577837,0.020039665,0.0147772245,0.03963531,0.5760377,0.047202054,0.023575775,0.040274374,0.043345284,0.049281806,0.045759395,0.058174577,0.02231905
+0.04962753,0.031602994,0.031358972,0.07433624,0.15786043,0.17700651,0.08936339,0.05545165,0.042340003,0.16477123,0.02115324,0.059756365,0.045371544
+0.028013574,0.027042532,0.030052343,0.06491127,0.083546355,0.118885174,0.099732846,0.06744788,0.054070134,0.14321429,0.2222415,0.029767558,0.03107451
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index f8f40bba..2ebebefb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.0342464,0.038354147,0.0311817,0.06441956,0.055262636,0.10589919,0.09309136,0.071602516,0.12141403,0.13439308,0.04130904,0.15468548,0.054140862
-0.0143472245,0.022048492,0.020232834,0.09900767,0.2333686,0.08585151,0.05457054,0.04249958,0.13348985,0.1477598,0.06448847,0.047783565,0.034551915
-0.015537152,0.014036647,0.01318621,0.09705379,0.33219895,0.034673247,0.06452561,0.04620453,0.1406619,0.06882144,0.033279873,0.10210958,0.037711035
+0.034384813,0.038509164,0.031307727,0.06452221,0.055054195,0.106119744,0.09328523,0.07180421,0.12095608,0.13388617,0.041476,0.15470518,0.05398935
+0.014236114,0.022049326,0.0202336,0.099011414,0.2333774,0.08577096,0.054572605,0.04250119,0.1334949,0.14776537,0.06464855,0.047785368,0.034553222
+0.015531263,0.014031327,0.013181212,0.09739672,0.33207306,0.034660105,0.06450115,0.0460519,0.14060858,0.06892985,0.03326726,0.10207088,0.037696745
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 8f216025..295be473 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.03765263,0.032458596,0.023470555,0.16874741,0.20756285,0.13245134,0.058517206,0.05027286,0.09747293,0.02659562,0.028421698,0.04208667,0.09428967
-0.01482675,0.025518453,0.02444518,0.04046103,0.08994222,0.11892158,0.036410987,0.1750695,0.14627604,0.042735364,0.020108057,0.18062681,0.08465808
-0.010264666,0.017460784,0.026728721,0.06693403,0.021815348,0.07467031,0.06196434,0.12652327,0.14004847,0.03016953,0.022773158,0.36467987,0.03596743
+0.037636146,0.032444384,0.023552097,0.16867347,0.20747192,0.13239335,0.05846303,0.050250847,0.09762073,0.026583975,0.028409252,0.04206824,0.09443264
+0.014826758,0.025618346,0.024445197,0.04046106,0.089766786,0.11938711,0.03626906,0.17506963,0.14627615,0.04265201,0.02010807,0.18062693,0.08449295
+0.010266415,0.017463759,0.026733276,0.06694543,0.021734,0.0745373,0.06203545,0.12654482,0.14007233,0.03017467,0.022777036,0.36474198,0.035973556

From b4c8d031c5d600722773a4f35862fb1d6c7323b0 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 8 Jun 2026 13:41:35 +0200
Subject: [PATCH 14/81] Add [mask] input column

---
 src/sequifier/preprocess.py   | 135 ++++++++++++++++++++++++++++++----
 tests/unit/test_preprocess.py |  96 ++++++++++++++++++++++++
 2 files changed, 216 insertions(+), 15 deletions(-)

diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 5e4f3ed7..519ce497 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -23,6 +23,12 @@
     write_data,
 )
 
+RESERVED_MASK_COLUMN = "[mask]"
+INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
+RESERVED_INPUT_COLUMNS = (*INPUT_METADATA_COLUMNS, RESERVED_MASK_COLUMN)
+CATEGORICAL_MASK_ID = 2
+REAL_MASK_VALUE = 0.0
+
 
 @beartype
 def preprocess(args: Any, args_config: dict[str, Any]) -> None:
@@ -131,6 +137,10 @@ def __init__(
 
         if selected_columns is not None:
             selected_columns = ["sequenceId", "itemPosition"] + selected_columns
+            if RESERVED_MASK_COLUMN in selected_columns:
+                raise ValueError(
+                    f"'{RESERVED_MASK_COLUMN}' is not allowed to be in 'selected_columns'"
+                )
 
         self._setup_split_paths(write_format, len(split_ratios))
 
@@ -161,9 +171,7 @@ def __init__(
             data = _load_and_preprocess_data(
                 data_path, read_format, selected_columns, max_rows
             )
-            data_columns = [
-                col for col in data.columns if col not in ["sequenceId", "itemPosition"]
-            ]
+            data_columns = _get_data_columns(data)
             if self.metadata_config_path:
                 metadata_path = os.path.join(
                     self.project_root, self.metadata_config_path
@@ -206,6 +214,7 @@ def __init__(
                 n_classes=n_classes,
                 col_types=col_types,
             )
+            data = _apply_reserved_mask_column(data, data_columns, col_types)
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
@@ -269,9 +278,7 @@ def __init__(
 
                 # Reconstruct data_columns from the provided col_types
                 data_columns = [
-                    col
-                    for col in col_types.keys()
-                    if col not in ["sequenceId", "itemPosition"]
+                    col for col in col_types.keys() if col not in RESERVED_INPUT_COLUMNS
                 ]
 
                 # We still need to find the files to process
@@ -451,11 +458,7 @@ def _get_column_metadata_across_files(
                     )
 
                     # Get columns for the current file (excluding metadata cols)
-                    current_file_cols = [
-                        col
-                        for col in data.columns
-                        if col not in ["sequenceId", "itemPosition"]
-                    ]
+                    current_file_cols = _get_data_columns(data)
 
                     if col_types is None:
                         data_columns = current_file_cols
@@ -880,6 +883,88 @@ def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> No
             json.dump(metadata, f, indent=4)
 
 
+@beartype
+def _get_data_columns(data: pl.DataFrame) -> list[str]:
+    return [col for col in data.columns if col not in RESERVED_INPUT_COLUMNS]
+
+
+@beartype
+def _deduplicate_columns(columns: list[str]) -> list[str]:
+    return list(dict.fromkeys(columns))
+
+
+@beartype
+def _selected_columns_with_optional_mask(
+    data_path: str, read_format: str, selected_columns: Optional[list[str]]
+) -> Optional[list[str]]:
+    if selected_columns is None or read_format != "parquet":
+        return selected_columns
+
+    schema_columns = pq.read_schema(data_path).names
+    if RESERVED_MASK_COLUMN in schema_columns:
+        return selected_columns + [RESERVED_MASK_COLUMN]
+    return selected_columns
+
+
+@beartype
+def _mask_column_expr(mask_dtype: Any) -> pl.Expr:
+    mask_col = pl.col(RESERVED_MASK_COLUMN)
+
+    if isinstance(mask_dtype, pl.Boolean):
+        return mask_col
+
+    if isinstance(
+        mask_dtype,
+        (
+            pl.Int8,
+            pl.Int16,
+            pl.Int32,
+            pl.Int64,
+            pl.UInt8,
+            pl.UInt16,
+            pl.UInt32,
+            pl.UInt64,
+            pl.Float32,
+            pl.Float64,
+        ),
+    ):
+        return mask_col == 1
+
+    if isinstance(mask_dtype, (pl.String, pl.Utf8)):
+        return mask_col.str.to_lowercase().is_in(["1", "true"])
+
+    raise ValueError(
+        f"Column {RESERVED_MASK_COLUMN} must be boolean, numeric, or string, got {mask_dtype}"
+    )
+
+
+@beartype
+def _apply_reserved_mask_column(
+    data: pl.DataFrame, data_columns: list[str], col_types: dict[str, str]
+) -> pl.DataFrame:
+    if RESERVED_MASK_COLUMN not in data.columns:
+        return data
+
+    mask_expr = _mask_column_expr(data.schema[RESERVED_MASK_COLUMN])
+    updates = []
+    for col in data_columns:
+        mask_value = (
+            CATEGORICAL_MASK_ID if col_types[col] == "Int64" else REAL_MASK_VALUE
+        )
+        updates.append(
+            pl.when(mask_expr)
+            .then(pl.lit(mask_value))
+            .otherwise(pl.col(col))
+            .cast(data.schema[col])
+            .alias(col)
+        )
+
+    if updates:
+        data = data.with_columns(updates)
+
+    return data.drop(RESERVED_MASK_COLUMN)
+
+
 @beartype
 def _apply_column_statistics(
     data: pl.DataFrame,
@@ -1042,6 +1127,10 @@ def _get_column_statistics(
         ValueError: If a column has an unsupported data type (neither
             string, integer, nor float).
     """
+    if RESERVED_MASK_COLUMN in data.columns:
+        mask_expr = _mask_column_expr(data.schema[RESERVED_MASK_COLUMN])
+        data = data.filter(~mask_expr)
+
     for data_col in data_columns:
         dtype = data.schema[data_col]
         if isinstance(
@@ -1117,16 +1206,22 @@ def _load_and_preprocess_data(
         prepared data.
     """
     logger.info(f"Reading data from '{data_path}'...")
-    data = read_data(data_path, read_format, columns=selected_columns)
+    columns_to_read = _selected_columns_with_optional_mask(
+        data_path, read_format, selected_columns
+    )
+    data = read_data(data_path, read_format, columns=columns_to_read)
 
     if data.null_count().sum().sum_horizontal().item() != 0:
         raise ValueError(f"NaN or null values not accepted: {data.null_count()}")
 
     if selected_columns:
         selected_columns_filtered = [
-            col for col in selected_columns if col not in ["sequenceId", "itemPosition"]
+            col for col in selected_columns if col not in INPUT_METADATA_COLUMNS
         ]
-        data = data.select(["sequenceId", "itemPosition"] + selected_columns_filtered)
+        columns_to_select = list(INPUT_METADATA_COLUMNS) + selected_columns_filtered
+        if RESERVED_MASK_COLUMN in data.columns:
+            columns_to_select.append(RESERVED_MASK_COLUMN)
+        data = data.select(_deduplicate_columns(columns_to_select))
 
     if max_rows:
         data = data.slice(0, int(max_rows))
@@ -1287,6 +1382,7 @@ def _process_batches_multiple_files_inner(
                 n_classes,
                 col_types,
             )
+            data = _apply_reserved_mask_column(data, data_columns, col_types)
 
             data_name_root_inner = f"{data_name_root}-{process_id}-{file_index_str}"
 
@@ -1549,8 +1645,17 @@ def combine_maps(
     Returns:
         A new, combined, and re-indexed ID map.
     """
-    combined_keys = sorted(list(set(list(map1.keys())).union(list(set(map2.keys())))))
+    keys1 = {k for k in map1.keys() if k not in ["[unknown]", "[other]", "[mask]"]}
+    keys2 = {k for k in map2.keys() if k not in ["[unknown]", "[other]", "[mask]"]}
+
+    combined_keys = sorted(list(keys1.union(keys2)))
     id_map = {id_: i + 3 for i, id_ in enumerate(combined_keys)}
+
+    # Re-apply special mapping rules if these are string keys
+    if combined_keys and isinstance(combined_keys[0], str):
+        id_map["[unknown]"] = 0
+        id_map["[other]"] = 1
+
     return id_map
 
 
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 82b56c35..56f6595b 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -1,10 +1,16 @@
+import json
+
 import numpy as np
 import polars as pl
 import pytest
 import torch
 
 from sequifier.preprocess import (
+    RESERVED_MASK_COLUMN,
+    Preprocessor,
+    _apply_reserved_mask_column,
     _get_column_statistics,
+    _get_data_columns,
     create_id_map,
     extract_sequences,
     extract_subsequences,
@@ -133,6 +139,96 @@ def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
     )
 
 
+def test_reserved_mask_column_is_not_a_data_column():
+    data = pl.DataFrame(
+        {
+            "sequenceId": [1],
+            "itemPosition": [0],
+            "itemId": [101],
+            RESERVED_MASK_COLUMN: [1],
+        }
+    )
+
+    assert _get_data_columns(data) == ["itemId"]
+
+
+def test_apply_reserved_mask_column_replaces_values_and_drops_column():
+    data = pl.DataFrame(
+        {
+            "cat_col": [3, 4, 5, 6],
+            "num_col": [-1.0, 2.5, 3.5, 4.5],
+            RESERVED_MASK_COLUMN: ["0", "1", "0", "1"],
+        }
+    )
+
+    masked = _apply_reserved_mask_column(
+        data,
+        ["cat_col", "num_col"],
+        {"cat_col": "Int64", "num_col": "Float64"},
+    )
+
+    assert RESERVED_MASK_COLUMN not in masked.columns
+    assert masked["cat_col"].to_list() == [3, 2, 5, 2]
+    assert masked["num_col"].to_list() == [-1.0, 0.0, 3.5, 0.0]
+
+
+def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
+    project_root = tmp_path / "project"
+    project_root.mkdir()
+    data_path = tmp_path / "masked-input.csv"
+    pl.DataFrame(
+        {
+            "sequenceId": [0, 0, 0],
+            "itemPosition": [0, 1, 2],
+            "itemId": ["a", "b", "c"],
+            "itemValue": [10.0, 100.0, 70.0],
+            RESERVED_MASK_COLUMN: [0, 1, 0],
+        }
+    ).write_csv(data_path)
+
+    Preprocessor(
+        project_root=str(project_root),
+        continue_preprocessing=False,
+        data_path=str(data_path),
+        read_format="csv",
+        write_format="parquet",
+        merge_output=True,
+        selected_columns=["itemId", "itemValue"],
+        split_ratios=[1.0],
+        seq_length=2,
+        stride_by_split=[1],
+        max_rows=None,
+        seed=1010,
+        n_cores=1,
+        batches_per_file=1024,
+        process_by_file=True,
+        subsequence_start_mode="distribute",
+        use_precomputed_maps=None,
+        metadata_config_path=None,
+    )
+
+    output = pl.read_parquet(project_root / "data" / "masked-input-split0.parquet")
+    metadata_path = project_root / "configs" / "metadata_configs" / "masked-input.json"
+
+    item_sequence = (
+        output.filter(pl.col("inputCol") == "itemId").select(["2", "1", "0"]).row(0)
+    )
+    value_sequence = (
+        output.filter(pl.col("inputCol") == "itemValue").select(["2", "1", "0"]).row(0)
+    )
+
+    assert RESERVED_MASK_COLUMN not in output.columns
+    assert item_sequence == (3.0, 2.0, 4.0)
+    assert value_sequence[1] == 0.0
+    assert value_sequence[0] != 0.0
+    assert value_sequence[2] != 0.0
+
+    metadata = json.loads(metadata_path.read_text())
+    assert RESERVED_MASK_COLUMN not in metadata["column_types"]
+    assert RESERVED_MASK_COLUMN not in metadata["id_maps"]
+    assert RESERVED_MASK_COLUMN not in metadata["selected_columns_statistics"]
+
+
 @pytest.mark.parametrize("mode", ["distribute", "exact"])
 def test_extract_subsequences_modes(mode):
     """Tests logic for 'distribute' vs 'exact' modes."""

From 3753839cd74b99aa8fe5bbddb20dd9b1580ecbef Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 8 Jun 2026 14:04:36 +0200
Subject: [PATCH 15/81] add reserved_mask_column

---
 documentation/configs/preprocess.md           |   1 +
 documentation/consolidated-docs.md            |   1 +
 src/sequifier/config/preprocess_config.py     |  19 ++-
 src/sequifier/make.py                         |   1 +
 src/sequifier/preprocess.py                   | 128 ++++++++++++++----
 ...ss-test-categorical-precomputed-stats.yaml |   1 +
 tests/unit/test_preprocess.py                 | 115 +++++++++++++++-
 7 files changed, 233 insertions(+), 33 deletions(-)

diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index f58c542c..ac5911f0 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -38,6 +38,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | `selected_columns` | `list[str]` | No | `null` | A specific list of columns to process. If `null`, all columns (except metadata) are processed. |
 | `max_rows` | `int` | No | `null` | Limits processing to the first N rows. Useful for rapid debugging. |
 | `metadata_config_path` | `Optional[str]` | No | `null` | use a preexisting metadata config path for tokenizing discrete columns and standardising real-valued columns |
+| `reserved_mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
 | `use_precomputed_maps`| `list[str]` | No | `null` | If not `null`, enforces the use of precomputed maps for the variables in the list. |
 
 ### 3\. Sequence Logic & Splitting
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 65315961..ec13be11 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -246,6 +246,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | `selected_columns` | `list[str]` | No | `null` | A specific list of columns to process. If `null`, all columns (except metadata) are processed. |
 | `max_rows` | `int` | No | `null` | Limits processing to the first N rows. Useful for rapid debugging. |
 | `metadata_config_path` | `Optional[str]` | No | `null` | use a preexisting metadata config path for tokenizing discrete columns and standardising real-valued columns |
+| `reserved_mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
 | `use_precomputed_maps`| `list[str]` | No | `null` | If not `null`, enforces the use of precomputed maps for the variables in the list. |
 
 ### 3\. Sequence Logic & Splitting
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index 94f72009..493c09c2 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -5,7 +5,13 @@
 import numpy as np
 import yaml
 from beartype import beartype
-from pydantic import BaseModel, ConfigDict, ValidationInfo, field_validator
+from pydantic import (
+    BaseModel,
+    ConfigDict,
+    ValidationInfo,
+    field_validator,
+    model_validator,
+)
 
 from sequifier.helpers import try_catch_excess_keys
 
@@ -56,6 +62,8 @@ class PreprocessorModel(BaseModel):
         process_by_file: A flag to indicate if processing should be done file by file.
         continue_preprocessing: Continue preprocessing job that was interrupted while writing to temp folder.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
+        reserved_mask_column: Optional input column used to mask all model input columns.
+                              Requires metadata_config_path when set.
     """
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
@@ -79,6 +87,7 @@ class PreprocessorModel(BaseModel):
     subsequence_start_mode: str = "distribute"
     use_precomputed_maps: Optional[list[str]] = None
     metadata_config_path: Optional[str] = None
+    reserved_mask_column: Optional[str] = None
 
     @field_validator("data_path")
     @classmethod
@@ -176,6 +185,14 @@ def validate_subsequence_start_mode(cls, v: str) -> str:
             )
         return v
 
+    @model_validator(mode="after")
+    def validate_reserved_mask_column_requires_metadata(self) -> "PreprocessorModel":
+        if self.reserved_mask_column is not None and self.metadata_config_path is None:
+            raise ValueError(
+                "metadata_config_path must be set when reserved_mask_column is set"
+            )
+        return self
+
     def __init__(self, **kwargs):
         default_stride_for_split = [kwargs["seq_length"]] * len(kwargs["split_ratios"])
         kwargs["stride_by_split"] = kwargs.get(
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 9ebf9383..1d20e3ab 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -5,6 +5,7 @@
 read_format: csv
 write_format: parquet
 selected_columns: [EXAMPLE_INPUT_COLUMN_NAME] # should include all target column, can include additional columns
+reserved_mask_column: null
 
 split_ratios:
 - 0.8
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 519ce497..532b15eb 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -25,7 +25,6 @@
 
 RESERVED_MASK_COLUMN = "[mask]"
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
-RESERVED_INPUT_COLUMNS = (*INPUT_METADATA_COLUMNS, RESERVED_MASK_COLUMN)
 CATEGORICAL_MASK_ID = 2
 REAL_MASK_VALUE = 0.0
 
@@ -92,6 +91,7 @@ def __init__(
         subsequence_start_mode: str,
         use_precomputed_maps: Optional[list[str]],
         metadata_config_path: Optional[str],
+        reserved_mask_column: Optional[str] = None,
     ):
         """Initializes the Preprocessor with the given parameters.
 
@@ -112,6 +112,7 @@ def __init__(
             process_by_file: A flag to indicate if processing should be done file by file.
             use_precomputed_maps: An optional list of columns for which to enforce precomputed maps
             metadata_config_path: Optional path to a precomputed metadata config
+            reserved_mask_column: Optional input column used to mask all data columns
         """
         self.project_root = project_root
         self.batches_per_file = batches_per_file
@@ -129,6 +130,12 @@ def __init__(
 
         self.use_precomputed_maps = use_precomputed_maps
         self.metadata_config_path = metadata_config_path
+        self.reserved_mask_column = reserved_mask_column
+        if self.reserved_mask_column is not None and self.metadata_config_path is None:
+            raise ValueError(
+                "metadata_config_path must be set when reserved_mask_column is set"
+            )
+
         self.seed = seed
         np.random.seed(seed)
         self.n_cores = n_cores or multiprocessing.cpu_count()
@@ -137,9 +144,12 @@ def __init__(
 
         if selected_columns is not None:
             selected_columns = ["sequenceId", "itemPosition"] + selected_columns
-            if RESERVED_MASK_COLUMN in selected_columns:
+            if (
+                self.reserved_mask_column is not None
+                and self.reserved_mask_column in selected_columns
+            ):
                 raise ValueError(
-                    f"'{RESERVED_MASK_COLUMN}' is not allowed to be in 'selected_columns'"
+                    f"'{self.reserved_mask_column}' is not allowed to be in 'selected_columns'"
                 )
 
         self._setup_split_paths(write_format, len(split_ratios))
@@ -169,9 +179,13 @@ def __init__(
 
         if os.path.isfile(data_path):
             data = _load_and_preprocess_data(
-                data_path, read_format, selected_columns, max_rows
+                data_path,
+                read_format,
+                selected_columns,
+                max_rows,
+                self.reserved_mask_column,
             )
-            data_columns = _get_data_columns(data)
+            data_columns = _get_data_columns(data, self.reserved_mask_column)
             if self.metadata_config_path:
                 metadata_path = os.path.join(
                     self.project_root, self.metadata_config_path
@@ -200,6 +214,7 @@ def __init__(
                     selected_columns_statistics,
                     0,
                     precomputed_id_maps,
+                    self.reserved_mask_column,
                 )
 
                 id_maps = id_maps | precomputed_id_maps
@@ -214,7 +229,9 @@ def __init__(
                 n_classes=n_classes,
                 col_types=col_types,
             )
-            data = _apply_reserved_mask_column(data, data_columns, col_types)
+            data = _apply_reserved_mask_column(
+                data, data_columns, col_types, self.reserved_mask_column
+            )
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
@@ -278,7 +295,9 @@ def __init__(
 
                 # Reconstruct data_columns from the provided col_types
                 data_columns = [
-                    col for col in col_types.keys() if col not in RESERVED_INPUT_COLUMNS
+                    col
+                    for col in col_types.keys()
+                    if col not in _reserved_input_columns(self.reserved_mask_column)
                 ]
 
                 # We still need to find the files to process
@@ -455,10 +474,13 @@ def _get_column_metadata_across_files(
                         read_format,
                         selected_columns,
                         max_rows_inner,
+                        self.reserved_mask_column,
                     )
 
                     # Get columns for the current file (excluding metadata cols)
-                    current_file_cols = _get_data_columns(data)
+                    current_file_cols = _get_data_columns(
+                        data, self.reserved_mask_column
+                    )
 
                     if col_types is None:
                         data_columns = current_file_cols
@@ -498,6 +520,7 @@ def _get_column_metadata_across_files(
                         selected_columns_statistics,
                         n_rows_running_count,
                         precomputed_id_maps,
+                        self.reserved_mask_column,
                     )
                     n_rows_running_count += data.shape[0]
 
@@ -584,6 +607,7 @@ def _process_batches_multiple_files(
         write_format: str,
         process_by_file: bool = True,
         subsequence_start_mode: str = "distribute",
+        reserved_mask_column: Optional[str] = None,
     ) -> None:
         """Processes batches of data from multiple files.
 
@@ -605,7 +629,11 @@ def _process_batches_multiple_files(
             write_format: The file format for the output files.
             process_by_file: A flag to indicate if processing should be done file by file.
             subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
+            reserved_mask_column: Optional input column used to mask all data columns.
         """
+        if reserved_mask_column is None:
+            reserved_mask_column = self.reserved_mask_column
+
         if process_by_file:
             _process_batches_multiple_files_inner(
                 project_root=self.project_root,
@@ -632,6 +660,7 @@ def _process_batches_multiple_files(
                 merge_output=self.merge_output,
                 continue_preprocessing=self.continue_preprocessing,
                 subsequence_start_mode=subsequence_start_mode,
+                reserved_mask_column=reserved_mask_column,
             )
             input_files = create_file_paths_for_multiple_files2(
                 self.project_root,
@@ -676,6 +705,7 @@ def _process_batches_multiple_files(
                 "merge_output": self.merge_output,
                 "continue_preprocessing": self.continue_preprocessing,
                 "subsequence_start_mode": subsequence_start_mode,
+                "reserved_mask_column": reserved_mask_column,
             }
 
             job_params = [
@@ -884,8 +914,21 @@ def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> No
 
 
 @beartype
-def _get_data_columns(data: pl.DataFrame) -> list[str]:
-    return [col for col in data.columns if col not in RESERVED_INPUT_COLUMNS]
+def _reserved_input_columns(reserved_mask_column: Optional[str]) -> tuple[str, ...]:
+    if reserved_mask_column is None:
+        return INPUT_METADATA_COLUMNS
+    return (*INPUT_METADATA_COLUMNS, reserved_mask_column)
+
+
+@beartype
+def _get_data_columns(
+    data: pl.DataFrame, reserved_mask_column: Optional[str] = None
+) -> list[str]:
+    return [
+        col
+        for col in data.columns
+        if col not in _reserved_input_columns(reserved_mask_column)
+    ]
 
 
 @beartype
@@ -895,20 +938,27 @@ def _deduplicate_columns(columns: list[str]) -> list[str]:
 
 @beartype
 def _selected_columns_with_optional_mask(
-    data_path: str, read_format: str, selected_columns: Optional[list[str]]
+    data_path: str,
+    read_format: str,
+    selected_columns: Optional[list[str]],
+    reserved_mask_column: Optional[str] = None,
 ) -> Optional[list[str]]:
-    if selected_columns is None or read_format != "parquet":
+    if (
+        selected_columns is None
+        or read_format != "parquet"
+        or reserved_mask_column is None
+    ):
         return selected_columns
 
     schema_columns = pq.read_schema(data_path).names
-    if RESERVED_MASK_COLUMN in schema_columns:
-        return selected_columns + [RESERVED_MASK_COLUMN]
+    if reserved_mask_column in schema_columns:
+        return selected_columns + [reserved_mask_column]
     return selected_columns
 
 
 @beartype
-def _mask_column_expr(mask_dtype: Any) -> pl.Expr:
-    mask_col = pl.col(RESERVED_MASK_COLUMN)
+def _mask_column_expr(mask_dtype: Any, reserved_mask_column: str) -> pl.Expr:
+    mask_col = pl.col(reserved_mask_column)
 
     if isinstance(mask_dtype, pl.Boolean):
         return mask_col
@@ -934,18 +984,23 @@ def _mask_column_expr(mask_dtype: Any) -> pl.Expr:
         return mask_col.str.to_lowercase().is_in(["1", "true"])
 
     raise ValueError(
-        f"Column {RESERVED_MASK_COLUMN} must be boolean, numeric, or string, got {mask_dtype}"
+        f"Column {reserved_mask_column} must be boolean, numeric, or string, got {mask_dtype}"
     )
 
 
 @beartype
 def _apply_reserved_mask_column(
-    data: pl.DataFrame, data_columns: list[str], col_types: dict[str, str]
+    data: pl.DataFrame,
+    data_columns: list[str],
+    col_types: dict[str, str],
+    reserved_mask_column: Optional[str] = None,
 ) -> pl.DataFrame:
-    if RESERVED_MASK_COLUMN not in data.columns:
+    if reserved_mask_column is None or reserved_mask_column not in data.columns:
         return data
 
-    mask_expr = _mask_column_expr(data.schema[RESERVED_MASK_COLUMN])
+    mask_expr = _mask_column_expr(
+        data.schema[reserved_mask_column], reserved_mask_column
+    )
     updates = []
     for col in data_columns:
         mask_value = (
@@ -962,7 +1017,7 @@ def _apply_reserved_mask_column(
     if updates:
         data = data.with_columns(updates)
 
-    return data.drop(RESERVED_MASK_COLUMN)
+    return data.drop(reserved_mask_column)
 
 
 @beartype
@@ -1091,6 +1146,7 @@ def _get_column_statistics(
     selected_columns_statistics: dict[str, dict[str, float]],
     n_rows_running_count: int,
     precomputed_id_maps: dict[str, dict[Union[str, int], int]],
+    reserved_mask_column: Optional[str] = None,
 ) -> tuple[
     dict[str, dict[Union[str, int], int]],
     dict[str, dict[str, float]],
@@ -1127,8 +1183,10 @@ def _get_column_statistics(
         ValueError: If a column has an unsupported data type (neither
             string, integer, nor float).
     """
-    if RESERVED_MASK_COLUMN in data.columns:
-        mask_expr = _mask_column_expr(data.schema[RESERVED_MASK_COLUMN])
+    if reserved_mask_column is not None and reserved_mask_column in data.columns:
+        mask_expr = _mask_column_expr(
+            data.schema[reserved_mask_column], reserved_mask_column
+        )
         data = data.filter(~mask_expr)
 
     for data_col in data_columns:
@@ -1184,6 +1242,7 @@ def _load_and_preprocess_data(
     read_format: str,
     selected_columns: Optional[list[str]],
     max_rows: Optional[int],
+    reserved_mask_column: Optional[str] = None,
 ) -> pl.DataFrame:
     """Loads data from a file and performs initial preparation.
 
@@ -1207,7 +1266,7 @@ def _load_and_preprocess_data(
     """
     logger.info(f"Reading data from '{data_path}'...")
     columns_to_read = _selected_columns_with_optional_mask(
-        data_path, read_format, selected_columns
+        data_path, read_format, selected_columns, reserved_mask_column
     )
     data = read_data(data_path, read_format, columns=columns_to_read)
 
@@ -1219,8 +1278,8 @@ def _load_and_preprocess_data(
             col for col in selected_columns if col not in INPUT_METADATA_COLUMNS
         ]
         columns_to_select = list(INPUT_METADATA_COLUMNS) + selected_columns_filtered
-        if RESERVED_MASK_COLUMN in data.columns:
-            columns_to_select.append(RESERVED_MASK_COLUMN)
+        if reserved_mask_column is not None and reserved_mask_column in data.columns:
+            columns_to_select.append(reserved_mask_column)
         data = data.select(_deduplicate_columns(columns_to_select))
 
     if max_rows:
@@ -1305,6 +1364,7 @@ def _process_batches_multiple_files_inner(
     merge_output: bool,
     continue_preprocessing: bool,
     subsequence_start_mode: str,
+    reserved_mask_column: Optional[str],
 ):
     """Inner function for processing batches of data from multiple files.
 
@@ -1366,13 +1426,21 @@ def _process_batches_multiple_files_inner(
                     logger.info(f"Skipping already processed file: {path}")
                     if max_rows is not None:
                         data = _load_and_preprocess_data(
-                            path, read_format, selected_columns, max_rows_inner
+                            path,
+                            read_format,
+                            selected_columns,
+                            max_rows_inner,
+                            reserved_mask_column,
                         )
                         n_rows_running_count += data.shape[0]
                     continue
 
             data = _load_and_preprocess_data(
-                path, read_format, selected_columns, max_rows_inner
+                path,
+                read_format,
+                selected_columns,
+                max_rows_inner,
+                reserved_mask_column,
             )
             data, _, _ = _apply_column_statistics(
                 data,
@@ -1382,7 +1450,9 @@ def _process_batches_multiple_files_inner(
                 n_classes,
                 col_types,
             )
-            data = _apply_reserved_mask_column(data, data_columns, col_types)
+            data = _apply_reserved_mask_column(
+                data, data_columns, col_types, reserved_mask_column
+            )
 
             data_name_root_inner = f"{data_name_root}-{process_id}-{file_index_str}"
 
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
index 15adf0e8..2c0b797f 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
@@ -12,3 +12,4 @@ stride_by_split:
 - 1
 max_rows: null
 metadata_config_path: configs/metadata_configs/test-data-categorical-multitarget-5.json
+reserved_mask_column: "[mask]"
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 56f6595b..e7ac8483 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -4,7 +4,9 @@
 import polars as pl
 import pytest
 import torch
+from pydantic import ValidationError
 
+from sequifier.config.preprocess_config import PreprocessorModel
 from sequifier.preprocess import (
     RESERVED_MASK_COLUMN,
     Preprocessor,
@@ -149,7 +151,7 @@ def test_reserved_mask_column_is_not_a_data_column():
         }
     )
 
-    assert _get_data_columns(data) == ["itemId"]
+    assert _get_data_columns(data, RESERVED_MASK_COLUMN) == ["itemId"]
 
 
 def test_apply_reserved_mask_column_replaces_values_and_drops_column():
@@ -165,6 +167,7 @@ def test_apply_reserved_mask_column_replaces_values_and_drops_column():
         data,
         ["cat_col", "num_col"],
         {"cat_col": "Int64", "num_col": "Float64"},
+        RESERVED_MASK_COLUMN,
     )
 
     assert RESERVED_MASK_COLUMN not in masked.columns
@@ -175,6 +178,22 @@ def test_apply_reserved_mask_column_replaces_values_and_drops_column():
 def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
     project_root = tmp_path / "project"
     project_root.mkdir()
+    metadata_dir = project_root / "configs" / "metadata_configs"
+    metadata_dir.mkdir(parents=True)
+    metadata_config_path = "configs/metadata_configs/masked-input-base.json"
+    (project_root / metadata_config_path).write_text(
+        json.dumps(
+            {
+                "n_classes": {"itemId": 6},
+                "id_maps": {"itemId": {"a": 3, "b": 4, "c": 5}},
+                "split_paths": [],
+                "column_types": {"itemId": "Int64", "itemValue": "Float64"},
+                "selected_columns_statistics": {
+                    "itemValue": {"mean": 60.0, "std": 10.0}
+                },
+            }
+        )
+    )
     data_path = tmp_path / "masked-input.csv"
     pl.DataFrame(
         {
@@ -204,7 +223,8 @@ def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
         process_by_file=True,
         subsequence_start_mode="distribute",
         use_precomputed_maps=None,
-        metadata_config_path=None,
+        metadata_config_path=metadata_config_path,
+        reserved_mask_column=RESERVED_MASK_COLUMN,
     )
 
     output = pl.read_parquet(project_root / "data" / "masked-input-split0.parquet")
@@ -218,7 +238,7 @@ def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
     )
 
     assert RESERVED_MASK_COLUMN not in output.columns
-    assert item_sequence == (3.0, 2.0, 4.0)
+    assert item_sequence == (3.0, 2.0, 5.0)
     assert value_sequence[1] == 0.0
     assert value_sequence[0] != 0.0
     assert value_sequence[2] != 0.0
@@ -229,6 +249,95 @@ def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
     assert RESERVED_MASK_COLUMN not in metadata["selected_columns_statistics"]
 
 
+def test_preprocessor_requires_metadata_config_for_reserved_mask_column(tmp_path):
+    project_root = tmp_path / "project"
+    project_root.mkdir()
+    data_path = tmp_path / "masked-input.csv"
+    pl.DataFrame(
+        {
+            "sequenceId": [0],
+            "itemPosition": [0],
+            "itemId": ["a"],
+            RESERVED_MASK_COLUMN: [0],
+        }
+    ).write_csv(data_path)
+
+    with pytest.raises(
+        ValueError,
+        match="metadata_config_path must be set when reserved_mask_column is set",
+    ):
+        Preprocessor(
+            project_root=str(project_root),
+            continue_preprocessing=False,
+            data_path=str(data_path),
+            read_format="csv",
+            write_format="parquet",
+            merge_output=True,
+            selected_columns=["itemId"],
+            split_ratios=[1.0],
+            seq_length=1,
+            stride_by_split=[1],
+            max_rows=None,
+            seed=1010,
+            n_cores=1,
+            batches_per_file=1024,
+            process_by_file=True,
+            subsequence_start_mode="distribute",
+            use_precomputed_maps=None,
+            metadata_config_path=None,
+            reserved_mask_column=RESERVED_MASK_COLUMN,
+        )
+
+
+def test_preprocessor_config_defaults_reserved_mask_column_to_none(tmp_path):
+    data_path = tmp_path / "input.csv"
+    pl.DataFrame(
+        {
+            "sequenceId": [0],
+            "itemPosition": [0],
+            "itemId": ["a"],
+            RESERVED_MASK_COLUMN: [0],
+        }
+    ).write_csv(data_path)
+
+    config = PreprocessorModel(
+        project_root=str(tmp_path),
+        data_path=str(data_path),
+        split_ratios=[1.0],
+        seq_length=1,
+        seed=1010,
+    )
+
+    assert config.reserved_mask_column is None
+
+
+def test_preprocessor_config_requires_metadata_config_for_reserved_mask_column(
+    tmp_path,
+):
+    data_path = tmp_path / "input.csv"
+    pl.DataFrame(
+        {
+            "sequenceId": [0],
+            "itemPosition": [0],
+            "itemId": ["a"],
+            RESERVED_MASK_COLUMN: [0],
+        }
+    ).write_csv(data_path)
+
+    with pytest.raises(
+        ValidationError,
+        match="metadata_config_path must be set when reserved_mask_column is set",
+    ):
+        PreprocessorModel(
+            project_root=str(tmp_path),
+            data_path=str(data_path),
+            split_ratios=[1.0],
+            seq_length=1,
+            seed=1010,
+            reserved_mask_column=RESERVED_MASK_COLUMN,
+        )
+
+
 @pytest.mark.parametrize("mode", ["distribute", "exact"])
 def test_extract_subsequences_modes(mode):
     """Tests logic for 'distribute' vs 'exact' modes."""

From 04fdba6b4cd952d2a8e6ff7a5697d1ff5b0b29d9 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 8 Jun 2026 15:06:34 +0200
Subject: [PATCH 16/81] Rename reserved_mask_column to mask_column

---
 documentation/configs/preprocess.md           |   2 +-
 documentation/consolidated-docs.md            |   2 +-
 src/sequifier/config/preprocess_config.py     |  12 +-
 src/sequifier/make.py                         |   2 +-
 src/sequifier/preprocess.py                   | 121 ++++++++----------
 ...ss-test-categorical-precomputed-stats.yaml |   2 +-
 tests/unit/test_preprocess.py                 |  28 ++--
 7 files changed, 73 insertions(+), 96 deletions(-)

diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index ac5911f0..74ef6c5d 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -38,7 +38,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | `selected_columns` | `list[str]` | No | `null` | A specific list of columns to process. If `null`, all columns (except metadata) are processed. |
 | `max_rows` | `int` | No | `null` | Limits processing to the first N rows. Useful for rapid debugging. |
 | `metadata_config_path` | `Optional[str]` | No | `null` | use a preexisting metadata config path for tokenizing discrete columns and standardising real-valued columns |
-| `reserved_mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
+| `mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
 | `use_precomputed_maps`| `list[str]` | No | `null` | If not `null`, enforces the use of precomputed maps for the variables in the list. |
 
 ### 3\. Sequence Logic & Splitting
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index ec13be11..b1563d6a 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -246,7 +246,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | `selected_columns` | `list[str]` | No | `null` | A specific list of columns to process. If `null`, all columns (except metadata) are processed. |
 | `max_rows` | `int` | No | `null` | Limits processing to the first N rows. Useful for rapid debugging. |
 | `metadata_config_path` | `Optional[str]` | No | `null` | use a preexisting metadata config path for tokenizing discrete columns and standardising real-valued columns |
-| `reserved_mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
+| `mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
 | `use_precomputed_maps`| `list[str]` | No | `null` | If not `null`, enforces the use of precomputed maps for the variables in the list. |
 
 ### 3\. Sequence Logic & Splitting
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index 493c09c2..4777d26d 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -62,7 +62,7 @@ class PreprocessorModel(BaseModel):
         process_by_file: A flag to indicate if processing should be done file by file.
         continue_preprocessing: Continue preprocessing job that was interrupted while writing to temp folder.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-        reserved_mask_column: Optional input column used to mask all model input columns.
+        mask_column: Optional input column used to mask all model input columns.
                               Requires metadata_config_path when set.
     """
 
@@ -87,7 +87,7 @@ class PreprocessorModel(BaseModel):
     subsequence_start_mode: str = "distribute"
     use_precomputed_maps: Optional[list[str]] = None
     metadata_config_path: Optional[str] = None
-    reserved_mask_column: Optional[str] = None
+    mask_column: Optional[str] = None
 
     @field_validator("data_path")
     @classmethod
@@ -186,11 +186,9 @@ def validate_subsequence_start_mode(cls, v: str) -> str:
         return v
 
     @model_validator(mode="after")
-    def validate_reserved_mask_column_requires_metadata(self) -> "PreprocessorModel":
-        if self.reserved_mask_column is not None and self.metadata_config_path is None:
-            raise ValueError(
-                "metadata_config_path must be set when reserved_mask_column is set"
-            )
+    def validate_mask_column_requires_metadata(self) -> "PreprocessorModel":
+        if self.mask_column is not None and self.metadata_config_path is None:
+            raise ValueError("metadata_config_path must be set when mask_column is set")
         return self
 
     def __init__(self, **kwargs):
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 1d20e3ab..8777ac97 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -5,7 +5,7 @@
 read_format: csv
 write_format: parquet
 selected_columns: [EXAMPLE_INPUT_COLUMN_NAME] # should include all target column, can include additional columns
-reserved_mask_column: null
+mask_column: null
 
 split_ratios:
 - 0.8
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 532b15eb..a1b6125e 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -91,7 +91,7 @@ def __init__(
         subsequence_start_mode: str,
         use_precomputed_maps: Optional[list[str]],
         metadata_config_path: Optional[str],
-        reserved_mask_column: Optional[str] = None,
+        mask_column: Optional[str] = None,
     ):
         """Initializes the Preprocessor with the given parameters.
 
@@ -112,7 +112,7 @@ def __init__(
             process_by_file: A flag to indicate if processing should be done file by file.
             use_precomputed_maps: An optional list of columns for which to enforce precomputed maps
             metadata_config_path: Optional path to a precomputed metadata config
-            reserved_mask_column: Optional input column used to mask all data columns
+            mask_column: Optional input column used to mask all data columns
         """
         self.project_root = project_root
         self.batches_per_file = batches_per_file
@@ -130,11 +130,9 @@ def __init__(
 
         self.use_precomputed_maps = use_precomputed_maps
         self.metadata_config_path = metadata_config_path
-        self.reserved_mask_column = reserved_mask_column
-        if self.reserved_mask_column is not None and self.metadata_config_path is None:
-            raise ValueError(
-                "metadata_config_path must be set when reserved_mask_column is set"
-            )
+        self.mask_column = mask_column
+        if self.mask_column is not None and self.metadata_config_path is None:
+            raise ValueError("metadata_config_path must be set when mask_column is set")
 
         self.seed = seed
         np.random.seed(seed)
@@ -144,12 +142,9 @@ def __init__(
 
         if selected_columns is not None:
             selected_columns = ["sequenceId", "itemPosition"] + selected_columns
-            if (
-                self.reserved_mask_column is not None
-                and self.reserved_mask_column in selected_columns
-            ):
+            if self.mask_column is not None and self.mask_column in selected_columns:
                 raise ValueError(
-                    f"'{self.reserved_mask_column}' is not allowed to be in 'selected_columns'"
+                    f"'{self.mask_column}' is not allowed to be in 'selected_columns'"
                 )
 
         self._setup_split_paths(write_format, len(split_ratios))
@@ -183,9 +178,9 @@ def __init__(
                 read_format,
                 selected_columns,
                 max_rows,
-                self.reserved_mask_column,
+                self.mask_column,
             )
-            data_columns = _get_data_columns(data, self.reserved_mask_column)
+            data_columns = _get_data_columns(data, self.mask_column)
             if self.metadata_config_path:
                 metadata_path = os.path.join(
                     self.project_root, self.metadata_config_path
@@ -214,7 +209,7 @@ def __init__(
                     selected_columns_statistics,
                     0,
                     precomputed_id_maps,
-                    self.reserved_mask_column,
+                    self.mask_column,
                 )
 
                 id_maps = id_maps | precomputed_id_maps
@@ -229,9 +224,7 @@ def __init__(
                 n_classes=n_classes,
                 col_types=col_types,
             )
-            data = _apply_reserved_mask_column(
-                data, data_columns, col_types, self.reserved_mask_column
-            )
+            data = _apply_mask_column(data, data_columns, col_types, self.mask_column)
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
@@ -297,7 +290,7 @@ def __init__(
                 data_columns = [
                     col
                     for col in col_types.keys()
-                    if col not in _reserved_input_columns(self.reserved_mask_column)
+                    if col not in _reserved_input_columns(self.mask_column)
                 ]
 
                 # We still need to find the files to process
@@ -474,13 +467,11 @@ def _get_column_metadata_across_files(
                         read_format,
                         selected_columns,
                         max_rows_inner,
-                        self.reserved_mask_column,
+                        self.mask_column,
                     )
 
                     # Get columns for the current file (excluding metadata cols)
-                    current_file_cols = _get_data_columns(
-                        data, self.reserved_mask_column
-                    )
+                    current_file_cols = _get_data_columns(data, self.mask_column)
 
                     if col_types is None:
                         data_columns = current_file_cols
@@ -520,7 +511,7 @@ def _get_column_metadata_across_files(
                         selected_columns_statistics,
                         n_rows_running_count,
                         precomputed_id_maps,
-                        self.reserved_mask_column,
+                        self.mask_column,
                     )
                     n_rows_running_count += data.shape[0]
 
@@ -607,7 +598,7 @@ def _process_batches_multiple_files(
         write_format: str,
         process_by_file: bool = True,
         subsequence_start_mode: str = "distribute",
-        reserved_mask_column: Optional[str] = None,
+        mask_column: Optional[str] = None,
     ) -> None:
         """Processes batches of data from multiple files.
 
@@ -629,10 +620,10 @@ def _process_batches_multiple_files(
             write_format: The file format for the output files.
             process_by_file: A flag to indicate if processing should be done file by file.
             subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-            reserved_mask_column: Optional input column used to mask all data columns.
+            mask_column: Optional input column used to mask all data columns.
         """
-        if reserved_mask_column is None:
-            reserved_mask_column = self.reserved_mask_column
+        if mask_column is None:
+            mask_column = self.mask_column
 
         if process_by_file:
             _process_batches_multiple_files_inner(
@@ -660,7 +651,7 @@ def _process_batches_multiple_files(
                 merge_output=self.merge_output,
                 continue_preprocessing=self.continue_preprocessing,
                 subsequence_start_mode=subsequence_start_mode,
-                reserved_mask_column=reserved_mask_column,
+                mask_column=mask_column,
             )
             input_files = create_file_paths_for_multiple_files2(
                 self.project_root,
@@ -705,7 +696,7 @@ def _process_batches_multiple_files(
                 "merge_output": self.merge_output,
                 "continue_preprocessing": self.continue_preprocessing,
                 "subsequence_start_mode": subsequence_start_mode,
-                "reserved_mask_column": reserved_mask_column,
+                "mask_column": mask_column,
             }
 
             job_params = [
@@ -914,20 +905,18 @@ def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> No
 
 
 @beartype
-def _reserved_input_columns(reserved_mask_column: Optional[str]) -> tuple[str, ...]:
-    if reserved_mask_column is None:
+def _reserved_input_columns(mask_column: Optional[str]) -> tuple[str, ...]:
+    if mask_column is None:
         return INPUT_METADATA_COLUMNS
-    return (*INPUT_METADATA_COLUMNS, reserved_mask_column)
+    return (*INPUT_METADATA_COLUMNS, mask_column)
 
 
 @beartype
 def _get_data_columns(
-    data: pl.DataFrame, reserved_mask_column: Optional[str] = None
+    data: pl.DataFrame, mask_column: Optional[str] = None
 ) -> list[str]:
     return [
-        col
-        for col in data.columns
-        if col not in _reserved_input_columns(reserved_mask_column)
+        col for col in data.columns if col not in _reserved_input_columns(mask_column)
     ]
 
 
@@ -941,24 +930,20 @@ def _selected_columns_with_optional_mask(
     data_path: str,
     read_format: str,
     selected_columns: Optional[list[str]],
-    reserved_mask_column: Optional[str] = None,
+    mask_column: Optional[str] = None,
 ) -> Optional[list[str]]:
-    if (
-        selected_columns is None
-        or read_format != "parquet"
-        or reserved_mask_column is None
-    ):
+    if selected_columns is None or read_format != "parquet" or mask_column is None:
         return selected_columns
 
     schema_columns = pq.read_schema(data_path).names
-    if reserved_mask_column in schema_columns:
-        return selected_columns + [reserved_mask_column]
+    if mask_column in schema_columns:
+        return selected_columns + [mask_column]
     return selected_columns
 
 
 @beartype
-def _mask_column_expr(mask_dtype: Any, reserved_mask_column: str) -> pl.Expr:
-    mask_col = pl.col(reserved_mask_column)
+def _mask_column_expr(mask_dtype: Any, mask_column: str) -> pl.Expr:
+    mask_col = pl.col(mask_column)
 
     if isinstance(mask_dtype, pl.Boolean):
         return mask_col
@@ -981,26 +966,24 @@ def _mask_column_expr(mask_dtype: Any, reserved_mask_column: str) -> pl.Expr:
         return mask_col == 1
 
     if isinstance(mask_dtype, (pl.String, pl.Utf8)):
-        return mask_col.str.to_lowercase().is_in(["1", "true"])
+        return mask_col.str.to_lowercase().is_in(["1"])
 
     raise ValueError(
-        f"Column {reserved_mask_column} must be boolean, numeric, or string, got {mask_dtype}"
+        f"Column {mask_column} must be boolean, numeric, or string, got {mask_dtype}"
     )
 
 
 @beartype
-def _apply_reserved_mask_column(
+def _apply_mask_column(
     data: pl.DataFrame,
     data_columns: list[str],
     col_types: dict[str, str],
-    reserved_mask_column: Optional[str] = None,
+    mask_column: Optional[str] = None,
 ) -> pl.DataFrame:
-    if reserved_mask_column is None or reserved_mask_column not in data.columns:
+    if mask_column is None or mask_column not in data.columns:
         return data
 
-    mask_expr = _mask_column_expr(
-        data.schema[reserved_mask_column], reserved_mask_column
-    )
+    mask_expr = _mask_column_expr(data.schema[mask_column], mask_column)
     updates = []
     for col in data_columns:
         mask_value = (
@@ -1017,7 +1000,7 @@ def _apply_reserved_mask_column(
     if updates:
         data = data.with_columns(updates)
 
-    return data.drop(reserved_mask_column)
+    return data.drop(mask_column)
 
 
 @beartype
@@ -1146,7 +1129,7 @@ def _get_column_statistics(
     selected_columns_statistics: dict[str, dict[str, float]],
     n_rows_running_count: int,
     precomputed_id_maps: dict[str, dict[Union[str, int], int]],
-    reserved_mask_column: Optional[str] = None,
+    mask_column: Optional[str] = None,
 ) -> tuple[
     dict[str, dict[Union[str, int], int]],
     dict[str, dict[str, float]],
@@ -1183,10 +1166,8 @@ def _get_column_statistics(
         ValueError: If a column has an unsupported data type (neither
             string, integer, nor float).
     """
-    if reserved_mask_column is not None and reserved_mask_column in data.columns:
-        mask_expr = _mask_column_expr(
-            data.schema[reserved_mask_column], reserved_mask_column
-        )
+    if mask_column is not None and mask_column in data.columns:
+        mask_expr = _mask_column_expr(data.schema[mask_column], mask_column)
         data = data.filter(~mask_expr)
 
     for data_col in data_columns:
@@ -1242,7 +1223,7 @@ def _load_and_preprocess_data(
     read_format: str,
     selected_columns: Optional[list[str]],
     max_rows: Optional[int],
-    reserved_mask_column: Optional[str] = None,
+    mask_column: Optional[str] = None,
 ) -> pl.DataFrame:
     """Loads data from a file and performs initial preparation.
 
@@ -1266,7 +1247,7 @@ def _load_and_preprocess_data(
     """
     logger.info(f"Reading data from '{data_path}'...")
     columns_to_read = _selected_columns_with_optional_mask(
-        data_path, read_format, selected_columns, reserved_mask_column
+        data_path, read_format, selected_columns, mask_column
     )
     data = read_data(data_path, read_format, columns=columns_to_read)
 
@@ -1278,8 +1259,8 @@ def _load_and_preprocess_data(
             col for col in selected_columns if col not in INPUT_METADATA_COLUMNS
         ]
         columns_to_select = list(INPUT_METADATA_COLUMNS) + selected_columns_filtered
-        if reserved_mask_column is not None and reserved_mask_column in data.columns:
-            columns_to_select.append(reserved_mask_column)
+        if mask_column is not None and mask_column in data.columns:
+            columns_to_select.append(mask_column)
         data = data.select(_deduplicate_columns(columns_to_select))
 
     if max_rows:
@@ -1364,7 +1345,7 @@ def _process_batches_multiple_files_inner(
     merge_output: bool,
     continue_preprocessing: bool,
     subsequence_start_mode: str,
-    reserved_mask_column: Optional[str],
+    mask_column: Optional[str],
 ):
     """Inner function for processing batches of data from multiple files.
 
@@ -1430,7 +1411,7 @@ def _process_batches_multiple_files_inner(
                             read_format,
                             selected_columns,
                             max_rows_inner,
-                            reserved_mask_column,
+                            mask_column,
                         )
                         n_rows_running_count += data.shape[0]
                     continue
@@ -1440,7 +1421,7 @@ def _process_batches_multiple_files_inner(
                 read_format,
                 selected_columns,
                 max_rows_inner,
-                reserved_mask_column,
+                mask_column,
             )
             data, _, _ = _apply_column_statistics(
                 data,
@@ -1450,9 +1431,7 @@ def _process_batches_multiple_files_inner(
                 n_classes,
                 col_types,
             )
-            data = _apply_reserved_mask_column(
-                data, data_columns, col_types, reserved_mask_column
-            )
+            data = _apply_mask_column(data, data_columns, col_types, mask_column)
 
             data_name_root_inner = f"{data_name_root}-{process_id}-{file_index_str}"
 
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
index 2c0b797f..11b3853c 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
@@ -12,4 +12,4 @@ stride_by_split:
 - 1
 max_rows: null
 metadata_config_path: configs/metadata_configs/test-data-categorical-multitarget-5.json
-reserved_mask_column: "[mask]"
+mask_column: "[mask]"
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index e7ac8483..b57bb27a 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -10,7 +10,7 @@
 from sequifier.preprocess import (
     RESERVED_MASK_COLUMN,
     Preprocessor,
-    _apply_reserved_mask_column,
+    _apply_mask_column,
     _get_column_statistics,
     _get_data_columns,
     create_id_map,
@@ -141,7 +141,7 @@ def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
     )
 
 
-def test_reserved_mask_column_is_not_a_data_column():
+def test_mask_column_is_not_a_data_column():
     data = pl.DataFrame(
         {
             "sequenceId": [1],
@@ -154,7 +154,7 @@ def test_reserved_mask_column_is_not_a_data_column():
     assert _get_data_columns(data, RESERVED_MASK_COLUMN) == ["itemId"]
 
 
-def test_apply_reserved_mask_column_replaces_values_and_drops_column():
+def test_apply_mask_column_replaces_values_and_drops_column():
     data = pl.DataFrame(
         {
             "cat_col": [3, 4, 5, 6],
@@ -163,7 +163,7 @@ def test_apply_reserved_mask_column_replaces_values_and_drops_column():
         }
     )
 
-    masked = _apply_reserved_mask_column(
+    masked = _apply_mask_column(
         data,
         ["cat_col", "num_col"],
         {"cat_col": "Int64", "num_col": "Float64"},
@@ -175,7 +175,7 @@ def test_apply_reserved_mask_column_replaces_values_and_drops_column():
     assert masked["num_col"].to_list() == [-1.0, 0.0, 3.5, 0.0]
 
 
-def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
+def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
     project_root = tmp_path / "project"
     project_root.mkdir()
     metadata_dir = project_root / "configs" / "metadata_configs"
@@ -224,7 +224,7 @@ def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
         subsequence_start_mode="distribute",
         use_precomputed_maps=None,
         metadata_config_path=metadata_config_path,
-        reserved_mask_column=RESERVED_MASK_COLUMN,
+        mask_column=RESERVED_MASK_COLUMN,
     )
 
     output = pl.read_parquet(project_root / "data" / "masked-input-split0.parquet")
@@ -249,7 +249,7 @@ def test_preprocessor_applies_reserved_mask_column_end_to_end(tmp_path):
     assert RESERVED_MASK_COLUMN not in metadata["selected_columns_statistics"]
 
 
-def test_preprocessor_requires_metadata_config_for_reserved_mask_column(tmp_path):
+def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
     project_root = tmp_path / "project"
     project_root.mkdir()
     data_path = tmp_path / "masked-input.csv"
@@ -264,7 +264,7 @@ def test_preprocessor_requires_metadata_config_for_reserved_mask_column(tmp_path
 
     with pytest.raises(
         ValueError,
-        match="metadata_config_path must be set when reserved_mask_column is set",
+        match="metadata_config_path must be set when mask_column is set",
     ):
         Preprocessor(
             project_root=str(project_root),
@@ -285,11 +285,11 @@ def test_preprocessor_requires_metadata_config_for_reserved_mask_column(tmp_path
             subsequence_start_mode="distribute",
             use_precomputed_maps=None,
             metadata_config_path=None,
-            reserved_mask_column=RESERVED_MASK_COLUMN,
+            mask_column=RESERVED_MASK_COLUMN,
         )
 
 
-def test_preprocessor_config_defaults_reserved_mask_column_to_none(tmp_path):
+def test_preprocessor_config_defaults_mask_column_to_none(tmp_path):
     data_path = tmp_path / "input.csv"
     pl.DataFrame(
         {
@@ -308,10 +308,10 @@ def test_preprocessor_config_defaults_reserved_mask_column_to_none(tmp_path):
         seed=1010,
     )
 
-    assert config.reserved_mask_column is None
+    assert config.mask_column is None
 
 
-def test_preprocessor_config_requires_metadata_config_for_reserved_mask_column(
+def test_preprocessor_config_requires_metadata_config_for_mask_column(
     tmp_path,
 ):
     data_path = tmp_path / "input.csv"
@@ -326,7 +326,7 @@ def test_preprocessor_config_requires_metadata_config_for_reserved_mask_column(
 
     with pytest.raises(
         ValidationError,
-        match="metadata_config_path must be set when reserved_mask_column is set",
+        match="metadata_config_path must be set when mask_column is set",
     ):
         PreprocessorModel(
             project_root=str(tmp_path),
@@ -334,7 +334,7 @@ def test_preprocessor_config_requires_metadata_config_for_reserved_mask_column(
             split_ratios=[1.0],
             seq_length=1,
             seed=1010,
-            reserved_mask_column=RESERVED_MASK_COLUMN,
+            mask_column=RESERVED_MASK_COLUMN,
         )
 
 

From 95fa63bd3442d781b8649fdf658412a130dd7103 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 8 Jun 2026 18:08:33 +0200
Subject: [PATCH 17/81] Bump version to 2.0.0.0

---
 README.md                          | 2 +-
 docs/source/conf.py                | 2 +-
 documentation/consolidated-docs.md | 2 +-
 pyproject.toml                     | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index 2cbb9d99..85f64440 100644
--- a/README.md
+++ b/README.md
@@ -199,7 +199,7 @@ Please cite with:
   title = {sequifier - causal transformer models for multivariate sequence modelling},
   year = {2025},
   publisher = {GitHub},
-  version = {v1.2.0.0},
+  version = {v2.0.0.0},
   url = {[https://github.com/0xideas/sequifier](https://github.com/0xideas/sequifier)}
 }
 
diff --git a/docs/source/conf.py b/docs/source/conf.py
index 91f64673..b0b88606 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -15,7 +15,7 @@
 project = 'sequifier'
 copyright = '2025, Leon Luithlen'
 author = 'Leon Luithlen'
-release = 'v1.2.0.0'
+release = 'v2.0.0.0'
 html_baseurl = 'https://www.sequifier.com/'
 
 # -- General configuration ---------------------------------------------------
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index b1563d6a..657cf59c 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -199,7 +199,7 @@ Please cite with:
   title = {sequifier - causal transformer models for multivariate sequence modelling},
   year = {2025},
   publisher = {GitHub},
-  version = {v1.2.0.0},
+  version = {v2.0.0.0},
   url = {[https://github.com/0xideas/sequifier](https://github.com/0xideas/sequifier)}
 }
 
diff --git a/pyproject.toml b/pyproject.toml
index 7a561108..b6550e5c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sequifier"
-version = "v1.2.0.0"
+version = "v2.0.0.0"
 authors = [
     { name = "Leon Luithlen", email = "leontimnaluithlen@gmail.com" },
 ]

From ea71ce029eba4fb479f3bb6a3c9fde67da1018f7 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 11:55:09 +0200
Subject: [PATCH 18/81] fix subsequence_starts problems

---
 src/sequifier/preprocess.py   |  9 ++++-----
 tests/unit/test_preprocess.py | 10 +++++-----
 2 files changed, 9 insertions(+), 10 deletions(-)

diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index a1b6125e..a90d28bc 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -23,7 +23,6 @@
     write_data,
 )
 
-RESERVED_MASK_COLUMN = "[mask]"
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
 CATEGORICAL_MASK_ID = 2
 REAL_MASK_VALUE = 0.0
@@ -2054,7 +2053,7 @@ def extract_sequences(
     for in_row in raw_sequences.iter_rows(named=True):
         in_seq_lists_only = {col: in_row[col] for col in columns}
 
-        subsequences, left_pad_lengths = extract_subsequences(
+        subsequences, left_pad_lengths, subsequence_starts = extract_subsequences(
             in_seq_lists_only,
             seq_length,
             stride_for_split,
@@ -2067,7 +2066,7 @@ def extract_sequences(
                 row = [
                     in_row["sequenceId"],
                     subsequence_id,
-                    subsequence_id * stride_for_split,
+                    int(subsequence_starts[subsequence_id]) - left_pad_lengths[subsequence_id],
                     left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
@@ -2142,7 +2141,7 @@ def extract_subsequences(
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
-) -> tuple[dict[str, list[list[Union[float, int]]]], list[int]]:
+) -> tuple[dict[str, list[list[Union[float, int]]]], list[int], np.ndarray]:
     """Extracts subsequences from a dictionary of sequence lists.
 
     This function takes a dictionary `in_seq` where keys are column
@@ -2186,7 +2185,7 @@ def extract_subsequences(
     }
     left_pad_lengths = [pad_len] * len(subsequence_starts)
 
-    return result, left_pad_lengths
+    return result, left_pad_lengths, subsequence_starts
 
 
 @beartype
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index b57bb27a..75fbb50d 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -36,7 +36,7 @@ def test_extract_subsequences_basic():
     # Expected behavior: Window size is seq_length + 1 (history + target)
     # Windows: [10,11,12,13], [11,12,13,14], [12,13,14,15]
 
-    result, left_pad_lengths = extract_subsequences(
+    result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data, seq_length, stride, columns, subsequence_start_mode="distribute"
     )
 
@@ -54,7 +54,7 @@ def test_extract_subsequences_padding():
 
     # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
 
-    result, left_pad_lengths = extract_subsequences(
+    result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data, seq_length, stride, columns, subsequence_start_mode="distribute"
     )
 
@@ -64,7 +64,7 @@ def test_extract_subsequences_padding():
 
 def test_extract_subsequences_returns_left_pad_lengths_when_requested():
     input_data = {"col1": [0.0, 1.5]}
-    result, left_pad_lengths = extract_subsequences(
+    result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
         seq_length=4,
         stride_for_split=1,
@@ -351,7 +351,7 @@ def test_extract_subsequences_modes(mode):
     if mode == "distribute":
         stride = 4
         # distribute might adjust indices to maximize coverage
-        result, left_pad_lengths = extract_subsequences(
+        result, left_pad_lengths, subsequence_starts = extract_subsequences(
             input_data, seq_length, stride, columns, mode
         )
         assert len(result["col1"]) > 0
@@ -367,7 +367,7 @@ def test_extract_subsequences_modes(mode):
         # Testing a passing exact case
         # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
         input_data_exact = {"col1": list(range(11))}
-        result, left_pad_lengths = extract_subsequences(
+        result, left_pad_lengths, subsequence_starts = extract_subsequences(
             input_data_exact, seq_length, 4, columns, mode
         )
         assert len(result["col1"]) > 0

From 3b3ee9610349dc174c25316e597d8110a05dc484 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 14:54:40 +0200
Subject: [PATCH 19/81] Add hyperparameter search for bert vars

---
 .gitignore                                    |   2 +-
 .../config/hyperparameter_search_config.py    | 148 +++++++++++++----
 src/sequifier/io/yaml.py                      |   2 +
 tests/unit/data/empty.parquet                 |   0
 .../unit/test_hyperparameter_search_config.py | 149 ++++++++++++++++++
 tests/unit/test_infer.py                      |   2 +-
 tests/unit/test_preprocess.py                 |   4 +-
 7 files changed, 268 insertions(+), 39 deletions(-)
 create mode 100644 tests/unit/data/empty.parquet
 create mode 100644 tests/unit/test_hyperparameter_search_config.py

diff --git a/.gitignore b/.gitignore
index d2bfad92..3a23070a 100644
--- a/.gitignore
+++ b/.gitignore
@@ -5,7 +5,7 @@ logs/
 metadata_configs/
 outputs/
 project_folder/
-
+!tests/unit/data
 *\~
 *.DS_Store
 
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 0659abf6..ee75b1fb 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -8,9 +8,12 @@
 from loguru import logger
 from pydantic import BaseModel, ConfigDict, Field, field_validator, model_validator
 
+from sequifier.config.probabilities import ProbabilityDistribution
 from sequifier.config.train_config import (
+    BERTSpecModel,
     DotDict,
     ModelSpecModel,
+    ReplacementDistribution,
     TrainingSpecModel,
     TrainModel,
 )
@@ -70,6 +73,66 @@ def validate_step_and_log(self):
 OptunaInt = Union[list[int], IntDistribution]
 
 
+def sample_param(
+    trial: Any,
+    name: str,
+    space: Union[list, FloatDistribution, IntDistribution],
+):
+    if isinstance(space, list):
+        return trial.suggest_categorical(name, space)
+    if isinstance(space, FloatDistribution):
+        return trial.suggest_float(
+            name, space.low, space.high, step=space.step, log=space.log
+        )
+    if isinstance(space, IntDistribution):
+        return trial.suggest_int(
+            name, space.low, space.high, step=space.step, log=space.log
+        )
+    raise TypeError(f"Unsupported hyperparameter search space for {name}: {space}")
+
+
+class BERTSpecHyperparameterSampling(BaseModel):
+    """Pydantic model for BERT objective hyperparameter sampling.
+
+    Each BERT spec field is sampled independently so Optuna can track the
+    conditional search space without receiving nested objects as categorical
+    values.
+    """
+
+    model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
+
+    masking_probability: OptunaFloat
+    replacement_distribution: list[ReplacementDistribution]
+    span_masking: list[ProbabilityDistribution]
+
+    def sample_trial(self, trial: Any) -> BERTSpecModel:
+        masking_probability = sample_param(
+            trial, "bert_masking_probability", self.masking_probability
+        )
+        replacement_distribution_index = trial.suggest_categorical(
+            "bert_replacement_distribution_index",
+            list(range(len(self.replacement_distribution))),
+        )
+        span_masking_index = trial.suggest_categorical(
+            "bert_span_masking_index", list(range(len(self.span_masking)))
+        )
+
+        replacement_distribution = self.replacement_distribution[
+            replacement_distribution_index
+        ].model_copy(deep=True)
+        span_masking = self.span_masking[span_masking_index].model_copy(deep=True)
+
+        logger.info(
+            f"{masking_probability = } - {replacement_distribution = } - {span_masking = }"
+        )
+
+        return BERTSpecModel(
+            masking_probability=masking_probability,
+            replacement_distribution=replacement_distribution,
+            span_masking=span_masking,
+        )
+
+
 @beartype
 def load_hyperparameter_search_config(
     config_path: str, skip_metadata: bool
@@ -165,8 +228,10 @@ class TrainingSpecHyperparameterSampling(BaseModel):
         save_batch_interval_minutes: the time interval in which a checkpoint is written to a unique checkpoint path
         save_batch_interval_minutes_val_loss: calculate val loss at the moment of batch interval saving
         calculate_validation_loss_on_initialization: calculate val loss on weight initialization
+        training_objective: Training objective choices, either 'causal' or 'bert'.
         batch_size: A list of possible batch sizes.
         learning_rate: A list of possible learning rates.
+        bert_spec: Optional BERT hyperparameter search space. Required if 'bert' can be sampled.
         criterion: A dictionary mapping target columns to loss functions.
         class_weights: Optional dictionary mapping columns to class weights.
         accumulation_steps: A list of possible gradient accumulation steps.
@@ -198,8 +263,10 @@ class TrainingSpecHyperparameterSampling(BaseModel):
     save_batch_interval_minutes_val_loss: bool = True
     calculate_validation_loss_on_initialization: bool = False
 
+    training_objective: list[str] = Field(default_factory=lambda: ["causal"])
     batch_size: OptunaInt
     learning_rate: list[float]  # Kept as list to preserve coupling with epochs
+    bert_spec: Optional[BERTSpecHyperparameterSampling] = None
     criterion: dict[str, str]
     class_weights: Optional[dict[str, list[float]]] = None
     accumulation_steps: OptunaInt
@@ -259,6 +326,32 @@ def __init__(self, **kwargs):
             DotDict(scheduler_config) for scheduler_config in kwargs["scheduler"]
         ]
 
+    @field_validator("training_objective", mode="before")
+    @classmethod
+    def normalize_training_objective(cls, v):
+        if isinstance(v, str):
+            return [v]
+        return v
+
+    @field_validator("training_objective")
+    @classmethod
+    def validate_training_objective(cls, v):
+        allowed = {"causal", "bert"}
+        invalid = set(v).difference(allowed)
+        if invalid:
+            raise ValueError(
+                f"Only 'causal' and 'bert' are allowed, found {invalid}"
+            )
+        return v
+
+    @model_validator(mode="after")
+    def validate_bert_spec(self):
+        if "bert" in self.training_objective and self.bert_spec is None:
+            raise ValueError(
+                "If 'bert' is in training_objective, bert_spec must be configured."
+            )
+        return self
+
     @field_validator("layer_type_dtypes")
     @classmethod
     def validate_layer_type_dtypes(cls, v):
@@ -343,29 +436,27 @@ def sample_trial(self, trial: Any) -> TrainingSpecModel:
         )
         optimizer = self.optimizer[opt_index]
 
-        def sample_param(
-            name: str, space: Union[list, FloatDistribution, IntDistribution]
-        ):
-            if isinstance(space, list):
-                return trial.suggest_categorical(name, space)
-            elif isinstance(space, FloatDistribution):
-                return trial.suggest_float(
-                    name, space.low, space.high, step=space.step, log=space.log
-                )
-            elif isinstance(space, IntDistribution):
-                return trial.suggest_int(
-                    name, space.low, space.high, step=space.step, log=space.log
-                )
+        training_objective = trial.suggest_categorical(
+            "training_objective", self.training_objective
+        )
+        bert_spec = (
+            self.bert_spec.sample_trial(trial)
+            if training_objective == "bert" and self.bert_spec is not None
+            else None
+        )
 
-        batch_size = sample_param("batch_size", self.batch_size)
-        dropout = sample_param("dropout", self.dropout)
-        accumulation_steps = sample_param("accumulation_steps", self.accumulation_steps)
+        batch_size = sample_param(trial, "batch_size", self.batch_size)
+        dropout = sample_param(trial, "dropout", self.dropout)
+        accumulation_steps = sample_param(
+            trial, "accumulation_steps", self.accumulation_steps
+        )
 
         logger.info(
-            f"{learning_rate = } - {batch_size = } - {dropout = } - {optimizer = }"
+            f"{training_objective = } - {learning_rate = } - {batch_size = } - {dropout = } - {optimizer = }"
         )
 
         return TrainingSpecModel(
+            training_objective=training_objective,
             device=self.device,
             epochs=epochs,
             log_interval=self.log_interval,
@@ -380,6 +471,7 @@ def sample_param(
             learning_rate=learning_rate,
             criterion=self.criterion,
             class_weights=self.class_weights,
+            bert_spec=bert_spec,
             accumulation_steps=accumulation_steps,
             dropout=dropout,
             loss_weights=self.loss_weights,
@@ -498,23 +590,11 @@ def sample_trial(self, trial: Any) -> ModelSpecModel:
             else self.feature_embedding_dims[dim_model_idx]
         )
 
-        def sample_param(
-            name: str, space: Union[list, FloatDistribution, IntDistribution]
-        ):
-            if isinstance(space, list):
-                return trial.suggest_categorical(name, space)
-            elif isinstance(space, FloatDistribution):
-                return trial.suggest_float(
-                    name, space.low, space.high, step=space.step, log=space.log
-                )
-            elif isinstance(space, IntDistribution):
-                return trial.suggest_int(
-                    name, space.low, space.high, step=space.step, log=space.log
-                )
-
-        dim_feedforward = sample_param("dim_feedforward", self.dim_feedforward)
-        num_layers = sample_param("num_layers", self.num_layers)
-        rope_theta = sample_param("rope_theta", self.rope_theta)
+        dim_feedforward = sample_param(
+            trial, "dim_feedforward", self.dim_feedforward
+        )
+        num_layers = sample_param(trial, "num_layers", self.num_layers)
+        rope_theta = sample_param(trial, "rope_theta", self.rope_theta)
 
         activation_fn = trial.suggest_categorical("activation_fn", self.activation_fn)
         normalization = trial.suggest_categorical("normalization", self.normalization)
diff --git a/src/sequifier/io/yaml.py b/src/sequifier/io/yaml.py
index c275050a..f8cdcd63 100644
--- a/src/sequifier/io/yaml.py
+++ b/src/sequifier/io/yaml.py
@@ -1,5 +1,6 @@
 import numpy
 import yaml
+from pydantic import BaseModel
 
 from sequifier.config.train_config import (
     DotDict,
@@ -77,6 +78,7 @@ def increase_indent(self, flow=False, indentless=False):
 TrainModelDumper.add_representer(TrainModel, represent_sequifier_object)
 TrainModelDumper.add_representer(ModelSpecModel, represent_sequifier_object)
 TrainModelDumper.add_representer(TrainingSpecModel, represent_sequifier_object)
+TrainModelDumper.add_multi_representer(BaseModel, represent_sequifier_object)
 TrainModelDumper.add_representer(DotDict, represent_dot_dict)
 TrainModelDumper.add_representer(numpy.float64, represent_numpy_float)
 TrainModelDumper.add_representer(
diff --git a/tests/unit/data/empty.parquet b/tests/unit/data/empty.parquet
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/unit/test_hyperparameter_search_config.py b/tests/unit/test_hyperparameter_search_config.py
new file mode 100644
index 00000000..b09eed26
--- /dev/null
+++ b/tests/unit/test_hyperparameter_search_config.py
@@ -0,0 +1,149 @@
+import pytest
+import yaml
+from pydantic import ValidationError
+
+from sequifier.config.hyperparameter_search_config import (
+    TrainingSpecHyperparameterSampling,
+)
+from sequifier.io.yaml import TrainModelDumper
+
+
+class RecordingTrial:
+    def __init__(self, overrides=None):
+        self.overrides = overrides or {}
+        self.params = {}
+
+    def suggest_categorical(self, name, choices):
+        value = self.overrides.get(name, choices[0])
+        assert value in choices
+        self.params[name] = value
+        return value
+
+    def suggest_float(self, name, low, high, step=None, log=False):
+        value = self.overrides.get(name, low)
+        self.params[name] = value
+        return value
+
+    def suggest_int(self, name, low, high, step=1, log=False):
+        value = self.overrides.get(name, low)
+        self.params[name] = value
+        return value
+
+
+def make_training_sampling(**overrides):
+    config = {
+        "device": "cpu",
+        "epochs": [1],
+        "save_interval_epochs": 1,
+        "batch_size": [2],
+        "learning_rate": [0.001],
+        "criterion": {"itemId": "CrossEntropyLoss"},
+        "accumulation_steps": [1],
+        "dropout": [0.0],
+        "optimizer": [{"name": "Adam"}],
+        "scheduler": [{"name": "StepLR", "step_size": 1, "gamma": 0.99}],
+        "continue_training": False,
+    }
+    config.update(overrides)
+    return TrainingSpecHyperparameterSampling(**config)
+
+
+def bert_spec_sampling_config():
+    return {
+        "masking_probability": [0.15, 0.30],
+        "replacement_distribution": [
+            {"masked": 0.8, "random": 0.1, "identical": 0.1},
+            {"masked": 0.6, "random": 0.2, "identical": 0.2},
+        ],
+        "span_masking": [
+            {"type": "GeometricDistribution", "p": 1.0},
+            {"type": "PoissonDistributionFloor", "rate": 2.0},
+        ],
+    }
+
+
+def test_training_objective_defaults_to_causal_without_bert_spec():
+    sampling = make_training_sampling()
+    trial = RecordingTrial()
+
+    training_spec = sampling.sample_trial(trial)
+
+    assert training_spec.training_objective == "causal"
+    assert training_spec.bert_spec is None
+    assert trial.params["training_objective"] == "causal"
+
+
+def test_bert_spec_fields_are_sampled_separately_for_bert_objective():
+    sampling = make_training_sampling(
+        training_objective=["causal", "bert"],
+        bert_spec=bert_spec_sampling_config(),
+    )
+    trial = RecordingTrial(
+        {
+            "training_objective": "bert",
+            "bert_masking_probability": 0.30,
+            "bert_replacement_distribution_index": 1,
+            "bert_span_masking_index": 1,
+        }
+    )
+
+    training_spec = sampling.sample_trial(trial)
+
+    assert training_spec.training_objective == "bert"
+    assert training_spec.bert_spec.masking_probability == 0.30
+    assert training_spec.bert_spec.replacement_distribution.masked == 0.6
+    assert training_spec.bert_spec.replacement_distribution.random == 0.2
+    assert training_spec.bert_spec.replacement_distribution.identical == 0.2
+    assert training_spec.bert_spec.span_masking.type == "PoissonDistributionFloor"
+    assert set(trial.params).issuperset(
+        {
+            "training_objective",
+            "bert_masking_probability",
+            "bert_replacement_distribution_index",
+            "bert_span_masking_index",
+        }
+    )
+
+
+def test_causal_objective_does_not_sample_bert_fields():
+    sampling = make_training_sampling(
+        training_objective=["causal", "bert"],
+        bert_spec=bert_spec_sampling_config(),
+    )
+    trial = RecordingTrial({"training_objective": "causal"})
+
+    training_spec = sampling.sample_trial(trial)
+
+    assert training_spec.training_objective == "causal"
+    assert training_spec.bert_spec is None
+    assert "bert_masking_probability" not in trial.params
+    assert "bert_replacement_distribution_index" not in trial.params
+    assert "bert_span_masking_index" not in trial.params
+
+
+def test_bert_objective_requires_bert_spec_sampling_config():
+    with pytest.raises(ValidationError, match="bert_spec"):
+        make_training_sampling(training_objective=["bert"])
+
+
+def test_bert_training_spec_dumps_to_plain_yaml():
+    sampling = make_training_sampling(
+        training_objective=["bert"],
+        bert_spec=bert_spec_sampling_config(),
+    )
+    training_spec = sampling.sample_trial(RecordingTrial())
+
+    dumped = yaml.dump(training_spec, Dumper=TrainModelDumper, sort_keys=False)
+    loaded = yaml.safe_load(dumped)
+
+    assert loaded["training_objective"] == "bert"
+    assert loaded["bert_spec"]["masking_probability"] == 0.15
+    assert loaded["bert_spec"]["replacement_distribution"] == {
+        "masked": 0.8,
+        "random": 0.1,
+        "identical": 0.1,
+    }
+    assert loaded["bert_spec"]["span_masking"] == {
+        "type": "GeometricDistribution",
+        "p": 1.0,
+    }
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index ecb3628b..998dbab8 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -164,7 +164,7 @@ def ar_config():
         model_path="dummy.onnx",
         model_type="generative",
         training_objective="causal",
-        data_path="tests/unit/data/dummy.parquet",
+        data_path="tests/unit/data/empty.parquet",
         input_columns=["target_col"],
         categorical_columns=[],
         real_columns=["target_col"],
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 75fbb50d..481abc3d 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -8,7 +8,6 @@
 
 from sequifier.config.preprocess_config import PreprocessorModel
 from sequifier.preprocess import (
-    RESERVED_MASK_COLUMN,
     Preprocessor,
     _apply_mask_column,
     _get_column_statistics,
@@ -24,8 +23,7 @@
 # ==========================================
 # 1. Test Sequence Extraction (Sliding Window)
 # ==========================================
-
-
+RESERVED_MASK_COLUMN = "[mask]"
 def test_extract_subsequences_basic():
     """Tests basic sliding window extraction with sufficient length."""
     input_data = {"col1": [10, 11, 12, 13, 14, 15]}

From dbc355a82d48c88f0c3e2b7c75026d88cea4a4d8 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 14:56:09 +0200
Subject: [PATCH 20/81] Add hyperparameter search for bert vars

---
 pyrightconfig.json                                   | 2 +-
 src/sequifier/config/hyperparameter_search_config.py | 8 ++------
 src/sequifier/preprocess.py                          | 3 ++-
 tests/unit/test_preprocess.py                        | 2 ++
 4 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/pyrightconfig.json b/pyrightconfig.json
index c2481558..c31d3b15 100644
--- a/pyrightconfig.json
+++ b/pyrightconfig.json
@@ -1,4 +1,4 @@
 {
     "reportMissingImports": false,
     "reportMissingModuleSource": false
-}
\ No newline at end of file
+}
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index ee75b1fb..cec47b2f 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -339,9 +339,7 @@ def validate_training_objective(cls, v):
         allowed = {"causal", "bert"}
         invalid = set(v).difference(allowed)
         if invalid:
-            raise ValueError(
-                f"Only 'causal' and 'bert' are allowed, found {invalid}"
-            )
+            raise ValueError(f"Only 'causal' and 'bert' are allowed, found {invalid}")
         return v
 
     @model_validator(mode="after")
@@ -590,9 +588,7 @@ def sample_trial(self, trial: Any) -> ModelSpecModel:
             else self.feature_embedding_dims[dim_model_idx]
         )
 
-        dim_feedforward = sample_param(
-            trial, "dim_feedforward", self.dim_feedforward
-        )
+        dim_feedforward = sample_param(trial, "dim_feedforward", self.dim_feedforward)
         num_layers = sample_param(trial, "num_layers", self.num_layers)
         rope_theta = sample_param(trial, "rope_theta", self.rope_theta)
 
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index a90d28bc..a9401479 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -2066,7 +2066,8 @@ def extract_sequences(
                 row = [
                     in_row["sequenceId"],
                     subsequence_id,
-                    int(subsequence_starts[subsequence_id]) - left_pad_lengths[subsequence_id],
+                    int(subsequence_starts[subsequence_id])
+                    - left_pad_lengths[subsequence_id],
                     left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 481abc3d..f478c3a8 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -24,6 +24,8 @@
 # 1. Test Sequence Extraction (Sliding Window)
 # ==========================================
 RESERVED_MASK_COLUMN = "[mask]"
+
+
 def test_extract_subsequences_basic():
     """Tests basic sliding window extraction with sufficient length."""
     input_data = {"col1": [10, 11, 12, 13, 14, 15]}

From c10f9eafdd03d918301464428ae9f62efcec716b Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 15:08:57 +0200
Subject: [PATCH 21/81] Update itemPosition in test outputs

---
 ...orical-1-best-embedding-3-0-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-1-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-2-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-3-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-4-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-5-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-6-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-7-embeddings.csv |   2 +-
 ...inf-size-best-embedding-3-0-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-1-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-2-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-3-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-4-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-5-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-6-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-7-embeddings.csv |   6 +-
 ...-1-best-3-autoregression-0-predictions.csv |   8 +-
 ...-1-best-3-autoregression-1-predictions.csv |   8 +-
 ...-1-best-3-autoregression-2-predictions.csv |  12 +-
 ...-1-best-3-autoregression-3-predictions.csv |  22 +-
 ...-1-best-3-autoregression-4-predictions.csv |  12 +-
 ...-1-best-3-autoregression-5-predictions.csv |  18 +-
 ...-1-best-3-autoregression-6-predictions.csv |  20 +-
 ...-1-best-3-autoregression-7-predictions.csv |  10 +-
 ...del-categorical-1-best-3-0-predictions.csv |   2 +-
 ...del-categorical-1-best-3-1-predictions.csv |   2 +-
 ...del-categorical-1-best-3-2-predictions.csv |   2 +-
 ...del-categorical-1-best-3-3-predictions.csv |   4 +-
 ...del-categorical-1-best-3-4-predictions.csv |   2 +-
 ...del-categorical-1-best-3-5-predictions.csv |   4 +-
 ...del-categorical-1-best-3-6-predictions.csv |   2 +-
 ...del-categorical-1-best-3-7-predictions.csv |   2 +-
 ...orical-1-inf-size-best-3-0-predictions.csv |   6 +-
 ...orical-1-inf-size-best-3-1-predictions.csv |   6 +-
 ...orical-1-inf-size-best-3-2-predictions.csv |   6 +-
 ...orical-1-inf-size-best-3-3-predictions.csv |  12 +-
 ...orical-1-inf-size-best-3-4-predictions.csv |   6 +-
 ...orical-1-inf-size-best-3-5-predictions.csv |  12 +-
 ...orical-1-inf-size-best-3-6-predictions.csv |   6 +-
 ...orical-1-inf-size-best-3-7-predictions.csv |   6 +-
 ...del-categorical-3-best-3-0-predictions.csv |   2 +-
 ...del-categorical-3-best-3-1-predictions.csv |   2 +-
 ...del-categorical-3-best-3-2-predictions.csv |   2 +-
 ...del-categorical-3-best-3-3-predictions.csv |   4 +-
 ...del-categorical-3-best-3-4-predictions.csv |   2 +-
 ...del-categorical-3-best-3-5-predictions.csv |   4 +-
 ...del-categorical-3-best-3-6-predictions.csv |   2 +-
 ...del-categorical-3-best-3-7-predictions.csv |   2 +-
 ...orical-3-inf-size-best-3-0-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-1-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-2-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-3-predictions.csv |  12 +-
 ...orical-3-inf-size-best-3-4-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-5-predictions.csv |  12 +-
 ...orical-3-inf-size-best-3-6-predictions.csv |   6 +-
 ...orical-3-inf-size-best-3-7-predictions.csv |   6 +-
 ...del-categorical-5-best-3-0-predictions.csv |   2 +-
 ...del-categorical-5-best-3-1-predictions.csv |   2 +-
 ...del-categorical-5-best-3-2-predictions.csv |   2 +-
 ...del-categorical-5-best-3-3-predictions.csv |   4 +-
 ...del-categorical-5-best-3-4-predictions.csv |   2 +-
 ...del-categorical-5-best-3-5-predictions.csv |   4 +-
 ...del-categorical-5-best-3-6-predictions.csv |   2 +-
 ...del-categorical-5-best-3-7-predictions.csv |   2 +-
 ...el-categorical-50-best-3-0-predictions.csv |   2 +-
 ...el-categorical-50-best-3-1-predictions.csv |   2 +-
 ...el-categorical-50-best-3-2-predictions.csv |   2 +-
 ...el-categorical-50-best-3-3-predictions.csv |   4 +-
 ...el-categorical-50-best-3-4-predictions.csv |   2 +-
 ...el-categorical-50-best-3-5-predictions.csv |   4 +-
 ...el-categorical-50-best-3-6-predictions.csv |   2 +-
 ...el-categorical-50-best-3-7-predictions.csv |   2 +-
 ...rical-distributed-best-3-0-predictions.csv |   2 +-
 ...rical-distributed-best-3-1-predictions.csv |   2 +-
 ...rical-distributed-best-3-2-predictions.csv |   2 +-
 ...rical-distributed-best-3-3-predictions.csv |   4 +-
 ...rical-distributed-best-3-4-predictions.csv |   2 +-
 ...rical-distributed-best-3-5-predictions.csv |   4 +-
 ...rical-distributed-best-3-6-predictions.csv |   2 +-
 ...rical-distributed-best-3-7-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-0-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-1-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-2-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-3-predictions.csv |   4 +-
 ...-categorical-lazy-best-3-4-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-5-predictions.csv |   4 +-
 ...-categorical-lazy-best-3-6-predictions.csv |   2 +-
 ...-categorical-lazy-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-7-predictions.csv |   2 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...uifier-model-real-1-best-3-predictions.csv |   4 +-
 ...uifier-model-real-3-best-3-predictions.csv |   4 +-
 ...uifier-model-real-5-best-3-predictions.csv |   4 +-
 ...ifier-model-real-50-best-3-predictions.csv |   4 +-
 ...custom-eval-run-0-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-0-predictions.csv |  28 +-
 ...custom-eval-run-3-best-1-1-predictions.csv |  28 +-
 ...custom-eval-run-3-best-1-2-predictions.csv |  20 +-
 ...custom-eval-run-3-best-1-3-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-4-predictions.csv |  46 +-
 ...custom-eval-run-3-best-1-5-predictions.csv |  54 +--
 ...custom-eval-run-3-best-1-6-predictions.csv |  24 +-
 ...custom-eval-run-3-best-1-7-predictions.csv |  30 +-
 141 files changed, 1488 insertions(+), 1488 deletions(-)

diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
index 69fa9fda..b6602347 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,7,0.129906,-0.5716,0.18384615,0.48382017,-0.95800894,-0.06879192,-1.5230852,0.32024476,-0.032266557,-0.2324689,1.2026423,1.2646347,-1.6631967,0.8923384,1.0081543,2.146518,1.260503,-0.024123436,-0.4261092,0.71811444,0.6665925,-0.83603954,0.7274643,0.081591725,-0.46056688,-0.7607759,-0.5974563,0.5428709,0.04133952,0.6946457,0.54585093,-1.3706633,0.9921805,-1.1147716,-1.1846336,-0.6535198,1.6928166,-0.6169712,1.1777917,-0.33404595,-1.2148309,0.16786538,1.4557567,-0.7030484,0.40415078,-2.0602548,-0.44774872,-0.9443908,-0.37846982,-1.4340259,-1.0351791,-0.54817986,-1.3410672,1.057497,0.6667623,1.8221252,-1.484842,1.5296896,0.2569867,-2.118917,0.27047974,-0.802363,-0.34508198,-0.7845012,0.29995131,-1.5465921,0.3353841,-0.85972494,0.9633959,0.1057641,0.77337104,1.4520825,-2.3267972,0.8137759,-0.9976677,0.25013983,-0.27935335,0.8346441,-0.33092296,-1.0873228,0.49370176,-0.4727684,-0.7608272,-0.8022482,1.3239352,-2.2216249,0.55208087,-1.1424916,1.1881708,-0.4595066,-0.09527821,0.9202227,0.57056504,-2.0961475,1.643413,-1.046783,-0.116270036,-0.3729071,-0.61354154,-0.8894295,-1.1429787,-0.78733057,0.57443416,1.4698808,0.43613648,-0.075808406,-0.14003257,-2.9505506,-0.016151994,0.6177055,0.68450046,0.5607075,0.87257403,-1.3993292,0.35658643,-0.0063092983,-0.25653496,-0.31633425,-3.5138404,0.88777286,1.6519107,0.40932092,-0.027439872,-0.79843307,-1.1049063,0.07724614,0.56875324,-1.0056511,0.48330793,-0.157245,-1.6178644,0.1934022,-1.9760464,0.0589674,0.5110818,0.089602806,-0.968479,-1.0943018,-0.7229412,1.4727569,0.47599703,-0.6288432,-2.055541,-1.011629,0.9921429,0.5582884,0.50531685,-0.097561255,0.731814,0.6783084,0.3472921,-1.7636684,0.27717423,0.28878227,0.5271399,1.5502092,0.1620326,-0.50224847,1.5970831,-0.6694094,-0.53928703,-1.924573,-0.013257886,-0.10813448,-2.0357018,-1.6700076,0.2103319,1.2394764,-1.1461264,0.036453854,0.40960938,-0.83274007,-0.008255889,-0.30468202,0.5700774,0.6194162,-1.6036439,0.2115395,0.6522949,-0.39073604,0.21115232,-0.46797746,-0.018942848,0.09517513,-0.61526686,0.67094487,-0.45691407,-0.42524102,-0.8716745,0.07637033,-0.72497505,-1.2465461,-0.4429114,1.5433344,-0.8214027,1.8114822,-0.90004414,-0.6661404,0.8775225,0.91526055
+0,0,3,0.129906,-0.5716,0.18384615,0.48382017,-0.95800894,-0.06879192,-1.5230852,0.32024476,-0.032266557,-0.2324689,1.2026423,1.2646347,-1.6631967,0.8923384,1.0081543,2.146518,1.260503,-0.024123436,-0.4261092,0.71811444,0.6665925,-0.83603954,0.7274643,0.081591725,-0.46056688,-0.7607759,-0.5974563,0.5428709,0.04133952,0.6946457,0.54585093,-1.3706633,0.9921805,-1.1147716,-1.1846336,-0.6535198,1.6928166,-0.6169712,1.1777917,-0.33404595,-1.2148309,0.16786538,1.4557567,-0.7030484,0.40415078,-2.0602548,-0.44774872,-0.9443908,-0.37846982,-1.4340259,-1.0351791,-0.54817986,-1.3410672,1.057497,0.6667623,1.8221252,-1.484842,1.5296896,0.2569867,-2.118917,0.27047974,-0.802363,-0.34508198,-0.7845012,0.29995131,-1.5465921,0.3353841,-0.85972494,0.9633959,0.1057641,0.77337104,1.4520825,-2.3267972,0.8137759,-0.9976677,0.25013983,-0.27935335,0.8346441,-0.33092296,-1.0873228,0.49370176,-0.4727684,-0.7608272,-0.8022482,1.3239352,-2.2216249,0.55208087,-1.1424916,1.1881708,-0.4595066,-0.09527821,0.9202227,0.57056504,-2.0961475,1.643413,-1.046783,-0.116270036,-0.3729071,-0.61354154,-0.8894295,-1.1429787,-0.78733057,0.57443416,1.4698808,0.43613648,-0.075808406,-0.14003257,-2.9505506,-0.016151994,0.6177055,0.68450046,0.5607075,0.87257403,-1.3993292,0.35658643,-0.0063092983,-0.25653496,-0.31633425,-3.5138404,0.88777286,1.6519107,0.40932092,-0.027439872,-0.79843307,-1.1049063,0.07724614,0.56875324,-1.0056511,0.48330793,-0.157245,-1.6178644,0.1934022,-1.9760464,0.0589674,0.5110818,0.089602806,-0.968479,-1.0943018,-0.7229412,1.4727569,0.47599703,-0.6288432,-2.055541,-1.011629,0.9921429,0.5582884,0.50531685,-0.097561255,0.731814,0.6783084,0.3472921,-1.7636684,0.27717423,0.28878227,0.5271399,1.5502092,0.1620326,-0.50224847,1.5970831,-0.6694094,-0.53928703,-1.924573,-0.013257886,-0.10813448,-2.0357018,-1.6700076,0.2103319,1.2394764,-1.1461264,0.036453854,0.40960938,-0.83274007,-0.008255889,-0.30468202,0.5700774,0.6194162,-1.6036439,0.2115395,0.6522949,-0.39073604,0.21115232,-0.46797746,-0.018942848,0.09517513,-0.61526686,0.67094487,-0.45691407,-0.42524102,-0.8716745,0.07637033,-0.72497505,-1.2465461,-0.4429114,1.5433344,-0.8214027,1.8114822,-0.90004414,-0.6661404,0.8775225,0.91526055
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
index 31fbd49b..03e75139 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,7,0.6310362,0.334674,0.16728853,0.36186692,-0.9478092,-0.3926634,-1.3003972,-0.044596065,-0.20526434,-0.026398035,0.19380486,0.8118285,-2.0392873,1.2787219,1.2542659,1.4919833,1.088667,-0.010551525,-0.8448257,0.43219802,0.56950766,-0.8026797,0.74732757,-0.42725757,-1.0895562,-0.21274805,-0.7626764,0.026918262,-0.5983657,0.28117272,0.73443246,-0.9515915,0.81785864,-1.6458837,-0.71132445,0.23738682,1.6034894,0.019240104,1.3241774,-0.20347075,-0.7043628,0.38013214,1.2755842,-1.4004871,0.46275398,-1.9880224,-0.9198009,-1.0473716,-0.65675557,-0.84835905,-1.3224105,-0.70712143,-0.7341193,1.1134466,0.6568125,1.3994768,-0.954168,1.3808511,0.251891,-1.9545913,0.2308793,-0.26487717,-0.64339316,-0.2622317,0.603964,-1.4150456,0.6588127,0.08482099,1.1321739,-0.099473916,0.010979951,1.831232,-1.8722236,0.88451487,-0.76108027,0.080021404,-0.06698817,1.07159,-0.9351867,-1.0435191,0.026436348,-0.41959018,-0.52513933,-0.5412578,2.0294697,-2.7480896,0.53006375,-1.0863824,0.33880177,-0.14623298,0.6506472,1.0939751,-0.48882413,-1.810358,1.3499824,-0.89643645,-0.008765998,-0.06027395,-0.48046795,-1.2652203,-1.4560691,-0.946743,-0.27704963,1.9105397,0.20772824,-0.3049073,-0.34610513,-3.3094442,-0.46867853,0.65010065,0.4332434,0.5593448,0.8796511,-0.8775561,0.56195873,0.2979048,0.54411083,-0.49651542,-2.4436908,1.2963794,1.9646289,0.41595295,0.077845775,-0.8488377,-1.2093579,-0.2594095,0.12893897,-0.5105848,0.64491534,-0.017900532,-2.1271238,0.27657434,-1.7817888,-0.6613088,0.2365224,-0.07435218,-1.0712526,-0.80908316,-0.77087396,1.1399101,0.7127807,-0.94809335,-2.0319622,-0.7065952,1.1998774,0.56364894,0.62384087,-0.25050914,-0.12355562,0.011005514,0.4171477,-2.2401168,1.0053111,0.10580753,0.5745275,1.6763166,0.3952029,-0.3734366,2.0440457,-0.5067878,-0.59976774,-2.560222,0.28748652,-0.068856396,-1.7209245,-1.5318912,0.7257471,0.8677974,-0.9443491,0.32327485,0.09680152,-1.0634658,0.2343228,-0.73152024,0.6149891,0.50479555,-1.6433326,0.36874586,1.5886227,-0.48504025,-0.5036868,-0.24184056,-0.20931806,-0.046914995,-0.6832404,0.3294942,0.5522635,-0.77527386,-1.4231143,0.22487257,-0.16668221,-1.7301935,0.025722886,1.6323115,-1.0889417,1.2082901,-0.3673339,0.19397202,0.32195246,1.1962756
+1,0,3,0.6310362,0.334674,0.16728853,0.36186692,-0.9478092,-0.3926634,-1.3003972,-0.044596065,-0.20526434,-0.026398035,0.19380486,0.8118285,-2.0392873,1.2787219,1.2542659,1.4919833,1.088667,-0.010551525,-0.8448257,0.43219802,0.56950766,-0.8026797,0.74732757,-0.42725757,-1.0895562,-0.21274805,-0.7626764,0.026918262,-0.5983657,0.28117272,0.73443246,-0.9515915,0.81785864,-1.6458837,-0.71132445,0.23738682,1.6034894,0.019240104,1.3241774,-0.20347075,-0.7043628,0.38013214,1.2755842,-1.4004871,0.46275398,-1.9880224,-0.9198009,-1.0473716,-0.65675557,-0.84835905,-1.3224105,-0.70712143,-0.7341193,1.1134466,0.6568125,1.3994768,-0.954168,1.3808511,0.251891,-1.9545913,0.2308793,-0.26487717,-0.64339316,-0.2622317,0.603964,-1.4150456,0.6588127,0.08482099,1.1321739,-0.099473916,0.010979951,1.831232,-1.8722236,0.88451487,-0.76108027,0.080021404,-0.06698817,1.07159,-0.9351867,-1.0435191,0.026436348,-0.41959018,-0.52513933,-0.5412578,2.0294697,-2.7480896,0.53006375,-1.0863824,0.33880177,-0.14623298,0.6506472,1.0939751,-0.48882413,-1.810358,1.3499824,-0.89643645,-0.008765998,-0.06027395,-0.48046795,-1.2652203,-1.4560691,-0.946743,-0.27704963,1.9105397,0.20772824,-0.3049073,-0.34610513,-3.3094442,-0.46867853,0.65010065,0.4332434,0.5593448,0.8796511,-0.8775561,0.56195873,0.2979048,0.54411083,-0.49651542,-2.4436908,1.2963794,1.9646289,0.41595295,0.077845775,-0.8488377,-1.2093579,-0.2594095,0.12893897,-0.5105848,0.64491534,-0.017900532,-2.1271238,0.27657434,-1.7817888,-0.6613088,0.2365224,-0.07435218,-1.0712526,-0.80908316,-0.77087396,1.1399101,0.7127807,-0.94809335,-2.0319622,-0.7065952,1.1998774,0.56364894,0.62384087,-0.25050914,-0.12355562,0.011005514,0.4171477,-2.2401168,1.0053111,0.10580753,0.5745275,1.6763166,0.3952029,-0.3734366,2.0440457,-0.5067878,-0.59976774,-2.560222,0.28748652,-0.068856396,-1.7209245,-1.5318912,0.7257471,0.8677974,-0.9443491,0.32327485,0.09680152,-1.0634658,0.2343228,-0.73152024,0.6149891,0.50479555,-1.6433326,0.36874586,1.5886227,-0.48504025,-0.5036868,-0.24184056,-0.20931806,-0.046914995,-0.6832404,0.3294942,0.5522635,-0.77527386,-1.4231143,0.22487257,-0.16668221,-1.7301935,0.025722886,1.6323115,-1.0889417,1.2082901,-0.3673339,0.19397202,0.32195246,1.1962756
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
index c085fdb1..ae9e5907 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,7,0.14896812,-0.7527326,0.098616265,1.5360665,-0.83212125,-0.8661572,0.87993956,0.7707245,0.8990906,0.35979807,1.5049816,0.9981651,-2.2045355,0.46189427,1.3522851,0.81136656,-0.3787586,-0.41562492,-1.6569216,0.9181141,0.8893558,-0.6614972,0.35581687,1.0705252,0.61813873,-1.6933419,-1.6523572,0.4721115,-0.6415743,-0.8052874,-1.1905364,-1.3483986,0.7520653,-1.4446422,0.032763932,-0.77299255,0.8698881,-0.9012206,-0.105312526,-0.03295452,-1.7241203,0.40647608,1.3450831,0.00073365227,0.23433386,-0.9771891,-0.1076481,1.2662536,-0.7311288,-0.6664411,0.006385311,-0.31623417,0.108964175,0.43619278,0.284846,0.34632137,0.13046862,1.7500304,-0.31573698,-2.1822793,-0.073760144,-0.63938767,-2.4171443,-1.3825034,1.4659554,0.061223067,-2.0762868,-1.1751351,-0.081808135,-1.6416469,0.61133504,1.4318693,-1.5806884,1.4651628,-0.42543516,0.27195463,-1.9691073,0.5124442,0.7785031,-1.2194265,-1.2140163,-0.21277447,-0.64433247,-0.34592688,1.480311,-1.6828288,0.5201631,-0.45541105,-0.17800705,-0.29835147,0.06779409,0.17683165,-0.5690362,-0.4795994,0.24317117,0.7038561,0.47528592,0.38569337,-0.92215085,0.12652376,-0.5222174,0.79115415,0.1062228,1.2142566,0.57600933,0.3640244,1.4705232,-1.8675983,-0.151888,0.84477943,0.5522926,0.25622973,0.50753284,-0.87316334,-1.7905984,-0.8826149,-0.80747306,1.2090178,-2.0752993,-0.18883346,0.35716313,0.12954345,-0.5863821,-0.3407679,-0.8663318,0.8249107,1.2127439,-1.0448643,-0.53810304,-0.6317458,-0.50353926,0.4781922,-0.91658026,-0.21077591,0.49530765,0.66140443,-0.061057545,-0.011890421,-1.4830853,-0.5170973,1.2860987,-0.48179677,-1.2840593,-0.12392463,-0.5612262,0.5283732,1.6894128,-0.015182488,1.6194154,0.6165103,0.846802,-1.259089,-0.19867773,0.05881237,0.87158996,1.7141718,-2.1852305,-0.17253563,1.7481165,0.7287171,-0.9333205,-0.5228718,-1.1484395,-1.0014697,-1.5860218,-0.9670557,0.5662043,-0.6358826,0.6285491,1.4611297,-1.5053117,-0.73541224,-1.1782357,0.4759477,0.56338716,0.29015008,-0.5925466,0.4408311,-0.23582521,-1.1164836,1.0048937,-0.39400414,-0.7058568,0.99132824,-1.3044466,2.0163782,-1.8661258,0.013690879,-0.98518753,-0.2840235,-1.4385029,-0.70052123,-1.5944667,1.7985247,0.9840921,1.857999,-0.74875927,1.5471251,0.73758185,0.14302142
+2,0,2,0.14896812,-0.7527326,0.098616265,1.5360665,-0.83212125,-0.8661572,0.87993956,0.7707245,0.8990906,0.35979807,1.5049816,0.9981651,-2.2045355,0.46189427,1.3522851,0.81136656,-0.3787586,-0.41562492,-1.6569216,0.9181141,0.8893558,-0.6614972,0.35581687,1.0705252,0.61813873,-1.6933419,-1.6523572,0.4721115,-0.6415743,-0.8052874,-1.1905364,-1.3483986,0.7520653,-1.4446422,0.032763932,-0.77299255,0.8698881,-0.9012206,-0.105312526,-0.03295452,-1.7241203,0.40647608,1.3450831,0.00073365227,0.23433386,-0.9771891,-0.1076481,1.2662536,-0.7311288,-0.6664411,0.006385311,-0.31623417,0.108964175,0.43619278,0.284846,0.34632137,0.13046862,1.7500304,-0.31573698,-2.1822793,-0.073760144,-0.63938767,-2.4171443,-1.3825034,1.4659554,0.061223067,-2.0762868,-1.1751351,-0.081808135,-1.6416469,0.61133504,1.4318693,-1.5806884,1.4651628,-0.42543516,0.27195463,-1.9691073,0.5124442,0.7785031,-1.2194265,-1.2140163,-0.21277447,-0.64433247,-0.34592688,1.480311,-1.6828288,0.5201631,-0.45541105,-0.17800705,-0.29835147,0.06779409,0.17683165,-0.5690362,-0.4795994,0.24317117,0.7038561,0.47528592,0.38569337,-0.92215085,0.12652376,-0.5222174,0.79115415,0.1062228,1.2142566,0.57600933,0.3640244,1.4705232,-1.8675983,-0.151888,0.84477943,0.5522926,0.25622973,0.50753284,-0.87316334,-1.7905984,-0.8826149,-0.80747306,1.2090178,-2.0752993,-0.18883346,0.35716313,0.12954345,-0.5863821,-0.3407679,-0.8663318,0.8249107,1.2127439,-1.0448643,-0.53810304,-0.6317458,-0.50353926,0.4781922,-0.91658026,-0.21077591,0.49530765,0.66140443,-0.061057545,-0.011890421,-1.4830853,-0.5170973,1.2860987,-0.48179677,-1.2840593,-0.12392463,-0.5612262,0.5283732,1.6894128,-0.015182488,1.6194154,0.6165103,0.846802,-1.259089,-0.19867773,0.05881237,0.87158996,1.7141718,-2.1852305,-0.17253563,1.7481165,0.7287171,-0.9333205,-0.5228718,-1.1484395,-1.0014697,-1.5860218,-0.9670557,0.5662043,-0.6358826,0.6285491,1.4611297,-1.5053117,-0.73541224,-1.1782357,0.4759477,0.56338716,0.29015008,-0.5925466,0.4408311,-0.23582521,-1.1164836,1.0048937,-0.39400414,-0.7058568,0.99132824,-1.3044466,2.0163782,-1.8661258,0.013690879,-0.98518753,-0.2840235,-1.4385029,-0.70052123,-1.5944667,1.7985247,0.9840921,1.857999,-0.74875927,1.5471251,0.73758185,0.14302142
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
index ecec6901..26935d23 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,7,0.3853767,-0.7623611,0.5445376,1.1489445,-0.93219095,-1.0841124,-0.27961808,0.19042599,0.06699983,0.6020505,0.7307122,0.9754021,-2.318475,0.97368747,1.6034338,0.94340336,1.2762036,-0.30378067,-0.97008324,0.62717986,1.3900626,-0.9464742,0.4837369,-0.030570975,-0.4290407,-1.1276956,-0.7923079,0.563961,-0.40978292,-0.23345281,-0.27136502,-0.8576837,0.69629353,-1.6003033,0.11132374,-0.5791753,1.9399358,-0.92576635,0.53490555,-0.49527624,-1.3456327,-0.176038,1.349483,-1.1282389,0.20037875,-0.955993,-0.8822724,-0.07519414,-0.77170074,-0.9350284,-0.5532876,-0.94019765,-0.10448791,1.1707093,0.6575927,1.3083346,-0.6112351,1.6713382,0.16979486,-2.0898654,0.00984007,-0.44183394,-1.1921827,-1.2657421,1.6432866,-1.130562,-0.46437404,-0.8430221,0.7453206,-0.65720564,0.35852155,1.6237744,-2.0383017,1.1427602,-0.8917945,0.20761068,-1.3489536,0.62074554,-0.44713894,-1.8621219,-0.5244267,-0.45581624,-0.39755735,-0.43458134,1.8005577,-2.0100255,0.3588848,-1.2393866,0.95991385,-0.65878445,0.741961,0.43640068,-0.37811747,-1.3537853,1.0124421,0.12569562,0.45669925,-0.062139705,-0.6899904,-0.7770923,-1.2683299,-0.42139253,-0.4123218,1.7224118,0.38184428,-0.5022623,0.1969878,-2.748319,-0.18802647,1.3159317,0.4502178,-0.26388726,0.5703611,-0.85277075,-0.46127146,0.23827219,-0.4298961,0.41151482,-2.6187694,0.5646321,1.1139392,-0.032928437,0.1021941,-0.6601091,-1.3054987,0.09395422,0.43201578,-1.4673975,0.08933542,-0.71540993,-1.6057761,0.4075759,-1.6277498,0.05748504,-0.18920064,0.1056629,-0.8918468,-0.17913246,-1.4436157,0.077475116,0.6109399,-0.83478075,-1.3056538,-0.87130284,0.28275952,0.87232274,0.8705629,-0.14861424,0.91572547,0.6586051,0.6749446,-1.7701945,0.1629751,0.4728509,1.0638493,1.7032028,-0.6542733,-0.017882407,2.1696196,-0.047487322,-0.77344674,-2.259672,-1.1815661,-0.5823015,-1.638453,-1.8893389,0.9176421,0.32815483,0.4039307,0.17879207,-0.32735565,-1.1835169,-1.0626798,-0.12413608,-0.13082603,0.5998312,-1.7939893,0.5362146,1.2261001,-1.4384091,0.4796257,-0.45061567,-0.100991376,0.51506233,-1.2538948,1.3633755,-0.58895797,-0.21253212,-1.3333261,0.50792605,-0.97876984,-1.5163478,-0.6666502,1.8272516,-1.4814882,2.3199954,-0.6350161,0.1867251,0.83364874,0.9136821
-4,0,7,1.8327771,0.9652215,-0.25314748,0.21621384,1.9031243,0.7036974,-0.930624,-0.8883423,0.55273986,0.5570372,1.0132973,-0.96464247,-0.51950616,-0.05579906,-0.37479147,0.44072837,0.16671647,0.5790387,0.41375738,0.80011296,-1.5359617,-0.22593148,1.7854865,-0.71542114,-1.0921632,-0.041372035,-0.21795599,-0.3369114,-1.2671196,-0.7473329,0.18677177,-1.8774186,2.0318227,0.2733341,-0.9697609,-0.58609706,-0.36415994,0.22731349,0.4721085,-1.3540337,-0.25599056,0.65862226,1.5556433,-0.81543595,1.168724,-2.2140863,0.42116505,0.47876942,-0.6478898,0.44233415,0.20755047,-0.5501464,-0.5956206,-0.60861623,-1.3033564,-0.9219007,1.9484233,-0.02818083,0.9555862,-0.6107661,1.3185904,0.4262676,-1.1561952,-0.5161904,-0.8352084,0.22098036,-0.32627368,-1.1956955,1.7369198,-1.1766452,0.07778335,-1.4228516,-2.164931,0.61946803,-0.4467436,-0.01088463,-0.43634295,1.737324,0.6609657,-0.20005353,-0.56503403,0.29499987,0.11585043,-0.05404717,1.7766248,-1.624335,0.4819137,1.0377663,-1.2305183,0.8004889,-1.9219416,1.6361125,0.13642022,-1.2401025,0.56233984,-0.15820779,1.5966985,-0.25904042,-0.5514964,0.43703973,0.4001095,-0.12492117,1.6866364,2.0976322,-0.49593046,0.8581702,0.2179227,-0.6346643,-1.7050065,-1.6924847,-0.08375938,-0.38175428,-0.32693657,-0.9640046,-0.9026502,-0.5941344,0.19496141,-0.8319495,-0.55289805,-0.80228716,1.650307,1.2991756,-1.0835283,0.39504835,0.29286614,-0.13108593,-0.38219267,-0.59767234,-1.3788447,-0.26507112,0.34214637,1.0462471,1.5181249,-0.59406966,0.58952206,2.7725644,-1.0286809,1.18978,0.3369609,-1.0014182,-0.02982089,-0.5898692,-0.99967504,0.01042568,1.4441023,-0.87508506,0.57642156,2.5007522,-1.5390786,-1.6458069,-1.7197639,-0.74723464,1.1642295,-0.31345525,-0.0055794707,1.5905838,-0.9592406,-0.4174035,-0.6339656,0.8530592,1.2996758,-0.11439483,0.3914311,0.46125937,-3.0907335,-0.27514952,0.20107323,-1.4204425,-1.7638282,-0.48746726,-1.2546072,-0.2341572,-0.10110028,-0.24766739,1.7387944,-0.7589158,0.75030494,0.5836463,0.10834081,0.7989048,-0.4846295,-0.62584215,0.5027704,1.1866003,-0.28132582,-0.80156577,0.44897035,1.1347668,0.07279313,-0.61313444,0.8892234,0.71706593,0.65837866,0.33229896,1.174505,-0.7435528,0.5451965,0.36238024,1.4138215,0.37747207
+3,0,2,0.3853767,-0.7623611,0.5445376,1.1489445,-0.93219095,-1.0841124,-0.27961808,0.19042599,0.06699983,0.6020505,0.7307122,0.9754021,-2.318475,0.97368747,1.6034338,0.94340336,1.2762036,-0.30378067,-0.97008324,0.62717986,1.3900626,-0.9464742,0.4837369,-0.030570975,-0.4290407,-1.1276956,-0.7923079,0.563961,-0.40978292,-0.23345281,-0.27136502,-0.8576837,0.69629353,-1.6003033,0.11132374,-0.5791753,1.9399358,-0.92576635,0.53490555,-0.49527624,-1.3456327,-0.176038,1.349483,-1.1282389,0.20037875,-0.955993,-0.8822724,-0.07519414,-0.77170074,-0.9350284,-0.5532876,-0.94019765,-0.10448791,1.1707093,0.6575927,1.3083346,-0.6112351,1.6713382,0.16979486,-2.0898654,0.00984007,-0.44183394,-1.1921827,-1.2657421,1.6432866,-1.130562,-0.46437404,-0.8430221,0.7453206,-0.65720564,0.35852155,1.6237744,-2.0383017,1.1427602,-0.8917945,0.20761068,-1.3489536,0.62074554,-0.44713894,-1.8621219,-0.5244267,-0.45581624,-0.39755735,-0.43458134,1.8005577,-2.0100255,0.3588848,-1.2393866,0.95991385,-0.65878445,0.741961,0.43640068,-0.37811747,-1.3537853,1.0124421,0.12569562,0.45669925,-0.062139705,-0.6899904,-0.7770923,-1.2683299,-0.42139253,-0.4123218,1.7224118,0.38184428,-0.5022623,0.1969878,-2.748319,-0.18802647,1.3159317,0.4502178,-0.26388726,0.5703611,-0.85277075,-0.46127146,0.23827219,-0.4298961,0.41151482,-2.6187694,0.5646321,1.1139392,-0.032928437,0.1021941,-0.6601091,-1.3054987,0.09395422,0.43201578,-1.4673975,0.08933542,-0.71540993,-1.6057761,0.4075759,-1.6277498,0.05748504,-0.18920064,0.1056629,-0.8918468,-0.17913246,-1.4436157,0.077475116,0.6109399,-0.83478075,-1.3056538,-0.87130284,0.28275952,0.87232274,0.8705629,-0.14861424,0.91572547,0.6586051,0.6749446,-1.7701945,0.1629751,0.4728509,1.0638493,1.7032028,-0.6542733,-0.017882407,2.1696196,-0.047487322,-0.77344674,-2.259672,-1.1815661,-0.5823015,-1.638453,-1.8893389,0.9176421,0.32815483,0.4039307,0.17879207,-0.32735565,-1.1835169,-1.0626798,-0.12413608,-0.13082603,0.5998312,-1.7939893,0.5362146,1.2261001,-1.4384091,0.4796257,-0.45061567,-0.100991376,0.51506233,-1.2538948,1.3633755,-0.58895797,-0.21253212,-1.3333261,0.50792605,-0.97876984,-1.5163478,-0.6666502,1.8272516,-1.4814882,2.3199954,-0.6350161,0.1867251,0.83364874,0.9136821
+4,0,2,1.8327771,0.9652215,-0.25314748,0.21621384,1.9031243,0.7036974,-0.930624,-0.8883423,0.55273986,0.5570372,1.0132973,-0.96464247,-0.51950616,-0.05579906,-0.37479147,0.44072837,0.16671647,0.5790387,0.41375738,0.80011296,-1.5359617,-0.22593148,1.7854865,-0.71542114,-1.0921632,-0.041372035,-0.21795599,-0.3369114,-1.2671196,-0.7473329,0.18677177,-1.8774186,2.0318227,0.2733341,-0.9697609,-0.58609706,-0.36415994,0.22731349,0.4721085,-1.3540337,-0.25599056,0.65862226,1.5556433,-0.81543595,1.168724,-2.2140863,0.42116505,0.47876942,-0.6478898,0.44233415,0.20755047,-0.5501464,-0.5956206,-0.60861623,-1.3033564,-0.9219007,1.9484233,-0.02818083,0.9555862,-0.6107661,1.3185904,0.4262676,-1.1561952,-0.5161904,-0.8352084,0.22098036,-0.32627368,-1.1956955,1.7369198,-1.1766452,0.07778335,-1.4228516,-2.164931,0.61946803,-0.4467436,-0.01088463,-0.43634295,1.737324,0.6609657,-0.20005353,-0.56503403,0.29499987,0.11585043,-0.05404717,1.7766248,-1.624335,0.4819137,1.0377663,-1.2305183,0.8004889,-1.9219416,1.6361125,0.13642022,-1.2401025,0.56233984,-0.15820779,1.5966985,-0.25904042,-0.5514964,0.43703973,0.4001095,-0.12492117,1.6866364,2.0976322,-0.49593046,0.8581702,0.2179227,-0.6346643,-1.7050065,-1.6924847,-0.08375938,-0.38175428,-0.32693657,-0.9640046,-0.9026502,-0.5941344,0.19496141,-0.8319495,-0.55289805,-0.80228716,1.650307,1.2991756,-1.0835283,0.39504835,0.29286614,-0.13108593,-0.38219267,-0.59767234,-1.3788447,-0.26507112,0.34214637,1.0462471,1.5181249,-0.59406966,0.58952206,2.7725644,-1.0286809,1.18978,0.3369609,-1.0014182,-0.02982089,-0.5898692,-0.99967504,0.01042568,1.4441023,-0.87508506,0.57642156,2.5007522,-1.5390786,-1.6458069,-1.7197639,-0.74723464,1.1642295,-0.31345525,-0.0055794707,1.5905838,-0.9592406,-0.4174035,-0.6339656,0.8530592,1.2996758,-0.11439483,0.3914311,0.46125937,-3.0907335,-0.27514952,0.20107323,-1.4204425,-1.7638282,-0.48746726,-1.2546072,-0.2341572,-0.10110028,-0.24766739,1.7387944,-0.7589158,0.75030494,0.5836463,0.10834081,0.7989048,-0.4846295,-0.62584215,0.5027704,1.1866003,-0.28132582,-0.80156577,0.44897035,1.1347668,0.07279313,-0.61313444,0.8892234,0.71706593,0.65837866,0.33229896,1.174505,-0.7435528,0.5451965,0.36238024,1.4138215,0.37747207
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
index 5278ebbd..0777c948 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,7,0.8678487,-1.1492083,-0.2957783,0.9786945,-1.9542453,-0.07664241,-0.039730284,-0.09718514,-0.465507,0.088307,0.014937709,1.1039451,-1.9457144,0.7994327,1.8773037,1.5537523,0.39788526,-0.4953133,-1.4370357,1.0003625,0.84108806,-0.99132866,0.15745479,0.4423289,-1.5118519,-1.9748083,-1.3705087,0.11752797,0.04718379,0.7227247,0.15624402,-1.3321223,0.8274153,-0.5020909,-0.8579634,-0.8208209,1.370573,-0.32510355,0.9213625,-0.45679018,-2.0167282,0.77819085,1.3592292,-1.6879606,0.35116172,-1.6433353,-0.8796014,0.47101834,-0.49356297,-1.540319,-0.56203103,-0.74793,0.114850305,0.5929445,1.0806793,0.8520335,-0.28449142,2.3426988,-0.71767795,-2.077216,0.4469343,-0.51118195,-1.8134437,-0.31995004,0.5275096,-0.7924808,-0.90600455,-0.5945146,0.62720716,-0.76678073,1.4406855,0.8350828,-1.3691337,0.43675637,-0.93423057,-0.58701897,-0.53982204,1.3742961,-0.3684079,-1.3804277,-1.3538327,-1.2029366,-0.6361641,-0.2061667,1.1179498,-0.8696951,0.46374846,-0.9316529,0.36905092,0.18406926,0.5726551,-0.097014,-0.3084513,-1.0851082,1.2294934,-0.43448213,0.4714254,0.29762262,-0.66758883,-0.2411063,-0.8684391,-0.15433899,0.3217168,1.7343627,0.6421659,-0.39611214,1.2057977,-3.0774305,0.8041049,1.0713868,0.3196744,0.885012,0.5413774,-1.8904048,-0.6433149,-0.7131405,-0.89828914,0.49162254,-1.7892698,1.266022,1.2143357,0.16972247,-0.4222612,-0.9209112,-1.2874361,-0.30553934,0.78019506,-1.3130422,-0.54638046,-0.06269859,-1.2766076,0.3067186,-1.1979859,-0.5041537,0.24134871,0.9985403,-0.20058915,-0.19003288,-0.9156221,0.73883057,0.60628974,-0.39632013,-1.401826,0.26617125,1.2859261,1.7502289,1.1291437,-0.2593595,1.014177,-0.002415313,-0.0126089025,-1.6622946,0.84452736,0.40896028,0.24172997,0.849172,-1.2459072,0.4409478,2.3070042,-0.20285252,-0.9055849,-1.9905902,-0.6291581,-0.70495003,-1.5236276,-1.6411536,0.28731304,0.23528825,-0.85808456,0.8726922,-0.8224612,-1.3814898,-0.8996288,0.0075938655,1.4490063,0.62249285,-1.3583359,-0.39259228,0.55505204,-1.1764418,1.2858897,-0.5120127,-0.3620197,0.028119635,-1.0583314,0.6261959,-0.93444824,-0.68522364,-0.5574861,0.12772661,-1.1170981,-1.7487799,-0.62091166,1.1936655,-0.55242705,1.6852115,-1.5539129,0.93077433,0.54523623,-0.32935345
+5,0,2,0.8678487,-1.1492083,-0.2957783,0.9786945,-1.9542453,-0.07664241,-0.039730284,-0.09718514,-0.465507,0.088307,0.014937709,1.1039451,-1.9457144,0.7994327,1.8773037,1.5537523,0.39788526,-0.4953133,-1.4370357,1.0003625,0.84108806,-0.99132866,0.15745479,0.4423289,-1.5118519,-1.9748083,-1.3705087,0.11752797,0.04718379,0.7227247,0.15624402,-1.3321223,0.8274153,-0.5020909,-0.8579634,-0.8208209,1.370573,-0.32510355,0.9213625,-0.45679018,-2.0167282,0.77819085,1.3592292,-1.6879606,0.35116172,-1.6433353,-0.8796014,0.47101834,-0.49356297,-1.540319,-0.56203103,-0.74793,0.114850305,0.5929445,1.0806793,0.8520335,-0.28449142,2.3426988,-0.71767795,-2.077216,0.4469343,-0.51118195,-1.8134437,-0.31995004,0.5275096,-0.7924808,-0.90600455,-0.5945146,0.62720716,-0.76678073,1.4406855,0.8350828,-1.3691337,0.43675637,-0.93423057,-0.58701897,-0.53982204,1.3742961,-0.3684079,-1.3804277,-1.3538327,-1.2029366,-0.6361641,-0.2061667,1.1179498,-0.8696951,0.46374846,-0.9316529,0.36905092,0.18406926,0.5726551,-0.097014,-0.3084513,-1.0851082,1.2294934,-0.43448213,0.4714254,0.29762262,-0.66758883,-0.2411063,-0.8684391,-0.15433899,0.3217168,1.7343627,0.6421659,-0.39611214,1.2057977,-3.0774305,0.8041049,1.0713868,0.3196744,0.885012,0.5413774,-1.8904048,-0.6433149,-0.7131405,-0.89828914,0.49162254,-1.7892698,1.266022,1.2143357,0.16972247,-0.4222612,-0.9209112,-1.2874361,-0.30553934,0.78019506,-1.3130422,-0.54638046,-0.06269859,-1.2766076,0.3067186,-1.1979859,-0.5041537,0.24134871,0.9985403,-0.20058915,-0.19003288,-0.9156221,0.73883057,0.60628974,-0.39632013,-1.401826,0.26617125,1.2859261,1.7502289,1.1291437,-0.2593595,1.014177,-0.002415313,-0.0126089025,-1.6622946,0.84452736,0.40896028,0.24172997,0.849172,-1.2459072,0.4409478,2.3070042,-0.20285252,-0.9055849,-1.9905902,-0.6291581,-0.70495003,-1.5236276,-1.6411536,0.28731304,0.23528825,-0.85808456,0.8726922,-0.8224612,-1.3814898,-0.8996288,0.0075938655,1.4490063,0.62249285,-1.3583359,-0.39259228,0.55505204,-1.1764418,1.2858897,-0.5120127,-0.3620197,0.028119635,-1.0583314,0.6261959,-0.93444824,-0.68522364,-0.5574861,0.12772661,-1.1170981,-1.7487799,-0.62091166,1.1936655,-0.55242705,1.6852115,-1.5539129,0.93077433,0.54523623,-0.32935345
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
index 257f845b..79e33b1c 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,7,0.47420236,1.2638448,-0.2942875,0.6063466,0.32506454,-0.011651637,-1.3833891,-1.2344038,-0.45170528,1.0377885,1.8840404,-0.33119497,-1.0696628,0.9726884,0.80814546,0.9354395,0.7167119,0.42845112,-0.015223959,0.81312424,-1.2162951,-1.1050558,0.6366257,-1.6387163,-0.9819105,-0.4953311,0.12158322,-0.67469686,-1.7174512,-0.2108714,1.0931984,-1.6311551,1.6655928,-1.1295462,-0.82908773,-0.27834365,0.35710472,0.7679168,1.2862792,-1.5655477,-0.47575817,0.38371342,0.8257985,-0.6541342,-0.06021277,-1.688667,0.31571636,-1.3439896,-1.3430765,0.0015745121,-0.87187445,-1.2474327,-0.82622105,0.53904265,0.4325758,0.7165552,-0.013134802,1.0078027,1.2916025,-1.6230769,0.1835245,1.4117191,-0.9933948,-0.17660141,-0.85399544,0.1709921,1.0725634,-0.23263937,1.7940853,0.49845877,-0.58278996,0.9100562,-2.1336787,-0.0805108,-0.15371813,0.5364138,0.036808733,0.27225935,0.43129516,-0.5169288,0.8617008,0.37777528,-0.3340564,-0.50667363,2.3349266,-3.2664785,0.23907632,0.13636121,-0.1356525,-0.2975334,-0.56251806,1.5495608,-0.25081053,-2.012842,1.4139545,-0.16621536,0.7844478,0.5935697,-0.85539633,-0.5690532,-0.5415988,-0.8971024,1.8330035,2.7651486,1.039496,-0.23426563,-0.57880294,-1.180363,-1.095461,-0.30837286,0.20563056,-0.45784765,0.799281,-0.585805,0.84253716,0.11244704,0.7201907,-1.2291319,-1.666961,-0.44462907,2.2034056,1.0664692,0.81338817,-0.4559743,-1.2628946,-0.22252318,-1.3892785,-0.74601096,0.09792413,-0.9173109,-1.1128796,0.4023642,-0.7147697,-0.77360827,0.39257395,2.3405876,-0.2756536,0.4053474,-0.8130045,-0.097918354,0.619593,-1.450402,-1.562418,-1.1680082,1.2857003,-1.2578353,0.41585723,1.0423979,-0.53722227,-0.38153183,-0.51949257,-1.5859616,1.3256049,-0.097610995,0.5610497,2.6576302,1.24299,-0.7566359,0.91003966,0.73991305,-0.10965724,-2.0155516,-0.009600935,-0.68587035,-1.7821821,-1.0589889,-0.13436046,0.4220983,-1.1379086,0.48460454,-0.61625963,-0.3349504,0.53863937,0.18685772,0.26678517,-0.20175351,-0.7204558,0.21726036,1.1120167,0.79319596,-0.36998117,-0.6713484,0.72676253,-0.18856229,0.46767554,-1.068567,0.32332298,-0.06708165,-1.2248515,-0.7845498,0.5460865,-0.61680746,1.510683,1.7673895,-0.68418354,0.44412625,0.7936944,-0.8888349,0.03981725,1.6578218
-7,0,7,-0.087797716,-0.72064584,0.3919503,0.30277744,-1.0405561,-0.8779837,-0.7976403,0.071234636,-0.29967022,0.2773496,-0.10566512,0.77108765,-1.7671878,0.824929,1.5373951,1.5010269,1.4846264,-0.7896349,-0.88700837,0.6919234,0.6787521,-0.4330691,0.6862605,-0.2904699,-0.9619546,-0.36133736,-0.9279964,0.18862396,-0.13149612,0.33348706,0.35246968,-0.86275643,0.7647198,-1.5118189,-0.24293631,-0.62797725,1.9434764,-0.15230915,1.2640054,-0.53051674,-1.571733,-0.23882791,1.2569784,-1.5155659,0.31609696,-1.4623228,-1.5276749,-0.63017756,-0.15530244,-1.2590177,-1.1136601,-0.8283698,-0.33160418,1.6386752,0.7145629,1.4172399,-0.9240341,1.155421,-0.20543078,-2.2690709,0.032772336,-0.6074186,-0.8408421,-0.43909267,1.1674483,-1.5343087,0.394567,-0.6319556,0.69497925,-0.17052399,0.66387993,1.6770401,-1.6856111,0.8501759,-1.0100725,-0.06956196,-0.49753228,0.45815924,-0.8491857,-1.5388991,-0.21994229,-0.61590695,-0.43685135,-0.54839456,1.844041,-1.9056602,0.23943688,-0.8901972,1.0178715,-0.4182901,0.47147426,0.28248206,-0.42557982,-1.3693498,1.378879,-0.832486,0.2925802,0.03857666,-0.99315274,-0.88979,-1.3541446,-0.38844082,-0.56667507,2.0802116,0.20410404,-0.5688037,-0.25536713,-2.8470826,0.22910759,1.339881,0.41405326,0.3723613,1.0213392,-0.61211425,0.20765173,0.22666255,-0.052260146,0.20032822,-2.3570492,1.4302993,1.4743407,-0.2883624,0.44256923,-0.7141606,-0.87645143,-0.20459794,0.5313736,-1.1576004,0.2882262,-0.3185675,-2.041501,0.55573565,-1.3157725,0.2728311,0.18704554,-0.34522107,-0.7096739,-0.43224367,-1.601502,1.0732054,0.91898465,-0.68341035,-1.5872612,-0.59699357,0.90725124,1.1939424,0.6279298,-0.49085775,0.9603635,0.668255,0.66679686,-1.698892,0.26019427,0.57732755,0.5378443,1.2200284,0.32803506,-0.15348187,2.141097,-0.014183752,-1.0378197,-2.6077065,-0.3066731,-0.06714382,-1.3892245,-1.7354015,0.53556174,0.9041617,-0.32855996,0.12410693,-0.40722385,-1.3432457,-0.73200125,-0.25400352,0.17272082,1.1034756,-2.2850537,0.6256181,1.7550697,-1.3452938,-0.017986335,-0.54053277,0.05288191,0.28705782,-0.98898697,0.7631376,0.15638618,-0.6830087,-1.3387133,0.62448514,-0.7181843,-1.9862368,0.070950225,1.8409837,-1.5815738,1.639628,-0.8540574,0.19265085,0.4421182,0.67777854
+6,0,3,0.47420236,1.2638448,-0.2942875,0.6063466,0.32506454,-0.011651637,-1.3833891,-1.2344038,-0.45170528,1.0377885,1.8840404,-0.33119497,-1.0696628,0.9726884,0.80814546,0.9354395,0.7167119,0.42845112,-0.015223959,0.81312424,-1.2162951,-1.1050558,0.6366257,-1.6387163,-0.9819105,-0.4953311,0.12158322,-0.67469686,-1.7174512,-0.2108714,1.0931984,-1.6311551,1.6655928,-1.1295462,-0.82908773,-0.27834365,0.35710472,0.7679168,1.2862792,-1.5655477,-0.47575817,0.38371342,0.8257985,-0.6541342,-0.06021277,-1.688667,0.31571636,-1.3439896,-1.3430765,0.0015745121,-0.87187445,-1.2474327,-0.82622105,0.53904265,0.4325758,0.7165552,-0.013134802,1.0078027,1.2916025,-1.6230769,0.1835245,1.4117191,-0.9933948,-0.17660141,-0.85399544,0.1709921,1.0725634,-0.23263937,1.7940853,0.49845877,-0.58278996,0.9100562,-2.1336787,-0.0805108,-0.15371813,0.5364138,0.036808733,0.27225935,0.43129516,-0.5169288,0.8617008,0.37777528,-0.3340564,-0.50667363,2.3349266,-3.2664785,0.23907632,0.13636121,-0.1356525,-0.2975334,-0.56251806,1.5495608,-0.25081053,-2.012842,1.4139545,-0.16621536,0.7844478,0.5935697,-0.85539633,-0.5690532,-0.5415988,-0.8971024,1.8330035,2.7651486,1.039496,-0.23426563,-0.57880294,-1.180363,-1.095461,-0.30837286,0.20563056,-0.45784765,0.799281,-0.585805,0.84253716,0.11244704,0.7201907,-1.2291319,-1.666961,-0.44462907,2.2034056,1.0664692,0.81338817,-0.4559743,-1.2628946,-0.22252318,-1.3892785,-0.74601096,0.09792413,-0.9173109,-1.1128796,0.4023642,-0.7147697,-0.77360827,0.39257395,2.3405876,-0.2756536,0.4053474,-0.8130045,-0.097918354,0.619593,-1.450402,-1.562418,-1.1680082,1.2857003,-1.2578353,0.41585723,1.0423979,-0.53722227,-0.38153183,-0.51949257,-1.5859616,1.3256049,-0.097610995,0.5610497,2.6576302,1.24299,-0.7566359,0.91003966,0.73991305,-0.10965724,-2.0155516,-0.009600935,-0.68587035,-1.7821821,-1.0589889,-0.13436046,0.4220983,-1.1379086,0.48460454,-0.61625963,-0.3349504,0.53863937,0.18685772,0.26678517,-0.20175351,-0.7204558,0.21726036,1.1120167,0.79319596,-0.36998117,-0.6713484,0.72676253,-0.18856229,0.46767554,-1.068567,0.32332298,-0.06708165,-1.2248515,-0.7845498,0.5460865,-0.61680746,1.510683,1.7673895,-0.68418354,0.44412625,0.7936944,-0.8888349,0.03981725,1.6578218
+7,0,2,-0.087797716,-0.72064584,0.3919503,0.30277744,-1.0405561,-0.8779837,-0.7976403,0.071234636,-0.29967022,0.2773496,-0.10566512,0.77108765,-1.7671878,0.824929,1.5373951,1.5010269,1.4846264,-0.7896349,-0.88700837,0.6919234,0.6787521,-0.4330691,0.6862605,-0.2904699,-0.9619546,-0.36133736,-0.9279964,0.18862396,-0.13149612,0.33348706,0.35246968,-0.86275643,0.7647198,-1.5118189,-0.24293631,-0.62797725,1.9434764,-0.15230915,1.2640054,-0.53051674,-1.571733,-0.23882791,1.2569784,-1.5155659,0.31609696,-1.4623228,-1.5276749,-0.63017756,-0.15530244,-1.2590177,-1.1136601,-0.8283698,-0.33160418,1.6386752,0.7145629,1.4172399,-0.9240341,1.155421,-0.20543078,-2.2690709,0.032772336,-0.6074186,-0.8408421,-0.43909267,1.1674483,-1.5343087,0.394567,-0.6319556,0.69497925,-0.17052399,0.66387993,1.6770401,-1.6856111,0.8501759,-1.0100725,-0.06956196,-0.49753228,0.45815924,-0.8491857,-1.5388991,-0.21994229,-0.61590695,-0.43685135,-0.54839456,1.844041,-1.9056602,0.23943688,-0.8901972,1.0178715,-0.4182901,0.47147426,0.28248206,-0.42557982,-1.3693498,1.378879,-0.832486,0.2925802,0.03857666,-0.99315274,-0.88979,-1.3541446,-0.38844082,-0.56667507,2.0802116,0.20410404,-0.5688037,-0.25536713,-2.8470826,0.22910759,1.339881,0.41405326,0.3723613,1.0213392,-0.61211425,0.20765173,0.22666255,-0.052260146,0.20032822,-2.3570492,1.4302993,1.4743407,-0.2883624,0.44256923,-0.7141606,-0.87645143,-0.20459794,0.5313736,-1.1576004,0.2882262,-0.3185675,-2.041501,0.55573565,-1.3157725,0.2728311,0.18704554,-0.34522107,-0.7096739,-0.43224367,-1.601502,1.0732054,0.91898465,-0.68341035,-1.5872612,-0.59699357,0.90725124,1.1939424,0.6279298,-0.49085775,0.9603635,0.668255,0.66679686,-1.698892,0.26019427,0.57732755,0.5378443,1.2200284,0.32803506,-0.15348187,2.141097,-0.014183752,-1.0378197,-2.6077065,-0.3066731,-0.06714382,-1.3892245,-1.7354015,0.53556174,0.9041617,-0.32855996,0.12410693,-0.40722385,-1.3432457,-0.73200125,-0.25400352,0.17272082,1.1034756,-2.2850537,0.6256181,1.7550697,-1.3452938,-0.017986335,-0.54053277,0.05288191,0.28705782,-0.98898697,0.7631376,0.15638618,-0.6830087,-1.3387133,0.62448514,-0.7181843,-1.9862368,0.070950225,1.8409837,-1.5815738,1.639628,-0.8540574,0.19265085,0.4421182,0.67777854
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
index e1edc7ec..902c22b2 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,7,1.6896031,0.2908513,-0.23292185,1.3335884,0.4545921,0.34372616,-0.4165884,-0.42207703,0.60733074,1.2122438,0.8359352,0.5889715,-1.7851549,1.4177665,0.7491191,0.5095494,0.26941696,0.21018945,-0.880237,0.5611264,0.49656537,-0.8069379,1.1553068,-0.0126349665,-0.10190317,-1.7098439,-0.7785349,0.08457157,-1.0872965,-0.46575686,0.17051879,-1.0981392,1.2801948,-0.20283222,-0.37779662,-1.045028,0.8350199,-0.4449086,0.37728134,-1.3267673,-1.3647467,0.49804264,1.072318,-1.5584799,0.62113297,-1.7105299,0.936378,0.96202284,-0.5543043,-0.22153261,0.5113075,-0.5170268,0.11910198,-0.16186874,0.33424348,-0.8031384,1.6559254,0.47857162,0.5240287,-0.85450006,1.1488028,-0.12382944,-2.3143659,-1.7480177,0.34907168,0.3368916,-0.96712863,-1.1980493,1.7140709,-1.4368651,0.66559404,-0.27778557,-1.7404891,1.0585526,-0.3457661,-0.026605887,-1.9635391,2.3253458,0.7187862,-0.6399079,-2.111302,-0.7639529,-0.3719801,-0.0044646794,1.4715577,-1.4121687,0.046070594,0.049432546,-0.67208856,0.77220047,-0.78486633,1.6651013,-0.5048683,-1.0169152,0.19552465,0.6060275,1.0623702,0.15507454,-0.3246709,0.2723275,-0.9548158,-0.060560215,1.6082474,1.8419513,-0.6165263,0.5530547,0.53137124,-1.6408414,-0.083764516,-0.6223381,0.053751897,0.17250697,-0.5837834,-1.9890918,-2.1878116,-0.63507724,-1.3977062,-0.09614903,-0.99611247,-0.8769,0.8914697,1.3250514,-1.3865021,0.070221275,0.14449106,-0.4635786,0.22937016,-1.06633,-1.124088,0.17105861,0.38474694,1.002527,0.062628046,-1.4390727,-0.07946947,2.1103234,-1.7293836,1.1202102,-0.1935348,-1.0497704,-0.1828823,-0.8191538,-0.9199905,-0.0011521083,1.4552091,0.028564112,1.5781827,1.5662959,-0.37651274,-0.75987285,-0.96915853,-1.9078714,1.1331102,-0.114896916,0.53727424,2.130492,-2.0829363,0.2657546,0.7349715,-0.30037367,0.7455469,-1.2468303,-0.94141734,-0.96966463,-2.640029,-1.1223298,0.8763038,-0.96782756,-0.92439544,1.2501682,-1.4089605,-0.8066179,-0.87894547,0.5093231,0.8661666,-1.0657727,-0.44514516,0.16784558,-0.82460415,-0.62026,0.22096872,-0.5744515,-0.05113996,0.9549697,-0.7317686,0.2157382,-0.73088646,1.3564851,-0.56419,-0.3550247,-1.0255395,-0.57220864,-0.9036402,0.36711925,-0.09569928,0.6653445,-0.5859867,1.2525154,1.7926836,0.031228034
+8,0,3,1.6896031,0.2908513,-0.23292185,1.3335884,0.4545921,0.34372616,-0.4165884,-0.42207703,0.60733074,1.2122438,0.8359352,0.5889715,-1.7851549,1.4177665,0.7491191,0.5095494,0.26941696,0.21018945,-0.880237,0.5611264,0.49656537,-0.8069379,1.1553068,-0.0126349665,-0.10190317,-1.7098439,-0.7785349,0.08457157,-1.0872965,-0.46575686,0.17051879,-1.0981392,1.2801948,-0.20283222,-0.37779662,-1.045028,0.8350199,-0.4449086,0.37728134,-1.3267673,-1.3647467,0.49804264,1.072318,-1.5584799,0.62113297,-1.7105299,0.936378,0.96202284,-0.5543043,-0.22153261,0.5113075,-0.5170268,0.11910198,-0.16186874,0.33424348,-0.8031384,1.6559254,0.47857162,0.5240287,-0.85450006,1.1488028,-0.12382944,-2.3143659,-1.7480177,0.34907168,0.3368916,-0.96712863,-1.1980493,1.7140709,-1.4368651,0.66559404,-0.27778557,-1.7404891,1.0585526,-0.3457661,-0.026605887,-1.9635391,2.3253458,0.7187862,-0.6399079,-2.111302,-0.7639529,-0.3719801,-0.0044646794,1.4715577,-1.4121687,0.046070594,0.049432546,-0.67208856,0.77220047,-0.78486633,1.6651013,-0.5048683,-1.0169152,0.19552465,0.6060275,1.0623702,0.15507454,-0.3246709,0.2723275,-0.9548158,-0.060560215,1.6082474,1.8419513,-0.6165263,0.5530547,0.53137124,-1.6408414,-0.083764516,-0.6223381,0.053751897,0.17250697,-0.5837834,-1.9890918,-2.1878116,-0.63507724,-1.3977062,-0.09614903,-0.99611247,-0.8769,0.8914697,1.3250514,-1.3865021,0.070221275,0.14449106,-0.4635786,0.22937016,-1.06633,-1.124088,0.17105861,0.38474694,1.002527,0.062628046,-1.4390727,-0.07946947,2.1103234,-1.7293836,1.1202102,-0.1935348,-1.0497704,-0.1828823,-0.8191538,-0.9199905,-0.0011521083,1.4552091,0.028564112,1.5781827,1.5662959,-0.37651274,-0.75987285,-0.96915853,-1.9078714,1.1331102,-0.114896916,0.53727424,2.130492,-2.0829363,0.2657546,0.7349715,-0.30037367,0.7455469,-1.2468303,-0.94141734,-0.96966463,-2.640029,-1.1223298,0.8763038,-0.96782756,-0.92439544,1.2501682,-1.4089605,-0.8066179,-0.87894547,0.5093231,0.8661666,-1.0657727,-0.44514516,0.16784558,-0.82460415,-0.62026,0.22096872,-0.5744515,-0.05113996,0.9549697,-0.7317686,0.2157382,-0.73088646,1.3564851,-0.56419,-0.3550247,-1.0255395,-0.57220864,-0.9036402,0.36711925,-0.09569928,0.6653445,-0.5859867,1.2525154,1.7926836,0.031228034
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
index 7c1f56ff..f9df3070 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,7,-0.16720563,-0.95600957,0.28606522,0.6061068,-0.3845801,-0.20699826,-1.1377771,0.20950346,-0.13692231,-0.39809802,1.1865551,1.4303648,-1.0548229,0.47550696,0.9458502,2.1207294,2.1566582,-0.24435441,-0.64201504,1.7221576,0.93151706,-0.7352646,0.34909317,0.72475654,-0.53412133,-0.38585177,-0.7807762,0.55336505,0.0011219378,1.0769404,0.178836,-1.3198038,1.3214653,-1.3146976,-0.7111258,-0.48764378,1.6924084,-0.51583934,1.3672884,-0.5570951,-0.8911707,-0.7621178,1.6342849,-1.217147,0.21610765,-2.0004432,-0.9046609,-0.66080266,0.099175125,-1.5156087,-1.408444,-1.4207674,-1.6178051,1.1580858,0.932027,1.5196209,-1.0533147,1.2685812,0.0013257095,-1.983875,-0.02037545,-0.64693433,-1.3806919,-1.0592809,0.54943085,-1.3961786,-0.15649292,-0.5929407,0.9686789,-0.40543142,1.0557822,0.863148,-2.2666905,0.6786515,-0.98612607,-0.46339324,-0.5806146,1.359029,-0.97610635,-1.6902065,0.015555903,-1.0601696,-0.68368477,-0.4881073,1.0633672,-2.1536195,0.27614263,-0.94285184,0.8829792,-0.1332724,-0.69135016,0.057943374,0.29546717,-1.8476412,1.6050953,-0.77256185,-0.34552783,-0.39992267,-0.9135076,-1.4197912,-1.4342732,0.05444667,0.38957366,1.6343316,-0.018683603,0.20149979,-0.5184369,-2.7988343,-0.23812824,0.39474156,0.7759951,0.26147765,0.7695459,-0.8783734,0.24983662,0.22944757,-0.50273734,0.17921253,-2.96453,0.8493479,0.8776484,0.17855343,-0.3165259,-0.49537572,-1.4034132,-0.5197445,0.6516528,-1.1936586,0.4376411,-0.31979737,-1.1554341,-0.094374776,-1.4360906,-0.666431,0.19690529,0.22096215,-0.77558017,-0.6480959,-1.0999115,1.1237246,0.32265925,-0.7196827,-2.2545595,-1.4951689,0.5562803,1.1043608,1.1391095,0.12772338,0.39196157,0.68036157,-0.48576477,-1.285276,0.81914747,0.30907387,-0.07296551,1.534054,0.30991095,-0.007137307,1.7125996,-0.8783491,-0.07996723,-2.1059916,-0.3130489,-0.31512117,-1.8248315,-1.5771933,0.51070344,0.7182178,-0.2582489,-0.45623493,-0.2644527,-0.9527642,-0.7283463,0.17178611,0.21193407,0.9493434,-1.3720604,0.15322249,0.701293,-0.551722,-0.15245207,0.06241879,-0.6320952,0.5529451,-0.4287735,1.0122428,-0.07743517,-0.28508037,-1.175873,0.21220748,-1.3312136,-1.4357239,0.21157013,0.5917681,-1.1742185,1.9612663,-1.2883047,0.06013697,1.073086,0.5143654
+9,0,2,-0.16720563,-0.95600957,0.28606522,0.6061068,-0.3845801,-0.20699826,-1.1377771,0.20950346,-0.13692231,-0.39809802,1.1865551,1.4303648,-1.0548229,0.47550696,0.9458502,2.1207294,2.1566582,-0.24435441,-0.64201504,1.7221576,0.93151706,-0.7352646,0.34909317,0.72475654,-0.53412133,-0.38585177,-0.7807762,0.55336505,0.0011219378,1.0769404,0.178836,-1.3198038,1.3214653,-1.3146976,-0.7111258,-0.48764378,1.6924084,-0.51583934,1.3672884,-0.5570951,-0.8911707,-0.7621178,1.6342849,-1.217147,0.21610765,-2.0004432,-0.9046609,-0.66080266,0.099175125,-1.5156087,-1.408444,-1.4207674,-1.6178051,1.1580858,0.932027,1.5196209,-1.0533147,1.2685812,0.0013257095,-1.983875,-0.02037545,-0.64693433,-1.3806919,-1.0592809,0.54943085,-1.3961786,-0.15649292,-0.5929407,0.9686789,-0.40543142,1.0557822,0.863148,-2.2666905,0.6786515,-0.98612607,-0.46339324,-0.5806146,1.359029,-0.97610635,-1.6902065,0.015555903,-1.0601696,-0.68368477,-0.4881073,1.0633672,-2.1536195,0.27614263,-0.94285184,0.8829792,-0.1332724,-0.69135016,0.057943374,0.29546717,-1.8476412,1.6050953,-0.77256185,-0.34552783,-0.39992267,-0.9135076,-1.4197912,-1.4342732,0.05444667,0.38957366,1.6343316,-0.018683603,0.20149979,-0.5184369,-2.7988343,-0.23812824,0.39474156,0.7759951,0.26147765,0.7695459,-0.8783734,0.24983662,0.22944757,-0.50273734,0.17921253,-2.96453,0.8493479,0.8776484,0.17855343,-0.3165259,-0.49537572,-1.4034132,-0.5197445,0.6516528,-1.1936586,0.4376411,-0.31979737,-1.1554341,-0.094374776,-1.4360906,-0.666431,0.19690529,0.22096215,-0.77558017,-0.6480959,-1.0999115,1.1237246,0.32265925,-0.7196827,-2.2545595,-1.4951689,0.5562803,1.1043608,1.1391095,0.12772338,0.39196157,0.68036157,-0.48576477,-1.285276,0.81914747,0.30907387,-0.07296551,1.534054,0.30991095,-0.007137307,1.7125996,-0.8783491,-0.07996723,-2.1059916,-0.3130489,-0.31512117,-1.8248315,-1.5771933,0.51070344,0.7182178,-0.2582489,-0.45623493,-0.2644527,-0.9527642,-0.7283463,0.17178611,0.21193407,0.9493434,-1.3720604,0.15322249,0.701293,-0.551722,-0.15245207,0.06241879,-0.6320952,0.5529451,-0.4287735,1.0122428,-0.07743517,-0.28508037,-1.175873,0.21220748,-1.3312136,-1.4357239,0.21157013,0.5917681,-1.1742185,1.9612663,-1.2883047,0.06013697,1.073086,0.5143654
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
index d4e16b52..b7a6d23e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,5,0.25,-0.0029144287,-1.1171875,-0.29296875,1.234375,0.76953125,-0.5703125,1.0390625,0.96875,-0.13964844,0.67578125,2.8125,0.44726562,0.28320312,-0.90234375,-0.04638672,-0.67578125,0.5625,3.265625,1.6875,1.015625,-0.90625,0.640625,0.2578125,0.030883789,0.79296875,-0.56640625,-0.4375,-0.921875,-0.87890625,0.09814453,0.05859375,0.16503906,1.140625,-1.125,0.0069274902,0.2890625,0.23144531,0.19335938,0.017700195,-0.37890625,-0.42578125,-0.25585938,-1.0703125,0.119628906,0.7578125,0.14355469,-1.484375,0.13476562,0.1328125,0.5,-1.5390625,0.068847656,0.119628906,-1.3515625,0.05126953,-0.33789062,-0.44726562,1.2265625,-1.2734375,-0.068359375,-0.36132812,2.1875,-0.40820312,-0.6953125,-1.1640625,0.025756836,-1.3984375,-0.2578125,0.8203125,0.047607422,1.4140625,-0.53125,-0.78515625,1.09375,2.859375,-0.90234375,-0.13671875,-2.34375,0.025390625,0.49414062,-0.068359375,0.54296875,0.31054688,0.04711914,-0.296875,-1.15625,-1.3359375,0.390625,1.65625,1.421875,0.66796875,-1.625,1.640625,-0.013427734,-0.026367188,-0.32617188,0.5390625,-0.765625,-0.25195312,-0.16503906,0.24121094,1.0078125,-1.234375,-1.4609375,0.27148438,-0.640625,0.6953125,-0.45507812,-0.14746094,1.203125,0.5,0.14160156,-0.24804688,1.4375,-1.5390625,-1.0,-0.41992188,0.8203125,0.18652344,-0.21972656,0.39257812,0.47460938,-0.125,-1.484375,2.40625,-0.49804688,-1.2890625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40625,-1.6953125,0.546875,-1.3671875,-0.62109375,-0.08544922,-0.78515625,1.4140625,-1.640625,-0.875,2.03125,-0.671875,-0.375,-0.14160156,0.99609375,-0.76953125,0.38671875,1.1015625,-0.94140625,0.55078125,-0.875,-1.03125,-0.088378906,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.3984375,0.73046875,2.28125,0.068847656,0.079589844,-1.28125,-2.375,1.7734375,0.033935547,0.3828125,-0.79296875,-0.33398438,0.7421875,-0.94921875,-0.19433594,-0.90234375,-1.125,-0.6953125,0.64453125,-0.765625,-0.052490234,0.4140625,1.109375,-1.4453125,0.9375,-0.9609375,-1.2890625,1.796875,-0.9375,-1.4140625,-0.5390625,0.43554688,-0.58203125,-0.20507812,1.3828125,-0.4140625,0.30859375,0.33789062,1.2265625
-0,0,6,0.29882812,0.50390625,0.65625,-0.48828125,0.25976562,-0.027954102,0.671875,-0.6015625,0.59375,-0.66796875,0.68359375,0.09765625,1.609375,-0.73046875,0.0064086914,-0.31640625,-0.19628906,0.00049972534,2.625,0.16210938,-0.15332031,-1.3984375,0.66796875,0.05883789,-0.703125,-1.0078125,0.296875,-0.3125,-0.3828125,1.609375,0.64453125,-1.0703125,1.5078125,-1.40625,0.19335938,0.20996094,0.06640625,-1.046875,-0.84765625,0.8203125,-2.15625,0.29101562,-0.828125,-0.6484375,-0.27539062,-0.51953125,0.3125,-0.97265625,1.3359375,0.67578125,-0.78515625,0.34570312,-0.93359375,1.5390625,0.20117188,-0.30078125,0.13769531,-0.42382812,-1.140625,-0.83203125,1.3515625,0.5078125,-0.05419922,1.2578125,-0.18945312,0.4453125,-0.06201172,-1.8046875,-1.265625,1.546875,-0.66796875,0.5859375,0.64453125,0.061279297,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.015563965,-0.80859375,-0.6640625,0.5625,-0.1796875,-0.27734375,-0.609375,-1.234375,-1.9375,1.484375,-0.5078125,1.1953125,0.94921875,-0.96484375,0.328125,-1.1015625,-0.73828125,-1.921875,-0.75390625,-1.296875,0.042236328,-1.375,1.125,1.796875,-0.018066406,0.44335938,2.609375,-1.0390625,0.59375,0.28710938,-0.62109375,0.703125,1.1484375,0.64453125,-0.075683594,0.46484375,0.31054688,-0.21679688,0.24121094,-0.06591797,1.3046875,0.54296875,-0.61328125,1.1953125,-0.91015625,-1.1171875,1.1171875,-1.8125,-1.3984375,-0.19140625,-0.10058594,1.109375,-0.609375,-0.54296875,-1.359375,0.3828125,0.40234375,0.39453125,-0.40429688,0.21484375,-1.875,2.5,0.5703125,-0.67578125,2.328125,-0.25,-1.5234375,2.21875,-1.0859375,-1.34375,0.82421875,-0.9609375,0.203125,1.03125,-1.4140625,-0.32421875,-0.034423828,1.2578125,-0.82421875,1.09375,0.31835938,-0.39257812,-0.53515625,2.390625,1.78125,-0.953125,-0.7734375,-1.1484375,1.015625,1.1015625,-0.39453125,-0.30859375,-0.5,-1.7265625,-0.9140625,-0.890625,0.265625,0.98046875,-1.4375,0.50390625,0.26171875,-1.921875,-1.015625,1.171875,0.2734375,-1.5390625,-0.41796875,0.43164062,-0.20996094,-0.65234375,-0.05810547,-1.703125,1.78125,1.1796875,-0.53125,-1.0859375,0.90234375,1.796875,-0.3515625,1.3671875,1.0703125
-0,0,7,0.66796875,1.21875,-1.203125,-1.3125,-0.515625,-1.90625,0.31640625,1.1875,-0.52734375,-0.67578125,-0.053466797,-1.203125,-1.828125,-0.33203125,-0.9296875,1.0625,-0.27734375,0.91796875,0.1640625,0.020629883,0.90234375,-0.234375,-0.03466797,1.3359375,-1.71875,-0.96484375,-2.375,0.6640625,1.9921875,0.9765625,0.76953125,1.3984375,-1.5234375,-0.38671875,0.734375,-0.017211914,0.9375,1.3671875,-0.049804688,0.62109375,-2.15625,-0.27539062,0.072753906,1.7578125,0.71484375,2.015625,0.43164062,0.59765625,0.1875,1.25,0.28320312,-0.22949219,-1.3515625,-0.796875,-0.13183594,-0.43359375,-0.98046875,0.012390137,-0.06347656,-0.61328125,-0.1640625,0.53515625,0.34960938,-0.38085938,1.328125,0.23535156,0.16503906,0.06347656,-1.0703125,0.59765625,0.049316406,-0.171875,0.390625,-0.8359375,0.6796875,0.08544922,2.09375,-0.043945312,-0.42773438,0.03857422,1.2734375,0.609375,1.5703125,0.9140625,-0.08544922,0.98828125,-1.328125,0.515625,-1.7421875,0.3515625,0.95703125,2.21875,1.796875,-0.41015625,-0.068847656,-1.40625,-0.328125,0.013549805,0.96875,1.0390625,0.875,-0.18457031,1.03125,-0.17382812,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0077819824,-0.37890625,-0.22558594,2.546875,-0.8828125,-0.4765625,-0.88671875,-0.35351562,-1.0390625,0.98828125,0.68359375,2.046875,-1.875,-0.86328125,-2.953125,-0.21582031,0.18847656,-0.98046875,1.25,0.17382812,-1.9765625,0.90234375,-0.37695312,0.67578125,-1.0703125,-1.8828125,0.58984375,0.53515625,-0.55078125,0.111816406,-2.1875,0.32421875,0.89453125,1.7109375,1.2109375,-0.6171875,-0.49804688,-1.625,1.0703125,-0.37695312,0.13085938,-0.041992188,-1.3515625,-0.7890625,1.4375,0.19433594,-0.7109375,-0.6015625,1.0625,-0.37695312,1.90625,-2.0625,0.9609375,-0.32617188,0.078125,-1.3046875,-0.15039062,-0.26953125,0.29296875,0.46679688,-0.5859375,-0.32421875,0.46875,-1.1015625,0.984375,0.953125,0.25,0.99609375,-1.546875,0.041992188,-0.1015625,1.0,0.53515625,0.62109375,-0.8828125,-0.50390625,-0.32617188,-0.24804688,0.064453125,-0.5390625,-1.0546875,-1.234375,0.31640625,-0.037841797,-0.87890625,1.296875,0.1171875,1.8984375,-0.2265625,0.09667969,0.42578125
+0,0,1,0.25,-0.0029144287,-1.1171875,-0.29296875,1.234375,0.76953125,-0.5703125,1.0390625,0.96875,-0.13964844,0.67578125,2.8125,0.44726562,0.28320312,-0.90234375,-0.04638672,-0.67578125,0.5625,3.265625,1.6875,1.015625,-0.90625,0.640625,0.2578125,0.030883789,0.79296875,-0.56640625,-0.4375,-0.921875,-0.87890625,0.09814453,0.05859375,0.16503906,1.140625,-1.125,0.0069274902,0.2890625,0.23144531,0.19335938,0.017700195,-0.37890625,-0.42578125,-0.25585938,-1.0703125,0.119628906,0.7578125,0.14355469,-1.484375,0.13476562,0.1328125,0.5,-1.5390625,0.068847656,0.119628906,-1.3515625,0.05126953,-0.33789062,-0.44726562,1.2265625,-1.2734375,-0.068359375,-0.36132812,2.1875,-0.40820312,-0.6953125,-1.1640625,0.025756836,-1.3984375,-0.2578125,0.8203125,0.047607422,1.4140625,-0.53125,-0.78515625,1.09375,2.859375,-0.90234375,-0.13671875,-2.34375,0.025390625,0.49414062,-0.068359375,0.54296875,0.31054688,0.04711914,-0.296875,-1.15625,-1.3359375,0.390625,1.65625,1.421875,0.66796875,-1.625,1.640625,-0.013427734,-0.026367188,-0.32617188,0.5390625,-0.765625,-0.25195312,-0.16503906,0.24121094,1.0078125,-1.234375,-1.4609375,0.27148438,-0.640625,0.6953125,-0.45507812,-0.14746094,1.203125,0.5,0.14160156,-0.24804688,1.4375,-1.5390625,-1.0,-0.41992188,0.8203125,0.18652344,-0.21972656,0.39257812,0.47460938,-0.125,-1.484375,2.40625,-0.49804688,-1.2890625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40625,-1.6953125,0.546875,-1.3671875,-0.62109375,-0.08544922,-0.78515625,1.4140625,-1.640625,-0.875,2.03125,-0.671875,-0.375,-0.14160156,0.99609375,-0.76953125,0.38671875,1.1015625,-0.94140625,0.55078125,-0.875,-1.03125,-0.088378906,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.3984375,0.73046875,2.28125,0.068847656,0.079589844,-1.28125,-2.375,1.7734375,0.033935547,0.3828125,-0.79296875,-0.33398438,0.7421875,-0.94921875,-0.19433594,-0.90234375,-1.125,-0.6953125,0.64453125,-0.765625,-0.052490234,0.4140625,1.109375,-1.4453125,0.9375,-0.9609375,-1.2890625,1.796875,-0.9375,-1.4140625,-0.5390625,0.43554688,-0.58203125,-0.20507812,1.3828125,-0.4140625,0.30859375,0.33789062,1.2265625
+0,0,2,0.29882812,0.50390625,0.65625,-0.48828125,0.25976562,-0.027954102,0.671875,-0.6015625,0.59375,-0.66796875,0.68359375,0.09765625,1.609375,-0.73046875,0.0064086914,-0.31640625,-0.19628906,0.00049972534,2.625,0.16210938,-0.15332031,-1.3984375,0.66796875,0.05883789,-0.703125,-1.0078125,0.296875,-0.3125,-0.3828125,1.609375,0.64453125,-1.0703125,1.5078125,-1.40625,0.19335938,0.20996094,0.06640625,-1.046875,-0.84765625,0.8203125,-2.15625,0.29101562,-0.828125,-0.6484375,-0.27539062,-0.51953125,0.3125,-0.97265625,1.3359375,0.67578125,-0.78515625,0.34570312,-0.93359375,1.5390625,0.20117188,-0.30078125,0.13769531,-0.42382812,-1.140625,-0.83203125,1.3515625,0.5078125,-0.05419922,1.2578125,-0.18945312,0.4453125,-0.06201172,-1.8046875,-1.265625,1.546875,-0.66796875,0.5859375,0.64453125,0.061279297,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.015563965,-0.80859375,-0.6640625,0.5625,-0.1796875,-0.27734375,-0.609375,-1.234375,-1.9375,1.484375,-0.5078125,1.1953125,0.94921875,-0.96484375,0.328125,-1.1015625,-0.73828125,-1.921875,-0.75390625,-1.296875,0.042236328,-1.375,1.125,1.796875,-0.018066406,0.44335938,2.609375,-1.0390625,0.59375,0.28710938,-0.62109375,0.703125,1.1484375,0.64453125,-0.075683594,0.46484375,0.31054688,-0.21679688,0.24121094,-0.06591797,1.3046875,0.54296875,-0.61328125,1.1953125,-0.91015625,-1.1171875,1.1171875,-1.8125,-1.3984375,-0.19140625,-0.10058594,1.109375,-0.609375,-0.54296875,-1.359375,0.3828125,0.40234375,0.39453125,-0.40429688,0.21484375,-1.875,2.5,0.5703125,-0.67578125,2.328125,-0.25,-1.5234375,2.21875,-1.0859375,-1.34375,0.82421875,-0.9609375,0.203125,1.03125,-1.4140625,-0.32421875,-0.034423828,1.2578125,-0.82421875,1.09375,0.31835938,-0.39257812,-0.53515625,2.390625,1.78125,-0.953125,-0.7734375,-1.1484375,1.015625,1.1015625,-0.39453125,-0.30859375,-0.5,-1.7265625,-0.9140625,-0.890625,0.265625,0.98046875,-1.4375,0.50390625,0.26171875,-1.921875,-1.015625,1.171875,0.2734375,-1.5390625,-0.41796875,0.43164062,-0.20996094,-0.65234375,-0.05810547,-1.703125,1.78125,1.1796875,-0.53125,-1.0859375,0.90234375,1.796875,-0.3515625,1.3671875,1.0703125
+0,0,3,0.66796875,1.21875,-1.203125,-1.3125,-0.515625,-1.90625,0.31640625,1.1875,-0.52734375,-0.67578125,-0.053466797,-1.203125,-1.828125,-0.33203125,-0.9296875,1.0625,-0.27734375,0.91796875,0.1640625,0.020629883,0.90234375,-0.234375,-0.03466797,1.3359375,-1.71875,-0.96484375,-2.375,0.6640625,1.9921875,0.9765625,0.76953125,1.3984375,-1.5234375,-0.38671875,0.734375,-0.017211914,0.9375,1.3671875,-0.049804688,0.62109375,-2.15625,-0.27539062,0.072753906,1.7578125,0.71484375,2.015625,0.43164062,0.59765625,0.1875,1.25,0.28320312,-0.22949219,-1.3515625,-0.796875,-0.13183594,-0.43359375,-0.98046875,0.012390137,-0.06347656,-0.61328125,-0.1640625,0.53515625,0.34960938,-0.38085938,1.328125,0.23535156,0.16503906,0.06347656,-1.0703125,0.59765625,0.049316406,-0.171875,0.390625,-0.8359375,0.6796875,0.08544922,2.09375,-0.043945312,-0.42773438,0.03857422,1.2734375,0.609375,1.5703125,0.9140625,-0.08544922,0.98828125,-1.328125,0.515625,-1.7421875,0.3515625,0.95703125,2.21875,1.796875,-0.41015625,-0.068847656,-1.40625,-0.328125,0.013549805,0.96875,1.0390625,0.875,-0.18457031,1.03125,-0.17382812,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0077819824,-0.37890625,-0.22558594,2.546875,-0.8828125,-0.4765625,-0.88671875,-0.35351562,-1.0390625,0.98828125,0.68359375,2.046875,-1.875,-0.86328125,-2.953125,-0.21582031,0.18847656,-0.98046875,1.25,0.17382812,-1.9765625,0.90234375,-0.37695312,0.67578125,-1.0703125,-1.8828125,0.58984375,0.53515625,-0.55078125,0.111816406,-2.1875,0.32421875,0.89453125,1.7109375,1.2109375,-0.6171875,-0.49804688,-1.625,1.0703125,-0.37695312,0.13085938,-0.041992188,-1.3515625,-0.7890625,1.4375,0.19433594,-0.7109375,-0.6015625,1.0625,-0.37695312,1.90625,-2.0625,0.9609375,-0.32617188,0.078125,-1.3046875,-0.15039062,-0.26953125,0.29296875,0.46679688,-0.5859375,-0.32421875,0.46875,-1.1015625,0.984375,0.953125,0.25,0.99609375,-1.546875,0.041992188,-0.1015625,1.0,0.53515625,0.62109375,-0.8828125,-0.50390625,-0.32617188,-0.24804688,0.064453125,-0.5390625,-1.0546875,-1.234375,0.31640625,-0.037841797,-0.87890625,1.296875,0.1171875,1.8984375,-0.2265625,0.09667969,0.42578125
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
index a0974d2a..9633bd19 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,5,1.0546875,0.89453125,-0.55078125,0.83203125,-0.65234375,-0.37890625,0.77734375,1.6484375,0.5625,-1.6171875,-0.53515625,0.54296875,0.76953125,-0.1015625,-1.4453125,0.828125,-0.05859375,0.58203125,0.34179688,-0.37304688,0.83984375,-0.9375,-0.032226562,-0.9140625,0.15820312,-0.953125,-1.5078125,-0.14160156,-1.1484375,-0.87890625,1.3203125,-1.3359375,1.71875,0.76171875,0.22167969,-1.578125,-0.122558594,-1.7265625,-0.7265625,-1.125,-1.2421875,-1.296875,-1.28125,0.017822266,-1.8828125,0.9296875,0.48046875,1.0078125,1.125,1.546875,-1.3046875,-0.13378906,-0.73046875,1.7421875,-0.671875,0.5078125,1.15625,0.66796875,-0.15234375,-0.23046875,0.05883789,0.51171875,0.08251953,-0.12207031,0.72265625,-0.5859375,-0.90625,-0.40429688,-0.7109375,0.25195312,1.5859375,-1.8125,0.57421875,-1.0390625,2.296875,2.609375,1.5390625,0.82421875,-0.34179688,-1.1328125,-0.38085938,-1.25,-0.42578125,-1.2734375,0.9296875,-1.1875,1.28125,0.59765625,-0.3046875,1.4921875,2.25,-1.3359375,0.16992188,0.734375,-0.45703125,0.57421875,0.19042969,-0.67578125,-0.21484375,0.5859375,-1.5703125,1.15625,1.03125,-0.02734375,1.9921875,-0.55859375,0.40429688,0.57421875,0.87890625,1.59375,-0.31835938,0.0119018555,0.296875,-0.61328125,1.0,-0.2890625,-1.6953125,0.16992188,-0.15234375,-0.28125,-0.66796875,-0.40039062,0.22363281,-0.026977539,-0.57421875,0.5859375,0.5703125,-0.59375,-0.48828125,-0.515625,-0.38476562,0.2109375,-0.25195312,-1.28125,-1.7578125,-0.46484375,-0.609375,0.37695312,-0.49609375,-2.328125,2.15625,-0.66796875,-0.5234375,-1.9453125,-0.30859375,-0.19140625,-0.7265625,-1.09375,0.61328125,3.640625,-0.44921875,1.0625,-1.7890625,0.828125,0.19335938,-0.9921875,0.14355469,1.2109375,0.29492188,-0.8515625,0.2578125,0.36523438,-0.029418945,0.12792969,-0.83984375,0.53125,0.15625,0.8359375,1.9765625,-2.328125,-0.33203125,0.29882812,-0.41601562,-0.25195312,0.27539062,1.1328125,1.0078125,1.203125,-0.42773438,0.38476562,-0.83203125,1.4375,0.875,0.13183594,0.32617188,-1.203125,1.4296875,0.91015625,-1.328125,-0.1953125,-0.41015625,-0.8203125,0.8203125,0.29101562,-1.6875,1.109375,0.049072266,-0.037353516,-0.34960938,-0.3828125
-1,0,6,-0.5546875,1.265625,-1.3671875,-1.5078125,0.30664062,-1.9453125,0.34375,1.6328125,1.609375,-0.58984375,0.21875,1.03125,-0.9453125,0.515625,-1.21875,0.53515625,0.2734375,0.21972656,-0.37109375,-0.07373047,0.33984375,-0.39453125,1.3359375,-0.0390625,-0.7421875,-0.921875,-2.421875,1.1015625,1.578125,-0.4765625,1.3671875,0.21191406,-1.1171875,0.14941406,1.5703125,-1.0859375,0.6640625,0.86328125,-0.7734375,1.046875,-1.546875,0.43164062,-0.22070312,-0.26953125,-0.32421875,2.625,-0.06201172,1.8125,1.140625,1.609375,-1.03125,1.0859375,-0.640625,-0.09033203,-0.82421875,-0.08984375,-0.51953125,0.48828125,-1.109375,1.40625,0.3984375,0.94140625,-0.16015625,0.609375,0.54296875,-1.203125,-0.12695312,0.71875,-1.65625,1.265625,0.52734375,0.7890625,2.046875,-0.53515625,0.55859375,2.09375,0.51171875,0.703125,-0.8359375,-0.40039062,0.48242188,-0.36523438,0.76171875,-0.13378906,-0.119628906,0.94140625,-1.421875,-0.27148438,-1.1640625,-0.6796875,0.5859375,1.53125,0.98046875,-0.25976562,0.7890625,-1.875,-0.53515625,-0.26171875,0.40625,-0.55859375,0.14160156,0.49414062,0.6953125,0.15527344,-0.76953125,0.068847656,-1.046875,-0.40039062,0.041259766,0.56640625,-1.6484375,-0.068359375,1.6640625,-2.3125,0.6484375,-0.14941406,-1.5625,-0.3125,0.91796875,0.08203125,1.171875,-2.34375,0.026367188,-1.359375,0.37890625,0.96484375,-0.26171875,-1.359375,1.0390625,-0.24902344,-1.328125,-0.91796875,-0.5390625,0.578125,-1.8984375,1.5390625,1.1875,-0.1953125,0.91015625,-2.03125,1.296875,0.123535156,0.6484375,0.8359375,-1.2421875,-1.3359375,-0.81640625,-1.3203125,-1.453125,-0.703125,0.60546875,-0.61328125,-0.48242188,0.875,-0.73828125,1.828125,0.5625,0.39453125,0.45898438,0.29101562,-0.27148438,1.1171875,0.73046875,0.36914062,0.026733398,-0.72265625,-0.38671875,0.7578125,-0.49414062,-1.9921875,-1.03125,-0.05883789,-1.578125,-0.6953125,0.8515625,-0.13183594,1.0703125,0.84375,-0.31445312,0.021362305,0.08154297,-1.1171875,0.30664062,0.16894531,0.33203125,0.06640625,-0.82421875,-0.8515625,-1.7734375,-1.0,0.2890625,0.81640625,-0.7265625,0.68359375,-0.81640625,2.640625,2.03125,0.09814453,-0.8984375,-0.48242188
-1,0,7,1.2578125,0.5703125,-0.9921875,0.103515625,-1.359375,-0.11767578,0.609375,1.0078125,-1.1640625,-1.4453125,-0.21582031,-0.37890625,-0.42382812,0.453125,-1.7421875,-0.8203125,-1.0625,0.28320312,-0.953125,-0.53125,0.59375,-0.21972656,-0.57421875,-0.46679688,-0.107421875,-2.171875,-2.390625,-0.4375,-0.008850098,0.19433594,-0.38085938,-1.4453125,0.2890625,0.87890625,-0.328125,-0.546875,-0.09423828,-2.390625,0.45507812,-0.08203125,0.40820312,-0.20996094,-0.859375,1.2421875,0.9375,0.3515625,-1.0234375,0.60546875,-0.025146484,1.1640625,0.16015625,-0.19824219,-1.65625,0.70703125,-0.90625,-0.67578125,-0.63671875,1.25,-1.0625,-1.03125,-0.28125,0.765625,1.265625,0.7734375,1.921875,-0.34960938,-0.28125,-0.5078125,-1.9296875,0.5,1.21875,-0.36132812,1.6640625,-0.6953125,0.8828125,0.828125,2.53125,1.234375,0.609375,0.099609375,-0.41210938,-0.3828125,0.37109375,-1.3359375,-0.27734375,0.99609375,0.31835938,-0.60546875,0.24121094,0.984375,0.5390625,1.34375,0.21972656,0.58203125,-0.95703125,-1.2890625,-0.734375,-0.13476562,1.1484375,1.0859375,0.38085938,1.2578125,-0.296875,-1.2265625,1.2578125,-1.078125,-0.71484375,0.69140625,0.41015625,-0.33398438,-0.96875,1.015625,1.578125,0.26367188,0.119628906,0.46679688,-0.3046875,0.515625,0.122558594,1.921875,0.30664062,-2.109375,-1.1484375,-0.59765625,-0.056152344,-1.25,0.171875,-1.046875,0.001335144,-0.16796875,0.421875,-0.5703125,-0.18945312,-0.022338867,-1.046875,2.4375,-0.9140625,-1.6875,1.40625,-2.3125,0.64453125,-0.671875,-0.28320312,-0.65234375,0.8671875,0.30664062,0.375,0.20214844,2.203125,3.59375,0.41992188,0.26367188,-1.609375,1.015625,-0.0071411133,0.29492188,0.65625,1.1484375,-1.359375,0.62890625,0.97265625,1.75,0.8046875,-0.62109375,-0.51171875,-0.55078125,-0.3515625,1.6796875,0.5234375,-1.7578125,-0.62109375,-0.94140625,0.23242188,-0.55078125,0.703125,1.3984375,-0.3125,1.1171875,1.3125,-0.14257812,0.064941406,0.122558594,1.3671875,-0.64453125,0.49023438,-0.2890625,0.640625,1.4140625,-0.0033111572,-1.515625,0.21582031,-1.6015625,0.5390625,-1.7421875,-0.81640625,1.2109375,0.30273438,-1.09375,-1.234375,-0.36328125
+1,0,1,1.0546875,0.89453125,-0.55078125,0.83203125,-0.65234375,-0.37890625,0.77734375,1.6484375,0.5625,-1.6171875,-0.53515625,0.54296875,0.76953125,-0.1015625,-1.4453125,0.828125,-0.05859375,0.58203125,0.34179688,-0.37304688,0.83984375,-0.9375,-0.032226562,-0.9140625,0.15820312,-0.953125,-1.5078125,-0.14160156,-1.1484375,-0.87890625,1.3203125,-1.3359375,1.71875,0.76171875,0.22167969,-1.578125,-0.122558594,-1.7265625,-0.7265625,-1.125,-1.2421875,-1.296875,-1.28125,0.017822266,-1.8828125,0.9296875,0.48046875,1.0078125,1.125,1.546875,-1.3046875,-0.13378906,-0.73046875,1.7421875,-0.671875,0.5078125,1.15625,0.66796875,-0.15234375,-0.23046875,0.05883789,0.51171875,0.08251953,-0.12207031,0.72265625,-0.5859375,-0.90625,-0.40429688,-0.7109375,0.25195312,1.5859375,-1.8125,0.57421875,-1.0390625,2.296875,2.609375,1.5390625,0.82421875,-0.34179688,-1.1328125,-0.38085938,-1.25,-0.42578125,-1.2734375,0.9296875,-1.1875,1.28125,0.59765625,-0.3046875,1.4921875,2.25,-1.3359375,0.16992188,0.734375,-0.45703125,0.57421875,0.19042969,-0.67578125,-0.21484375,0.5859375,-1.5703125,1.15625,1.03125,-0.02734375,1.9921875,-0.55859375,0.40429688,0.57421875,0.87890625,1.59375,-0.31835938,0.0119018555,0.296875,-0.61328125,1.0,-0.2890625,-1.6953125,0.16992188,-0.15234375,-0.28125,-0.66796875,-0.40039062,0.22363281,-0.026977539,-0.57421875,0.5859375,0.5703125,-0.59375,-0.48828125,-0.515625,-0.38476562,0.2109375,-0.25195312,-1.28125,-1.7578125,-0.46484375,-0.609375,0.37695312,-0.49609375,-2.328125,2.15625,-0.66796875,-0.5234375,-1.9453125,-0.30859375,-0.19140625,-0.7265625,-1.09375,0.61328125,3.640625,-0.44921875,1.0625,-1.7890625,0.828125,0.19335938,-0.9921875,0.14355469,1.2109375,0.29492188,-0.8515625,0.2578125,0.36523438,-0.029418945,0.12792969,-0.83984375,0.53125,0.15625,0.8359375,1.9765625,-2.328125,-0.33203125,0.29882812,-0.41601562,-0.25195312,0.27539062,1.1328125,1.0078125,1.203125,-0.42773438,0.38476562,-0.83203125,1.4375,0.875,0.13183594,0.32617188,-1.203125,1.4296875,0.91015625,-1.328125,-0.1953125,-0.41015625,-0.8203125,0.8203125,0.29101562,-1.6875,1.109375,0.049072266,-0.037353516,-0.34960938,-0.3828125
+1,0,2,-0.5546875,1.265625,-1.3671875,-1.5078125,0.30664062,-1.9453125,0.34375,1.6328125,1.609375,-0.58984375,0.21875,1.03125,-0.9453125,0.515625,-1.21875,0.53515625,0.2734375,0.21972656,-0.37109375,-0.07373047,0.33984375,-0.39453125,1.3359375,-0.0390625,-0.7421875,-0.921875,-2.421875,1.1015625,1.578125,-0.4765625,1.3671875,0.21191406,-1.1171875,0.14941406,1.5703125,-1.0859375,0.6640625,0.86328125,-0.7734375,1.046875,-1.546875,0.43164062,-0.22070312,-0.26953125,-0.32421875,2.625,-0.06201172,1.8125,1.140625,1.609375,-1.03125,1.0859375,-0.640625,-0.09033203,-0.82421875,-0.08984375,-0.51953125,0.48828125,-1.109375,1.40625,0.3984375,0.94140625,-0.16015625,0.609375,0.54296875,-1.203125,-0.12695312,0.71875,-1.65625,1.265625,0.52734375,0.7890625,2.046875,-0.53515625,0.55859375,2.09375,0.51171875,0.703125,-0.8359375,-0.40039062,0.48242188,-0.36523438,0.76171875,-0.13378906,-0.119628906,0.94140625,-1.421875,-0.27148438,-1.1640625,-0.6796875,0.5859375,1.53125,0.98046875,-0.25976562,0.7890625,-1.875,-0.53515625,-0.26171875,0.40625,-0.55859375,0.14160156,0.49414062,0.6953125,0.15527344,-0.76953125,0.068847656,-1.046875,-0.40039062,0.041259766,0.56640625,-1.6484375,-0.068359375,1.6640625,-2.3125,0.6484375,-0.14941406,-1.5625,-0.3125,0.91796875,0.08203125,1.171875,-2.34375,0.026367188,-1.359375,0.37890625,0.96484375,-0.26171875,-1.359375,1.0390625,-0.24902344,-1.328125,-0.91796875,-0.5390625,0.578125,-1.8984375,1.5390625,1.1875,-0.1953125,0.91015625,-2.03125,1.296875,0.123535156,0.6484375,0.8359375,-1.2421875,-1.3359375,-0.81640625,-1.3203125,-1.453125,-0.703125,0.60546875,-0.61328125,-0.48242188,0.875,-0.73828125,1.828125,0.5625,0.39453125,0.45898438,0.29101562,-0.27148438,1.1171875,0.73046875,0.36914062,0.026733398,-0.72265625,-0.38671875,0.7578125,-0.49414062,-1.9921875,-1.03125,-0.05883789,-1.578125,-0.6953125,0.8515625,-0.13183594,1.0703125,0.84375,-0.31445312,0.021362305,0.08154297,-1.1171875,0.30664062,0.16894531,0.33203125,0.06640625,-0.82421875,-0.8515625,-1.7734375,-1.0,0.2890625,0.81640625,-0.7265625,0.68359375,-0.81640625,2.640625,2.03125,0.09814453,-0.8984375,-0.48242188
+1,0,3,1.2578125,0.5703125,-0.9921875,0.103515625,-1.359375,-0.11767578,0.609375,1.0078125,-1.1640625,-1.4453125,-0.21582031,-0.37890625,-0.42382812,0.453125,-1.7421875,-0.8203125,-1.0625,0.28320312,-0.953125,-0.53125,0.59375,-0.21972656,-0.57421875,-0.46679688,-0.107421875,-2.171875,-2.390625,-0.4375,-0.008850098,0.19433594,-0.38085938,-1.4453125,0.2890625,0.87890625,-0.328125,-0.546875,-0.09423828,-2.390625,0.45507812,-0.08203125,0.40820312,-0.20996094,-0.859375,1.2421875,0.9375,0.3515625,-1.0234375,0.60546875,-0.025146484,1.1640625,0.16015625,-0.19824219,-1.65625,0.70703125,-0.90625,-0.67578125,-0.63671875,1.25,-1.0625,-1.03125,-0.28125,0.765625,1.265625,0.7734375,1.921875,-0.34960938,-0.28125,-0.5078125,-1.9296875,0.5,1.21875,-0.36132812,1.6640625,-0.6953125,0.8828125,0.828125,2.53125,1.234375,0.609375,0.099609375,-0.41210938,-0.3828125,0.37109375,-1.3359375,-0.27734375,0.99609375,0.31835938,-0.60546875,0.24121094,0.984375,0.5390625,1.34375,0.21972656,0.58203125,-0.95703125,-1.2890625,-0.734375,-0.13476562,1.1484375,1.0859375,0.38085938,1.2578125,-0.296875,-1.2265625,1.2578125,-1.078125,-0.71484375,0.69140625,0.41015625,-0.33398438,-0.96875,1.015625,1.578125,0.26367188,0.119628906,0.46679688,-0.3046875,0.515625,0.122558594,1.921875,0.30664062,-2.109375,-1.1484375,-0.59765625,-0.056152344,-1.25,0.171875,-1.046875,0.001335144,-0.16796875,0.421875,-0.5703125,-0.18945312,-0.022338867,-1.046875,2.4375,-0.9140625,-1.6875,1.40625,-2.3125,0.64453125,-0.671875,-0.28320312,-0.65234375,0.8671875,0.30664062,0.375,0.20214844,2.203125,3.59375,0.41992188,0.26367188,-1.609375,1.015625,-0.0071411133,0.29492188,0.65625,1.1484375,-1.359375,0.62890625,0.97265625,1.75,0.8046875,-0.62109375,-0.51171875,-0.55078125,-0.3515625,1.6796875,0.5234375,-1.7578125,-0.62109375,-0.94140625,0.23242188,-0.55078125,0.703125,1.3984375,-0.3125,1.1171875,1.3125,-0.14257812,0.064941406,0.122558594,1.3671875,-0.64453125,0.49023438,-0.2890625,0.640625,1.4140625,-0.0033111572,-1.515625,0.21582031,-1.6015625,0.5390625,-1.7421875,-0.81640625,1.2109375,0.30273438,-1.09375,-1.234375,-0.36328125
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
index 6031742c..2451ec01 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,5,1.203125,-0.09814453,-1.3359375,-0.6015625,0.96484375,-1.4765625,-0.8203125,3.140625,-0.5234375,-1.671875,0.023925781,1.5859375,-0.53125,-0.74609375,-0.57421875,1.7421875,-0.26953125,-1.3984375,0.059326172,0.48828125,-1.3671875,-0.6796875,-0.44921875,1.234375,-0.09033203,0.32617188,-2.5,-0.15136719,-0.8828125,-0.06982422,-0.022216797,0.78515625,0.69140625,0.7734375,-0.07470703,-1.0546875,1.2578125,-0.17285156,0.515625,0.6875,-0.0625,-0.008483887,0.11669922,-0.7890625,-0.11816406,0.65625,-0.46875,0.96875,-0.43945312,2.21875,-0.60546875,-2.21875,-1.3359375,-0.04736328,-0.15039062,-0.33203125,0.91796875,0.296875,0.40039062,-1.046875,-0.36523438,1.484375,-0.1328125,0.53515625,-0.265625,-0.03125,-0.002380371,1.4921875,0.16699219,-0.11767578,-0.4140625,-0.29296875,0.8984375,-0.671875,1.625,2.140625,0.625,0.2734375,-0.2421875,-0.75,1.671875,0.75,0.53515625,-0.62890625,-0.44140625,-1.1640625,1.328125,1.828125,0.49414062,1.65625,1.7421875,-0.7734375,-1.046875,1.3671875,-0.28710938,0.03100586,-1.7265625,-0.734375,-1.0859375,-0.21777344,-1.0625,-0.24414062,1.4375,0.94140625,-0.47265625,0.25195312,-1.640625,-0.62109375,-0.29101562,0.0234375,-1.140625,-1.9296875,1.4140625,-2.265625,0.84765625,0.27929688,-2.046875,1.2734375,-0.30078125,-1.21875,1.25,-0.4765625,-0.92578125,-0.36523438,0.4140625,2.75,-0.43164062,-0.40625,0.44335938,-1.8671875,-1.4765625,-1.640625,0.36914062,-1.1015625,-0.890625,-0.012817383,1.328125,0.34570312,0.43554688,-1.3515625,1.46875,-1.8203125,-0.62890625,-0.15527344,0.16113281,0.46875,-0.32421875,0.83203125,-0.73828125,0.83203125,-0.6171875,-0.018188477,0.8125,-0.4453125,0.045898438,-0.7578125,0.3203125,0.015563965,1.65625,1.015625,-1.1484375,1.875,1.2890625,0.8203125,-0.57421875,0.77734375,0.16894531,1.3125,1.2578125,0.76953125,-1.296875,-0.17089844,0.060546875,1.0859375,-1.28125,0.43359375,0.3515625,-0.69140625,-0.059326172,-1.171875,0.45898438,0.38476562,0.30273438,-1.09375,0.22363281,-0.24414062,-1.3984375,0.0134887695,-0.83984375,-1.3515625,0.20996094,1.421875,-0.11621094,-0.43164062,0.12207031,1.1953125,0.49609375,0.6015625,-1.0703125,0.4609375
-2,0,6,0.31445312,-0.33398438,-0.72265625,-0.4453125,0.40429688,-0.546875,-0.006439209,1.5859375,-0.13769531,-0.828125,-0.53515625,-0.375,1.2265625,0.49804688,-1.6484375,0.44726562,-0.5234375,-1.984375,0.42578125,-0.6875,0.5625,-0.32421875,-0.27734375,-0.06640625,-1.328125,-1.8203125,-1.640625,-0.08984375,-2.078125,0.49023438,-0.6875,-1.1171875,0.5625,0.07519531,0.546875,1.6796875,0.62109375,-1.453125,-1.8671875,1.1953125,-1.4765625,0.3515625,-0.54296875,-0.043701172,0.23144531,1.0625,0.5625,-0.546875,0.24023438,1.71875,-2.109375,-1.328125,-1.40625,0.73046875,-1.9921875,0.62109375,0.125,-0.66796875,-0.19140625,-2.140625,1.1796875,-0.16113281,0.42578125,-0.072265625,0.17675781,0.055908203,-1.0859375,0.18847656,-0.671875,2.0,0.90625,-1.6015625,-0.4609375,-0.49609375,0.044189453,1.921875,1.1171875,1.109375,-0.61328125,-0.8828125,1.5546875,0.31835938,0.27734375,-0.5078125,-0.67578125,-0.030639648,0.28710938,0.9609375,-0.66796875,0.67578125,2.125,-0.17285156,0.14941406,0.41796875,-1.390625,-0.19921875,-0.30273438,-0.53125,1.4140625,-0.328125,-2.25,0.14160156,2.25,1.15625,-0.46679688,-0.22851562,1.484375,-0.9296875,0.40039062,0.71484375,0.2734375,0.36328125,0.8515625,-1.4296875,0.30664062,-0.46875,0.18554688,-0.27148438,-0.65234375,-0.65625,1.328125,-0.55078125,0.4140625,-1.1875,-0.24609375,1.2734375,-0.44921875,-2.296875,-0.005645752,-0.041015625,0.671875,-0.061767578,0.19238281,-1.3046875,-1.34375,-0.19921875,-0.35351562,-0.47070312,-0.019042969,-1.8046875,3.0625,-0.76953125,0.39257812,-0.053955078,-0.06933594,1.828125,-1.234375,-0.01977539,-1.9375,1.921875,0.23632812,1.125,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.4375,1.6171875,1.4296875,0.671875,-0.71484375,1.3515625,1.375,0.49804688,-1.3359375,0.53515625,-1.1171875,-0.33984375,2.375,-0.25195312,0.2265625,-0.14550781,-0.66796875,-0.31835938,0.118652344,0.453125,-0.9375,-0.91796875,0.93359375,-0.58984375,-0.18164062,-0.10888672,1.265625,0.48828125,0.6640625,-0.30273438,1.3203125,0.82421875,-0.5390625,-1.0390625,-0.23242188,0.53125,2.15625,0.024658203,0.83203125,-0.15625,1.40625,-0.20996094,1.21875,0.27148438
-2,0,7,0.76953125,0.111328125,-0.36523438,-0.22363281,0.20703125,-0.7578125,-0.2734375,1.4140625,0.5546875,-1.09375,0.5859375,1.4140625,0.4375,0.54296875,-1.6015625,0.8046875,0.19335938,-1.2109375,-0.11230469,0.119628906,0.98828125,-0.5390625,-0.39257812,0.578125,-0.29296875,-0.73046875,-0.51171875,0.06201172,-1.0078125,-0.059814453,-0.68359375,-0.51171875,1.78125,-0.45117188,1.296875,1.59375,0.83203125,-0.16015625,-2.21875,1.7265625,-1.8359375,0.375,-0.59765625,-0.05078125,0.6328125,0.81640625,-0.33984375,-0.1328125,-0.4453125,1.90625,-1.375,-0.11328125,-0.6875,-0.08935547,-0.796875,0.71484375,0.62109375,0.60546875,-0.07861328,-1.2421875,1.609375,-0.21484375,1.8671875,-0.34375,-1.7421875,0.8515625,-1.171875,0.53515625,-0.53125,1.265625,-0.10449219,-1.1640625,-0.09814453,0.6171875,-0.15527344,1.9296875,1.390625,0.38671875,-1.0,-1.46875,1.3828125,-0.20019531,0.59765625,-1.2890625,-1.3671875,-0.3359375,0.19433594,1.40625,-0.81640625,0.36132812,1.9609375,-0.03540039,-0.8203125,0.62109375,-1.546875,0.93359375,-0.17773438,0.890625,0.734375,0.9921875,-2.46875,0.013305664,2.0,1.03125,-1.3984375,0.35546875,0.91015625,-1.6640625,-1.0703125,0.18652344,0.578125,-0.36132812,1.546875,-1.5078125,-0.359375,-1.21875,0.95703125,-0.045410156,-0.3046875,-1.296875,2.125,0.44140625,-0.59765625,-1.2734375,0.60546875,0.41992188,-0.73046875,-2.1875,0.26367188,-0.36523438,0.18261719,-1.1640625,-0.54296875,-0.79296875,-1.84375,1.65625,0.30859375,-0.62109375,-0.53125,-1.34375,1.8203125,-1.6953125,-0.68359375,-0.3828125,-0.17871094,1.3125,-0.58203125,0.51953125,-2.6875,-0.1171875,0.25976562,1.125,0.69921875,-1.0390625,-0.22753906,-0.78515625,0.90234375,0.47851562,1.1171875,0.54296875,-0.41015625,0.3671875,-0.26171875,0.91796875,-1.5703125,1.28125,-1.109375,-0.58203125,1.3984375,0.08935547,-1.6796875,-0.79296875,-0.64453125,-0.34960938,0.16503906,0.36328125,0.80078125,-0.4609375,0.31054688,-0.828125,-0.49804688,-0.03466797,0.640625,-0.48828125,1.234375,-1.5546875,0.47070312,0.61328125,0.24902344,-0.85546875,1.09375,1.546875,0.86328125,-0.7421875,-0.7109375,-0.67578125,2.21875,1.6484375,1.1953125,-0.23925781
+2,0,0,1.203125,-0.09814453,-1.3359375,-0.6015625,0.96484375,-1.4765625,-0.8203125,3.140625,-0.5234375,-1.671875,0.023925781,1.5859375,-0.53125,-0.74609375,-0.57421875,1.7421875,-0.26953125,-1.3984375,0.059326172,0.48828125,-1.3671875,-0.6796875,-0.44921875,1.234375,-0.09033203,0.32617188,-2.5,-0.15136719,-0.8828125,-0.06982422,-0.022216797,0.78515625,0.69140625,0.7734375,-0.07470703,-1.0546875,1.2578125,-0.17285156,0.515625,0.6875,-0.0625,-0.008483887,0.11669922,-0.7890625,-0.11816406,0.65625,-0.46875,0.96875,-0.43945312,2.21875,-0.60546875,-2.21875,-1.3359375,-0.04736328,-0.15039062,-0.33203125,0.91796875,0.296875,0.40039062,-1.046875,-0.36523438,1.484375,-0.1328125,0.53515625,-0.265625,-0.03125,-0.002380371,1.4921875,0.16699219,-0.11767578,-0.4140625,-0.29296875,0.8984375,-0.671875,1.625,2.140625,0.625,0.2734375,-0.2421875,-0.75,1.671875,0.75,0.53515625,-0.62890625,-0.44140625,-1.1640625,1.328125,1.828125,0.49414062,1.65625,1.7421875,-0.7734375,-1.046875,1.3671875,-0.28710938,0.03100586,-1.7265625,-0.734375,-1.0859375,-0.21777344,-1.0625,-0.24414062,1.4375,0.94140625,-0.47265625,0.25195312,-1.640625,-0.62109375,-0.29101562,0.0234375,-1.140625,-1.9296875,1.4140625,-2.265625,0.84765625,0.27929688,-2.046875,1.2734375,-0.30078125,-1.21875,1.25,-0.4765625,-0.92578125,-0.36523438,0.4140625,2.75,-0.43164062,-0.40625,0.44335938,-1.8671875,-1.4765625,-1.640625,0.36914062,-1.1015625,-0.890625,-0.012817383,1.328125,0.34570312,0.43554688,-1.3515625,1.46875,-1.8203125,-0.62890625,-0.15527344,0.16113281,0.46875,-0.32421875,0.83203125,-0.73828125,0.83203125,-0.6171875,-0.018188477,0.8125,-0.4453125,0.045898438,-0.7578125,0.3203125,0.015563965,1.65625,1.015625,-1.1484375,1.875,1.2890625,0.8203125,-0.57421875,0.77734375,0.16894531,1.3125,1.2578125,0.76953125,-1.296875,-0.17089844,0.060546875,1.0859375,-1.28125,0.43359375,0.3515625,-0.69140625,-0.059326172,-1.171875,0.45898438,0.38476562,0.30273438,-1.09375,0.22363281,-0.24414062,-1.3984375,0.0134887695,-0.83984375,-1.3515625,0.20996094,1.421875,-0.11621094,-0.43164062,0.12207031,1.1953125,0.49609375,0.6015625,-1.0703125,0.4609375
+2,0,1,0.31445312,-0.33398438,-0.72265625,-0.4453125,0.40429688,-0.546875,-0.006439209,1.5859375,-0.13769531,-0.828125,-0.53515625,-0.375,1.2265625,0.49804688,-1.6484375,0.44726562,-0.5234375,-1.984375,0.42578125,-0.6875,0.5625,-0.32421875,-0.27734375,-0.06640625,-1.328125,-1.8203125,-1.640625,-0.08984375,-2.078125,0.49023438,-0.6875,-1.1171875,0.5625,0.07519531,0.546875,1.6796875,0.62109375,-1.453125,-1.8671875,1.1953125,-1.4765625,0.3515625,-0.54296875,-0.043701172,0.23144531,1.0625,0.5625,-0.546875,0.24023438,1.71875,-2.109375,-1.328125,-1.40625,0.73046875,-1.9921875,0.62109375,0.125,-0.66796875,-0.19140625,-2.140625,1.1796875,-0.16113281,0.42578125,-0.072265625,0.17675781,0.055908203,-1.0859375,0.18847656,-0.671875,2.0,0.90625,-1.6015625,-0.4609375,-0.49609375,0.044189453,1.921875,1.1171875,1.109375,-0.61328125,-0.8828125,1.5546875,0.31835938,0.27734375,-0.5078125,-0.67578125,-0.030639648,0.28710938,0.9609375,-0.66796875,0.67578125,2.125,-0.17285156,0.14941406,0.41796875,-1.390625,-0.19921875,-0.30273438,-0.53125,1.4140625,-0.328125,-2.25,0.14160156,2.25,1.15625,-0.46679688,-0.22851562,1.484375,-0.9296875,0.40039062,0.71484375,0.2734375,0.36328125,0.8515625,-1.4296875,0.30664062,-0.46875,0.18554688,-0.27148438,-0.65234375,-0.65625,1.328125,-0.55078125,0.4140625,-1.1875,-0.24609375,1.2734375,-0.44921875,-2.296875,-0.005645752,-0.041015625,0.671875,-0.061767578,0.19238281,-1.3046875,-1.34375,-0.19921875,-0.35351562,-0.47070312,-0.019042969,-1.8046875,3.0625,-0.76953125,0.39257812,-0.053955078,-0.06933594,1.828125,-1.234375,-0.01977539,-1.9375,1.921875,0.23632812,1.125,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.4375,1.6171875,1.4296875,0.671875,-0.71484375,1.3515625,1.375,0.49804688,-1.3359375,0.53515625,-1.1171875,-0.33984375,2.375,-0.25195312,0.2265625,-0.14550781,-0.66796875,-0.31835938,0.118652344,0.453125,-0.9375,-0.91796875,0.93359375,-0.58984375,-0.18164062,-0.10888672,1.265625,0.48828125,0.6640625,-0.30273438,1.3203125,0.82421875,-0.5390625,-1.0390625,-0.23242188,0.53125,2.15625,0.024658203,0.83203125,-0.15625,1.40625,-0.20996094,1.21875,0.27148438
+2,0,2,0.76953125,0.111328125,-0.36523438,-0.22363281,0.20703125,-0.7578125,-0.2734375,1.4140625,0.5546875,-1.09375,0.5859375,1.4140625,0.4375,0.54296875,-1.6015625,0.8046875,0.19335938,-1.2109375,-0.11230469,0.119628906,0.98828125,-0.5390625,-0.39257812,0.578125,-0.29296875,-0.73046875,-0.51171875,0.06201172,-1.0078125,-0.059814453,-0.68359375,-0.51171875,1.78125,-0.45117188,1.296875,1.59375,0.83203125,-0.16015625,-2.21875,1.7265625,-1.8359375,0.375,-0.59765625,-0.05078125,0.6328125,0.81640625,-0.33984375,-0.1328125,-0.4453125,1.90625,-1.375,-0.11328125,-0.6875,-0.08935547,-0.796875,0.71484375,0.62109375,0.60546875,-0.07861328,-1.2421875,1.609375,-0.21484375,1.8671875,-0.34375,-1.7421875,0.8515625,-1.171875,0.53515625,-0.53125,1.265625,-0.10449219,-1.1640625,-0.09814453,0.6171875,-0.15527344,1.9296875,1.390625,0.38671875,-1.0,-1.46875,1.3828125,-0.20019531,0.59765625,-1.2890625,-1.3671875,-0.3359375,0.19433594,1.40625,-0.81640625,0.36132812,1.9609375,-0.03540039,-0.8203125,0.62109375,-1.546875,0.93359375,-0.17773438,0.890625,0.734375,0.9921875,-2.46875,0.013305664,2.0,1.03125,-1.3984375,0.35546875,0.91015625,-1.6640625,-1.0703125,0.18652344,0.578125,-0.36132812,1.546875,-1.5078125,-0.359375,-1.21875,0.95703125,-0.045410156,-0.3046875,-1.296875,2.125,0.44140625,-0.59765625,-1.2734375,0.60546875,0.41992188,-0.73046875,-2.1875,0.26367188,-0.36523438,0.18261719,-1.1640625,-0.54296875,-0.79296875,-1.84375,1.65625,0.30859375,-0.62109375,-0.53125,-1.34375,1.8203125,-1.6953125,-0.68359375,-0.3828125,-0.17871094,1.3125,-0.58203125,0.51953125,-2.6875,-0.1171875,0.25976562,1.125,0.69921875,-1.0390625,-0.22753906,-0.78515625,0.90234375,0.47851562,1.1171875,0.54296875,-0.41015625,0.3671875,-0.26171875,0.91796875,-1.5703125,1.28125,-1.109375,-0.58203125,1.3984375,0.08935547,-1.6796875,-0.79296875,-0.64453125,-0.34960938,0.16503906,0.36328125,0.80078125,-0.4609375,0.31054688,-0.828125,-0.49804688,-0.03466797,0.640625,-0.48828125,1.234375,-1.5546875,0.47070312,0.61328125,0.24902344,-0.85546875,1.09375,1.546875,0.86328125,-0.7421875,-0.7109375,-0.67578125,2.21875,1.6484375,1.1953125,-0.23925781
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
index 1f353e1c..d4d5cdcd 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,5,1.0859375,-0.07324219,-1.484375,0.24511719,0.85546875,-1.703125,-0.57421875,2.953125,-0.0625,-0.76171875,0.23046875,2.65625,0.26757812,-0.36328125,-0.69921875,1.734375,0.2734375,-1.328125,0.359375,0.92578125,-0.24414062,-0.73828125,-0.65625,0.5390625,0.9375,1.3359375,-2.703125,0.38671875,-0.59375,-1.0234375,0.34375,-0.045166016,1.6953125,1.34375,0.54296875,-0.62890625,0.99609375,-0.021362305,-0.40820312,0.10644531,0.04736328,0.30859375,-0.17773438,-1.6484375,-0.13574219,0.8828125,-1.5625,0.44726562,0.07519531,0.79296875,-0.84765625,-1.421875,-1.171875,-1.046875,-0.19042969,-0.20800781,0.74609375,0.30078125,-0.24511719,-1.0625,0.21191406,0.71484375,1.3203125,0.8125,-1.265625,-0.32226562,0.30859375,0.27929688,-0.57421875,-0.40234375,0.41601562,-0.42382812,0.49609375,-0.47070312,2.28125,3.09375,1.046875,0.02355957,-0.81640625,-0.84765625,1.453125,0.6875,0.82421875,-1.6015625,0.04321289,-0.3984375,0.47070312,1.328125,0.578125,2.203125,1.875,-0.765625,-1.015625,1.359375,-0.61328125,-0.08691406,-1.3984375,0.0048828125,-0.14941406,0.6796875,-0.51171875,0.16015625,0.052734375,1.265625,-1.046875,-0.18261719,-1.125,0.028930664,-1.3828125,-0.35351562,-1.125,-1.53125,1.765625,-1.8984375,0.61328125,-0.03857422,-1.4609375,0.69140625,-0.0063476562,-0.39453125,1.2734375,0.11425781,-0.890625,-0.11328125,-0.04321289,2.421875,-0.75,-0.62109375,1.3828125,-1.5390625,-1.171875,-2.828125,0.103027344,-0.890625,-1.2109375,0.33984375,1.15625,-0.27539062,0.18066406,-1.171875,0.92578125,-1.2734375,-1.3984375,-0.54296875,-0.19824219,0.54296875,-0.82421875,0.765625,-0.34960938,0.026489258,0.41992188,-0.5390625,0.21484375,-1.2109375,0.28125,-0.12695312,0.96875,0.041503906,1.3359375,0.58984375,-0.18261719,2.46875,0.421875,0.94921875,-0.05883789,0.40039062,0.18945312,0.4765625,1.2734375,1.265625,-1.0234375,-0.640625,-0.88671875,0.91015625,-1.2890625,0.29492188,0.14355469,-0.30078125,0.23828125,-0.30078125,-0.5390625,-0.328125,0.49023438,-1.2109375,0.546875,-0.5234375,-1.1015625,-0.91015625,-0.515625,-0.97265625,0.57421875,0.96484375,-0.42773438,-0.8046875,-0.16699219,1.921875,-0.18457031,0.93359375,-1.453125,-0.171875
-3,0,6,-0.31054688,0.4609375,-1.6953125,-1.1796875,-0.61328125,0.16113281,-0.45703125,2.109375,0.5625,-0.8203125,1.734375,1.125,1.453125,-0.36328125,0.07763672,0.9609375,-0.07128906,-0.9375,0.76171875,-0.53515625,0.62109375,-0.012634277,0.8671875,0.96875,-0.055664062,0.13867188,0.2265625,-0.30859375,-1.2890625,-1.7265625,-0.31054688,1.6953125,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.2421875,0.53515625,0.75,-0.20214844,-0.45703125,0.16113281,-1.5,0.27539062,0.84375,0.23828125,0.55078125,0.3984375,-0.30273438,-2.21875,0.80078125,-0.9140625,0.35546875,-0.85546875,-1.015625,-0.24023438,-0.61328125,0.28125,-0.050048828,-0.12060547,0.86328125,0.69140625,1.5546875,1.40625,0.08886719,-1.28125,0.5390625,-0.765625,0.5390625,-0.40625,0.12109375,0.484375,0.09765625,-0.62109375,2.421875,-0.609375,0.017333984,-0.14648438,-0.03149414,-0.88671875,0.09033203,0.99609375,-0.8359375,0.88671875,0.76171875,-0.6171875,-0.4140625,1.0859375,-0.34570312,1.390625,-0.24902344,-0.875,0.8359375,-0.1171875,-1.125,0.6796875,-0.83203125,-1.3984375,-1.2109375,-1.4140625,0.6796875,3.15625,-0.81640625,-1.234375,-1.1640625,-0.6796875,0.30664062,-0.15429688,0.91796875,-0.546875,-0.47070312,0.8828125,-1.25,1.7734375,-0.30273438,-0.5546875,-0.36523438,0.011047363,-0.64453125,-0.0018310547,-0.828125,0.45507812,-0.44140625,-1.03125,1.28125,-0.88671875,-1.6640625,1.6015625,0.53125,-0.93359375,-0.78125,0.37304688,-0.984375,-1.734375,1.5,0.69140625,0.3359375,-0.43359375,-1.96875,2.484375,-1.375,-1.0078125,-0.061523438,-0.18261719,-0.26171875,0.32421875,-1.59375,-1.1953125,0.24121094,-0.068359375,1.59375,1.421875,-2.140625,0.106933594,-0.18359375,0.41210938,-1.0234375,1.6796875,-0.58203125,0.42382812,-0.27734375,0.90234375,-0.71484375,-0.013977051,2.234375,-0.953125,-0.110839844,1.5859375,-0.2578125,-1.6171875,-1.1015625,-2.0625,-1.5546875,-0.734375,0.47460938,-0.030151367,0.001625061,-1.328125,-1.2734375,-0.12890625,-0.48632812,0.7421875,-0.03100586,0.734375,-1.0546875,-0.45507812,0.15820312,-1.09375,1.0703125,1.734375,-0.21289062,0.6953125,-0.46679688,-0.95703125,0.75,1.90625,1.609375,0.66015625,1.421875
-3,0,7,0.578125,0.41992188,-0.9765625,-0.27929688,-1.546875,-1.1796875,0.515625,1.3671875,-0.12890625,-0.91015625,1.6640625,1.390625,0.76171875,0.19921875,0.42773438,0.95703125,-0.8359375,-0.28320312,1.0625,-0.7265625,0.80078125,-0.29101562,2.28125,2.0,-0.28125,0.5546875,-0.17382812,-1.1640625,-0.57421875,-0.32617188,0.35546875,0.6484375,1.078125,2.125,-0.0016326904,-0.47460938,-0.20605469,0.5,0.60546875,-1.2734375,-1.0546875,-1.84375,1.0078125,-0.4765625,0.66796875,0.19921875,0.3515625,-0.19433594,-2.375,-0.13085938,-0.34765625,0.14355469,-1.6640625,0.95703125,1.625,-0.106933594,-0.53515625,-0.020629883,-0.53125,0.026367188,0.2421875,0.25976562,-1.40625,-1.1796875,1.1484375,1.53125,-1.078125,-0.23925781,1.0234375,0.092285156,-1.4296875,0.88671875,-0.88671875,0.49609375,0.9453125,0.65625,1.0703125,0.61328125,-0.67578125,0.39257812,-0.76953125,0.546875,1.7734375,0.037597656,1.0390625,0.057128906,-0.39453125,1.265625,0.06640625,0.828125,1.2265625,-0.90625,-1.3046875,0.52734375,-0.42578125,-0.75390625,1.5546875,-0.31640625,-0.5078125,-0.5078125,-0.921875,-1.0234375,2.40625,-0.26367188,-1.9453125,-1.6953125,0.7109375,-1.5234375,0.86328125,0.64453125,-0.26171875,-1.6171875,0.921875,-1.234375,0.27539062,-0.53125,0.028320312,0.421875,0.3515625,-1.3515625,1.34375,0.4765625,-0.31445312,-2.140625,0.45507812,2.015625,-0.734375,-0.119628906,1.796875,-1.59375,0.453125,1.234375,1.2109375,-2.625,-2.296875,-0.28710938,-1.015625,0.5625,-0.030273438,-1.4765625,0.89453125,-0.09814453,-0.28710938,0.296875,-0.40234375,0.54296875,-0.9921875,-0.30273438,-0.8046875,0.640625,-0.43945312,0.29492188,-0.69140625,-0.15136719,-0.29101562,-2.109375,-0.78125,-0.55859375,0.53515625,1.4609375,-0.80859375,-1.3828125,0.078125,-0.8515625,0.90625,-0.040039062,-2.0,-1.515625,2.046875,-0.59765625,-0.30664062,0.5390625,-1.5234375,-0.47265625,-0.22363281,-0.09326172,1.5390625,0.22265625,-0.34960938,0.11425781,0.4765625,1.7578125,0.47851562,-0.25390625,0.63671875,-0.7109375,-0.37695312,0.31640625,0.52734375,1.546875,0.8359375,0.60546875,-0.21386719,-0.50390625,0.40820312,-1.53125,0.76953125,0.12988281,1.8046875,0.21191406
-4,0,5,2.046875,0.015991211,-1.2265625,1.3203125,1.1015625,-1.015625,-0.7109375,1.546875,-0.13476562,-0.93359375,1.03125,1.0546875,1.0390625,0.23828125,-1.75,0.83984375,-0.37109375,-0.0859375,0.375,0.013183594,0.875,0.49804688,0.625,0.36523438,0.40429688,0.08935547,-1.7578125,0.33007812,-1.6640625,0.45507812,0.16796875,0.79296875,1.34375,1.765625,0.7109375,0.41015625,0.24316406,-1.2421875,0.734375,0.91015625,-0.80078125,0.203125,-1.0546875,-0.734375,0.40429688,-0.106933594,1.2890625,-0.21386719,-0.81640625,-0.78125,-0.4453125,-1.1953125,-2.203125,-1.5390625,-1.6796875,-0.8203125,0.27929688,0.29296875,0.3828125,-0.6953125,0.09716797,0.23046875,0.049072266,0.76171875,-0.21875,-1.2578125,-0.049560547,0.20996094,0.13867188,0.92578125,1.1875,-2.8125,0.037597656,-0.033935547,0.62109375,2.046875,0.40039062,-0.6484375,0.1484375,-1.9375,-0.48046875,0.29492188,0.114746094,-1.640625,1.1328125,0.6484375,0.25195312,-0.3671875,1.2734375,-0.0703125,1.9609375,0.92578125,0.002456665,0.47460938,0.49804688,-1.546875,-0.19433594,-0.63671875,-0.67578125,-0.064941406,-0.90625,1.3125,0.94140625,-0.29296875,-0.6328125,-1.875,-1.2578125,-0.1796875,-0.12109375,1.3984375,-0.37695312,0.5078125,0.38476562,-0.3203125,1.0,1.296875,-0.78125,-0.859375,0.071777344,-0.265625,-0.12597656,-0.7265625,-1.0859375,0.8359375,-1.3828125,1.5234375,-0.8984375,-1.0859375,0.1484375,0.29101562,0.6171875,-0.7890625,-0.3984375,-0.62109375,-0.84375,0.6953125,0.13183594,-1.1640625,0.48632812,-2.28125,1.96875,-1.984375,-1.1015625,0.44726562,0.76953125,0.31054688,-1.3359375,0.671875,1.4296875,2.46875,-0.029663086,0.85546875,0.45703125,0.2890625,1.53125,1.0859375,-0.953125,0.27148438,0.78125,-0.98046875,-0.18945312,1.4921875,1.0,1.15625,-0.6640625,0.88671875,-0.26757812,-0.15527344,1.125,-1.1328125,-1.203125,-1.375,0.09814453,-1.828125,-1.6484375,-0.82421875,0.35546875,-1.8515625,0.5390625,-0.09033203,-0.49023438,-0.88671875,0.5234375,-1.9921875,1.5,0.39648438,0.58984375,-0.5,-0.052001953,-2.0625,1.1171875,-0.055664062,1.15625,-0.47265625,-0.171875,2.65625,0.6171875,0.014770508,0.3046875,1.015625
-4,0,6,0.38867188,1.015625,-0.84375,-1.046875,1.4921875,0.36523438,-1.2578125,1.28125,0.83203125,-0.13476562,0.16894531,0.7109375,-0.65234375,0.14550781,-0.390625,-0.63671875,-1.1953125,0.2109375,1.8671875,1.171875,0.6640625,-0.39257812,0.9375,0.6796875,-0.84375,-0.7734375,-2.078125,-1.0625,-2.015625,-0.031982422,-0.38671875,0.07421875,0.22460938,1.0546875,-0.57421875,0.09472656,0.38085938,-0.24316406,0.546875,0.86328125,-1.6796875,0.47265625,-0.044433594,0.08251953,0.26953125,0.37890625,0.7890625,-0.875,0.703125,0.65234375,0.984375,-1.2734375,0.107910156,0.28320312,-1.9921875,-0.6484375,-0.34570312,-1.0078125,0.62109375,-1.390625,0.2890625,0.76171875,2.03125,-0.73046875,0.47851562,-1.0546875,-0.05053711,-1.2734375,-0.16992188,1.484375,0.49804688,1.4140625,0.044189453,-1.046875,-0.04296875,0.83203125,-0.37695312,0.3046875,-0.71484375,0.064941406,0.51171875,-0.9140625,0.46484375,-0.32617188,-0.37890625,-0.26367188,-0.50390625,-1.3359375,1.5078125,0.38671875,1.3125,1.75,-1.0,1.453125,-0.53515625,-0.7421875,-0.4140625,-0.028686523,-0.98046875,-0.09326172,-0.62890625,1.234375,1.6640625,-1.609375,-1.1015625,0.5859375,-0.9296875,0.11279297,0.31054688,0.021728516,0.35351562,0.3984375,0.106933594,0.38867188,1.0,-0.53125,-0.5703125,-0.3203125,0.67578125,0.07324219,0.25195312,-0.22070312,-1.0,0.22167969,-1.15625,2.078125,-0.1875,-2.203125,1.359375,2.84375,0.62109375,-1.625,-2.03125,0.47851562,-1.6875,-0.69921875,-1.1640625,-1.4765625,0.41015625,-1.6875,2.140625,-2.296875,-0.98828125,1.875,0.796875,-0.75390625,0.7734375,0.22265625,-0.37695312,1.75,0.375,-0.22753906,1.1640625,-0.37695312,0.051513672,0.83203125,1.6875,0.35351562,0.7578125,-2.046875,0.875,1.203125,0.9765625,2.203125,-1.03125,1.03125,-1.6171875,-1.421875,2.03125,-0.81640625,0.39453125,-1.515625,0.11816406,-0.1875,-0.72265625,-1.0390625,-0.49023438,-1.3828125,0.22851562,0.0050354004,-0.8203125,-0.61328125,0.8359375,0.6015625,-0.35351562,1.2265625,-0.06347656,-0.56640625,1.0625,-0.9765625,-1.109375,0.034423828,0.45898438,-0.5859375,-1.03125,1.953125,0.79296875,0.66796875,-0.421875,0.6640625
-4,0,7,1.265625,-0.80859375,-0.62890625,0.19921875,-0.29101562,-0.51171875,-0.76953125,0.8828125,-0.671875,-0.6328125,0.32617188,1.1875,1.0546875,0.5234375,-0.66015625,1.515625,-0.31445312,-0.44921875,2.265625,-0.46289062,1.6953125,0.15039062,-0.20507812,0.05078125,2.1875,-0.5390625,-0.98046875,-1.3125,-0.5546875,0.49023438,-0.29882812,0.54296875,1.03125,-0.008605957,-1.984375,-0.13183594,0.5390625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.86328125,-1.0625,-0.23925781,-0.099609375,0.921875,0.19726562,-0.6796875,-1.2578125,-0.103515625,-0.44140625,-1.0078125,-1.59375,-0.59375,-0.50390625,1.9765625,0.48632812,0.014526367,0.23632812,-0.7109375,1.328125,-0.578125,0.5234375,1.5234375,0.58203125,0.18652344,-0.48632812,-0.4609375,2.265625,-0.023071289,0.17675781,0.029907227,-0.37890625,0.28515625,1.0859375,2.375,-0.25,0.083984375,-1.6484375,-0.34375,0.88671875,-0.51953125,-0.45703125,0.111328125,1.0,-0.15722656,0.31445312,1.578125,0.55859375,1.6015625,1.578125,-0.40429688,0.28710938,0.97265625,-1.7265625,1.5703125,-0.90625,-0.034179688,0.375,-0.22167969,-0.83984375,0.19335938,-0.22460938,0.28710938,-0.92578125,-0.29882812,-0.48828125,-2.6875,-0.64453125,0.328125,0.48046875,-0.953125,0.65234375,-0.6796875,-0.61328125,-0.100097656,-0.79296875,0.96875,-0.24511719,0.21386719,0.77734375,1.15625,-0.25195312,-1.7265625,-1.2109375,0.8984375,-1.421875,-0.94921875,0.022949219,0.23046875,0.97265625,-0.30273438,-0.2890625,-1.28125,-0.60546875,-0.20898438,1.8515625,-0.62890625,-1.6796875,-2.625,2.171875,0.64453125,-1.078125,-0.12597656,-0.83984375,-0.20507812,0.671875,-0.87109375,0.296875,1.9296875,-0.36523438,-0.111816406,1.3125,-2.09375,-0.30859375,-1.6015625,1.609375,1.265625,1.40625,-0.84765625,-0.48242188,-0.8046875,0.37890625,-0.28710938,-1.140625,0.18652344,0.096191406,-1.0859375,1.8359375,-0.390625,0.30859375,-0.28125,-1.2890625,0.4609375,-0.51171875,1.5859375,0.21289062,-0.12695312,0.96484375,0.09326172,-1.328125,0.22460938,-0.020507812,0.5,0.018676758,-0.546875,1.0,-0.44335938,1.5625,-0.89453125,-0.15917969,-0.53515625,1.25,-0.67578125,-0.81640625,0.48632812,0.7109375,2.765625,2.171875,0.546875
+3,0,0,1.0859375,-0.07324219,-1.484375,0.24511719,0.85546875,-1.703125,-0.57421875,2.953125,-0.0625,-0.76171875,0.23046875,2.65625,0.26757812,-0.36328125,-0.69921875,1.734375,0.2734375,-1.328125,0.359375,0.92578125,-0.24414062,-0.73828125,-0.65625,0.5390625,0.9375,1.3359375,-2.703125,0.38671875,-0.59375,-1.0234375,0.34375,-0.045166016,1.6953125,1.34375,0.54296875,-0.62890625,0.99609375,-0.021362305,-0.40820312,0.10644531,0.04736328,0.30859375,-0.17773438,-1.6484375,-0.13574219,0.8828125,-1.5625,0.44726562,0.07519531,0.79296875,-0.84765625,-1.421875,-1.171875,-1.046875,-0.19042969,-0.20800781,0.74609375,0.30078125,-0.24511719,-1.0625,0.21191406,0.71484375,1.3203125,0.8125,-1.265625,-0.32226562,0.30859375,0.27929688,-0.57421875,-0.40234375,0.41601562,-0.42382812,0.49609375,-0.47070312,2.28125,3.09375,1.046875,0.02355957,-0.81640625,-0.84765625,1.453125,0.6875,0.82421875,-1.6015625,0.04321289,-0.3984375,0.47070312,1.328125,0.578125,2.203125,1.875,-0.765625,-1.015625,1.359375,-0.61328125,-0.08691406,-1.3984375,0.0048828125,-0.14941406,0.6796875,-0.51171875,0.16015625,0.052734375,1.265625,-1.046875,-0.18261719,-1.125,0.028930664,-1.3828125,-0.35351562,-1.125,-1.53125,1.765625,-1.8984375,0.61328125,-0.03857422,-1.4609375,0.69140625,-0.0063476562,-0.39453125,1.2734375,0.11425781,-0.890625,-0.11328125,-0.04321289,2.421875,-0.75,-0.62109375,1.3828125,-1.5390625,-1.171875,-2.828125,0.103027344,-0.890625,-1.2109375,0.33984375,1.15625,-0.27539062,0.18066406,-1.171875,0.92578125,-1.2734375,-1.3984375,-0.54296875,-0.19824219,0.54296875,-0.82421875,0.765625,-0.34960938,0.026489258,0.41992188,-0.5390625,0.21484375,-1.2109375,0.28125,-0.12695312,0.96875,0.041503906,1.3359375,0.58984375,-0.18261719,2.46875,0.421875,0.94921875,-0.05883789,0.40039062,0.18945312,0.4765625,1.2734375,1.265625,-1.0234375,-0.640625,-0.88671875,0.91015625,-1.2890625,0.29492188,0.14355469,-0.30078125,0.23828125,-0.30078125,-0.5390625,-0.328125,0.49023438,-1.2109375,0.546875,-0.5234375,-1.1015625,-0.91015625,-0.515625,-0.97265625,0.57421875,0.96484375,-0.42773438,-0.8046875,-0.16699219,1.921875,-0.18457031,0.93359375,-1.453125,-0.171875
+3,0,1,-0.31054688,0.4609375,-1.6953125,-1.1796875,-0.61328125,0.16113281,-0.45703125,2.109375,0.5625,-0.8203125,1.734375,1.125,1.453125,-0.36328125,0.07763672,0.9609375,-0.07128906,-0.9375,0.76171875,-0.53515625,0.62109375,-0.012634277,0.8671875,0.96875,-0.055664062,0.13867188,0.2265625,-0.30859375,-1.2890625,-1.7265625,-0.31054688,1.6953125,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.2421875,0.53515625,0.75,-0.20214844,-0.45703125,0.16113281,-1.5,0.27539062,0.84375,0.23828125,0.55078125,0.3984375,-0.30273438,-2.21875,0.80078125,-0.9140625,0.35546875,-0.85546875,-1.015625,-0.24023438,-0.61328125,0.28125,-0.050048828,-0.12060547,0.86328125,0.69140625,1.5546875,1.40625,0.08886719,-1.28125,0.5390625,-0.765625,0.5390625,-0.40625,0.12109375,0.484375,0.09765625,-0.62109375,2.421875,-0.609375,0.017333984,-0.14648438,-0.03149414,-0.88671875,0.09033203,0.99609375,-0.8359375,0.88671875,0.76171875,-0.6171875,-0.4140625,1.0859375,-0.34570312,1.390625,-0.24902344,-0.875,0.8359375,-0.1171875,-1.125,0.6796875,-0.83203125,-1.3984375,-1.2109375,-1.4140625,0.6796875,3.15625,-0.81640625,-1.234375,-1.1640625,-0.6796875,0.30664062,-0.15429688,0.91796875,-0.546875,-0.47070312,0.8828125,-1.25,1.7734375,-0.30273438,-0.5546875,-0.36523438,0.011047363,-0.64453125,-0.0018310547,-0.828125,0.45507812,-0.44140625,-1.03125,1.28125,-0.88671875,-1.6640625,1.6015625,0.53125,-0.93359375,-0.78125,0.37304688,-0.984375,-1.734375,1.5,0.69140625,0.3359375,-0.43359375,-1.96875,2.484375,-1.375,-1.0078125,-0.061523438,-0.18261719,-0.26171875,0.32421875,-1.59375,-1.1953125,0.24121094,-0.068359375,1.59375,1.421875,-2.140625,0.106933594,-0.18359375,0.41210938,-1.0234375,1.6796875,-0.58203125,0.42382812,-0.27734375,0.90234375,-0.71484375,-0.013977051,2.234375,-0.953125,-0.110839844,1.5859375,-0.2578125,-1.6171875,-1.1015625,-2.0625,-1.5546875,-0.734375,0.47460938,-0.030151367,0.001625061,-1.328125,-1.2734375,-0.12890625,-0.48632812,0.7421875,-0.03100586,0.734375,-1.0546875,-0.45507812,0.15820312,-1.09375,1.0703125,1.734375,-0.21289062,0.6953125,-0.46679688,-0.95703125,0.75,1.90625,1.609375,0.66015625,1.421875
+3,0,2,0.578125,0.41992188,-0.9765625,-0.27929688,-1.546875,-1.1796875,0.515625,1.3671875,-0.12890625,-0.91015625,1.6640625,1.390625,0.76171875,0.19921875,0.42773438,0.95703125,-0.8359375,-0.28320312,1.0625,-0.7265625,0.80078125,-0.29101562,2.28125,2.0,-0.28125,0.5546875,-0.17382812,-1.1640625,-0.57421875,-0.32617188,0.35546875,0.6484375,1.078125,2.125,-0.0016326904,-0.47460938,-0.20605469,0.5,0.60546875,-1.2734375,-1.0546875,-1.84375,1.0078125,-0.4765625,0.66796875,0.19921875,0.3515625,-0.19433594,-2.375,-0.13085938,-0.34765625,0.14355469,-1.6640625,0.95703125,1.625,-0.106933594,-0.53515625,-0.020629883,-0.53125,0.026367188,0.2421875,0.25976562,-1.40625,-1.1796875,1.1484375,1.53125,-1.078125,-0.23925781,1.0234375,0.092285156,-1.4296875,0.88671875,-0.88671875,0.49609375,0.9453125,0.65625,1.0703125,0.61328125,-0.67578125,0.39257812,-0.76953125,0.546875,1.7734375,0.037597656,1.0390625,0.057128906,-0.39453125,1.265625,0.06640625,0.828125,1.2265625,-0.90625,-1.3046875,0.52734375,-0.42578125,-0.75390625,1.5546875,-0.31640625,-0.5078125,-0.5078125,-0.921875,-1.0234375,2.40625,-0.26367188,-1.9453125,-1.6953125,0.7109375,-1.5234375,0.86328125,0.64453125,-0.26171875,-1.6171875,0.921875,-1.234375,0.27539062,-0.53125,0.028320312,0.421875,0.3515625,-1.3515625,1.34375,0.4765625,-0.31445312,-2.140625,0.45507812,2.015625,-0.734375,-0.119628906,1.796875,-1.59375,0.453125,1.234375,1.2109375,-2.625,-2.296875,-0.28710938,-1.015625,0.5625,-0.030273438,-1.4765625,0.89453125,-0.09814453,-0.28710938,0.296875,-0.40234375,0.54296875,-0.9921875,-0.30273438,-0.8046875,0.640625,-0.43945312,0.29492188,-0.69140625,-0.15136719,-0.29101562,-2.109375,-0.78125,-0.55859375,0.53515625,1.4609375,-0.80859375,-1.3828125,0.078125,-0.8515625,0.90625,-0.040039062,-2.0,-1.515625,2.046875,-0.59765625,-0.30664062,0.5390625,-1.5234375,-0.47265625,-0.22363281,-0.09326172,1.5390625,0.22265625,-0.34960938,0.11425781,0.4765625,1.7578125,0.47851562,-0.25390625,0.63671875,-0.7109375,-0.37695312,0.31640625,0.52734375,1.546875,0.8359375,0.60546875,-0.21386719,-0.50390625,0.40820312,-1.53125,0.76953125,0.12988281,1.8046875,0.21191406
+4,0,0,2.046875,0.015991211,-1.2265625,1.3203125,1.1015625,-1.015625,-0.7109375,1.546875,-0.13476562,-0.93359375,1.03125,1.0546875,1.0390625,0.23828125,-1.75,0.83984375,-0.37109375,-0.0859375,0.375,0.013183594,0.875,0.49804688,0.625,0.36523438,0.40429688,0.08935547,-1.7578125,0.33007812,-1.6640625,0.45507812,0.16796875,0.79296875,1.34375,1.765625,0.7109375,0.41015625,0.24316406,-1.2421875,0.734375,0.91015625,-0.80078125,0.203125,-1.0546875,-0.734375,0.40429688,-0.106933594,1.2890625,-0.21386719,-0.81640625,-0.78125,-0.4453125,-1.1953125,-2.203125,-1.5390625,-1.6796875,-0.8203125,0.27929688,0.29296875,0.3828125,-0.6953125,0.09716797,0.23046875,0.049072266,0.76171875,-0.21875,-1.2578125,-0.049560547,0.20996094,0.13867188,0.92578125,1.1875,-2.8125,0.037597656,-0.033935547,0.62109375,2.046875,0.40039062,-0.6484375,0.1484375,-1.9375,-0.48046875,0.29492188,0.114746094,-1.640625,1.1328125,0.6484375,0.25195312,-0.3671875,1.2734375,-0.0703125,1.9609375,0.92578125,0.002456665,0.47460938,0.49804688,-1.546875,-0.19433594,-0.63671875,-0.67578125,-0.064941406,-0.90625,1.3125,0.94140625,-0.29296875,-0.6328125,-1.875,-1.2578125,-0.1796875,-0.12109375,1.3984375,-0.37695312,0.5078125,0.38476562,-0.3203125,1.0,1.296875,-0.78125,-0.859375,0.071777344,-0.265625,-0.12597656,-0.7265625,-1.0859375,0.8359375,-1.3828125,1.5234375,-0.8984375,-1.0859375,0.1484375,0.29101562,0.6171875,-0.7890625,-0.3984375,-0.62109375,-0.84375,0.6953125,0.13183594,-1.1640625,0.48632812,-2.28125,1.96875,-1.984375,-1.1015625,0.44726562,0.76953125,0.31054688,-1.3359375,0.671875,1.4296875,2.46875,-0.029663086,0.85546875,0.45703125,0.2890625,1.53125,1.0859375,-0.953125,0.27148438,0.78125,-0.98046875,-0.18945312,1.4921875,1.0,1.15625,-0.6640625,0.88671875,-0.26757812,-0.15527344,1.125,-1.1328125,-1.203125,-1.375,0.09814453,-1.828125,-1.6484375,-0.82421875,0.35546875,-1.8515625,0.5390625,-0.09033203,-0.49023438,-0.88671875,0.5234375,-1.9921875,1.5,0.39648438,0.58984375,-0.5,-0.052001953,-2.0625,1.1171875,-0.055664062,1.15625,-0.47265625,-0.171875,2.65625,0.6171875,0.014770508,0.3046875,1.015625
+4,0,1,0.38867188,1.015625,-0.84375,-1.046875,1.4921875,0.36523438,-1.2578125,1.28125,0.83203125,-0.13476562,0.16894531,0.7109375,-0.65234375,0.14550781,-0.390625,-0.63671875,-1.1953125,0.2109375,1.8671875,1.171875,0.6640625,-0.39257812,0.9375,0.6796875,-0.84375,-0.7734375,-2.078125,-1.0625,-2.015625,-0.031982422,-0.38671875,0.07421875,0.22460938,1.0546875,-0.57421875,0.09472656,0.38085938,-0.24316406,0.546875,0.86328125,-1.6796875,0.47265625,-0.044433594,0.08251953,0.26953125,0.37890625,0.7890625,-0.875,0.703125,0.65234375,0.984375,-1.2734375,0.107910156,0.28320312,-1.9921875,-0.6484375,-0.34570312,-1.0078125,0.62109375,-1.390625,0.2890625,0.76171875,2.03125,-0.73046875,0.47851562,-1.0546875,-0.05053711,-1.2734375,-0.16992188,1.484375,0.49804688,1.4140625,0.044189453,-1.046875,-0.04296875,0.83203125,-0.37695312,0.3046875,-0.71484375,0.064941406,0.51171875,-0.9140625,0.46484375,-0.32617188,-0.37890625,-0.26367188,-0.50390625,-1.3359375,1.5078125,0.38671875,1.3125,1.75,-1.0,1.453125,-0.53515625,-0.7421875,-0.4140625,-0.028686523,-0.98046875,-0.09326172,-0.62890625,1.234375,1.6640625,-1.609375,-1.1015625,0.5859375,-0.9296875,0.11279297,0.31054688,0.021728516,0.35351562,0.3984375,0.106933594,0.38867188,1.0,-0.53125,-0.5703125,-0.3203125,0.67578125,0.07324219,0.25195312,-0.22070312,-1.0,0.22167969,-1.15625,2.078125,-0.1875,-2.203125,1.359375,2.84375,0.62109375,-1.625,-2.03125,0.47851562,-1.6875,-0.69921875,-1.1640625,-1.4765625,0.41015625,-1.6875,2.140625,-2.296875,-0.98828125,1.875,0.796875,-0.75390625,0.7734375,0.22265625,-0.37695312,1.75,0.375,-0.22753906,1.1640625,-0.37695312,0.051513672,0.83203125,1.6875,0.35351562,0.7578125,-2.046875,0.875,1.203125,0.9765625,2.203125,-1.03125,1.03125,-1.6171875,-1.421875,2.03125,-0.81640625,0.39453125,-1.515625,0.11816406,-0.1875,-0.72265625,-1.0390625,-0.49023438,-1.3828125,0.22851562,0.0050354004,-0.8203125,-0.61328125,0.8359375,0.6015625,-0.35351562,1.2265625,-0.06347656,-0.56640625,1.0625,-0.9765625,-1.109375,0.034423828,0.45898438,-0.5859375,-1.03125,1.953125,0.79296875,0.66796875,-0.421875,0.6640625
+4,0,2,1.265625,-0.80859375,-0.62890625,0.19921875,-0.29101562,-0.51171875,-0.76953125,0.8828125,-0.671875,-0.6328125,0.32617188,1.1875,1.0546875,0.5234375,-0.66015625,1.515625,-0.31445312,-0.44921875,2.265625,-0.46289062,1.6953125,0.15039062,-0.20507812,0.05078125,2.1875,-0.5390625,-0.98046875,-1.3125,-0.5546875,0.49023438,-0.29882812,0.54296875,1.03125,-0.008605957,-1.984375,-0.13183594,0.5390625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.86328125,-1.0625,-0.23925781,-0.099609375,0.921875,0.19726562,-0.6796875,-1.2578125,-0.103515625,-0.44140625,-1.0078125,-1.59375,-0.59375,-0.50390625,1.9765625,0.48632812,0.014526367,0.23632812,-0.7109375,1.328125,-0.578125,0.5234375,1.5234375,0.58203125,0.18652344,-0.48632812,-0.4609375,2.265625,-0.023071289,0.17675781,0.029907227,-0.37890625,0.28515625,1.0859375,2.375,-0.25,0.083984375,-1.6484375,-0.34375,0.88671875,-0.51953125,-0.45703125,0.111328125,1.0,-0.15722656,0.31445312,1.578125,0.55859375,1.6015625,1.578125,-0.40429688,0.28710938,0.97265625,-1.7265625,1.5703125,-0.90625,-0.034179688,0.375,-0.22167969,-0.83984375,0.19335938,-0.22460938,0.28710938,-0.92578125,-0.29882812,-0.48828125,-2.6875,-0.64453125,0.328125,0.48046875,-0.953125,0.65234375,-0.6796875,-0.61328125,-0.100097656,-0.79296875,0.96875,-0.24511719,0.21386719,0.77734375,1.15625,-0.25195312,-1.7265625,-1.2109375,0.8984375,-1.421875,-0.94921875,0.022949219,0.23046875,0.97265625,-0.30273438,-0.2890625,-1.28125,-0.60546875,-0.20898438,1.8515625,-0.62890625,-1.6796875,-2.625,2.171875,0.64453125,-1.078125,-0.12597656,-0.83984375,-0.20507812,0.671875,-0.87109375,0.296875,1.9296875,-0.36523438,-0.111816406,1.3125,-2.09375,-0.30859375,-1.6015625,1.609375,1.265625,1.40625,-0.84765625,-0.48242188,-0.8046875,0.37890625,-0.28710938,-1.140625,0.18652344,0.096191406,-1.0859375,1.8359375,-0.390625,0.30859375,-0.28125,-1.2890625,0.4609375,-0.51171875,1.5859375,0.21289062,-0.12695312,0.96484375,0.09326172,-1.328125,0.22460938,-0.020507812,0.5,0.018676758,-0.546875,1.0,-0.44335938,1.5625,-0.89453125,-0.15917969,-0.53515625,1.25,-0.67578125,-0.81640625,0.48632812,0.7109375,2.765625,2.171875,0.546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
index f76bfe98..254cb8d4 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,5,0.58984375,0.99609375,-0.34570312,-1.3515625,-0.484375,-0.91015625,-0.88671875,0.4609375,0.94921875,-1.25,-0.1953125,-0.58203125,-1.09375,1.546875,-1.6640625,0.671875,-0.33007812,0.453125,-0.81640625,0.87109375,1.671875,0.70703125,0.06347656,0.34570312,0.41015625,-0.58984375,-1.8828125,-0.15136719,-1.0,1.2109375,-0.92578125,-0.22558594,0.83984375,0.20800781,-0.22753906,0.91015625,0.40039062,-0.060546875,-0.58984375,0.6171875,-1.8046875,0.38085938,-0.12109375,0.40625,-0.0045166016,1.34375,0.703125,0.13867188,-0.40039062,1.25,-0.41796875,-0.66796875,-1.515625,-0.021972656,-0.045654297,1.8203125,0.018310547,-0.23730469,0.40429688,-1.546875,0.29101562,0.39453125,1.2890625,0.2421875,0.9765625,1.1875,-1.5859375,1.4765625,0.013549805,-0.22753906,-0.80078125,-1.2109375,-0.765625,-0.07080078,-0.58203125,0.013427734,0.4921875,0.55859375,0.3125,0.64453125,1.078125,-1.046875,1.0,0.42382812,-0.53125,0.49804688,0.34375,0.62890625,-0.5859375,-0.50390625,3.578125,0.421875,0.79296875,1.2734375,-0.52734375,1.03125,-0.97265625,0.31640625,-0.578125,0.34179688,-0.82421875,0.17285156,1.375,-1.140625,0.04663086,0.17480469,-0.38671875,-1.859375,-0.4453125,0.828125,-0.15234375,-0.40429688,0.43554688,-0.8125,-0.6171875,-0.609375,-0.34570312,0.6953125,0.42773438,-0.125,0.73046875,-1.0703125,-0.62890625,-1.8359375,-0.82421875,1.6484375,-0.35742188,-1.6328125,-1.1015625,-0.96484375,1.3125,0.25585938,-2.375,-0.58984375,-1.5625,2.0,0.13769531,-0.45507812,-1.3515625,-2.5625,1.1015625,-2.25,0.5859375,-0.092285156,0.64453125,0.35351562,-0.20507812,-0.76171875,-0.087402344,1.5546875,-1.578125,0.59375,0.60546875,-0.10253906,1.2890625,-0.061279297,0.41601562,0.9375,0.796875,0.41992188,-0.87890625,-0.6640625,-0.37109375,0.09082031,-2.6875,2.40625,0.22851562,-0.012207031,1.421875,-0.91796875,-1.9140625,-0.8671875,-1.1171875,-0.92578125,-0.44921875,1.5078125,1.1875,-1.0078125,-0.18554688,-0.91015625,0.44921875,1.3203125,0.50390625,-0.041992188,1.0078125,-1.109375,-0.13476562,1.609375,0.58203125,-1.3125,-0.23535156,1.3671875,1.2265625,0.3828125,-0.55078125,0.44140625,3.265625,-0.100097656,-0.18652344,0.44726562
-5,0,6,-0.43945312,-0.69921875,-0.38476562,-1.4296875,0.18261719,-1.203125,-0.9453125,1.6484375,0.064453125,0.8125,-0.8671875,-0.53125,1.4140625,0.87890625,-2.140625,0.92578125,-0.515625,-0.86328125,0.10058594,0.5078125,2.453125,-0.18164062,-0.84765625,0.23925781,0.484375,-0.37890625,-2.0625,-1.1328125,-0.421875,1.7421875,-0.26367188,0.50390625,-0.98046875,0.546875,-1.2578125,0.49609375,-0.37304688,-0.49609375,-0.27734375,-0.6015625,-0.94921875,0.33203125,-0.6953125,-0.28515625,0.5390625,2.5,-0.10595703,-0.31640625,1.390625,1.0625,-1.0,-2.421875,-2.328125,0.4375,-0.609375,1.671875,-0.28125,-0.80859375,-0.26367188,-1.5546875,-0.87890625,-0.80859375,1.5859375,0.9609375,1.4296875,2.015625,-1.671875,1.6328125,-0.96875,0.62109375,1.296875,-1.296875,-0.69921875,-0.26367188,-0.23632812,0.24609375,0.26367188,-0.33789062,0.7265625,0.73046875,1.5703125,-0.43359375,0.27929688,1.75,-0.84375,0.64453125,-0.8203125,0.049316406,0.13769531,-0.11621094,1.4765625,-0.36132812,1.7734375,-0.103515625,-0.14453125,0.025878906,-1.84375,0.78125,-0.5390625,0.375,-0.28710938,-0.26953125,0.35742188,0.17578125,0.3671875,1.3125,-1.671875,-2.625,-1.0703125,-0.62109375,-0.19824219,0.7734375,-1.0546875,-0.53125,-0.100097656,-2.375,-0.09667969,0.81640625,0.33007812,1.5859375,0.11230469,-0.3828125,-0.15625,-1.65625,-1.265625,0.63671875,-1.359375,-0.77734375,-0.78515625,0.4765625,2.0625,-0.75390625,-1.71875,-0.5546875,-0.28515625,0.83984375,-0.13671875,0.079589844,-0.100097656,-2.046875,-0.328125,1.609375,0.58984375,0.95703125,0.52734375,0.15820312,-0.5234375,0.20507812,-0.59375,1.0390625,0.034423828,-0.90234375,0.54296875,-0.53515625,0.4375,-0.1796875,1.40625,1.0078125,0.55078125,0.359375,-0.19433594,0.875,0.41601562,0.49804688,-1.7890625,1.7265625,0.734375,0.20996094,0.45898438,0.037597656,-0.49609375,0.63671875,-0.039794922,0.18457031,-1.1171875,1.9921875,0.18261719,-1.6640625,-0.5703125,-0.09667969,-0.66015625,0.875,0.91796875,0.3828125,0.28515625,-0.28515625,0.27734375,0.3984375,0.32226562,-0.7890625,-1.4609375,0.7578125,1.5,0.15039062,-0.55078125,0.27734375,2.375,-0.39453125,1.015625,1.109375
-5,0,7,0.90234375,0.56640625,0.828125,0.19921875,-1.140625,-0.07861328,-1.1953125,-0.034179688,2.609375,-1.453125,1.8359375,-0.03930664,0.34960938,1.3671875,-0.625,-0.42578125,-0.46484375,0.18164062,2.265625,1.1484375,0.24023438,-0.51171875,-0.15625,0.578125,-0.48242188,-0.53515625,0.6484375,-0.52734375,0.90625,0.5078125,-0.40625,-0.008117676,0.78125,-0.6796875,0.38085938,0.91796875,0.953125,-0.19042969,-0.21191406,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24707031,-0.234375,0.50390625,-0.29101562,0.87890625,0.6796875,-0.16015625,-0.265625,-0.74609375,0.95703125,0.9375,0.41015625,-0.65234375,-0.49804688,-0.578125,0.7734375,1.1953125,-0.37890625,1.2578125,0.099609375,0.640625,0.51953125,-0.20703125,-1.2578125,-1.46875,0.41210938,-0.111328125,1.3828125,0.78515625,-0.640625,-0.65234375,0.47851562,-0.52734375,0.48046875,-1.09375,1.2578125,-1.8828125,-0.70703125,0.9375,-1.1640625,-0.6953125,0.5078125,-0.36132812,-1.515625,0.48242188,0.28710938,1.5390625,1.2109375,-0.46875,0.87109375,-0.453125,0.16308594,-1.7890625,0.73828125,-0.640625,-0.36914062,-1.6875,0.609375,0.9140625,-1.296875,-0.34960938,2.171875,-0.6328125,-0.7890625,-0.375,0.78515625,-0.47265625,0.25976562,1.2265625,-0.58203125,0.26757812,0.34960938,-0.625,0.453125,-0.41210938,0.55859375,0.6875,-0.0054626465,0.4453125,-1.6796875,-1.9140625,0.55859375,-0.16894531,-0.19921875,-0.34765625,1.140625,1.0703125,0.61328125,-1.5078125,-0.265625,-1.09375,1.578125,0.16503906,-0.734375,-0.41601562,-1.75,1.6875,-0.26953125,-0.62890625,1.5078125,-0.28710938,-0.075683594,1.4609375,-0.73046875,-1.1875,0.22167969,-1.0859375,-0.16503906,2.15625,-0.578125,-0.609375,0.19628906,1.40625,-0.39648438,0.65625,0.390625,-0.75,-0.3046875,1.390625,1.2421875,-1.171875,-0.14550781,-0.6640625,-0.6171875,0.6484375,-1.6953125,-2.109375,-0.25585938,-2.984375,-0.40234375,-0.578125,2.484375,0.80078125,-1.0546875,-0.4765625,0.36132812,-0.33007812,-0.092285156,-0.068359375,-0.65625,-1.15625,0.20800781,1.1171875,0.58984375,-0.38671875,-0.8203125,-1.5234375,0.296875,0.546875,-0.8671875,-2.8125,1.2578125,3.171875,0.20019531,0.34375,0.85546875
+5,0,0,0.58984375,0.99609375,-0.34570312,-1.3515625,-0.484375,-0.91015625,-0.88671875,0.4609375,0.94921875,-1.25,-0.1953125,-0.58203125,-1.09375,1.546875,-1.6640625,0.671875,-0.33007812,0.453125,-0.81640625,0.87109375,1.671875,0.70703125,0.06347656,0.34570312,0.41015625,-0.58984375,-1.8828125,-0.15136719,-1.0,1.2109375,-0.92578125,-0.22558594,0.83984375,0.20800781,-0.22753906,0.91015625,0.40039062,-0.060546875,-0.58984375,0.6171875,-1.8046875,0.38085938,-0.12109375,0.40625,-0.0045166016,1.34375,0.703125,0.13867188,-0.40039062,1.25,-0.41796875,-0.66796875,-1.515625,-0.021972656,-0.045654297,1.8203125,0.018310547,-0.23730469,0.40429688,-1.546875,0.29101562,0.39453125,1.2890625,0.2421875,0.9765625,1.1875,-1.5859375,1.4765625,0.013549805,-0.22753906,-0.80078125,-1.2109375,-0.765625,-0.07080078,-0.58203125,0.013427734,0.4921875,0.55859375,0.3125,0.64453125,1.078125,-1.046875,1.0,0.42382812,-0.53125,0.49804688,0.34375,0.62890625,-0.5859375,-0.50390625,3.578125,0.421875,0.79296875,1.2734375,-0.52734375,1.03125,-0.97265625,0.31640625,-0.578125,0.34179688,-0.82421875,0.17285156,1.375,-1.140625,0.04663086,0.17480469,-0.38671875,-1.859375,-0.4453125,0.828125,-0.15234375,-0.40429688,0.43554688,-0.8125,-0.6171875,-0.609375,-0.34570312,0.6953125,0.42773438,-0.125,0.73046875,-1.0703125,-0.62890625,-1.8359375,-0.82421875,1.6484375,-0.35742188,-1.6328125,-1.1015625,-0.96484375,1.3125,0.25585938,-2.375,-0.58984375,-1.5625,2.0,0.13769531,-0.45507812,-1.3515625,-2.5625,1.1015625,-2.25,0.5859375,-0.092285156,0.64453125,0.35351562,-0.20507812,-0.76171875,-0.087402344,1.5546875,-1.578125,0.59375,0.60546875,-0.10253906,1.2890625,-0.061279297,0.41601562,0.9375,0.796875,0.41992188,-0.87890625,-0.6640625,-0.37109375,0.09082031,-2.6875,2.40625,0.22851562,-0.012207031,1.421875,-0.91796875,-1.9140625,-0.8671875,-1.1171875,-0.92578125,-0.44921875,1.5078125,1.1875,-1.0078125,-0.18554688,-0.91015625,0.44921875,1.3203125,0.50390625,-0.041992188,1.0078125,-1.109375,-0.13476562,1.609375,0.58203125,-1.3125,-0.23535156,1.3671875,1.2265625,0.3828125,-0.55078125,0.44140625,3.265625,-0.100097656,-0.18652344,0.44726562
+5,0,1,-0.43945312,-0.69921875,-0.38476562,-1.4296875,0.18261719,-1.203125,-0.9453125,1.6484375,0.064453125,0.8125,-0.8671875,-0.53125,1.4140625,0.87890625,-2.140625,0.92578125,-0.515625,-0.86328125,0.10058594,0.5078125,2.453125,-0.18164062,-0.84765625,0.23925781,0.484375,-0.37890625,-2.0625,-1.1328125,-0.421875,1.7421875,-0.26367188,0.50390625,-0.98046875,0.546875,-1.2578125,0.49609375,-0.37304688,-0.49609375,-0.27734375,-0.6015625,-0.94921875,0.33203125,-0.6953125,-0.28515625,0.5390625,2.5,-0.10595703,-0.31640625,1.390625,1.0625,-1.0,-2.421875,-2.328125,0.4375,-0.609375,1.671875,-0.28125,-0.80859375,-0.26367188,-1.5546875,-0.87890625,-0.80859375,1.5859375,0.9609375,1.4296875,2.015625,-1.671875,1.6328125,-0.96875,0.62109375,1.296875,-1.296875,-0.69921875,-0.26367188,-0.23632812,0.24609375,0.26367188,-0.33789062,0.7265625,0.73046875,1.5703125,-0.43359375,0.27929688,1.75,-0.84375,0.64453125,-0.8203125,0.049316406,0.13769531,-0.11621094,1.4765625,-0.36132812,1.7734375,-0.103515625,-0.14453125,0.025878906,-1.84375,0.78125,-0.5390625,0.375,-0.28710938,-0.26953125,0.35742188,0.17578125,0.3671875,1.3125,-1.671875,-2.625,-1.0703125,-0.62109375,-0.19824219,0.7734375,-1.0546875,-0.53125,-0.100097656,-2.375,-0.09667969,0.81640625,0.33007812,1.5859375,0.11230469,-0.3828125,-0.15625,-1.65625,-1.265625,0.63671875,-1.359375,-0.77734375,-0.78515625,0.4765625,2.0625,-0.75390625,-1.71875,-0.5546875,-0.28515625,0.83984375,-0.13671875,0.079589844,-0.100097656,-2.046875,-0.328125,1.609375,0.58984375,0.95703125,0.52734375,0.15820312,-0.5234375,0.20507812,-0.59375,1.0390625,0.034423828,-0.90234375,0.54296875,-0.53515625,0.4375,-0.1796875,1.40625,1.0078125,0.55078125,0.359375,-0.19433594,0.875,0.41601562,0.49804688,-1.7890625,1.7265625,0.734375,0.20996094,0.45898438,0.037597656,-0.49609375,0.63671875,-0.039794922,0.18457031,-1.1171875,1.9921875,0.18261719,-1.6640625,-0.5703125,-0.09667969,-0.66015625,0.875,0.91796875,0.3828125,0.28515625,-0.28515625,0.27734375,0.3984375,0.32226562,-0.7890625,-1.4609375,0.7578125,1.5,0.15039062,-0.55078125,0.27734375,2.375,-0.39453125,1.015625,1.109375
+5,0,2,0.90234375,0.56640625,0.828125,0.19921875,-1.140625,-0.07861328,-1.1953125,-0.034179688,2.609375,-1.453125,1.8359375,-0.03930664,0.34960938,1.3671875,-0.625,-0.42578125,-0.46484375,0.18164062,2.265625,1.1484375,0.24023438,-0.51171875,-0.15625,0.578125,-0.48242188,-0.53515625,0.6484375,-0.52734375,0.90625,0.5078125,-0.40625,-0.008117676,0.78125,-0.6796875,0.38085938,0.91796875,0.953125,-0.19042969,-0.21191406,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24707031,-0.234375,0.50390625,-0.29101562,0.87890625,0.6796875,-0.16015625,-0.265625,-0.74609375,0.95703125,0.9375,0.41015625,-0.65234375,-0.49804688,-0.578125,0.7734375,1.1953125,-0.37890625,1.2578125,0.099609375,0.640625,0.51953125,-0.20703125,-1.2578125,-1.46875,0.41210938,-0.111328125,1.3828125,0.78515625,-0.640625,-0.65234375,0.47851562,-0.52734375,0.48046875,-1.09375,1.2578125,-1.8828125,-0.70703125,0.9375,-1.1640625,-0.6953125,0.5078125,-0.36132812,-1.515625,0.48242188,0.28710938,1.5390625,1.2109375,-0.46875,0.87109375,-0.453125,0.16308594,-1.7890625,0.73828125,-0.640625,-0.36914062,-1.6875,0.609375,0.9140625,-1.296875,-0.34960938,2.171875,-0.6328125,-0.7890625,-0.375,0.78515625,-0.47265625,0.25976562,1.2265625,-0.58203125,0.26757812,0.34960938,-0.625,0.453125,-0.41210938,0.55859375,0.6875,-0.0054626465,0.4453125,-1.6796875,-1.9140625,0.55859375,-0.16894531,-0.19921875,-0.34765625,1.140625,1.0703125,0.61328125,-1.5078125,-0.265625,-1.09375,1.578125,0.16503906,-0.734375,-0.41601562,-1.75,1.6875,-0.26953125,-0.62890625,1.5078125,-0.28710938,-0.075683594,1.4609375,-0.73046875,-1.1875,0.22167969,-1.0859375,-0.16503906,2.15625,-0.578125,-0.609375,0.19628906,1.40625,-0.39648438,0.65625,0.390625,-0.75,-0.3046875,1.390625,1.2421875,-1.171875,-0.14550781,-0.6640625,-0.6171875,0.6484375,-1.6953125,-2.109375,-0.25585938,-2.984375,-0.40234375,-0.578125,2.484375,0.80078125,-1.0546875,-0.4765625,0.36132812,-0.33007812,-0.092285156,-0.068359375,-0.65625,-1.15625,0.20800781,1.1171875,0.58984375,-0.38671875,-0.8203125,-1.5234375,0.296875,0.546875,-0.8671875,-2.8125,1.2578125,3.171875,0.20019531,0.34375,0.85546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
index a0c6dbf9..76476f56 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,5,0.13964844,0.10888672,0.27539062,-0.88671875,1.390625,0.61328125,0.48828125,1.0234375,-0.4921875,-1.046875,0.12597656,-0.26757812,0.57421875,0.35351562,-1.25,0.48632812,-0.546875,-1.9140625,1.1328125,-0.53125,-0.087402344,-0.6640625,1.28125,0.5234375,-1.234375,-1.5078125,-0.953125,-0.8671875,-1.7890625,0.44335938,-0.64453125,-0.26171875,0.010559082,-0.4609375,-0.5234375,0.953125,0.91796875,-0.73828125,-1.4609375,1.578125,-1.8046875,0.030517578,-0.07519531,0.09814453,0.38671875,0.5234375,0.63671875,-0.69921875,0.38085938,1.8515625,-1.1953125,-1.6015625,-0.98046875,1.1015625,-2.421875,0.091796875,-0.21289062,-0.80859375,1.328125,-1.6953125,1.734375,0.18359375,0.29296875,0.30273438,-0.5078125,-0.06640625,-1.0703125,0.16210938,0.546875,1.84375,0.05517578,-0.31835938,0.57421875,-0.13574219,-0.66796875,1.8515625,0.76953125,1.1796875,-1.7734375,-0.5234375,1.28125,0.89453125,-0.11816406,0.12207031,-0.63671875,-0.20117188,-0.46484375,1.4453125,-0.36132812,0.98828125,1.3828125,0.21386719,-0.921875,0.36523438,-1.4296875,0.22363281,-0.59375,-0.640625,0.625,-0.890625,-2.375,-0.18066406,2.515625,0.78125,-0.79296875,0.0051574707,0.063964844,-1.0703125,0.49609375,0.98828125,0.703125,0.036376953,0.56640625,-0.796875,0.49414062,-0.31640625,-0.47851562,-0.34765625,-0.484375,-1.1796875,0.71484375,-0.78515625,0.60546875,-1.5703125,-0.22460938,1.265625,-1.3203125,-2.40625,0.013122559,0.17382812,-0.18652344,0.044921875,-0.025512695,-1.3671875,-1.03125,-0.13671875,-0.65234375,0.25390625,0.20507812,-1.7265625,3.375,-0.40429688,1.3125,1.109375,-1.234375,1.0703125,-0.6015625,-0.671875,-2.515625,1.1015625,-0.30859375,-0.03564453,1.0625,-0.484375,-1.078125,-1.2109375,0.38085938,1.359375,1.015625,1.4375,-0.26367188,-0.011962891,1.8203125,1.2421875,-1.78125,1.6640625,-1.671875,-0.24511719,2.125,-0.20117188,0.36914062,-0.40234375,0.099609375,0.62109375,0.68359375,0.041992188,-0.86328125,-0.84765625,0.859375,-0.61328125,-0.3046875,-0.890625,0.13964844,0.86328125,0.1484375,0.54296875,0.59375,0.037841797,0.87109375,-1.8046875,0.09863281,1.375,1.2578125,0.28710938,0.43554688,-0.17871094,1.6875,0.15625,1.40625,0.51953125
-6,0,6,0.828125,0.83203125,-0.3984375,0.84375,1.40625,0.62890625,-0.3515625,0.39257812,-0.10253906,-1.4375,1.109375,-0.19335938,0.53125,-0.54296875,-0.99609375,-0.37109375,-0.81640625,-0.39648438,1.1171875,1.25,0.20410156,-1.8984375,0.36132812,0.35742188,-0.90234375,-1.609375,-1.3828125,-0.34960938,0.22753906,0.057861328,-0.23632812,-0.328125,1.71875,0.20703125,-0.33007812,-0.16992188,0.31445312,-0.6015625,-0.26367188,0.78125,-0.6015625,0.71875,-0.26757812,1.2578125,-0.27929688,0.25390625,-0.734375,-1.0234375,0.43554688,1.7109375,1.140625,-1.7734375,-0.49414062,-0.26757812,-1.0859375,-2.140625,0.3828125,1.3046875,-0.11279297,-1.015625,1.609375,1.59375,1.734375,-0.34570312,-1.0390625,-0.71875,0.05029297,-2.078125,-0.30078125,0.85546875,0.38476562,0.17382812,1.3515625,-1.4921875,0.45898438,0.51953125,0.18359375,-0.20214844,-1.71875,0.39648438,-0.2734375,0.71875,-0.016235352,0.32226562,-0.76171875,0.734375,-2.828125,0.45507812,1.578125,1.2109375,1.4296875,1.203125,1.125,0.22558594,-0.6328125,-0.9375,-0.87109375,1.25,-0.22753906,0.123535156,-0.54296875,-0.40039062,0.890625,-0.11279297,-1.0390625,0.78125,-0.98828125,-0.18164062,-0.34765625,-0.328125,0.18261719,1.96875,0.13378906,-0.63671875,-0.57421875,0.6796875,-0.4140625,0.33398438,-0.46484375,0.83984375,0.052490234,-1.09375,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.59765625,-0.24511719,0.765625,-0.04272461,-2.5,-0.93359375,0.73828125,-1.3125,2.546875,-0.31445312,0.013977051,1.8203125,-1.3203125,1.4296875,-0.75,0.12158203,1.171875,0.6640625,-0.20117188,-0.22558594,-0.28320312,-0.56640625,0.203125,0.4296875,-0.32226562,-1.34375,1.0234375,-1.140625,1.2265625,0.69921875,-0.36523438,-0.39257812,1.0546875,0.78515625,1.1484375,1.0078125,3.03125,-2.671875,1.6171875,-2.25,1.1796875,0.16796875,-0.29492188,-1.7734375,-1.84375,0.7265625,-0.6484375,0.65625,0.27734375,-0.66796875,-1.96875,1.40625,-0.35351562,0.09667969,0.15332031,0.83984375,-0.11230469,-1.015625,1.578125,-0.49609375,-0.26757812,0.5078125,-0.36914062,0.65625,1.546875,-0.31835938,-1.046875,-0.8125,-0.46289062,-0.46875,-0.19726562,0.19726562,0.51953125
-6,0,7,-0.9140625,0.041748047,-0.34179688,1.09375,1.3671875,-0.032226562,-0.43945312,1.0625,0.59765625,-0.66015625,0.27539062,1.84375,-0.2890625,-0.65625,-0.97265625,0.22949219,-0.00076293945,-1.046875,1.96875,-0.48046875,-0.123046875,-2.0625,-0.8125,0.53515625,-0.11816406,-0.62890625,0.71484375,0.73828125,0.68359375,0.40820312,0.44140625,0.2734375,1.328125,-0.515625,2.125,0.88671875,1.171875,-0.82421875,-1.015625,0.6640625,-0.40429688,-0.80859375,-1.8671875,-1.875,-0.828125,0.890625,-0.3984375,-0.94140625,0.86328125,1.75,1.140625,-0.5703125,-1.1484375,0.31054688,-0.12109375,-2.03125,0.35546875,1.515625,0.43359375,0.15722656,0.78125,0.83203125,-0.3515625,1.2890625,-1.3203125,0.5546875,-1.3359375,-1.9140625,0.47851562,-0.049316406,0.8046875,1.4453125,1.765625,-1.7734375,0.8203125,0.921875,-0.08251953,-0.69140625,-0.2890625,0.92578125,-1.03125,0.5703125,-0.28710938,-0.91015625,0.04296875,-1.359375,-0.4375,-0.19238281,1.5625,0.859375,1.4453125,-0.13574219,-0.20117188,1.0859375,-0.3203125,-0.27734375,-0.8203125,-0.9609375,-1.0234375,-0.21289062,0.083984375,0.40625,1.390625,1.1875,-1.0703125,1.3671875,0.61328125,0.19824219,-0.049560547,0.5234375,0.74609375,0.84375,2.484375,-1.578125,0.77734375,-0.35546875,-1.53125,0.66796875,-0.83984375,-0.7890625,1.328125,0.4609375,-0.671875,-0.64453125,-1.296875,2.015625,-0.06689453,0.7578125,0.8203125,-1.28125,1.484375,-2.328125,0.15722656,0.19335938,-0.110839844,0.8984375,0.72265625,-0.12988281,0.41210938,-1.0,1.5703125,0.26367188,-1.4453125,1.3828125,-0.36523438,-1.015625,0.49804688,-0.31640625,-0.1875,-1.25,0.7890625,0.22851562,2.671875,-0.921875,-0.8125,0.0390625,0.140625,-1.59375,1.1953125,0.12207031,0.63671875,0.59765625,1.1328125,0.47851562,0.0546875,0.73046875,0.020141602,0.98046875,-0.33398438,-0.005218506,-1.1953125,-1.40625,-2.359375,-1.2890625,-0.41796875,0.99609375,-0.49804688,-1.53125,-1.640625,-0.953125,0.7109375,0.16601562,0.25585938,-1.234375,-0.83984375,-1.21875,-0.25,0.19042969,-1.2578125,1.078125,-1.2109375,-0.030517578,-0.8671875,-1.1875,-1.609375,0.578125,-0.1875,0.21386719,0.203125,1.609375
-7,0,5,0.8671875,1.2109375,-0.828125,-0.984375,-1.28125,0.32617188,-0.5,1.5234375,-0.7890625,-0.71484375,1.625,0.010192871,-0.060791016,-0.28125,-1.15625,1.640625,-0.28515625,-1.5234375,0.15039062,-0.66796875,1.34375,0.84375,0.47851562,1.125,0.88671875,-0.85546875,-0.9296875,-1.9453125,-1.6171875,-0.62890625,-1.2578125,2.453125,1.015625,2.515625,-0.16992188,0.14550781,-0.40234375,-0.59765625,0.86328125,0.5546875,-0.62890625,0.28125,0.640625,0.20019531,1.3046875,0.36328125,0.58203125,0.27539062,0.6484375,-1.1796875,-0.83984375,-0.5078125,-1.890625,-1.609375,-1.03125,0.59375,0.08642578,-0.33789062,0.31445312,-0.75,-0.19042969,-0.515625,1.8515625,1.0,1.1796875,0.7109375,-0.890625,-0.13964844,0.22460938,0.24707031,-0.03857422,-0.36132812,-0.6796875,-0.18847656,-0.47265625,-0.15332031,-0.23339844,-0.13085938,-0.014709473,0.45898438,-0.48046875,-0.5390625,1.1328125,0.4296875,1.109375,1.2890625,-0.97265625,0.53125,1.359375,-1.015625,1.3359375,0.40625,0.546875,1.2109375,0.21875,-1.3125,0.765625,1.0,-1.71875,-0.29882812,-0.64453125,0.328125,2.234375,-1.7734375,-1.2265625,-1.484375,-2.0,-1.7265625,-0.48242188,0.93359375,0.21289062,-1.0078125,0.09082031,0.57421875,0.52734375,0.203125,0.008850098,-0.984375,-0.265625,0.22753906,-0.36914062,-0.31054688,-1.4375,-1.0,-1.46875,1.0625,-1.7578125,-0.55078125,1.0390625,0.98828125,0.90625,-0.484375,-1.140625,-0.64453125,-1.7578125,0.83203125,1.109375,-0.18359375,-1.6640625,-2.421875,1.4765625,-0.55078125,-0.640625,0.59765625,0.4765625,0.18261719,-0.24902344,-0.52734375,0.71875,1.5390625,-0.75390625,0.27539062,1.2734375,-1.2109375,0.67578125,0.3125,0.59375,-0.86328125,0.77734375,-1.1328125,0.796875,-0.40234375,0.64453125,-0.6484375,-1.140625,3.46875,-0.7578125,-1.1953125,0.33007812,-1.0234375,-1.0,-0.47460938,-1.28125,-1.1171875,-0.38085938,1.171875,0.78515625,-0.72265625,1.0390625,-0.49804688,0.39453125,0.3515625,0.73046875,-1.2890625,1.1484375,-0.053710938,0.08935547,0.28515625,0.25,0.3828125,2.03125,-0.14257812,-0.033203125,-0.7265625,-0.011657715,0.84375,2.0625,1.796875,0.84765625,0.66796875
-7,0,6,0.58984375,1.3984375,-0.609375,-0.51171875,-1.9140625,-0.7421875,1.0234375,0.9765625,0.18652344,0.0099487305,0.11669922,0.44335938,-0.01940918,-0.82421875,-0.51953125,1.796875,-1.2578125,-0.82421875,0.34960938,0.27929688,0.45703125,-0.6015625,1.046875,0.3671875,1.2734375,-0.50390625,-1.984375,-0.71875,0.5,-0.6328125,-0.73828125,-0.008544922,2.21875,-0.43359375,-0.62109375,-0.07519531,0.25976562,-1.28125,-0.515625,-0.6875,-0.42773438,0.18847656,0.29882812,-0.09863281,-0.73046875,0.15234375,0.37890625,0.6640625,-0.27734375,-0.20019531,0.89453125,0.546875,-3.765625,-0.15625,1.15625,1.71875,-0.15332031,0.34375,0.47070312,-0.100097656,-1.0546875,-1.1171875,1.0390625,0.328125,1.6015625,-0.375,-0.23828125,-0.34765625,0.04638672,0.35546875,-0.32617188,1.125,-0.045654297,-0.9921875,2.4375,0.83203125,0.88671875,-0.12792969,-1.4453125,0.34179688,0.59765625,-0.27929688,1.5234375,0.123046875,0.37304688,0.003616333,-1.203125,1.125,1.59375,-0.24609375,0.87890625,-0.34570312,1.0,0.52734375,-1.2734375,0.57421875,-0.84375,1.8203125,-1.3984375,0.35742188,0.33789062,0.9765625,-0.5546875,0.71484375,-0.625,-0.21484375,-1.6875,-1.75,-0.29296875,0.33398438,-1.4609375,-2.3125,0.31445312,-0.20410156,-1.4140625,1.15625,-0.66796875,1.0390625,-1.140625,-0.54296875,-0.13085938,0.41210938,-1.203125,-1.328125,0.484375,0.765625,-1.6328125,-0.09716797,-0.18652344,-1.1328125,1.1484375,-0.23730469,-0.110839844,-2.1875,-0.765625,-0.6640625,1.65625,0.048095703,-1.421875,-2.78125,1.453125,0.15136719,-1.3984375,-1.203125,-0.328125,-0.44921875,0.40625,0.77734375,2.21875,1.9765625,-0.5390625,-1.0703125,0.17578125,-0.3203125,-0.23925781,-1.2734375,1.234375,0.08105469,0.99609375,0.09765625,-0.27148438,-0.072753906,0.22949219,-0.40234375,-1.15625,0.62890625,-0.515625,-1.375,0.58203125,0.46875,0.84375,0.61328125,-0.796875,0.80859375,-1.6015625,1.1328125,2.75,-0.35546875,1.6796875,1.90625,-0.25390625,0.65234375,0.625,-1.0234375,0.047607422,-1.03125,0.49023438,-0.087890625,0.84375,-0.04272461,0.18847656,0.96875,-0.0099487305,0.3125,-1.09375,-0.20214844,0.014343262,1.9921875,1.0703125,0.028198242
-7,0,7,0.9609375,0.671875,-1.0234375,-0.95703125,0.057128906,-0.6953125,-0.15332031,1.9921875,0.82421875,0.625,-1.3359375,1.84375,-0.36328125,1.15625,0.025634766,0.765625,0.62890625,-0.35546875,0.87890625,1.4140625,0.22949219,-0.049072266,-0.74609375,-0.33203125,-0.6171875,1.3828125,-0.64453125,0.028442383,0.8984375,-2.203125,0.5703125,1.796875,1.1796875,0.16113281,-0.59375,-0.4140625,0.22558594,1.4609375,0.080078125,-0.107421875,-0.9765625,-0.07910156,-0.53125,0.49023438,0.26171875,0.05053711,-0.24121094,-0.33203125,0.071777344,0.18066406,0.09472656,-1.390625,-0.35742188,-0.73828125,0.07324219,0.49804688,-1.6484375,0.15332031,0.28515625,-0.64453125,-1.1484375,-0.6875,1.6171875,-0.33789062,0.45117188,0.55859375,0.25195312,-0.18261719,-0.29492188,1.046875,0.22167969,2.078125,-0.859375,-0.7421875,0.86328125,1.5859375,0.16601562,-0.81640625,-2.84375,0.70703125,1.6015625,0.578125,1.859375,2.0625,-0.66796875,0.80078125,-0.6796875,-0.65625,-0.55859375,1.375,1.140625,1.0,-0.16992188,1.2890625,0.020141602,0.111816406,-1.640625,1.03125,-0.12890625,0.45507812,0.66796875,-0.50390625,-0.484375,-0.34375,-0.59765625,0.9375,-0.3359375,-0.9140625,-1.71875,-0.875,-0.17871094,-0.47265625,1.59375,-0.83203125,-0.15039062,-1.078125,0.21484375,0.19238281,0.8984375,0.8046875,1.234375,0.06225586,-1.0703125,-0.78515625,-0.46679688,0.73828125,-0.17675781,-0.23925781,1.2265625,-0.53125,0.8984375,-1.859375,0.29492188,-1.1640625,-1.359375,0.765625,-1.578125,0.16601562,0.22460938,-2.046875,0.071777344,-0.13378906,-1.375,1.234375,-0.23828125,0.06689453,-0.53515625,1.7109375,0.21679688,-1.875,1.5859375,-2.109375,0.34375,0.20410156,0.15039062,0.46679688,1.421875,0.68359375,-0.54296875,0.23535156,-1.578125,2.234375,-0.8359375,0.796875,1.1484375,-0.78515625,-0.484375,-2.875,0.23535156,1.25,-0.55859375,0.73828125,-1.1171875,1.3203125,0.091308594,1.703125,0.23144531,-1.4921875,-0.765625,0.82421875,-0.61328125,0.51953125,-0.41601562,0.47070312,-1.7109375,0.29882812,-1.6796875,-2.046875,1.484375,-0.9765625,-1.5703125,-0.92578125,-1.5234375,0.48242188,0.06982422,0.064941406,0.24121094,-0.37890625,-0.859375,1.265625
+6,0,1,0.13964844,0.10888672,0.27539062,-0.88671875,1.390625,0.61328125,0.48828125,1.0234375,-0.4921875,-1.046875,0.12597656,-0.26757812,0.57421875,0.35351562,-1.25,0.48632812,-0.546875,-1.9140625,1.1328125,-0.53125,-0.087402344,-0.6640625,1.28125,0.5234375,-1.234375,-1.5078125,-0.953125,-0.8671875,-1.7890625,0.44335938,-0.64453125,-0.26171875,0.010559082,-0.4609375,-0.5234375,0.953125,0.91796875,-0.73828125,-1.4609375,1.578125,-1.8046875,0.030517578,-0.07519531,0.09814453,0.38671875,0.5234375,0.63671875,-0.69921875,0.38085938,1.8515625,-1.1953125,-1.6015625,-0.98046875,1.1015625,-2.421875,0.091796875,-0.21289062,-0.80859375,1.328125,-1.6953125,1.734375,0.18359375,0.29296875,0.30273438,-0.5078125,-0.06640625,-1.0703125,0.16210938,0.546875,1.84375,0.05517578,-0.31835938,0.57421875,-0.13574219,-0.66796875,1.8515625,0.76953125,1.1796875,-1.7734375,-0.5234375,1.28125,0.89453125,-0.11816406,0.12207031,-0.63671875,-0.20117188,-0.46484375,1.4453125,-0.36132812,0.98828125,1.3828125,0.21386719,-0.921875,0.36523438,-1.4296875,0.22363281,-0.59375,-0.640625,0.625,-0.890625,-2.375,-0.18066406,2.515625,0.78125,-0.79296875,0.0051574707,0.063964844,-1.0703125,0.49609375,0.98828125,0.703125,0.036376953,0.56640625,-0.796875,0.49414062,-0.31640625,-0.47851562,-0.34765625,-0.484375,-1.1796875,0.71484375,-0.78515625,0.60546875,-1.5703125,-0.22460938,1.265625,-1.3203125,-2.40625,0.013122559,0.17382812,-0.18652344,0.044921875,-0.025512695,-1.3671875,-1.03125,-0.13671875,-0.65234375,0.25390625,0.20507812,-1.7265625,3.375,-0.40429688,1.3125,1.109375,-1.234375,1.0703125,-0.6015625,-0.671875,-2.515625,1.1015625,-0.30859375,-0.03564453,1.0625,-0.484375,-1.078125,-1.2109375,0.38085938,1.359375,1.015625,1.4375,-0.26367188,-0.011962891,1.8203125,1.2421875,-1.78125,1.6640625,-1.671875,-0.24511719,2.125,-0.20117188,0.36914062,-0.40234375,0.099609375,0.62109375,0.68359375,0.041992188,-0.86328125,-0.84765625,0.859375,-0.61328125,-0.3046875,-0.890625,0.13964844,0.86328125,0.1484375,0.54296875,0.59375,0.037841797,0.87109375,-1.8046875,0.09863281,1.375,1.2578125,0.28710938,0.43554688,-0.17871094,1.6875,0.15625,1.40625,0.51953125
+6,0,2,0.828125,0.83203125,-0.3984375,0.84375,1.40625,0.62890625,-0.3515625,0.39257812,-0.10253906,-1.4375,1.109375,-0.19335938,0.53125,-0.54296875,-0.99609375,-0.37109375,-0.81640625,-0.39648438,1.1171875,1.25,0.20410156,-1.8984375,0.36132812,0.35742188,-0.90234375,-1.609375,-1.3828125,-0.34960938,0.22753906,0.057861328,-0.23632812,-0.328125,1.71875,0.20703125,-0.33007812,-0.16992188,0.31445312,-0.6015625,-0.26367188,0.78125,-0.6015625,0.71875,-0.26757812,1.2578125,-0.27929688,0.25390625,-0.734375,-1.0234375,0.43554688,1.7109375,1.140625,-1.7734375,-0.49414062,-0.26757812,-1.0859375,-2.140625,0.3828125,1.3046875,-0.11279297,-1.015625,1.609375,1.59375,1.734375,-0.34570312,-1.0390625,-0.71875,0.05029297,-2.078125,-0.30078125,0.85546875,0.38476562,0.17382812,1.3515625,-1.4921875,0.45898438,0.51953125,0.18359375,-0.20214844,-1.71875,0.39648438,-0.2734375,0.71875,-0.016235352,0.32226562,-0.76171875,0.734375,-2.828125,0.45507812,1.578125,1.2109375,1.4296875,1.203125,1.125,0.22558594,-0.6328125,-0.9375,-0.87109375,1.25,-0.22753906,0.123535156,-0.54296875,-0.40039062,0.890625,-0.11279297,-1.0390625,0.78125,-0.98828125,-0.18164062,-0.34765625,-0.328125,0.18261719,1.96875,0.13378906,-0.63671875,-0.57421875,0.6796875,-0.4140625,0.33398438,-0.46484375,0.83984375,0.052490234,-1.09375,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.59765625,-0.24511719,0.765625,-0.04272461,-2.5,-0.93359375,0.73828125,-1.3125,2.546875,-0.31445312,0.013977051,1.8203125,-1.3203125,1.4296875,-0.75,0.12158203,1.171875,0.6640625,-0.20117188,-0.22558594,-0.28320312,-0.56640625,0.203125,0.4296875,-0.32226562,-1.34375,1.0234375,-1.140625,1.2265625,0.69921875,-0.36523438,-0.39257812,1.0546875,0.78515625,1.1484375,1.0078125,3.03125,-2.671875,1.6171875,-2.25,1.1796875,0.16796875,-0.29492188,-1.7734375,-1.84375,0.7265625,-0.6484375,0.65625,0.27734375,-0.66796875,-1.96875,1.40625,-0.35351562,0.09667969,0.15332031,0.83984375,-0.11230469,-1.015625,1.578125,-0.49609375,-0.26757812,0.5078125,-0.36914062,0.65625,1.546875,-0.31835938,-1.046875,-0.8125,-0.46289062,-0.46875,-0.19726562,0.19726562,0.51953125
+6,0,3,-0.9140625,0.041748047,-0.34179688,1.09375,1.3671875,-0.032226562,-0.43945312,1.0625,0.59765625,-0.66015625,0.27539062,1.84375,-0.2890625,-0.65625,-0.97265625,0.22949219,-0.00076293945,-1.046875,1.96875,-0.48046875,-0.123046875,-2.0625,-0.8125,0.53515625,-0.11816406,-0.62890625,0.71484375,0.73828125,0.68359375,0.40820312,0.44140625,0.2734375,1.328125,-0.515625,2.125,0.88671875,1.171875,-0.82421875,-1.015625,0.6640625,-0.40429688,-0.80859375,-1.8671875,-1.875,-0.828125,0.890625,-0.3984375,-0.94140625,0.86328125,1.75,1.140625,-0.5703125,-1.1484375,0.31054688,-0.12109375,-2.03125,0.35546875,1.515625,0.43359375,0.15722656,0.78125,0.83203125,-0.3515625,1.2890625,-1.3203125,0.5546875,-1.3359375,-1.9140625,0.47851562,-0.049316406,0.8046875,1.4453125,1.765625,-1.7734375,0.8203125,0.921875,-0.08251953,-0.69140625,-0.2890625,0.92578125,-1.03125,0.5703125,-0.28710938,-0.91015625,0.04296875,-1.359375,-0.4375,-0.19238281,1.5625,0.859375,1.4453125,-0.13574219,-0.20117188,1.0859375,-0.3203125,-0.27734375,-0.8203125,-0.9609375,-1.0234375,-0.21289062,0.083984375,0.40625,1.390625,1.1875,-1.0703125,1.3671875,0.61328125,0.19824219,-0.049560547,0.5234375,0.74609375,0.84375,2.484375,-1.578125,0.77734375,-0.35546875,-1.53125,0.66796875,-0.83984375,-0.7890625,1.328125,0.4609375,-0.671875,-0.64453125,-1.296875,2.015625,-0.06689453,0.7578125,0.8203125,-1.28125,1.484375,-2.328125,0.15722656,0.19335938,-0.110839844,0.8984375,0.72265625,-0.12988281,0.41210938,-1.0,1.5703125,0.26367188,-1.4453125,1.3828125,-0.36523438,-1.015625,0.49804688,-0.31640625,-0.1875,-1.25,0.7890625,0.22851562,2.671875,-0.921875,-0.8125,0.0390625,0.140625,-1.59375,1.1953125,0.12207031,0.63671875,0.59765625,1.1328125,0.47851562,0.0546875,0.73046875,0.020141602,0.98046875,-0.33398438,-0.005218506,-1.1953125,-1.40625,-2.359375,-1.2890625,-0.41796875,0.99609375,-0.49804688,-1.53125,-1.640625,-0.953125,0.7109375,0.16601562,0.25585938,-1.234375,-0.83984375,-1.21875,-0.25,0.19042969,-1.2578125,1.078125,-1.2109375,-0.030517578,-0.8671875,-1.1875,-1.609375,0.578125,-0.1875,0.21386719,0.203125,1.609375
+7,0,0,0.8671875,1.2109375,-0.828125,-0.984375,-1.28125,0.32617188,-0.5,1.5234375,-0.7890625,-0.71484375,1.625,0.010192871,-0.060791016,-0.28125,-1.15625,1.640625,-0.28515625,-1.5234375,0.15039062,-0.66796875,1.34375,0.84375,0.47851562,1.125,0.88671875,-0.85546875,-0.9296875,-1.9453125,-1.6171875,-0.62890625,-1.2578125,2.453125,1.015625,2.515625,-0.16992188,0.14550781,-0.40234375,-0.59765625,0.86328125,0.5546875,-0.62890625,0.28125,0.640625,0.20019531,1.3046875,0.36328125,0.58203125,0.27539062,0.6484375,-1.1796875,-0.83984375,-0.5078125,-1.890625,-1.609375,-1.03125,0.59375,0.08642578,-0.33789062,0.31445312,-0.75,-0.19042969,-0.515625,1.8515625,1.0,1.1796875,0.7109375,-0.890625,-0.13964844,0.22460938,0.24707031,-0.03857422,-0.36132812,-0.6796875,-0.18847656,-0.47265625,-0.15332031,-0.23339844,-0.13085938,-0.014709473,0.45898438,-0.48046875,-0.5390625,1.1328125,0.4296875,1.109375,1.2890625,-0.97265625,0.53125,1.359375,-1.015625,1.3359375,0.40625,0.546875,1.2109375,0.21875,-1.3125,0.765625,1.0,-1.71875,-0.29882812,-0.64453125,0.328125,2.234375,-1.7734375,-1.2265625,-1.484375,-2.0,-1.7265625,-0.48242188,0.93359375,0.21289062,-1.0078125,0.09082031,0.57421875,0.52734375,0.203125,0.008850098,-0.984375,-0.265625,0.22753906,-0.36914062,-0.31054688,-1.4375,-1.0,-1.46875,1.0625,-1.7578125,-0.55078125,1.0390625,0.98828125,0.90625,-0.484375,-1.140625,-0.64453125,-1.7578125,0.83203125,1.109375,-0.18359375,-1.6640625,-2.421875,1.4765625,-0.55078125,-0.640625,0.59765625,0.4765625,0.18261719,-0.24902344,-0.52734375,0.71875,1.5390625,-0.75390625,0.27539062,1.2734375,-1.2109375,0.67578125,0.3125,0.59375,-0.86328125,0.77734375,-1.1328125,0.796875,-0.40234375,0.64453125,-0.6484375,-1.140625,3.46875,-0.7578125,-1.1953125,0.33007812,-1.0234375,-1.0,-0.47460938,-1.28125,-1.1171875,-0.38085938,1.171875,0.78515625,-0.72265625,1.0390625,-0.49804688,0.39453125,0.3515625,0.73046875,-1.2890625,1.1484375,-0.053710938,0.08935547,0.28515625,0.25,0.3828125,2.03125,-0.14257812,-0.033203125,-0.7265625,-0.011657715,0.84375,2.0625,1.796875,0.84765625,0.66796875
+7,0,1,0.58984375,1.3984375,-0.609375,-0.51171875,-1.9140625,-0.7421875,1.0234375,0.9765625,0.18652344,0.0099487305,0.11669922,0.44335938,-0.01940918,-0.82421875,-0.51953125,1.796875,-1.2578125,-0.82421875,0.34960938,0.27929688,0.45703125,-0.6015625,1.046875,0.3671875,1.2734375,-0.50390625,-1.984375,-0.71875,0.5,-0.6328125,-0.73828125,-0.008544922,2.21875,-0.43359375,-0.62109375,-0.07519531,0.25976562,-1.28125,-0.515625,-0.6875,-0.42773438,0.18847656,0.29882812,-0.09863281,-0.73046875,0.15234375,0.37890625,0.6640625,-0.27734375,-0.20019531,0.89453125,0.546875,-3.765625,-0.15625,1.15625,1.71875,-0.15332031,0.34375,0.47070312,-0.100097656,-1.0546875,-1.1171875,1.0390625,0.328125,1.6015625,-0.375,-0.23828125,-0.34765625,0.04638672,0.35546875,-0.32617188,1.125,-0.045654297,-0.9921875,2.4375,0.83203125,0.88671875,-0.12792969,-1.4453125,0.34179688,0.59765625,-0.27929688,1.5234375,0.123046875,0.37304688,0.003616333,-1.203125,1.125,1.59375,-0.24609375,0.87890625,-0.34570312,1.0,0.52734375,-1.2734375,0.57421875,-0.84375,1.8203125,-1.3984375,0.35742188,0.33789062,0.9765625,-0.5546875,0.71484375,-0.625,-0.21484375,-1.6875,-1.75,-0.29296875,0.33398438,-1.4609375,-2.3125,0.31445312,-0.20410156,-1.4140625,1.15625,-0.66796875,1.0390625,-1.140625,-0.54296875,-0.13085938,0.41210938,-1.203125,-1.328125,0.484375,0.765625,-1.6328125,-0.09716797,-0.18652344,-1.1328125,1.1484375,-0.23730469,-0.110839844,-2.1875,-0.765625,-0.6640625,1.65625,0.048095703,-1.421875,-2.78125,1.453125,0.15136719,-1.3984375,-1.203125,-0.328125,-0.44921875,0.40625,0.77734375,2.21875,1.9765625,-0.5390625,-1.0703125,0.17578125,-0.3203125,-0.23925781,-1.2734375,1.234375,0.08105469,0.99609375,0.09765625,-0.27148438,-0.072753906,0.22949219,-0.40234375,-1.15625,0.62890625,-0.515625,-1.375,0.58203125,0.46875,0.84375,0.61328125,-0.796875,0.80859375,-1.6015625,1.1328125,2.75,-0.35546875,1.6796875,1.90625,-0.25390625,0.65234375,0.625,-1.0234375,0.047607422,-1.03125,0.49023438,-0.087890625,0.84375,-0.04272461,0.18847656,0.96875,-0.0099487305,0.3125,-1.09375,-0.20214844,0.014343262,1.9921875,1.0703125,0.028198242
+7,0,2,0.9609375,0.671875,-1.0234375,-0.95703125,0.057128906,-0.6953125,-0.15332031,1.9921875,0.82421875,0.625,-1.3359375,1.84375,-0.36328125,1.15625,0.025634766,0.765625,0.62890625,-0.35546875,0.87890625,1.4140625,0.22949219,-0.049072266,-0.74609375,-0.33203125,-0.6171875,1.3828125,-0.64453125,0.028442383,0.8984375,-2.203125,0.5703125,1.796875,1.1796875,0.16113281,-0.59375,-0.4140625,0.22558594,1.4609375,0.080078125,-0.107421875,-0.9765625,-0.07910156,-0.53125,0.49023438,0.26171875,0.05053711,-0.24121094,-0.33203125,0.071777344,0.18066406,0.09472656,-1.390625,-0.35742188,-0.73828125,0.07324219,0.49804688,-1.6484375,0.15332031,0.28515625,-0.64453125,-1.1484375,-0.6875,1.6171875,-0.33789062,0.45117188,0.55859375,0.25195312,-0.18261719,-0.29492188,1.046875,0.22167969,2.078125,-0.859375,-0.7421875,0.86328125,1.5859375,0.16601562,-0.81640625,-2.84375,0.70703125,1.6015625,0.578125,1.859375,2.0625,-0.66796875,0.80078125,-0.6796875,-0.65625,-0.55859375,1.375,1.140625,1.0,-0.16992188,1.2890625,0.020141602,0.111816406,-1.640625,1.03125,-0.12890625,0.45507812,0.66796875,-0.50390625,-0.484375,-0.34375,-0.59765625,0.9375,-0.3359375,-0.9140625,-1.71875,-0.875,-0.17871094,-0.47265625,1.59375,-0.83203125,-0.15039062,-1.078125,0.21484375,0.19238281,0.8984375,0.8046875,1.234375,0.06225586,-1.0703125,-0.78515625,-0.46679688,0.73828125,-0.17675781,-0.23925781,1.2265625,-0.53125,0.8984375,-1.859375,0.29492188,-1.1640625,-1.359375,0.765625,-1.578125,0.16601562,0.22460938,-2.046875,0.071777344,-0.13378906,-1.375,1.234375,-0.23828125,0.06689453,-0.53515625,1.7109375,0.21679688,-1.875,1.5859375,-2.109375,0.34375,0.20410156,0.15039062,0.46679688,1.421875,0.68359375,-0.54296875,0.23535156,-1.578125,2.234375,-0.8359375,0.796875,1.1484375,-0.78515625,-0.484375,-2.875,0.23535156,1.25,-0.55859375,0.73828125,-1.1171875,1.3203125,0.091308594,1.703125,0.23144531,-1.4921875,-0.765625,0.82421875,-0.61328125,0.51953125,-0.41601562,0.47070312,-1.7109375,0.29882812,-1.6796875,-2.046875,1.484375,-0.9765625,-1.5703125,-0.92578125,-1.5234375,0.48242188,0.06982422,0.064941406,0.24121094,-0.37890625,-0.859375,1.265625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
index 0aac125f..439e2692 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,5,1.078125,0.25195312,-0.16601562,-0.296875,0.109375,-0.42578125,-1.28125,0.7890625,0.23632812,0.46484375,0.13574219,0.921875,1.2421875,-0.13476562,-0.55859375,-0.13183594,-1.078125,-2.25,0.9375,0.21386719,1.7890625,-1.0234375,-1.2890625,-0.2109375,1.40625,-0.30078125,-0.71484375,-0.013305664,0.2578125,-0.14648438,-0.84765625,-0.61328125,1.9453125,-0.94921875,-0.93359375,-0.41601562,0.09472656,-1.140625,-2.578125,-0.453125,0.44726562,-0.87890625,-0.71484375,-0.29882812,-1.21875,0.5234375,-0.3203125,0.5,1.765625,1.4296875,-0.27148438,0.36328125,-1.6640625,1.2578125,-0.10595703,0.38671875,-0.5,1.8828125,0.71484375,-0.82421875,-0.69140625,-1.9453125,0.76953125,0.80078125,1.328125,1.46875,-1.4609375,-1.875,0.41601562,0.9375,1.421875,1.2265625,-0.41992188,-0.94921875,2.90625,1.78125,-0.3515625,-0.11669922,-0.5,0.26953125,-0.34765625,0.26757812,0.0007362366,2.34375,-0.48046875,-0.58984375,-0.41601562,2.5,1.0546875,0.91015625,0.8125,0.5390625,0.8203125,0.06933594,-0.84765625,-0.13769531,0.5546875,1.0390625,0.5078125,-0.9453125,0.19824219,1.0390625,-0.46484375,0.37109375,-0.16796875,0.76953125,-1.171875,-1.15625,0.375,0.33398438,-1.015625,-0.15820312,-0.55859375,-1.9453125,-1.78125,0.10253906,-0.625,1.2265625,-1.2578125,0.44140625,0.61328125,0.41210938,-2.203125,-1.359375,-0.8046875,0.18847656,-0.5625,0.43945312,0.6640625,-0.62890625,1.7578125,-0.25390625,0.14550781,-1.5078125,-0.65234375,0.5703125,0.55859375,0.19726562,-0.75,-2.5,1.0703125,2.59375,-0.7109375,-1.21875,-1.265625,1.234375,0.31640625,-0.15332031,0.31640625,0.6015625,0.56640625,0.98046875,0.6328125,-0.6171875,-0.77734375,-1.4609375,0.29296875,-0.049316406,1.171875,-0.64453125,0.35546875,0.57421875,0.080078125,-1.0703125,0.5859375,-0.28515625,0.57421875,-0.74609375,-1.2890625,1.2890625,0.14550781,0.65234375,-0.8984375,-0.52734375,-1.6953125,1.84375,-0.17871094,-0.58203125,0.07861328,-0.48632812,1.25,0.40625,0.828125,-0.68359375,-1.0703125,-1.5546875,0.18652344,0.36328125,0.6953125,0.625,-0.30078125,0.36328125,0.796875,-0.93359375,-0.6328125,-1.8828125,-0.6953125,1.25,1.03125,0.72265625
-8,0,6,0.87109375,0.984375,-0.17871094,0.12597656,0.08203125,-0.60546875,0.765625,1.0703125,1.078125,-0.80859375,1.3203125,1.5546875,0.053710938,-1.03125,-0.14355469,0.30078125,-0.67578125,-1.0078125,0.8359375,0.3359375,0.026489258,-0.36328125,-0.24609375,1.640625,-0.18164062,0.14550781,-2.78125,0.21191406,0.3203125,0.7890625,-0.5078125,-0.90625,1.3671875,-0.78125,0.57421875,0.59375,0.53515625,-0.93359375,-0.21972656,1.0859375,-1.2109375,1.109375,-0.34179688,-2.421875,-0.8671875,-0.7890625,0.25976562,-0.22949219,1.3046875,1.6015625,0.5546875,-0.47070312,-0.92578125,0.6796875,0.46875,0.26757812,0.19921875,-0.5078125,-1.0703125,-0.036376953,1.4765625,0.20703125,0.6328125,0.859375,0.50390625,-0.43554688,-0.14160156,0.55859375,-1.3671875,0.66796875,-0.09863281,0.5390625,0.84765625,-0.42578125,1.8359375,0.8046875,1.3984375,1.046875,-0.35546875,0.36914062,-0.0859375,-1.125,0.83984375,-1.4375,-0.45507812,0.3671875,-0.03173828,-0.036132812,1.75,0.1875,1.34375,-0.5390625,-1.03125,1.2421875,-1.09375,-0.59765625,-2.75,0.49023438,-0.6015625,-0.3671875,-1.453125,1.171875,0.38085938,0.19140625,-0.58984375,1.84375,-0.75,0.16601562,-1.1640625,0.30859375,-0.56640625,-0.42578125,0.91796875,-0.26367188,0.36328125,1.765625,-0.5078125,0.5546875,-0.15917969,-0.44335938,0.57421875,-0.62890625,-0.4921875,-0.94140625,-1.328125,2.109375,-1.2578125,-1.0546875,0.71875,-1.265625,1.3203125,-1.2109375,-0.6875,-0.76171875,-1.59375,0.65234375,1.09375,-0.52734375,0.057373047,-2.46875,1.703125,-0.9453125,-1.7578125,1.3671875,0.71484375,-0.04272461,0.875,-0.89453125,-1.171875,0.9140625,-1.296875,-0.70703125,0.68359375,-1.0234375,1.109375,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.07128906,1.65625,2.453125,1.4609375,-1.53125,0.100097656,-0.5390625,0.084472656,0.859375,-0.16113281,-1.015625,-0.296875,-1.6328125,-0.14746094,-1.5390625,1.0390625,0.9140625,-1.1640625,0.29492188,0.046875,-1.2734375,-1.4140625,0.54296875,-1.0,-0.45703125,0.24902344,0.022705078,0.56640625,-1.6171875,-1.625,-0.234375,2.46875,-0.3671875,-0.9609375,-1.546875,1.03125,1.4921875,1.09375,-0.421875,-0.59375
-8,0,7,1.25,1.5546875,0.10205078,-0.26171875,-1.609375,-0.016479492,-1.328125,1.4375,1.40625,-2.640625,1.7578125,-0.69140625,0.03100586,1.1171875,-0.49609375,0.46679688,0.29101562,0.41210938,1.859375,-0.016845703,-0.8046875,-0.034179688,-0.39453125,-0.25195312,-0.8671875,-1.0546875,-0.71484375,-0.22363281,0.09765625,-0.20605469,-0.25195312,0.01928711,1.0546875,0.71875,0.099609375,0.28710938,1.421875,0.14550781,-0.8828125,1.6875,-2.3125,-0.22265625,0.114746094,2.140625,0.26367188,0.921875,-0.87109375,0.71484375,-0.0625,0.87890625,-0.6484375,-0.09033203,-0.34960938,-0.25390625,-0.4140625,-0.45703125,0.19628906,0.8125,-1.1328125,0.16796875,0.546875,0.65234375,1.65625,-0.41601562,1.359375,0.43359375,-0.36328125,-1.2578125,-0.98046875,0.076660156,-0.2734375,-0.234375,0.6171875,0.04272461,-0.9140625,0.734375,0.14160156,0.34375,-1.53125,1.3203125,-0.81640625,-0.66796875,0.95703125,-1.2421875,-0.17773438,0.3359375,0.24609375,-1.3046875,0.10058594,0.51171875,1.953125,1.5859375,-0.35546875,1.0546875,-0.89453125,-0.14648438,-0.6796875,0.0028839111,0.19433594,-0.703125,-2.390625,0.87890625,1.71875,-1.7421875,0.03149414,1.5390625,-1.109375,-0.46484375,0.053466797,0.20800781,-0.08691406,-0.3125,1.328125,-0.7265625,0.06738281,0.080078125,-0.37890625,0.24707031,-0.25390625,0.859375,1.640625,-0.56640625,-0.15332031,-1.796875,-0.51171875,0.20214844,0.39257812,-1.5546875,-0.11767578,1.3828125,-0.890625,1.1171875,-1.3984375,0.6484375,-2.09375,1.515625,0.9140625,-0.59765625,-0.7109375,-2.328125,1.5859375,-1.1640625,-0.765625,0.03930664,-0.18945312,0.203125,0.48046875,-0.5859375,-0.76171875,1.2421875,-1.40625,0.29492188,0.578125,-0.114746094,-0.31835938,0.09765625,1.1953125,0.34179688,0.53125,0.03564453,-0.66796875,0.16796875,0.44140625,0.13085938,-1.2578125,0.84375,-0.6015625,0.67578125,0.671875,-2.015625,-2.109375,-1.328125,-2.171875,-0.49414062,0.34765625,2.28125,0.36914062,-0.5234375,0.036865234,-0.44726562,0.64453125,-0.27734375,-0.25390625,0.21191406,-0.38085938,0.14160156,0.91796875,1.453125,-0.47460938,0.048583984,0.084472656,0.2578125,0.77734375,-1.421875,-3.140625,1.265625,3.359375,0.6484375,-0.44726562,0.4921875
+8,0,1,1.078125,0.25195312,-0.16601562,-0.296875,0.109375,-0.42578125,-1.28125,0.7890625,0.23632812,0.46484375,0.13574219,0.921875,1.2421875,-0.13476562,-0.55859375,-0.13183594,-1.078125,-2.25,0.9375,0.21386719,1.7890625,-1.0234375,-1.2890625,-0.2109375,1.40625,-0.30078125,-0.71484375,-0.013305664,0.2578125,-0.14648438,-0.84765625,-0.61328125,1.9453125,-0.94921875,-0.93359375,-0.41601562,0.09472656,-1.140625,-2.578125,-0.453125,0.44726562,-0.87890625,-0.71484375,-0.29882812,-1.21875,0.5234375,-0.3203125,0.5,1.765625,1.4296875,-0.27148438,0.36328125,-1.6640625,1.2578125,-0.10595703,0.38671875,-0.5,1.8828125,0.71484375,-0.82421875,-0.69140625,-1.9453125,0.76953125,0.80078125,1.328125,1.46875,-1.4609375,-1.875,0.41601562,0.9375,1.421875,1.2265625,-0.41992188,-0.94921875,2.90625,1.78125,-0.3515625,-0.11669922,-0.5,0.26953125,-0.34765625,0.26757812,0.0007362366,2.34375,-0.48046875,-0.58984375,-0.41601562,2.5,1.0546875,0.91015625,0.8125,0.5390625,0.8203125,0.06933594,-0.84765625,-0.13769531,0.5546875,1.0390625,0.5078125,-0.9453125,0.19824219,1.0390625,-0.46484375,0.37109375,-0.16796875,0.76953125,-1.171875,-1.15625,0.375,0.33398438,-1.015625,-0.15820312,-0.55859375,-1.9453125,-1.78125,0.10253906,-0.625,1.2265625,-1.2578125,0.44140625,0.61328125,0.41210938,-2.203125,-1.359375,-0.8046875,0.18847656,-0.5625,0.43945312,0.6640625,-0.62890625,1.7578125,-0.25390625,0.14550781,-1.5078125,-0.65234375,0.5703125,0.55859375,0.19726562,-0.75,-2.5,1.0703125,2.59375,-0.7109375,-1.21875,-1.265625,1.234375,0.31640625,-0.15332031,0.31640625,0.6015625,0.56640625,0.98046875,0.6328125,-0.6171875,-0.77734375,-1.4609375,0.29296875,-0.049316406,1.171875,-0.64453125,0.35546875,0.57421875,0.080078125,-1.0703125,0.5859375,-0.28515625,0.57421875,-0.74609375,-1.2890625,1.2890625,0.14550781,0.65234375,-0.8984375,-0.52734375,-1.6953125,1.84375,-0.17871094,-0.58203125,0.07861328,-0.48632812,1.25,0.40625,0.828125,-0.68359375,-1.0703125,-1.5546875,0.18652344,0.36328125,0.6953125,0.625,-0.30078125,0.36328125,0.796875,-0.93359375,-0.6328125,-1.8828125,-0.6953125,1.25,1.03125,0.72265625
+8,0,2,0.87109375,0.984375,-0.17871094,0.12597656,0.08203125,-0.60546875,0.765625,1.0703125,1.078125,-0.80859375,1.3203125,1.5546875,0.053710938,-1.03125,-0.14355469,0.30078125,-0.67578125,-1.0078125,0.8359375,0.3359375,0.026489258,-0.36328125,-0.24609375,1.640625,-0.18164062,0.14550781,-2.78125,0.21191406,0.3203125,0.7890625,-0.5078125,-0.90625,1.3671875,-0.78125,0.57421875,0.59375,0.53515625,-0.93359375,-0.21972656,1.0859375,-1.2109375,1.109375,-0.34179688,-2.421875,-0.8671875,-0.7890625,0.25976562,-0.22949219,1.3046875,1.6015625,0.5546875,-0.47070312,-0.92578125,0.6796875,0.46875,0.26757812,0.19921875,-0.5078125,-1.0703125,-0.036376953,1.4765625,0.20703125,0.6328125,0.859375,0.50390625,-0.43554688,-0.14160156,0.55859375,-1.3671875,0.66796875,-0.09863281,0.5390625,0.84765625,-0.42578125,1.8359375,0.8046875,1.3984375,1.046875,-0.35546875,0.36914062,-0.0859375,-1.125,0.83984375,-1.4375,-0.45507812,0.3671875,-0.03173828,-0.036132812,1.75,0.1875,1.34375,-0.5390625,-1.03125,1.2421875,-1.09375,-0.59765625,-2.75,0.49023438,-0.6015625,-0.3671875,-1.453125,1.171875,0.38085938,0.19140625,-0.58984375,1.84375,-0.75,0.16601562,-1.1640625,0.30859375,-0.56640625,-0.42578125,0.91796875,-0.26367188,0.36328125,1.765625,-0.5078125,0.5546875,-0.15917969,-0.44335938,0.57421875,-0.62890625,-0.4921875,-0.94140625,-1.328125,2.109375,-1.2578125,-1.0546875,0.71875,-1.265625,1.3203125,-1.2109375,-0.6875,-0.76171875,-1.59375,0.65234375,1.09375,-0.52734375,0.057373047,-2.46875,1.703125,-0.9453125,-1.7578125,1.3671875,0.71484375,-0.04272461,0.875,-0.89453125,-1.171875,0.9140625,-1.296875,-0.70703125,0.68359375,-1.0234375,1.109375,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.07128906,1.65625,2.453125,1.4609375,-1.53125,0.100097656,-0.5390625,0.084472656,0.859375,-0.16113281,-1.015625,-0.296875,-1.6328125,-0.14746094,-1.5390625,1.0390625,0.9140625,-1.1640625,0.29492188,0.046875,-1.2734375,-1.4140625,0.54296875,-1.0,-0.45703125,0.24902344,0.022705078,0.56640625,-1.6171875,-1.625,-0.234375,2.46875,-0.3671875,-0.9609375,-1.546875,1.03125,1.4921875,1.09375,-0.421875,-0.59375
+8,0,3,1.25,1.5546875,0.10205078,-0.26171875,-1.609375,-0.016479492,-1.328125,1.4375,1.40625,-2.640625,1.7578125,-0.69140625,0.03100586,1.1171875,-0.49609375,0.46679688,0.29101562,0.41210938,1.859375,-0.016845703,-0.8046875,-0.034179688,-0.39453125,-0.25195312,-0.8671875,-1.0546875,-0.71484375,-0.22363281,0.09765625,-0.20605469,-0.25195312,0.01928711,1.0546875,0.71875,0.099609375,0.28710938,1.421875,0.14550781,-0.8828125,1.6875,-2.3125,-0.22265625,0.114746094,2.140625,0.26367188,0.921875,-0.87109375,0.71484375,-0.0625,0.87890625,-0.6484375,-0.09033203,-0.34960938,-0.25390625,-0.4140625,-0.45703125,0.19628906,0.8125,-1.1328125,0.16796875,0.546875,0.65234375,1.65625,-0.41601562,1.359375,0.43359375,-0.36328125,-1.2578125,-0.98046875,0.076660156,-0.2734375,-0.234375,0.6171875,0.04272461,-0.9140625,0.734375,0.14160156,0.34375,-1.53125,1.3203125,-0.81640625,-0.66796875,0.95703125,-1.2421875,-0.17773438,0.3359375,0.24609375,-1.3046875,0.10058594,0.51171875,1.953125,1.5859375,-0.35546875,1.0546875,-0.89453125,-0.14648438,-0.6796875,0.0028839111,0.19433594,-0.703125,-2.390625,0.87890625,1.71875,-1.7421875,0.03149414,1.5390625,-1.109375,-0.46484375,0.053466797,0.20800781,-0.08691406,-0.3125,1.328125,-0.7265625,0.06738281,0.080078125,-0.37890625,0.24707031,-0.25390625,0.859375,1.640625,-0.56640625,-0.15332031,-1.796875,-0.51171875,0.20214844,0.39257812,-1.5546875,-0.11767578,1.3828125,-0.890625,1.1171875,-1.3984375,0.6484375,-2.09375,1.515625,0.9140625,-0.59765625,-0.7109375,-2.328125,1.5859375,-1.1640625,-0.765625,0.03930664,-0.18945312,0.203125,0.48046875,-0.5859375,-0.76171875,1.2421875,-1.40625,0.29492188,0.578125,-0.114746094,-0.31835938,0.09765625,1.1953125,0.34179688,0.53125,0.03564453,-0.66796875,0.16796875,0.44140625,0.13085938,-1.2578125,0.84375,-0.6015625,0.67578125,0.671875,-2.015625,-2.109375,-1.328125,-2.171875,-0.49414062,0.34765625,2.28125,0.36914062,-0.5234375,0.036865234,-0.44726562,0.64453125,-0.27734375,-0.25390625,0.21191406,-0.38085938,0.14160156,0.91796875,1.453125,-0.47460938,0.048583984,0.084472656,0.2578125,0.77734375,-1.421875,-3.140625,1.265625,3.359375,0.6484375,-0.44726562,0.4921875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
index f8e891b1..90369015 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,5,0.78515625,1.0,-0.059326172,0.65234375,0.359375,-0.50390625,0.3359375,1.0859375,-0.15917969,-1.0546875,0.72265625,-0.24804688,-1.1328125,-0.8515625,-1.046875,0.14648438,-1.0859375,1.3671875,-0.43164062,0.30273438,-0.024780273,0.80078125,-0.33203125,-0.13867188,-0.61328125,-1.4765625,-2.765625,0.3125,-0.6328125,0.5078125,-1.0390625,-1.109375,-0.38671875,0.21972656,-0.63671875,-0.8984375,0.51171875,-1.265625,0.63671875,-0.10107422,-0.7578125,0.43554688,0.40039062,1.5859375,0.29882812,0.6953125,0.059570312,0.296875,-2.15625,0.80859375,1.28125,-1.2890625,-2.3125,-0.14160156,-0.92578125,-0.4453125,0.07861328,1.0078125,-0.27734375,-0.8125,-0.07714844,1.265625,-0.076660156,-0.171875,1.3046875,-0.58984375,0.33789062,-0.0034332275,0.22070312,0.94140625,0.18554688,-2.0,0.8984375,-1.1015625,-0.13476562,1.2734375,1.7578125,-0.1328125,0.22949219,-0.36328125,0.8984375,0.47265625,-0.17578125,-1.0859375,-0.46289062,1.0078125,0.07128906,-1.3203125,0.14355469,0.60546875,0.625,1.6875,0.88671875,0.59765625,-0.28125,-0.20898438,-1.1015625,0.80078125,0.12158203,0.6015625,1.6640625,0.671875,-0.90625,-0.21191406,0.5,-0.34960938,-0.9453125,-0.53125,0.42382812,-0.16308594,-1.59375,2.015625,0.078125,0.50390625,-0.359375,0.16796875,-0.4765625,1.5625,-0.3046875,1.4453125,1.46875,-1.296875,-0.78515625,-1.03125,0.025268555,0.20703125,0.3359375,-1.0546875,-2.203125,0.18457031,0.22167969,-0.78515625,-0.8515625,0.21289062,0.24511719,1.0,-0.44140625,-1.1015625,1.2421875,-1.890625,1.3515625,-0.54296875,-0.4375,-0.12060547,1.1796875,0.33398438,-0.96875,-0.115722656,2.21875,3.484375,-0.27734375,-0.74609375,-2.359375,2.203125,-0.48828125,0.23925781,0.20214844,1.265625,-1.5,0.4453125,-0.95703125,1.0546875,0.671875,1.75,-1.9375,0.234375,-1.2578125,0.828125,1.21875,-1.140625,-0.06640625,-1.0703125,1.8359375,-0.80078125,-0.48046875,0.0024719238,0.038330078,-0.7265625,1.2265625,0.17480469,-0.5625,0.31054688,1.53125,0.30273438,-0.06640625,0.43554688,1.3046875,0.50390625,0.953125,-2.96875,-0.12597656,-0.103027344,0.011413574,-1.2421875,-1.4140625,1.6640625,1.234375,-0.5390625,-0.890625,-0.50390625
-9,0,6,-0.16210938,1.3515625,-0.18554688,-0.9375,0.55859375,0.32421875,0.58203125,-0.81640625,1.578125,-1.453125,2.109375,-1.2890625,-0.4609375,-0.17871094,-0.43554688,-0.48242188,-1.0625,0.029785156,0.9296875,0.63671875,0.30664062,-0.98046875,1.2578125,0.07373047,-1.6171875,-2.0625,-0.64453125,-0.014831543,-0.08935547,1.328125,0.62109375,-0.2734375,0.703125,0.83984375,0.114746094,-0.17382812,0.59375,0.65234375,-0.30859375,0.00042533875,-0.45117188,0.56640625,1.0546875,0.90625,-0.72265625,0.99609375,0.07324219,-1.6484375,-0.703125,2.234375,0.578125,-1.109375,-0.609375,0.84765625,0.34960938,-0.515625,-0.0063476562,1.09375,0.55078125,-0.171875,0.96875,1.671875,0.20996094,0.49804688,0.99609375,0.24511719,-0.8515625,-0.90234375,0.24316406,0.9765625,-1.1015625,0.49609375,0.38476562,-0.53515625,-0.53515625,-0.34375,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.09033203,-0.040039062,2.171875,0.28320312,0.023925781,-2.203125,-0.55078125,0.671875,-0.13183594,2.203125,0.045898438,1.328125,-0.72265625,-0.11621094,-0.049804688,-1.2109375,0.20996094,-0.80859375,-1.5390625,-0.41210938,-0.6171875,1.4921875,-0.45507812,-0.54296875,1.109375,-2.046875,0.33007812,-0.21484375,-0.34179688,-0.4921875,2.109375,-0.43164062,-2.609375,-0.58203125,0.65625,-0.984375,0.84765625,-0.25,0.13378906,0.5625,-1.171875,0.026733398,-2.21875,-1.265625,2.296875,-1.46875,-0.70703125,-0.4140625,-0.38085938,1.0859375,0.07324219,-1.0546875,-0.03540039,-1.078125,2.375,-0.54296875,0.625,0.72265625,-0.8984375,1.421875,0.22167969,0.17089844,2.203125,0.06591797,-1.0703125,-0.053222656,-0.75,-1.265625,0.6484375,-0.49804688,-0.65234375,-0.68359375,0.052490234,-0.86328125,-0.05078125,0.19726562,-0.27148438,-1.2265625,2.296875,-0.69921875,-0.6484375,1.5,1.125,-2.421875,1.90625,-0.78515625,2.03125,-0.029541016,-0.14746094,-1.2421875,-0.02709961,-0.48242188,-1.125,0.6015625,0.3359375,-0.11328125,-2.6875,-0.5,-0.94140625,1.28125,0.578125,0.5078125,0.796875,-0.98828125,1.7734375,-1.28125,-0.25390625,-0.71484375,0.12695312,0.48632812,0.6640625,0.36523438,0.41796875,-0.12597656,-1.734375,0.2109375,-0.034423828,1.0,0.7265625
-9,0,7,-0.31445312,0.34179688,0.55859375,-2.0625,-0.025634766,0.67578125,1.8046875,-0.69921875,1.046875,-0.98046875,1.1171875,-1.359375,1.3671875,-0.6484375,-0.54296875,0.6953125,-0.045166016,-0.62890625,2.09375,-0.7421875,-0.78515625,-1.4140625,1.25,-0.15234375,-2.484375,-1.6171875,0.18164062,-0.7890625,-0.51171875,1.65625,0.52734375,-0.546875,-0.118652344,0.099121094,1.1015625,0.78125,-0.20898438,-0.13964844,-1.8203125,0.953125,-1.546875,0.51953125,0.94921875,0.97265625,0.6015625,1.375,-0.984375,-1.140625,-0.2265625,1.609375,-1.78125,0.25390625,-1.0078125,1.8671875,-0.25,0.625,0.20898438,0.27734375,-0.045166016,-0.34375,0.60546875,-0.045654297,-0.703125,0.52734375,0.6328125,1.4140625,-0.78515625,-1.1015625,-0.984375,2.109375,-1.1015625,-0.5703125,0.083496094,0.87109375,-0.40625,0.76953125,1.5703125,0.19335938,-0.21386719,0.22753906,1.125,0.4609375,-0.16113281,1.25,0.17089844,-0.82421875,-0.66015625,-1.2734375,-0.5703125,-0.328125,0.87109375,0.03540039,-0.83984375,-1.0,-1.546875,0.48242188,-0.5703125,-0.60546875,-0.25585938,-0.625,-1.3828125,0.0625,2.3125,1.1875,0.17480469,1.7890625,-0.92578125,-0.10107422,0.53125,-0.78515625,0.40039062,1.015625,0.90625,-2.578125,0.546875,-0.34179688,0.28710938,0.19042969,-0.76953125,0.5390625,1.75,-0.38671875,1.2421875,-2.0625,0.83203125,0.15625,-2.140625,-1.6015625,-0.19628906,0.24804688,0.7109375,0.6875,0.32421875,-1.5078125,0.578125,0.40429688,-0.25195312,0.6015625,-0.328125,-0.62890625,2.546875,1.671875,-0.25390625,1.71875,-0.875,-0.36132812,0.19042969,-0.15136719,-2.421875,0.66015625,-0.5390625,-0.140625,-0.4140625,-1.2578125,-1.234375,-0.92578125,0.33789062,-0.07519531,-0.055908203,2.09375,-2.078125,-0.67578125,1.1640625,-0.014465332,-1.6875,-0.109375,-0.17285156,1.3515625,0.024536133,0.46679688,0.34960938,0.50390625,-1.375,-0.30078125,0.035888672,-0.14941406,0.765625,-1.2890625,-0.25390625,0.111328125,-0.26953125,-1.140625,1.0546875,0.79296875,-0.640625,-0.17480469,0.3671875,0.0234375,-1.1484375,-0.17480469,-0.54296875,1.203125,1.265625,0.1328125,-0.029907227,-0.45703125,1.3359375,-0.6953125,2.046875,0.58203125
+9,0,0,0.78515625,1.0,-0.059326172,0.65234375,0.359375,-0.50390625,0.3359375,1.0859375,-0.15917969,-1.0546875,0.72265625,-0.24804688,-1.1328125,-0.8515625,-1.046875,0.14648438,-1.0859375,1.3671875,-0.43164062,0.30273438,-0.024780273,0.80078125,-0.33203125,-0.13867188,-0.61328125,-1.4765625,-2.765625,0.3125,-0.6328125,0.5078125,-1.0390625,-1.109375,-0.38671875,0.21972656,-0.63671875,-0.8984375,0.51171875,-1.265625,0.63671875,-0.10107422,-0.7578125,0.43554688,0.40039062,1.5859375,0.29882812,0.6953125,0.059570312,0.296875,-2.15625,0.80859375,1.28125,-1.2890625,-2.3125,-0.14160156,-0.92578125,-0.4453125,0.07861328,1.0078125,-0.27734375,-0.8125,-0.07714844,1.265625,-0.076660156,-0.171875,1.3046875,-0.58984375,0.33789062,-0.0034332275,0.22070312,0.94140625,0.18554688,-2.0,0.8984375,-1.1015625,-0.13476562,1.2734375,1.7578125,-0.1328125,0.22949219,-0.36328125,0.8984375,0.47265625,-0.17578125,-1.0859375,-0.46289062,1.0078125,0.07128906,-1.3203125,0.14355469,0.60546875,0.625,1.6875,0.88671875,0.59765625,-0.28125,-0.20898438,-1.1015625,0.80078125,0.12158203,0.6015625,1.6640625,0.671875,-0.90625,-0.21191406,0.5,-0.34960938,-0.9453125,-0.53125,0.42382812,-0.16308594,-1.59375,2.015625,0.078125,0.50390625,-0.359375,0.16796875,-0.4765625,1.5625,-0.3046875,1.4453125,1.46875,-1.296875,-0.78515625,-1.03125,0.025268555,0.20703125,0.3359375,-1.0546875,-2.203125,0.18457031,0.22167969,-0.78515625,-0.8515625,0.21289062,0.24511719,1.0,-0.44140625,-1.1015625,1.2421875,-1.890625,1.3515625,-0.54296875,-0.4375,-0.12060547,1.1796875,0.33398438,-0.96875,-0.115722656,2.21875,3.484375,-0.27734375,-0.74609375,-2.359375,2.203125,-0.48828125,0.23925781,0.20214844,1.265625,-1.5,0.4453125,-0.95703125,1.0546875,0.671875,1.75,-1.9375,0.234375,-1.2578125,0.828125,1.21875,-1.140625,-0.06640625,-1.0703125,1.8359375,-0.80078125,-0.48046875,0.0024719238,0.038330078,-0.7265625,1.2265625,0.17480469,-0.5625,0.31054688,1.53125,0.30273438,-0.06640625,0.43554688,1.3046875,0.50390625,0.953125,-2.96875,-0.12597656,-0.103027344,0.011413574,-1.2421875,-1.4140625,1.6640625,1.234375,-0.5390625,-0.890625,-0.50390625
+9,0,1,-0.16210938,1.3515625,-0.18554688,-0.9375,0.55859375,0.32421875,0.58203125,-0.81640625,1.578125,-1.453125,2.109375,-1.2890625,-0.4609375,-0.17871094,-0.43554688,-0.48242188,-1.0625,0.029785156,0.9296875,0.63671875,0.30664062,-0.98046875,1.2578125,0.07373047,-1.6171875,-2.0625,-0.64453125,-0.014831543,-0.08935547,1.328125,0.62109375,-0.2734375,0.703125,0.83984375,0.114746094,-0.17382812,0.59375,0.65234375,-0.30859375,0.00042533875,-0.45117188,0.56640625,1.0546875,0.90625,-0.72265625,0.99609375,0.07324219,-1.6484375,-0.703125,2.234375,0.578125,-1.109375,-0.609375,0.84765625,0.34960938,-0.515625,-0.0063476562,1.09375,0.55078125,-0.171875,0.96875,1.671875,0.20996094,0.49804688,0.99609375,0.24511719,-0.8515625,-0.90234375,0.24316406,0.9765625,-1.1015625,0.49609375,0.38476562,-0.53515625,-0.53515625,-0.34375,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.09033203,-0.040039062,2.171875,0.28320312,0.023925781,-2.203125,-0.55078125,0.671875,-0.13183594,2.203125,0.045898438,1.328125,-0.72265625,-0.11621094,-0.049804688,-1.2109375,0.20996094,-0.80859375,-1.5390625,-0.41210938,-0.6171875,1.4921875,-0.45507812,-0.54296875,1.109375,-2.046875,0.33007812,-0.21484375,-0.34179688,-0.4921875,2.109375,-0.43164062,-2.609375,-0.58203125,0.65625,-0.984375,0.84765625,-0.25,0.13378906,0.5625,-1.171875,0.026733398,-2.21875,-1.265625,2.296875,-1.46875,-0.70703125,-0.4140625,-0.38085938,1.0859375,0.07324219,-1.0546875,-0.03540039,-1.078125,2.375,-0.54296875,0.625,0.72265625,-0.8984375,1.421875,0.22167969,0.17089844,2.203125,0.06591797,-1.0703125,-0.053222656,-0.75,-1.265625,0.6484375,-0.49804688,-0.65234375,-0.68359375,0.052490234,-0.86328125,-0.05078125,0.19726562,-0.27148438,-1.2265625,2.296875,-0.69921875,-0.6484375,1.5,1.125,-2.421875,1.90625,-0.78515625,2.03125,-0.029541016,-0.14746094,-1.2421875,-0.02709961,-0.48242188,-1.125,0.6015625,0.3359375,-0.11328125,-2.6875,-0.5,-0.94140625,1.28125,0.578125,0.5078125,0.796875,-0.98828125,1.7734375,-1.28125,-0.25390625,-0.71484375,0.12695312,0.48632812,0.6640625,0.36523438,0.41796875,-0.12597656,-1.734375,0.2109375,-0.034423828,1.0,0.7265625
+9,0,2,-0.31445312,0.34179688,0.55859375,-2.0625,-0.025634766,0.67578125,1.8046875,-0.69921875,1.046875,-0.98046875,1.1171875,-1.359375,1.3671875,-0.6484375,-0.54296875,0.6953125,-0.045166016,-0.62890625,2.09375,-0.7421875,-0.78515625,-1.4140625,1.25,-0.15234375,-2.484375,-1.6171875,0.18164062,-0.7890625,-0.51171875,1.65625,0.52734375,-0.546875,-0.118652344,0.099121094,1.1015625,0.78125,-0.20898438,-0.13964844,-1.8203125,0.953125,-1.546875,0.51953125,0.94921875,0.97265625,0.6015625,1.375,-0.984375,-1.140625,-0.2265625,1.609375,-1.78125,0.25390625,-1.0078125,1.8671875,-0.25,0.625,0.20898438,0.27734375,-0.045166016,-0.34375,0.60546875,-0.045654297,-0.703125,0.52734375,0.6328125,1.4140625,-0.78515625,-1.1015625,-0.984375,2.109375,-1.1015625,-0.5703125,0.083496094,0.87109375,-0.40625,0.76953125,1.5703125,0.19335938,-0.21386719,0.22753906,1.125,0.4609375,-0.16113281,1.25,0.17089844,-0.82421875,-0.66015625,-1.2734375,-0.5703125,-0.328125,0.87109375,0.03540039,-0.83984375,-1.0,-1.546875,0.48242188,-0.5703125,-0.60546875,-0.25585938,-0.625,-1.3828125,0.0625,2.3125,1.1875,0.17480469,1.7890625,-0.92578125,-0.10107422,0.53125,-0.78515625,0.40039062,1.015625,0.90625,-2.578125,0.546875,-0.34179688,0.28710938,0.19042969,-0.76953125,0.5390625,1.75,-0.38671875,1.2421875,-2.0625,0.83203125,0.15625,-2.140625,-1.6015625,-0.19628906,0.24804688,0.7109375,0.6875,0.32421875,-1.5078125,0.578125,0.40429688,-0.25195312,0.6015625,-0.328125,-0.62890625,2.546875,1.671875,-0.25390625,1.71875,-0.875,-0.36132812,0.19042969,-0.15136719,-2.421875,0.66015625,-0.5390625,-0.140625,-0.4140625,-1.2578125,-1.234375,-0.92578125,0.33789062,-0.07519531,-0.055908203,2.09375,-2.078125,-0.67578125,1.1640625,-0.014465332,-1.6875,-0.109375,-0.17285156,1.3515625,0.024536133,0.46679688,0.34960938,0.50390625,-1.375,-0.30078125,0.035888672,-0.14941406,0.765625,-1.2890625,-0.25390625,0.111328125,-0.26953125,-1.140625,1.0546875,0.79296875,-0.640625,-0.17480469,0.3671875,0.0234375,-1.1484375,-0.17480469,-0.54296875,1.203125,1.265625,0.1328125,-0.029907227,-0.45703125,1.3359375,-0.6953125,2.046875,0.58203125
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv
index 4c3b4b21..b31efdb2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv
@@ -1,4 +1,8 @@
 sequenceId,itemPosition,itemId
+0,4,113
+0,5,113
+0,6,113
+0,7,113
 0,8,113
 0,9,113
 0,10,113
@@ -15,7 +19,3 @@ sequenceId,itemPosition,itemId
 0,21,113
 0,22,113
 0,23,113
-0,24,113
-0,25,113
-0,26,113
-0,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
index fb1845ae..72488290 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
@@ -1,4 +1,8 @@
 sequenceId,itemPosition,itemId
+1,4,113
+1,5,113
+1,6,113
+1,7,113
 1,8,113
 1,9,113
 1,10,113
@@ -15,7 +19,3 @@ sequenceId,itemPosition,itemId
 1,21,113
 1,22,113
 1,23,113
-1,24,113
-1,25,113
-1,26,113
-1,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
index b59ddd38..9966bc14 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
@@ -1,5 +1,10 @@
 sequenceId,itemPosition,itemId
-2,8,103
+2,3,103
+2,4,122
+2,5,122
+2,6,122
+2,7,122
+2,8,122
 2,9,122
 2,10,122
 2,11,122
@@ -14,8 +19,3 @@ sequenceId,itemPosition,itemId
 2,20,122
 2,21,122
 2,22,122
-2,23,122
-2,24,122
-2,25,122
-2,26,122
-2,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
index ec95ae85..bcce6828 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
@@ -1,5 +1,10 @@
 sequenceId,itemPosition,itemId
-3,8,103
+3,3,103
+3,4,124
+3,5,124
+3,6,124
+3,7,124
+3,8,124
 3,9,124
 3,10,124
 3,11,124
@@ -14,11 +19,11 @@ sequenceId,itemPosition,itemId
 3,20,124
 3,21,124
 3,22,124
-3,23,124
-3,24,124
-3,25,124
-3,26,124
-3,27,124
+4,3,122
+4,4,122
+4,5,122
+4,6,122
+4,7,122
 4,8,122
 4,9,122
 4,10,122
@@ -34,8 +39,3 @@ sequenceId,itemPosition,itemId
 4,20,122
 4,21,122
 4,22,122
-4,23,122
-4,24,122
-4,25,122
-4,26,122
-4,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
index e689f7f2..90a13a9d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
@@ -1,5 +1,10 @@
 sequenceId,itemPosition,itemId
-5,8,103
+5,3,103
+5,4,122
+5,5,122
+5,6,122
+5,7,122
+5,8,122
 5,9,122
 5,10,122
 5,11,122
@@ -14,8 +19,3 @@ sequenceId,itemPosition,itemId
 5,20,122
 5,21,122
 5,22,122
-5,23,122
-5,24,122
-5,25,122
-5,26,122
-5,27,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
index 0acf1d86..f87b5d08 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
@@ -1,4 +1,8 @@
 sequenceId,itemPosition,itemId
+6,4,113
+6,5,113
+6,6,113
+6,7,113
 6,8,113
 6,9,113
 6,10,113
@@ -15,10 +19,11 @@ sequenceId,itemPosition,itemId
 6,21,113
 6,22,113
 6,23,113
-6,24,113
-6,25,113
-6,26,113
-6,27,113
+7,3,113
+7,4,113
+7,5,113
+7,6,113
+7,7,113
 7,8,113
 7,9,113
 7,10,113
@@ -34,8 +39,3 @@ sequenceId,itemPosition,itemId
 7,20,113
 7,21,113
 7,22,113
-7,23,113
-7,24,113
-7,25,113
-7,26,113
-7,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
index a074f362..fcb6691b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
@@ -1,21 +1,21 @@
 sequenceId,itemPosition,itemId
+8,4,122
+8,5,122
+8,6,122
+8,7,122
 8,8,122
 8,9,122
-8,10,122
+8,10,113
 8,11,122
 8,12,122
 8,13,122
-8,14,113
+8,14,122
 8,15,122
 8,16,122
-8,17,122
-8,18,122
-8,19,122
-8,20,122
+8,17,113
+8,18,113
+8,19,113
+8,20,113
 8,21,113
 8,22,113
 8,23,113
-8,24,113
-8,25,113
-8,26,113
-8,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
index c97879eb..2b8cfd75 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
@@ -1,4 +1,9 @@
 sequenceId,itemPosition,itemId
+9,3,113
+9,4,113
+9,5,113
+9,6,113
+9,7,113
 9,8,113
 9,9,113
 9,10,113
@@ -14,8 +19,3 @@ sequenceId,itemPosition,itemId
 9,20,113
 9,21,113
 9,22,113
-9,23,113
-9,24,113
-9,25,113
-9,26,113
-9,27,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv
index 5aa808f6..783e9350 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,113
+0,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
index c3d900b3..9eb44d36 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,113
+1,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
index 460f1499..ac8d014f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,103
+2,3,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
index 90765ba3..19f45567 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,103
-4,8,122
+3,3,103
+4,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
index 60a3b97e..e7767b59 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,103
+5,3,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
index 953b3bbc..e09b589d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,113
-7,8,113
+6,4,113
+7,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
index f3ae7b5b..d9b241c6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,122
+8,4,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
index ba964099..d5752b57 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,113
+9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv
index b376d486..461de2e0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-0,6,119
-0,7,103
-0,8,111
+0,2,119
+0,3,103
+0,4,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
index 5158426c..d32d6aab 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-1,6,122
-1,7,124
-1,8,113
+1,2,122
+1,3,124
+1,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
index 613c7045..118f3041 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-2,6,121
-2,7,102
-2,8,126
+2,1,121
+2,2,102
+2,3,126
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
index a4583793..d7dc85f6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId
-3,6,121
-3,7,109
-3,8,102
-4,6,127
-4,7,119
-4,8,120
+3,1,121
+3,2,109
+3,3,102
+4,1,127
+4,2,119
+4,3,120
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv
index 670c736f..a133efa3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-5,6,126
-5,7,104
-5,8,112
+5,1,126
+5,2,104
+5,3,112
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
index 285c20cb..0a48eeb8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId
-6,6,102
-6,7,118
-6,8,113
-7,6,109
-7,7,103
-7,8,113
+6,2,102
+6,3,118
+6,4,113
+7,1,109
+7,2,103
+7,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
index 79312602..12b5fd01 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-8,6,104
-8,7,121
-8,8,112
+8,2,104
+8,3,121
+8,4,112
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
index 2ee41e48..de154d9e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId
-9,6,118
-9,7,118
-9,8,118
+9,1,118
+9,2,118
+9,3,118
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv
index 5aa808f6..783e9350 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,113
+0,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
index 1213b475..4ac41fe6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,127
+1,4,127
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv
index 7e8daf90..5b9010cb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,113
+2,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
index 90765ba3..19f45567 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,103
-4,8,122
+3,3,103
+4,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
index 60a3b97e..e7767b59 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,103
+5,3,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
index 132c76ec..6ed65b7f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,122
-7,8,122
+6,4,122
+7,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
index bc91b117..c2bb5aec 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,103
+8,4,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
index 824e36f4..40d6064e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,124
+9,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
index 799db1cf..4f60920f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-0,6,119,5,7
-0,7,103,4,5
-0,8,111,6,0
+0,2,119,5,7
+0,3,103,4,5
+0,4,111,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
index 84336320..79b88c4f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-1,6,122,0,1
-1,7,121,9,1
-1,8,108,9,1
+1,2,122,0,1
+1,3,121,9,1
+1,4,108,9,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
index f2206ac1..98346781 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-2,6,121,6,1
-2,7,127,7,5
-2,8,112,4,0
+2,1,121,6,1
+2,2,127,7,5
+2,3,112,4,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
index bba6a834..248bef03 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-3,6,121,6,1
-3,7,109,3,1
-3,8,110,0,2
-4,6,121,6,1
-4,7,121,3,1
-4,8,102,0,1
+3,1,121,6,1
+3,2,109,3,1
+3,3,110,0,2
+4,1,121,6,1
+4,2,121,3,1
+4,3,102,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
index a3007f68..2dbc2594 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-5,6,113,6,1
-5,7,113,4,0
-5,8,121,4,5
+5,1,113,6,1
+5,2,113,4,0
+5,3,121,4,5
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
index 3fe6eafb..36d5bae9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
@@ -1,7 +1,7 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-6,6,102,4,5
-6,7,118,9,1
-6,8,121,4,6
-7,6,109,3,1
-7,7,121,0,2
-7,8,108,5,7
+6,2,102,4,5
+6,3,118,9,1
+6,4,121,4,6
+7,1,109,3,1
+7,2,121,0,2
+7,3,108,5,7
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
index e453627c..3e0792ba 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-8,6,102,8,8
-8,7,121,4,1
-8,8,121,3,1
+8,2,102,8,8
+8,3,121,4,1
+8,4,121,3,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
index 62834e55..20f73580 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
@@ -1,4 +1,4 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2
-9,6,108,9,1
-9,7,118,4,8
-9,8,103,4,8
+9,1,108,9,1
+9,2,118,4,8
+9,3,103,4,8
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv
index 5aa808f6..783e9350 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,113
+0,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv
index c3d900b3..9eb44d36 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,113
+1,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
index 8302df2f..53c89ed7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,122
+2,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
index 3ee3129e..6b9ee33a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,122
-4,8,113
+3,3,122
+4,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv
index 4e9004da..2b757dfc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,113
+5,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
index c71ccfa4..579df4c3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,113
-7,8,122
+6,4,113
+7,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv
index 346e8242..3f1e127e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,113
+8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv
index ba964099..d5752b57 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,113
+9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
index 7240bc4f..d5e918cd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,124
+0,4,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
index dea53fa9..e87652fa 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,122
+1,4,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
index 1616d64b..8ace2dd0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,124
+2,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
index a467ff6d..47d128ca 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,122
-4,8,124
+3,3,122
+4,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
index 81899090..081fa079 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,124
+5,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
index 3e2fb799..12c49755 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,113
-7,8,128
+6,4,113
+7,3,128
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
index 346e8242..3f1e127e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,113
+8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
index 56680af1..d302645a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,121
+9,3,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
index f27b432d..0502ed89 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,100
+0,4,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
index 72fb79cf..79a8ecae 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,115
+1,4,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
index 53195e3d..20ab5369 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,102
+2,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
index b08ec867..e4054674 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,115
-4,8,102
+3,3,115
+4,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
index a620b023..f37176b4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,122
+5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
index ebfed34f..25b28cdd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,115
-7,8,111
+6,4,115
+7,3,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
index ae1ba6f4..8eb4c4a0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,111
+8,4,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
index 56680af1..d302645a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,121
+9,3,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
index f27b432d..0502ed89 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-0,8,100
+0,4,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv
index 72fb79cf..79a8ecae 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-1,8,115
+1,4,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
index 8302df2f..53c89ed7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-2,8,122
+2,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
index b08ec867..e4054674 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-3,8,115
-4,8,102
+3,3,115
+4,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
index a620b023..f37176b4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-5,8,122
+5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
index ebfed34f..25b28cdd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId
-6,8,115
-7,8,111
+6,4,115
+7,3,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
index a2894aea..59349bc1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,8,102
+8,4,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
index 56680af1..d302645a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-9,8,121
+9,3,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
index 31f237e5..ed59d730 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,8,109,[other],0.00014168024
+0,4,109,[other],0.00014168024
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
index 5875e518..98c7c6b4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,8,102,[unknown],-0.09726027
+1,4,102,[unknown],-0.09726027
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
index 7521792f..461de44b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,[mask],[unknown],-0.12791729
+2,3,[mask],[unknown],-0.12791729
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
index 9bf8759e..8cb7700d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,8,103,4,0.043316968
-4,8,110,4,0.032937437
+3,3,103,4,0.043316968
+4,3,110,4,0.032937437
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
index c93e224e..02507d44 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,8,110,9,-0.075533785
+5,3,110,9,-0.075533785
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
index ddd61771..93e77202 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,8,122,7,-0.04820269
-7,8,113,[other],0.057539806
+6,4,122,7,-0.04820269
+7,3,113,[other],0.057539806
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
index 2911647f..c2229265 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,110,[mask],-0.095947556
+8,4,110,[mask],-0.095947556
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
index 9ad6394e..ced3fa21 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,8,114,6,-0.0014844015
+9,3,114,6,-0.0014844015
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
index 09619471..626180f8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,8,102,4,0.016973361
+0,4,102,4,0.016973361
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
index 99102984..35dc1d9f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,8,100,[unknown],-0.09465362
+1,4,100,[unknown],-0.09465362
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
index b2187677..c0b5fba4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,8,100,[unknown],-0.12798095
+2,3,100,[unknown],-0.12798095
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
index 2fbedb5c..5fc0d368 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,8,111,4,0.058097668
-4,8,128,4,0.047720857
+3,3,111,4,0.058097668
+4,3,128,4,0.047720857
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
index 5063e0ad..8901171a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,8,102,6,-0.06665557
+5,3,102,6,-0.06665557
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
index 2740828e..5c04fea5 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,8,110,7,-0.029965512
-7,8,102,4,0.083941795
+6,4,110,7,-0.029965512
+7,3,102,4,0.083941795
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
index a3fb49ac..30927882 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,8,128,0,-0.07864175
+8,4,128,0,-0.07864175
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
index 5d07e206..69f195e6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,8,125,6,0.00074065477
+9,3,125,6,0.00074065477
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index 78d03599..d1af2939 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.03839054
-0,22,0.3938979
-0,23,0.48175326
-0,24,0.19821997
-0,25,0.3539636
-0,26,0.25612468
-0,27,0.24314602
-0,28,0.44780913
-0,29,0.01864058
-0,30,0.039980453
-0,31,0.22218053
-0,32,-0.2989616
-0,33,-0.10028865
-0,34,-0.07033796
-0,35,-0.002902084
-0,36,-0.01792423
-0,37,-0.036144238
-0,38,0.11984898
-0,39,0.051461555
-0,40,0.11685391
-1,19,0.24713944
-1,20,0.2920655
-1,21,0.13532351
-1,22,0.22817066
-1,23,0.090397455
-1,24,-0.13223606
-1,25,0.19821997
-1,26,-0.03339876
-1,27,0.072926216
-1,28,-0.11127058
-1,29,-0.33290574
-1,30,0.034239903
-1,31,-0.07483056
-1,32,-0.14122127
-1,33,0.06344183
-1,34,-0.11426565
-1,35,-0.09978948
-1,36,0.05919882
-1,37,-0.035395473
-1,38,0.0037588265
-2,17,0.2920655
-2,18,0.15029885
-2,19,0.07043032
-2,20,-0.050121233
-2,21,0.12184569
-2,22,-0.060354386
-2,23,0.20021668
-2,24,0.054955803
-2,25,-0.04438068
-2,26,-0.07233467
-2,27,-0.05062041
-2,28,-0.1981276
-2,29,0.012400852
-2,30,-0.03127725
-2,31,-0.121254146
-2,32,-0.04138561
-2,33,-0.122751676
-2,34,-0.11426565
-2,35,-0.052117944
-2,36,0.066936076
-3,17,0.03473908
-3,18,0.18623969
-3,19,0.32600963
-3,20,0.030496065
-3,21,0.06843361
-3,22,0.3938979
-3,23,0.25013453
-3,24,0.19222984
-3,25,-0.10577962
-3,26,-0.114764825
-3,27,0.04971443
-3,28,0.102377735
-3,29,-0.055861782
-3,30,-0.0040096357
-3,31,0.12533994
-3,32,-0.21709637
-3,33,-0.007753473
-3,34,0.021136472
-3,35,-0.055362605
-3,36,0.17326106
-4,14,0.16926762
-4,15,-0.10028865
-4,16,-0.042883147
-4,17,0.17725448
-4,18,0.21119861
-4,19,-0.054863427
-4,20,-0.018797792
-4,21,-0.0050235917
-4,22,-0.13523114
-4,23,-0.2670142
-4,24,0.048466485
-4,25,-0.037392184
-4,26,0.061694708
-4,27,0.13432515
-4,28,-0.06634453
-4,29,-0.065346174
-4,30,-0.21609803
-4,31,-0.03639383
-4,32,-0.04188479
-4,33,0.19222984
-5,14,0.106371164
-5,15,0.44581243
-5,16,0.24414438
-5,17,0.14730379
-5,18,0.40188473
-5,19,0.17625612
-5,20,0.17026599
-5,21,0.1403153
-5,22,-0.07383221
-5,23,0.03149442
-5,24,-0.22308652
-5,25,-0.0021962146
-5,26,0.042226754
-5,27,-0.021917658
-5,28,0.016394278
-5,29,-0.02690944
-5,30,-0.07832481
-5,31,-0.06684371
-5,32,-0.20910953
-5,33,-0.095296875
-6,18,0.024505924
-6,19,0.13831857
-6,20,0.15329392
-6,21,0.16427584
-6,22,0.07192786
-6,23,0.044972237
-6,24,0.0412284
-6,25,-0.10078783
-6,26,-0.13822621
-6,27,-0.24604872
-6,28,-0.061103154
-6,29,-0.07782563
-6,30,-0.17915882
-6,31,-0.13323443
-6,32,0.044972237
-6,33,-0.03489629
-6,34,0.018016607
-6,35,-0.012994845
-6,36,0.05520539
-6,37,-0.006505527
-7,15,0.15229557
-7,16,0.4877434
-7,17,0.18424298
-7,18,0.029622503
-7,19,0.24015094
-7,20,0.32001948
-7,21,0.0098425625
-7,22,-0.109273866
-7,23,-0.11626236
-7,24,-0.03364835
-7,25,-0.20611446
-7,26,0.056952514
-7,27,-0.20711282
-7,28,0.009031397
-7,29,-0.006973507
-7,30,-0.12025579
-7,31,-0.049622055
-7,32,-0.16917527
-7,33,-0.0004529903
-7,34,0.02163565
-8,20,0.076919645
-8,21,-0.09978948
-8,22,-0.011310118
-8,23,0.10187856
-8,24,0.1722627
-8,25,0.09538924
-8,26,0.061694708
-8,27,0.16727091
-8,28,-0.23107336
-8,29,0.07592129
-8,30,-0.0009950667
-8,31,-0.018298615
-8,32,-0.013369229
-8,33,0.0527095
-8,34,0.04721854
-8,35,-0.019047381
-8,36,0.000061787316
-8,37,0.18524134
-8,38,-0.07882399
-8,39,0.007346671
-9,16,0.23016739
-9,17,0.25412795
-9,18,0.23815423
-9,19,0.16028242
-9,20,0.10736952
-9,21,0.20720518
-9,22,0.06394101
-9,23,-0.095296875
-9,24,0.05894923
-9,25,-0.04837411
-9,26,-0.057359315
-9,27,0.03224319
-9,28,-0.13523114
-9,29,-0.046127804
-9,30,0.3180228
-9,31,-0.15519828
-9,32,-0.14221963
-9,33,0.08640403
-9,34,0.08390814
-9,35,-0.017425053
+0,21,-0.25203884
+0,22,0.07592129
+0,23,0.29406223
+0,24,0.2411493
+0,25,0.21718875
+0,26,0.29605892
+0,27,0.13432515
+0,28,0.20121504
+0,29,-0.016863476
+0,30,-0.21210459
+0,31,-0.022416836
+0,32,0.042226754
+0,33,0.08740239
+0,34,0.2900688
+0,35,-0.06784207
+0,36,-0.09479769
+0,37,0.01564551
+0,38,-0.12325086
+0,39,-0.25403556
+0,40,-0.032525197
+1,19,0.25113288
+1,20,0.16227913
+1,21,0.07142868
+1,22,0.27209836
+1,23,0.27209836
+1,24,-0.11975661
+1,25,0.1148572
+1,26,-0.029405331
+1,27,-0.008003062
+1,28,-0.02528711
+1,29,-0.089306735
+1,30,0.21020025
+1,31,0.07192786
+1,32,0.07642046
+1,33,0.11385884
+1,34,0.062443476
+1,35,0.20021668
+1,36,0.08340896
+1,37,0.18224627
+1,38,-0.05960562
+2,17,0.054456625
+2,18,0.001980504
+2,19,-0.08032152
+2,20,-0.015865121
+2,21,-0.05710973
+2,22,-0.11027222
+2,23,0.11186212
+2,24,-0.09030509
+2,25,0.029123325
+2,26,0.034239903
+2,27,0.035987027
+2,28,0.0001475836
+2,29,-0.021792863
+2,30,0.07642046
+2,31,-0.17217033
+2,32,0.16227913
+2,33,-0.14421634
+2,34,-0.102784544
+2,35,-0.09180263
+2,36,-0.22308652
+3,17,0.23116574
+3,18,0.20421012
+3,19,0.42384857
+3,20,0.16727091
+3,21,0.24713944
+3,22,0.11136295
+3,23,0.008781808
+3,24,0.088899925
+3,25,-0.07982235
+3,26,0.11136295
+3,27,-0.1522032
+3,28,-0.085313305
+3,29,0.029248118
+3,30,0.035987027
+3,31,0.012463248
+3,32,-0.101287015
+3,33,-0.101287015
+3,34,0.10886706
+3,35,-0.0046804063
+3,36,-0.043382324
+4,14,0.2900688
+4,15,0.17425941
+4,16,0.007783452
+4,17,-0.08181906
+4,18,0.016643867
+4,19,-0.04487986
+4,20,0.2780885
+4,21,-0.00597515
+4,22,-0.015677929
+4,23,-0.11726072
+4,24,-0.021044096
+4,25,0.054955803
+4,26,-0.14122127
+4,27,-0.11726072
+4,28,-0.10428208
+4,29,-0.10178619
+4,30,-0.14122127
+4,31,-0.028781358
+4,32,-0.21709637
+4,33,-0.15719499
+5,14,0.05520539
+5,15,0.33199978
+5,16,0.27209836
+5,17,0.2231789
+5,18,0.11735309
+5,19,0.11335966
+5,20,0.051461555
+5,21,0.2391526
+5,22,-0.03364835
+5,23,-0.20411775
+5,24,0.16427584
+5,25,-0.19213746
+5,26,0.037234973
+5,27,-0.09579605
+5,28,0.056702927
+5,29,0.045471415
+5,30,0.20421012
+5,31,-0.006349534
+5,32,0.100880206
+5,33,0.030246476
+6,18,0.29605892
+6,19,0.29605892
+6,20,0.1792512
+6,21,0.17425941
+6,22,0.19821997
+6,23,0.0434747
+6,24,0.0016997161
+6,25,-0.07483056
+6,26,-0.065845355
+6,27,0.015146332
+6,28,-0.026160672
+6,29,0.021136472
+6,30,-0.09629523
+6,31,-0.16318512
+6,32,-0.22907665
+6,33,0.090896636
+6,34,-0.06509659
+6,35,-0.07982235
+6,36,0.015271126
+6,37,-0.08481413
+7,15,0.072926216
+7,16,0.014896743
+7,17,-0.029654922
+7,18,-0.043631915
+7,19,-0.05261712
+7,20,0.2740951
+7,21,-0.11626236
+7,22,0.023881953
+7,23,-0.0000942059
+7,24,-0.064597405
+7,25,0.2780885
+7,26,0.0055371495
+7,27,0.054706212
+7,28,0.032991957
+7,29,0.078916356
+7,30,0.24514273
+7,31,0.090397455
+7,32,0.041977167
+7,33,-0.047874928
+7,34,-0.05910644
+8,20,-0.08281741
+8,21,0.21918546
+8,22,0.14131364
+8,23,0.04996402
+8,24,0.21119861
+8,25,0.07242704
+8,26,0.0065355063
+8,27,-0.018922588
+8,28,-0.051618766
+8,29,-0.18315226
+8,30,-0.22408487
+8,31,0.055704568
+8,32,-0.10178619
+8,33,-0.14022292
+8,34,0.13732022
+8,35,-0.048873287
+8,36,0.16327749
+8,37,0.004351601
+8,38,0.08690321
+8,39,-0.024538344
+9,16,0.3958946
+9,17,0.37193403
+9,18,0.082909785
+9,19,0.018765375
+9,20,0.34398004
+9,21,0.17425941
+9,22,0.084407315
+9,23,0.09938267
+9,24,0.017143045
+9,25,-0.1751654
+9,26,-0.107277155
+9,27,0.10886706
+9,28,0.26211482
+9,29,-0.026784645
+9,30,-0.05261712
+9,31,-0.14721142
+9,32,-0.11376647
+9,33,0.03149442
+9,34,0.019264553
+9,35,0.04671936
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
index 744668f3..1cf52011 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
@@ -11,8 +11,8 @@ sequenceId,itemPosition,itemValue
 2,9,0.11735309
 3,8,0.36993733
 3,9,0.29805565
-4,8,-0.13523114
-5,8,0.14131364
+4,7,-0.13523114
+5,7,0.14131364
 6,8,0.3359932
 6,9,0.26810494
 6,10,0.31402937
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
index 2d1901f7..68074514 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
@@ -11,8 +11,8 @@ sequenceId,itemPosition,itemValue
 2,9,-0.18891405
 3,8,0.08957863
 3,9,0.17450903
-4,8,0.37695938
-5,8,-0.18002598
+4,7,0.37695938
+5,7,-0.18002598
 6,8,0.2999295
 6,9,0.33745688
 6,10,0.007867463
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
index 27254750..7c11814b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
@@ -11,8 +11,8 @@ sequenceId,itemPosition,itemValue
 2,9,-0.29939193
 3,8,-0.1360021
 3,9,0.23850942
-4,8,0.104493044
-5,8,-0.3471238
+4,7,0.104493044
+5,7,-0.3471238
 6,8,0.09990344
 6,9,-0.11305408
 6,10,0.15406075
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
index e1bc7a5b..22ba483f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
@@ -11,8 +11,8 @@ sequenceId,itemPosition,itemValue
 2,9,-0.24058047
 3,8,0.01506035
 3,9,0.37075448
-4,8,0.3579036
-5,8,-0.12813523
+4,7,0.3579036
+5,7,-0.12813523
 6,8,0.3395452
 6,9,0.2771266
 6,10,0.31567925
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
index aaee17e3..9a24e80c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,127,1,1,2,1
-0,9,113,4,9,8,1
-0,10,113,1,1,8,1
-0,11,113,1,9,8,1
-0,12,113,1,9,8,1
-0,13,113,1,9,8,1
-0,14,113,1,8,8,1
-0,15,113,1,0,8,1
-0,16,113,1,9,7,1
-0,17,121,1,8,7,1
-0,18,105,1,8,7,1
-0,19,105,1,8,7,1
-0,20,105,1,8,7,1
-0,21,113,1,7,7,1
-0,22,113,6,7,7,1
-0,23,113,6,7,2,1
-0,24,113,1,7,8,1
-0,25,113,1,7,8,1
-0,26,113,1,7,8,1
-0,27,113,1,0,8,1
-0,28,113,1,0,8,1
-0,29,113,1,0,8,1
-0,30,113,1,0,8,1
-0,31,113,1,0,8,1
-0,32,113,1,0,7,1
-0,33,108,1,8,7,1
-0,34,108,1,0,7,1
-0,35,108,6,0,7,1
-0,36,108,1,0,7,1
-0,37,108,6,7,7,1
+0,4,127,1,1,2,1
+0,5,127,1,8,8,1
+0,6,113,1,1,7,0
+0,7,105,8,9,7,1
+0,8,127,1,1,0,1
+0,9,113,1,7,7,1
+0,10,113,8,9,1,1
+0,11,113,1,1,7,1
+0,12,105,8,5,7,1
+0,13,113,1,5,8,1
+0,14,113,1,[other],8,1
+0,15,113,1,5,7,1
+0,16,105,1,5,7,1
+0,17,108,1,8,7,1
+0,18,108,1,5,7,1
+0,19,124,1,5,7,1
+0,20,124,1,5,7,1
+0,21,105,8,5,7,1
+0,22,105,1,5,7,1
+0,23,110,1,5,7,1
+0,24,110,1,5,8,1
+0,25,110,1,7,8,1
+0,26,110,[other],7,7,1
+0,27,110,5,7,7,1
+0,28,103,5,7,0,1
+0,29,102,1,5,0,1
+0,30,103,1,7,7,1
+0,31,102,5,5,0,1
+0,32,103,1,7,0,1
+0,33,102,[mask],7,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
index 7ae8e23a..12e2155a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,121,6,0,8,0
-1,9,121,1,0,8,0
-1,10,121,1,0,8,0
-1,11,121,1,0,8,0
-1,12,121,1,0,8,0
-1,13,126,1,0,8,0
-1,14,127,1,0,8,0
-1,15,127,0,0,8,0
-1,16,127,0,9,8,0
-1,17,127,0,7,7,0
-1,18,105,8,9,7,1
-1,19,127,0,7,2,1
-1,20,113,4,7,8,1
-1,21,113,1,9,8,1
-1,22,113,1,7,8,1
-1,23,113,1,9,8,1
-1,24,113,1,0,8,1
-1,25,113,1,0,8,1
-1,26,113,1,0,8,1
-1,27,113,1,0,8,1
-1,28,113,1,0,7,1
-1,29,108,1,8,7,1
-1,30,108,1,8,7,1
-1,31,105,6,8,7,1
-1,32,113,1,8,7,1
-1,33,113,6,7,7,1
-1,34,113,6,7,7,1
-1,35,113,6,7,2,1
-1,36,113,1,7,8,1
-1,37,113,1,0,8,1
+1,4,121,1,0,7,0
+1,5,108,3,0,8,1
+1,6,103,1,0,7,0
+1,7,103,8,0,7,1
+1,8,108,1,0,2,1
+1,9,113,1,0,7,1
+1,10,108,8,7,7,1
+1,11,124,8,7,7,1
+1,12,105,8,7,7,1
+1,13,113,8,5,7,1
+1,14,113,1,[other],8,1
+1,15,113,1,[other],8,1
+1,16,113,1,[other],7,1
+1,17,108,1,5,7,1
+1,18,124,1,[other],7,1
+1,19,108,1,5,7,1
+1,20,124,1,5,7,1
+1,21,124,1,5,7,1
+1,22,105,8,5,7,1
+1,23,105,1,5,7,1
+1,24,124,8,5,7,1
+1,25,110,8,5,7,1
+1,26,110,1,5,8,1
+1,27,110,1,7,8,1
+1,28,110,6,7,7,1
+1,29,110,1,7,0,1
+1,30,113,1,7,0,1
+1,31,113,1,7,0,1
+1,32,113,1,7,0,1
+1,33,113,[mask],5,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
index 38fb6ff6..5cf7bea0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,121,1,8,0,0
-2,9,105,0,8,7,0
-2,10,121,1,8,2,0
-2,11,113,4,7,8,0
-2,12,121,1,9,8,0
-2,13,127,1,0,8,0
-2,14,127,4,0,8,1
-2,15,127,1,0,8,0
-2,16,127,4,7,8,1
-2,17,127,1,0,8,1
-2,18,103,1,0,8,1
-2,19,113,1,0,8,1
-2,20,103,1,9,7,1
-2,21,105,1,7,7,1
-2,22,113,1,9,7,1
-2,23,105,1,9,7,1
-2,24,113,1,7,7,1
-2,25,113,6,9,7,1
-2,26,113,1,7,7,1
-2,27,113,6,5,7,1
-2,28,113,1,8,8,1
-2,29,113,1,0,8,1
-2,30,113,1,0,8,1
-2,31,113,1,0,8,1
-2,32,113,1,0,8,1
-2,33,113,1,0,8,1
-2,34,113,1,0,7,1
-2,35,108,1,8,7,1
-2,36,105,1,8,7,1
-2,37,105,1,9,7,1
+2,3,128,1,8,6,0
+2,4,105,0,8,7,2
+2,5,105,1,8,2,1
+2,6,113,1,8,7,2
+2,7,105,1,9,7,1
+2,8,105,1,7,7,1
+2,9,113,1,7,7,1
+2,10,113,1,5,7,1
+2,11,113,1,[other],8,1
+2,12,113,1,[other],7,1
+2,13,113,8,5,7,1
+2,14,113,1,[other],8,1
+2,15,113,1,5,7,1
+2,16,113,1,5,7,1
+2,17,105,1,5,7,1
+2,18,108,1,5,7,1
+2,19,108,1,5,7,1
+2,20,108,1,5,7,1
+2,21,124,1,5,7,1
+2,22,105,8,5,7,1
+2,23,110,1,5,7,1
+2,24,110,8,5,8,1
+2,25,110,1,5,8,1
+2,26,110,[other],5,8,1
+2,27,127,1,7,8,1
+2,28,103,5,7,0,1
+2,29,103,1,5,7,1
+2,30,103,1,5,0,1
+2,31,103,[other],4,0,1
+2,32,127,1,4,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
index 932a6a31..4e1d63dd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,127,4,0,2,0
-3,9,127,4,0,8,1
-3,10,127,4,0,8,1
-3,11,127,1,0,2,1
-3,12,113,4,9,8,1
-3,13,121,1,1,2,1
-3,14,113,0,9,8,1
-3,15,113,1,1,8,1
-3,16,113,1,9,7,1
-3,17,105,1,9,7,1
-3,18,105,1,8,7,1
-3,19,105,1,9,7,1
-3,20,113,1,8,7,1
-3,21,105,8,9,7,1
-3,22,113,1,8,8,1
-3,23,113,1,9,8,1
-3,24,113,1,7,8,1
-3,25,113,1,9,7,1
-3,26,113,1,8,7,1
-3,27,113,1,8,7,1
-3,28,113,1,8,7,1
-3,29,113,1,8,7,1
-3,30,113,1,5,7,1
-3,31,113,1,5,8,1
-3,32,113,1,0,8,1
-3,33,113,1,0,8,1
-3,34,113,1,0,8,1
-3,35,113,1,0,8,1
-3,36,113,1,0,7,1
-3,37,108,1,8,7,1
-4,8,127,4,0,8,0
-4,9,127,4,0,8,1
-4,10,127,1,0,2,1
-4,11,127,4,7,8,1
-4,12,127,1,1,8,1
-4,13,127,1,0,8,1
-4,14,113,1,0,8,1
-4,15,113,1,9,8,1
-4,16,113,1,9,7,1
-4,17,105,1,9,7,1
-4,18,105,1,9,7,1
-4,19,105,1,9,7,1
-4,20,105,1,9,7,1
-4,21,113,1,9,7,1
-4,22,113,1,9,7,1
-4,23,113,1,9,7,1
-4,24,113,1,9,8,1
-4,25,113,1,9,8,1
-4,26,113,1,9,8,1
-4,27,113,1,9,8,1
-4,28,113,1,9,7,1
-4,29,113,1,8,7,1
-4,30,105,1,8,7,1
-4,31,113,1,8,7,1
-4,32,113,1,8,7,1
-4,33,113,1,8,7,1
-4,34,113,1,9,7,1
-4,35,113,1,7,8,1
-4,36,113,1,7,8,1
-4,37,113,1,7,8,1
+3,3,127,5,0,2,0
+3,4,127,4,0,8,0
+3,5,127,4,0,8,0
+3,6,127,4,0,8,0
+3,7,127,4,0,8,0
+3,8,127,4,0,8,1
+3,9,108,1,0,8,1
+3,10,103,1,7,7,1
+3,11,103,8,9,7,1
+3,12,103,1,7,7,1
+3,13,113,8,7,7,1
+3,14,113,8,5,7,1
+3,15,113,1,5,7,1
+3,16,113,1,5,7,1
+3,17,113,1,5,7,1
+3,18,113,1,[other],8,1
+3,19,113,1,[other],7,1
+3,20,108,8,5,7,1
+3,21,108,1,2,7,1
+3,22,108,1,5,7,1
+3,23,108,1,5,7,1
+3,24,124,1,5,7,1
+3,25,124,1,5,7,1
+3,26,124,8,5,7,1
+3,27,124,1,5,7,1
+3,28,124,8,5,7,1
+3,29,110,8,5,7,1
+3,30,110,1,5,8,1
+3,31,110,1,7,8,1
+3,32,110,[other],7,7,1
+4,3,127,0,9,[unknown],1
+4,4,127,0,1,8,1
+4,5,102,0,1,8,1
+4,6,127,1,4,8,0
+4,7,127,0,1,7,1
+4,8,127,0,4,0,1
+4,9,127,1,1,7,1
+4,10,124,1,9,0,1
+4,11,113,1,7,7,1
+4,12,113,8,5,0,1
+4,13,113,1,5,8,1
+4,14,113,1,7,8,1
+4,15,113,1,5,7,1
+4,16,113,1,5,7,1
+4,17,113,1,5,7,1
+4,18,113,1,5,7,1
+4,19,108,1,5,7,1
+4,20,108,1,5,7,1
+4,21,103,1,5,7,1
+4,22,103,1,5,7,1
+4,23,110,1,5,7,1
+4,24,110,1,5,8,1
+4,25,110,1,5,8,1
+4,26,110,1,5,0,1
+4,27,110,0,5,7,1
+4,28,110,[other],5,0,1
+4,29,127,0,4,7,1
+4,30,103,0,5,0,1
+4,31,103,[other],4,0,1
+4,32,127,1,4,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
index 0c148033..4f5b8222 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,121,1,5,4,1
-5,9,127,1,0,8,1
-5,10,127,1,0,2,1
-5,11,127,1,7,2,1
-5,12,127,1,1,0,1
-5,13,105,1,9,2,1
-5,14,113,1,7,0,1
-5,15,113,1,9,7,1
-5,16,105,8,9,7,1
-5,17,105,1,8,2,1
-5,18,113,0,7,8,1
-5,19,113,1,9,8,1
-5,20,113,1,9,8,1
-5,21,113,1,9,8,1
-5,22,113,1,0,8,1
-5,23,113,1,0,8,1
-5,24,113,1,0,8,1
-5,25,113,1,0,7,1
-5,26,108,1,8,7,1
-5,27,105,1,8,7,1
-5,28,105,1,8,7,1
-5,29,105,1,9,7,1
-5,30,113,1,7,7,1
-5,31,113,6,7,7,1
-5,32,113,6,7,2,1
-5,33,113,1,7,8,1
-5,34,113,1,7,8,1
-5,35,113,1,9,8,1
-5,36,113,1,0,8,1
-5,37,113,1,0,8,1
+5,3,128,1,5,4,1
+5,4,128,1,5,6,1
+5,5,105,1,5,6,1
+5,6,128,1,5,0,1
+5,7,105,1,5,7,1
+5,8,105,1,5,0,1
+5,9,113,1,7,7,1
+5,10,105,1,9,7,1
+5,11,103,1,9,7,1
+5,12,108,1,9,7,1
+5,13,113,1,5,7,1
+5,14,113,1,5,8,1
+5,15,113,1,5,8,1
+5,16,113,1,5,8,1
+5,17,113,1,5,7,1
+5,18,105,1,5,7,1
+5,19,105,1,5,7,1
+5,20,103,1,5,7,1
+5,21,103,1,5,7,1
+5,22,103,1,5,7,1
+5,23,103,1,5,7,1
+5,24,108,1,5,7,1
+5,25,124,8,5,7,1
+5,26,110,8,5,7,1
+5,27,110,1,5,8,1
+5,28,110,2,7,8,1
+5,29,110,2,5,0,1
+5,30,110,1,5,7,1
+5,31,124,[other],5,0,1
+5,32,127,1,8,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
index 1b1ae93e..929f2c8c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,121,4,0,8,0
-6,9,113,1,0,8,0
-6,10,113,1,0,8,1
-6,11,113,1,0,8,1
-6,12,113,1,0,8,1
-6,13,113,1,0,8,1
-6,14,113,1,0,8,1
-6,15,113,1,9,7,1
-6,16,105,8,9,7,1
-6,17,105,1,8,7,1
-6,18,105,1,9,7,1
-6,19,105,1,9,7,1
-6,20,113,1,7,7,1
-6,21,113,6,9,7,1
-6,22,113,1,3,2,1
-6,23,113,6,7,8,1
-6,24,113,1,9,8,1
-6,25,113,1,0,8,1
-6,26,113,1,0,8,1
-6,27,113,1,0,8,1
-6,28,113,1,0,8,1
-6,29,113,1,0,8,1
-6,30,113,1,0,7,1
-6,31,108,1,8,7,1
-6,32,105,1,8,7,1
-6,33,105,1,9,7,1
-6,34,105,1,7,7,1
-6,35,113,6,7,7,1
-6,36,113,6,7,7,1
-6,37,113,6,7,2,1
-7,8,105,4,9,8,0
-7,9,121,0,8,8,0
-7,10,113,0,8,8,0
-7,11,121,4,8,8,0
-7,12,127,0,0,8,0
-7,13,121,4,8,8,0
-7,14,127,4,0,8,0
-7,15,127,1,0,8,0
-7,16,127,4,0,8,1
-7,17,127,1,0,8,0
-7,18,127,1,0,8,1
-7,19,127,1,7,8,1
-7,20,113,1,7,8,1
-7,21,113,1,9,7,1
-7,22,103,1,7,7,1
-7,23,113,8,7,7,1
-7,24,113,1,9,7,1
-7,25,113,1,7,7,1
-7,26,113,6,5,7,1
-7,27,113,1,5,8,1
-7,28,113,1,0,8,1
-7,29,113,1,0,8,1
-7,30,113,1,0,8,1
-7,31,113,1,0,8,1
-7,32,113,1,0,7,1
-7,33,108,1,8,7,1
-7,34,108,1,8,7,1
-7,35,108,1,8,7,1
-7,36,122,6,7,7,1
-7,37,113,6,7,7,1
+6,4,109,4,8,8,1
+6,5,113,1,0,7,0
+6,6,113,1,7,7,1
+6,7,113,1,7,7,1
+6,8,113,1,7,7,1
+6,9,102,8,7,7,1
+6,10,102,5,7,7,1
+6,11,102,1,7,7,1
+6,12,113,1,7,7,1
+6,13,113,8,5,0,1
+6,14,102,1,5,2,1
+6,15,113,1,7,8,1
+6,16,113,1,1,0,1
+6,17,113,1,5,7,1
+6,18,105,1,5,7,1
+6,19,105,1,5,7,1
+6,20,108,1,7,7,1
+6,21,108,8,7,7,1
+6,22,105,5,5,7,1
+6,23,113,1,5,7,1
+6,24,113,1,5,8,1
+6,25,113,1,5,8,1
+6,26,113,1,5,8,1
+6,27,113,1,5,7,1
+6,28,108,1,5,7,1
+6,29,108,1,5,7,1
+6,30,103,1,5,7,1
+6,31,103,1,5,7,1
+6,32,103,1,5,7,1
+6,33,103,8,5,7,1
+7,3,105,0,9,8,0
+7,4,127,0,1,7,0
+7,5,105,0,9,8,1
+7,6,127,0,1,7,0
+7,7,105,0,9,8,1
+7,8,113,0,1,0,0
+7,9,113,1,9,8,1
+7,10,113,1,1,8,1
+7,11,113,1,9,8,1
+7,12,113,1,1,7,1
+7,13,113,1,9,7,1
+7,14,113,1,9,7,1
+7,15,105,1,5,7,1
+7,16,113,1,9,8,1
+7,17,113,1,5,7,1
+7,18,105,1,5,7,1
+7,19,105,1,5,7,1
+7,20,113,1,5,7,1
+7,21,113,1,5,7,1
+7,22,113,1,5,7,1
+7,23,113,1,5,7,1
+7,24,113,1,5,7,1
+7,25,113,1,5,7,1
+7,26,113,1,5,7,1
+7,27,113,1,5,7,1
+7,28,113,1,5,7,1
+7,29,113,1,5,7,1
+7,30,113,1,5,7,1
+7,31,113,1,5,7,1
+7,32,113,1,5,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
index 9dc6a76d..83cccde4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,113,1,0,8,0
-8,9,113,1,0,8,1
-8,10,113,1,0,2,1
-8,11,113,1,8,2,1
-8,12,113,1,0,8,1
-8,13,113,1,0,8,1
-8,14,113,1,0,2,1
-8,15,113,1,0,2,1
-8,16,113,1,8,7,1
-8,17,105,8,9,7,1
-8,18,105,1,8,2,1
-8,19,113,1,9,7,1
-8,20,105,1,9,7,1
-8,21,113,1,7,7,1
-8,22,113,1,9,7,1
-8,23,113,1,7,8,1
-8,24,113,1,9,8,1
-8,25,113,1,0,8,1
-8,26,113,1,9,8,1
-8,27,113,1,0,7,1
-8,28,113,1,8,7,1
-8,29,105,1,8,7,1
-8,30,113,1,8,7,1
-8,31,105,1,9,7,1
-8,32,113,1,8,7,1
-8,33,113,6,9,7,1
-8,34,113,1,8,2,1
-8,35,113,1,7,8,1
-8,36,113,1,7,8,1
-8,37,113,1,7,8,1
+8,4,109,1,0,7,0
+8,5,127,1,8,7,1
+8,6,124,1,7,0,1
+8,7,127,1,7,7,0
+8,8,127,8,9,0,1
+8,9,127,1,1,0,1
+8,10,124,1,7,0,1
+8,11,113,1,7,0,1
+8,12,113,1,7,7,1
+8,13,113,8,5,7,1
+8,14,113,1,5,8,1
+8,15,113,1,5,8,1
+8,16,113,1,5,7,1
+8,17,113,1,5,7,1
+8,18,108,1,5,7,1
+8,19,108,1,5,7,1
+8,20,108,1,5,7,1
+8,21,124,1,5,7,1
+8,22,103,8,5,7,1
+8,23,103,1,5,7,1
+8,24,110,8,5,7,1
+8,25,110,1,5,8,1
+8,26,110,2,5,8,1
+8,27,110,1,5,8,1
+8,28,110,[other],7,0,1
+8,29,127,1,7,7,1
+8,30,108,5,7,0,1
+8,31,127,5,5,0,1
+8,32,102,1,4,0,1
+8,33,103,[other],4,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
index 06013440..eebbcfd3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,121,1,8,8,1
-9,9,113,1,0,8,1
-9,10,113,1,8,8,1
-9,11,113,1,0,7,1
-9,12,105,1,8,7,1
-9,13,105,1,8,7,1
-9,14,105,1,7,7,1
-9,15,113,1,7,7,1
-9,16,113,1,7,7,1
-9,17,113,6,7,7,1
-9,18,113,6,5,2,1
-9,19,113,1,0,8,1
-9,20,113,1,0,8,1
-9,21,113,1,0,8,1
-9,22,113,1,0,8,1
-9,23,113,1,0,8,1
-9,24,113,1,0,8,1
-9,25,113,1,0,8,1
-9,26,113,1,0,7,1
-9,27,105,1,8,7,1
-9,28,105,1,8,7,1
-9,29,105,1,9,7,1
-9,30,105,1,7,7,1
-9,31,113,6,7,7,1
-9,32,113,6,7,2,1
-9,33,113,1,7,8,1
-9,34,113,1,7,8,1
-9,35,113,1,9,8,1
-9,36,113,1,0,8,1
-9,37,113,1,0,8,1
+9,3,108,1,8,7,1
+9,4,108,1,[other],7,1
+9,5,108,8,5,7,1
+9,6,108,1,[other],7,1
+9,7,108,8,[other],7,1
+9,8,108,8,[other],7,1
+9,9,124,8,[other],7,1
+9,10,113,8,[other],2,1
+9,11,113,1,[other],8,1
+9,12,113,1,[other],8,1
+9,13,113,1,[other],8,1
+9,14,113,1,[other],0,1
+9,15,113,1,9,2,1
+9,16,113,1,5,7,1
+9,17,105,1,5,7,1
+9,18,105,1,8,7,1
+9,19,103,1,5,7,1
+9,20,105,1,5,7,1
+9,21,113,1,5,7,1
+9,22,113,1,7,7,1
+9,23,113,8,5,7,1
+9,24,113,1,5,8,1
+9,25,113,1,5,8,1
+9,26,113,1,5,7,1
+9,27,108,1,5,7,1
+9,28,108,1,5,7,1
+9,29,108,1,5,7,1
+9,30,108,1,5,7,1
+9,31,124,1,5,7,1
+9,32,103,8,5,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index fd1bb4cf..a90c5af9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,104,7,[mask],[unknown],1
-0,9,104,7,[mask],[unknown],0
-0,10,104,7,1,[unknown],0
-0,11,101,7,1,[unknown],0
-0,12,106,7,1,[unknown],[other]
-0,13,112,9,1,[unknown],0
-0,14,129,9,1,5,[unknown]
-0,15,112,2,1,1,[other]
-0,16,112,7,1,[unknown],2
-0,17,129,2,1,[unknown],2
-0,18,129,7,1,5,2
-0,19,112,7,1,1,2
-0,20,129,7,1,[unknown],2
-0,21,129,7,1,1,[other]
-0,22,112,7,1,5,[mask]
-0,23,118,7,[mask],[unknown],[mask]
-0,24,129,9,1,5,[other]
-0,25,112,9,1,5,[mask]
-0,26,[other],7,[mask],[unknown],[mask]
-0,27,129,9,1,5,[other]
-0,28,112,9,1,5,[other]
-0,29,129,7,1,[unknown],[mask]
-0,30,104,0,1,[unknown],[other]
-0,31,112,7,1,[unknown],[other]
-0,32,104,7,1,[unknown],2
-0,33,129,7,1,[unknown],[other]
-0,34,112,9,1,5,[other]
-0,35,112,7,1,[unknown],2
-0,36,108,7,1,[unknown],2
-0,37,129,7,1,5,2
+0,4,104,7,[mask],[unknown],1
+0,5,104,7,9,2,0
+0,6,104,0,9,3,0
+0,7,103,0,[mask],8,0
+0,8,104,7,1,5,0
+0,9,103,5,1,[unknown],0
+0,10,112,0,1,5,0
+0,11,112,7,1,5,0
+0,12,112,5,1,5,0
+0,13,103,0,1,[unknown],0
+0,14,112,7,1,5,0
+0,15,112,0,1,5,0
+0,16,112,7,1,5,0
+0,17,112,7,1,5,0
+0,18,112,7,[mask],5,1
+0,19,119,7,[mask],5,1
+0,20,119,7,[mask],8,0
+0,21,104,0,1,5,1
+0,22,104,7,[mask],8,1
+0,23,104,7,[mask],[unknown],1
+0,24,104,7,[mask],8,1
+0,25,104,7,[mask],8,1
+0,26,108,9,[mask],3,2
+0,27,104,9,1,5,2
+0,28,112,9,[other],1,0
+0,29,111,8,1,4,2
+0,30,112,9,[other],1,2
+0,31,104,9,[other],5,2
+0,32,104,9,[other],1,[mask]
+0,33,104,9,[other],8,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index d23a5aeb..61720a4a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,104,7,[mask],1,1
-1,9,104,7,[mask],[unknown],[mask]
-1,10,[other],9,1,8,[mask]
-1,11,[other],9,1,5,[other]
-1,12,104,9,1,1,[other]
-1,13,112,9,1,[unknown],[other]
-1,14,112,7,1,[unknown],[other]
-1,15,112,2,1,5,[other]
-1,16,112,2,1,[unknown],[mask]
-1,17,129,2,1,[unknown],[mask]
-1,18,129,7,1,5,2
-1,19,129,2,1,1,2
-1,20,129,7,1,1,2
-1,21,129,7,1,1,2
-1,22,129,7,1,1,2
-1,23,129,7,1,1,2
-1,24,129,2,1,5,[mask]
-1,25,129,2,1,5,[mask]
-1,26,129,2,1,5,[mask]
-1,27,129,2,1,5,[mask]
-1,28,129,2,1,5,[mask]
-1,29,112,2,1,5,[mask]
-1,30,107,2,1,5,[mask]
-1,31,107,2,1,5,1
-1,32,107,2,[mask],[unknown],1
-1,33,107,2,[mask],[unknown],1
-1,34,107,2,[mask],[unknown],1
-1,35,107,7,[mask],[unknown],1
-1,36,107,7,[mask],[unknown],1
-1,37,107,[mask],[mask],[unknown],1
+1,4,104,7,9,1,1
+1,5,104,7,9,8,1
+1,6,104,0,9,8,1
+1,7,104,7,[mask],8,1
+1,8,104,7,9,8,1
+1,9,104,7,9,[unknown],0
+1,10,118,7,9,[unknown],1
+1,11,129,7,9,[unknown],[other]
+1,12,112,7,9,[unknown],0
+1,13,108,7,9,[unknown],2
+1,14,104,9,9,5,2
+1,15,129,9,8,8,[other]
+1,16,112,9,2,5,0
+1,17,104,9,1,5,[other]
+1,18,104,0,1,8,0
+1,19,112,0,1,8,[other]
+1,20,112,0,[mask],5,0
+1,21,112,0,1,5,[mask]
+1,22,104,7,[mask],5,0
+1,23,129,9,1,8,1
+1,24,104,7,1,8,0
+1,25,112,5,1,5,0
+1,26,112,0,1,5,0
+1,27,112,7,1,5,0
+1,28,112,0,1,5,1
+1,29,112,7,[mask],[unknown],0
+1,30,108,7,1,5,1
+1,31,119,7,[mask],[unknown],0
+1,32,119,0,1,5,1
+1,33,104,7,[mask],5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index cc5b8a64..32aad96e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,112,2,[mask],5,[mask]
-2,9,104,7,[mask],[unknown],[mask]
-2,10,104,7,1,8,[mask]
-2,11,112,7,1,[unknown],[other]
-2,12,112,7,1,[unknown],[mask]
-2,13,107,7,[mask],[unknown],[mask]
-2,14,107,7,[mask],[unknown],[mask]
-2,15,[other],7,[mask],[unknown],2
-2,16,129,9,[other],1,[other]
-2,17,112,9,[other],1,[mask]
-2,18,104,1,[other],5,[mask]
-2,19,104,2,[other],[unknown],[other]
-2,20,110,9,[mask],[unknown],[other]
-2,21,110,2,[mask],[unknown],[mask]
-2,22,[other],2,[mask],[unknown],[other]
-2,23,110,2,[mask],5,[mask]
-2,24,[other],2,[other],7,[mask]
-2,25,104,2,[other],8,[mask]
-2,26,107,2,[other],1,[mask]
-2,27,107,2,[other],1,[mask]
-2,28,107,2,[other],8,[mask]
-2,29,107,2,1,5,[mask]
-2,30,107,2,1,5,[unknown]
-2,31,129,2,1,5,[mask]
-2,32,129,9,1,5,[other]
-2,33,107,2,1,5,[mask]
-2,34,107,2,1,5,[mask]
-2,35,107,2,1,[unknown],1
-2,36,107,2,[mask],[unknown],1
-2,37,107,7,[mask],[unknown],1
+2,3,112,2,[mask],5,0
+2,4,129,7,1,5,1
+2,5,104,7,[mask],8,0
+2,6,129,7,1,5,1
+2,7,104,7,[mask],8,0
+2,8,112,7,1,5,1
+2,9,107,7,[mask],[unknown],1
+2,10,107,7,[mask],[unknown],1
+2,11,104,7,[mask],[unknown],1
+2,12,108,7,[mask],[unknown],1
+2,13,104,9,1,5,2
+2,14,108,7,[mask],[unknown],0
+2,15,104,9,1,5,2
+2,16,104,7,[mask],[unknown],2
+2,17,104,9,[mask],1,2
+2,18,112,9,[other],1,[mask]
+2,19,104,9,[other],5,[mask]
+2,20,112,9,8,8,[mask]
+2,21,112,9,6,5,[mask]
+2,22,112,9,6,8,[mask]
+2,23,112,9,6,5,[mask]
+2,24,104,9,6,8,[other]
+2,25,112,0,8,8,[other]
+2,26,112,0,[mask],8,0
+2,27,104,4,1,8,[mask]
+2,28,112,0,1,8,[other]
+2,29,112,0,[mask],8,0
+2,30,104,1,1,5,2
+2,31,112,0,1,8,0
+2,32,112,1,1,5,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index 2a4d06f3..f0a3eaa7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,104,[mask],[mask],[unknown],[mask]
-3,9,104,0,[mask],8,[other]
-3,10,104,[mask],[mask],6,[mask]
-3,11,104,0,[mask],8,[mask]
-3,12,104,0,[mask],8,[mask]
-3,13,104,1,[mask],8,[mask]
-3,14,[other],2,[mask],[unknown],[mask]
-3,15,129,1,[other],6,[mask]
-3,16,107,2,[other],1,[mask]
-3,17,107,2,[other],6,[mask]
-3,18,129,9,[other],6,[mask]
-3,19,129,9,[other],5,[mask]
-3,20,129,9,[other],5,[mask]
-3,21,129,9,[other],5,[mask]
-3,22,129,9,[other],5,[other]
-3,23,110,9,[other],5,[mask]
-3,24,107,2,[other],8,[other]
-3,25,107,2,1,5,[other]
-3,26,110,9,1,[unknown],[other]
-3,27,[other],9,1,[unknown],[other]
-3,28,[other],2,1,[unknown],[other]
-3,29,[other],2,1,[unknown],[other]
-3,30,107,2,1,[unknown],[unknown]
-3,31,129,2,1,[unknown],[mask]
-3,32,129,2,1,[unknown],2
-3,33,129,2,1,5,2
-3,34,129,2,1,5,2
-3,35,129,2,1,5,2
-3,36,129,2,1,5,2
-3,37,129,2,1,5,2
-4,8,104,9,1,[unknown],[mask]
-4,9,[other],7,1,[unknown],[other]
-4,10,104,9,1,[unknown],[other]
-4,11,[other],9,1,[unknown],[other]
-4,12,[other],9,1,[unknown],[other]
-4,13,[other],9,1,[unknown],[other]
-4,14,129,2,1,5,[unknown]
-4,15,129,2,1,1,2
-4,16,129,2,1,1,2
-4,17,129,7,1,5,2
-4,18,129,2,1,1,2
-4,19,129,7,1,5,2
-4,20,129,7,1,1,2
-4,21,129,7,1,1,[mask]
-4,22,129,7,1,5,[mask]
-4,23,129,2,1,5,[mask]
-4,24,112,2,1,5,[mask]
-4,25,107,2,[mask],[unknown],[mask]
-4,26,129,2,[mask],5,[mask]
-4,27,129,9,1,5,[mask]
-4,28,129,9,1,5,[mask]
-4,29,129,9,1,5,[mask]
-4,30,129,9,1,5,[mask]
-4,31,129,9,1,5,[mask]
-4,32,129,9,1,5,[mask]
-4,33,129,9,1,5,[other]
-4,34,112,9,1,5,[mask]
-4,35,107,7,[mask],[unknown],[mask]
-4,36,107,7,[mask],[unknown],[mask]
-4,37,104,7,[mask],[unknown],[mask]
+3,3,104,0,[mask],8,1
+3,4,104,0,[mask],8,0
+3,5,104,0,[mask],8,1
+3,6,104,0,[mask],8,0
+3,7,104,0,[mask],8,1
+3,8,104,0,[mask],6,0
+3,9,112,0,[mask],5,0
+3,10,112,1,1,5,2
+3,11,108,2,1,5,0
+3,12,129,9,1,5,2
+3,13,112,9,1,5,0
+3,14,129,9,1,5,0
+3,15,112,9,1,5,0
+3,16,129,0,1,5,1
+3,17,112,7,[mask],5,1
+3,18,119,7,[mask],[unknown],1
+3,19,104,7,[mask],5,0
+3,20,104,7,1,5,1
+3,21,104,7,[mask],8,0
+3,22,108,7,1,5,1
+3,23,104,7,[mask],8,0
+3,24,108,7,[mask],5,1
+3,25,104,9,[mask],8,2
+3,26,112,9,1,5,0
+3,27,112,9,1,5,2
+3,28,103,7,1,5,0
+3,29,112,9,1,5,0
+3,30,104,7,1,5,0
+3,31,112,0,1,8,0
+3,32,112,0,[mask],5,0
+4,3,108,9,1,5,1
+4,4,104,7,1,4,0
+4,5,108,0,1,4,0
+4,6,108,0,1,5,0
+4,7,112,7,1,5,0
+4,8,108,7,1,5,0
+4,9,129,0,1,5,2
+4,10,112,7,1,5,2
+4,11,112,7,[mask],5,0
+4,12,112,7,1,5,1
+4,13,107,7,[mask],[unknown],1
+4,14,107,7,[mask],[unknown],1
+4,15,107,7,[mask],[unknown],1
+4,16,107,7,[mask],[unknown],1
+4,17,104,7,[mask],[unknown],1
+4,18,104,7,[mask],[unknown],1
+4,19,104,7,[mask],[unknown],1
+4,20,118,7,[mask],[unknown],1
+4,21,104,9,[mask],[unknown],2
+4,22,104,9,[mask],1,2
+4,23,104,9,[mask],1,2
+4,24,104,9,[other],1,2
+4,25,104,9,[other],1,[mask]
+4,26,104,9,[other],[unknown],[mask]
+4,27,109,9,[other],[unknown],[mask]
+4,28,104,9,6,5,[mask]
+4,29,112,9,[other],1,[mask]
+4,30,118,2,6,1,[mask]
+4,31,110,1,[other],8,[other]
+4,32,112,[other],6,[unknown],[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index 529df285..5a908cad 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,102,7,[mask],9,1
-5,9,107,[mask],[mask],[unknown],1
-5,10,107,[mask],[mask],[unknown],1
-5,11,107,[mask],[mask],[unknown],1
-5,12,104,[mask],[mask],[unknown],1
-5,13,104,[mask],[mask],[unknown],1
-5,14,104,[mask],[mask],[unknown],1
-5,15,104,[mask],[mask],3,1
-5,16,104,[mask],[mask],3,[other]
-5,17,112,[unknown],[mask],3,[other]
-5,18,111,[mask],[mask],3,2
-5,19,104,9,[mask],3,2
-5,20,104,5,[mask],3,2
-5,21,104,5,[mask],3,2
-5,22,104,5,[mask],3,2
-5,23,104,[unknown],[mask],3,2
-5,24,104,9,[mask],3,[unknown]
-5,25,104,9,[mask],3,[mask]
-5,26,112,[unknown],[mask],3,[mask]
-5,27,112,[unknown],[mask],3,[mask]
-5,28,112,[mask],[mask],3,[mask]
-5,29,104,[mask],[mask],7,[mask]
-5,30,107,9,[mask],3,[mask]
-5,31,107,2,[mask],3,[mask]
-5,32,107,2,[mask],6,[mask]
-5,33,104,2,[mask],6,[mask]
-5,34,104,9,[mask],6,[mask]
-5,35,[other],9,[mask],6,[mask]
-5,36,[other],2,[other],6,[mask]
-5,37,107,9,[other],6,[mask]
+5,3,102,7,9,9,1
+5,4,102,7,9,9,1
+5,5,105,7,9,9,1
+5,6,105,7,9,9,1
+5,7,105,7,9,9,1
+5,8,105,7,9,9,1
+5,9,111,[mask],9,9,1
+5,10,104,7,9,0,1
+5,11,103,7,9,8,0
+5,12,105,7,9,9,1
+5,13,111,7,9,9,0
+5,14,103,7,9,3,1
+5,15,111,[mask],9,0,0
+5,16,104,7,9,8,1
+5,17,104,0,9,8,0
+5,18,103,0,9,8,1
+5,19,104,7,9,8,0
+5,20,103,0,9,8,0
+5,21,104,7,9,5,1
+5,22,104,7,9,8,[other]
+5,23,101,0,9,8,0
+5,24,105,7,9,[unknown],0
+5,25,103,7,9,8,1
+5,26,104,7,9,8,0
+5,27,101,0,8,8,0
+5,28,112,7,9,9,1
+5,29,104,7,9,8,[other]
+5,30,129,0,1,8,0
+5,31,104,7,[mask],5,1
+5,32,104,7,1,8,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index 819defac..58547214 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,129,9,1,5,[other]
-6,9,129,9,1,5,[mask]
-6,10,129,9,1,5,[other]
-6,11,129,9,1,5,[other]
-6,12,129,9,1,5,[other]
-6,13,129,2,1,5,[mask]
-6,14,129,2,1,8,[other]
-6,15,112,7,1,[unknown],[unknown]
-6,16,107,7,1,[unknown],2
-6,17,108,7,[mask],[unknown],2
-6,18,104,7,1,5,2
-6,19,108,7,1,[unknown],2
-6,20,104,7,1,5,2
-6,21,112,9,[mask],1,[mask]
-6,22,129,9,[other],5,[mask]
-6,23,104,9,[other],5,[mask]
-6,24,112,9,8,1,[mask]
-6,25,129,9,[other],5,[mask]
-6,26,104,9,[other],1,[mask]
-6,27,104,9,[other],1,[other]
-6,28,112,7,[other],1,[mask]
-6,29,[other],1,[other],8,[other]
-6,30,[other],2,[mask],[unknown],[other]
-6,31,[other],2,1,[unknown],[other]
-6,32,110,2,1,5,[other]
-6,33,[other],2,1,[unknown],[other]
-6,34,[other],2,1,5,[mask]
-6,35,129,2,1,8,[unknown]
-6,36,129,2,1,5,[unknown]
-6,37,129,2,1,5,2
-7,8,129,2,1,5,[mask]
-7,9,129,2,1,5,[mask]
-7,10,129,2,1,5,[mask]
-7,11,129,2,1,5,[mask]
-7,12,129,2,1,5,1
-7,13,107,2,1,5,[mask]
-7,14,107,2,1,[unknown],1
-7,15,107,2,[mask],[unknown],1
-7,16,107,2,[mask],[unknown],1
-7,17,107,7,[mask],[unknown],1
-7,18,107,7,[mask],[unknown],1
-7,19,107,7,[mask],[unknown],1
-7,20,107,7,[mask],[unknown],1
-7,21,107,[mask],[mask],[unknown],1
-7,22,104,[mask],[mask],[unknown],2
-7,23,104,9,[mask],1,[other]
-7,24,104,9,[mask],3,[mask]
-7,25,104,[unknown],[mask],3,[other]
-7,26,104,[mask],[mask],3,[mask]
-7,27,104,9,[mask],[unknown],[other]
-7,28,112,9,[mask],[unknown],[mask]
-7,29,104,2,[mask],[unknown],[mask]
-7,30,107,2,[other],[unknown],[mask]
-7,31,107,2,3,1,[mask]
-7,32,107,2,[other],[unknown],[mask]
-7,33,107,2,[other],[unknown],[mask]
-7,34,107,2,[other],[unknown],[mask]
-7,35,107,2,[other],8,[mask]
-7,36,[other],2,[other],8,[mask]
-7,37,[other],9,[other],5,[other]
+6,4,129,9,1,5,2
+6,5,129,9,1,5,0
+6,6,129,9,1,5,1
+6,7,112,7,1,5,0
+6,8,112,7,1,5,1
+6,9,112,7,[mask],[unknown],0
+6,10,129,7,1,5,1
+6,11,104,7,[mask],5,0
+6,12,112,7,[mask],5,1
+6,13,119,7,[mask],[unknown],1
+6,14,104,7,[mask],[unknown],1
+6,15,104,7,[mask],[unknown],1
+6,16,104,7,[mask],[unknown],1
+6,17,104,7,1,8,2
+6,18,108,7,[mask],1,2
+6,19,104,9,[other],1,2
+6,20,104,9,[mask],1,2
+6,21,118,9,[other],1,2
+6,22,112,9,[other],1,[mask]
+6,23,118,9,[other],5,[mask]
+6,24,112,9,[other],7,[mask]
+6,25,104,2,6,8,[mask]
+6,26,104,9,6,8,[other]
+6,27,112,9,6,8,[mask]
+6,28,104,1,6,8,[other]
+6,29,112,0,[mask],8,[other]
+6,30,112,1,[mask],8,0
+6,31,112,1,[mask],8,[mask]
+6,32,104,1,[mask],8,[mask]
+6,33,[other],9,[other],8,[mask]
+7,3,129,9,1,5,1
+7,4,112,2,1,5,1
+7,5,112,1,1,5,1
+7,6,112,1,[mask],4,1
+7,7,107,1,[mask],5,1
+7,8,107,2,[mask],4,1
+7,9,104,7,[mask],5,1
+7,10,104,7,[mask],0,1
+7,11,108,7,[mask],0,1
+7,12,104,7,[mask],0,1
+7,13,104,7,9,0,1
+7,14,104,0,9,3,1
+7,15,104,0,[mask],3,1
+7,16,104,[mask],[mask],3,1
+7,17,104,5,9,3,0
+7,18,104,0,9,3,0
+7,19,103,0,9,3,0
+7,20,111,6,9,3,1
+7,21,103,6,9,3,0
+7,22,105,6,[mask],[unknown],1
+7,23,105,7,[mask],[unknown],2
+7,24,108,7,9,[unknown],0
+7,25,104,0,9,5,0
+7,26,104,0,9,5,1
+7,27,104,0,[mask],8,0
+7,28,104,0,[mask],5,1
+7,29,104,7,9,8,0
+7,30,112,0,1,5,0
+7,31,112,7,1,5,0
+7,32,112,7,1,5,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index a98e349c..d05f5e62 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,104,2,1,5,[mask]
-8,9,129,9,1,5,[mask]
-8,10,129,9,1,5,[mask]
-8,11,129,9,1,5,[mask]
-8,12,129,9,1,5,[mask]
-8,13,129,9,1,5,[mask]
-8,14,112,9,1,5,[other]
-8,15,107,7,[mask],[unknown],[mask]
-8,16,107,7,[mask],[unknown],[mask]
-8,17,104,7,[mask],[unknown],[mask]
-8,18,104,0,1,[unknown],2
-8,19,104,7,[mask],[unknown],[mask]
-8,20,104,9,1,8,[mask]
-8,21,112,9,1,[unknown],[mask]
-8,22,104,9,1,1,[mask]
-8,23,129,9,[other],1,[other]
-8,24,112,2,[other],1,[mask]
-8,25,129,2,[other],5,[mask]
-8,26,129,2,[other],1,[mask]
-8,27,129,2,[other],1,[other]
-8,28,129,2,1,1,[mask]
-8,29,129,9,1,5,[other]
-8,30,129,2,1,5,[other]
-8,31,129,2,1,5,[other]
-8,32,129,2,1,5,[other]
-8,33,129,2,1,5,[unknown]
-8,34,129,2,1,5,2
-8,35,129,2,1,5,2
-8,36,129,2,1,5,2
-8,37,107,2,1,[unknown],2
+8,4,104,7,1,5,1
+8,5,104,7,1,5,1
+8,6,112,7,[mask],[unknown],1
+8,7,107,7,[mask],5,1
+8,8,104,7,[mask],[unknown],1
+8,9,108,7,[mask],[unknown],1
+8,10,104,9,[mask],5,1
+8,11,108,7,[mask],0,1
+8,12,108,7,9,0,2
+8,13,104,7,9,1,2
+8,14,104,7,9,1,1
+8,15,104,7,9,[unknown],0
+8,16,104,9,8,5,2
+8,17,104,9,8,8,[other]
+8,18,112,9,[mask],3,[mask]
+8,19,104,9,[mask],8,[mask]
+8,20,112,9,[other],8,[mask]
+8,21,112,1,[mask],5,0
+8,22,112,1,1,[unknown],[mask]
+8,23,112,1,1,8,[mask]
+8,24,112,2,1,5,[mask]
+8,25,104,9,1,8,1
+8,26,104,0,[mask],8,1
+8,27,104,0,[mask],8,1
+8,28,104,0,[mask],8,1
+8,29,104,7,[mask],[unknown],1
+8,30,104,7,9,[unknown],1
+8,31,104,7,9,[unknown],1
+8,32,103,5,9,[unknown],0
+8,33,108,5,9,3,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index 9cd75ca4..e30646b9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,104,7,[mask],[unknown],[other]
-9,9,104,7,[mask],[unknown],[other]
-9,10,104,7,[mask],[unknown],[other]
-9,11,104,0,[mask],[unknown],0
-9,12,104,7,[mask],[unknown],2
-9,13,104,9,[other],[unknown],2
-9,14,104,9,[other],[unknown],[mask]
-9,15,[other],9,[other],[unknown],[mask]
-9,16,107,2,[other],1,[mask]
-9,17,107,2,[other],1,[mask]
-9,18,107,2,[other],1,[mask]
-9,19,107,2,[other],[unknown],[mask]
-9,20,110,9,[other],[unknown],[other]
-9,21,110,2,[other],[unknown],[other]
-9,22,110,2,3,[unknown],[other]
-9,23,110,1,1,[unknown],[other]
-9,24,110,2,[mask],[unknown],[other]
-9,25,110,2,[mask],7,[mask]
-9,26,[other],2,[mask],[unknown],[mask]
-9,27,[other],2,[mask],5,[mask]
-9,28,129,2,[other],8,[mask]
-9,29,129,2,1,5,[mask]
-9,30,129,9,1,5,[mask]
-9,31,129,9,1,5,[mask]
-9,32,129,9,1,5,[unknown]
-9,33,129,9,1,5,[mask]
-9,34,129,9,1,5,2
-9,35,129,2,1,5,2
-9,36,112,2,1,5,2
-9,37,107,7,[mask],[unknown],[mask]
+9,3,104,7,9,0,1
+9,4,104,7,9,8,1
+9,5,104,7,[mask],8,1
+9,6,104,7,9,[unknown],1
+9,7,104,7,9,8,0
+9,8,104,7,9,8,1
+9,9,104,7,9,[unknown],0
+9,10,108,7,9,[unknown],0
+9,11,103,6,1,[unknown],2
+9,12,108,5,1,[unknown],2
+9,13,108,9,1,[unknown],0
+9,14,129,9,1,5,2
+9,15,112,9,1,8,0
+9,16,112,0,1,5,2
+9,17,112,7,1,5,0
+9,18,112,0,1,5,1
+9,19,112,7,[mask],[unknown],0
+9,20,102,4,1,5,1
+9,21,112,7,[mask],[unknown],0
+9,22,119,7,[mask],5,1
+9,23,104,7,[mask],8,1
+9,24,104,7,[mask],[unknown],1
+9,25,104,7,[mask],[unknown],1
+9,26,104,7,[mask],8,1
+9,27,104,9,[mask],1,2
+9,28,104,7,9,1,1
+9,29,104,9,8,1,2
+9,30,118,9,[mask],[unknown],2
+9,31,112,9,[other],1,2
+9,32,104,5,[other],1,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index fd1bb4cf..aa9878ce 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,104,7,[mask],[unknown],1
-0,9,104,7,[mask],[unknown],0
-0,10,104,7,1,[unknown],0
-0,11,101,7,1,[unknown],0
-0,12,106,7,1,[unknown],[other]
-0,13,112,9,1,[unknown],0
-0,14,129,9,1,5,[unknown]
-0,15,112,2,1,1,[other]
-0,16,112,7,1,[unknown],2
-0,17,129,2,1,[unknown],2
-0,18,129,7,1,5,2
-0,19,112,7,1,1,2
-0,20,129,7,1,[unknown],2
-0,21,129,7,1,1,[other]
-0,22,112,7,1,5,[mask]
-0,23,118,7,[mask],[unknown],[mask]
-0,24,129,9,1,5,[other]
-0,25,112,9,1,5,[mask]
-0,26,[other],7,[mask],[unknown],[mask]
-0,27,129,9,1,5,[other]
-0,28,112,9,1,5,[other]
-0,29,129,7,1,[unknown],[mask]
-0,30,104,0,1,[unknown],[other]
-0,31,112,7,1,[unknown],[other]
-0,32,104,7,1,[unknown],2
-0,33,129,7,1,[unknown],[other]
-0,34,112,9,1,5,[other]
-0,35,112,7,1,[unknown],2
-0,36,108,7,1,[unknown],2
-0,37,129,7,1,5,2
+0,4,121,1,1,9,1
+0,5,105,0,[other],8,2
+0,6,113,1,[other],[mask],0
+0,7,105,8,[other],1,1
+0,8,113,1,[other],2,[unknown]
+0,9,113,3,[other],1,1
+0,10,113,1,[other],8,1
+0,11,113,1,[other],[mask],1
+0,12,105,8,9,1,[other]
+0,13,113,1,8,0,1
+0,14,105,0,[other],1,[other]
+0,15,113,[other],8,9,1
+0,16,105,0,[other],1,1
+0,17,113,1,8,8,1
+0,18,113,1,[other],7,1
+0,19,105,8,[other],1,[other]
+0,20,113,1,8,8,[unknown]
+0,21,113,1,[other],[mask],[other]
+0,22,113,[mask],[other],8,1
+0,23,113,1,[other],2,1
+0,24,113,[mask],[other],1,1
+0,25,113,1,[other],2,1
+0,26,113,1,[other],1,1
+0,27,113,1,[other],1,1
+0,28,105,1,[other],1,1
+0,29,108,1,8,0,1
+0,30,103,1,0,7,1
+0,31,103,1,[other],7,1
+0,32,108,5,9,0,1
+0,33,105,0,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index d23a5aeb..5e0f7cd5 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,104,7,[mask],1,1
-1,9,104,7,[mask],[unknown],[mask]
-1,10,[other],9,1,8,[mask]
-1,11,[other],9,1,5,[other]
-1,12,104,9,1,1,[other]
-1,13,112,9,1,[unknown],[other]
-1,14,112,7,1,[unknown],[other]
-1,15,112,2,1,5,[other]
-1,16,112,2,1,[unknown],[mask]
-1,17,129,2,1,[unknown],[mask]
-1,18,129,7,1,5,2
-1,19,129,2,1,1,2
-1,20,129,7,1,1,2
-1,21,129,7,1,1,2
-1,22,129,7,1,1,2
-1,23,129,7,1,1,2
-1,24,129,2,1,5,[mask]
-1,25,129,2,1,5,[mask]
-1,26,129,2,1,5,[mask]
-1,27,129,2,1,5,[mask]
-1,28,129,2,1,5,[mask]
-1,29,112,2,1,5,[mask]
-1,30,107,2,1,5,[mask]
-1,31,107,2,1,5,1
-1,32,107,2,[mask],[unknown],1
-1,33,107,2,[mask],[unknown],1
-1,34,107,2,[mask],[unknown],1
-1,35,107,7,[mask],[unknown],1
-1,36,107,7,[mask],[unknown],1
-1,37,107,[mask],[mask],[unknown],1
+1,4,107,5,[other],[mask],[other]
+1,5,121,3,[other],0,[other]
+1,6,129,3,[other],8,[other]
+1,7,121,[other],[other],8,[other]
+1,8,[unknown],5,[other],[mask],[other]
+1,9,121,6,3,4,1
+1,10,105,1,[other],8,[unknown]
+1,11,121,1,7,[mask],[other]
+1,12,105,0,7,1,[other]
+1,13,105,[other],7,5,0
+1,14,105,1,1,1,1
+1,15,105,1,7,[other],[other]
+1,16,105,0,7,0,1
+1,17,105,1,7,1,1
+1,18,113,1,7,1,1
+1,19,113,1,7,1,1
+1,20,113,1,[other],1,1
+1,21,113,1,1,1,1
+1,22,113,0,[other],1,1
+1,23,113,1,1,0,1
+1,24,108,1,[other],1,1
+1,25,108,0,[other],7,1
+1,26,108,5,[other],0,1
+1,27,113,5,[other],7,1
+1,28,105,8,[other],0,[other]
+1,29,113,[other],8,2,[other]
+1,30,113,[mask],[other],[unknown],1
+1,31,113,1,[other],2,1
+1,32,113,1,[other],8,1
+1,33,113,1,[other],[mask],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index cc5b8a64..fa4d361a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,112,2,[mask],5,[mask]
-2,9,104,7,[mask],[unknown],[mask]
-2,10,104,7,1,8,[mask]
-2,11,112,7,1,[unknown],[other]
-2,12,112,7,1,[unknown],[mask]
-2,13,107,7,[mask],[unknown],[mask]
-2,14,107,7,[mask],[unknown],[mask]
-2,15,[other],7,[mask],[unknown],2
-2,16,129,9,[other],1,[other]
-2,17,112,9,[other],1,[mask]
-2,18,104,1,[other],5,[mask]
-2,19,104,2,[other],[unknown],[other]
-2,20,110,9,[mask],[unknown],[other]
-2,21,110,2,[mask],[unknown],[mask]
-2,22,[other],2,[mask],[unknown],[other]
-2,23,110,2,[mask],5,[mask]
-2,24,[other],2,[other],7,[mask]
-2,25,104,2,[other],8,[mask]
-2,26,107,2,[other],1,[mask]
-2,27,107,2,[other],1,[mask]
-2,28,107,2,[other],8,[mask]
-2,29,107,2,1,5,[mask]
-2,30,107,2,1,5,[unknown]
-2,31,129,2,1,5,[mask]
-2,32,129,9,1,5,[other]
-2,33,107,2,1,5,[mask]
-2,34,107,2,1,5,[mask]
-2,35,107,2,1,[unknown],1
-2,36,107,2,[mask],[unknown],1
-2,37,107,7,[mask],[unknown],1
+2,3,128,0,7,[unknown],2
+2,4,113,8,[other],6,2
+2,5,109,1,[other],[mask],1
+2,6,128,1,[other],7,2
+2,7,105,8,[other],6,[other]
+2,8,121,8,1,[mask],1
+2,9,114,1,[other],8,2
+2,10,113,1,[other],[mask],1
+2,11,104,1,[other],0,1
+2,12,113,1,7,0,1
+2,13,113,1,[other],1,1
+2,14,113,1,[other],1,1
+2,15,113,1,[other],1,1
+2,16,105,1,[other],1,1
+2,17,113,1,[other],0,1
+2,18,105,0,[other],1,1
+2,19,108,0,5,0,1
+2,20,113,0,[other],0,1
+2,21,108,0,9,2,1
+2,22,113,0,[other],0,1
+2,23,113,8,[other],2,1
+2,24,113,3,[other],8,1
+2,25,113,1,[other],8,1
+2,26,113,1,[other],8,1
+2,27,113,1,[other],0,1
+2,28,113,1,[other],7,1
+2,29,105,8,[other],0,1
+2,30,113,1,[other],2,1
+2,31,113,8,[other],1,1
+2,32,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index 2a4d06f3..865dd214 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,104,[mask],[mask],[unknown],[mask]
-3,9,104,0,[mask],8,[other]
-3,10,104,[mask],[mask],6,[mask]
-3,11,104,0,[mask],8,[mask]
-3,12,104,0,[mask],8,[mask]
-3,13,104,1,[mask],8,[mask]
-3,14,[other],2,[mask],[unknown],[mask]
-3,15,129,1,[other],6,[mask]
-3,16,107,2,[other],1,[mask]
-3,17,107,2,[other],6,[mask]
-3,18,129,9,[other],6,[mask]
-3,19,129,9,[other],5,[mask]
-3,20,129,9,[other],5,[mask]
-3,21,129,9,[other],5,[mask]
-3,22,129,9,[other],5,[other]
-3,23,110,9,[other],5,[mask]
-3,24,107,2,[other],8,[other]
-3,25,107,2,1,5,[other]
-3,26,110,9,1,[unknown],[other]
-3,27,[other],9,1,[unknown],[other]
-3,28,[other],2,1,[unknown],[other]
-3,29,[other],2,1,[unknown],[other]
-3,30,107,2,1,[unknown],[unknown]
-3,31,129,2,1,[unknown],[mask]
-3,32,129,2,1,[unknown],2
-3,33,129,2,1,5,2
-3,34,129,2,1,5,2
-3,35,129,2,1,5,2
-3,36,129,2,1,5,2
-3,37,129,2,1,5,2
-4,8,104,9,1,[unknown],[mask]
-4,9,[other],7,1,[unknown],[other]
-4,10,104,9,1,[unknown],[other]
-4,11,[other],9,1,[unknown],[other]
-4,12,[other],9,1,[unknown],[other]
-4,13,[other],9,1,[unknown],[other]
-4,14,129,2,1,5,[unknown]
-4,15,129,2,1,1,2
-4,16,129,2,1,1,2
-4,17,129,7,1,5,2
-4,18,129,2,1,1,2
-4,19,129,7,1,5,2
-4,20,129,7,1,1,2
-4,21,129,7,1,1,[mask]
-4,22,129,7,1,5,[mask]
-4,23,129,2,1,5,[mask]
-4,24,112,2,1,5,[mask]
-4,25,107,2,[mask],[unknown],[mask]
-4,26,129,2,[mask],5,[mask]
-4,27,129,9,1,5,[mask]
-4,28,129,9,1,5,[mask]
-4,29,129,9,1,5,[mask]
-4,30,129,9,1,5,[mask]
-4,31,129,9,1,5,[mask]
-4,32,129,9,1,5,[mask]
-4,33,129,9,1,5,[other]
-4,34,112,9,1,5,[mask]
-4,35,107,7,[mask],[unknown],[mask]
-4,36,107,7,[mask],[unknown],[mask]
-4,37,104,7,[mask],[unknown],[mask]
+3,3,128,5,1,1,2
+3,4,105,2,9,5,[unknown]
+3,5,128,1,1,[mask],1
+3,6,105,8,[other],1,2
+3,7,128,1,9,[mask],1
+3,8,128,0,9,1,1
+3,9,119,1,[other],[mask],2
+3,10,104,1,8,[mask],1
+3,11,104,1,9,0,[other]
+3,12,113,3,[other],1,[other]
+3,13,113,1,[other],[mask],[other]
+3,14,113,3,[other],[unknown],[other]
+3,15,113,3,[other],[mask],[other]
+3,16,105,3,[other],8,[other]
+3,17,105,1,7,8,[other]
+3,18,113,1,7,1,[other]
+3,19,113,1,[other],1,1
+3,20,113,1,[other],1,[other]
+3,21,113,1,[other],[mask],1
+3,22,120,1,[other],1,1
+3,23,113,1,7,7,1
+3,24,105,1,7,1,1
+3,25,113,1,7,1,1
+3,26,113,0,1,1,1
+3,27,105,0,[other],1,[other]
+3,28,108,0,8,0,1
+3,29,113,0,[other],7,1
+3,30,108,8,[other],2,1
+3,31,113,[mask],[other],7,1
+3,32,124,[mask],[other],2,1
+4,3,102,0,[other],[unknown],[unknown]
+4,4,102,8,[other],6,[unknown]
+4,5,105,8,[other],6,[mask]
+4,6,127,1,[other],2,1
+4,7,127,[mask],7,[unknown],1
+4,8,102,[mask],1,2,1
+4,9,127,[mask],[other],8,[unknown]
+4,10,106,[mask],1,0,1
+4,11,113,1,[other],1,1
+4,12,106,8,1,[mask],1
+4,13,105,1,[other],[mask],1
+4,14,106,8,9,0,1
+4,15,113,1,[other],0,1
+4,16,113,1,[other],2,1
+4,17,113,8,[other],1,1
+4,18,113,1,[other],2,1
+4,19,113,3,[other],1,2
+4,20,113,1,[other],[mask],1
+4,21,105,8,[other],1,1
+4,22,113,1,8,[mask],1
+4,23,105,3,[other],1,1
+4,24,113,1,8,0,1
+4,25,105,1,[other],1,1
+4,26,113,1,9,0,1
+4,27,113,1,[other],1,1
+4,28,113,1,[other],1,2
+4,29,113,1,[other],1,[other]
+4,30,113,0,[other],1,[other]
+4,31,113,0,[other],[mask],[other]
+4,32,121,0,[other],2,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index 529df285..67ddc700 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,102,7,[mask],9,1
-5,9,107,[mask],[mask],[unknown],1
-5,10,107,[mask],[mask],[unknown],1
-5,11,107,[mask],[mask],[unknown],1
-5,12,104,[mask],[mask],[unknown],1
-5,13,104,[mask],[mask],[unknown],1
-5,14,104,[mask],[mask],[unknown],1
-5,15,104,[mask],[mask],3,1
-5,16,104,[mask],[mask],3,[other]
-5,17,112,[unknown],[mask],3,[other]
-5,18,111,[mask],[mask],3,2
-5,19,104,9,[mask],3,2
-5,20,104,5,[mask],3,2
-5,21,104,5,[mask],3,2
-5,22,104,5,[mask],3,2
-5,23,104,[unknown],[mask],3,2
-5,24,104,9,[mask],3,[unknown]
-5,25,104,9,[mask],3,[mask]
-5,26,112,[unknown],[mask],3,[mask]
-5,27,112,[unknown],[mask],3,[mask]
-5,28,112,[mask],[mask],3,[mask]
-5,29,104,[mask],[mask],7,[mask]
-5,30,107,9,[mask],3,[mask]
-5,31,107,2,[mask],3,[mask]
-5,32,107,2,[mask],6,[mask]
-5,33,104,2,[mask],6,[mask]
-5,34,104,9,[mask],6,[mask]
-5,35,[other],9,[mask],6,[mask]
-5,36,[other],2,[other],6,[mask]
-5,37,107,9,[other],6,[mask]
+5,3,128,3,[other],4,1
+5,4,128,1,[other],6,1
+5,5,105,1,[other],7,1
+5,6,105,1,[other],0,1
+5,7,105,1,[other],0,1
+5,8,105,1,[other],0,1
+5,9,105,1,7,2,[other]
+5,10,113,[mask],7,0,1
+5,11,113,1,[other],2,1
+5,12,113,[mask],[other],1,1
+5,13,113,1,[other],2,1
+5,14,113,[mask],[other],1,1
+5,15,113,1,[other],2,1
+5,16,113,1,[other],1,1
+5,17,113,1,1,[mask],1
+5,18,108,1,9,1,1
+5,19,103,1,0,7,1
+5,20,108,1,[other],7,1
+5,21,108,8,[other],7,1
+5,22,108,5,[other],2,1
+5,23,113,[mask],[other],7,1
+5,24,105,8,[other],2,[other]
+5,25,113,8,[other],8,1
+5,26,113,1,[other],[mask],1
+5,27,113,8,[other],2,1
+5,28,113,1,[other],8,1
+5,29,113,1,[other],[mask],1
+5,30,113,8,[other],2,1
+5,31,113,1,[other],2,1
+5,32,113,1,[other],7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index 819defac..62b206ad 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,129,9,1,5,[other]
-6,9,129,9,1,5,[mask]
-6,10,129,9,1,5,[other]
-6,11,129,9,1,5,[other]
-6,12,129,9,1,5,[other]
-6,13,129,2,1,5,[mask]
-6,14,129,2,1,8,[other]
-6,15,112,7,1,[unknown],[unknown]
-6,16,107,7,1,[unknown],2
-6,17,108,7,[mask],[unknown],2
-6,18,104,7,1,5,2
-6,19,108,7,1,[unknown],2
-6,20,104,7,1,5,2
-6,21,112,9,[mask],1,[mask]
-6,22,129,9,[other],5,[mask]
-6,23,104,9,[other],5,[mask]
-6,24,112,9,8,1,[mask]
-6,25,129,9,[other],5,[mask]
-6,26,104,9,[other],1,[mask]
-6,27,104,9,[other],1,[other]
-6,28,112,7,[other],1,[mask]
-6,29,[other],1,[other],8,[other]
-6,30,[other],2,[mask],[unknown],[other]
-6,31,[other],2,1,[unknown],[other]
-6,32,110,2,1,5,[other]
-6,33,[other],2,1,[unknown],[other]
-6,34,[other],2,1,5,[mask]
-6,35,129,2,1,8,[unknown]
-6,36,129,2,1,5,[unknown]
-6,37,129,2,1,5,2
-7,8,129,2,1,5,[mask]
-7,9,129,2,1,5,[mask]
-7,10,129,2,1,5,[mask]
-7,11,129,2,1,5,[mask]
-7,12,129,2,1,5,1
-7,13,107,2,1,5,[mask]
-7,14,107,2,1,[unknown],1
-7,15,107,2,[mask],[unknown],1
-7,16,107,2,[mask],[unknown],1
-7,17,107,7,[mask],[unknown],1
-7,18,107,7,[mask],[unknown],1
-7,19,107,7,[mask],[unknown],1
-7,20,107,7,[mask],[unknown],1
-7,21,107,[mask],[mask],[unknown],1
-7,22,104,[mask],[mask],[unknown],2
-7,23,104,9,[mask],1,[other]
-7,24,104,9,[mask],3,[mask]
-7,25,104,[unknown],[mask],3,[other]
-7,26,104,[mask],[mask],3,[mask]
-7,27,104,9,[mask],[unknown],[other]
-7,28,112,9,[mask],[unknown],[mask]
-7,29,104,2,[mask],[unknown],[mask]
-7,30,107,2,[other],[unknown],[mask]
-7,31,107,2,3,1,[mask]
-7,32,107,2,[other],[unknown],[mask]
-7,33,107,2,[other],[unknown],[mask]
-7,34,107,2,[other],[unknown],[mask]
-7,35,107,2,[other],8,[mask]
-7,36,[other],2,[other],8,[mask]
-7,37,[other],9,[other],5,[other]
+6,4,122,2,[other],8,[unknown]
+6,5,113,[mask],[other],8,[other]
+6,6,113,1,[other],[unknown],1
+6,7,113,[mask],[other],2,1
+6,8,113,1,[other],2,1
+6,9,113,3,[other],1,1
+6,10,121,1,[other],[mask],1
+6,11,108,0,7,1,1
+6,12,113,1,1,[mask],1
+6,13,108,8,9,1,[other]
+6,14,113,0,1,[mask],1
+6,15,108,8,[other],1,[other]
+6,16,113,0,[other],7,[other]
+6,17,105,8,[other],2,[other]
+6,18,113,[mask],[other],8,[other]
+6,19,113,1,[other],[mask],1
+6,20,113,1,[other],1,1
+6,21,113,1,[other],[mask],1
+6,22,113,1,[other],[mask],1
+6,23,124,1,[other],2,1
+6,24,113,1,[other],7,[unknown]
+6,25,105,8,[other],1,1
+6,26,113,1,[other],2,1
+6,27,113,8,[other],1,1
+6,28,113,1,[other],2,1
+6,29,113,8,[other],1,1
+6,30,113,1,[other],2,1
+6,31,113,[mask],9,1,1
+6,32,113,1,[other],1,1
+6,33,113,1,9,1,1
+7,3,105,0,[other],8,[mask]
+7,4,105,0,[other],1,[other]
+7,5,105,0,9,6,[unknown]
+7,6,105,0,[other],1,[unknown]
+7,7,113,[other],[other],1,[other]
+7,8,113,0,[other],1,1
+7,9,113,0,[other],1,2
+7,10,113,1,[other],[mask],1
+7,11,113,3,[other],1,1
+7,12,113,1,[other],[mask],1
+7,13,108,3,[other],1,1
+7,14,113,1,[other],[mask],1
+7,15,124,3,9,2,1
+7,16,113,0,8,0,1
+7,17,105,1,[other],[mask],[other]
+7,18,108,0,7,0,[other]
+7,19,113,5,[other],7,[other]
+7,20,105,8,[other],[mask],[other]
+7,21,113,[mask],[other],8,[other]
+7,22,113,1,[other],[mask],1
+7,23,113,1,[other],8,1
+7,24,113,1,[other],8,1
+7,25,113,1,[other],8,1
+7,26,113,1,[other],[mask],1
+7,27,105,8,[other],2,1
+7,28,113,1,8,2,1
+7,29,113,1,[other],1,1
+7,30,113,8,[other],1,1
+7,31,113,1,[other],2,1
+7,32,113,3,[other],1,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index a98e349c..f7a88c52 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,8,104,2,1,5,[mask]
-8,9,129,9,1,5,[mask]
-8,10,129,9,1,5,[mask]
-8,11,129,9,1,5,[mask]
-8,12,129,9,1,5,[mask]
-8,13,129,9,1,5,[mask]
-8,14,112,9,1,5,[other]
-8,15,107,7,[mask],[unknown],[mask]
-8,16,107,7,[mask],[unknown],[mask]
-8,17,104,7,[mask],[unknown],[mask]
-8,18,104,0,1,[unknown],2
-8,19,104,7,[mask],[unknown],[mask]
-8,20,104,9,1,8,[mask]
-8,21,112,9,1,[unknown],[mask]
-8,22,104,9,1,1,[mask]
-8,23,129,9,[other],1,[other]
-8,24,112,2,[other],1,[mask]
-8,25,129,2,[other],5,[mask]
-8,26,129,2,[other],1,[mask]
-8,27,129,2,[other],1,[other]
-8,28,129,2,1,1,[mask]
-8,29,129,9,1,5,[other]
-8,30,129,2,1,5,[other]
-8,31,129,2,1,5,[other]
-8,32,129,2,1,5,[other]
-8,33,129,2,1,5,[unknown]
-8,34,129,2,1,5,2
-8,35,129,2,1,5,2
-8,36,129,2,1,5,2
-8,37,107,2,1,[unknown],2
+8,4,101,1,[other],8,2
+8,5,104,1,[other],[unknown],2
+8,6,104,3,[other],[other],[other]
+8,7,103,5,[other],0,1
+8,8,128,1,[other],0,[other]
+8,9,105,3,[other],6,[other]
+8,10,128,1,7,[mask],[other]
+8,11,120,[mask],7,[unknown],[other]
+8,12,113,1,[other],[mask],1
+8,13,105,1,[other],2,[other]
+8,14,113,1,7,2,1
+8,15,113,1,[other],1,1
+8,16,113,1,[other],[mask],1
+8,17,113,1,[other],2,1
+8,18,113,1,[other],1,1
+8,19,113,1,[other],1,1
+8,20,105,0,[other],1,1
+8,21,105,1,8,9,1
+8,22,103,0,9,1,1
+8,23,113,0,8,0,1
+8,24,113,0,[other],1,1
+8,25,113,0,[other],1,1
+8,26,113,0,[other],1,1
+8,27,113,0,[other],8,1
+8,28,113,1,[other],8,1
+8,29,113,1,[other],8,1
+8,30,113,1,[other],0,1
+8,31,113,1,[other],0,1
+8,32,108,1,[other],1,1
+8,33,113,1,[other],7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index 9cd75ca4..7f5bf564 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,104,7,[mask],[unknown],[other]
-9,9,104,7,[mask],[unknown],[other]
-9,10,104,7,[mask],[unknown],[other]
-9,11,104,0,[mask],[unknown],0
-9,12,104,7,[mask],[unknown],2
-9,13,104,9,[other],[unknown],2
-9,14,104,9,[other],[unknown],[mask]
-9,15,[other],9,[other],[unknown],[mask]
-9,16,107,2,[other],1,[mask]
-9,17,107,2,[other],1,[mask]
-9,18,107,2,[other],1,[mask]
-9,19,107,2,[other],[unknown],[mask]
-9,20,110,9,[other],[unknown],[other]
-9,21,110,2,[other],[unknown],[other]
-9,22,110,2,3,[unknown],[other]
-9,23,110,1,1,[unknown],[other]
-9,24,110,2,[mask],[unknown],[other]
-9,25,110,2,[mask],7,[mask]
-9,26,[other],2,[mask],[unknown],[mask]
-9,27,[other],2,[mask],5,[mask]
-9,28,129,2,[other],8,[mask]
-9,29,129,2,1,5,[mask]
-9,30,129,9,1,5,[mask]
-9,31,129,9,1,5,[mask]
-9,32,129,9,1,5,[unknown]
-9,33,129,9,1,5,[mask]
-9,34,129,9,1,5,2
-9,35,129,2,1,5,2
-9,36,112,2,1,5,2
-9,37,107,7,[mask],[unknown],[mask]
+9,3,108,5,[other],8,2
+9,4,122,5,[other],7,1
+9,5,105,5,[other],7,[other]
+9,6,122,5,7,7,[other]
+9,7,122,5,[other],7,[other]
+9,8,122,5,[other],7,1
+9,9,122,1,[other],2,1
+9,10,113,[mask],7,8,1
+9,11,113,1,[other],2,1
+9,12,113,[mask],7,2,1
+9,13,113,1,[other],2,1
+9,14,113,[mask],[other],2,1
+9,15,113,[mask],1,2,1
+9,16,113,1,[other],7,1
+9,17,108,8,1,7,1
+9,18,108,1,[other],7,1
+9,19,108,1,[other],7,1
+9,20,108,1,[other],7,1
+9,21,108,8,[other],7,1
+9,22,122,1,[other],2,1
+9,23,113,1,[other],2,1
+9,24,113,8,[other],8,1
+9,25,113,1,[other],8,1
+9,26,113,1,[other],0,1
+9,27,113,8,[other],2,1
+9,28,113,1,[other],2,1
+9,29,113,8,[other],1,1
+9,30,113,1,1,2,1
+9,31,113,[mask],[other],1,1
+9,32,113,1,1,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
index 8cab47cb..8c8e70ba 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
@@ -1,16 +1,20 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,8,102,3,9,5,1
+0,4,102,3,9,5,1
+0,5,102,4,3,0,1
+0,6,102,6,9,2,1
+0,7,102,4,3,0,1
+0,8,112,6,9,2,1
 0,9,102,4,3,0,1
-0,10,102,6,9,2,1
+0,10,112,6,9,0,0
 0,11,102,4,3,0,1
-0,12,112,6,9,2,1
-0,13,102,4,3,0,1
-0,14,112,6,9,0,0
-0,15,102,4,3,0,1
-0,16,112,6,9,0,0
-0,17,102,4,3,0,2
-0,18,112,6,3,0,0
-0,19,102,4,9,0,0
+0,12,112,6,9,0,0
+0,13,102,4,3,0,2
+0,14,112,6,3,0,0
+0,15,102,4,9,0,0
+0,16,112,4,3,0,0
+0,17,112,4,3,0,0
+0,18,112,4,3,0,0
+0,19,112,4,3,0,0
 0,20,112,4,3,0,0
 0,21,112,4,3,0,0
 0,22,112,4,3,0,0
@@ -25,7 +29,3 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 0,31,112,4,3,0,0
 0,32,112,4,3,0,0
 0,33,112,4,3,0,0
-0,34,112,4,3,0,0
-0,35,112,4,3,0,0
-0,36,112,4,3,0,0
-0,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
index 5429f965..60c86728 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
@@ -1,19 +1,23 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,8,102,4,0,5,0
-1,9,102,6,0,2,1
-1,10,111,4,2,2,1
-1,11,112,4,9,5,1
-1,12,112,4,9,8,1
-1,13,112,4,9,0,0
-1,14,112,7,9,0,1
+1,4,102,4,0,5,0
+1,5,102,6,0,2,1
+1,6,111,4,2,2,1
+1,7,112,4,9,5,1
+1,8,112,4,9,8,1
+1,9,112,4,9,0,0
+1,10,112,7,9,0,1
+1,11,112,4,3,0,0
+1,12,112,4,9,0,0
+1,13,112,4,9,0,1
+1,14,112,4,9,0,0
 1,15,112,4,3,0,0
 1,16,112,4,9,0,0
-1,17,112,4,9,0,1
+1,17,112,4,3,0,0
 1,18,112,4,9,0,0
 1,19,112,4,3,0,0
-1,20,112,4,9,0,0
+1,20,112,4,3,0,0
 1,21,112,4,3,0,0
-1,22,112,4,9,0,0
+1,22,112,4,3,0,0
 1,23,112,4,3,0,0
 1,24,112,4,3,0,0
 1,25,112,4,3,0,0
@@ -25,7 +29,3 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 1,31,112,4,3,0,0
 1,32,112,4,3,0,0
 1,33,112,4,3,0,0
-1,34,112,4,3,0,0
-1,35,112,4,3,0,0
-1,36,112,4,3,0,0
-1,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
index d61975e1..9c10138e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
@@ -1,12 +1,17 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,8,112,4,0,5,1
+2,3,112,4,0,5,1
+2,4,112,4,3,0,0
+2,5,112,4,9,0,1
+2,6,112,4,9,0,0
+2,7,112,4,3,0,0
+2,8,112,4,9,0,0
 2,9,112,4,3,0,0
-2,10,112,4,9,0,1
-2,11,112,4,9,0,0
+2,10,112,4,9,0,0
+2,11,112,4,3,0,0
 2,12,112,4,3,0,0
-2,13,112,4,9,0,0
+2,13,112,4,3,0,0
 2,14,112,4,3,0,0
-2,15,112,4,9,0,0
+2,15,112,4,3,0,0
 2,16,112,4,3,0,0
 2,17,112,4,3,0,0
 2,18,112,4,3,0,0
@@ -24,8 +29,3 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 2,30,112,4,3,0,0
 2,31,112,4,3,0,0
 2,32,112,4,3,0,0
-2,33,112,4,3,0,0
-2,34,112,4,3,0,0
-2,35,112,4,3,0,0
-2,36,112,4,3,0,0
-2,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
index dae0b6c2..8295c2ce 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
@@ -1,15 +1,20 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,8,112,3,3,5,0
-3,9,102,4,3,5,0
-3,10,112,3,3,0,0
-3,11,112,4,3,5,0
+3,3,112,3,3,5,0
+3,4,102,4,3,5,0
+3,5,112,3,3,0,0
+3,6,112,4,3,5,0
+3,7,112,4,3,0,0
+3,8,112,4,3,0,0
+3,9,112,4,9,0,0
+3,10,112,4,3,0,0
+3,11,112,4,9,0,0
 3,12,112,4,3,0,0
-3,13,112,4,3,0,0
-3,14,112,4,9,0,0
+3,13,112,4,9,0,0
+3,14,112,4,3,0,0
 3,15,112,4,3,0,0
-3,16,112,4,9,0,0
+3,16,112,4,3,0,0
 3,17,112,4,3,0,0
-3,18,112,4,9,0,0
+3,18,112,4,3,0,0
 3,19,112,4,3,0,0
 3,20,112,4,3,0,0
 3,21,112,4,3,0,0
@@ -24,38 +29,33 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 3,30,112,4,3,0,0
 3,31,112,4,3,0,0
 3,32,112,4,3,0,0
-3,33,112,4,3,0,0
-3,34,112,4,3,0,0
-3,35,112,4,3,0,0
-3,36,112,4,3,0,0
-3,37,112,4,3,0,0
-4,8,113,4,9,5,1
-4,9,112,4,9,8,1
-4,10,118,4,9,8,1
-4,11,102,4,9,0,1
+4,3,113,4,9,5,1
+4,4,112,4,9,8,1
+4,5,118,4,9,8,1
+4,6,102,4,9,0,1
+4,7,112,4,9,0,1
+4,8,112,4,9,0,1
+4,9,112,4,9,0,1
+4,10,112,4,9,0,1
+4,11,112,4,3,0,0
 4,12,112,4,9,0,1
-4,13,112,4,9,0,1
-4,14,112,4,9,0,1
-4,15,112,4,9,0,1
+4,13,112,4,3,0,0
+4,14,112,4,3,0,1
+4,15,112,4,9,0,0
 4,16,112,4,3,0,0
-4,17,112,4,9,0,1
+4,17,112,4,9,0,0
 4,18,112,4,3,0,0
-4,19,112,4,3,0,1
-4,20,112,4,9,0,0
+4,19,112,4,9,0,0
+4,20,112,4,3,0,0
 4,21,112,4,3,0,0
 4,22,112,4,9,0,0
 4,23,112,4,3,0,0
-4,24,112,4,9,0,0
+4,24,112,4,3,0,0
 4,25,112,4,3,0,0
 4,26,112,4,3,0,0
-4,27,112,4,9,0,0
+4,27,112,4,3,0,0
 4,28,112,4,3,0,0
 4,29,112,4,3,0,0
 4,30,112,4,3,0,0
 4,31,112,4,3,0,0
 4,32,112,4,3,0,0
-4,33,112,4,3,0,0
-4,34,112,4,3,0,0
-4,35,112,4,3,0,0
-4,36,112,4,3,0,0
-4,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
index 83cec41e..cdb9d0fa 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,8,108,4,9,8,1
-5,9,104,4,9,8,1
-5,10,119,4,9,8,1
-5,11,104,4,9,8,1
-5,12,119,4,9,8,1
-5,13,104,4,9,8,0
-5,14,102,4,9,5,1
-5,15,119,4,3,0,0
+5,3,108,4,9,8,1
+5,4,104,4,9,8,1
+5,5,119,4,9,8,1
+5,6,104,4,9,8,1
+5,7,119,4,9,8,1
+5,8,104,4,9,8,0
+5,9,102,4,9,5,1
+5,10,119,4,3,0,0
+5,11,112,4,3,6,0
+5,12,112,4,3,5,0
+5,13,112,4,3,5,0
+5,14,112,4,3,0,0
+5,15,112,4,3,0,0
 5,16,112,4,3,6,0
-5,17,112,4,3,5,0
-5,18,112,4,3,5,0
-5,19,112,4,3,0,0
+5,17,112,4,3,0,0
+5,18,112,4,9,0,0
+5,19,112,4,3,6,0
 5,20,112,4,3,0,0
-5,21,112,4,3,6,0
+5,21,112,4,9,0,0
 5,22,112,4,3,0,0
-5,23,112,4,9,0,0
+5,23,112,4,3,0,0
 5,24,112,4,3,6,0
 5,25,112,4,3,0,0
-5,26,112,4,9,0,0
-5,27,112,4,3,0,0
+5,26,112,4,3,0,0
+5,27,112,4,3,6,0
 5,28,112,4,3,0,0
-5,29,112,4,3,6,0
+5,29,112,4,3,0,0
 5,30,112,4,3,0,0
-5,31,112,4,3,0,0
-5,32,112,4,3,6,0
-5,33,112,4,3,0,0
-5,34,112,4,3,0,0
-5,35,112,4,3,0,0
-5,36,112,4,3,6,0
-5,37,112,4,3,0,0
+5,31,112,4,3,6,0
+5,32,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
index 2ce84b09..181f8b8a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
@@ -1,19 +1,23 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,8,102,4,9,5,1
-6,9,112,4,3,0,1
-6,10,112,4,9,5,1
-6,11,112,4,3,0,1
-6,12,112,4,9,0,1
+6,4,102,4,9,5,1
+6,5,112,4,3,0,1
+6,6,112,4,9,5,1
+6,7,112,4,3,0,1
+6,8,112,4,9,0,1
+6,9,112,4,3,0,0
+6,10,112,4,9,0,1
+6,11,112,4,3,0,0
+6,12,112,4,9,0,0
 6,13,112,4,3,0,0
-6,14,112,4,9,0,1
+6,14,112,4,9,0,0
 6,15,112,4,3,0,0
 6,16,112,4,9,0,0
 6,17,112,4,3,0,0
 6,18,112,4,9,0,0
 6,19,112,4,3,0,0
-6,20,112,4,9,0,0
+6,20,112,4,3,0,0
 6,21,112,4,3,0,0
-6,22,112,4,9,0,0
+6,22,112,4,3,0,0
 6,23,112,4,3,0,0
 6,24,112,4,3,0,0
 6,25,112,4,3,0,0
@@ -25,25 +29,26 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,31,112,4,3,0,0
 6,32,112,4,3,0,0
 6,33,112,4,3,0,0
-6,34,112,4,3,0,0
-6,35,112,4,3,0,0
-6,36,112,4,3,0,0
-6,37,112,4,3,0,0
-7,8,102,4,0,5,1
-7,9,112,4,0,8,1
-7,10,118,4,9,8,1
-7,11,112,4,9,0,1
+7,3,102,4,0,5,1
+7,4,112,4,0,8,1
+7,5,118,4,9,8,1
+7,6,112,4,9,0,1
+7,7,112,4,9,0,1
+7,8,112,4,9,0,1
+7,9,112,4,9,0,1
+7,10,112,4,9,0,0
+7,11,112,4,3,0,0
 7,12,112,4,9,0,1
-7,13,112,4,9,0,1
-7,14,112,4,9,0,1
+7,13,112,4,9,0,0
+7,14,112,4,3,0,0
 7,15,112,4,9,0,0
 7,16,112,4,3,0,0
-7,17,112,4,9,0,1
-7,18,112,4,9,0,0
+7,17,112,4,9,0,0
+7,18,112,4,3,0,0
 7,19,112,4,3,0,0
-7,20,112,4,9,0,0
+7,20,112,4,3,0,0
 7,21,112,4,3,0,0
-7,22,112,4,9,0,0
+7,22,112,4,3,0,0
 7,23,112,4,3,0,0
 7,24,112,4,3,0,0
 7,25,112,4,3,0,0
@@ -54,8 +59,3 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 7,30,112,4,3,0,0
 7,31,112,4,3,0,0
 7,32,112,4,3,0,0
-7,33,112,4,3,0,0
-7,34,112,4,3,0,0
-7,35,112,4,3,0,0
-7,36,112,4,3,0,0
-7,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
index c373d0f8..a4fd8639 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
@@ -1,18 +1,22 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+8,4,112,4,9,5,0
+8,5,112,4,0,5,0
+8,6,112,7,8,5,0
+8,7,112,4,9,5,0
 8,8,112,4,9,5,0
-8,9,112,4,0,5,0
-8,10,112,7,8,5,0
-8,11,112,4,9,5,0
-8,12,112,4,9,5,0
-8,13,112,4,9,5,1
-8,14,112,4,9,0,0
-8,15,112,4,9,0,0
+8,9,112,4,9,5,1
+8,10,112,4,9,0,0
+8,11,112,4,9,0,0
+8,12,112,4,3,0,0
+8,13,112,4,9,0,0
+8,14,112,4,3,0,0
+8,15,112,4,3,0,0
 8,16,112,4,3,0,0
 8,17,112,4,9,0,0
 8,18,112,4,3,0,0
 8,19,112,4,3,0,0
 8,20,112,4,3,0,0
-8,21,112,4,9,0,0
+8,21,112,4,3,0,0
 8,22,112,4,3,0,0
 8,23,112,4,3,0,0
 8,24,112,4,3,0,0
@@ -25,7 +29,3 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 8,31,112,4,3,0,0
 8,32,112,4,3,0,0
 8,33,112,4,3,0,0
-8,34,112,4,3,0,0
-8,35,112,4,3,0,0
-8,36,112,4,3,0,0
-8,37,112,4,3,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
index 3613e328..42cdbf62 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
@@ -1,18 +1,23 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,8,119,4,9,8,1
-9,9,102,4,9,8,1
-9,10,102,4,9,8,1
-9,11,102,4,9,0,1
-9,12,102,4,9,0,1
-9,13,102,6,9,0,1
-9,14,112,4,9,0,1
-9,15,112,4,9,0,0
-9,16,112,4,3,0,0
+9,3,119,4,9,8,1
+9,4,102,4,9,8,1
+9,5,102,4,9,8,1
+9,6,102,4,9,0,1
+9,7,102,4,9,0,1
+9,8,102,6,9,0,1
+9,9,112,4,9,0,1
+9,10,112,4,9,0,0
+9,11,112,4,3,0,0
+9,12,112,4,3,0,0
+9,13,112,4,3,0,0
+9,14,112,4,3,0,0
+9,15,112,4,3,0,0
+9,16,112,4,9,0,0
 9,17,112,4,3,0,0
 9,18,112,4,3,0,0
 9,19,112,4,3,0,0
 9,20,112,4,3,0,0
-9,21,112,4,9,0,0
+9,21,112,4,3,0,0
 9,22,112,4,3,0,0
 9,23,112,4,3,0,0
 9,24,112,4,3,0,0
@@ -24,8 +29,3 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 9,30,112,4,3,0,0
 9,31,112,4,3,0,0
 9,32,112,4,3,0,0
-9,33,112,4,3,0,0
-9,34,112,4,3,0,0
-9,35,112,4,3,0,0
-9,36,112,4,3,0,0
-9,37,112,4,3,0,0

From e34e398b7e592418179601bae7d6f3895701c303 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 15:38:56 +0200
Subject: [PATCH 22/81] Add bert integration tests

---
 src/sequifier/config/infer_config.py          |  3 +-
 tests/configs/hyperparameter-search-bert.yaml | 76 +++++++++++++++
 ...-test-categorical-bert-autoregression.yaml | 26 ++++++
 ...infer-test-categorical-bert-embedding.yaml | 25 +++++
 .../configs/infer-test-categorical-bert.yaml  | 25 +++++
 .../configs/train-test-categorical-bert.yaml  | 68 ++++++++++++++
 tests/configs/train-test-real-bert.yaml       | 68 ++++++++++++++
 tests/integration-test-log.txt                |  7 ++
 tests/integration/conftest.py                 | 92 +++++++++++++++++++
 .../integration/test_hyperparameter_search.py | 11 +++
 tests/integration/test_inference.py           | 77 ++++++++++++++++
 tests/integration/test_training.py            | 12 +++
 12 files changed, 489 insertions(+), 1 deletion(-)
 create mode 100644 tests/configs/hyperparameter-search-bert.yaml
 create mode 100644 tests/configs/infer-test-categorical-bert-autoregression.yaml
 create mode 100644 tests/configs/infer-test-categorical-bert-embedding.yaml
 create mode 100644 tests/configs/infer-test-categorical-bert.yaml
 create mode 100644 tests/configs/train-test-categorical-bert.yaml
 create mode 100644 tests/configs/train-test-real-bert.yaml

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index d67b04c6..0c1e3915 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -222,7 +222,8 @@ def validate_autoregression(cls, v: bool, info: ValidationInfo):
             )
 
         if (
-            info.data.get("training_objective") is not None
+            v
+            and info.data.get("training_objective") is not None
             and info.data.get("training_objective") == "bert"
         ):
             raise ValueError(
diff --git a/tests/configs/hyperparameter-search-bert.yaml b/tests/configs/hyperparameter-search-bert.yaml
new file mode 100644
index 00000000..84b49cf3
--- /dev/null
+++ b/tests/configs/hyperparameter-search-bert.yaml
@@ -0,0 +1,76 @@
+project_root: tests/project_folder
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+hp_search_name: test-hp-search-bert
+model_config_write_path: configs
+
+read_format: pt
+target_columns: [itemId]
+target_column_types:
+  itemId: categorical
+seq_length: [8]
+inference_batch_size: 10
+
+search_strategy: sample
+n_samples: 1
+prune_trials: false
+
+input_columns:
+  - [itemId]
+export_embedding_model: false
+export_generative_model: true
+export_onnx: false
+export_pt: true
+export_with_dropout: false
+
+model_hyperparameter_sampling:
+  initial_embedding_dim: [16]
+  joint_embedding_dim: [null]
+  feature_embedding_dims: null
+  dim_model: [16]
+  n_head: [2]
+  dim_feedforward: [8]
+  num_layers: [1]
+  activation_fn: ["relu"]
+  normalization: ["layer_norm"]
+  positional_encoding: ["learned"]
+  attention_type: ["mha"]
+  norm_first: [false]
+  n_kv_heads: [2]
+  rope_theta: [10000.0]
+  prediction_length: 8
+
+training_hyperparameter_sampling:
+  training_objective: ["bert"]
+  bert_spec:
+    masking_probability: [0.5]
+    replacement_distribution:
+      - masked: 0.8
+        random: 0.1
+        identical: 0.1
+    span_masking:
+      - type: GeometricDistribution
+        p: 1.0
+  device: cpu
+  torch_compile: none
+  epochs: [1]
+  save_interval_epochs: 10
+  batch_size: [5]
+  learning_rate: [0.001]
+  criterion:
+    itemId: CrossEntropyLoss
+  loss_weights:
+    itemId: 1.0
+  accumulation_steps: [1]
+  dropout: [0.0]
+  optimizer:
+    - name: Adam
+  scheduler:
+    - name: StepLR
+      step_size: 1
+      gamma: 0.99
+  log_interval: 5
+  calculate_validation_loss_on_initialization: false
+  continue_training: false
+  layer_autocast: false
+
+override_input: true
diff --git a/tests/configs/infer-test-categorical-bert-autoregression.yaml b/tests/configs/infer-test-categorical-bert-autoregression.yaml
new file mode 100644
index 00000000..552bb21f
--- /dev/null
+++ b/tests/configs/infer-test-categorical-bert-autoregression.yaml
@@ -0,0 +1,26 @@
+project_root: tests/project_folder
+training_config_path: tests/configs/train-test-categorical-bert.yaml
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+model_type: generative
+training_objective: bert
+model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-1.pt
+data_path: tests/project_folder/data/test-data-categorical-1-split2
+read_format: pt
+write_format: csv
+
+input_columns: [itemId]
+target_columns: [itemId]
+target_column_types:
+  itemId: categorical
+
+output_probabilities: false
+map_to_id: true
+device: cpu
+seq_length: 8
+inference_batch_size: 10
+prediction_length: null
+enforce_determinism: true
+
+sample_from_distribution_columns: null
+autoregression: true
+autoregression_total_steps: 1
diff --git a/tests/configs/infer-test-categorical-bert-embedding.yaml b/tests/configs/infer-test-categorical-bert-embedding.yaml
new file mode 100644
index 00000000..14e95fea
--- /dev/null
+++ b/tests/configs/infer-test-categorical-bert-embedding.yaml
@@ -0,0 +1,25 @@
+project_root: tests/project_folder
+training_config_path: tests/configs/train-test-categorical-bert.yaml
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+model_type: embedding
+training_objective: bert
+model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-embedding-1.pt
+data_path: tests/project_folder/data/test-data-categorical-1-split2
+read_format: pt
+write_format: csv
+
+input_columns: [itemId]
+target_columns: [itemId]
+target_column_types:
+  itemId: categorical
+
+output_probabilities: false
+map_to_id: true
+device: cpu
+seq_length: 8
+inference_batch_size: 10
+prediction_length: null
+enforce_determinism: true
+
+sample_from_distribution_columns: null
+autoregression: false
diff --git a/tests/configs/infer-test-categorical-bert.yaml b/tests/configs/infer-test-categorical-bert.yaml
new file mode 100644
index 00000000..c6ec607f
--- /dev/null
+++ b/tests/configs/infer-test-categorical-bert.yaml
@@ -0,0 +1,25 @@
+project_root: tests/project_folder
+training_config_path: tests/configs/train-test-categorical-bert.yaml
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+model_type: generative
+training_objective: bert
+model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-1.pt
+data_path: tests/project_folder/data/test-data-categorical-1-split2
+read_format: pt
+write_format: csv
+
+input_columns: [itemId]
+target_columns: [itemId]
+target_column_types:
+  itemId: categorical
+
+output_probabilities: true
+map_to_id: true
+device: cpu
+seq_length: 8
+inference_batch_size: 10
+prediction_length: null
+enforce_determinism: true
+
+sample_from_distribution_columns: null
+autoregression: false
diff --git a/tests/configs/train-test-categorical-bert.yaml b/tests/configs/train-test-categorical-bert.yaml
new file mode 100644
index 00000000..12adcc5a
--- /dev/null
+++ b/tests/configs/train-test-categorical-bert.yaml
@@ -0,0 +1,68 @@
+project_root: tests/project_folder
+model_name: model-categorical-bert
+read_format: pt
+metadata_config_path: configs/metadata_configs/test-data-categorical-1.json
+
+input_columns: [itemId]
+target_columns: [itemId]
+target_column_types:
+  itemId: categorical
+
+seq_length: 8
+inference_batch_size: 10
+
+export_generative_model: true
+export_embedding_model: true
+export_onnx: false
+export_pt: true
+export_with_dropout: false
+
+model_spec:
+  initial_embedding_dim: 16
+  dim_model: 16
+  n_head: 2
+  dim_feedforward: 8
+  num_layers: 1
+  prediction_length: 8
+  activation_fn: relu
+  normalization: layer_norm
+  positional_encoding: learned
+  attention_type: mha
+  norm_first: false
+  n_kv_heads: 2
+  rope_theta: 10000.0
+training_spec:
+  training_objective: bert
+  bert_spec:
+    masking_probability: 0.5
+    replacement_distribution:
+      masked: 0.8
+      random: 0.1
+      identical: 0.1
+    span_masking:
+      type: GeometricDistribution
+      p: 1.0
+  device: cpu
+  torch_compile: none
+  epochs: 1
+  save_interval_epochs: 1
+  batch_size: 5
+  log_interval: 1
+  learning_rate: 0.001
+  accumulation_steps: null
+  dropout: 0.0
+  criterion:
+    itemId: CrossEntropyLoss
+  loss_weights:
+    itemId: 1.0
+  optimizer:
+    name: Adam
+  scheduler:
+    name: StepLR
+    step_size: 1.0
+    gamma: 0.99
+  scheduler_step_on: epoch
+  calculate_validation_loss_on_initialization: false
+  continue_training: false
+  enforce_determinism: true
+  layer_autocast: false
diff --git a/tests/configs/train-test-real-bert.yaml b/tests/configs/train-test-real-bert.yaml
new file mode 100644
index 00000000..426a129c
--- /dev/null
+++ b/tests/configs/train-test-real-bert.yaml
@@ -0,0 +1,68 @@
+project_root: tests/project_folder
+model_name: model-real-bert
+read_format: parquet
+metadata_config_path: configs/metadata_configs/test-data-real-1.json
+
+input_columns: [itemValue]
+target_columns: [itemValue]
+target_column_types:
+  itemValue: real
+
+seq_length: 8
+inference_batch_size: 10
+
+export_generative_model: true
+export_embedding_model: false
+export_onnx: false
+export_pt: true
+export_with_dropout: false
+
+model_spec:
+  initial_embedding_dim: 16
+  dim_model: 16
+  n_head: 2
+  dim_feedforward: 8
+  num_layers: 1
+  prediction_length: 8
+  activation_fn: relu
+  normalization: layer_norm
+  positional_encoding: learned
+  attention_type: mha
+  norm_first: false
+  n_kv_heads: 2
+  rope_theta: 10000.0
+training_spec:
+  training_objective: bert
+  bert_spec:
+    masking_probability: 0.5
+    replacement_distribution:
+      masked: 0.8
+      random: 0.1
+      identical: 0.1
+    span_masking:
+      type: GeometricDistribution
+      p: 1.0
+  device: cpu
+  torch_compile: none
+  epochs: 1
+  save_interval_epochs: 1
+  batch_size: 5
+  log_interval: 1
+  learning_rate: 0.001
+  accumulation_steps: null
+  dropout: 0.0
+  criterion:
+    itemValue: MSELoss
+  loss_weights:
+    itemValue: 1.0
+  optimizer:
+    name: Adam
+  scheduler:
+    name: StepLR
+    step_size: 1.0
+    gamma: 0.99
+  scheduler_step_on: epoch
+  calculate_validation_loss_on_initialization: false
+  continue_training: false
+  enforce_determinism: true
+  layer_autocast: false
diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
index ef9fa309..e44783c3 100644
--- a/tests/integration-test-log.txt
+++ b/tests/integration-test-log.txt
@@ -13,6 +13,7 @@ sequifier preprocess --config-path tests/configs/preprocess-test-categorical-exa
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-exact-pt.yaml
 sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-grid.yaml
 sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-sample.yaml
+sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-bert.yaml
 sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-bayesian.yaml
 sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-custom-eval.yaml
 sequifier train --config-path tests/configs/train-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-1.json --model-name model-categorical-1 --input-columns itemId
@@ -25,6 +26,8 @@ sequifier train --config-path tests/configs/train-test-categorical.yaml --metada
 sequifier train --config-path tests/configs/train-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-50.json --model-name model-real-50 --input-columns None
 sequifier train --config-path tests/configs/train-test-categorical-inf-size-1.yaml
 sequifier train --config-path tests/configs/train-test-categorical-inf-size-3.yaml
+sequifier train --config-path tests/configs/train-test-categorical-bert.yaml
+sequifier train --config-path tests/configs/train-test-real-bert.yaml
 sequifier train --config-path tests/configs/train-test-categorical-multitarget.yaml
 sequifier train --config-path tests/configs/train-test-categorical-multitarget-eager.yaml
 sequifier train --config-path tests/configs/train-test-distributed.yaml
@@ -48,6 +51,8 @@ sequifier infer --config-path tests/configs/infer-test-lazy.yaml
 sequifier infer --config-path tests/configs/infer-test-categorical-autoregression.yaml  --input-columns itemId
 sequifier infer --config-path tests/configs/infer-test-categorical-embedding.yaml  --input-columns itemId
 sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+sequifier infer --config-path tests/configs/infer-test-categorical-bert.yaml
+sequifier infer --config-path tests/configs/infer-test-categorical-bert-embedding.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
 sequifier train --config-path tests/configs/train-test-resume-epoch.yaml
@@ -59,6 +64,7 @@ sequifier visualize-training model-categorical-3-from-mid-epoch-checkpoint --pro
 sequifier visualize-training model-categorical-3-inf-size --project-root tests/project_folder
 sequifier visualize-training model-categorical-5 --project-root tests/project_folder
 sequifier visualize-training model-categorical-50 --project-root tests/project_folder
+sequifier visualize-training model-categorical-bert --project-root tests/project_folder
 sequifier visualize-training model-categorical-distributed --project-root tests/project_folder
 sequifier visualize-training model-categorical-distributed-lazy-parquet --project-root tests/project_folder
 sequifier visualize-training model-categorical-lazy --project-root tests/project_folder
@@ -69,4 +75,5 @@ sequifier visualize-training model-real-1-from-epoch-checkpoint --project-root t
 sequifier visualize-training model-real-3 --project-root tests/project_folder
 sequifier visualize-training model-real-5 --project-root tests/project_folder
 sequifier visualize-training model-real-50 --project-root tests/project_folder
+sequifier visualize-training model-real-bert --project-root tests/project_folder
 sequifier visualize-training test-hp-search-grid-run-0,test-hp-search-grid-run-1,test-hp-search-grid-run-2,test-hp-search-grid-run-3 --project-root tests/project_folder --log-scale --bucket-training-batches 5
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index 9b193a7b..c634ee90 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -133,6 +133,16 @@ def training_config_path_cat_inf_size_3():
     return os.path.join("tests", "configs", "train-test-categorical-inf-size-3.yaml")
 
 
+@pytest.fixture(scope="session")
+def training_config_path_cat_bert():
+    return os.path.join("tests", "configs", "train-test-categorical-bert.yaml")
+
+
+@pytest.fixture(scope="session")
+def training_config_path_real_bert():
+    return os.path.join("tests", "configs", "train-test-real-bert.yaml")
+
+
 @pytest.fixture(scope="session")
 def training_config_path_distributed():
     return os.path.join("tests", "configs", "train-test-distributed.yaml")
@@ -183,6 +193,25 @@ def inference_config_path_cat_inf_size_3():
     return os.path.join("tests", "configs", "infer-test-categorical-inf-size-3.yaml")
 
 
+@pytest.fixture(scope="session")
+def inference_config_path_cat_bert():
+    return os.path.join("tests", "configs", "infer-test-categorical-bert.yaml")
+
+
+@pytest.fixture(scope="session")
+def inference_config_path_cat_bert_embedding():
+    return os.path.join(
+        "tests", "configs", "infer-test-categorical-bert-embedding.yaml"
+    )
+
+
+@pytest.fixture(scope="session")
+def inference_config_path_cat_bert_autoregression():
+    return os.path.join(
+        "tests", "configs", "infer-test-categorical-bert-autoregression.yaml"
+    )
+
+
 @pytest.fixture(scope="session")
 def inference_config_path_real_autoregression():
     return os.path.join("tests", "configs", "infer-test-real-autoregression.yaml")
@@ -240,6 +269,7 @@ def hp_search_configs():
     return {
         "grid": os.path.join("tests", "configs", "hyperparameter-search-grid.yaml"),
         "sample": os.path.join("tests", "configs", "hyperparameter-search-sample.yaml"),
+        "bert": os.path.join("tests", "configs", "hyperparameter-search-bert.yaml"),
         "bayesian": os.path.join(
             "tests", "configs", "hyperparameter-search-bayesian.yaml"
         ),
@@ -271,6 +301,8 @@ def format_configs_locally(
     training_config_path_real,
     training_config_path_cat_inf_size_1,
     training_config_path_cat_inf_size_3,
+    training_config_path_cat_bert,
+    training_config_path_real_bert,
     training_config_path_distributed,
     training_config_path_distributed_lazy_parquet,
     training_config_path_lazy,
@@ -284,6 +316,9 @@ def format_configs_locally(
     inference_config_path_categorical_autoregression,
     inference_config_path_cat_inf_size_1,
     inference_config_path_cat_inf_size_3,
+    inference_config_path_cat_bert,
+    inference_config_path_cat_bert_embedding,
+    inference_config_path_cat_bert_autoregression,
     inference_config_path_distributed,
     inference_config_path_distributed_parquet,
     inference_config_path_lazy,
@@ -303,6 +338,8 @@ def format_configs_locally(
             training_config_path_real,
             training_config_path_cat_inf_size_1,
             training_config_path_cat_inf_size_3,
+            training_config_path_cat_bert,
+            training_config_path_real_bert,
             training_config_path_distributed,
             training_config_path_distributed_lazy_parquet,
             training_config_path_lazy,
@@ -316,11 +353,15 @@ def format_configs_locally(
             inference_config_path_categorical_autoregression,
             inference_config_path_cat_inf_size_1,
             inference_config_path_cat_inf_size_3,
+            inference_config_path_cat_bert,
+            inference_config_path_cat_bert_embedding,
+            inference_config_path_cat_bert_autoregression,
             inference_config_path_distributed,
             inference_config_path_distributed_parquet,
             inference_config_path_lazy,
             hp_search_configs["grid"],
             hp_search_configs["sample"],
+            hp_search_configs["bert"],
             hp_search_configs["bayesian"],
             hp_search_configs["custom-eval"],
         ]
@@ -464,6 +505,8 @@ def run_training(
     training_config_path_real,
     training_config_path_cat_inf_size_1,
     training_config_path_cat_inf_size_3,
+    training_config_path_cat_bert,
+    training_config_path_real_bert,
     training_config_path_distributed,
     training_config_path_distributed_lazy_parquet,
     training_config_path_lazy,
@@ -491,6 +534,10 @@ def run_training(
 
     run_and_log(f"sequifier train --config-path {training_config_path_cat_inf_size_3}")
 
+    run_and_log(f"sequifier train --config-path {training_config_path_cat_bert}")
+
+    run_and_log(f"sequifier train --config-path {training_config_path_real_bert}")
+
     run_and_log(f"sequifier train --config-path {training_config_path_cat_multitarget}")
 
     run_and_log(
@@ -559,6 +606,10 @@ def run_hp_search(
         f"sequifier hyperparameter-search --config-path {hp_search_configs['sample']}"
     )
 
+    run_and_log(
+        f"sequifier hyperparameter-search --config-path {hp_search_configs['bert']}"
+    )
+
     run_and_log(
         f"sequifier hyperparameter-search --config-path {hp_search_configs['bayesian']}"
     )
@@ -594,6 +645,8 @@ def run_inference(
     inference_config_path_embedding,
     inference_config_path_cat_inf_size_1,
     inference_config_path_cat_inf_size_3,
+    inference_config_path_cat_bert,
+    inference_config_path_cat_bert_embedding,
     inference_config_path_distributed,
     inference_config_path_distributed_parquet,
     inference_config_path_lazy,
@@ -658,6 +711,12 @@ def run_inference(
         f"sequifier infer --config-path {inference_config_path_cat_inf_size_3_embedding}"
     )
 
+    run_and_log(f"sequifier infer --config-path {inference_config_path_cat_bert}")
+
+    run_and_log(
+        f"sequifier infer --config-path {inference_config_path_cat_bert_embedding}"
+    )
+
 
 @pytest.fixture()
 def model_names_preds():
@@ -813,3 +872,36 @@ def embeddings(run_inference, model_names_embeddings, project_root):
             embeddings_path, "categorical", "csv"
         )
     return embeds
+
+
+@pytest.fixture()
+def bert_predictions(run_inference, project_root):
+    prediction_path = os.path.join(
+        project_root,
+        "outputs",
+        "predictions",
+        "sequifier-model-categorical-bert-best-1-predictions",
+    )
+    return read_multi_file_preds(prediction_path, "categorical")
+
+
+@pytest.fixture()
+def bert_probabilities(run_inference, project_root):
+    probabilities_path = os.path.join(
+        project_root,
+        "outputs",
+        "probabilities",
+        "sequifier-model-categorical-bert-best-1-itemId-probabilities",
+    )
+    return read_multi_file_preds(probabilities_path, "categorical", "csv")
+
+
+@pytest.fixture()
+def bert_embeddings(run_inference, project_root):
+    embeddings_path = os.path.join(
+        project_root,
+        "outputs",
+        "embeddings",
+        "sequifier-model-categorical-bert-best-embedding-1-embeddings",
+    )
+    return read_multi_file_preds(embeddings_path, "categorical", "csv")
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index 00f42d25..187c8312 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -23,6 +23,16 @@ def test_hp_search_sample_outputs(run_hp_search, project_root):
     ), f"Expected 4 sample configs, found {len(generated_configs)}"
 
 
+def test_hp_search_bert_outputs(run_hp_search, project_root):
+    hp_name = "test-hp-search-bert"
+    config_dir = os.path.join(project_root, "configs")
+
+    generated_configs = glob.glob(os.path.join(config_dir, f"{hp_name}-run-*.yaml"))
+    assert (
+        len(generated_configs) == 1
+    ), f"Expected 1 BERT sample config, found {len(generated_configs)}"
+
+
 def test_hp_search_bayesian_outputs(run_hp_search, project_root):
     hp_name = "test-hp-search-bayesian"
     config_dir = os.path.join(project_root, "configs")
@@ -38,6 +48,7 @@ def test_hp_search_state(run_hp_search, project_root):
 
     assert os.path.exists(os.path.join(state_dir, "test-hp-search-sample.db"))
     assert os.path.exists(os.path.join(state_dir, "test-hp-search-grid.db"))
+    assert os.path.exists(os.path.join(state_dir, "test-hp-search-bert.db"))
     assert os.path.exists(os.path.join(state_dir, "test-hp-search-bayesian.db"))
     assert os.path.exists(os.path.join(state_dir, "test-hp-search-custom-eval.db"))
 
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 52bd0c7c..99f82f61 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -1,7 +1,21 @@
+import json
+import os
+import subprocess
+
 import numpy as np
 import polars as pl
 
 TARGET_VARIABLE_DICT = {"categorical": "itemId", "real": "itemValue"}
+BERT_SEQ_LENGTH = 8
+BERT_EMBEDDING_DIM = 16
+
+
+def _categorical_metadata(project_root):
+    metadata_path = os.path.join(
+        project_root, "configs", "metadata_configs", "test-data-categorical-1.json"
+    )
+    with open(metadata_path, "r") as f:
+        return json.load(f)
 
 
 def test_predictions_real(predictions):
@@ -97,6 +111,32 @@ def test_probabilities(probabilities):
         )
 
 
+def test_bert_generative_predictions_default_to_seq_length(
+    bert_predictions, project_root
+):
+    metadata = _categorical_metadata(project_root)
+    valid_values = {str(v) for v in metadata["id_maps"]["itemId"].keys()}
+
+    assert bert_predictions.height > 0
+    assert set(bert_predictions["itemId"].to_list()).issubset(valid_values)
+
+    rows_per_sequence = bert_predictions.group_by("sequenceId").len()
+    assert (rows_per_sequence["len"] == BERT_SEQ_LENGTH).all(), rows_per_sequence
+
+
+def test_bert_probabilities(bert_predictions, bert_probabilities, project_root):
+    metadata = _categorical_metadata(project_root)
+    expected_class_count = metadata["n_classes"]["itemId"]
+
+    assert bert_probabilities.height == bert_predictions.height
+    assert bert_probabilities.shape[1] == expected_class_count
+    np.testing.assert_almost_equal(
+        bert_probabilities.sum_horizontal(),
+        np.ones(bert_probabilities.shape[0]),
+        decimal=5,
+    )
+
+
 def test_multi_pred(predictions):
     multitarget_models = [name for name in predictions.keys() if "multitarget" in name]
 
@@ -132,6 +172,43 @@ def test_embeddings(embeddings):
             assert np.abs(model_embeddings[:, 1:].to_numpy().mean()) < 0.3
 
 
+def test_bert_embeddings(bert_predictions, bert_embeddings):
+    expected_embedding_cols = [str(i) for i in range(BERT_EMBEDDING_DIM)]
+    expected_columns = ["sequenceId", "subsequenceId", "itemPosition"]
+
+    assert bert_embeddings.height == bert_predictions.height
+    assert bert_embeddings.shape[1] == len(expected_columns) + BERT_EMBEDDING_DIM
+    assert all(col in bert_embeddings.columns for col in expected_columns)
+    assert all(col in bert_embeddings.columns for col in expected_embedding_cols)
+
+    embedding_values = bert_embeddings.select(expected_embedding_cols).to_numpy()
+    assert np.isfinite(embedding_values).all()
+
+    rows_per_sequence = bert_embeddings.group_by("sequenceId").len()
+    assert (rows_per_sequence["len"] == BERT_SEQ_LENGTH).all(), rows_per_sequence
+
+
+def test_bert_rejects_autoregression_end_to_end(
+    run_training, inference_config_path_cat_bert_autoregression
+):
+    result = subprocess.run(
+        [
+            "sequifier",
+            "infer",
+            "--config-path",
+            inference_config_path_cat_bert_autoregression,
+        ],
+        capture_output=True,
+        text=True,
+    )
+
+    assert result.returncode != 0
+    assert (
+        "Autoregressive inference is not possible with BERT-style models."
+        in result.stdout + result.stderr
+    )
+
+
 def test_predictions_item_position(predictions):
     """
     Checks if itemPosition increments correctly within each sequenceId.
diff --git a/tests/integration/test_training.py b/tests/integration/test_training.py
index 1fb9fc51..65128d30 100644
--- a/tests/integration/test_training.py
+++ b/tests/integration/test_training.py
@@ -33,6 +33,8 @@ def test_checkpoint_files_exists(
                 for j in [1, 3]
                 for i in range(1, 4)
             ]
+            + ["model-categorical-bert-epoch-1.pt"]
+            + ["model-real-bert-epoch-1.pt"]
             + [f"model-categorical-multitarget-5-epoch-{i}.pt" for i in range(1, 4)]
             + [
                 f"model-categorical-multitarget-5-eager-epoch-{i}.pt"
@@ -107,6 +109,12 @@ def test_model_files_exists(run_training, run_training_from_checkpoint, project_
                 "sequifier-model-categorical-3-inf-size-last-3.pt",
                 "sequifier-model-categorical-3-inf-size-best-embedding-3.pt",
                 "sequifier-model-categorical-3-inf-size-last-embedding-3.pt",
+                "sequifier-model-categorical-bert-best-1.pt",
+                "sequifier-model-categorical-bert-last-1.pt",
+                "sequifier-model-categorical-bert-best-embedding-1.pt",
+                "sequifier-model-categorical-bert-last-embedding-1.pt",
+                "sequifier-model-real-bert-best-1.pt",
+                "sequifier-model-real-bert-last-1.pt",
                 "sequifier-model-categorical-distributed-best-3.pt",
                 "sequifier-model-categorical-distributed-best-3.onnx",
                 "sequifier-model-categorical-distributed-last-3.pt",
@@ -125,6 +133,10 @@ def test_model_files_exists(run_training, run_training_from_checkpoint, project_
                 for i in range(4)
                 for suffix in ["best", "last"]
             ]
+            + [
+                f"sequifier-test-hp-search-bert-run-0-{suffix}-1.pt"
+                for suffix in ["best", "last"]
+            ]
             + [
                 f"sequifier-test-hp-search-grid-run-{i}-{suffix}-2.pt"
                 for i in range(4)

From c2ff935cb589b84fed8979f353885f8a2acb7c0a Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 16:39:13 +0200
Subject: [PATCH 23/81] WIP

---
 src/sequifier/config/infer_config.py |  21 +++-
 src/sequifier/config/train_config.py |  30 +++--
 src/sequifier/infer.py               | 157 +++++++++++++++++++++++++--
 src/sequifier/preprocess.py          | 102 +++++++++++++++--
 src/sequifier/train.py               |  19 +++-
 tests/unit/test_infer.py             | 111 +++++++++++++++++++
 tests/unit/test_preprocess.py        |  66 +++++++++++
 tests/unit/test_train.py             |  73 ++++++++++++-
 8 files changed, 538 insertions(+), 41 deletions(-)

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 0c1e3915..668252b2 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -5,7 +5,14 @@
 import numpy as np
 import yaml
 from beartype import beartype
-from pydantic import BaseModel, ConfigDict, Field, ValidationInfo, field_validator
+from pydantic import (
+    BaseModel,
+    ConfigDict,
+    Field,
+    ValidationInfo,
+    field_validator,
+    model_validator,
+)
 
 from sequifier.helpers import normalize_path, try_catch_excess_keys
 
@@ -138,6 +145,18 @@ class InfererModel(BaseModel):
     autoregression: bool = Field(default=False)
     autoregression_total_steps: Optional[int] = Field(default=None)
 
+    @model_validator(mode="after")
+    def validate_bert_prediction_length_matches_seq_length(self):
+        if self.training_objective == "bert":
+            if self.prediction_length is None:
+                self.prediction_length = self.seq_length
+            elif self.prediction_length != self.seq_length:
+                raise ValueError(
+                    "For BERT inference, prediction_length must be equal to seq_length "
+                    f"(got prediction_length={self.prediction_length}, seq_length={self.seq_length})."
+                )
+        return self
+
     @field_validator("training_objective")
     @classmethod
     def validate_training_objective(cls, v):
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 92117551..477a602c 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -119,9 +119,9 @@ def __setstate__(self, state):
 
 
 class ReplacementDistribution(BaseModel):
-    masked: float = Field(..., gt=0.0, le=1.0)
-    random: float = Field(..., gt=0.0, le=1.0)
-    identical: float = Field(..., gt=0.0, le=1.0)
+    masked: float = Field(..., ge=0.0, le=1.0)
+    random: float = Field(..., ge=0.0, le=1.0)
+    identical: float = Field(..., ge=0.0, le=1.0)
 
     @model_validator(mode="after")
     def validate_sum(self):
@@ -341,19 +341,17 @@ def validate_training_objective(cls, v):
             raise ValueError(f"Only 'causal' and 'bert' are allowed, found {v}")
         return v
 
-    @field_validator("bert_spec")
-    @classmethod
-    def validate_bert_spec(cls, v, info):
-        training_objective = info.data.get("training_objective")
-        if v and not training_objective == "bert":
+    @model_validator(mode="after")
+    def validate_bert_spec_matches_objective(self):
+        if self.bert_spec is not None and self.training_objective != "bert":
             raise ValueError(
                 "The BERT hyperparameters should only be configured if the training objective is 'bert'"
             )
-        if not v and training_objective == "bert":
+        if self.bert_spec is None and self.training_objective == "bert":
             raise ValueError(
                 "If the training_objective is 'bert', the BERT hyperparameters must be set"
             )
-        return v
+        return self
 
     @field_validator("sampling_strategy")
     @classmethod
@@ -568,6 +566,18 @@ class TrainModel(BaseModel):
     model_spec: ModelSpecModel
     training_spec: TrainingSpecModel
 
+    @model_validator(mode="after")
+    def validate_bert_prediction_length_matches_seq_length(self):
+        if (
+            self.training_spec.training_objective == "bert"
+            and self.model_spec.prediction_length != self.seq_length
+        ):
+            raise ValueError(
+                "For BERT training, model_spec.prediction_length must be equal to seq_length "
+                f"(got prediction_length={self.model_spec.prediction_length}, seq_length={self.seq_length})."
+            )
+        return self
+
     @field_validator("model_name")
     @classmethod
     def validate_model_name(cls, v):
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 1777cc4a..24b250ea 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -18,6 +18,7 @@
     configure_determinism,
     construct_index_maps,
     generate_padding_masks,
+    get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
@@ -296,6 +297,74 @@ def calculate_item_positions(
     return repeated_bases + tiled_offsets
 
 
+def _flatten_prediction_valid_mask(
+    metadata: Optional[dict[str, Any]], prediction_length: int
+) -> Optional[np.ndarray]:
+    if metadata is None or "target_valid_mask" not in metadata:
+        return None
+
+    mask = metadata["target_valid_mask"]
+    mask_np = mask.cpu().numpy() if isinstance(mask, torch.Tensor) else np.asarray(mask)
+    return mask_np[:, -prediction_length:].reshape(-1).astype(bool)
+
+
+def _prediction_valid_mask_from_frame(
+    config: "InfererModel", data: pl.DataFrame, prediction_length: int
+) -> Optional[np.ndarray]:
+    if config.training_objective != "bert":
+        return None
+
+    left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
+    if left_pad_lengths is None:
+        return None
+
+    metadata = generate_padding_masks(
+        left_pad_lengths,
+        config.seq_length,
+        data_offset=1,
+        target_offset=1,
+    )
+    return _flatten_prediction_valid_mask(metadata, prediction_length)
+
+
+def _filter_outputs_to_valid_targets(
+    prediction_valid_mask: Optional[np.ndarray],
+    sequence_ids_for_preds: np.ndarray,
+    item_positions_for_preds: np.ndarray,
+    preds: dict[str, np.ndarray],
+    probs: Optional[dict[str, np.ndarray]],
+) -> tuple[
+    np.ndarray, np.ndarray, dict[str, np.ndarray], Optional[dict[str, np.ndarray]]
+]:
+    if prediction_valid_mask is None:
+        return sequence_ids_for_preds, item_positions_for_preds, preds, probs
+
+    if prediction_valid_mask.shape[0] != len(sequence_ids_for_preds):
+        raise ValueError(
+            "target_valid_mask length does not match prediction metadata length "
+            f"({prediction_valid_mask.shape[0]} != {len(sequence_ids_for_preds)})."
+        )
+
+    filtered_preds = {
+        target_column: np.asarray(values)[prediction_valid_mask]
+        for target_column, values in preds.items()
+    }
+    filtered_probs = (
+        {
+            target_column: np.asarray(values)[prediction_valid_mask]
+            for target_column, values in probs.items()
+        }
+        if probs is not None
+        else None
+    )
+    return (
+        np.asarray(sequence_ids_for_preds)[prediction_valid_mask],
+        np.asarray(item_positions_for_preds)[prediction_valid_mask],
+        filtered_preds,
+        filtered_probs,
+    )
+
+
 @beartype
 def infer_embedding(
     config: "InfererModel",
@@ -502,6 +571,7 @@ def infer_generative(
         is_folder_input = os.path.isdir(
             normalize_path(config.data_path, config.project_root)
         )
+        prediction_valid_mask = None
 
         if config.read_format in ["parquet", "csv"] and not is_folder_input:
             if config.input_columns is not None:
@@ -521,6 +591,9 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
+                prediction_valid_mask = _prediction_valid_mask_from_frame(
+                    config, data, prediction_length
+                )
 
                 # Expand IDs to match model output shape
                 sequence_ids_for_preds = np.repeat(
@@ -570,6 +643,9 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
+                prediction_valid_mask = _prediction_valid_mask_from_frame(
+                    config, data, prediction_length
+                )
 
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
@@ -637,6 +713,9 @@ def infer_generative(
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
                 )
+                prediction_valid_mask = _flatten_prediction_valid_mask(
+                    metadata, prediction_length
+                )
 
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
@@ -674,6 +753,16 @@ def infer_generative(
                     predictions, target_column
                 )
 
+        sequence_ids_for_preds, item_positions_for_preds, preds, probs = (
+            _filter_outputs_to_valid_targets(
+                prediction_valid_mask,
+                sequence_ids_for_preds,
+                item_positions_for_preds,
+                preds,
+                probs,
+            )
+        )
+
         os.makedirs(
             os.path.join(config.project_root, "outputs", "predictions"),
             exist_ok=True,
@@ -1417,8 +1506,14 @@ def adjust_and_infer_embedding(
         if self.inference_model_type == "onnx":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=True)
+            metadata_adjusted = (
+                self.prepare_inference_batches(metadata, pad_to_batch_size=True)
+                if metadata
+                else [None] * len(x_adjusted)
+            )
             inference_batch_embeddings = [
-                self.infer_pure(x_sub)[0] for x_sub in x_adjusted
+                self.infer_pure(x_sub, metadata_sub)[0]
+                for x_sub, metadata_sub in zip(x_adjusted, metadata_adjusted)
             ]
             embeddings = np.concatenate(inference_batch_embeddings, axis=0)[:size]
         elif self.inference_model_type == "pt":
@@ -1466,9 +1561,14 @@ def adjust_and_infer_generative(
         if self.inference_model_type == "onnx":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=True)
+            metadata_adjusted = (
+                self.prepare_inference_batches(metadata, pad_to_batch_size=True)
+                if metadata
+                else [None] * len(x_adjusted)
+            )
             out_subs = [
-                dict(zip(self.target_columns, self.infer_pure(x_sub)))
-                for x_sub in x_adjusted
+                dict(zip(self.target_columns, self.infer_pure(x_sub, metadata_sub)))
+                for x_sub, metadata_sub in zip(x_adjusted, metadata_adjusted)
             ]
             outs = {
                 target_column: np.concatenate(
@@ -1544,7 +1644,11 @@ def prepare_inference_batches(
             return xs
 
     @beartype
-    def infer_pure(self, x: dict[str, np.ndarray]) -> list[np.ndarray]:
+    def infer_pure(
+        self,
+        x: dict[str, np.ndarray],
+        metadata: Optional[dict[str, np.ndarray]] = None,
+    ) -> list[np.ndarray]:
         """Performs a single inference pass using the ONNX session.
 
         This function assumes `x` is already a single, correctly-sized
@@ -1559,13 +1663,44 @@ def infer_pure(self, x: dict[str, np.ndarray]) -> list[np.ndarray]:
             A list of NumPy arrays, representing the raw outputs from the
             ONNX model.
         """
-        ort_inputs = {
-            session_input.name: self.expand_to_batch_size(x[col])
-            for session_input, col in zip(
-                self.ort_session.get_inputs(),
-                self.categorical_columns + self.real_columns,
-            )
-        }
+        metadata = metadata or {}
+        fallback_columns = sorted(x.keys())
+        used_feature_columns = set()
+        reference_shape = next(iter(x.values())).shape
+        ort_inputs = {}
+        for session_input in self.ort_session.get_inputs():
+            input_name = session_input.name
+            if input_name == "attention_valid_mask":
+                value = metadata.get(input_name)
+                if value is None:
+                    value = np.ones(reference_shape[:2], dtype=np.bool_)
+            elif input_name in metadata:
+                value = metadata[input_name]
+            elif input_name.endswith("_in") and input_name[:-3] in x:
+                feature_column = input_name[:-3]
+                value = x[feature_column]
+                used_feature_columns.add(feature_column)
+            elif input_name in x:
+                value = x[input_name]
+                used_feature_columns.add(input_name)
+            else:
+                fallback_column = next(
+                    (
+                        column
+                        for column in fallback_columns
+                        if column not in used_feature_columns
+                    ),
+                    None,
+                )
+                if fallback_column is None:
+                    raise ValueError(
+                        f"Could not map ONNX input '{input_name}' to a feature or metadata array."
+                    )
+                value = x[fallback_column]
+                used_feature_columns.add(fallback_column)
+
+            ort_inputs[input_name] = self.expand_to_batch_size(value)
+
         ort_outs = self.ort_session.run(None, ort_inputs)
         return [
             oo.transpose(1, 0, 2).reshape(oo.shape[0] * oo.shape[1], oo.shape[2])
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index a9401479..5888ada4 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -26,6 +26,8 @@
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
 CATEGORICAL_MASK_ID = 2
 REAL_MASK_VALUE = 0.0
+RESERVED_ID_KEYS = {"[unknown]", "[other]", "[mask]"}
+RESERVED_ID_VALUES = {0, 1, 2}
 
 
 @beartype
@@ -815,7 +817,10 @@ def _export_metadata(
                 for split_path in self.split_paths
             ],
             "column_types": col_types,
-            "selected_columns_statistics": selected_columns_statistics,
+            "selected_columns_statistics": {
+                col: {"mean": stats["mean"], "std": stats["std"]}
+                for col, stats in selected_columns_statistics.items()
+            },
         }
         os.makedirs(
             os.path.join(self.project_root, "configs", "metadata_configs"),
@@ -1048,6 +1053,17 @@ def _apply_column_statistics(
     if col_types is None:
         col_types = {col: str(data.schema[col]) for col in data_columns}
 
+    missing_columns = [
+        col
+        for col in data_columns
+        if col not in id_maps and col not in selected_columns_statistics
+    ]
+    if missing_columns:
+        raise ValueError(
+            "No unmasked examples found for columns: "
+            f"{missing_columns}. Check the mask column or provide precomputed metadata."
+        )
+
     for col in data_columns:
         if col in id_maps:
             data = data.with_columns(pl.col(col).replace(id_maps[col], default=1))
@@ -1103,11 +1119,53 @@ def load_precomputed_id_maps(
 
                     if not len(m) > 0:
                         raise ValueError(f"map in {file} does not contain any values")
-                    min_val = min(m.values())
+                    for reserved_key, expected_value in {
+                        "[unknown]": 0,
+                        "[other]": 1,
+                        "[mask]": 2,
+                    }.items():
+                        if reserved_key in m and m[reserved_key] != expected_value:
+                            raise ValueError(
+                                f"{reserved_key} in map {file} must map to {expected_value}"
+                            )
+
+                    user_values = [
+                        value for key, value in m.items() if key not in RESERVED_ID_KEYS
+                    ]
+                    if not user_values:
+                        raise ValueError(
+                            f"map in {file} does not contain any non-reserved values"
+                        )
+
+                    min_val = min(user_values)
+                    if min_val == 2:
+                        warnings.warn(
+                            f"Precomputed map {file} uses legacy user IDs starting at 2; shifting user IDs by 1 to reserve the BERT mask ID.",
+                            stacklevel=2,
+                        )
+                        m = {
+                            key: value + 1
+                            if key not in RESERVED_ID_KEYS and value >= 2
+                            else value
+                            for key, value in m.items()
+                        }
+                        user_values = [
+                            value
+                            for key, value in m.items()
+                            if key not in RESERVED_ID_KEYS
+                        ]
+                        min_val = min(user_values)
+
                     if min_val != 3:
                         raise ValueError(
-                            f"minimum value in map {file} is {min_val}, must be 3."
+                            f"minimum non-reserved value in map {file} is {min_val}, must be 3."
+                        )
+                    if any(value in RESERVED_ID_VALUES for value in user_values):
+                        raise ValueError(
+                            f"non-reserved values in map {file} must not use reserved IDs {RESERVED_ID_VALUES}"
                         )
+                    if len(set(m.values())) != len(m.values()):
+                        raise ValueError(f"map in {file} contains duplicate IDs")
                     custom_maps[col_name] = m
     if required_maps:
         missing_maps = [col for col in required_maps if col not in custom_maps]
@@ -1169,6 +1227,9 @@ def _get_column_statistics(
         mask_expr = _mask_column_expr(data.schema[mask_column], mask_column)
         data = data.filter(~mask_expr)
 
+    if data.is_empty():
+        return id_maps, selected_columns_statistics
+
     for data_col in data_columns:
         dtype = data.schema[data_col]
         if isinstance(
@@ -1197,18 +1258,29 @@ def _get_column_statistics(
                     f"Column {data_col} is not categorical, precomputed map is invalid."
                 )
 
-            combined_mean, combined_std = get_combined_statistics(
-                data.shape[0],
-                data.get_column(data_col).mean(),
-                data.get_column(data_col).std(),
-                n_rows_running_count,
-                selected_columns_statistics.get(data_col, {"mean": 0.0})["mean"],
-                selected_columns_statistics.get(data_col, {"std": 0.0})["std"],
-            )
+            chunk_mean = data.get_column(data_col).mean()
+            chunk_std = data.get_column(data_col).std() or 0.0
+            previous_stats = selected_columns_statistics.get(data_col)
+
+            if previous_stats is None:
+                combined_mean, combined_std = chunk_mean, chunk_std
+                combined_count = data.shape[0]
+            else:
+                previous_count = int(previous_stats.get("count", n_rows_running_count))
+                combined_mean, combined_std = get_combined_statistics(
+                    data.shape[0],
+                    chunk_mean,
+                    chunk_std,
+                    previous_count,
+                    previous_stats["mean"],
+                    previous_stats["std"],
+                )
+                combined_count = previous_count + data.shape[0]
 
             selected_columns_statistics[data_col] = {
                 "std": combined_std,
                 "mean": combined_mean,
+                "count": float(combined_count),
             }
         else:
             raise ValueError(f"Column {data_col} has unsupported dtype: {dtype}")
@@ -1577,9 +1649,17 @@ def get_combined_statistics(
         A tuple `(combined_mean, combined_std)` containing the combined
         mean and standard deviation of the two subsets.
     """
+    if n1 == 0:
+        return mean2, std2
+    if n2 == 0:
+        return mean1, std1
+
     # Step 1: Calculate the combined mean.
     combined_mean = (n1 * mean1 + n2 * mean2) / (n1 + n2)
 
+    if n1 + n2 <= 1:
+        return combined_mean, 0.0
+
     # Step 2: Calculate the pooled sum of squared differences.
     # This includes the internal variance of each subset and the variance
     # between the subset mean and the combined mean.
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 29e8121c..b6230c98 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -2099,14 +2099,21 @@ def _export_model(
             }
 
             input_dict = {**x_cat, **x_real}
+            metadata_dict = {
+                "attention_valid_mask": torch.ones(
+                    self.inference_batch_size,
+                    self.seq_length,
+                    dtype=torch.bool,
+                    device=export_device,
+                )
+            }
 
-            # Wrap in a tuple with an empty dict to prevent PyTorch from treating input_dict as kwargs
-            x = (input_dict, {})
+            # Wrap in a tuple to prevent PyTorch from treating dictionaries as kwargs.
+            x = (input_dict, metadata_dict)
 
-            # PyTree flattening sorts dictionary keys automatically, so we sort names to match
-            input_names = [
-                f"{col}_in" if col in sorted(list(input_dict.keys())) else col
-                for col in sorted(self.target_columns)
+            # PyTree flattening sorts dictionary keys automatically, so names must match that order.
+            input_names = [f"{col}_in" for col in sorted(input_dict.keys())] + [
+                "attention_valid_mask"
             ]
 
             # Determine output names based on the model type
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 998dbab8..7069cbb7 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -7,6 +7,7 @@
 from sequifier.config.infer_config import InfererModel
 from sequifier.infer import (
     Inferer,
+    _filter_outputs_to_valid_targets,
     get_probs_preds_from_dict,
     normalize,
     sample_with_cumsum,
@@ -150,6 +151,116 @@ def test_inferer_prepare_inference_batches_split(mock_inferer):
     np.testing.assert_array_equal(batches[2]["cat_col"], [[5]])
 
 
+def test_infer_config_defaults_bert_prediction_length_to_seq_length():
+    config = InfererModel(
+        project_root=".",
+        metadata_config_path="dummy.json",
+        model_path="dummy.onnx",
+        model_type="generative",
+        training_objective="bert",
+        data_path="tests/unit/data/empty.parquet",
+        input_columns=["target_col"],
+        categorical_columns=[],
+        real_columns=["target_col"],
+        target_columns=["target_col"],
+        column_types={"target_col": "float64"},
+        target_column_types={"target_col": "real"},
+        seed=42,
+        device="cpu",
+        prediction_length=None,
+        seq_length=3,
+        inference_batch_size=2,
+        output_probabilities=False,
+        map_to_id=False,
+        autoregression=False,
+    )
+
+    assert config.prediction_length == config.seq_length
+
+
+def test_infer_config_rejects_bert_prediction_length_mismatch():
+    with pytest.raises(ValueError, match="prediction_length must be equal"):
+        InfererModel(
+            project_root=".",
+            metadata_config_path="dummy.json",
+            model_path="dummy.onnx",
+            model_type="generative",
+            training_objective="bert",
+            data_path="tests/unit/data/empty.parquet",
+            input_columns=["target_col"],
+            categorical_columns=[],
+            real_columns=["target_col"],
+            target_columns=["target_col"],
+            column_types={"target_col": "float64"},
+            target_column_types={"target_col": "real"},
+            seed=42,
+            device="cpu",
+            prediction_length=1,
+            seq_length=3,
+            inference_batch_size=2,
+            output_probabilities=False,
+            map_to_id=False,
+            autoregression=False,
+        )
+
+
+def test_infer_pure_passes_attention_valid_mask_to_onnx(mock_inferer):
+    class Input:
+        def __init__(self, name):
+            self.name = name
+
+    class Session:
+        def __init__(self):
+            self.ort_inputs = None
+
+        def get_inputs(self):
+            return [
+                Input("cat_col_in"),
+                Input("real_col_in"),
+                Input("attention_valid_mask"),
+            ]
+
+        def run(self, _, ort_inputs):
+            self.ort_inputs = ort_inputs
+            return [np.zeros((1, 2, 3), dtype=np.float32)]
+
+    session = Session()
+    mock_inferer.ort_session = session
+    mock_inferer.inference_batch_size = 2
+
+    x = {
+        "cat_col": np.array([[1, 2], [3, 4]], dtype=np.int64),
+        "real_col": np.array([[1.0, 2.0], [3.0, 4.0]], dtype=np.float32),
+    }
+    metadata = {
+        "attention_valid_mask": np.array([[False, True], [True, True]], dtype=np.bool_)
+    }
+
+    mock_inferer.infer_pure(x, metadata)
+    assert session.ort_inputs is not None
+    np.testing.assert_array_equal(
+        session.ort_inputs["attention_valid_mask"],
+        metadata["attention_valid_mask"],
+    )
+
+
+def test_filter_outputs_to_valid_targets_filters_predictions_and_probabilities():
+    mask = np.array([False, True, True, False])
+    sequence_ids, positions, preds, probs = _filter_outputs_to_valid_targets(
+        mask,
+        np.array([10, 10, 11, 11]),
+        np.array([0, 1, 0, 1]),
+        {"target_col": np.array([1, 2, 3, 4])},
+        {"target_col": np.array([[0.7, 0.3], [0.2, 0.8], [0.4, 0.6], [0.9, 0.1]])},
+    )
+
+    assert probs is not None
+    np.testing.assert_array_equal(sequence_ids, [10, 11])
+    np.testing.assert_array_equal(positions, [1, 0])
+    np.testing.assert_array_equal(preds["target_col"], [2, 3])
+    np.testing.assert_array_equal(probs["target_col"], [[0.2, 0.8], [0.4, 0.6]])
+
+
 # ==========================================
 # Test Autoregressive Tensor Inference
 # ==========================================
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index f478c3a8..c7c736a7 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -9,6 +9,7 @@
 from sequifier.config.preprocess_config import PreprocessorModel
 from sequifier.preprocess import (
     Preprocessor,
+    _apply_column_statistics,
     _apply_mask_column,
     _get_column_statistics,
     _get_data_columns,
@@ -17,6 +18,7 @@
     extract_subsequences,
     get_batch_limits,
     get_combined_statistics,
+    load_precomputed_id_maps,
     process_and_write_data_pt,
 )
 
@@ -490,6 +492,36 @@ def test_get_column_statistics_state_accumulation():
     np.testing.assert_almost_equal(stats["num_col"]["std"], expected_std)
 
 
+def test_get_column_statistics_skips_all_masked_chunks():
+    data = pl.DataFrame(
+        {
+            "cat_col": ["a", "b"],
+            "num_col": [1.0, 2.0],
+            RESERVED_MASK_COLUMN: [1, 1],
+        }
+    )
+
+    id_maps, stats = _get_column_statistics(
+        data,
+        ["cat_col", "num_col"],
+        {},
+        {},
+        0,
+        {},
+        mask_column=RESERVED_MASK_COLUMN,
+    )
+
+    assert id_maps == {}
+    assert stats == {}
+
+
+def test_apply_column_statistics_errors_when_all_rows_masked():
+    data = pl.DataFrame({"cat_col": ["a"], "num_col": [1.0]})
+
+    with pytest.raises(ValueError, match="No unmasked examples"):
+        _apply_column_statistics(data, ["cat_col", "num_col"], {}, {})
+
+
 def test_create_id_map():
     """Tests basic ID mapping creation."""
     df = pl.DataFrame({"A": ["z", "x", "y", "x"]})
@@ -499,3 +531,37 @@ def test_create_id_map():
     assert mapping["x"] == 3
     assert mapping["y"] == 4
     assert mapping["z"] == 5
+
+
+def test_load_precomputed_id_maps_allows_reserved_entries(tmp_path):
+    project_root = tmp_path / "project"
+    id_map_dir = project_root / "configs" / "id_maps"
+    id_map_dir.mkdir(parents=True)
+    (id_map_dir / "itemId.json").write_text(
+        json.dumps(
+            {
+                "[unknown]": 0,
+                "[other]": 1,
+                "[mask]": 2,
+                "a": 3,
+                "b": 4,
+            }
+        )
+    )
+
+    id_maps = load_precomputed_id_maps(str(project_root), ["itemId"])
+
+    assert id_maps["itemId"]["[mask]"] == 2
+    assert id_maps["itemId"]["a"] == 3
+
+
+def test_load_precomputed_id_maps_migrates_legacy_user_ids(tmp_path):
+    project_root = tmp_path / "project"
+    id_map_dir = project_root / "configs" / "id_maps"
+    id_map_dir.mkdir(parents=True)
+    (id_map_dir / "itemId.json").write_text(json.dumps({"a": 2, "b": 3}))
+
+    with pytest.warns(UserWarning, match="legacy user IDs"):
+        id_maps = load_precomputed_id_maps(str(project_root), ["itemId"])
+
+    assert id_maps["itemId"] == {"a": 3, "b": 4}
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index f581b40b..ee5ae7ab 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -2,12 +2,69 @@
 
 import pytest
 import torch
-
-from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
+from pydantic import ValidationError
+
+from sequifier.config.train_config import (
+    BERTSpecModel,
+    ModelSpecModel,
+    ReplacementDistribution,
+    TrainingSpecModel,
+    TrainModel,
+)
 from sequifier.helpers import infer_valid_mask_from_data
 from sequifier.train import TransformerModel
 
 
+def _training_spec_kwargs(**overrides):
+    values = {
+        "training_objective": "causal",
+        "device": "cpu",
+        "epochs": 1,
+        "save_interval_epochs": 1,
+        "batch_size": 4,
+        "learning_rate": 0.001,
+        "criterion": {"cat_col": "CrossEntropyLoss", "real_col": "MSELoss"},
+        "optimizer": {"name": "Adam"},
+        "scheduler": {"name": "StepLR", "step_size": 1, "gamma": 0.1},
+        "loss_weights": {"cat_col": 1.0, "real_col": 1.0},
+    }
+    values.update(overrides)
+    return values
+
+
+def _bert_spec():
+    return BERTSpecModel(
+        masking_probability=0.5,
+        replacement_distribution={
+            "masked": 1.0,
+            "random": 0.0,
+            "identical": 0.0,
+        },
+        span_masking={"type": "GeometricDistribution", "p": 1.0},
+    )
+
+
+def test_replacement_distribution_allows_zero_probabilities():
+    distribution = ReplacementDistribution(masked=1.0, random=0.0, identical=0.0)
+
+    assert distribution.masked == 1.0
+    assert distribution.random == 0.0
+    assert distribution.identical == 0.0
+
+
+def test_training_spec_model_requires_bert_spec_for_bert_objective():
+    with pytest.raises(ValidationError, match="BERT hyperparameters must be set"):
+        TrainingSpecModel(**_training_spec_kwargs(training_objective="bert"))
+
+
+def test_training_spec_model_rejects_bert_spec_for_causal_objective():
+    with pytest.raises(
+        ValidationError,
+        match="BERT hyperparameters should only be configured",
+    ):
+        TrainingSpecModel(**_training_spec_kwargs(bert_spec=_bert_spec()))
+
+
 @pytest.fixture
 def model_config(tmp_path):
     """Creates a valid TrainModel configuration for testing."""
@@ -113,6 +170,18 @@ def test_transformer_model_initialization(model, model_config):
         assert "real_col" in model.encoder
 
 
+def test_train_model_requires_bert_prediction_length_to_equal_seq_length(model_config):
+    config_values = model_config.model_dump()
+    config_values["model_spec"]["prediction_length"] = model_config.seq_length - 1
+    config_values["training_spec"] = _training_spec_kwargs(
+        training_objective="bert",
+        bert_spec=_bert_spec(),
+    )
+
+    with pytest.raises(ValidationError, match="prediction_length must be equal"):
+        TrainModel(**config_values)
+
+
 def test_forward_train_shapes(model, model_config):
     """Tests the output shapes of the forward_train method."""
     batch_size = model_config.training_spec.batch_size

From 8eaf770d57b3ffb973ee2b77657a6212af274f87 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 16:51:26 +0200
Subject: [PATCH 24/81] WIP

---
 src/sequifier/infer.py                |  33 ++++-----
 tests/integration-test-log.txt        |  14 ----
 tests/integration/test_onnx_export.py | 102 ++++++++++++++++++++++++++
 tests/unit/test_infer.py              |  63 ++++++++++++++++
 4 files changed, 180 insertions(+), 32 deletions(-)
 create mode 100644 tests/integration/test_onnx_export.py

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 24b250ea..465103f0 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -316,7 +316,10 @@ def _prediction_valid_mask_from_frame(
 
     left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
     if left_pad_lengths is None:
-        return None
+        raise ValueError(
+            "BERT inference with full-sequence prediction requires leftPadLength metadata "
+            "so padded target positions can be filtered."
+        )
 
     metadata = generate_padding_masks(
         left_pad_lengths,
@@ -706,6 +709,15 @@ def infer_generative(
             prediction_length = inferer.prediction_length  # Get prediction_length
 
             if total_steps == 1:
+                if (
+                    config.training_objective == "bert"
+                    and "target_valid_mask" not in metadata
+                ):
+                    raise ValueError(
+                        "BERT inference with full-sequence prediction requires target_valid_mask metadata "
+                        "so padded target positions can be filtered."
+                    )
+
                 # Non-autoregressive path: Apply prediction_length logic
                 sequence_ids_for_preds_base = sequence_ids_tensor.numpy()
                 item_positions_base_raw = start_positions_tensor.numpy()
@@ -1664,8 +1676,6 @@ def infer_pure(
             ONNX model.
         """
         metadata = metadata or {}
-        fallback_columns = sorted(x.keys())
-        used_feature_columns = set()
         reference_shape = next(iter(x.values())).shape
         ort_inputs = {}
         for session_input in self.ort_session.get_inputs():
@@ -1679,25 +1689,12 @@ def infer_pure(
             elif input_name.endswith("_in") and input_name[:-3] in x:
                 feature_column = input_name[:-3]
                 value = x[feature_column]
-                used_feature_columns.add(feature_column)
             elif input_name in x:
                 value = x[input_name]
-                used_feature_columns.add(input_name)
             else:
-                fallback_column = next(
-                    (
-                        column
-                        for column in fallback_columns
-                        if column not in used_feature_columns
-                    ),
-                    None,
+                raise ValueError(
+                    f"Could not map ONNX input '{input_name}' to a feature or metadata array."
                 )
-                if fallback_column is None:
-                    raise ValueError(
-                        f"Could not map ONNX input '{input_name}' to a feature or metadata array."
-                    )
-                value = x[fallback_column]
-                used_feature_columns.add(fallback_column)
 
             ort_inputs[input_name] = self.expand_to_batch_size(value)
 
diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
index e44783c3..4b0c9730 100644
--- a/tests/integration-test-log.txt
+++ b/tests/integration-test-log.txt
@@ -39,20 +39,6 @@ sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metada
 sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-3.json --model-path models/sequifier-model-real-3-best-3.pt --data-path data/test-data-real-3-split1.parquet --input-columns None
 sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-5.json --model-path models/sequifier-model-categorical-5-best-3.onnx --data-path data/test-data-categorical-5-split2 --input-columns itemId supCat1 supCat2 supCat4
 sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-5.json --model-path models/sequifier-model-real-5-best-3.pt --data-path data/test-data-real-5-split1.parquet --input-columns None
-sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-50.json --model-path models/sequifier-model-categorical-50-best-3.onnx --data-path data/test-data-categorical-50-split2 --input-columns itemId supCat1 supCat2 supCat3 supCat4 supCat5 supCat6 supCat7 supCat8 supCat9 supCat10 supCat11 supCat12 supCat13 supCat14 supCat15 supCat16 supCat17 supCat18 supCat19 supCat20 supCat21 supCat22 supCat23 supCat24 supCat25 supCat26 supCat27 supCat28 supCat29 supCat30 supCat31 supCat32 supCat33 supCat34 supCat35 supCat36 supCat37 supCat38 supCat39 supCat40 supCat41 supCat42 supCat43 supCat44 supCat45 supCat46 supCat47 supCat48 supCat49
-sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-50.json --model-path models/sequifier-model-real-50-best-3.pt --data-path data/test-data-real-50-split1.parquet --input-columns None
-sequifier infer --config-path tests/configs/infer-test-categorical-multitarget.yaml
-sequifier infer --config-path tests/configs/infer-test-real-autoregression.yaml --input-columns itemValue --randomize
-sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-1.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3.yaml
-sequifier infer --config-path tests/configs/infer-test-distributed.yaml
-sequifier infer --config-path tests/configs/infer-test-distributed-parquet.yaml
-sequifier infer --config-path tests/configs/infer-test-lazy.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-autoregression.yaml  --input-columns itemId
-sequifier infer --config-path tests/configs/infer-test-categorical-embedding.yaml  --input-columns itemId
-sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-bert.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-bert-embedding.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
 sequifier train --config-path tests/configs/train-test-resume-epoch.yaml
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
new file mode 100644
index 00000000..f4c5a736
--- /dev/null
+++ b/tests/integration/test_onnx_export.py
@@ -0,0 +1,102 @@
+import os
+
+import numpy as np
+import onnxruntime
+
+from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
+from sequifier.train import TransformerModel
+
+
+def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
+    project_root = str(tmp_path)
+    (tmp_path / "logs").mkdir()
+
+    seq_length = 4
+    inference_batch_size = 2
+    config = TrainModel(
+        project_root=project_root,
+        model_name="bert-onnx-mask",
+        metadata_config_path="metadata.json",
+        training_data_path="data/train.pt",
+        validation_data_path="data/val.pt",
+        input_columns=["cat_col", "real_col"],
+        target_columns=["cat_col", "real_col"],
+        target_column_types={"cat_col": "categorical", "real_col": "real"},
+        column_types={"cat_col": "int64", "real_col": "float64"},
+        categorical_columns=["cat_col"],
+        real_columns=["real_col"],
+        id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
+        n_classes={"cat_col": 6},
+        seq_length=seq_length,
+        inference_batch_size=inference_batch_size,
+        seed=42,
+        export_generative_model=True,
+        export_embedding_model=False,
+        export_onnx=True,
+        export_pt=False,
+        model_spec=ModelSpecModel(
+            initial_embedding_dim=8,
+            dim_model=8,
+            n_head=2,
+            dim_feedforward=8,
+            num_layers=1,
+            prediction_length=seq_length,
+            feature_embedding_dims={"cat_col": 7, "real_col": 1},
+            activation_fn="relu",
+            normalization="layer_norm",
+            positional_encoding="learned",
+            attention_type="mha",
+            norm_first=False,
+            n_kv_heads=2,
+        ),
+        training_spec=TrainingSpecModel(
+            training_objective="bert",
+            bert_spec={
+                "masking_probability": 0.5,
+                "replacement_distribution": {
+                    "masked": 1.0,
+                    "random": 0.0,
+                    "identical": 0.0,
+                },
+                "span_masking": {"type": "GeometricDistribution", "p": 1.0},
+            },
+            device="cpu",
+            epochs=1,
+            save_interval_epochs=1,
+            batch_size=2,
+            learning_rate=0.001,
+            criterion={"cat_col": "CrossEntropyLoss", "real_col": "MSELoss"},
+            optimizer={"name": "Adam"},
+            scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
+            loss_weights={"cat_col": 1.0, "real_col": 1.0},
+            torch_compile="none",
+            layer_autocast=False,
+        ),
+    )
+    model = TransformerModel(config)
+
+    model._export_model(model, "best", 1)
+    export_path = os.path.join(
+        project_root, "models", "sequifier-bert-onnx-mask-best-1.onnx"
+    )
+
+    session = onnxruntime.InferenceSession(
+        export_path, providers=["CPUExecutionProvider"]
+    )
+    input_names = [session_input.name for session_input in session.get_inputs()]
+
+    assert "attention_valid_mask" in input_names
+
+    ort_inputs = {
+        "cat_col_in": np.array([[3, 4, 5, 3], [0, 0, 3, 4]], dtype=np.int64),
+        "real_col_in": np.array(
+            [[0.1, 0.2, 0.3, 0.4], [0.0, 0.0, 0.5, 0.6]], dtype=np.float32
+        ),
+        "attention_valid_mask": np.array(
+            [[True, True, True, True], [False, False, True, True]], dtype=np.bool_
+        ),
+    }
+    outputs = session.run(None, ort_inputs)
+
+    assert len(outputs) == 2
+    assert outputs[0].shape[1] == inference_batch_size
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 7069cbb7..b349d073 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -1,6 +1,7 @@
 from unittest.mock import patch
 
 import numpy as np
+import polars as pl
 import pytest
 import torch
 
@@ -8,6 +9,7 @@
 from sequifier.infer import (
     Inferer,
     _filter_outputs_to_valid_targets,
+    _prediction_valid_mask_from_frame,
     get_probs_preds_from_dict,
     normalize,
     sample_with_cumsum,
@@ -244,6 +246,67 @@ def run(self, _, ort_inputs):
     )
 
 
+def test_infer_pure_rejects_unknown_onnx_input(mock_inferer):
+    class Input:
+        def __init__(self, name):
+            self.name = name
+
+    class Session:
+        def get_inputs(self):
+            return [Input("unexpected_input")]
+
+    mock_inferer.ort_session = Session()
+
+    with pytest.raises(ValueError, match="Could not map ONNX input"):
+        mock_inferer.infer_pure(
+            {
+                "cat_col": np.array([[1, 2]], dtype=np.int64),
+                "real_col": np.array([[1.0, 2.0]], dtype=np.float32),
+            }
+        )
+
+
+def test_prediction_valid_mask_from_frame_requires_left_pad_length_for_bert():
+    config = InfererModel(
+        project_root=".",
+        metadata_config_path="dummy.json",
+        model_path="dummy.onnx",
+        model_type="generative",
+        training_objective="bert",
+        data_path="tests/unit/data/empty.parquet",
+        input_columns=["target_col"],
+        categorical_columns=[],
+        real_columns=["target_col"],
+        target_columns=["target_col"],
+        column_types={"target_col": "float64"},
+        target_column_types={"target_col": "real"},
+        seed=42,
+        device="cpu",
+        prediction_length=None,
+        seq_length=3,
+        inference_batch_size=2,
+        output_probabilities=False,
+        map_to_id=False,
+        autoregression=False,
+    )
+    data = pl.DataFrame(
+        {
+            "sequenceId": [0],
+            "subsequenceId": [0],
+            "startItemPosition": [0],
+            "inputCol": ["target_col"],
+            "3": [0.0],
+            "2": [0.0],
+            "1": [1.0],
+            "0": [2.0],
+        }
+    )
+
+    assert config.prediction_length is not None
+    with pytest.raises(ValueError, match="leftPadLength"):
+        _prediction_valid_mask_from_frame(config, data, config.prediction_length)
+
+
 def test_filter_outputs_to_valid_targets_filters_predictions_and_probabilities():
     mask = np.array([False, True, True, False])
     sequence_ids, positions, preds, probs = _filter_outputs_to_valid_targets(

From 6e287a0c20919ec6ba4bdd18572f54929b92208a Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 17:10:58 +0200
Subject: [PATCH 25/81] WIP

---
 src/sequifier/config/train_config.py  |  5 ++
 src/sequifier/train.py                | 45 ++++++++++----
 tests/integration-test-log.txt        | 14 +++++
 tests/integration/test_onnx_export.py | 88 +++++++++++++++++++++++++++
 4 files changed, 139 insertions(+), 13 deletions(-)

diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 477a602c..2b1666f0 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -16,6 +16,7 @@
     ConfigDict,
     Field,
     computed_field,
+    field_serializer,
     field_validator,
     model_validator,
 )
@@ -233,6 +234,10 @@ def __init__(self, **kwargs):
         self.validate_scheduler_config(kwargs["scheduler"], kwargs)
         self.scheduler = DotDict(kwargs["scheduler"])
 
+    @field_serializer("optimizer", "scheduler")
+    def serialize_dotdict(self, value: DotDict) -> dict[str, Any]:
+        return dict(value)
+
     @computed_field
     @property
     def data_offset(self) -> int:
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index b6230c98..900fb027 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -533,6 +533,22 @@ def forward(
         return self.transformer_model.forward_embed(src, metadata=metadata)
 
 
+class _OnnxExportWrapper(nn.Module):
+    def __init__(
+        self,
+        model: Union["TransformerModel", TransformerEmbeddingModel],
+        feature_columns: list[str],
+    ):
+        super().__init__()
+        self.model = model
+        self.feature_columns = feature_columns
+
+    def forward(self, *inputs: Tensor):
+        features = dict(zip(self.feature_columns, inputs[:-1]))
+        metadata = {"attention_valid_mask": inputs[-1]}
+        return self.model(features, metadata=metadata)
+
+
 class TransformerModel(nn.Module):
     """The main Transformer model for the sequifier.
 
@@ -2067,6 +2083,8 @@ def _export_model(
             suffix: A string suffix for the filename (e.g., "best", "last-embedding").
             epoch: The current epoch number, included in the filename.
         """
+        os.makedirs(os.path.join(self.project_root, "models"), exist_ok=True)
+
         if self.export_onnx:
             is_different_type = any(
                 p.dtype in [torch.float16, torch.bfloat16, torch.float64]
@@ -2099,20 +2117,21 @@ def _export_model(
             }
 
             input_dict = {**x_cat, **x_real}
-            metadata_dict = {
-                "attention_valid_mask": torch.ones(
-                    self.inference_batch_size,
-                    self.seq_length,
-                    dtype=torch.bool,
-                    device=export_device,
-                )
-            }
+            attention_valid_mask = torch.ones(
+                self.inference_batch_size,
+                self.seq_length,
+                dtype=torch.bool,
+                device=export_device,
+            )
+            attention_valid_mask[0, 0] = False
 
-            # Wrap in a tuple to prevent PyTorch from treating dictionaries as kwargs.
-            x = (input_dict, metadata_dict)
+            feature_columns = list(input_dict.keys())
+            x = tuple(input_dict[col] for col in feature_columns) + (
+                attention_valid_mask,
+            )
+            export_wrapper = _OnnxExportWrapper(model_to_export, feature_columns)
 
-            # PyTree flattening sorts dictionary keys automatically, so names must match that order.
-            input_names = [f"{col}_in" for col in sorted(input_dict.keys())] + [
+            input_names = [f"{col}_in" for col in input_dict.keys()] + [
                 "attention_valid_mask"
             ]
 
@@ -2158,7 +2177,7 @@ def _export_model(
                 warnings.filterwarnings("ignore", category=FutureWarning)
 
                 torch.onnx.export(
-                    model_to_export,
+                    export_wrapper,
                     x,
                     export_path,
                     export_params=True,
diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
index 4b0c9730..e44783c3 100644
--- a/tests/integration-test-log.txt
+++ b/tests/integration-test-log.txt
@@ -39,6 +39,20 @@ sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metada
 sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-3.json --model-path models/sequifier-model-real-3-best-3.pt --data-path data/test-data-real-3-split1.parquet --input-columns None
 sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-5.json --model-path models/sequifier-model-categorical-5-best-3.onnx --data-path data/test-data-categorical-5-split2 --input-columns itemId supCat1 supCat2 supCat4
 sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-5.json --model-path models/sequifier-model-real-5-best-3.pt --data-path data/test-data-real-5-split1.parquet --input-columns None
+sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-50.json --model-path models/sequifier-model-categorical-50-best-3.onnx --data-path data/test-data-categorical-50-split2 --input-columns itemId supCat1 supCat2 supCat3 supCat4 supCat5 supCat6 supCat7 supCat8 supCat9 supCat10 supCat11 supCat12 supCat13 supCat14 supCat15 supCat16 supCat17 supCat18 supCat19 supCat20 supCat21 supCat22 supCat23 supCat24 supCat25 supCat26 supCat27 supCat28 supCat29 supCat30 supCat31 supCat32 supCat33 supCat34 supCat35 supCat36 supCat37 supCat38 supCat39 supCat40 supCat41 supCat42 supCat43 supCat44 supCat45 supCat46 supCat47 supCat48 supCat49
+sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-50.json --model-path models/sequifier-model-real-50-best-3.pt --data-path data/test-data-real-50-split1.parquet --input-columns None
+sequifier infer --config-path tests/configs/infer-test-categorical-multitarget.yaml
+sequifier infer --config-path tests/configs/infer-test-real-autoregression.yaml --input-columns itemValue --randomize
+sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-1.yaml
+sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3.yaml
+sequifier infer --config-path tests/configs/infer-test-distributed.yaml
+sequifier infer --config-path tests/configs/infer-test-distributed-parquet.yaml
+sequifier infer --config-path tests/configs/infer-test-lazy.yaml
+sequifier infer --config-path tests/configs/infer-test-categorical-autoregression.yaml  --input-columns itemId
+sequifier infer --config-path tests/configs/infer-test-categorical-embedding.yaml  --input-columns itemId
+sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+sequifier infer --config-path tests/configs/infer-test-categorical-bert.yaml
+sequifier infer --config-path tests/configs/infer-test-categorical-bert-embedding.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
 sequifier train --config-path tests/configs/train-test-resume-epoch.yaml
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index f4c5a736..e26d3d8e 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -100,3 +100,91 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
 
     assert len(outputs) == 2
     assert outputs[0].shape[1] == inference_batch_size
+
+
+def test_onnx_export_preserves_feature_name_order(tmp_path):
+    project_root = str(tmp_path)
+    (tmp_path / "logs").mkdir()
+
+    seq_length = 4
+    inference_batch_size = 2
+    config = TrainModel(
+        project_root=project_root,
+        model_name="onnx-feature-order",
+        metadata_config_path="metadata.json",
+        training_data_path="data/train.pt",
+        validation_data_path="data/val.pt",
+        input_columns=["cat2", "cat10"],
+        target_columns=["cat2"],
+        target_column_types={"cat2": "categorical"},
+        column_types={"cat2": "int64", "cat10": "int64"},
+        categorical_columns=["cat2", "cat10"],
+        real_columns=[],
+        id_maps={
+            "cat2": {"a": 3, "b": 4, "c": 5, "d": 6},
+            "cat10": {"x": 3, "y": 4, "z": 5},
+        },
+        n_classes={"cat2": 7, "cat10": 6},
+        seq_length=seq_length,
+        inference_batch_size=inference_batch_size,
+        seed=42,
+        export_generative_model=True,
+        export_embedding_model=False,
+        export_onnx=True,
+        export_pt=False,
+        model_spec=ModelSpecModel(
+            initial_embedding_dim=8,
+            dim_model=8,
+            n_head=2,
+            dim_feedforward=8,
+            num_layers=1,
+            prediction_length=1,
+            feature_embedding_dims={"cat2": 4, "cat10": 4},
+            activation_fn="relu",
+            normalization="layer_norm",
+            positional_encoding="learned",
+            attention_type="mha",
+            norm_first=False,
+            n_kv_heads=2,
+        ),
+        training_spec=TrainingSpecModel(
+            training_objective="causal",
+            device="cpu",
+            epochs=1,
+            save_interval_epochs=1,
+            batch_size=2,
+            learning_rate=0.001,
+            criterion={"cat2": "CrossEntropyLoss"},
+            optimizer={"name": "Adam"},
+            scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
+            loss_weights={"cat2": 1.0},
+            torch_compile="none",
+            layer_autocast=False,
+        ),
+    )
+    model = TransformerModel(config)
+
+    model._export_model(model, "best", 1)
+    export_path = os.path.join(
+        project_root, "models", "sequifier-onnx-feature-order-best-1.onnx"
+    )
+
+    session = onnxruntime.InferenceSession(
+        export_path, providers=["CPUExecutionProvider"]
+    )
+    input_names = [session_input.name for session_input in session.get_inputs()]
+
+    assert input_names == ["cat2_in", "cat10_in", "attention_valid_mask"]
+
+    outputs = session.run(
+        None,
+        {
+            "cat2_in": np.array([[3, 4, 5, 6], [6, 5, 4, 3]], dtype=np.int64),
+            "cat10_in": np.array([[3, 4, 5, 3], [5, 4, 3, 5]], dtype=np.int64),
+            "attention_valid_mask": np.ones(
+                (inference_batch_size, seq_length), dtype=np.bool_
+            ),
+        },
+    )
+
+    assert outputs[0].shape == (1, inference_batch_size, config.n_classes["cat2"])

From 34a3e7f416e1f262fafdc3b3e7f833a5f1978182 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 10 Jun 2026 17:41:06 +0200
Subject: [PATCH 26/81] Clean up for tests

---
 src/sequifier/infer.py   | 101 ---------------------------------------
 tests/unit/test_infer.py |  66 ++++++-------------------
 2 files changed, 14 insertions(+), 153 deletions(-)

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 465103f0..b89002b6 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -18,7 +18,6 @@
     configure_determinism,
     construct_index_maps,
     generate_padding_masks,
-    get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
@@ -297,77 +296,6 @@ def calculate_item_positions(
     return repeated_bases + tiled_offsets
 
 
-def _flatten_prediction_valid_mask(
-    metadata: Optional[dict[str, Any]], prediction_length: int
-) -> Optional[np.ndarray]:
-    if metadata is None or "target_valid_mask" not in metadata:
-        return None
-
-    mask = metadata["target_valid_mask"]
-    mask_np = mask.cpu().numpy() if isinstance(mask, torch.Tensor) else np.asarray(mask)
-    return mask_np[:, -prediction_length:].reshape(-1).astype(bool)
-
-
-def _prediction_valid_mask_from_frame(
-    config: "InfererModel", data: pl.DataFrame, prediction_length: int
-) -> Optional[np.ndarray]:
-    if config.training_objective != "bert":
-        return None
-
-    left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
-    if left_pad_lengths is None:
-        raise ValueError(
-            "BERT inference with full-sequence prediction requires leftPadLength metadata "
-            "so padded target positions can be filtered."
-        )
-
-    metadata = generate_padding_masks(
-        left_pad_lengths,
-        config.seq_length,
-        data_offset=1,
-        target_offset=1,
-    )
-    return _flatten_prediction_valid_mask(metadata, prediction_length)
-
-
-def _filter_outputs_to_valid_targets(
-    prediction_valid_mask: Optional[np.ndarray],
-    sequence_ids_for_preds: np.ndarray,
-    item_positions_for_preds: np.ndarray,
-    preds: dict[str, np.ndarray],
-    probs: Optional[dict[str, np.ndarray]],
-) -> tuple[
-    np.ndarray, np.ndarray, dict[str, np.ndarray], Optional[dict[str, np.ndarray]]
-]:
-    if prediction_valid_mask is None:
-        return sequence_ids_for_preds, item_positions_for_preds, preds, probs
-
-    if prediction_valid_mask.shape[0] != len(sequence_ids_for_preds):
-        raise ValueError(
-            "target_valid_mask length does not match prediction metadata length "
-            f"({prediction_valid_mask.shape[0]} != {len(sequence_ids_for_preds)})."
-        )
-
-    filtered_preds = {
-        target_column: np.asarray(values)[prediction_valid_mask]
-        for target_column, values in preds.items()
-    }
-    filtered_probs = (
-        {
-            target_column: np.asarray(values)[prediction_valid_mask]
-            for target_column, values in probs.items()
-        }
-        if probs is not None
-        else None
-    )
-    return (
-        np.asarray(sequence_ids_for_preds)[prediction_valid_mask],
-        np.asarray(item_positions_for_preds)[prediction_valid_mask],
-        filtered_preds,
-        filtered_probs,
-    )
-
-
 @beartype
 def infer_embedding(
     config: "InfererModel",
@@ -574,7 +502,6 @@ def infer_generative(
         is_folder_input = os.path.isdir(
             normalize_path(config.data_path, config.project_root)
         )
-        prediction_valid_mask = None
 
         if config.read_format in ["parquet", "csv"] and not is_folder_input:
             if config.input_columns is not None:
@@ -594,9 +521,6 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
-                prediction_valid_mask = _prediction_valid_mask_from_frame(
-                    config, data, prediction_length
-                )
 
                 # Expand IDs to match model output shape
                 sequence_ids_for_preds = np.repeat(
@@ -646,9 +570,6 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
-                prediction_valid_mask = _prediction_valid_mask_from_frame(
-                    config, data, prediction_length
-                )
 
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
@@ -709,15 +630,6 @@ def infer_generative(
             prediction_length = inferer.prediction_length  # Get prediction_length
 
             if total_steps == 1:
-                if (
-                    config.training_objective == "bert"
-                    and "target_valid_mask" not in metadata
-                ):
-                    raise ValueError(
-                        "BERT inference with full-sequence prediction requires target_valid_mask metadata "
-                        "so padded target positions can be filtered."
-                    )
-
                 # Non-autoregressive path: Apply prediction_length logic
                 sequence_ids_for_preds_base = sequence_ids_tensor.numpy()
                 item_positions_base_raw = start_positions_tensor.numpy()
@@ -725,9 +637,6 @@ def infer_generative(
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
                 )
-                prediction_valid_mask = _flatten_prediction_valid_mask(
-                    metadata, prediction_length
-                )
 
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
@@ -765,16 +674,6 @@ def infer_generative(
                     predictions, target_column
                 )
 
-        sequence_ids_for_preds, item_positions_for_preds, preds, probs = (
-            _filter_outputs_to_valid_targets(
-                prediction_valid_mask,
-                sequence_ids_for_preds,
-                item_positions_for_preds,
-                preds,
-                probs,
-            )
-        )
-
         os.makedirs(
             os.path.join(config.project_root, "outputs", "predictions"),
             exist_ok=True,
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index b349d073..6642ba9e 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -1,15 +1,13 @@
 from unittest.mock import patch
 
 import numpy as np
-import polars as pl
 import pytest
 import torch
 
 from sequifier.config.infer_config import InfererModel
 from sequifier.infer import (
     Inferer,
-    _filter_outputs_to_valid_targets,
-    _prediction_valid_mask_from_frame,
+    calculate_item_positions,
     get_probs_preds_from_dict,
     normalize,
     sample_with_cumsum,
@@ -266,62 +264,26 @@ def get_inputs(self):
         )
 
 
-def test_prediction_valid_mask_from_frame_requires_left_pad_length_for_bert():
-    config = InfererModel(
-        project_root=".",
-        metadata_config_path="dummy.json",
-        model_path="dummy.onnx",
-        model_type="generative",
+def test_calculate_item_positions_bert_uses_full_input_window():
+    positions = calculate_item_positions(
+        np.array([10, 20]),
+        seq_length=4,
+        prediction_length=4,
         training_objective="bert",
-        data_path="tests/unit/data/empty.parquet",
-        input_columns=["target_col"],
-        categorical_columns=[],
-        real_columns=["target_col"],
-        target_columns=["target_col"],
-        column_types={"target_col": "float64"},
-        target_column_types={"target_col": "real"},
-        seed=42,
-        device="cpu",
-        prediction_length=None,
-        seq_length=3,
-        inference_batch_size=2,
-        output_probabilities=False,
-        map_to_id=False,
-        autoregression=False,
-    )
-    data = pl.DataFrame(
-        {
-            "sequenceId": [0],
-            "subsequenceId": [0],
-            "startItemPosition": [0],
-            "inputCol": ["target_col"],
-            "3": [0.0],
-            "2": [0.0],
-            "1": [1.0],
-            "0": [2.0],
-        }
     )
 
-    assert config.prediction_length is not None
-    with pytest.raises(ValueError, match="leftPadLength"):
-        _prediction_valid_mask_from_frame(config, data, config.prediction_length)
+    np.testing.assert_array_equal(positions, [10, 11, 12, 13, 20, 21, 22, 23])
 
 
-def test_filter_outputs_to_valid_targets_filters_predictions_and_probabilities():
-    mask = np.array([False, True, True, False])
-    sequence_ids, positions, preds, probs = _filter_outputs_to_valid_targets(
-        mask,
-        np.array([10, 10, 11, 11]),
-        np.array([0, 1, 0, 1]),
-        {"target_col": np.array([1, 2, 3, 4])},
-        {"target_col": np.array([[0.7, 0.3], [0.2, 0.8], [0.4, 0.6], [0.9, 0.1]])},
+def test_calculate_item_positions_causal_uses_future_window_tail():
+    positions = calculate_item_positions(
+        np.array([10, 20]),
+        seq_length=4,
+        prediction_length=2,
+        training_objective="causal",
     )
 
-    assert probs is not None
-    np.testing.assert_array_equal(sequence_ids, [10, 11])
-    np.testing.assert_array_equal(positions, [1, 0])
-    np.testing.assert_array_equal(preds["target_col"], [2, 3])
-    np.testing.assert_array_equal(probs["target_col"], [[0.2, 0.8], [0.4, 0.6]])
+    np.testing.assert_array_equal(positions, [13, 14, 23, 24])
 
 
 # ==========================================

From cc87def73e5075d6f981bf5618ca58d4b6604ad0 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 15:48:08 +0200
Subject: [PATCH 27/81] Address BERT shortcomings

---
 src/sequifier/config/infer_config.py  |  10 +-
 src/sequifier/config/probabilities.py |   2 +-
 src/sequifier/config/train_config.py  |   3 -
 src/sequifier/infer.py                | 130 +++++++++++++++++++++
 src/sequifier/preprocess.py           |   7 +-
 src/sequifier/train.py                |   7 +-
 tests/integration/test_inference.py   |  80 ++++++++++++-
 tests/unit/test_infer.py              | 160 ++++++++++++++++++++++++++
 tests/unit/test_preprocess.py         |  42 +++++++
 tests/unit/test_train.py              |  56 ++++++++-
 10 files changed, 480 insertions(+), 17 deletions(-)

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 668252b2..b89b54fb 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -146,11 +146,13 @@ class InfererModel(BaseModel):
     autoregression_total_steps: Optional[int] = Field(default=None)
 
     @model_validator(mode="after")
-    def validate_bert_prediction_length_matches_seq_length(self):
+    def normalize_prediction_length(self):
+        if self.prediction_length is None:
+            self.prediction_length = (
+                self.seq_length if self.training_objective == "bert" else 1
+            )
         if self.training_objective == "bert":
-            if self.prediction_length is None:
-                self.prediction_length = self.seq_length
-            elif self.prediction_length != self.seq_length:
+            if self.prediction_length != self.seq_length:
                 raise ValueError(
                     "For BERT inference, prediction_length must be equal to seq_length "
                     f"(got prediction_length={self.prediction_length}, seq_length={self.seq_length})."
diff --git a/src/sequifier/config/probabilities.py b/src/sequifier/config/probabilities.py
index d18a8de4..a7656f1e 100644
--- a/src/sequifier/config/probabilities.py
+++ b/src/sequifier/config/probabilities.py
@@ -67,7 +67,7 @@ class PoissonDistributionFloor(BaseModel, ProbabilityDistributionBaseClass):
 
     def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
         m = torch.distributions.Poisson(rate=torch.tensor([self.rate], device=device))
-        return m.sample(shape).squeeze(-1).long()
+        return m.sample(shape).squeeze(-1).long() + 1
 
 
 ProbabilityDistribution = Annotated[
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 2b1666f0..fd0dfa22 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -15,7 +15,6 @@
     BaseModel,
     ConfigDict,
     Field,
-    computed_field,
     field_serializer,
     field_validator,
     model_validator,
@@ -238,12 +237,10 @@ def __init__(self, **kwargs):
     def serialize_dotdict(self, value: DotDict) -> dict[str, Any]:
         return dict(value)
 
-    @computed_field
     @property
     def data_offset(self) -> int:
         return 1
 
-    @computed_field
     @property
     def target_offset(self) -> int:
         return 0 if self.training_objective == "causal" else 1
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index b89002b6..16a6c176 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -18,6 +18,7 @@
     configure_determinism,
     construct_index_maps,
     generate_padding_masks,
+    get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
@@ -296,6 +297,88 @@ def calculate_item_positions(
     return repeated_bases + tiled_offsets
 
 
+@beartype
+def _flatten_bert_target_valid_mask(
+    config: InfererModel,
+    metadata: Optional[dict[str, Any]],
+    prediction_length: int,
+) -> Optional[np.ndarray]:
+    if config.training_objective != "bert" or not metadata:
+        return None
+    if "target_valid_mask" not in metadata:
+        return None
+
+    valid_mask = metadata["target_valid_mask"]
+    if isinstance(valid_mask, torch.Tensor):
+        valid_mask = valid_mask.detach().cpu().numpy()
+    valid_mask = np.asarray(valid_mask, dtype=bool)
+
+    if valid_mask.ndim != 2:
+        raise ValueError(f"target_valid_mask must be 2D, got shape {valid_mask.shape}.")
+    if valid_mask.shape[1] != prediction_length:
+        raise ValueError(
+            "target_valid_mask width must match prediction_length "
+            f"(got {valid_mask.shape[1]} and {prediction_length})."
+        )
+
+    return valid_mask.reshape(-1)
+
+
+@beartype
+def _bert_target_valid_mask_from_preprocessed_data(
+    config: InfererModel,
+    data: pl.DataFrame,
+    prediction_length: int,
+) -> Optional[np.ndarray]:
+    if config.training_objective != "bert":
+        return None
+
+    left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
+    if left_pad_lengths is None:
+        return None
+
+    metadata = generate_padding_masks(
+        left_pad_lengths,
+        config.seq_length,
+        data_offset=1,
+        target_offset=1,
+    )
+    return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
+
+
+@beartype
+def _apply_valid_prediction_mask(
+    values: np.ndarray,
+    valid_prediction_mask: Optional[np.ndarray],
+    label: str,
+) -> np.ndarray:
+    values = np.asarray(values)
+    if valid_prediction_mask is None:
+        return values
+    if values.shape[0] != valid_prediction_mask.shape[0]:
+        raise ValueError(
+            f"{label} has {values.shape[0]} rows, but target_valid_mask has "
+            f"{valid_prediction_mask.shape[0]} rows."
+        )
+    return values[valid_prediction_mask]
+
+
+@beartype
+def _apply_valid_prediction_mask_to_dict(
+    values: Optional[dict[str, np.ndarray]],
+    valid_prediction_mask: Optional[np.ndarray],
+    label: str,
+) -> Optional[dict[str, np.ndarray]]:
+    if values is None:
+        return None
+    return {
+        key: _apply_valid_prediction_mask(
+            value, valid_prediction_mask, f"{label}.{key}"
+        )
+        for key, value in values.items()
+    }
+
+
 @beartype
 def infer_embedding(
     config: "InfererModel",
@@ -326,6 +409,7 @@ def infer_embedding(
     """
     for data_id, data in enumerate(dataset):
         prediction_length = inferer.prediction_length
+        valid_prediction_mask = None
 
         # Step 1: Get embeddings and base position/ID data
         is_folder_input = os.path.isdir(
@@ -343,6 +427,9 @@ def infer_embedding(
             mask = pl.arange(0, data.height, eager=True) % n_input_cols == 0
 
             embeddings = get_embeddings(config, inferer, data, column_types)
+            valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
+                config, data, prediction_length
+            )
 
             sequence_ids_for_preds = data.get_column("sequenceId").filter(mask)
             subsequence_ids_for_preds = data.get_column("subsequenceId").filter(mask)
@@ -358,6 +445,9 @@ def infer_embedding(
             mask = pl.arange(0, data.height, eager=True) % n_input_cols == 0
 
             embeddings = get_embeddings(config, inferer, data, column_types)
+            valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
+                config, data, prediction_length
+            )
 
             sequence_ids_for_preds = (
                 data.get_column("sequenceId").filter(mask).to_numpy()
@@ -389,6 +479,9 @@ def infer_embedding(
             embeddings = get_embeddings_pt(
                 config, inferer, sequences_dict, metadata=metadata
             )
+            valid_prediction_mask = _flatten_bert_target_valid_mask(
+                config, metadata, prediction_length
+            )
 
             sequence_ids_for_preds = sequence_ids_tensor.numpy()
             subsequence_ids_for_preds = subsequence_ids_tensor.numpy()
@@ -420,6 +513,18 @@ def infer_embedding(
         subsequence_ids_repeated = np.repeat(
             subsequence_ids_for_preds, prediction_length
         )
+        embeddings = _apply_valid_prediction_mask(
+            embeddings, valid_prediction_mask, "embeddings"
+        )
+        final_positions = _apply_valid_prediction_mask(
+            final_positions, valid_prediction_mask, "itemPosition"
+        )
+        sequence_ids_repeated = _apply_valid_prediction_mask(
+            sequence_ids_repeated, valid_prediction_mask, "sequenceId"
+        )
+        subsequence_ids_repeated = _apply_valid_prediction_mask(
+            subsequence_ids_repeated, valid_prediction_mask, "subsequenceId"
+        )
 
         # Step 3: Build the final DataFrame
         embeddings_df = pl.DataFrame(
@@ -498,6 +603,8 @@ def infer_generative(
             `torch.dtype`.
     """
     for data_id, data in enumerate(dataset):
+        valid_prediction_mask = None
+
         # Step 1: Adapt Data Subsetting (now works on Polars DF)
         is_folder_input = os.path.isdir(
             normalize_path(config.data_path, config.project_root)
@@ -521,6 +628,9 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
+                valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
+                    config, data, prediction_length
+                )
 
                 # Expand IDs to match model output shape
                 sequence_ids_for_preds = np.repeat(
@@ -570,6 +680,9 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
+                valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
+                    config, data, prediction_length
+                )
 
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_for_preds_base, prediction_length
@@ -628,6 +741,9 @@ def infer_generative(
             )
 
             prediction_length = inferer.prediction_length  # Get prediction_length
+            valid_prediction_mask = _flatten_bert_target_valid_mask(
+                config, metadata, prediction_length
+            )
 
             if total_steps == 1:
                 # Non-autoregressive path: Apply prediction_length logic
@@ -674,6 +790,19 @@ def infer_generative(
                     predictions, target_column
                 )
 
+        sequence_ids_for_preds = _apply_valid_prediction_mask(
+            sequence_ids_for_preds, valid_prediction_mask, "sequenceId"
+        )
+        item_positions_for_preds = _apply_valid_prediction_mask(
+            item_positions_for_preds, valid_prediction_mask, "itemPosition"
+        )
+        preds = _apply_valid_prediction_mask_to_dict(
+            preds, valid_prediction_mask, "preds"
+        )
+        probs = _apply_valid_prediction_mask_to_dict(
+            probs, valid_prediction_mask, "probs"
+        )
+
         os.makedirs(
             os.path.join(config.project_root, "outputs", "predictions"),
             exist_ok=True,
@@ -723,6 +852,7 @@ def infer_generative(
 
         n_input_cols = len(config.input_columns)
 
+        assert preds is not None
         predictions = pl.DataFrame(
             {
                 "sequenceId": sequence_ids_for_preds,
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 5888ada4..bfe486a1 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -936,12 +936,15 @@ def _selected_columns_with_optional_mask(
     selected_columns: Optional[list[str]],
     mask_column: Optional[str] = None,
 ) -> Optional[list[str]]:
-    if selected_columns is None or read_format != "parquet" or mask_column is None:
+    if selected_columns is None or mask_column is None:
         return selected_columns
 
+    if read_format != "parquet":
+        return _deduplicate_columns(selected_columns + [mask_column])
+
     schema_columns = pq.read_schema(data_path).names
     if mask_column in schema_columns:
-        return selected_columns + [mask_column]
+        return _deduplicate_columns(selected_columns + [mask_column])
     return selected_columns
 
 
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 900fb027..c24a0ea7 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1712,8 +1712,13 @@ def _calculate_loss(
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
+        categorical_target_columns = [
+            target_name
+            for target_name in target_names
+            if self.target_column_types[target_name] == "categorical"
+        ]
         valid_mask = infer_valid_mask_from_data(
-            targets, self.categorical_columns, "target_valid_mask", metadata
+            targets, categorical_target_columns, "target_valid_mask", metadata
         )
 
         if metadata is not None and "bert_mask" in metadata:
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 99f82f61..e668b1b5 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -4,6 +4,7 @@
 
 import numpy as np
 import polars as pl
+import torch
 
 TARGET_VARIABLE_DICT = {"categorical": "itemId", "real": "itemValue"}
 BERT_SEQ_LENGTH = 8
@@ -18,6 +19,63 @@ def _categorical_metadata(project_root):
         return json.load(f)
 
 
+def _bert_inference_metadata(project_root):
+    data_path = os.path.join(project_root, "data", "test-data-categorical-1-split2")
+    contents = []
+    for root, _, files in os.walk(data_path):
+        for file in sorted(files):
+            if not file.endswith(".pt"):
+                continue
+
+            loaded = torch.load(os.path.join(root, file), weights_only=False)
+            if len(loaded) == 5:
+                _, sequence_ids, subsequence_ids, _, left_pad_lengths = loaded
+            else:
+                _, sequence_ids, subsequence_ids, _ = loaded
+                left_pad_lengths = None
+
+            if left_pad_lengths is None:
+                left_pad_lengths = torch.zeros_like(sequence_ids)
+
+            valid_counts = torch.clamp(
+                BERT_SEQ_LENGTH - left_pad_lengths, min=0, max=BERT_SEQ_LENGTH
+            )
+            contents.append(
+                pl.DataFrame(
+                    {
+                        "sequenceId": sequence_ids.detach().cpu().numpy(),
+                        "subsequenceId": subsequence_ids.detach().cpu().numpy(),
+                        "valid_count": valid_counts.detach().cpu().numpy(),
+                    }
+                )
+            )
+
+    assert len(contents) > 0, f"no files found for {data_path}"
+    return pl.concat(contents, how="vertical")
+
+
+def _expected_bert_valid_counts(project_root, group_columns):
+    metadata = _bert_inference_metadata(project_root)
+
+    return (
+        metadata.group_by(group_columns)
+        .agg(pl.col("valid_count").sum().alias("expected_len"))
+        .filter(pl.col("expected_len") > 0)
+    )
+
+
+def _assert_counts_match_expected(actual, expected, group_columns):
+    actual_keys = set(actual.select(group_columns).iter_rows())
+    expected_keys = set(expected.select(group_columns).iter_rows())
+
+    assert actual_keys == expected_keys
+
+    comparison = actual.join(expected, on=group_columns)
+    assert comparison.select(
+        (pl.col("len") == pl.col("expected_len")).all()
+    ).item(), comparison
+
+
 def test_predictions_real(predictions):
     for model_name, model_predictions in predictions.items():
         if "categorical" not in model_name or "multitarget" in model_name:
@@ -121,7 +179,12 @@ def test_bert_generative_predictions_default_to_seq_length(
     assert set(bert_predictions["itemId"].to_list()).issubset(valid_values)
 
     rows_per_sequence = bert_predictions.group_by("sequenceId").len()
-    assert (rows_per_sequence["len"] == BERT_SEQ_LENGTH).all(), rows_per_sequence
+    expected_rows_per_sequence = _expected_bert_valid_counts(
+        project_root, ["sequenceId"]
+    )
+    _assert_counts_match_expected(
+        rows_per_sequence, expected_rows_per_sequence, ["sequenceId"]
+    )
 
 
 def test_bert_probabilities(bert_predictions, bert_probabilities, project_root):
@@ -172,7 +235,7 @@ def test_embeddings(embeddings):
             assert np.abs(model_embeddings[:, 1:].to_numpy().mean()) < 0.3
 
 
-def test_bert_embeddings(bert_predictions, bert_embeddings):
+def test_bert_embeddings(bert_predictions, bert_embeddings, project_root):
     expected_embedding_cols = [str(i) for i in range(BERT_EMBEDDING_DIM)]
     expected_columns = ["sequenceId", "subsequenceId", "itemPosition"]
 
@@ -184,8 +247,17 @@ def test_bert_embeddings(bert_predictions, bert_embeddings):
     embedding_values = bert_embeddings.select(expected_embedding_cols).to_numpy()
     assert np.isfinite(embedding_values).all()
 
-    rows_per_sequence = bert_embeddings.group_by("sequenceId").len()
-    assert (rows_per_sequence["len"] == BERT_SEQ_LENGTH).all(), rows_per_sequence
+    rows_per_subsequence = bert_embeddings.group_by(
+        ["sequenceId", "subsequenceId"]
+    ).len()
+    expected_rows_per_subsequence = _expected_bert_valid_counts(
+        project_root, ["sequenceId", "subsequenceId"]
+    )
+    _assert_counts_match_expected(
+        rows_per_subsequence,
+        expected_rows_per_subsequence,
+        ["sequenceId", "subsequenceId"],
+    )
 
 
 def test_bert_rejects_autoregression_end_to_end(
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 6642ba9e..1d370b26 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -1,6 +1,7 @@
 from unittest.mock import patch
 
 import numpy as np
+import polars as pl
 import pytest
 import torch
 
@@ -9,6 +10,8 @@
     Inferer,
     calculate_item_positions,
     get_probs_preds_from_dict,
+    infer_embedding,
+    infer_generative,
     normalize,
     sample_with_cumsum,
 )
@@ -178,6 +181,33 @@ def test_infer_config_defaults_bert_prediction_length_to_seq_length():
     assert config.prediction_length == config.seq_length
 
 
+def test_infer_config_defaults_causal_prediction_length_to_one():
+    config = InfererModel(
+        project_root=".",
+        metadata_config_path="dummy.json",
+        model_path="dummy.onnx",
+        model_type="generative",
+        training_objective="causal",
+        data_path="tests/unit/data/empty.parquet",
+        input_columns=["target_col"],
+        categorical_columns=[],
+        real_columns=["target_col"],
+        target_columns=["target_col"],
+        column_types={"target_col": "float64"},
+        target_column_types={"target_col": "real"},
+        seed=42,
+        device="cpu",
+        prediction_length=None,
+        seq_length=3,
+        inference_batch_size=2,
+        output_probabilities=False,
+        map_to_id=False,
+        autoregression=False,
+    )
+
+    assert config.prediction_length == 1
+
+
 def test_infer_config_rejects_bert_prediction_length_mismatch():
     with pytest.raises(ValueError, match="prediction_length must be equal"):
         InfererModel(
@@ -286,6 +316,136 @@ def test_calculate_item_positions_causal_uses_future_window_tail():
     np.testing.assert_array_equal(positions, [13, 14, 23, 24])
 
 
+def _bert_inference_config(tmp_path, model_type="generative"):
+    data_path = tmp_path / "data.parquet"
+    data_path.touch()
+
+    return InfererModel(
+        project_root=str(tmp_path),
+        metadata_config_path="dummy.json",
+        model_path="dummy.onnx",
+        model_type=model_type,
+        training_objective="bert",
+        data_path=str(data_path),
+        input_columns=["target_col"],
+        categorical_columns=["target_col"],
+        real_columns=[],
+        target_columns=["target_col"],
+        column_types={"target_col": "int64"},
+        target_column_types={"target_col": "categorical"},
+        seed=42,
+        device="cpu",
+        prediction_length=None,
+        seq_length=3,
+        inference_batch_size=4,
+        output_probabilities=False,
+        map_to_id=True,
+        autoregression=False,
+    )
+
+
+def _bert_preprocessed_frame():
+    return pl.DataFrame(
+        {
+            "sequenceId": [7],
+            "subsequenceId": [0],
+            "startItemPosition": [100],
+            "leftPadLength": [2],
+            "inputCol": ["target_col"],
+            "3": [0],
+            "2": [0],
+            "1": [3],
+            "0": [4],
+        }
+    )
+
+
+def _mock_bert_inferer(config):
+    return Inferer(
+        model_type=config.model_type,
+        model_path=config.model_path,
+        project_root=config.project_root,
+        id_maps={"target_col": {"A": 3, "B": 4}},
+        selected_columns_statistics={},
+        map_to_id=config.map_to_id,
+        categorical_columns=config.categorical_columns,
+        real_columns=config.real_columns,
+        input_columns=config.input_columns,
+        target_columns=config.target_columns,
+        target_column_types=config.target_column_types,
+        sample_from_distribution_columns=config.sample_from_distribution_columns,
+        infer_with_dropout=config.infer_with_dropout,
+        prediction_length=config.prediction_length,
+        inference_batch_size=config.inference_batch_size,
+        device=config.device,
+        args_config={},
+        training_config_path=config.training_config_path,
+    )
+
+
+def test_bert_generative_inference_filters_left_padded_positions(tmp_path):
+    config = _bert_inference_config(tmp_path)
+    config.output_probabilities = True
+    data = _bert_preprocessed_frame()
+    written = []
+
+    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
+        "sequifier.infer.load_inference_model"
+    ):
+        inferer = _mock_bert_inferer(config)
+
+    def capture_write(dataframe, path, write_format):
+        written.append((path, dataframe, write_format))
+
+    probs = {"target_col": np.full((3, 5), 0.2)}
+    preds = {"target_col": np.array([3, 4, 3])}
+    with patch(
+        "sequifier.infer.get_probs_preds_from_df", return_value=(probs, preds)
+    ), patch("sequifier.infer.write_data", side_effect=capture_write):
+        infer_generative(
+            config, inferer, "bert-model", [data], {"target_col": torch.int64}
+        )
+
+    predictions = next(frame for path, frame, _ in written if "predictions" in path)
+    probabilities = next(frame for path, frame, _ in written if "probabilities" in path)
+
+    assert predictions.height == 1
+    assert predictions.get_column("sequenceId").to_list() == [7]
+    assert predictions.get_column("itemPosition").to_list() == [102]
+    assert predictions.get_column("target_col").to_list() == ["A"]
+    assert probabilities.height == 1
+
+
+def test_bert_embedding_inference_filters_left_padded_positions(tmp_path):
+    config = _bert_inference_config(tmp_path, model_type="embedding")
+    data = _bert_preprocessed_frame()
+    written = []
+
+    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
+        "sequifier.infer.load_inference_model"
+    ):
+        inferer = _mock_bert_inferer(config)
+
+    def capture_write(dataframe, path, write_format):
+        written.append((path, dataframe, write_format))
+
+    embeddings = np.array([[0.1, 1.1], [0.2, 1.2], [0.3, 1.3]])
+    with patch("sequifier.infer.get_embeddings", return_value=embeddings), patch(
+        "sequifier.infer.write_data", side_effect=capture_write
+    ):
+        infer_embedding(
+            config, inferer, "bert-model", [data], {"target_col": torch.int64}
+        )
+
+    embeddings_df = next(frame for _, frame, _ in written)
+
+    assert embeddings_df.height == 1
+    assert embeddings_df.get_column("sequenceId").to_list() == [7]
+    assert embeddings_df.get_column("subsequenceId").to_list() == [0]
+    assert embeddings_df.get_column("itemPosition").to_list() == [102]
+    assert embeddings_df.select(["0", "1"]).row(0) == (0.3, 1.3)
+
+
 # ==========================================
 # Test Autoregressive Tensor Inference
 # ==========================================
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index c7c736a7..a592e758 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -13,6 +13,7 @@
     _apply_mask_column,
     _get_column_statistics,
     _get_data_columns,
+    _load_and_preprocess_data,
     create_id_map,
     extract_sequences,
     extract_subsequences,
@@ -251,6 +252,47 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
     assert RESERVED_MASK_COLUMN not in metadata["selected_columns_statistics"]
 
 
+def test_load_and_preprocess_data_requests_mask_column_for_csv_projection(tmp_path):
+    data_path = tmp_path / "masked-input.csv"
+    captured_columns = None
+
+    def projected_read_data(path, read_format, columns=None):
+        nonlocal captured_columns
+        captured_columns = columns
+        assert path == str(data_path)
+        assert read_format == "csv"
+        assert columns is not None
+        assert RESERVED_MASK_COLUMN in columns
+        return pl.DataFrame(
+            {
+                "sequenceId": [0],
+                "itemPosition": [0],
+                "itemId": ["a"],
+                "itemValue": [1.0],
+                RESERVED_MASK_COLUMN: [1],
+            }
+        )
+
+    with pytest.MonkeyPatch.context() as monkeypatch:
+        monkeypatch.setattr("sequifier.preprocess.read_data", projected_read_data)
+        data = _load_and_preprocess_data(
+            str(data_path),
+            "csv",
+            ["itemId", "itemValue"],
+            None,
+            RESERVED_MASK_COLUMN,
+        )
+
+    assert captured_columns == ["itemId", "itemValue", RESERVED_MASK_COLUMN]
+    assert data.columns == [
+        "sequenceId",
+        "itemPosition",
+        "itemId",
+        "itemValue",
+        RESERVED_MASK_COLUMN,
+    ]
+
+
 def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
     project_root = tmp_path / "project"
     project_root.mkdir()
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index ee5ae7ab..67d642f9 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -4,6 +4,7 @@
 import torch
 from pydantic import ValidationError
 
+from sequifier.config.probabilities import PoissonDistributionFloor
 from sequifier.config.train_config import (
     BERTSpecModel,
     ModelSpecModel,
@@ -65,6 +66,24 @@ def test_training_spec_model_rejects_bert_spec_for_causal_objective():
         TrainingSpecModel(**_training_spec_kwargs(bert_spec=_bert_spec()))
 
 
+def test_training_spec_model_dump_excludes_runtime_offsets():
+    training_spec = TrainingSpecModel(**_training_spec_kwargs())
+
+    dumped = training_spec.model_dump()
+
+    assert "data_offset" not in dumped
+    assert "target_offset" not in dumped
+    assert TrainingSpecModel(**dumped).target_offset == 0
+
+
+def test_poisson_span_masking_samples_at_least_one_token():
+    distribution = PoissonDistributionFloor(rate=0.1)
+
+    samples = distribution.sample((1000,), device=torch.device("cpu"))
+
+    assert samples.min().item() >= 1
+
+
 @pytest.fixture
 def model_config(tmp_path):
     """Creates a valid TrainModel configuration for testing."""
@@ -138,8 +157,13 @@ def causal_model(model_config):
 
 @pytest.fixture
 def bert_model(model_config):
-    config = copy.deepcopy(model_config)
-    config.training_spec.training_objective = "bert"
+    config_values = model_config.model_dump()
+    config_values["model_spec"]["prediction_length"] = model_config.seq_length
+    config_values["training_spec"] = _training_spec_kwargs(
+        training_objective="bert",
+        bert_spec=_bert_spec(),
+    )
+    config = TrainModel(**config_values)
     return TransformerModel(config)
 
 
@@ -302,6 +326,34 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
     assert torch.isclose(component_losses["real_col"], torch.tensor(2.0))
 
 
+def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_column_types = {"real_target": "real"}
+    model.criterion = {"real_target": torch.nn.MSELoss(reduction="none")}
+    model.loss_weights = None
+    model.categorical_columns = ["context_cat"]
+    outputs = {
+        "real_target": torch.tensor(
+            [
+                [[10.0]],
+                [[20.0]],
+                [[3.0]],
+            ]
+        )
+    }
+    targets = {
+        "real_target": torch.tensor([[0.0, 0.0, 2.0]]),
+    }
+
+    with pytest.warns(UserWarning):
+        total_loss, component_losses = TransformerModel._calculate_loss(
+            model, outputs, targets
+        )
+
+    assert torch.isclose(total_loss, torch.tensor(1.0))
+    assert torch.isclose(component_losses["real_target"], torch.tensor(1.0))
+
+
 def test_infer_valid_mask_from_data():
     model = TransformerModel.__new__(TransformerModel)
     model.categorical_columns = []

From c9f0514606b7668fe439e0d220245c3fd362a3df Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 16:17:52 +0200
Subject: [PATCH 28/81] infer bert valid column

---
 src/sequifier/infer.py   | 108 +++++++++++++++++++++++++++++++++++----
 tests/unit/test_infer.py |  73 ++++++++++++++++++++++++++
 2 files changed, 171 insertions(+), 10 deletions(-)

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 16a6c176..6cd28bc3 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -18,7 +18,6 @@
     configure_determinism,
     construct_index_maps,
     generate_padding_masks,
-    get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
@@ -324,6 +323,70 @@ def _flatten_bert_target_valid_mask(
     return valid_mask.reshape(-1)
 
 
+@beartype
+def _is_categorical_column(config: InfererModel, column_name: str) -> bool:
+    column_type = str(config.column_types.get(column_name, "")).lower()
+    return (
+        column_name in config.categorical_columns
+        or config.target_column_types.get(column_name) == "categorical"
+        or column_type.startswith("int")
+    )
+
+
+@beartype
+def _infer_bert_valid_mask_from_values(
+    config: InfererModel,
+    column_name: str,
+    values: np.ndarray,
+    prediction_length: int,
+) -> Optional[np.ndarray]:
+    if _is_categorical_column(config, column_name):
+        valid_mask = values != 0
+    else:
+        valid_mask = np.cumsum(values != 0.0, axis=1) > 0
+
+    return _flatten_bert_target_valid_mask(
+        config, {"target_valid_mask": valid_mask}, prediction_length
+    )
+
+
+@beartype
+def _bert_reference_column(config: InfererModel, data_columns: set[str]) -> str:
+    preferred_columns = (
+        [col for col in config.target_columns if col in config.categorical_columns]
+        + [col for col in config.input_columns if col in config.categorical_columns]
+        + [col for col in config.target_columns if col in data_columns]
+        + [col for col in config.input_columns if col in data_columns]
+    )
+    for column_name in preferred_columns:
+        if column_name in data_columns:
+            return column_name
+
+    raise ValueError("Could not find a reference column for BERT valid-mask inference.")
+
+
+@beartype
+def _flatten_legacy_bert_target_valid_mask_from_sequences(
+    config: InfererModel,
+    sequences: dict[str, Union[torch.Tensor, np.ndarray]],
+    prediction_length: int,
+) -> Optional[np.ndarray]:
+    if config.training_objective != "bert":
+        return None
+
+    column_name = _bert_reference_column(config, set(sequences.keys()))
+    values = sequences[column_name]
+    if isinstance(values, torch.Tensor):
+        values = values.detach().cpu().numpy()  # type: ignore
+    values = np.asarray(values)
+    if values.ndim == 2 and values.shape[1] > prediction_length:
+        values = values[:, :prediction_length]
+
+    return _infer_bert_valid_mask_from_values(
+        config, column_name, values, prediction_length
+    )
+
+
 @beartype
 def _bert_target_valid_mask_from_preprocessed_data(
     config: InfererModel,
@@ -333,17 +396,30 @@ def _bert_target_valid_mask_from_preprocessed_data(
     if config.training_objective != "bert":
         return None
 
-    left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
-    if left_pad_lengths is None:
-        return None
+    data_columns = set(data.get_column("inputCol").unique())
+    column_name = _bert_reference_column(config, data_columns)
+    reference_rows = data.filter(pl.col("inputCol") == column_name)
 
-    metadata = generate_padding_masks(
-        left_pad_lengths,
-        config.seq_length,
-        data_offset=1,
-        target_offset=1,
+    if "leftPadLength" in reference_rows.columns:
+        left_pad_lengths = torch.tensor(
+            reference_rows.get_column("leftPadLength").to_numpy(), dtype=torch.int64
+        )
+        metadata = generate_padding_masks(
+            left_pad_lengths,
+            config.seq_length,
+            data_offset=1,
+            target_offset=1,
+        )
+        return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
+
+    target_seq_cols = [
+        str(c)
+        for c in range(config.seq_length, config.seq_length - prediction_length, -1)
+    ]
+    values = reference_rows.select(target_seq_cols).to_numpy()
+    return _infer_bert_valid_mask_from_values(
+        config, column_name, values, prediction_length
     )
-    return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
 
 
 @beartype
@@ -482,6 +558,12 @@ def infer_embedding(
             valid_prediction_mask = _flatten_bert_target_valid_mask(
                 config, metadata, prediction_length
             )
+            if valid_prediction_mask is None:
+                valid_prediction_mask = (
+                    _flatten_legacy_bert_target_valid_mask_from_sequences(
+                        config, sequences_dict, prediction_length
+                    )
+                )
 
             sequence_ids_for_preds = sequence_ids_tensor.numpy()
             subsequence_ids_for_preds = subsequence_ids_tensor.numpy()
@@ -744,6 +826,12 @@ def infer_generative(
             valid_prediction_mask = _flatten_bert_target_valid_mask(
                 config, metadata, prediction_length
             )
+            if valid_prediction_mask is None:
+                valid_prediction_mask = (
+                    _flatten_legacy_bert_target_valid_mask_from_sequences(
+                        config, sequences_dict, prediction_length
+                    )
+                )
 
             if total_steps == 1:
                 # Non-autoregressive path: Apply prediction_length logic
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 1d370b26..d858fbe5 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -360,6 +360,22 @@ def _bert_preprocessed_frame():
     )
 
 
+def _bert_preprocessed_frame_out_of_order():
+    return pl.DataFrame(
+        {
+            "sequenceId": [2, 1],
+            "subsequenceId": [0, 0],
+            "startItemPosition": [200, 100],
+            "leftPadLength": [1, 2],
+            "inputCol": ["target_col", "target_col"],
+            "3": [0, 0],
+            "2": [3, 0],
+            "1": [4, 3],
+            "0": [3, 4],
+        }
+    )
+
+
 def _mock_bert_inferer(config):
     return Inferer(
         model_type=config.model_type,
@@ -416,6 +432,63 @@ def capture_write(dataframe, path, write_format):
     assert probabilities.height == 1
 
 
+def test_bert_generative_inference_uses_dataframe_order_for_padding_masks(tmp_path):
+    config = _bert_inference_config(tmp_path)
+    data = _bert_preprocessed_frame_out_of_order()
+    written = []
+
+    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
+        "sequifier.infer.load_inference_model"
+    ):
+        inferer = _mock_bert_inferer(config)
+
+    def capture_write(dataframe, path, write_format):
+        written.append((path, dataframe, write_format))
+
+    preds = {"target_col": np.array([3, 4, 3, 4, 3, 4])}
+    with patch(
+        "sequifier.infer.get_probs_preds_from_df", return_value=(None, preds)
+    ), patch("sequifier.infer.write_data", side_effect=capture_write):
+        infer_generative(
+            config, inferer, "bert-model", [data], {"target_col": torch.int64}
+        )
+
+    predictions = next(frame for path, frame, _ in written if "predictions" in path)
+
+    assert predictions.get_column("sequenceId").to_list() == [2, 2, 1]
+    assert predictions.get_column("itemPosition").to_list() == [201, 202, 102]
+    assert predictions.get_column("target_col").to_list() == ["B", "A", "B"]
+
+
+def test_bert_generative_inference_filters_legacy_data_without_left_pad(tmp_path):
+    config = _bert_inference_config(tmp_path)
+    data = _bert_preprocessed_frame().drop("leftPadLength")
+    written = []
+
+    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
+        "sequifier.infer.load_inference_model"
+    ):
+        inferer = _mock_bert_inferer(config)
+
+    def capture_write(dataframe, path, write_format):
+        written.append((path, dataframe, write_format))
+
+    preds = {"target_col": np.array([3, 4, 3])}
+    with patch(
+        "sequifier.infer.get_probs_preds_from_df", return_value=(None, preds)
+    ), patch("sequifier.infer.write_data", side_effect=capture_write):
+        infer_generative(
+            config, inferer, "bert-model", [data], {"target_col": torch.int64}
+        )
+
+    predictions = next(frame for path, frame, _ in written if "predictions" in path)
+
+    assert predictions.height == 1
+    assert predictions.get_column("sequenceId").to_list() == [7]
+    assert predictions.get_column("itemPosition").to_list() == [102]
+    assert predictions.get_column("target_col").to_list() == ["A"]
+
+
 def test_bert_embedding_inference_filters_left_padded_positions(tmp_path):
     config = _bert_inference_config(tmp_path, model_type="embedding")
     data = _bert_preprocessed_frame()

From 6c593f566dbdf1b6d2265a88f7d7acbd8cc53efc Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 16:42:55 +0200
Subject: [PATCH 29/81] Small fixes

---
 src/sequifier/helpers.py    |  2 +-
 src/sequifier/preprocess.py | 20 +++-----------------
 2 files changed, 4 insertions(+), 18 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 85f46d41..3bff44fe 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -93,7 +93,7 @@ def construct_index_maps(
     A special mapping for these indices are added:
     - If original IDs are strings, 0 maps to "[unknown]".
     - If original IDs are strings, 1 maps to "[other]".
-    - If original IDs are strings, 2 maps to "[unknown]".
+    - If original IDs are strings, 2 maps to "[mask]".
     - If original IDs are integers, 0 maps to (minimum original ID) - 3.
     - If original IDs are integers, 1 maps to (minimum original ID) - 2.
     - If original IDs are integers, 2 maps to (minimum original ID) - 1.
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index bfe486a1..bb5de7c0 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -952,27 +952,13 @@ def _selected_columns_with_optional_mask(
 def _mask_column_expr(mask_dtype: Any, mask_column: str) -> pl.Expr:
     mask_col = pl.col(mask_column)
 
-    if isinstance(mask_dtype, pl.Boolean):
+    if mask_dtype == pl.Boolean:
         return mask_col
 
-    if isinstance(
-        mask_dtype,
-        (
-            pl.Int8,
-            pl.Int16,
-            pl.Int32,
-            pl.Int64,
-            pl.UInt8,
-            pl.UInt16,
-            pl.UInt32,
-            pl.UInt64,
-            pl.Float32,
-            pl.Float64,
-        ),
-    ):
+    if mask_dtype.is_numeric():
         return mask_col == 1
 
-    if isinstance(mask_dtype, (pl.String, pl.Utf8)):
+    if mask_dtype in (pl.String, pl.Utf8):
         return mask_col.str.to_lowercase().is_in(["1"])
 
     raise ValueError(

From 2e8c2a423f66f8b413defcaebd020b6a3d631376 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 18:02:28 +0200
Subject: [PATCH 30/81] Clean up internal abstractions

---
 src/sequifier/config/probabilities.py         |   8 +-
 src/sequifier/config/train_config.py          |   7 +
 src/sequifier/helpers.py                      |  33 +-
 src/sequifier/io/batch.py                     |  21 +
 .../io/sequifier_dataset_from_file.py         |  27 +-
 .../sequifier_dataset_from_folder_parquet.py  |  19 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  31 +-
 .../io/sequifier_dataset_from_folder_pt.py    |  19 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  31 +-
 src/sequifier/preprocess.py                   |  95 +++--
 src/sequifier/special_tokens.py               |  29 ++
 src/sequifier/train.py                        |  44 +-
 tests/integration/test_preprocessing.py       |   6 +
 ...est-data-categorical-precomputed-stats.csv | 402 +++++++++---------
 tests/unit/test_helpers.py                    |  10 +-
 tests/unit/test_preprocess.py                 |  35 ++
 16 files changed, 473 insertions(+), 344 deletions(-)
 create mode 100644 src/sequifier/io/batch.py
 create mode 100644 src/sequifier/special_tokens.py

diff --git a/src/sequifier/config/probabilities.py b/src/sequifier/config/probabilities.py
index a7656f1e..231dd6a1 100644
--- a/src/sequifier/config/probabilities.py
+++ b/src/sequifier/config/probabilities.py
@@ -43,11 +43,11 @@ def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
         return torch.clamp(torch.round(val), min=0).long() + 1
 
 
-class LogNormalDistributionDistcretizedFloor(
+class LogNormalDistributionDiscretizedFloor(
     BaseModel, ProbabilityDistributionBaseClass
 ):
-    type: Literal["LogNormalDistributionDistcretizedFloor"] = (
-        "LogNormalDistributionDistcretizedFloor"
+    type: Literal["LogNormalDistributionDiscretizedFloor",] = (
+        "LogNormalDistributionDiscretizedFloor"
     )
     mean: float
     standard_deviation: float = Field(..., gt=0.0)
@@ -74,7 +74,7 @@ def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
     Union[
         GeometricDistribution,
         NormalDistributionDiscretizedFloor,
-        LogNormalDistributionDistcretizedFloor,
+        LogNormalDistributionDiscretizedFloor,
         PoissonDistributionFloor,
     ],
     Field(discriminator="type"),
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index fd0dfa22..1a7f8078 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -23,6 +23,7 @@
 import sequifier
 from sequifier.config.probabilities import ProbabilityDistribution
 from sequifier.helpers import normalize_path, try_catch_excess_keys
+from sequifier.special_tokens import SPECIAL_TOKEN_IDS
 
 AnyType = str | int | float
 
@@ -97,6 +98,9 @@ def load_train_config(
         )
 
         config_values["id_maps"] = metadata_config["id_maps"]
+        config_values["special_token_ids"] = metadata_config.get(
+            "special_token_ids", SPECIAL_TOKEN_IDS.ids_by_label
+        )
 
     return try_catch_excess_keys(config_path, TrainModel, config_values)
 
@@ -553,6 +557,9 @@ class TrainModel(BaseModel):
     target_columns: list[str]
     target_column_types: dict[str, str]
     id_maps: dict[str, dict[str | int, int]]
+    special_token_ids: dict[str, int] = Field(
+        default_factory=lambda: SPECIAL_TOKEN_IDS.ids_by_label
+    )
 
     seq_length: int
     n_classes: dict[str, int]
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 3bff44fe..3f463757 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -15,6 +15,8 @@
 from pydantic import ValidationError
 from torch import Tensor
 
+from sequifier.special_tokens import SPECIAL_TOKEN_IDS
+
 PANDAS_TO_TORCH_TYPES = {
     "Float64": torch.float32,
     "float64": torch.float32,
@@ -90,13 +92,8 @@ def construct_index_maps(
     the original string or integer identifiers. It only performs this
     operation if `map_to_id` is True and `id_maps` is provided.
 
-    A special mapping for these indices are added:
-    - If original IDs are strings, 0 maps to "[unknown]".
-    - If original IDs are strings, 1 maps to "[other]".
-    - If original IDs are strings, 2 maps to "[mask]".
-    - If original IDs are integers, 0 maps to (minimum original ID) - 3.
-    - If original IDs are integers, 1 maps to (minimum original ID) - 2.
-    - If original IDs are integers, 2 maps to (minimum original ID) - 1.
+    Special reserved IDs are always decoded as their sentinel labels:
+    0 maps to "[unknown]", 1 maps to "[other]", and 2 maps to "[mask]".
 
     Args:
         id_maps: A nested dictionary mapping column names to their
@@ -126,17 +123,11 @@ def construct_index_maps(
         for target_column in target_columns_index_map:
             map_ = {v: k for k, v in id_maps[target_column].items()}
             val = next(iter(map_.values()))
-            if isinstance(val, str):
-                map_[0] = "[unknown]"
-                map_[1] = "[other]"
-                map_[2] = "[mask]"
-            else:
-                if not isinstance(val, int):
-                    raise TypeError(f"Expected integer ID in map, got {type(val)}")
-                min_id = int(min(map_.values()))
-                map_[0] = min_id - 3  # type: ignore
-                map_[1] = min_id - 2  # type: ignore
-                map_[2] = min_id - 1  # type: ignore
+            if not isinstance(val, (str, int)):
+                raise TypeError(
+                    f"Expected string or integer ID in map, got {type(val)}"
+                )
+            map_.update(SPECIAL_TOKEN_IDS.labels_by_id)
             index_map[target_column] = map_
     return index_map
 
@@ -738,17 +729,15 @@ def apply_bert_masking(
     # 5. Apply corruption to data_batch
     for col, tensor in data_batch.items():
         if col in config.categorical_columns:
-            mask_id = 2
-
             random_tokens = torch.randint(
-                low=3,
+                low=SPECIAL_TOKEN_IDS.user_start,
                 high=config.n_classes[col],
                 size=(batch_size, seq_len),
                 device=device,
                 dtype=tensor.dtype,
             )
 
-            tensor[mask_token_mask] = mask_id
+            tensor[mask_token_mask] = SPECIAL_TOKEN_IDS.mask
             tensor[random_token_mask] = random_tokens[random_token_mask]
 
         elif col in config.real_columns:
diff --git a/src/sequifier/io/batch.py b/src/sequifier/io/batch.py
new file mode 100644
index 00000000..ba2d7330
--- /dev/null
+++ b/src/sequifier/io/batch.py
@@ -0,0 +1,21 @@
+from dataclasses import dataclass
+from typing import Iterator, Optional
+
+import torch
+
+
+@dataclass(frozen=True)
+class SequifierBatch:
+    inputs: dict[str, torch.Tensor]
+    targets: dict[str, torch.Tensor]
+    metadata: dict[str, torch.Tensor]
+    sequence_ids: Optional[torch.Tensor] = None
+    subsequence_ids: Optional[torch.Tensor] = None
+    start_positions: Optional[torch.Tensor] = None
+
+    def __iter__(self) -> Iterator[object]:
+        yield self.inputs
+        yield self.targets
+        yield self.metadata
+        yield self.sequence_ids
+        yield self.subsequence_ids
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index ffc7d2aa..87203d2e 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -1,5 +1,5 @@
 import math
-from typing import Dict, Iterator, Tuple
+from typing import Iterator
 
 import torch
 import torch.distributed as dist
@@ -8,6 +8,7 @@
 
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import PANDAS_TO_TORCH_TYPES, numpy_to_pytorch, read_data
+from sequifier.io.batch import SequifierBatch
 
 
 class SequifierDatasetFromFile(IterableDataset):
@@ -84,28 +85,14 @@ def __len__(self) -> int:
 
     def __iter__(
         self,
-    ) -> Iterator[
-        Tuple[
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            None,
-            None,
-        ]
-    ]:
+    ) -> Iterator[SequifierBatch]:
         """Yields batches of data.
 
         Handles shuffling (if enabled) and slicing data based on distributed
         rank and worker ID.
 
         Yields:
-            Iterator[Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None]]:
-            An iterator where each item is a tuple containing:
-                - data_batch (dict): Dictionary of feature tensors for the batch.
-                - targets_batch (dict): Dictionary of target tensors for the batch.
-                - metadata_batch (dict): Dictionary of metadata tensors for the batch.
-                - None: Placeholder for sequence_id (not used in this dataset type).
-                - None: Placeholder for subsequence_id (not used in this dataset type).
+            An iterator where each item is a SequifierBatch.
         """
         worker_info = torch.utils.data.get_worker_info()
         world_size = dist.get_world_size() if dist.is_initialized() else 1
@@ -150,4 +137,8 @@ def __iter__(
                 for key, tensor in self.metadata_tensors.items()
             }
 
-            yield data_batch, targets_batch, metadata_batch, None, None
+            yield SequifierBatch(
+                inputs=data_batch,
+                targets=targets_batch,
+                metadata=metadata_batch,
+            )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 99e58efc..2d2cae63 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -1,7 +1,7 @@
 import json
 import math
 import os
-from typing import Dict, Iterator, Tuple
+from typing import Dict, Iterator
 
 import polars as pl
 import torch
@@ -16,6 +16,7 @@
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
 )
+from sequifier.io.batch import SequifierBatch
 
 
 class SequifierDatasetFromFolderParquet(IterableDataset):
@@ -182,15 +183,7 @@ def __len__(self) -> int:
 
     def __iter__(
         self,
-    ) -> Iterator[
-        Tuple[
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            None,
-            None,
-        ]
-    ]:
+    ) -> Iterator[SequifierBatch]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
 
@@ -243,4 +236,8 @@ def __iter__(
                     self.config.training_spec.target_offset,
                 )
 
-            yield data_batch, targets_batch, metadata_batch, None, None
+            yield SequifierBatch(
+                inputs=data_batch,
+                targets=targets_batch,
+                metadata=metadata_batch,
+            )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index f4120cc3..c922d88c 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -1,7 +1,7 @@
 import json
 import math
 import os
-from typing import Dict, Iterator, Tuple
+from typing import Dict, Iterator
 
 import numpy as np
 import polars as pl
@@ -17,6 +17,7 @@
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
 )
+from sequifier.io.batch import SequifierBatch
 
 
 class SequifierDatasetFromFolderParquetLazy(IterableDataset):
@@ -34,9 +35,9 @@ class SequifierDatasetFromFolderParquetLazy(IterableDataset):
             sample indices per epoch. Defaults to True.
 
     Yields:
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None]:
-            A batch tuple containing sequence dictionaries, target dictionaries,
-            metadata dictionaries, and two `None` placeholders (for API compatibility).
+        SequifierBatch:
+            A named batch containing sequence dictionaries, target dictionaries,
+            metadata dictionaries, and optional identifiers.
 
     Raises:
         FileNotFoundError: If `metadata.json` is missing.
@@ -155,15 +156,7 @@ def __len__(self) -> int:
 
     def __iter__(
         self,
-    ) -> Iterator[
-        Tuple[
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            None,
-            None,
-        ]
-    ]:
+    ) -> Iterator[SequifierBatch]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
 
@@ -366,7 +359,11 @@ def __iter__(
                 batch_tgt = {k: v[: self.batch_size] for k, v in tgt_buffer.items()}
                 batch_meta = {k: v[: self.batch_size] for k, v in meta_buffer.items()}
 
-                yield batch_seq, batch_tgt, batch_meta, None, None
+                yield SequifierBatch(
+                    inputs=batch_seq,
+                    targets=batch_tgt,
+                    metadata=batch_meta,
+                )
                 yielded_samples += self.batch_size
 
                 # Keep the remainder in the buffer for the next loop/file
@@ -384,4 +381,8 @@ def __iter__(
             batch_tgt = {k: v[:final_yield_size] for k, v in tgt_buffer.items()}
             batch_meta = {k: v[:final_yield_size] for k, v in meta_buffer.items()}
 
-            yield batch_seq, batch_tgt, batch_meta, None, None
+            yield SequifierBatch(
+                inputs=batch_seq,
+                targets=batch_tgt,
+                metadata=batch_meta,
+            )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 85a5d258..f689ac73 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -1,7 +1,7 @@
 import json
 import math
 import os
-from typing import Dict, Iterator, Tuple
+from typing import Dict, Iterator
 
 import torch
 import torch.distributed as dist
@@ -14,6 +14,7 @@
     normalize_path,
     unpack_preprocessed_pt_tuple,
 )
+from sequifier.io.batch import SequifierBatch
 
 
 class SequifierDatasetFromFolderPt(IterableDataset):
@@ -132,15 +133,7 @@ def __len__(self) -> int:
 
     def __iter__(
         self,
-    ) -> Iterator[
-        Tuple[
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            None,
-            None,
-        ]
-    ]:
+    ) -> Iterator[SequifierBatch]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
 
@@ -208,4 +201,8 @@ def __iter__(
                     target_offset,
                 )
 
-            yield data_batch, targets_batch, metadata_batch, None, None
+            yield SequifierBatch(
+                inputs=data_batch,
+                targets=targets_batch,
+                metadata=metadata_batch,
+            )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 41d5f115..37d7940f 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -1,7 +1,7 @@
 import json
 import math
 import os
-from typing import Dict, Iterator, Tuple
+from typing import Dict, Iterator
 
 import numpy as np
 import torch
@@ -15,6 +15,7 @@
     normalize_path,
     unpack_preprocessed_pt_tuple,
 )
+from sequifier.io.batch import SequifierBatch
 
 
 class SequifierDatasetFromFolderPtLazy(IterableDataset):
@@ -32,9 +33,9 @@ class SequifierDatasetFromFolderPtLazy(IterableDataset):
             sample indices per epoch. Defaults to True.
 
     Yields:
-        Tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor], None, None]:
-            A batch tuple containing sequence dictionaries, target dictionaries,
-            metadata dictionaries, and two `None` placeholders (for API compatibility).
+        SequifierBatch:
+            A named batch containing sequence dictionaries, target dictionaries,
+            metadata dictionaries, and optional identifiers.
 
     Raises:
         FileNotFoundError: If `metadata.json` is missing.
@@ -148,15 +149,7 @@ def __len__(self) -> int:
 
     def __iter__(
         self,
-    ) -> Iterator[
-        Tuple[
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            Dict[str, torch.Tensor],
-            None,
-            None,
-        ]
-    ]:
+    ) -> Iterator[SequifierBatch]:
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
 
@@ -331,7 +324,11 @@ def __iter__(
                 batch_tgt = {k: v[: self.batch_size] for k, v in tgt_buffer.items()}
                 batch_meta = {k: v[: self.batch_size] for k, v in meta_buffer.items()}
 
-                yield batch_seq, batch_tgt, batch_meta, None, None
+                yield SequifierBatch(
+                    inputs=batch_seq,
+                    targets=batch_tgt,
+                    metadata=batch_meta,
+                )
                 yielded_samples += self.batch_size
 
                 # Keep the remainder in the buffer for the next loop/file
@@ -349,4 +346,8 @@ def __iter__(
             batch_tgt = {k: v[:final_yield_size] for k, v in tgt_buffer.items()}
             batch_meta = {k: v[:final_yield_size] for k, v in meta_buffer.items()}
 
-            yield batch_seq, batch_tgt, batch_meta, None, None
+            yield SequifierBatch(
+                inputs=batch_seq,
+                targets=batch_tgt,
+                metadata=batch_meta,
+            )
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index bb5de7c0..46747d4a 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -22,12 +22,14 @@
     unpack_preprocessed_pt_tuple,
     write_data,
 )
+from sequifier.special_tokens import (
+    SPECIAL_TOKEN_ID_VALUES,
+    SPECIAL_TOKEN_IDS,
+    SPECIAL_TOKEN_LABELS,
+)
 
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
-CATEGORICAL_MASK_ID = 2
 REAL_MASK_VALUE = 0.0
-RESERVED_ID_KEYS = {"[unknown]", "[other]", "[mask]"}
-RESERVED_ID_VALUES = {0, 1, 2}
 
 
 @beartype
@@ -812,6 +814,7 @@ def _export_metadata(
         data_driven_config = {
             "n_classes": n_classes,
             "id_maps": id_maps,
+            "special_token_ids": SPECIAL_TOKEN_IDS.ids_by_label,
             "split_paths": [
                 os.path.splitext(split_path)[0] if not self.merge_output else split_path
                 for split_path in self.split_paths
@@ -945,7 +948,7 @@ def _selected_columns_with_optional_mask(
     schema_columns = pq.read_schema(data_path).names
     if mask_column in schema_columns:
         return _deduplicate_columns(selected_columns + [mask_column])
-    return selected_columns
+    raise ValueError(f"mask_column '{mask_column}' not found in {data_path}")
 
 
 @beartype
@@ -973,14 +976,16 @@ def _apply_mask_column(
     col_types: dict[str, str],
     mask_column: Optional[str] = None,
 ) -> pl.DataFrame:
-    if mask_column is None or mask_column not in data.columns:
+    if mask_column is None:
         return data
+    if mask_column not in data.columns:
+        raise ValueError(f"mask_column '{mask_column}' not found in input data")
 
     mask_expr = _mask_column_expr(data.schema[mask_column], mask_column)
     updates = []
     for col in data_columns:
         mask_value = (
-            CATEGORICAL_MASK_ID if col_types[col] == "Int64" else REAL_MASK_VALUE
+            SPECIAL_TOKEN_IDS.mask if col_types[col] == "Int64" else REAL_MASK_VALUE
         )
         updates.append(
             pl.when(mask_expr)
@@ -1108,18 +1113,19 @@ def load_precomputed_id_maps(
 
                     if not len(m) > 0:
                         raise ValueError(f"map in {file} does not contain any values")
-                    for reserved_key, expected_value in {
-                        "[unknown]": 0,
-                        "[other]": 1,
-                        "[mask]": 2,
-                    }.items():
+                    for (
+                        reserved_key,
+                        expected_value,
+                    ) in SPECIAL_TOKEN_IDS.ids_by_label.items():
                         if reserved_key in m and m[reserved_key] != expected_value:
                             raise ValueError(
                                 f"{reserved_key} in map {file} must map to {expected_value}"
                             )
 
                     user_values = [
-                        value for key, value in m.items() if key not in RESERVED_ID_KEYS
+                        value
+                        for key, value in m.items()
+                        if key not in SPECIAL_TOKEN_LABELS
                     ]
                     if not user_values:
                         raise ValueError(
@@ -1134,24 +1140,25 @@ def load_precomputed_id_maps(
                         )
                         m = {
                             key: value + 1
-                            if key not in RESERVED_ID_KEYS and value >= 2
+                            if key not in SPECIAL_TOKEN_LABELS
+                            and value >= SPECIAL_TOKEN_IDS.mask
                             else value
                             for key, value in m.items()
                         }
                         user_values = [
                             value
                             for key, value in m.items()
-                            if key not in RESERVED_ID_KEYS
+                            if key not in SPECIAL_TOKEN_LABELS
                         ]
                         min_val = min(user_values)
 
-                    if min_val != 3:
+                    if min_val != SPECIAL_TOKEN_IDS.user_start:
                         raise ValueError(
-                            f"minimum non-reserved value in map {file} is {min_val}, must be 3."
+                            f"minimum non-reserved value in map {file} is {min_val}, must be {SPECIAL_TOKEN_IDS.user_start}."
                         )
-                    if any(value in RESERVED_ID_VALUES for value in user_values):
+                    if any(value in SPECIAL_TOKEN_ID_VALUES for value in user_values):
                         raise ValueError(
-                            f"non-reserved values in map {file} must not use reserved IDs {RESERVED_ID_VALUES}"
+                            f"non-reserved values in map {file} must not use reserved IDs {SPECIAL_TOKEN_ID_VALUES}"
                         )
                     if len(set(m.values())) != len(m.values()):
                         raise ValueError(f"map in {file} contains duplicate IDs")
@@ -1311,6 +1318,9 @@ def _load_and_preprocess_data(
     )
     data = read_data(data_path, read_format, columns=columns_to_read)
 
+    if mask_column is not None and mask_column not in data.columns:
+        raise ValueError(f"mask_column '{mask_column}' not found in {data_path}")
+
     if data.null_count().sum().sum_horizontal().item() != 0:
         raise ValueError(f"NaN or null values not accepted: {data.null_count()}")
 
@@ -1682,20 +1692,37 @@ def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]
     )  # type: ignore
 
     if isinstance(ids[0], str):
-        if "[mask]" in ids:
-            raise ValueError(f"Found value '[mask]' in {column}, this is invalid")
+        if SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.mask] in ids:
+            raise ValueError(
+                f"Found value '{SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.mask]}' in {column}, this is invalid"
+            )
 
-        for special_val in ["[unknown]", "[other]"]:
+        for special_val in [
+            SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.unknown],
+            SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.other],
+        ]:
             if special_val in ids:
                 warnings.warn(
                     f"Found special value {special_val} in {column}, these will be combined with the sequifier-internal special value {special_val}"
                 )
-        ids = [id_ for id_ in ids if id_ not in ["[unknown]", "[other]"]]
-        id_map = {id_: i + 3 for i, id_ in enumerate(ids)}
-        id_map["[unknown]"] = 0
-        id_map["[other]"] = 1
+        ids = [
+            id_
+            for id_ in ids
+            if id_
+            not in [
+                SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.unknown],
+                SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.other],
+            ]
+        ]
+        id_map = {id_: i + SPECIAL_TOKEN_IDS.user_start for i, id_ in enumerate(ids)}
+        id_map[SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.unknown]] = (
+            SPECIAL_TOKEN_IDS.unknown
+        )
+        id_map[SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.other]] = (
+            SPECIAL_TOKEN_IDS.other
+        )
     else:
-        id_map = {id_: i + 3 for i, id_ in enumerate(ids)}
+        id_map = {id_: i + SPECIAL_TOKEN_IDS.user_start for i, id_ in enumerate(ids)}
     return dict(id_map)
 
 
@@ -1762,16 +1789,22 @@ def combine_maps(
     Returns:
         A new, combined, and re-indexed ID map.
     """
-    keys1 = {k for k in map1.keys() if k not in ["[unknown]", "[other]", "[mask]"]}
-    keys2 = {k for k in map2.keys() if k not in ["[unknown]", "[other]", "[mask]"]}
+    keys1 = {k for k in map1.keys() if k not in SPECIAL_TOKEN_LABELS}
+    keys2 = {k for k in map2.keys() if k not in SPECIAL_TOKEN_LABELS}
 
     combined_keys = sorted(list(keys1.union(keys2)))
-    id_map = {id_: i + 3 for i, id_ in enumerate(combined_keys)}
+    id_map = {
+        id_: i + SPECIAL_TOKEN_IDS.user_start for i, id_ in enumerate(combined_keys)
+    }
 
     # Re-apply special mapping rules if these are string keys
     if combined_keys and isinstance(combined_keys[0], str):
-        id_map["[unknown]"] = 0
-        id_map["[other]"] = 1
+        id_map[SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.unknown]] = (
+            SPECIAL_TOKEN_IDS.unknown
+        )
+        id_map[SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.other]] = (
+            SPECIAL_TOKEN_IDS.other
+        )
 
     return id_map
 
diff --git a/src/sequifier/special_tokens.py b/src/sequifier/special_tokens.py
new file mode 100644
index 00000000..9f1cd4ba
--- /dev/null
+++ b/src/sequifier/special_tokens.py
@@ -0,0 +1,29 @@
+from dataclasses import asdict, dataclass
+
+
+@dataclass(frozen=True)
+class SpecialTokenIds:
+    unknown: int = 0
+    other: int = 1
+    mask: int = 2
+
+    @property
+    def user_start(self) -> int:
+        return max(asdict(self).values()) + 1
+
+    @property
+    def labels_by_id(self) -> dict[int, str]:
+        return {
+            self.unknown: "[unknown]",
+            self.other: "[other]",
+            self.mask: "[mask]",
+        }
+
+    @property
+    def ids_by_label(self) -> dict[str, int]:
+        return {label: id_ for id_, label in self.labels_by_id.items()}
+
+
+SPECIAL_TOKEN_IDS = SpecialTokenIds()
+SPECIAL_TOKEN_LABELS = frozenset(SPECIAL_TOKEN_IDS.ids_by_label.keys())
+SPECIAL_TOKEN_ID_VALUES = frozenset(SPECIAL_TOKEN_IDS.labels_by_id.keys())
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index c24a0ea7..b3f0b284 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -64,6 +64,7 @@
     infer_valid_mask_from_data,
     normalize_path,
 )
+from sequifier.io.batch import SequifierBatch  # noqa: E402
 from sequifier.io.sequifier_dataset_from_file import (  # noqa: E402
     SequifierDatasetFromFile,
 )
@@ -83,6 +84,20 @@
 from sequifier.optimizers.optimizers import get_optimizer_class  # noqa: E402
 
 
+def _as_sequifier_batch(batch: Any) -> SequifierBatch:
+    if isinstance(batch, SequifierBatch):
+        return batch
+
+    data, targets, metadata, sequence_ids, subsequence_ids = batch
+    return SequifierBatch(
+        inputs=data,
+        targets=targets,
+        metadata=metadata or {},
+        sequence_ids=sequence_ids,
+        subsequence_ids=subsequence_ids,
+    )
+
+
 def cleanup():
     """Cleans up the distributed training environment."""
     dist.destroy_process_group()
@@ -1496,9 +1511,7 @@ def _train_epoch(
 
         Iterates through the training DataLoader, computes loss, performs
         backpropagation, and updates model parameters. The DataLoader is expected
-        to yield tuples of
-        (sequences_dict, targets_dict, metadata_dict, sequence_ids, subsequence_ids).
-        The IDs are currently unused in this training loop.
+        to yield SequifierBatch objects. IDs are currently unused in this loop.
 
         Args:
             train_loader: DataLoader for the training dataset.
@@ -1516,8 +1529,12 @@ def _train_epoch(
 
         model_to_call.train()
 
-        for batch_count, (data, targets, metadata, _, _) in enumerate(train_loader):
+        for batch_count, raw_batch in enumerate(train_loader):
+            batch = _as_sequifier_batch(raw_batch)
             if batch_count >= start_batch:
+                data = batch.inputs
+                targets = batch.targets
+                metadata = batch.metadata
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
@@ -1828,10 +1845,9 @@ def _evaluate(
 
         Iterates through the validation data, calculates the total loss,
         and aggregates results across all processes if in distributed mode.
-        Also calculates a one-time baseline loss on the first call.
-        The DataLoader is expected to yield tuples of
-        (sequences_dict, targets_dict, metadata_dict, sequence_ids, subsequence_ids).
-        The IDs are currently unused during evaluation.
+        Also calculates a one-time baseline loss on the first call. The
+        DataLoader is expected to yield SequifierBatch objects. IDs are
+        currently unused during evaluation.
 
         Args:
             valid_loader: DataLoader for the validation dataset.
@@ -1854,7 +1870,11 @@ def _evaluate(
         model_to_call.eval()
 
         with torch.no_grad():
-            for data, targets, metadata, _, _ in valid_loader:
+            for raw_batch in valid_loader:
+                batch = _as_sequifier_batch(raw_batch)
+                data = batch.inputs
+                targets = batch.targets
+                metadata = batch.metadata
                 # Move data to the current process's assigned GPU
                 data = {
                     k: v.to(self.device, non_blocking=True)
@@ -1954,7 +1974,11 @@ def _evaluate(
             baseline_losses_local_collect = {col: [] for col in self.target_columns}
 
             # Iterate over the sharded validation loader
-            for data, targets, metadata, _, _ in valid_loader:
+            for raw_batch in valid_loader:
+                batch = _as_sequifier_batch(raw_batch)
+                data = batch.inputs
+                targets = batch.targets
+                metadata = batch.metadata
                 data = {
                     k: v.to(self.device, non_blocking=True)
                     for k, v in data.items()
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index cdd42c52..95061cf3 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -32,12 +32,18 @@ def test_metadata_config(metadata_configs):
                 [
                     "n_classes",
                     "id_maps",
+                    "special_token_ids",
                     "split_paths",
                     "column_types",
                     "selected_columns_statistics",
                 ]
             )
         ), list(metadata_config.keys())
+        assert metadata_config["special_token_ids"] == {
+            "[unknown]": 0,
+            "[other]": 1,
+            "[mask]": 2,
+        }
 
         assert metadata_config["split_paths"][0].endswith(
             "split0.parquet"
diff --git a/tests/resources/source_data/test-data-categorical-precomputed-stats.csv b/tests/resources/source_data/test-data-categorical-precomputed-stats.csv
index 02a630c2..734bd449 100644
--- a/tests/resources/source_data/test-data-categorical-precomputed-stats.csv
+++ b/tests/resources/source_data/test-data-categorical-precomputed-stats.csv
@@ -1,201 +1,201 @@
-sequenceId,itemPosition,itemId,supCat1,supReal2,supReal3,supCat4
-0,0,113,3,0.23756075956718,-1.1640173812934225,3
-0,1,105,9,-0.5630867234310606,-1.8178494433325032,1
-0,2,129,7,-0.6081001558815576,-1.0308743424505562,1
-0,3,121,9,-0.7466738515871298,-1.759307333132087,1
-0,4,109,3,-0.4835408143281817,-0.1893670050382644,3
-0,5,111,7,0.14757929805884176,1.146907612603052,1
-0,6,121,9,-0.4128405939888088,-1.3452711173264285,1
-0,7,109,3,-0.13133273866941814,1.4676755703847157,1
-0,8,127,5,0.6564099027980262,0.8247687731058516,1
-0,9,113,3,-0.08622570845242584,2.0005516149382276,3
-0,10,105,5,0.4992279385752224,-3.358606755136373,1
-0,11,103,9,0.4120884761752859,1.0384301395677145,1
-0,12,119,3,0.5762410188436872,-0.1369896760325402,3
-0,13,103,1,-0.22661438614011545,-2.025340358480932,1
-0,14,105,7,0.37034788642948163,0.9196011438572373,3
-0,15,121,7,-0.1523339501381893,4.457098431914344,1
-0,16,107,1,-0.0490514356525169,3.161523505159044,3
-0,17,129,5,-0.1693437999308125,2.054921627292583,1
-0,18,129,7,-0.2963164638064198,1.4211553800634344,3
-0,19,117,9,-0.43418074401427853,1.4000470430791148,1
-0,20,107,3,-0.11119720264051855,-0.2842816816898542,3
-0,21,103,9,0.1119258339375208,1.5342499637561926,1
-0,22,119,7,-0.2397209205593547,0.568539328380014,1
-0,23,105,9,-0.1043511841935035,1.073596528818807,1
-0,24,101,5,0.0636689964273863,1.8286076605291732,1
-1,25,113,9,-0.13283334035174255,0.9612448513077018,1
-1,26,117,1,-0.4877130507537292,0.5881920065149198,1
-1,27,105,5,-1.0319853596529749,2.4266792951807745,1
-1,28,113,7,0.36599244564499156,-0.8458755927714368,1
-1,29,109,1,-0.04541393204777325,1.1332991030875792,3
-1,30,119,5,0.14270870818902634,1.2019717678164248,1
-1,31,103,3,0.1007341413628698,3.4185328489051545,1
-1,32,103,1,0.0726528173264286,0.6798517808089142,1
-1,33,121,7,0.4047583876998109,-0.215857840126759,1
-1,34,105,3,0.7950212886366006,1.258841265676203,3
-1,35,111,1,-0.40894101251224185,2.9266389229634195,1
-1,36,111,7,-0.5905624213594342,1.1007581452294355,1
-1,37,103,5,-0.27605549137493063,0.4576222503937258,3
-1,38,103,7,0.19748743112184536,3.089575720800873,1
-1,39,121,1,-0.3231988066797896,0.3260653118807056,3
-1,40,109,7,-0.6402978908699791,-2.3652965225104947,1
-1,41,101,9,0.2754450193286252,-0.3343460991871334,1
-1,42,111,1,0.04265434446667645,-0.0265267517674288,1
-1,43,127,3,-0.1093848556929905,-1.4039708736281316,3
-1,44,105,1,-0.3098861853313818,4.277223053793332,1
-1,45,115,7,-0.5623720760818156,-0.9282928823756458,3
-1,46,105,5,-0.0362778238543955,-0.785870762785741,1
-2,47,113,3,0.6699513286275102,3.756180727360066,1
-2,48,125,5,-0.0121909350376935,1.0632670323412912,1
-2,49,121,1,-0.4985428652715131,-2.767024017505266,3
-2,50,125,9,-0.41674804550372435,-1.8232361265594708,1
-2,51,129,7,0.1922800647827131,0.3614326479800766,1
-2,52,103,3,-0.1147113810473422,0.6955419567131546,1
-2,53,119,7,0.5084886555692772,-1.182177344128451,1
-2,54,119,5,0.1681706731512443,-2.0002345186543655,1
-2,55,111,1,0.00489090849851505,-1.2119677670104652,1
-2,56,103,3,0.2950952608133617,0.7392696815922633,1
-2,57,121,9,-0.1664963816784592,-4.3184593675736345,1
-2,58,109,5,0.2734968675602151,0.2317458237066366,1
-2,59,101,7,0.7193973857758358,1.020369498449154,3
-2,60,103,5,-0.7964210408690493,-2.0363883178208826,1
-2,61,111,9,0.2072449425144576,-4.202452168223529,3
-2,62,107,9,0.4813276271478582,-0.5282806781620062,1
-2,63,103,5,0.29694766144249257,2.252849978564028,1
-2,64,111,5,-0.34264442506928156,-0.5576317791480824,1
-2,65,109,5,-0.901152134284954,-0.7768569166101674,1
-2,66,115,3,-0.00647098093573455,-1.939123138639253,1
-3,67,123,9,0.3640480787779695,3.15721060624824,1
-3,68,129,5,0.3578139552426438,-0.9705160338846076,1
-3,69,101,1,0.6773419994294364,2.623083513295643,1
-3,70,117,3,-0.6161806906782396,-1.8750394780263255,1
-3,71,121,1,-0.02179363892395065,-2.986785300875957,1
-3,72,127,9,0.2742262667237867,3.33396269826164,1
-3,73,123,5,0.653923817402728,0.8431916125568506,1
-3,74,113,5,-1.309266107419916,3.2749345537056085,3
-3,75,127,7,-0.4248256569891429,0.5290052043252962,1
-3,76,115,9,-0.3760627852144511,1.5052140738898157,1
-3,77,117,1,-1.063352767838854,0.4364134544435038,1
-3,78,103,5,0.43763228075223376,0.5460770179634292,3
-3,79,121,7,0.17808332843594354,-2.3590410762810787,3
-3,80,109,7,-0.2604307066744918,-0.600994152922156,1
-3,81,127,9,0.06496062311791385,-1.6810104306763307,3
-3,82,117,7,-0.1523655063137391,1.4713636745100194,3
-3,83,103,5,0.9557698802026362,-0.4777913312818128,1
-3,84,121,1,-0.4283939310570219,3.286310865387594,3
-3,85,119,9,0.4683163969196768,1.0639713340534525,1
-3,86,105,5,0.05918350344033825,-4.265334468962421,1
-4,87,113,5,-0.07365771506054584,-3.2085100052938498,1
-4,88,127,3,-0.6528270195518846,-1.199488174092495,1
-4,89,105,7,1.5137471671937042,-0.1041592978187344,1
-4,90,105,9,0.18086471642780114,-0.176760115308051,1
-4,91,107,9,-0.22268250180214355,-1.6668522895897546,1
-4,92,119,1,0.31414268525117434,2.700642938436344,3
-4,93,113,1,0.4821281052755843,-1.9984526206862492,1
-4,94,123,1,-0.41654505373555256,-1.5270811322998106,3
-4,95,105,1,0.3990717968896191,-1.3333943830007091,3
-4,96,121,7,0.39975842804009404,-7.460240076256136,3
-4,97,119,9,-0.2389382834262071,-3.6036268736043153,1
-4,98,127,1,0.8865821752236414,-5.933173903688966,1
-4,99,121,9,0.6200500637175907,1.9526039321781257,1
-4,100,103,7,-0.3086399573460153,-3.868319973856011,1
-4,101,117,1,0.3763432482016413,0.0230803370202396,1
-4,102,115,7,-0.19610200435266445,-3.5946735502342606,1
-5,103,121,5,0.6582222235652958,-1.1565601648186674,1
-5,104,123,5,0.210010355223184,-0.3700868001061204,1
-5,105,127,1,0.44609895345157075,-0.466451727035951,1
-5,106,105,5,0.21395314818454306,-3.079000434320467,1
-5,107,113,7,-0.09045776647608085,-2.21295241292887,1
-5,108,109,5,0.26679432858260427,-4.258487041288806,1
-5,109,123,9,-0.3912608511206187,0.9536287033277712,1
-5,110,109,7,-0.6648528537220284,-0.3446684601197502,1
-5,111,129,9,-0.4949563299989164,-1.2691154528617874,3
-5,112,123,1,-0.6120796408055832,-2.7176268721322505,1
-5,113,117,7,-0.8040575605720383,-2.230618406212464,1
-5,114,123,3,0.34466826874757156,3.320167025422368,1
-5,115,109,3,0.6592255898925946,1.1963021486138496,1
-5,116,113,1,0.1265463872076663,0.3524305338372076,3
-5,117,127,9,-0.36090671060617907,0.1991192313196762,1
-5,118,121,9,-0.00346023525880035,-0.269946254030518,3
-6,119,127,1,-0.2069274087283918,-1.4983857095036925,1
-6,120,119,9,-0.4608668090772096,4.141066849383603,3
-6,121,127,5,-0.543241871075917,-1.429667411117142,1
-6,122,103,3,0.7371442736167835,0.3262826938556598,1
-6,123,113,7,0.7937790994016082,2.054537435734958,1
-6,124,119,5,-0.9494328754610578,3.7617736489864257,3
-6,125,103,3,-0.2025407208280338,0.8874555973490652,3
-6,126,127,1,0.4363248490262944,0.2481060743864992,1
-6,127,129,5,0.9112368560642352,0.8419542731986736,3
-6,128,113,5,0.1646877914705859,1.2879311306352885,1
-6,129,111,7,0.5676868132357109,-0.1265805632469282,1
-6,130,117,9,-1.044462070351678,1.5747984413324092,3
-6,131,125,9,0.5888200685806166,0.3311529110566518,3
-6,132,101,3,0.1550655289088061,-1.087772015925998,1
-6,133,119,5,0.02448451869814495,-2.7168204655897465,1
-6,134,117,5,-0.5612827926145505,-0.3608931971219676,1
-6,135,103,7,-0.06895955594486534,2.0089156749900283,1
-6,136,111,5,0.2386687846790385,1.5074623202579414,3
-6,137,109,7,-0.0103767345335778,1.1682437308875186,1
-6,138,125,5,0.4674416024517608,-1.4452965786865868,1
-6,139,107,9,-0.402113804564477,-1.1143202352937496,1
-7,140,111,9,0.07159698573628046,0.0845048113931306,3
-7,141,113,3,-0.23027849149205426,-0.8651366428920254,1
-7,142,113,7,-0.5276752352570413,1.6931263448599303,1
-7,143,113,1,0.36030134407682074,0.5742417223803036,3
-7,144,113,9,-0.8591185434489855,-1.573618378735019,1
-7,145,121,1,0.2637783382546025,-3.863689604625783,1
-7,146,101,5,0.7364946010807949,1.6783637541403809,1
-7,147,129,7,-0.5029781169022207,0.0039304229456184,1
-7,148,101,5,0.8152784745739302,2.1903743861120266,1
-7,149,125,7,1.2736852756936394,-2.0417356426375637,1
-7,150,119,9,1.016662275589319,-0.2309438830049552,1
-7,151,121,7,0.2021641470325514,-1.7414085189195898,3
-7,152,127,9,0.2095640210815988,0.1915711997670132,1
-7,153,121,1,-0.7793534531754494,0.3583296644699246,1
-7,154,129,1,-0.39154291689852044,0.1297020398076264,1
-7,155,113,9,0.0685599894350678,0.4323507757122852,1
-7,156,111,3,0.7749898540861211,2.0689152027698867,1
-8,157,101,7,0.909287892921888,-3.989987895098015,3
-8,158,119,5,0.0591576225354796,-2.0688934315523086,1
-8,159,103,5,0.38263311649619963,-2.003805618784399,1
-8,160,113,7,0.7843252853779198,-1.1533873434116515,3
-8,161,105,7,0.06279078494571524,-1.8773457835545853,3
-8,162,111,9,0.2169085162118607,-2.033257200716173,1
-8,163,125,7,0.724906768076498,2.5248601944100124,1
-8,164,109,9,0.43106360552415107,-1.0569306831210186,1
-8,165,119,5,0.1617027306762413,2.323383376870655,1
-8,166,125,3,0.25320441824075834,-0.930273413013715,1
-8,167,109,1,0.7456736036850744,0.598252461847501,1
-8,168,121,9,0.5118838476588286,-0.110896649044434,1
-8,169,111,5,-0.7218907289288162,1.198758968337366,1
-8,170,121,5,-0.6157395896817728,0.7802693999604342,1
-8,171,115,5,0.38369214592817863,1.4790116390747352,3
-8,172,107,9,0.8423963059206273,-0.5237249538681794,3
-8,173,101,3,0.22574490676417786,-0.8904664351067142,1
-8,174,111,7,0.7851195032274458,1.5226194800279191,1
-8,175,107,9,-0.7116886776940682,-4.5883589812374606,3
-8,176,109,5,-0.41740781716411335,0.6369119820796147,1
-8,177,115,1,0.82150802012012,-1.6210202504559137,1
-8,178,103,7,0.29659052088695614,-0.8867954709909982,3
-8,179,127,3,0.1466956467187299,-0.4310364763509028,1
-8,180,115,3,0.5223660767025526,3.348885082673797,3
-9,181,129,1,0.3090525876709156,-2.7416777723376646,1
-9,182,121,1,-0.21520439985795284,0.6307442663117814,1
-9,183,103,5,0.21220901158898275,-1.8188459028762447,3
-9,184,117,5,-0.02897385433854525,-1.440215891320799,1
-9,185,123,7,0.19749026974006945,-0.779497479448969,1
-9,186,101,7,-0.8052615487503672,-1.4344762234821997,1
-9,187,129,1,-1.121699959433816,0.4896249408465866,1
-9,188,103,7,0.4649808838118618,-0.198806681337086,1
-9,189,123,7,-0.7073291167356331,-3.043576509789646,1
-9,190,111,5,-0.3740773149679098,0.8608798093746408,1
-9,191,123,7,0.4989177413451416,1.6835410358613347,1
-9,192,107,5,-0.6634732155934616,-0.3061356639836578,3
-9,193,123,3,-0.7887391302965693,-1.3438864368895593,3
-9,194,103,9,-0.04733828356775615,3.656344892439264,1
-9,195,129,9,-0.5213560739887506,-2.4135537875363897,1
-9,196,107,7,-0.43137296420290955,-2.2356462270857462,1
-9,197,109,3,0.04976548623382745,1.2900131060668418,1
-9,198,119,5,-0.285600783170167,5.764139626487539,1
-9,199,129,5,-0.5218301693993399,-1.1335640211899563,1
+sequenceId,itemPosition,itemId,supCat1,supReal2,supReal3,supCat4,[mask]
+0,0,113,3,0.23756075956718,-1.1640173812934225,3,0
+0,1,105,9,-0.5630867234310606,-1.8178494433325032,1,0
+0,2,129,7,-0.6081001558815576,-1.0308743424505562,1,0
+0,3,121,9,-0.7466738515871298,-1.759307333132087,1,0
+0,4,109,3,-0.4835408143281817,-0.1893670050382644,3,0
+0,5,111,7,0.1475792980588417,1.146907612603052,1,0
+0,6,121,9,-0.4128405939888088,-1.3452711173264285,1,0
+0,7,109,3,-0.1313327386694181,1.4676755703847155,1,0
+0,8,127,5,0.6564099027980262,0.8247687731058516,1,0
+0,9,113,3,-0.0862257084524258,2.0005516149382276,3,0
+0,10,105,5,0.4992279385752224,-3.358606755136373,1,0
+0,11,103,9,0.4120884761752859,1.0384301395677145,1,0
+0,12,119,3,0.5762410188436872,-0.1369896760325402,3,0
+0,13,103,1,-0.2266143861401154,-2.025340358480932,1,0
+0,14,105,7,0.3703478864294816,0.9196011438572372,3,0
+0,15,121,7,-0.1523339501381893,4.457098431914344,1,0
+0,16,107,1,-0.0490514356525169,3.161523505159044,3,0
+0,17,129,5,-0.1693437999308125,2.054921627292583,1,0
+0,18,129,7,-0.2963164638064198,1.4211553800634344,3,0
+0,19,117,9,-0.4341807440142785,1.4000470430791148,1,0
+0,20,107,3,-0.1111972026405185,-0.2842816816898542,3,0
+0,21,103,9,0.1119258339375208,1.5342499637561926,1,0
+0,22,119,7,-0.2397209205593547,0.568539328380014,1,0
+0,23,105,9,-0.1043511841935035,1.073596528818807,1,0
+0,24,101,5,0.0636689964273863,1.8286076605291732,1,0
+1,25,113,9,-0.1328333403517425,0.9612448513077018,1,0
+1,26,117,1,-0.4877130507537292,0.5881920065149198,1,0
+1,27,105,5,-1.0319853596529749,2.4266792951807745,1,0
+1,28,113,7,0.3659924456449915,-0.8458755927714368,1,0
+1,29,109,1,-0.0454139320477732,1.1332991030875792,3,0
+1,30,119,5,0.1427087081890263,1.2019717678164248,1,0
+1,31,103,3,0.1007341413628698,3.4185328489051545,1,0
+1,32,103,1,0.0726528173264286,0.6798517808089142,1,0
+1,33,121,7,0.4047583876998109,-0.215857840126759,1,0
+1,34,105,3,0.7950212886366006,1.258841265676203,3,0
+1,35,111,1,-0.4089410125122418,2.9266389229634195,1,0
+1,36,111,7,-0.5905624213594342,1.1007581452294355,1,0
+1,37,103,5,-0.2760554913749306,0.4576222503937258,3,0
+1,38,103,7,0.1974874311218453,3.089575720800873,1,0
+1,39,121,1,-0.3231988066797896,0.3260653118807056,3,0
+1,40,109,7,-0.6402978908699791,-2.3652965225104947,1,0
+1,41,101,9,0.2754450193286252,-0.3343460991871334,1,0
+1,42,111,1,0.0426543444666764,-0.0265267517674288,1,0
+1,43,127,3,-0.1093848556929905,-1.4039708736281316,3,0
+1,44,105,1,-0.3098861853313818,4.277223053793332,1,0
+1,45,115,7,-0.5623720760818156,-0.9282928823756458,3,0
+1,46,105,5,-0.0362778238543955,-0.785870762785741,1,0
+2,47,113,3,0.6699513286275102,3.756180727360066,1,0
+2,48,125,5,-0.0121909350376935,1.0632670323412912,1,0
+2,49,121,1,-0.4985428652715131,-2.767024017505266,3,0
+2,50,125,9,-0.4167480455037243,-1.8232361265594708,1,0
+2,51,129,7,0.1922800647827131,0.3614326479800766,1,0
+2,52,103,3,-0.1147113810473422,0.6955419567131546,1,0
+2,53,119,7,0.5084886555692772,-1.182177344128451,1,0
+2,54,119,5,0.1681706731512443,-2.0002345186543655,1,0
+2,55,111,1,0.004890908498515,-1.2119677670104652,1,0
+2,56,103,3,0.2950952608133617,0.7392696815922633,1,0
+2,57,121,9,-0.1664963816784592,-4.3184593675736345,1,0
+2,58,109,5,0.2734968675602151,0.2317458237066366,1,0
+2,59,101,7,0.7193973857758358,1.020369498449154,3,0
+2,60,103,5,-0.7964210408690493,-2.0363883178208826,1,0
+2,61,111,9,0.2072449425144576,-4.202452168223529,3,0
+2,62,107,9,0.4813276271478582,-0.5282806781620062,1,0
+2,63,103,5,0.2969476614424925,2.252849978564028,1,0
+2,64,111,5,-0.3426444250692815,-0.5576317791480824,1,0
+2,65,109,5,-0.901152134284954,-0.7768569166101674,1,0
+2,66,115,3,-0.0064709809357345,-1.939123138639253,1,0
+3,67,123,9,0.3640480787779695,3.15721060624824,1,0
+3,68,129,5,0.3578139552426438,-0.9705160338846076,1,0
+3,69,101,1,0.6773419994294364,2.623083513295643,1,0
+3,70,117,3,-0.6161806906782396,-1.8750394780263255,1,0
+3,71,121,1,-0.0217936389239506,-2.986785300875957,1,0
+3,72,127,9,0.2742262667237867,3.33396269826164,1,0
+3,73,123,5,0.653923817402728,0.8431916125568506,1,0
+3,74,113,5,-1.309266107419916,3.2749345537056085,3,0
+3,75,127,7,-0.4248256569891429,0.5290052043252962,1,0
+3,76,115,9,-0.3760627852144511,1.5052140738898157,1,0
+3,77,117,1,-1.063352767838854,0.4364134544435038,1,0
+3,78,103,5,0.4376322807522337,0.5460770179634292,3,0
+3,79,121,7,0.1780833284359435,-2.3590410762810787,3,0
+3,80,109,7,-0.2604307066744918,-0.600994152922156,1,0
+3,81,127,9,0.0649606231179138,-1.6810104306763307,3,0
+3,82,117,7,-0.1523655063137391,1.4713636745100194,3,0
+3,83,103,5,0.9557698802026362,-0.4777913312818128,1,0
+3,84,121,1,-0.4283939310570219,3.286310865387594,3,0
+3,85,119,9,0.4683163969196768,1.0639713340534525,1,0
+3,86,105,5,0.0591835034403382,-4.265334468962421,1,0
+4,87,113,5,-0.0736577150605458,-3.2085100052938498,1,0
+4,88,127,3,-0.6528270195518846,-1.199488174092495,1,0
+4,89,105,7,1.5137471671937042,-0.1041592978187344,1,0
+4,90,105,9,0.1808647164278011,-0.176760115308051,1,0
+4,91,107,9,-0.2226825018021435,-1.6668522895897546,1,0
+4,92,119,1,0.3141426852511743,2.700642938436344,3,0
+4,93,113,1,0.4821281052755843,-1.9984526206862492,1,0
+4,94,123,1,-0.4165450537355525,-1.5270811322998106,3,0
+4,95,105,1,0.3990717968896191,-1.3333943830007091,3,0
+4,96,121,7,0.399758428040094,-7.460240076256136,3,0
+4,97,119,9,-0.2389382834262071,-3.6036268736043153,1,0
+4,98,127,1,0.8865821752236414,-5.933173903688966,1,0
+4,99,121,9,0.6200500637175907,1.9526039321781257,1,0
+4,100,103,7,-0.3086399573460153,-3.868319973856011,1,0
+4,101,117,1,0.3763432482016413,0.0230803370202396,1,0
+4,102,115,7,-0.1961020043526644,-3.5946735502342606,1,0
+5,103,121,5,0.6582222235652958,-1.1565601648186674,1,0
+5,104,123,5,0.210010355223184,-0.3700868001061204,1,0
+5,105,127,1,0.4460989534515707,-0.466451727035951,1,0
+5,106,105,5,0.213953148184543,-3.079000434320467,1,0
+5,107,113,7,-0.0904577664760808,-2.21295241292887,1,0
+5,108,109,5,0.2667943285826042,-4.258487041288806,1,0
+5,109,123,9,-0.3912608511206187,0.9536287033277712,1,0
+5,110,109,7,-0.6648528537220284,-0.3446684601197502,1,0
+5,111,129,9,-0.4949563299989164,-1.2691154528617874,3,0
+5,112,123,1,-0.6120796408055832,-2.7176268721322505,1,0
+5,113,117,7,-0.8040575605720383,-2.230618406212464,1,0
+5,114,123,3,0.3446682687475715,3.320167025422368,1,0
+5,115,109,3,0.6592255898925946,1.1963021486138496,1,0
+5,116,113,1,0.1265463872076663,0.3524305338372076,3,0
+5,117,127,9,-0.360906710606179,0.1991192313196762,1,0
+5,118,121,9,-0.0034602352588003,-0.269946254030518,3,0
+6,119,127,1,-0.2069274087283918,-1.4983857095036923,1,0
+6,120,119,9,-0.4608668090772096,4.141066849383603,3,0
+6,121,127,5,-0.543241871075917,-1.429667411117142,1,0
+6,122,103,3,0.7371442736167835,0.3262826938556598,1,0
+6,123,113,7,0.7937790994016082,2.054537435734958,1,0
+6,124,119,5,-0.9494328754610578,3.7617736489864257,3,0
+6,125,103,3,-0.2025407208280338,0.8874555973490652,3,0
+6,126,127,1,0.4363248490262944,0.2481060743864992,1,0
+6,127,129,5,0.9112368560642352,0.8419542731986736,3,0
+6,128,113,5,0.1646877914705859,1.2879311306352883,1,0
+6,129,111,7,0.5676868132357109,-0.1265805632469282,1,0
+6,130,117,9,-1.044462070351678,1.5747984413324092,3,0
+6,131,125,9,0.5888200685806166,0.3311529110566518,3,0
+6,132,101,3,0.1550655289088061,-1.087772015925998,1,0
+6,133,119,5,0.0244845186981449,-2.7168204655897465,1,0
+6,134,117,5,-0.5612827926145505,-0.3608931971219676,1,0
+6,135,103,7,-0.0689595559448653,2.0089156749900283,1,0
+6,136,111,5,0.2386687846790385,1.5074623202579414,3,0
+6,137,109,7,-0.0103767345335778,1.1682437308875186,1,0
+6,138,125,5,0.4674416024517608,-1.4452965786865868,1,0
+6,139,107,9,-0.402113804564477,-1.1143202352937496,1,0
+7,140,111,9,0.0715969857362804,0.0845048113931306,3,0
+7,141,113,3,-0.2302784914920542,-0.8651366428920254,1,0
+7,142,113,7,-0.5276752352570413,1.6931263448599303,1,0
+7,143,113,1,0.3603013440768207,0.5742417223803036,3,0
+7,144,113,9,-0.8591185434489855,-1.573618378735019,1,0
+7,145,121,1,0.2637783382546025,-3.863689604625783,1,0
+7,146,101,5,0.7364946010807949,1.6783637541403809,1,0
+7,147,129,7,-0.5029781169022207,0.0039304229456184,1,0
+7,148,101,5,0.8152784745739302,2.1903743861120266,1,0
+7,149,125,7,1.2736852756936394,-2.0417356426375637,1,0
+7,150,119,9,1.016662275589319,-0.2309438830049552,1,0
+7,151,121,7,0.2021641470325514,-1.7414085189195898,3,0
+7,152,127,9,0.2095640210815988,0.1915711997670132,1,0
+7,153,121,1,-0.7793534531754494,0.3583296644699246,1,0
+7,154,129,1,-0.3915429168985204,0.1297020398076264,1,0
+7,155,113,9,0.0685599894350678,0.4323507757122852,1,0
+7,156,111,3,0.7749898540861211,2.0689152027698867,1,0
+8,157,101,7,0.909287892921888,-3.989987895098015,3,0
+8,158,119,5,0.0591576225354796,-2.0688934315523086,1,0
+8,159,103,5,0.3826331164961996,-2.003805618784399,1,0
+8,160,113,7,0.7843252853779198,-1.1533873434116515,3,0
+8,161,105,7,0.0627907849457152,-1.8773457835545853,3,0
+8,162,111,9,0.2169085162118607,-2.033257200716173,1,0
+8,163,125,7,0.724906768076498,2.5248601944100124,1,0
+8,164,109,9,0.431063605524151,-1.0569306831210186,1,0
+8,165,119,5,0.1617027306762413,2.323383376870655,1,0
+8,166,125,3,0.2532044182407583,-0.930273413013715,1,0
+8,167,109,1,0.7456736036850744,0.598252461847501,1,0
+8,168,121,9,0.5118838476588286,-0.110896649044434,1,0
+8,169,111,5,-0.7218907289288162,1.198758968337366,1,0
+8,170,121,5,-0.6157395896817728,0.7802693999604342,1,0
+8,171,115,5,0.3836921459281786,1.4790116390747352,3,0
+8,172,107,9,0.8423963059206273,-0.5237249538681794,3,0
+8,173,101,3,0.2257449067641778,-0.8904664351067142,1,0
+8,174,111,7,0.7851195032274458,1.5226194800279191,1,0
+8,175,107,9,-0.7116886776940682,-4.5883589812374606,3,0
+8,176,109,5,-0.4174078171641133,0.6369119820796147,1,0
+8,177,115,1,0.82150802012012,-1.6210202504559137,1,0
+8,178,103,7,0.2965905208869561,-0.8867954709909982,3,0
+8,179,127,3,0.1466956467187299,-0.4310364763509028,1,0
+8,180,115,3,0.5223660767025526,3.348885082673797,3,0
+9,181,129,1,0.3090525876709156,-2.7416777723376646,1,0
+9,182,121,1,-0.2152043998579528,0.6307442663117814,1,0
+9,183,103,5,0.2122090115889827,-1.8188459028762447,3,0
+9,184,117,5,-0.0289738543385452,-1.440215891320799,1,0
+9,185,123,7,0.1974902697400694,-0.779497479448969,1,0
+9,186,101,7,-0.8052615487503672,-1.4344762234821995,1,0
+9,187,129,1,-1.121699959433816,0.4896249408465866,1,0
+9,188,103,7,0.4649808838118618,-0.198806681337086,1,0
+9,189,123,7,-0.7073291167356331,-3.043576509789646,1,0
+9,190,111,5,-0.3740773149679098,0.8608798093746408,1,0
+9,191,123,7,0.4989177413451416,1.6835410358613347,1,0
+9,192,107,5,-0.6634732155934616,-0.3061356639836578,3,0
+9,193,123,3,-0.7887391302965693,-1.343886436889559,3,0
+9,194,103,9,-0.0473382835677561,3.656344892439264,1,0
+9,195,129,9,-0.5213560739887506,-2.4135537875363897,1,0
+9,196,107,7,-0.4313729642029095,-2.2356462270857462,1,0
+9,197,109,3,0.0497654862338274,1.2900131060668418,1,0
+9,198,119,5,-0.285600783170167,5.764139626487539,1,0
+9,199,129,5,-0.5218301693993399,-1.1335640211899565,1,0
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index f046e9e5..dd1a23c4 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -33,7 +33,7 @@ def test_construct_index_maps_string():
 
 
 def test_construct_index_maps_integer():
-    """Tests reversing an int-to-int map, ensuring 0 maps to min_id - 1."""
+    """Tests reversing an int-to-int map with special IDs as sentinels."""
     id_maps: dict[str, dict[str | int, int]] = {"storeId": {100: 3, 101: 4}}
     target_cols = ["storeId"]
 
@@ -42,11 +42,9 @@ def test_construct_index_maps_integer():
     assert result["storeId"][3] == 100
     assert result["storeId"][4] == 101
 
-    # Check the special 0 index for integers (min value - 2)
-    # Min value is 100, so 2 -> 98, 1 -> 99
-    assert result["storeId"][0] == 97
-    assert result["storeId"][1] == 98
-    assert result["storeId"][2] == 99
+    assert result["storeId"][0] == "[unknown]"
+    assert result["storeId"][1] == "[other]"
+    assert result["storeId"][2] == "[mask]"
 
 
 def test_construct_index_maps_flag_false():
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index a592e758..416d13d9 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -250,6 +250,11 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
     assert RESERVED_MASK_COLUMN not in metadata["column_types"]
     assert RESERVED_MASK_COLUMN not in metadata["id_maps"]
     assert RESERVED_MASK_COLUMN not in metadata["selected_columns_statistics"]
+    assert metadata["special_token_ids"] == {
+        "[unknown]": 0,
+        "[other]": 1,
+        "[mask]": 2,
+    }
 
 
 def test_load_and_preprocess_data_requests_mask_column_for_csv_projection(tmp_path):
@@ -293,6 +298,36 @@ def projected_read_data(path, read_format, columns=None):
     ]
 
 
+def test_load_and_preprocess_data_fails_when_mask_column_is_missing(tmp_path):
+    csv_path = tmp_path / "input.csv"
+    pl.DataFrame({"sequenceId": [0], "itemPosition": [0], "itemId": ["a"]}).write_csv(
+        csv_path
+    )
+
+    with pytest.raises(ValueError, match="mask_column '.+' not found"):
+        _load_and_preprocess_data(
+            str(csv_path),
+            "csv",
+            None,
+            None,
+            RESERVED_MASK_COLUMN,
+        )
+
+    parquet_path = tmp_path / "input.parquet"
+    pl.DataFrame(
+        {"sequenceId": [0], "itemPosition": [0], "itemId": ["a"]}
+    ).write_parquet(parquet_path)
+
+    with pytest.raises(ValueError, match="mask_column '.+' not found"):
+        _load_and_preprocess_data(
+            str(parquet_path),
+            "parquet",
+            ["itemId"],
+            None,
+            RESERVED_MASK_COLUMN,
+        )
+
+
 def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
     project_root = tmp_path / "project"
     project_root.mkdir()

From f739c0aa9b178da0c09f431a1caf2334dfe145f0 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 18:19:33 +0200
Subject: [PATCH 31/81] Remove backward compatibility for missing mask tensors

---
 src/sequifier/infer.py   | 72 ++--------------------------------------
 tests/unit/test_infer.py | 10 +++---
 2 files changed, 7 insertions(+), 75 deletions(-)

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 6cd28bc3..41b491cb 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -323,33 +323,6 @@ def _flatten_bert_target_valid_mask(
     return valid_mask.reshape(-1)
 
 
-@beartype
-def _is_categorical_column(config: InfererModel, column_name: str) -> bool:
-    column_type = str(config.column_types.get(column_name, "")).lower()
-    return (
-        column_name in config.categorical_columns
-        or config.target_column_types.get(column_name) == "categorical"
-        or column_type.startswith("int")
-    )
-
-
-@beartype
-def _infer_bert_valid_mask_from_values(
-    config: InfererModel,
-    column_name: str,
-    values: np.ndarray,
-    prediction_length: int,
-) -> Optional[np.ndarray]:
-    if _is_categorical_column(config, column_name):
-        valid_mask = values != 0
-    else:
-        valid_mask = np.cumsum(values != 0.0, axis=1) > 0
-
-    return _flatten_bert_target_valid_mask(
-        config, {"target_valid_mask": valid_mask}, prediction_length
-    )
-
-
 @beartype
 def _bert_reference_column(config: InfererModel, data_columns: set[str]) -> str:
     preferred_columns = (
@@ -362,29 +335,7 @@ def _bert_reference_column(config: InfererModel, data_columns: set[str]) -> str:
         if column_name in data_columns:
             return column_name
 
-    raise ValueError("Could not find a reference column for BERT valid-mask inference.")
-
-
-@beartype
-def _flatten_legacy_bert_target_valid_mask_from_sequences(
-    config: InfererModel,
-    sequences: dict[str, Union[torch.Tensor, np.ndarray]],
-    prediction_length: int,
-) -> Optional[np.ndarray]:
-    if config.training_objective != "bert":
-        return None
-
-    column_name = _bert_reference_column(config, set(sequences.keys()))
-    values = sequences[column_name]
-    if isinstance(values, torch.Tensor):
-        values = values.detach().cpu().numpy()  # type: ignore
-    values = np.asarray(values)
-    if values.ndim == 2 and values.shape[1] > prediction_length:
-        values = values[:, :prediction_length]
-
-    return _infer_bert_valid_mask_from_values(
-        config, column_name, values, prediction_length
-    )
+    raise ValueError("Could not find a reference column for BERT padding metadata.")
 
 
 @beartype
@@ -412,14 +363,7 @@ def _bert_target_valid_mask_from_preprocessed_data(
         )
         return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
 
-    target_seq_cols = [
-        str(c)
-        for c in range(config.seq_length, config.seq_length - prediction_length, -1)
-    ]
-    values = reference_rows.select(target_seq_cols).to_numpy()
-    return _infer_bert_valid_mask_from_values(
-        config, column_name, values, prediction_length
-    )
+    return None
 
 
 @beartype
@@ -558,12 +502,6 @@ def infer_embedding(
             valid_prediction_mask = _flatten_bert_target_valid_mask(
                 config, metadata, prediction_length
             )
-            if valid_prediction_mask is None:
-                valid_prediction_mask = (
-                    _flatten_legacy_bert_target_valid_mask_from_sequences(
-                        config, sequences_dict, prediction_length
-                    )
-                )
 
             sequence_ids_for_preds = sequence_ids_tensor.numpy()
             subsequence_ids_for_preds = subsequence_ids_tensor.numpy()
@@ -826,12 +764,6 @@ def infer_generative(
             valid_prediction_mask = _flatten_bert_target_valid_mask(
                 config, metadata, prediction_length
             )
-            if valid_prediction_mask is None:
-                valid_prediction_mask = (
-                    _flatten_legacy_bert_target_valid_mask_from_sequences(
-                        config, sequences_dict, prediction_length
-                    )
-                )
 
             if total_steps == 1:
                 # Non-autoregressive path: Apply prediction_length logic
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index d858fbe5..5416dc41 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -460,7 +460,7 @@ def capture_write(dataframe, path, write_format):
     assert predictions.get_column("target_col").to_list() == ["B", "A", "B"]
 
 
-def test_bert_generative_inference_filters_legacy_data_without_left_pad(tmp_path):
+def test_bert_generative_inference_keeps_data_without_left_pad(tmp_path):
     config = _bert_inference_config(tmp_path)
     data = _bert_preprocessed_frame().drop("leftPadLength")
     written = []
@@ -483,10 +483,10 @@ def capture_write(dataframe, path, write_format):
 
     predictions = next(frame for path, frame, _ in written if "predictions" in path)
 
-    assert predictions.height == 1
-    assert predictions.get_column("sequenceId").to_list() == [7]
-    assert predictions.get_column("itemPosition").to_list() == [102]
-    assert predictions.get_column("target_col").to_list() == ["A"]
+    assert predictions.height == 3
+    assert predictions.get_column("sequenceId").to_list() == [7, 7, 7]
+    assert predictions.get_column("itemPosition").to_list() == [100, 101, 102]
+    assert predictions.get_column("target_col").to_list() == ["A", "B", "A"]
 
 
 def test_bert_embedding_inference_filters_left_padded_positions(tmp_path):

From 9ad5a018c5e8b57a11ce392e43cd9a6c98136cbe Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 18:56:05 +0200
Subject: [PATCH 32/81] remove default value for training_objective

---
 src/sequifier/config/infer_config.py                           | 2 +-
 src/sequifier/config/train_config.py                           | 2 +-
 tests/configs/hyperparameter-search-bayesian.yaml              | 1 +
 tests/configs/hyperparameter-search-custom-eval-inference.yaml | 1 +
 tests/configs/hyperparameter-search-custom-eval.yaml           | 1 +
 tests/configs/hyperparameter-search-grid.yaml                  | 1 +
 tests/configs/hyperparameter-search-sample.yaml                | 1 +
 tests/configs/infer-test-categorical-autoregression.yaml       | 1 +
 tests/configs/infer-test-categorical-embedding.yaml            | 1 +
 tests/configs/infer-test-categorical-inf-size-1.yaml           | 1 +
 tests/configs/infer-test-categorical-inf-size-3-embedding.yaml | 1 +
 tests/configs/infer-test-categorical-inf-size-3.yaml           | 1 +
 tests/configs/infer-test-categorical-multitarget.yaml          | 1 +
 tests/configs/infer-test-categorical.yaml                      | 1 +
 tests/configs/infer-test-distributed-parquet.yaml              | 1 +
 tests/configs/infer-test-distributed.yaml                      | 1 +
 tests/configs/infer-test-lazy.yaml                             | 1 +
 tests/configs/infer-test-real-autoregression.yaml              | 1 +
 tests/configs/infer-test-real.yaml                             | 1 +
 tests/configs/train-test-categorical-inf-size-1.yaml           | 1 +
 tests/configs/train-test-categorical-inf-size-3.yaml           | 1 +
 tests/configs/train-test-categorical-multitarget-eager.yaml    | 1 +
 tests/configs/train-test-categorical-multitarget.yaml          | 1 +
 tests/configs/train-test-categorical.yaml                      | 1 +
 tests/configs/train-test-distributed-lazy-parquet.yaml         | 1 +
 tests/configs/train-test-distributed.yaml                      | 1 +
 tests/configs/train-test-lazy.yaml                             | 1 +
 tests/configs/train-test-real.yaml                             | 1 +
 tests/configs/train-test-resume-epoch.yaml                     | 1 +
 tests/configs/train-test-resume-mid-epoch.yaml                 | 1 +
 30 files changed, 30 insertions(+), 2 deletions(-)

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index b89b54fb..62277c76 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -118,7 +118,7 @@ class InfererModel(BaseModel):
     metadata_config_path: str
     model_path: Union[str, list[str]]
     model_type: str
-    training_objective: str = "causal"
+    training_objective: str
     data_path: str
     training_config_path: str = Field(default="configs/train.yaml")
     read_format: str = Field(default="parquet")
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 1a7f8078..d387e7cd 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -183,7 +183,7 @@ class TrainingSpecModel(BaseModel):
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
-    training_objective: str = "causal"
+    training_objective: str
     device: str
     device_max_concat_length: int = 12
     epochs: int
diff --git a/tests/configs/hyperparameter-search-bayesian.yaml b/tests/configs/hyperparameter-search-bayesian.yaml
index 990869c7..e15a2b84 100644
--- a/tests/configs/hyperparameter-search-bayesian.yaml
+++ b/tests/configs/hyperparameter-search-bayesian.yaml
@@ -43,6 +43,7 @@ model_hyperparameter_sampling:
 
 # Training Hyperparameter Search Space
 training_hyperparameter_sampling:
+  training_objective: ["causal"]
   device: cpu
   epochs: [1, 1]
   save_interval_epochs: 10
diff --git a/tests/configs/hyperparameter-search-custom-eval-inference.yaml b/tests/configs/hyperparameter-search-custom-eval-inference.yaml
index 7c11e091..ef23b3a0 100644
--- a/tests/configs/hyperparameter-search-custom-eval-inference.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval-inference.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 metadata_config_path: configs/metadata_configs/test-data-categorical-5.json
 
 model_type: generative
+training_objective: causal
 model_path: models/sequifier-test-hp-search-custom-eval-run-0-best-1.pt
 
 data_path: data/test-data-categorical-5-split2
diff --git a/tests/configs/hyperparameter-search-custom-eval.yaml b/tests/configs/hyperparameter-search-custom-eval.yaml
index be3403d4..6d870395 100644
--- a/tests/configs/hyperparameter-search-custom-eval.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval.yaml
@@ -52,6 +52,7 @@ model_hyperparameter_sampling:
 
 # Training Hyperparameter Search Space
 training_hyperparameter_sampling:
+  training_objective: ["causal"]
   device: cpu
   epochs: [1, 1]
   save_interval_epochs: 10
diff --git a/tests/configs/hyperparameter-search-grid.yaml b/tests/configs/hyperparameter-search-grid.yaml
index 625d2538..df79f038 100644
--- a/tests/configs/hyperparameter-search-grid.yaml
+++ b/tests/configs/hyperparameter-search-grid.yaml
@@ -43,6 +43,7 @@ model_hyperparameter_sampling:
 
 # Training Hyperparameter Search Space
 training_hyperparameter_sampling:
+  training_objective: ["causal"]
   device: cpu
   epochs: [2]
   save_interval_epochs: 10
diff --git a/tests/configs/hyperparameter-search-sample.yaml b/tests/configs/hyperparameter-search-sample.yaml
index 87bf2db6..5a0085ba 100644
--- a/tests/configs/hyperparameter-search-sample.yaml
+++ b/tests/configs/hyperparameter-search-sample.yaml
@@ -53,6 +53,7 @@ model_hyperparameter_sampling:
 
 # Training Hyperparameter Search Space using Distributions
 training_hyperparameter_sampling:
+  training_objective: ["causal"]
   device: cpu
   epochs: [1]
   save_interval_epochs: 10
diff --git a/tests/configs/infer-test-categorical-autoregression.yaml b/tests/configs/infer-test-categorical-autoregression.yaml
index 59db37ad..76d9d06b 100644
--- a/tests/configs/infer-test-categorical-autoregression.yaml
+++ b/tests/configs/infer-test-categorical-autoregression.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-1-best-3-autoregression.onnx
 data_path: tests/project_folder/data/test-data-categorical-1-split2
 read_format: pt
diff --git a/tests/configs/infer-test-categorical-embedding.yaml b/tests/configs/infer-test-categorical-embedding.yaml
index a302b6e3..9ea82e5a 100644
--- a/tests/configs/infer-test-categorical-embedding.yaml
+++ b/tests/configs/infer-test-categorical-embedding.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
 model_type: embedding
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-1-best-embedding-3.pt
 data_path: tests/project_folder/data/test-data-categorical-1-split2
 read_format: pt
diff --git a/tests/configs/infer-test-categorical-inf-size-1.yaml b/tests/configs/infer-test-categorical-inf-size-1.yaml
index 724226ea..37280c9c 100644
--- a/tests/configs/infer-test-categorical-inf-size-1.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-1.yaml
@@ -1,6 +1,7 @@
 project_root: tests/project_folder
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-1-inf-size-best-3.onnx
 data_path: tests/project_folder/data/test-data-categorical-1-split2
 read_format: pt
diff --git a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
index d26322e3..6f0e3085 100644
--- a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical-inf-size-3.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-3.json
 model_type: embedding
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-3-inf-size-best-embedding-3.pt
 data_path: tests/project_folder/data/test-data-categorical-3-split2
 read_format: pt
diff --git a/tests/configs/infer-test-categorical-inf-size-3.yaml b/tests/configs/infer-test-categorical-inf-size-3.yaml
index 8cf3c83d..2a99c2be 100644
--- a/tests/configs/infer-test-categorical-inf-size-3.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical-inf-size-3.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-3.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-3-inf-size-best-3.pt
 data_path: tests/project_folder/data/test-data-categorical-3-split2
 read_format: pt
diff --git a/tests/configs/infer-test-categorical-multitarget.yaml b/tests/configs/infer-test-categorical-multitarget.yaml
index d65fd107..29c4d8d8 100644
--- a/tests/configs/infer-test-categorical-multitarget.yaml
+++ b/tests/configs/infer-test-categorical-multitarget.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical-multitarget.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-multitarget-5.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-multitarget-5-best-3.onnx
 data_path: tests/project_folder/data/test-data-categorical-multitarget-5-split2
 read_format: parquet
diff --git a/tests/configs/infer-test-categorical.yaml b/tests/configs/infer-test-categorical.yaml
index 7f89e1e4..eeca3e19 100644
--- a/tests/configs/infer-test-categorical.yaml
+++ b/tests/configs/infer-test-categorical.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-5-best-3.pt
 data_path: tests/project_folder/data/test-data-categorical5-split2
 read_format: pt
diff --git a/tests/configs/infer-test-distributed-parquet.yaml b/tests/configs/infer-test-distributed-parquet.yaml
index 67935b4a..6e01ac6a 100644
--- a/tests/configs/infer-test-distributed-parquet.yaml
+++ b/tests/configs/infer-test-distributed-parquet.yaml
@@ -1,6 +1,7 @@
 project_root: tests/project_folder
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-multitarget-5.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-categorical-multitarget-5-last-3.onnx
 data_path: tests/project_folder/data/test-data-categorical-multitarget-5-split2
 read_format: parquet
diff --git a/tests/configs/infer-test-distributed.yaml b/tests/configs/infer-test-distributed.yaml
index 3a7f7faa..2e468055 100644
--- a/tests/configs/infer-test-distributed.yaml
+++ b/tests/configs/infer-test-distributed.yaml
@@ -4,6 +4,7 @@ metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
 # Point to the model generated by the distributed training test
 model_path: models/sequifier-model-categorical-distributed-best-3.onnx
 model_type: generative
+training_objective: causal
 
 # Use the split2 data generated by preprocessing
 data_path: data/test-data-categorical-3-split2
diff --git a/tests/configs/infer-test-lazy.yaml b/tests/configs/infer-test-lazy.yaml
index 1c8f4137..472b1dfa 100644
--- a/tests/configs/infer-test-lazy.yaml
+++ b/tests/configs/infer-test-lazy.yaml
@@ -3,6 +3,7 @@ metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
 
 model_path: models/sequifier-model-categorical-lazy-best-3.onnx
 model_type: generative
+training_objective: causal
 
 data_path: data/test-data-categorical-3-split2
 read_format: pt
diff --git a/tests/configs/infer-test-real-autoregression.yaml b/tests/configs/infer-test-real-autoregression.yaml
index c965b6f5..72c4128e 100644
--- a/tests/configs/infer-test-real-autoregression.yaml
+++ b/tests/configs/infer-test-real-autoregression.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-real.yaml
 metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-real-1.json
 model_type: generative
+training_objective: causal
 model_path: tests/project_folder/models/sequifier-model-real-1-best-3-autoregression.pt
 data_path: tests/project_folder/data/test-data-real-1-split1-autoregression.csv
 read_format: csv
diff --git a/tests/configs/infer-test-real.yaml b/tests/configs/infer-test-real.yaml
index 0fac3579..009823e0 100644
--- a/tests/configs/infer-test-real.yaml
+++ b/tests/configs/infer-test-real.yaml
@@ -2,6 +2,7 @@ project_root: tests/project_folder
 training_config_path: tests/configs/train-test-real.yaml
 metadata_config_path: configs/metadata_configs/test-data-real-5.json
 model_type: generative
+training_objective: causal
 model_path: models/sequifier-model-real-5-best-3.pt
 data_path: data/test-data-real5-split2.csv
 
diff --git a/tests/configs/train-test-categorical-inf-size-1.yaml b/tests/configs/train-test-categorical-inf-size-1.yaml
index 540dc963..92cae64c 100644
--- a/tests/configs/train-test-categorical-inf-size-1.yaml
+++ b/tests/configs/train-test-categorical-inf-size-1.yaml
@@ -31,6 +31,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   save_interval_epochs: 1
diff --git a/tests/configs/train-test-categorical-inf-size-3.yaml b/tests/configs/train-test-categorical-inf-size-3.yaml
index 8a8575a0..c20d9ac0 100644
--- a/tests/configs/train-test-categorical-inf-size-3.yaml
+++ b/tests/configs/train-test-categorical-inf-size-3.yaml
@@ -38,6 +38,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 10000.0
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   save_interval_epochs: 1
diff --git a/tests/configs/train-test-categorical-multitarget-eager.yaml b/tests/configs/train-test-categorical-multitarget-eager.yaml
index 65e93add..55d55d83 100644
--- a/tests/configs/train-test-categorical-multitarget-eager.yaml
+++ b/tests/configs/train-test-categorical-multitarget-eager.yaml
@@ -38,6 +38,7 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   save_interval_epochs: 1
diff --git a/tests/configs/train-test-categorical-multitarget.yaml b/tests/configs/train-test-categorical-multitarget.yaml
index 395fb71b..6066730a 100644
--- a/tests/configs/train-test-categorical-multitarget.yaml
+++ b/tests/configs/train-test-categorical-multitarget.yaml
@@ -39,6 +39,7 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   save_interval_epochs: 1
diff --git a/tests/configs/train-test-categorical.yaml b/tests/configs/train-test-categorical.yaml
index 468aab70..eafdbd8d 100644
--- a/tests/configs/train-test-categorical.yaml
+++ b/tests/configs/train-test-categorical.yaml
@@ -32,6 +32,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
+  training_objective: causal
   device: cpu
   torch_compile: "outer"
   epochs: 3
diff --git a/tests/configs/train-test-distributed-lazy-parquet.yaml b/tests/configs/train-test-distributed-lazy-parquet.yaml
index 3ac5fa17..871fa0c1 100644
--- a/tests/configs/train-test-distributed-lazy-parquet.yaml
+++ b/tests/configs/train-test-distributed-lazy-parquet.yaml
@@ -25,6 +25,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   log_interval: 1
diff --git a/tests/configs/train-test-distributed.yaml b/tests/configs/train-test-distributed.yaml
index 1cd9872e..23e92988 100644
--- a/tests/configs/train-test-distributed.yaml
+++ b/tests/configs/train-test-distributed.yaml
@@ -27,6 +27,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   log_interval: 1
diff --git a/tests/configs/train-test-lazy.yaml b/tests/configs/train-test-lazy.yaml
index c771a614..e53b9e2f 100644
--- a/tests/configs/train-test-lazy.yaml
+++ b/tests/configs/train-test-lazy.yaml
@@ -27,6 +27,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   log_interval: 1
diff --git a/tests/configs/train-test-real.yaml b/tests/configs/train-test-real.yaml
index fd1a4e00..0b02444a 100644
--- a/tests/configs/train-test-real.yaml
+++ b/tests/configs/train-test-real.yaml
@@ -32,6 +32,7 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
+  training_objective: causal
   device: cpu
   torch_compile: "inner"
   epochs: 3
diff --git a/tests/configs/train-test-resume-epoch.yaml b/tests/configs/train-test-resume-epoch.yaml
index dbaef582..ce93f148 100644
--- a/tests/configs/train-test-resume-epoch.yaml
+++ b/tests/configs/train-test-resume-epoch.yaml
@@ -32,6 +32,7 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
+  training_objective: causal
   device: cpu
   epochs: 3
   save_interval_epochs: 1
diff --git a/tests/configs/train-test-resume-mid-epoch.yaml b/tests/configs/train-test-resume-mid-epoch.yaml
index b0449d91..6398cb81 100644
--- a/tests/configs/train-test-resume-mid-epoch.yaml
+++ b/tests/configs/train-test-resume-mid-epoch.yaml
@@ -32,6 +32,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
+  training_objective: causal
   device: cpu
   torch_compile: "outer"
   epochs: 3

From c0cd86eb9b5e517d14e4baef1aad3e3e93aa5bf9 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 18:57:02 +0200
Subject: [PATCH 33/81] remove negative int test

---
 ...-test-categorical-bert-autoregression.yaml | 26 -------------------
 tests/integration/conftest.py                 |  9 -------
 tests/integration/test_inference.py           | 22 ----------------
 3 files changed, 57 deletions(-)
 delete mode 100644 tests/configs/infer-test-categorical-bert-autoregression.yaml

diff --git a/tests/configs/infer-test-categorical-bert-autoregression.yaml b/tests/configs/infer-test-categorical-bert-autoregression.yaml
deleted file mode 100644
index 552bb21f..00000000
--- a/tests/configs/infer-test-categorical-bert-autoregression.yaml
+++ /dev/null
@@ -1,26 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical-bert.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
-model_type: generative
-training_objective: bert
-model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-1.pt
-data_path: tests/project_folder/data/test-data-categorical-1-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-seq_length: 8
-inference_batch_size: 10
-prediction_length: null
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: true
-autoregression_total_steps: 1
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index c634ee90..83208921 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -205,13 +205,6 @@ def inference_config_path_cat_bert_embedding():
     )
 
 
-@pytest.fixture(scope="session")
-def inference_config_path_cat_bert_autoregression():
-    return os.path.join(
-        "tests", "configs", "infer-test-categorical-bert-autoregression.yaml"
-    )
-
-
 @pytest.fixture(scope="session")
 def inference_config_path_real_autoregression():
     return os.path.join("tests", "configs", "infer-test-real-autoregression.yaml")
@@ -318,7 +311,6 @@ def format_configs_locally(
     inference_config_path_cat_inf_size_3,
     inference_config_path_cat_bert,
     inference_config_path_cat_bert_embedding,
-    inference_config_path_cat_bert_autoregression,
     inference_config_path_distributed,
     inference_config_path_distributed_parquet,
     inference_config_path_lazy,
@@ -355,7 +347,6 @@ def format_configs_locally(
             inference_config_path_cat_inf_size_3,
             inference_config_path_cat_bert,
             inference_config_path_cat_bert_embedding,
-            inference_config_path_cat_bert_autoregression,
             inference_config_path_distributed,
             inference_config_path_distributed_parquet,
             inference_config_path_lazy,
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index e668b1b5..36d5bbdd 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -1,6 +1,5 @@
 import json
 import os
-import subprocess
 
 import numpy as np
 import polars as pl
@@ -260,27 +259,6 @@ def test_bert_embeddings(bert_predictions, bert_embeddings, project_root):
     )
 
 
-def test_bert_rejects_autoregression_end_to_end(
-    run_training, inference_config_path_cat_bert_autoregression
-):
-    result = subprocess.run(
-        [
-            "sequifier",
-            "infer",
-            "--config-path",
-            inference_config_path_cat_bert_autoregression,
-        ],
-        capture_output=True,
-        text=True,
-    )
-
-    assert result.returncode != 0
-    assert (
-        "Autoregressive inference is not possible with BERT-style models."
-        in result.stdout + result.stderr
-    )
-
-
 def test_predictions_item_position(predictions):
     """
     Checks if itemPosition increments correctly within each sequenceId.

From bd077080f157f708bd12661a3742067f2eb9c50c Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 19:07:31 +0200
Subject: [PATCH 34/81] remove padding inference again

---
 src/sequifier/helpers.py | 27 +-------------
 src/sequifier/train.py   | 26 ++++----------
 tests/unit/test_train.py | 76 +++++++++++++++++++++++++---------------
 3 files changed, 55 insertions(+), 74 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 3f463757..4641a18c 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -3,7 +3,6 @@
 import random
 import re
 import sys
-import warnings
 from datetime import datetime
 from typing import Any, Dict, Optional, Union
 
@@ -607,24 +606,6 @@ def get_last_training_batch_timedelta(
     return (t2 - t1).total_seconds()
 
 
-def infer_valid_mask_from_data(
-    data_batch: Dict[str, torch.Tensor],
-    categorical_columns: list[str],
-    default_key: str,
-    metadata: Optional[dict[str, Tensor]] = None,
-):
-    if metadata is not None and default_key in metadata:
-        return metadata[default_key].bool()
-
-    if len(categorical_columns):
-        ref_col = categorical_columns[0]
-        return data_batch[ref_col] != 0
-    else:
-        warnings.warn(EXPLICIT_PADDING_MASK_FALLBACK_WARNING, stacklevel=2)
-        ref_col = list(data_batch.keys())[0]
-        return (data_batch[ref_col] != 0.0).long().cumsum(dim=1) > 0
-
-
 def apply_bert_masking(
     data_batch: Dict[str, torch.Tensor],
     targets_batch: Dict[str, torch.Tensor],
@@ -643,13 +624,7 @@ def apply_bert_masking(
         else {}
     )
 
-    # 1. Identify valid tokens (Renamed from padding_mask to valid_mask for clarity)
-    if "attention_valid_mask" in metadata_batch:
-        valid_mask = metadata_batch["attention_valid_mask"].bool()
-    else:
-        valid_mask = infer_valid_mask_from_data(
-            data_batch, config.categorical_columns, "attention_valid_mask"
-        )
+    valid_mask = metadata_batch["attention_valid_mask"].bool()
 
     batch_size, seq_len = valid_mask.shape
     device = valid_mask.device
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index b3f0b284..ea39e6e5 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -61,7 +61,6 @@
     configure_logger,
     construct_index_maps,
     get_torch_dtype,
-    infer_valid_mask_from_data,
     normalize_path,
 )
 from sequifier.io.batch import SequifierBatch  # noqa: E402
@@ -534,9 +533,7 @@ def _copy_model(self):
         return model_copy
 
     @conditional_beartype
-    def forward(
-        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
-    ):
+    def forward(self, src: dict[str, Tensor], metadata: dict[str, Tensor]):
         """Forward pass for the embedding model.
 
         Args:
@@ -1052,7 +1049,7 @@ def _zero_padding_positions(self, x: Tensor, valid_mask: Tensor) -> Tensor:
 
     @conditional_beartype
     def forward_inner(
-        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+        self, src: dict[str, Tensor], metadata: dict[str, Tensor]
     ) -> Tensor:
         """The inner forward pass of the model.
 
@@ -1122,9 +1119,7 @@ def forward_inner(
         if self.joint_embedding_layer is not None:
             src2 = self.joint_embedding_layer(src2)
 
-        valid_mask = infer_valid_mask_from_data(
-            src, self.hparams.categorical_columns, "attention_valid_mask", metadata
-        ).to(device=src2.device)
+        valid_mask = metadata["attention_valid_mask"].bool()  # type: ignore
         if valid_mask.shape != src2.shape[:2]:
             raise ValueError(
                 f"Invalid attention mask shape: got {tuple(valid_mask.shape)}, "
@@ -1146,7 +1141,7 @@ def forward_inner(
 
     @conditional_beartype
     def forward_embed(
-        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+        self, src: dict[str, Tensor], metadata: dict[str, Tensor]
     ) -> Tensor:
         """Forward pass for the embedding model.
 
@@ -1164,7 +1159,7 @@ def forward_embed(
 
     @conditional_beartype
     def forward_train(
-        self, src: dict[str, Tensor], metadata: Optional[dict[str, Tensor]] = None
+        self, src: dict[str, Tensor], metadata: dict[str, Tensor]
     ) -> dict[str, Tensor]:
         """Forward pass for training.
 
@@ -1231,7 +1226,7 @@ def apply_softmax(self, target_column: str, output: Tensor) -> Tensor:
     def forward(
         self,
         src: dict[str, Tensor],
-        metadata: Optional[dict[str, Tensor]] = None,
+        metadata: dict[str, Tensor],
         return_logits: Union[bool, Tensor] = False,
     ) -> dict[str, Tensor]:
         """The main forward pass of the model.
@@ -1729,14 +1724,7 @@ def _calculate_loss(
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
-        categorical_target_columns = [
-            target_name
-            for target_name in target_names
-            if self.target_column_types[target_name] == "categorical"
-        ]
-        valid_mask = infer_valid_mask_from_data(
-            targets, categorical_target_columns, "target_valid_mask", metadata
-        )
+        valid_mask = metadata["attention_valid_mask"].bool()  # type: ignore
 
         if metadata is not None and "bert_mask" in metadata:
             valid_mask = valid_mask & metadata["bert_mask"].bool()
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 67d642f9..fb451533 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -12,7 +12,6 @@
     TrainingSpecModel,
     TrainModel,
 )
-from sequifier.helpers import infer_valid_mask_from_data
 from sequifier.train import TransformerModel
 
 
@@ -167,6 +166,19 @@ def bert_model(model_config):
     return TransformerModel(config)
 
 
+def _all_valid_metadata(batch_size, seq_len, device=None):
+    valid_mask = torch.ones(
+        batch_size,
+        seq_len,
+        dtype=torch.bool,
+        device=device,
+    )
+    return {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+
 def test_transformer_model_initialization(model, model_config):
     """Tests that the model initializes with the correct layers."""
     # Check if encoder dicts were created
@@ -218,9 +230,10 @@ def test_forward_train_shapes(model, model_config):
     x_real = torch.randn(batch_size, seq_len)
 
     src = {"cat_col": x_cat, "real_col": x_real}
+    metadata = _all_valid_metadata(batch_size, seq_len, device=x_cat.device)
 
     # forward_train returns a dict of tensors
-    outputs = model.forward_train(src)
+    outputs = model.forward_train(src, metadata)
 
     assert "cat_col" in outputs
     assert "real_col" in outputs
@@ -245,10 +258,11 @@ def test_forward_inference_shapes(model, model_config):
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
     x_real = torch.randn(batch_size, seq_len)
     src = {"cat_col": x_cat, "real_col": x_real}
+    metadata = _all_valid_metadata(batch_size, seq_len, device=x_cat.device)
 
     # forward returns predictions for the *last* prediction_length tokens
     # And applies softmax to categorical outputs
-    outputs = model.forward(src)
+    outputs = model.forward(src, metadata)
 
     # Expected shape: (prediction_length, batch_size, n_classes_or_1)
     # If prediction_length is 1, dim 0 is size 1.
@@ -282,12 +296,13 @@ def test_calculate_loss(model, model_config):
     y_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
     y_real = torch.randn(batch_size, seq_len)
     targets = {"cat_col": y_cat, "real_col": y_real}
+    metadata = _all_valid_metadata(batch_size, seq_len, device=x_cat.device)
 
     # Run forward pass
-    outputs = model.forward_train(src)
+    outputs = model.forward_train(src, metadata)
 
     # Calculate loss
-    total_loss, component_losses = model._calculate_loss(outputs, targets)
+    total_loss, component_losses = model._calculate_loss(outputs, targets, metadata)
 
     # Assertions
     assert total_loss.dim() == 0  # Scalar
@@ -315,6 +330,7 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
         "real_col": torch.tensor([[0.0, 0.0, 2.0]]),
     }
     metadata = {
+        "attention_valid_mask": torch.tensor([[True, True, True]]),
         "target_valid_mask": torch.tensor([[True, True, True]]),
     }
 
@@ -344,33 +360,17 @@ def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
     targets = {
         "real_target": torch.tensor([[0.0, 0.0, 2.0]]),
     }
-
-    with pytest.warns(UserWarning):
-        total_loss, component_losses = TransformerModel._calculate_loss(
-            model, outputs, targets
-        )
-
-    assert torch.isclose(total_loss, torch.tensor(1.0))
-    assert torch.isclose(component_losses["real_target"], torch.tensor(1.0))
-
-
-def test_infer_valid_mask_from_data():
-    model = TransformerModel.__new__(TransformerModel)
-    model.categorical_columns = []
-    model.input_columns = ["real_col"]
-
-    src = {
-        "real_col": torch.tensor([[0.0, 0.0, 1.0]]),
-    }
     metadata = {
-        "attention_valid_mask": torch.tensor([[True, True, True]]),
+        "attention_valid_mask": torch.tensor([[False, False, True]]),
+        "target_valid_mask": torch.tensor([[False, False, True]]),
     }
 
-    mask = infer_valid_mask_from_data(
-        src, model.categorical_columns, "attention_valid_mask", metadata
+    total_loss, component_losses = TransformerModel._calculate_loss(
+        model, outputs, targets, metadata
     )
 
-    assert torch.equal(mask, torch.tensor([[True, True, True]]))
+    assert torch.isclose(total_loss, torch.tensor(1.0))
+    assert torch.isclose(component_losses["real_target"], torch.tensor(1.0))
 
 
 def test_padding_keys_are_masked(bert_model):
@@ -458,8 +458,26 @@ def batch(model):
     }
 
 
-def test_forward_no_nan_with_padding(model, batch):
-    out = model.forward_train(batch)
+@pytest.fixture
+def batch_metadata(model):
+    seq_len = model.seq_length
+    valid_mask = torch.ones(
+        2,
+        seq_len,
+        dtype=torch.bool,
+        device=model.src_mask.device,
+    )
+    valid_mask[0, :2] = False
+    valid_mask[1, :1] = False
+
+    return {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+
+def test_forward_no_nan_with_padding(model, batch, batch_metadata):
+    out = model.forward_train(batch, batch_metadata)
 
     for tensor in out.values():
         assert torch.isfinite(tensor).all()

From 6113f548ecfc4cd2ecc69051f12f123da83bd349 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 19:57:21 +0200
Subject: [PATCH 35/81] Make metadata mandatory

---
 src/sequifier/helpers.py                      | 11 +++++
 src/sequifier/io/batch.py                     |  1 -
 src/sequifier/train.py                        | 41 +++++++++++++++----
 ...test-data-real-1-split1-autoregression.csv | 40 +++++++++---------
 4 files changed, 65 insertions(+), 28 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 4641a18c..fd3fdb90 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -611,6 +611,7 @@ def apply_bert_masking(
     targets_batch: Dict[str, torch.Tensor],
     metadata_batch: Optional[Dict[str, torch.Tensor]],
     config: Any,  # TrainConfig
+    eval_seed: Optional[int] = None,
 ) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
     """
     Applies BERT-style span corruption to the input data using custom distributions.
@@ -629,6 +630,12 @@ def apply_bert_masking(
     batch_size, seq_len = valid_mask.shape
     device = valid_mask.device
 
+    if eval_seed is not None:
+        cpu_rng_state = torch.get_rng_state()
+        if device.type == "cuda":
+            gpu_rng_state = torch.cuda.get_rng_state(device)
+        torch.manual_seed(eval_seed)
+
     # Calculate exact number of tokens to mask per sequence based on valid length
     masking_prob = config.training_spec.bert_spec.masking_probability
     budgets = (valid_mask.sum(dim=1) * masking_prob).long()
@@ -728,4 +735,8 @@ def apply_bert_masking(
     metadata_batch["bert_mask"] = bert_mask
     metadata_batch["attention_valid_mask"] = valid_mask.detach()
 
+    if eval_seed is not None:
+        torch.set_rng_state(cpu_rng_state)  # type: ignore
+        if device.type == "cuda":
+            torch.cuda.set_rng_state(gpu_rng_state, device)  # type: ignore
     return data_batch, targets_batch, metadata_batch
diff --git a/src/sequifier/io/batch.py b/src/sequifier/io/batch.py
index ba2d7330..36906103 100644
--- a/src/sequifier/io/batch.py
+++ b/src/sequifier/io/batch.py
@@ -11,7 +11,6 @@ class SequifierBatch:
     metadata: dict[str, torch.Tensor]
     sequence_ids: Optional[torch.Tensor] = None
     subsequence_ids: Optional[torch.Tensor] = None
-    start_positions: Optional[torch.Tensor] = None
 
     def __iter__(self) -> Iterator[object]:
         yield self.inputs
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index ea39e6e5..c05be463 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -81,6 +81,7 @@
 )
 from sequifier.model.layers import RMSNorm, SequifierEncoderLayer  # noqa: E402
 from sequifier.optimizers.optimizers import get_optimizer_class  # noqa: E402
+from sequifier.special_tokens import SPECIAL_TOKEN_IDS  # noqa: E402
 
 
 def _as_sequifier_batch(batch: Any) -> SequifierBatch:
@@ -1858,7 +1859,7 @@ def _evaluate(
         model_to_call.eval()
 
         with torch.no_grad():
-            for raw_batch in valid_loader:
+            for batch_idx, raw_batch in enumerate(valid_loader):
                 batch = _as_sequifier_batch(raw_batch)
                 data = batch.inputs
                 targets = batch.targets
@@ -1880,7 +1881,11 @@ def _evaluate(
                 }
                 if self.hparams.training_spec.training_objective == "bert":
                     data, targets, metadata = apply_bert_masking(
-                        data, targets, metadata, self.hparams
+                        data,
+                        targets,
+                        metadata,
+                        self.hparams,
+                        eval_seed=self.hparams.seed + batch_idx,
                     )
 
                 if (
@@ -1962,7 +1967,7 @@ def _evaluate(
             baseline_losses_local_collect = {col: [] for col in self.target_columns}
 
             # Iterate over the sharded validation loader
-            for raw_batch in valid_loader:
+            for batch_idx, raw_batch in enumerate(valid_loader):
                 batch = _as_sequifier_batch(raw_batch)
                 data = batch.inputs
                 targets = batch.targets
@@ -1982,13 +1987,35 @@ def _evaluate(
                     for k, v in (metadata or {}).items()
                 }
 
+                if self.hparams.training_spec.training_objective == "bert":
+                    _, _, metadata = apply_bert_masking(
+                        data,
+                        targets,
+                        metadata,
+                        self.hparams,
+                        eval_seed=self.hparams.seed + batch_idx,
+                    )
+
                 pseudo_output = {}
                 targets_for_baseline = {}
                 for col in self.target_columns:
-                    if col in data:
-                        pseudo_output[col] = self._transform_val(
-                            col, data[col].transpose(0, 1)
-                        )
+                    if col in targets:
+                        if self.hparams.training_spec.training_objective == "causal":
+                            pseudo_output[col] = self._transform_val(
+                                col, data[col].transpose(0, 1)
+                            )
+                        elif self.hparams.training_spec.training_objective == "bert":
+                            shifted_targets = torch.roll(targets[col], shifts=1, dims=1)
+
+                            if self.target_column_types[col] == "categorical":
+                                shifted_targets[:, 0] = SPECIAL_TOKEN_IDS.unknown
+                            else:
+                                shifted_targets[:, 0] = 0.0
+                            pseudo_output[col] = self._transform_val(
+                                col, shifted_targets.transpose(0, 1)
+                            )
+                        else:
+                            raise ValueError("Impossible")
                         targets_for_baseline[col] = targets[col]
 
                 if len(pseudo_output) > 0:
diff --git a/tests/resources/source_data/test-data-real-1-split1-autoregression.csv b/tests/resources/source_data/test-data-real-1-split1-autoregression.csv
index 841267f4..5174b6ba 100644
--- a/tests/resources/source_data/test-data-real-1-split1-autoregression.csv
+++ b/tests/resources/source_data/test-data-real-1-split1-autoregression.csv
@@ -1,20 +1,20 @@
-sequenceId,subsequenceId,startItemPosition,inputCol,8,7,6,5,4,3,2,1,0
-0,13,13,itemValue,-1.2113837237467866,1.1957454087349089,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-0.5880064200454953
-0,14,14,itemValue,1.1957454087349089,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-inf,-1.6146336831548016
-0,15,15,itemValue,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-inf,-inf,-0.5557687212758077
-1,11,11,itemValue,-0.3979099218270587,0.1472006972746456,1.0744329431771849,2.098645946368935,-1.3233038228636609,0.3701880711691477,1.6509216329279448,0.695362468181269,-0.3227447762317759
-1,12,12,itemValue,0.1472006972746456,1.0744329431771849,2.098645946368935,-1.3233038228636609,0.3701880711691477,1.6509216329279448,0.695362468181269,-inf,-2.737064338735829
-2,9,9,itemValue,-1.9364972938429672,0.381774018107753,0.7895053840838986,0.071840823669383,-0.3274088457533277,0.8707377962958142,1.4490373682705944,1.0572550888530108,0.0765690379670138
-2,10,10,itemValue,0.381774018107753,0.7895053840838986,0.071840823669383,-0.3274088457533277,0.8707377962958142,1.4490373682705944,1.0572550888530108,-inf,0.476280644045447
-3,9,9,itemValue,-0.1291893885530777,-0.8243404591242441,-1.6194781231152326,0.1530304968195626,-0.4118502341991996,-0.7016026835022849,0.2909377683444284,1.9914944653082405,0.2070996307773291
-3,10,10,itemValue,-0.8243404591242441,-1.6194781231152326,0.1530304968195626,-0.4118502341991996,-0.7016026835022849,0.2909377683444284,1.9914944653082405,-inf,-0.0923048918846115
-4,6,6,itemValue,-0.3473441905826143,-0.9526372634835768,0.3152110308127366,1.0910573998928226,0.4164594405061328,0.142507586519504,1.0387354362777137,0.4461925335963277,1.6935115804711565
-5,6,6,itemValue,0.607763764789741,-2.458568025269152,-0.8214502614252478,-1.5396443182660882,-0.3113729142601391,-1.7463565527326972,0.441061415952748,-0.7445626882829559,-0.4361008211094133
-6,10,10,itemValue,1.5641135703291782,0.8372376866758972,-0.927489932456818,1.1789898362189837,-1.7300169078644538,-0.0513293403405074,0.013139305117334,0.6322625662411738,-1.0028799371032169
-6,11,11,itemValue,0.8372376866758972,-0.927489932456818,1.1789898362189837,-1.7300169078644538,-0.0513293403405074,0.013139305117334,0.6322625662411738,-inf,0.1685537328210896
-7,7,7,itemValue,-1.7680402671702622,-1.1479572969796248,0.366441270630024,0.9519780800395158,0.3286902962731986,1.3789080143759465,0.3820587792655491,1.5380644602548283,0.4910444675915603
-8,12,12,itemValue,-0.7095652049712206,0.116749400524584,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,1.2543512347210088
-8,13,13,itemValue,0.116749400524584,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,-inf,-0.6162672551654905
-8,14,14,itemValue,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,-inf,-inf,1.0886264665030447
-9,8,8,itemValue,0.2102757881460243,-0.756823800221343,1.0186924371356785,0.618472230767961,0.2340549739792952,-0.2583133906283459,-1.781087365783252,0.5179010995396939,-0.2841126785745299
-9,9,9,itemValue,-0.756823800221343,1.0186924371356785,0.618472230767961,0.2340549739792952,-0.2583133906283459,-1.781087365783252,0.5179010995396939,-inf,1.2530292363758264
+sequenceId,subsequenceId,startItemPosition,leftPadLength,inputCol,8,7,6,5,4,3,2,1,0
+0,13,13,0,itemValue,-1.2113837237467866,1.1957454087349089,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-0.5880064200454953
+0,14,14,0,itemValue,1.1957454087349089,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-inf,-1.6146336831548016
+0,15,15,0,itemValue,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-inf,-inf,-0.5557687212758077
+1,11,11,0,itemValue,-0.3979099218270587,0.1472006972746456,1.0744329431771849,2.098645946368935,-1.3233038228636609,0.3701880711691477,1.6509216329279448,0.695362468181269,-0.3227447762317759
+1,12,12,0,itemValue,0.1472006972746456,1.0744329431771849,2.098645946368935,-1.3233038228636609,0.3701880711691477,1.6509216329279448,0.695362468181269,-inf,-2.737064338735829
+2,9,9,0,itemValue,-1.9364972938429672,0.381774018107753,0.7895053840838986,0.071840823669383,-0.3274088457533277,0.8707377962958142,1.4490373682705944,1.0572550888530108,0.0765690379670138
+2,10,10,0,itemValue,0.381774018107753,0.7895053840838986,0.071840823669383,-0.3274088457533277,0.8707377962958142,1.4490373682705944,1.0572550888530108,-inf,0.476280644045447
+3,9,9,0,itemValue,-0.1291893885530777,-0.8243404591242441,-1.6194781231152326,0.1530304968195626,-0.4118502341991996,-0.7016026835022849,0.2909377683444284,1.9914944653082405,0.2070996307773291
+3,10,10,0,itemValue,-0.8243404591242441,-1.6194781231152326,0.1530304968195626,-0.4118502341991996,-0.7016026835022849,0.2909377683444284,1.9914944653082405,-inf,-0.0923048918846115
+4,6,6,0,itemValue,-0.3473441905826143,-0.9526372634835768,0.3152110308127366,1.0910573998928226,0.4164594405061328,0.142507586519504,1.0387354362777137,0.4461925335963277,1.6935115804711565
+5,6,6,0,itemValue,0.607763764789741,-2.458568025269152,-0.8214502614252478,-1.5396443182660882,-0.3113729142601391,-1.7463565527326972,0.441061415952748,-0.7445626882829559,-0.4361008211094133
+6,10,10,0,itemValue,1.5641135703291782,0.8372376866758972,-0.927489932456818,1.1789898362189837,-1.7300169078644538,-0.0513293403405074,0.013139305117334,0.6322625662411738,-1.0028799371032169
+6,11,11,0,itemValue,0.8372376866758972,-0.927489932456818,1.1789898362189837,-1.7300169078644538,-0.0513293403405074,0.013139305117334,0.6322625662411738,-inf,0.1685537328210896
+7,7,7,0,itemValue,-1.7680402671702622,-1.1479572969796248,0.366441270630024,0.9519780800395158,0.3286902962731986,1.3789080143759465,0.3820587792655491,1.5380644602548283,0.4910444675915603
+8,12,12,0,itemValue,-0.7095652049712206,0.116749400524584,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,1.2543512347210088
+8,13,13,0,itemValue,0.116749400524584,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,-inf,-0.6162672551654905
+8,14,14,0,itemValue,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,-inf,-inf,1.0886264665030447
+9,8,8,0,itemValue,0.2102757881460243,-0.756823800221343,1.0186924371356785,0.618472230767961,0.2340549739792952,-0.2583133906283459,-1.781087365783252,0.5179010995396939,-0.2841126785745299
+9,9,9,0,itemValue,-0.756823800221343,1.0186924371356785,0.618472230767961,0.2340549739792952,-0.2583133906283459,-1.781087365783252,0.5179010995396939,-inf,1.2530292363758264

From d15cb58752b95ba1c0bc2a6b881191d1d67cf697 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Thu, 11 Jun 2026 19:58:26 +0200
Subject: [PATCH 36/81] Update outputs

---
 ...cal-bert-best-embedding-1-0-embeddings.csv |   5 +
 ...cal-bert-best-embedding-1-1-embeddings.csv |   5 +
 ...cal-bert-best-embedding-1-2-embeddings.csv |   4 +
 ...cal-bert-best-embedding-1-3-embeddings.csv |   7 +
 ...cal-bert-best-embedding-1-4-embeddings.csv |   4 +
 ...cal-bert-best-embedding-1-5-embeddings.csv |   8 +
 ...cal-bert-best-embedding-1-6-embeddings.csv |   5 +
 ...cal-bert-best-embedding-1-7-embeddings.csv |   4 +
 ...-categorical-bert-best-1-0-predictions.csv |   5 +
 ...-categorical-bert-best-1-1-predictions.csv |   5 +
 ...-categorical-bert-best-1-2-predictions.csv |   4 +
 ...-categorical-bert-best-1-3-predictions.csv |   7 +
 ...-categorical-bert-best-1-4-predictions.csv |   4 +
 ...-categorical-bert-best-1-5-predictions.csv |   8 +
 ...-categorical-bert-best-1-6-predictions.csv |   5 +
 ...-categorical-bert-best-1-7-predictions.csv |   4 +
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...uifier-model-real-1-best-3-predictions.csv |  48 +--
 ...uifier-model-real-3-best-3-predictions.csv |  48 +--
 ...uifier-model-real-5-best-3-predictions.csv |  48 +--
 ...ifier-model-real-50-best-3-predictions.csv |  48 +--
 ...custom-eval-run-0-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-0-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-0-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-2-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-7-predictions.csv |  60 +--
 ...ategorical-bert-best-1-0-probabilities.csv |   5 +
 ...ategorical-bert-best-1-1-probabilities.csv |   5 +
 ...ategorical-bert-best-1-2-probabilities.csv |   4 +
 ...ategorical-bert-best-1-3-probabilities.csv |   7 +
 ...ategorical-bert-best-1-4-probabilities.csv |   4 +
 ...ategorical-bert-best-1-5-probabilities.csv |   8 +
 ...ategorical-bert-best-1-6-probabilities.csv |   5 +
 ...ategorical-bert-best-1-7-probabilities.csv |   4 +
 53 files changed, 1322 insertions(+), 1196 deletions(-)
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
 create mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
 create mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv

diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
new file mode 100644
index 00000000..79067fd4
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
@@ -0,0 +1,5 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+0,0,0,-0.46009058,0.45777348,-0.22698353,-0.25634044,0.05071858,-0.7203443,-0.5872563,-0.7930882,1.2241583,-0.81149936,-0.7057305,0.83234805,1.888106,-0.4073704,2.095673,-1.5786989
+0,0,1,1.2350427,-0.07666658,1.6687232,-0.10760293,-0.34233996,-0.6405651,-0.34899667,1.3310788,-1.7401463,0.48453188,-0.39308575,-0.38906077,0.09585311,0.9867051,0.3474074,-2.1432836
+0,0,2,-0.04062133,-1.1088247,-1.103962,-0.7512283,-0.3894962,0.40089244,1.3842655,-0.45180556,1.6632154,0.7094805,-1.2707394,1.2500856,-1.5106493,1.293199,0.25951874,-0.33003083
+0,0,3,0.95941216,0.026796307,0.18043526,-0.27425033,-0.8116876,0.59967464,0.15953179,0.6978008,0.6560124,-2.234662,-0.5438349,-1.4812955,-0.07771553,2.3249497,0.22156315,-0.4244198
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
new file mode 100644
index 00000000..f1ad3c83
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
@@ -0,0 +1,5 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+1,0,0,0.5496252,-0.29538283,1.4190812,1.0952207,-1.0380151,-0.95464176,0.25532782,2.3793674,0.15770668,-0.4943242,-0.23793282,-0.93078727,-0.4377435,0.48551127,-1.7345339,-0.24836543
+1,0,1,1.1703843,-1.293456,-0.009221493,1.1512711,-1.2613713,-0.14320165,0.21250723,1.3096029,-0.19192691,-1.8777579,1.3773452,0.84090453,0.026053404,0.5442735,-1.1615742,-0.7001723
+1,0,2,1.3515635,0.47997367,-0.15122601,0.20858693,-0.96445537,0.32341665,-0.17822078,0.5217958,0.63256186,-1.831718,-0.63131493,-1.3727977,0.07998669,2.4652927,-0.28591567,-0.66949725
+1,0,3,1.2581677,-1.3701472,1.279228,0.34158385,-0.1695098,-0.12549305,-0.9352961,-0.7049133,0.7994457,-0.7667792,0.9320275,1.0522128,-1.8637831,-0.15688421,-0.91187423,1.3544698
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
new file mode 100644
index 00000000..d9c4ea92
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
@@ -0,0 +1,4 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+2,0,0,-0.7215402,-0.9107281,-1.3260528,-1.056419,0.45294917,-1.4000003,1.2015996,0.9110033,0.082641706,1.9949383,0.7476521,-0.21124426,1.3691758,-0.049398102,-0.21682854,-0.8714193
+2,0,1,1.7914528,0.28032902,1.0305467,0.54616165,-1.1185598,-0.72510433,-0.10695522,-0.7386186,-0.97949564,1.8118858,-0.005990093,-0.68710846,-0.26387492,0.34162205,-1.9024037,0.6979735
+2,0,2,-0.30499086,-0.8624588,-0.21858704,0.46042088,1.684188,-0.8045112,0.051364895,0.14763248,-0.15275371,-0.62137896,0.52078897,-2.5685115,1.6713971,0.9936811,0.5037575,-0.5022072
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
new file mode 100644
index 00000000..41d3d3c6
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
@@ -0,0 +1,7 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+3,0,0,-0.7998329,-1.009516,-1.2457482,-0.76004726,0.5837586,-1.667505,0.9730014,0.9463384,0.092550665,2.0262837,0.91990626,-0.16828287,1.2223057,-0.15285197,-0.15041813,-0.8101929
+3,0,1,-0.44488773,-0.768482,-0.96725094,-1.0061802,-1.1290205,0.5468874,1.4315809,-0.18544558,0.2409695,-0.26420698,0.2942724,2.9089172,0.037383337,0.20568404,0.09013668,-0.9878235
+3,0,2,0.9197474,-0.54972196,0.19757852,-1.6962574,-1.6389741,0.1349557,1.9117059,-0.8651552,0.5093372,-1.4415438,0.5049666,0.5232043,0.6729788,0.5089508,0.9317495,-0.64518726
+4,0,0,1.0214102,-0.8599426,1.1313342,-0.9132186,-1.1094761,0.36796033,-0.3840757,0.69376415,0.1755591,-0.18347333,-2.1679997,-0.50493,-0.43210468,2.0920951,0.5312533,0.5269235
+4,0,1,1.975615,0.48146883,1.191331,0.069339134,-0.15332681,-0.8564395,0.082964964,0.59752154,-1.4309397,0.55613095,-0.887867,-0.40480894,0.031874895,1.017457,-0.07405214,-2.231385
+4,0,2,0.24375618,-1.9693801,-0.9987763,-1.4291574,0.28681424,0.22939153,0.74847555,1.6184663,0.56282,0.12442069,0.036241677,-1.4214271,-0.19361049,0.4269336,-0.029789643,1.7619773
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
new file mode 100644
index 00000000..bc0d81ae
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
@@ -0,0 +1,4 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+5,0,0,0.050885476,-0.68763536,-0.32008207,1.0336747,1.0007741,-0.9633068,-0.68598473,0.54518497,-0.34585357,-0.47822043,0.52911806,-2.0803304,1.9946527,1.3150282,0.1371507,-1.0467436
+5,0,1,1.5032598,0.98465127,-0.9370364,0.1338618,-0.023070391,-0.055669162,-0.92458737,0.0777592,-0.114300214,0.7075389,0.89388293,-1.1466067,0.17746806,1.8015935,-1.9533355,-1.140061
+5,0,2,0.92082304,-0.24809152,-0.097492896,-0.5940675,-1.1684945,0.82807684,-0.26342613,0.24813169,-0.61301243,-0.017792376,1.3554065,1.2064362,0.016166179,-0.95746326,1.6580712,-2.2975802
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
new file mode 100644
index 00000000..32b2488a
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
@@ -0,0 +1,8 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+6,0,0,1.4568412,-0.14638022,1.5296284,-0.23393637,0.47851744,-0.61866605,-0.3450205,0.9253853,-1.4923376,1.1160238,-0.5999185,-0.7552432,-0.13930646,0.5775359,0.43673426,-2.2207265
+6,0,1,1.5268029,-0.8329019,1.6105913,0.68216175,-1.6239593,-0.7086337,-1.3626277,-0.081963986,-1.1570084,1.6663798,0.41809285,0.0707222,-0.10685151,0.3952702,-0.6428383,0.13019788
+6,0,2,1.0811658,0.20269495,-0.8124527,-0.00006336486,-1.4045546,0.38837618,-0.56309855,0.12770331,1.2022637,0.8725209,0.38372105,-2.5436192,0.303661,-0.14655858,1.5541319,-0.6847054
+6,0,3,0.03786793,-1.276104,1.9573607,0.4641471,-0.90920424,-0.64407074,0.8226515,1.1558217,-0.020683609,-0.8913047,-0.7226361,-0.8886756,-0.007999741,-0.83854043,-0.30502415,2.0437944
+7,0,0,-0.4584146,-1.3955376,-0.52467126,-0.618622,-1.2942654,0.7003769,0.48657182,0.09509239,-0.07182534,-0.2584199,0.48787862,2.825618,0.3362462,0.20020695,0.8225655,-1.3267877
+7,0,1,1.4503253,-0.6273112,-0.99216473,1.4233189,1.0060959,0.67193794,-1.6410915,-0.6428859,-0.934798,1.4429418,0.1590132,-1.4877084,0.1463788,0.6280047,-0.5242633,-0.079930626
+7,0,2,1.1836636,-0.09071134,-0.14702518,1.2998279,-1.8555058,0.77598166,-0.72920847,0.43935195,0.4997686,1.223445,0.761577,-2.2151723,-0.3662264,0.030556075,0.093911,-0.94706964
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
new file mode 100644
index 00000000..c949fdbf
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
@@ -0,0 +1,5 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+8,0,0,-0.15959328,-0.6185797,-0.4608494,0.8537297,1.7962078,-1.1127523,-0.532915,0.41857973,-0.025692305,-0.11058396,0.43107894,-2.3876948,1.7336528,0.75202924,0.051930904,-0.62754744
+8,0,1,-0.0847378,-0.80747044,0.54135734,0.21987116,-0.27892733,-0.6947288,-1.0170838,-0.3269096,1.2616061,1.8441427,2.0002346,0.7380944,-0.06132242,-0.8849537,-1.2508907,-1.1942315
+8,0,2,-0.060736626,-0.2758699,-1.7476931,-0.90112156,0.66965175,-1.3700024,1.6178493,0.32909736,0.27642158,1.9732589,0.37533468,-0.48581138,1.1548799,0.011515356,-0.70278364,-0.87338156
+8,0,3,0.7022285,-0.33732405,-0.057168916,-0.8344231,-0.868286,0.98070174,-0.23612636,0.26457375,-0.65804225,0.08511807,1.3956434,1.1746727,-0.08018996,-1.348229,1.8052337,-2.011606
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
new file mode 100644
index 00000000..13a6a356
--- /dev/null
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
@@ -0,0 +1,4 @@
+sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
+9,0,0,-0.5791086,0.37848184,-0.3573193,-0.2355006,-0.14353162,-0.3610847,-0.8467316,-0.7029914,0.79546255,-0.9406641,-0.41330278,1.1095508,1.7990614,-0.31147677,2.3270056,-1.514815
+9,0,1,0.80497956,0.5582172,-0.92235243,0.12620161,-1.2371104,0.36691037,-0.58676964,0.394082,1.040764,0.61476994,0.54693735,-2.5416214,-0.021673568,-0.45668596,1.8399619,-0.5678765
+9,0,2,-0.72852725,-1.68056,-0.4684366,-1.2692064,0.31585234,0.8645183,1.0582336,-0.34532753,1.2095294,0.25657162,-0.7247522,1.3270385,-1.5651994,0.36106008,1.513459,-0.11602255
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
new file mode 100644
index 00000000..a6555801
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
@@ -0,0 +1,5 @@
+sequenceId,itemPosition,itemId
+0,0,116
+0,1,125
+0,2,111
+0,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
new file mode 100644
index 00000000..63cca8be
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
@@ -0,0 +1,5 @@
+sequenceId,itemPosition,itemId
+1,0,100
+1,1,120
+1,2,102
+1,3,104
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
new file mode 100644
index 00000000..78e69770
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
@@ -0,0 +1,4 @@
+sequenceId,itemPosition,itemId
+2,0,103
+2,1,115
+2,2,125
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
new file mode 100644
index 00000000..aa1d014b
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
@@ -0,0 +1,7 @@
+sequenceId,itemPosition,itemId
+3,0,103
+3,1,119
+3,2,119
+4,0,122
+4,1,125
+4,2,118
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
new file mode 100644
index 00000000..d0661d18
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
@@ -0,0 +1,4 @@
+sequenceId,itemPosition,itemId
+5,0,125
+5,1,123
+5,2,117
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
new file mode 100644
index 00000000..9b747406
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
@@ -0,0 +1,8 @@
+sequenceId,itemPosition,itemId
+6,0,125
+6,1,115
+6,2,109
+6,3,106
+7,0,117
+7,1,125
+7,2,109
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
new file mode 100644
index 00000000..69194925
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
@@ -0,0 +1,5 @@
+sequenceId,itemPosition,itemId
+8,0,125
+8,1,126
+8,2,123
+8,3,117
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
new file mode 100644
index 00000000..5be21105
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
@@ -0,0 +1,4 @@
+sequenceId,itemPosition,itemId
+9,0,116
+9,1,128
+9,2,114
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index d1af2939..05a3e9a6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.25203884
-0,22,0.07592129
-0,23,0.29406223
-0,24,0.2411493
-0,25,0.21718875
-0,26,0.29605892
-0,27,0.13432515
-0,28,0.20121504
-0,29,-0.016863476
-0,30,-0.21210459
-0,31,-0.022416836
-0,32,0.042226754
-0,33,0.08740239
-0,34,0.2900688
-0,35,-0.06784207
-0,36,-0.09479769
-0,37,0.01564551
-0,38,-0.12325086
-0,39,-0.25403556
-0,40,-0.032525197
-1,19,0.25113288
-1,20,0.16227913
-1,21,0.07142868
-1,22,0.27209836
-1,23,0.27209836
-1,24,-0.11975661
-1,25,0.1148572
-1,26,-0.029405331
-1,27,-0.008003062
-1,28,-0.02528711
-1,29,-0.089306735
-1,30,0.21020025
-1,31,0.07192786
-1,32,0.07642046
-1,33,0.11385884
-1,34,0.062443476
-1,35,0.20021668
-1,36,0.08340896
-1,37,0.18224627
-1,38,-0.05960562
-2,17,0.054456625
-2,18,0.001980504
-2,19,-0.08032152
-2,20,-0.015865121
-2,21,-0.05710973
-2,22,-0.11027222
-2,23,0.11186212
-2,24,-0.09030509
-2,25,0.029123325
-2,26,0.034239903
-2,27,0.035987027
-2,28,0.0001475836
-2,29,-0.021792863
-2,30,0.07642046
-2,31,-0.17217033
-2,32,0.16227913
-2,33,-0.14421634
-2,34,-0.102784544
-2,35,-0.09180263
-2,36,-0.22308652
-3,17,0.23116574
-3,18,0.20421012
-3,19,0.42384857
-3,20,0.16727091
-3,21,0.24713944
-3,22,0.11136295
-3,23,0.008781808
-3,24,0.088899925
-3,25,-0.07982235
-3,26,0.11136295
-3,27,-0.1522032
-3,28,-0.085313305
-3,29,0.029248118
-3,30,0.035987027
-3,31,0.012463248
-3,32,-0.101287015
-3,33,-0.101287015
-3,34,0.10886706
-3,35,-0.0046804063
-3,36,-0.043382324
-4,14,0.2900688
-4,15,0.17425941
-4,16,0.007783452
-4,17,-0.08181906
-4,18,0.016643867
-4,19,-0.04487986
-4,20,0.2780885
-4,21,-0.00597515
-4,22,-0.015677929
-4,23,-0.11726072
-4,24,-0.021044096
-4,25,0.054955803
-4,26,-0.14122127
-4,27,-0.11726072
-4,28,-0.10428208
-4,29,-0.10178619
-4,30,-0.14122127
-4,31,-0.028781358
-4,32,-0.21709637
-4,33,-0.15719499
-5,14,0.05520539
-5,15,0.33199978
-5,16,0.27209836
-5,17,0.2231789
-5,18,0.11735309
-5,19,0.11335966
-5,20,0.051461555
-5,21,0.2391526
-5,22,-0.03364835
-5,23,-0.20411775
-5,24,0.16427584
-5,25,-0.19213746
-5,26,0.037234973
-5,27,-0.09579605
-5,28,0.056702927
-5,29,0.045471415
-5,30,0.20421012
-5,31,-0.006349534
-5,32,0.100880206
-5,33,0.030246476
-6,18,0.29605892
-6,19,0.29605892
-6,20,0.1792512
-6,21,0.17425941
-6,22,0.19821997
-6,23,0.0434747
-6,24,0.0016997161
-6,25,-0.07483056
-6,26,-0.065845355
-6,27,0.015146332
-6,28,-0.026160672
-6,29,0.021136472
-6,30,-0.09629523
-6,31,-0.16318512
-6,32,-0.22907665
-6,33,0.090896636
-6,34,-0.06509659
-6,35,-0.07982235
-6,36,0.015271126
-6,37,-0.08481413
-7,15,0.072926216
-7,16,0.014896743
-7,17,-0.029654922
-7,18,-0.043631915
-7,19,-0.05261712
-7,20,0.2740951
-7,21,-0.11626236
-7,22,0.023881953
-7,23,-0.0000942059
-7,24,-0.064597405
-7,25,0.2780885
-7,26,0.0055371495
-7,27,0.054706212
-7,28,0.032991957
-7,29,0.078916356
-7,30,0.24514273
-7,31,0.090397455
-7,32,0.041977167
-7,33,-0.047874928
-7,34,-0.05910644
-8,20,-0.08281741
-8,21,0.21918546
-8,22,0.14131364
-8,23,0.04996402
-8,24,0.21119861
-8,25,0.07242704
-8,26,0.0065355063
-8,27,-0.018922588
-8,28,-0.051618766
-8,29,-0.18315226
-8,30,-0.22408487
-8,31,0.055704568
-8,32,-0.10178619
-8,33,-0.14022292
-8,34,0.13732022
-8,35,-0.048873287
-8,36,0.16327749
-8,37,0.004351601
-8,38,0.08690321
-8,39,-0.024538344
-9,16,0.3958946
-9,17,0.37193403
-9,18,0.082909785
-9,19,0.018765375
-9,20,0.34398004
-9,21,0.17425941
-9,22,0.084407315
-9,23,0.09938267
-9,24,0.017143045
-9,25,-0.1751654
-9,26,-0.107277155
-9,27,0.10886706
-9,28,0.26211482
-9,29,-0.026784645
-9,30,-0.05261712
-9,31,-0.14721142
-9,32,-0.11376647
-9,33,0.03149442
-9,34,0.019264553
-9,35,0.04671936
+0,21,-0.24294445
+0,22,0.2182963
+0,23,0.24425358
+0,24,0.276201
+0,25,0.0940009
+0,26,0.51380986
+0,27,0.13742942
+0,28,0.19333738
+0,29,0.14042449
+0,30,-0.110163026
+0,31,-0.12064577
+0,32,-0.17106278
+0,33,-0.007956264
+0,34,0.05157075
+0,35,-0.14410715
+0,36,-0.038530935
+0,37,0.07652966
+0,38,-0.1730595
+0,39,-0.068232045
+0,40,-0.04052765
+1,19,0.2532388
+1,20,0.2412585
+1,21,0.1653834
+1,22,0.23926179
+1,23,0.33210897
+1,24,0.11496639
+1,25,-0.025427505
+1,26,-0.054255053
+1,27,-0.20999868
+1,28,-0.045269843
+1,29,-0.11365727
+1,30,-0.116153166
+1,31,0.18535054
+1,32,0.15639819
+1,33,0.025363889
+1,34,0.17836204
+1,35,-0.06773287
+1,36,-0.096186034
+1,37,-0.16507263
+1,38,0.009327785
+2,17,0.40199393
+2,18,-0.024678737
+2,19,0.033101153
+2,20,0.15040804
+2,21,0.23327164
+2,22,0.08850994
+2,23,0.16138998
+2,24,0.013258814
+2,25,0.0075182635
+2,26,-0.09369014
+2,27,0.04258554
+2,28,-0.0762189
+2,29,-0.10716796
+2,30,-0.14510551
+2,31,0.12744585
+2,32,-0.072724655
+2,33,0.016253883
+2,34,-0.086202465
+2,35,0.018500187
+2,36,0.13643105
+3,17,0.0030256584
+3,18,0.49384275
+3,19,0.3081484
+3,20,0.16238832
+3,21,0.117462285
+3,22,0.11546557
+3,23,0.045830198
+3,24,0.01562991
+3,25,0.09699597
+3,26,0.033849917
+3,27,-0.032291207
+3,28,0.12744585
+3,29,-0.122143306
+3,30,-0.0832074
+3,31,-0.11515481
+3,32,-0.021808462
+3,33,-0.12314166
+3,34,0.06355102
+3,35,-0.020810105
+3,36,-0.17605457
+4,14,0.0077678524
+4,15,0.050821982
+4,16,0.0075182635
+4,17,-0.116153166
+4,18,0.24325521
+4,19,0.13942613
+4,20,0.12594831
+4,21,0.0665461
+4,22,-0.04252436
+4,23,0.06754445
+4,24,-0.021059694
+4,25,0.13243763
+4,26,0.07652966
+4,27,-0.24993296
+4,28,-0.1361203
+4,29,0.06030637
+4,30,0.070040345
+4,31,-0.002878685
+4,32,-0.06573616
+4,33,-0.16107921
+5,14,0.26821414
+5,15,0.25623387
+5,16,0.36405638
+5,17,0.22129136
+5,18,0.3061517
+5,19,0.077528015
+5,20,0.07503212
+5,21,0.09499926
+5,22,-0.08121069
+5,23,-0.18404141
+5,24,-0.08270822
+5,25,-0.10117782
+5,26,-0.10117782
+5,27,-0.018064626
+5,28,-0.1910299
+5,29,0.33210897
+5,30,-0.16207758
+5,31,0.025987862
+5,32,-0.035535865
+5,33,-0.049762446
+6,18,0.20631602
+6,19,0.27420428
+6,20,0.082020625
+6,21,0.28219113
+6,22,0.3940071
+6,23,0.16638176
+6,24,-0.08370657
+6,25,-0.12613674
+6,26,0.14641462
+6,27,-0.012511266
+6,28,-0.08520411
+6,29,-0.15708579
+6,30,-0.044521075
+6,31,0.069541164
+6,32,-0.0036664505
+6,33,-0.09269179
+6,34,-0.13711865
+6,35,-0.027549013
+6,36,-0.15309235
+6,37,0.07403377
+7,15,0.19832917
+7,16,0.28019443
+7,17,0.052569106
+7,18,-0.13412358
+7,19,-0.039279703
+7,20,0.06704527
+7,21,0.10049022
+7,22,0.055564176
+7,23,0.024365531
+7,24,0.018375391
+7,25,-0.11365727
+7,26,0.003056857
+7,27,0.027360601
+7,28,0.08401734
+7,29,0.07253624
+7,30,-0.12763427
+7,31,-0.1910299
+7,32,-0.012760855
+7,33,0.058559246
+7,34,0.026611833
+8,20,0.34808266
+8,21,-0.02392997
+8,22,-0.16806771
+8,23,0.09499926
+8,24,0.3361024
+8,25,0.017127445
+8,26,-0.15608743
+8,27,-0.10067864
+8,28,-0.12414002
+8,29,-0.013821609
+8,30,0.17137355
+8,31,-0.114156455
+8,32,0.03859211
+8,33,-0.009578593
+8,34,0.06454938
+8,35,-0.11116138
+8,36,0.05132116
+8,37,-0.12713508
+8,38,0.06305185
+8,39,-0.033539154
+9,16,0.19134067
+9,17,0.32811555
+9,18,0.09000748
+9,19,0.09000748
+9,20,-0.13212687
+9,21,0.055564176
+9,22,0.0665461
+9,23,-0.23395924
+9,24,0.15240477
+9,25,-0.021683667
+9,26,-0.090195894
+9,27,-0.06299067
+9,28,0.009639772
+9,29,-0.061992317
+9,30,-0.27688858
+9,31,0.118959814
+9,32,0.18934396
+9,33,-0.066734515
+9,34,0.06704527
+9,35,0.13742942
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
index 1cf52011..8cc53618 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,0.2900688
-0,9,-0.17616376
-0,10,0.3799209
-0,11,0.26411152
-0,12,-0.3009583
-1,8,0.28208193
-1,9,-0.0039628376
-1,10,0.33798993
-2,8,-0.14820977
-2,9,0.11735309
-3,8,0.36993733
-3,9,0.29805565
-4,7,-0.13523114
-5,7,0.14131364
-6,8,0.3359932
-6,9,0.26810494
-6,10,0.31402937
-7,8,-0.15419991
-8,8,0.16826928
-8,9,-0.25703064
-8,10,0.41586173
-8,11,0.40787488
-9,8,0.31402937
-9,9,0.25812137
+0,8,0.30016157
+0,9,-0.1820447
+0,10,0.39201036
+0,11,0.27420428
+0,12,-0.30883598
+1,8,0.290178
+1,9,-0.1890332
+1,10,0.3500794
+2,8,-0.15508908
+2,9,0.12445078
+3,8,0.3800301
+3,9,0.3061517
+4,7,-0.14011373
+5,7,0.14641462
+6,8,0.34808266
+6,9,-0.13911536
+6,10,0.3221254
+7,8,-0.16307592
+8,8,0.1743686
+8,9,-0.13711865
+8,10,0.43593806
+8,11,0.4259545
+9,8,0.33011225
+9,9,-0.12813345
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
index 68074514..405ee2f4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.19088916
-0,9,-0.078307025
-0,10,0.020881303
-0,11,0.13500652
-0,12,0.30980512
-1,8,0.0876035
-1,9,-0.026953768
-1,10,-0.2353295
-2,8,0.046619654
-2,9,-0.18891405
-3,8,0.08957863
-3,9,0.17450903
-4,7,0.37695938
-5,7,-0.18002598
-6,8,0.2999295
-6,9,0.33745688
-6,10,0.007867463
-7,8,0.66730285
-8,8,0.20117323
-8,9,0.19031003
-8,10,-0.03213847
-8,11,-0.012880999
-9,8,0.0023336397
-9,9,-0.2531056
+0,8,-0.19243994
+0,9,-0.010234638
+0,10,0.020318083
+0,11,0.18382144
+0,12,0.47416487
+1,8,0.09296566
+1,9,-0.059118994
+1,10,-0.2941589
+2,8,0.047784667
+2,9,-0.19342752
+3,8,0.09247188
+3,9,0.17888363
+4,7,0.37935886
+5,7,-0.14503694
+6,8,0.30035383
+6,9,0.33393097
+6,10,0.00545835
+7,8,0.6716774
+8,8,0.21542345
+8,9,0.19073437
+8,10,-0.041836645
+8,11,-0.02529497
+9,8,0.0025419537
+9,9,-0.2546564
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
index 7c11814b..5e697aea 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.07564883
-0,9,-0.34528795
-0,10,-0.06342901
-0,11,-0.22871205
-0,12,-0.23238373
-1,8,-0.2268762
-1,9,-0.10984136
-1,10,-0.016901925
-2,8,-0.3618105
-2,9,-0.29939193
-3,8,-0.1360021
-3,9,0.23850942
-4,7,0.104493044
-5,7,-0.3471238
-6,8,0.09990344
-6,9,-0.11305408
-6,10,0.15406075
-7,8,0.12652314
-8,8,-0.17822644
-8,9,-0.11075929
-8,10,-0.29755607
-8,11,-0.29204854
-9,8,-0.14472234
-9,9,-0.019196726
+0,8,-0.083931625
+0,9,-0.35219386
+0,10,-0.061499946
+0,11,-0.23102838
+0,12,-0.24387926
+1,8,-0.27600646
+1,9,-0.108486
+1,10,-0.01715292
+2,8,-0.36320892
+2,9,-0.30262616
+3,8,-0.13740048
+3,9,0.2316035
+4,7,0.106766336
+5,7,-0.35219386
+6,8,0.06591888
+6,9,-0.11628832
+6,10,0.15358028
+7,8,0.11869929
+8,8,-0.17595315
+8,9,-0.14336696
+8,10,-0.3081337
+8,11,-0.3007903
+9,8,-0.1419901
+9,9,-0.01715292
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
index 22ba483f..d6da3019 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.08269817
-0,9,-0.22038624
-0,10,-0.060151752
-0,11,-0.066634566
-0,12,-0.0973849
-1,8,-0.055131875
-1,9,-0.02945879
-1,10,-0.12446355
-2,8,-0.25710306
-2,9,-0.24058047
-3,8,0.01506035
-3,9,0.37075448
-4,7,0.3579036
-5,7,-0.12813523
-6,8,0.3395452
-6,9,0.2771266
-6,10,0.31567925
-7,8,0.35423192
-8,8,-0.039154325
-8,9,-0.015919466
-8,10,-0.053353403
-8,11,-0.12629938
-9,8,-0.0007737763
-9,9,0.19084209
+0,8,-0.08198105
+0,9,-0.215309
+0,10,-0.075096644
+0,11,-0.0692449
+0,12,-0.101486854
+1,8,-0.049674552
+1,9,-0.034134448
+1,10,-0.122140065
+2,8,-0.2749738
+2,9,-0.2575333
+3,8,-0.0012040511
+3,9,0.36940628
+4,7,0.35104787
+5,7,-0.06190154
+6,8,0.34003285
+6,9,0.2757784
+6,10,0.33085364
+7,8,0.35471958
+8,8,0.011417352
+8,9,0.100455634
+8,10,-0.057369307
+8,11,-0.12718862
+9,8,-0.0424531
+9,9,0.18031469
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
index 9a24e80c..4e65dc53 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,127,1,1,2,1
-0,5,127,1,8,8,1
-0,6,113,1,1,7,0
-0,7,105,8,9,7,1
-0,8,127,1,1,0,1
-0,9,113,1,7,7,1
-0,10,113,8,9,1,1
-0,11,113,1,1,7,1
-0,12,105,8,5,7,1
-0,13,113,1,5,8,1
-0,14,113,1,[other],8,1
-0,15,113,1,5,7,1
-0,16,105,1,5,7,1
-0,17,108,1,8,7,1
-0,18,108,1,5,7,1
-0,19,124,1,5,7,1
-0,20,124,1,5,7,1
-0,21,105,8,5,7,1
-0,22,105,1,5,7,1
-0,23,110,1,5,7,1
-0,24,110,1,5,8,1
-0,25,110,1,7,8,1
-0,26,110,[other],7,7,1
-0,27,110,5,7,7,1
-0,28,103,5,7,0,1
-0,29,102,1,5,0,1
-0,30,103,1,7,7,1
-0,31,102,5,5,0,1
-0,32,103,1,7,0,1
-0,33,102,[mask],7,0,1
+0,4,104,4,9,7,0
+0,5,112,4,9,7,0
+0,6,104,9,9,7,0
+0,7,112,4,9,5,0
+0,8,112,9,9,5,0
+0,9,112,9,9,5,0
+0,10,112,9,9,5,0
+0,11,112,9,9,5,0
+0,12,112,9,9,5,0
+0,13,112,9,9,5,0
+0,14,112,9,9,5,0
+0,15,112,9,9,2,0
+0,16,112,9,9,7,0
+0,17,112,9,9,7,0
+0,18,112,9,9,7,0
+0,19,112,9,9,7,0
+0,20,112,9,9,7,0
+0,21,112,9,9,7,0
+0,22,112,9,9,7,0
+0,23,112,9,9,7,0
+0,24,112,9,9,7,0
+0,25,112,9,9,7,0
+0,26,112,9,9,7,0
+0,27,112,9,9,7,0
+0,28,112,9,9,7,0
+0,29,112,9,9,7,0
+0,30,112,9,9,7,0
+0,31,112,9,9,7,0
+0,32,112,9,9,7,0
+0,33,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
index 12e2155a..bf364d0d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,121,1,0,7,0
-1,5,108,3,0,8,1
-1,6,103,1,0,7,0
-1,7,103,8,0,7,1
-1,8,108,1,0,2,1
-1,9,113,1,0,7,1
-1,10,108,8,7,7,1
-1,11,124,8,7,7,1
-1,12,105,8,7,7,1
-1,13,113,8,5,7,1
-1,14,113,1,[other],8,1
-1,15,113,1,[other],8,1
-1,16,113,1,[other],7,1
-1,17,108,1,5,7,1
-1,18,124,1,[other],7,1
-1,19,108,1,5,7,1
-1,20,124,1,5,7,1
-1,21,124,1,5,7,1
-1,22,105,8,5,7,1
-1,23,105,1,5,7,1
-1,24,124,8,5,7,1
-1,25,110,8,5,7,1
-1,26,110,1,5,8,1
-1,27,110,1,7,8,1
-1,28,110,6,7,7,1
-1,29,110,1,7,0,1
-1,30,113,1,7,0,1
-1,31,113,1,7,0,1
-1,32,113,1,7,0,1
-1,33,113,[mask],5,0,1
+1,4,111,4,9,7,1
+1,5,111,4,9,7,1
+1,6,111,4,9,7,1
+1,7,111,4,9,7,1
+1,8,111,4,9,7,1
+1,9,111,4,9,7,1
+1,10,111,4,9,7,1
+1,11,111,4,9,7,1
+1,12,102,4,9,7,1
+1,13,102,4,9,7,1
+1,14,102,4,9,7,1
+1,15,102,4,9,7,1
+1,16,122,4,9,5,1
+1,17,122,9,9,2,0
+1,18,121,9,9,5,0
+1,19,121,9,9,5,0
+1,20,112,9,9,2,0
+1,21,112,9,9,2,0
+1,22,112,9,9,2,0
+1,23,112,9,9,2,0
+1,24,112,9,9,7,0
+1,25,112,9,9,7,0
+1,26,112,9,9,7,0
+1,27,112,9,9,7,0
+1,28,112,9,9,7,0
+1,29,112,9,9,7,0
+1,30,112,9,9,7,0
+1,31,112,9,9,7,0
+1,32,112,9,9,7,0
+1,33,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
index 5cf7bea0..ab10690c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,128,1,8,6,0
-2,4,105,0,8,7,2
-2,5,105,1,8,2,1
-2,6,113,1,8,7,2
-2,7,105,1,9,7,1
-2,8,105,1,7,7,1
-2,9,113,1,7,7,1
-2,10,113,1,5,7,1
-2,11,113,1,[other],8,1
-2,12,113,1,[other],7,1
-2,13,113,8,5,7,1
-2,14,113,1,[other],8,1
-2,15,113,1,5,7,1
-2,16,113,1,5,7,1
-2,17,105,1,5,7,1
-2,18,108,1,5,7,1
-2,19,108,1,5,7,1
-2,20,108,1,5,7,1
-2,21,124,1,5,7,1
-2,22,105,8,5,7,1
-2,23,110,1,5,7,1
-2,24,110,8,5,8,1
-2,25,110,1,5,8,1
-2,26,110,[other],5,8,1
-2,27,127,1,7,8,1
-2,28,103,5,7,0,1
-2,29,103,1,5,7,1
-2,30,103,1,5,0,1
-2,31,103,[other],4,0,1
-2,32,127,1,4,7,1
+2,3,112,9,9,7,0
+2,4,112,9,9,2,0
+2,5,112,9,9,2,0
+2,6,112,9,9,2,0
+2,7,112,9,9,2,0
+2,8,112,9,9,2,0
+2,9,112,9,9,7,0
+2,10,112,9,9,7,0
+2,11,112,9,9,7,0
+2,12,112,9,9,7,0
+2,13,112,9,9,7,0
+2,14,112,9,9,7,0
+2,15,112,9,9,7,0
+2,16,112,9,9,7,0
+2,17,112,9,9,7,0
+2,18,112,9,9,7,0
+2,19,112,9,9,7,0
+2,20,112,9,9,7,0
+2,21,112,9,9,7,0
+2,22,112,9,9,7,0
+2,23,112,9,9,7,0
+2,24,112,9,9,7,0
+2,25,112,9,9,7,0
+2,26,112,9,9,7,0
+2,27,112,9,9,7,0
+2,28,112,9,9,7,0
+2,29,112,9,9,7,0
+2,30,112,9,9,7,0
+2,31,112,9,9,7,0
+2,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
index 4e1d63dd..c3e044e5 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,127,5,0,2,0
-3,4,127,4,0,8,0
-3,5,127,4,0,8,0
-3,6,127,4,0,8,0
-3,7,127,4,0,8,0
-3,8,127,4,0,8,1
-3,9,108,1,0,8,1
-3,10,103,1,7,7,1
-3,11,103,8,9,7,1
-3,12,103,1,7,7,1
-3,13,113,8,7,7,1
-3,14,113,8,5,7,1
-3,15,113,1,5,7,1
-3,16,113,1,5,7,1
-3,17,113,1,5,7,1
-3,18,113,1,[other],8,1
-3,19,113,1,[other],7,1
-3,20,108,8,5,7,1
-3,21,108,1,2,7,1
-3,22,108,1,5,7,1
-3,23,108,1,5,7,1
-3,24,124,1,5,7,1
-3,25,124,1,5,7,1
-3,26,124,8,5,7,1
-3,27,124,1,5,7,1
-3,28,124,8,5,7,1
-3,29,110,8,5,7,1
-3,30,110,1,5,8,1
-3,31,110,1,7,8,1
-3,32,110,[other],7,7,1
-4,3,127,0,9,[unknown],1
-4,4,127,0,1,8,1
-4,5,102,0,1,8,1
-4,6,127,1,4,8,0
-4,7,127,0,1,7,1
-4,8,127,0,4,0,1
-4,9,127,1,1,7,1
-4,10,124,1,9,0,1
-4,11,113,1,7,7,1
-4,12,113,8,5,0,1
-4,13,113,1,5,8,1
-4,14,113,1,7,8,1
-4,15,113,1,5,7,1
-4,16,113,1,5,7,1
-4,17,113,1,5,7,1
-4,18,113,1,5,7,1
-4,19,108,1,5,7,1
-4,20,108,1,5,7,1
-4,21,103,1,5,7,1
-4,22,103,1,5,7,1
-4,23,110,1,5,7,1
-4,24,110,1,5,8,1
-4,25,110,1,5,8,1
-4,26,110,1,5,0,1
-4,27,110,0,5,7,1
-4,28,110,[other],5,0,1
-4,29,127,0,4,7,1
-4,30,103,0,5,0,1
-4,31,103,[other],4,0,1
-4,32,127,1,4,7,1
+3,3,112,9,9,5,0
+3,4,112,9,9,2,0
+3,5,112,9,9,2,0
+3,6,112,9,9,2,0
+3,7,112,9,9,2,0
+3,8,112,9,9,2,0
+3,9,112,9,9,2,0
+3,10,112,9,9,7,0
+3,11,112,9,9,7,0
+3,12,112,9,9,7,0
+3,13,112,9,9,7,0
+3,14,112,9,9,7,0
+3,15,112,9,9,7,0
+3,16,112,9,9,7,0
+3,17,112,9,9,7,0
+3,18,112,9,9,7,0
+3,19,112,9,9,7,0
+3,20,112,9,9,7,0
+3,21,112,9,9,7,0
+3,22,112,9,9,7,0
+3,23,112,9,9,7,0
+3,24,112,9,9,7,0
+3,25,112,9,9,7,0
+3,26,112,9,9,7,0
+3,27,112,9,9,7,0
+3,28,112,9,9,7,0
+3,29,112,9,9,7,0
+3,30,112,9,9,7,0
+3,31,112,9,9,7,0
+3,32,112,9,9,7,0
+4,3,112,9,9,7,0
+4,4,112,9,9,7,0
+4,5,112,9,9,7,0
+4,6,112,9,9,7,0
+4,7,112,9,9,7,0
+4,8,112,9,9,7,0
+4,9,112,9,9,5,0
+4,10,112,9,9,5,0
+4,11,112,9,9,5,0
+4,12,112,9,9,5,0
+4,13,112,9,9,5,0
+4,14,112,9,9,5,0
+4,15,112,9,9,7,0
+4,16,112,9,9,7,0
+4,17,112,9,9,7,0
+4,18,112,9,9,7,0
+4,19,112,9,9,7,0
+4,20,112,9,9,7,0
+4,21,112,9,9,7,0
+4,22,112,9,9,7,0
+4,23,112,9,9,7,0
+4,24,112,9,9,7,0
+4,25,112,9,9,7,0
+4,26,112,9,9,7,0
+4,27,112,9,9,7,0
+4,28,112,9,9,7,0
+4,29,112,9,9,7,0
+4,30,112,9,9,7,0
+4,31,112,9,9,7,0
+4,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
index 4f5b8222..c38b58c0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,128,1,5,4,1
-5,4,128,1,5,6,1
-5,5,105,1,5,6,1
-5,6,128,1,5,0,1
-5,7,105,1,5,7,1
-5,8,105,1,5,0,1
-5,9,113,1,7,7,1
-5,10,105,1,9,7,1
-5,11,103,1,9,7,1
-5,12,108,1,9,7,1
-5,13,113,1,5,7,1
-5,14,113,1,5,8,1
-5,15,113,1,5,8,1
-5,16,113,1,5,8,1
-5,17,113,1,5,7,1
-5,18,105,1,5,7,1
-5,19,105,1,5,7,1
-5,20,103,1,5,7,1
-5,21,103,1,5,7,1
-5,22,103,1,5,7,1
-5,23,103,1,5,7,1
-5,24,108,1,5,7,1
-5,25,124,8,5,7,1
-5,26,110,8,5,7,1
-5,27,110,1,5,8,1
-5,28,110,2,7,8,1
-5,29,110,2,5,0,1
-5,30,110,1,5,7,1
-5,31,124,[other],5,0,1
-5,32,127,1,8,0,1
+5,3,111,3,2,2,0
+5,4,111,4,9,8,1
+5,5,111,4,9,7,1
+5,6,111,4,9,7,1
+5,7,111,4,9,7,1
+5,8,111,4,9,7,1
+5,9,111,4,9,7,1
+5,10,102,4,9,7,1
+5,11,102,4,9,7,1
+5,12,102,4,9,7,1
+5,13,102,4,9,7,1
+5,14,122,4,9,5,1
+5,15,122,9,9,2,0
+5,16,121,9,9,5,0
+5,17,121,9,9,5,0
+5,18,112,9,9,2,0
+5,19,112,9,9,2,0
+5,20,112,9,9,2,0
+5,21,112,9,9,2,0
+5,22,112,9,9,7,0
+5,23,112,9,9,7,0
+5,24,112,9,9,7,0
+5,25,112,9,9,7,0
+5,26,112,9,9,7,0
+5,27,112,9,9,7,0
+5,28,112,9,9,7,0
+5,29,112,9,9,7,0
+5,30,112,9,9,7,0
+5,31,112,9,9,7,0
+5,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
index 929f2c8c..d8447e35 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,109,4,8,8,1
-6,5,113,1,0,7,0
-6,6,113,1,7,7,1
-6,7,113,1,7,7,1
-6,8,113,1,7,7,1
-6,9,102,8,7,7,1
-6,10,102,5,7,7,1
-6,11,102,1,7,7,1
-6,12,113,1,7,7,1
-6,13,113,8,5,0,1
-6,14,102,1,5,2,1
-6,15,113,1,7,8,1
-6,16,113,1,1,0,1
-6,17,113,1,5,7,1
-6,18,105,1,5,7,1
-6,19,105,1,5,7,1
-6,20,108,1,7,7,1
-6,21,108,8,7,7,1
-6,22,105,5,5,7,1
-6,23,113,1,5,7,1
-6,24,113,1,5,8,1
-6,25,113,1,5,8,1
-6,26,113,1,5,8,1
-6,27,113,1,5,7,1
-6,28,108,1,5,7,1
-6,29,108,1,5,7,1
-6,30,103,1,5,7,1
-6,31,103,1,5,7,1
-6,32,103,1,5,7,1
-6,33,103,8,5,7,1
-7,3,105,0,9,8,0
-7,4,127,0,1,7,0
-7,5,105,0,9,8,1
-7,6,127,0,1,7,0
-7,7,105,0,9,8,1
-7,8,113,0,1,0,0
-7,9,113,1,9,8,1
-7,10,113,1,1,8,1
-7,11,113,1,9,8,1
-7,12,113,1,1,7,1
-7,13,113,1,9,7,1
-7,14,113,1,9,7,1
-7,15,105,1,5,7,1
-7,16,113,1,9,8,1
-7,17,113,1,5,7,1
-7,18,105,1,5,7,1
-7,19,105,1,5,7,1
-7,20,113,1,5,7,1
-7,21,113,1,5,7,1
-7,22,113,1,5,7,1
-7,23,113,1,5,7,1
-7,24,113,1,5,7,1
-7,25,113,1,5,7,1
-7,26,113,1,5,7,1
-7,27,113,1,5,7,1
-7,28,113,1,5,7,1
-7,29,113,1,5,7,1
-7,30,113,1,5,7,1
-7,31,113,1,5,7,1
-7,32,113,1,5,7,1
+6,4,111,4,9,7,1
+6,5,111,4,9,7,1
+6,6,111,4,9,7,1
+6,7,111,4,9,7,1
+6,8,111,4,9,7,1
+6,9,111,4,9,7,1
+6,10,102,4,9,7,1
+6,11,102,4,9,7,1
+6,12,102,4,9,7,1
+6,13,102,4,9,7,1
+6,14,122,4,9,5,1
+6,15,122,9,9,2,0
+6,16,121,9,9,5,0
+6,17,121,9,9,5,0
+6,18,112,9,9,2,0
+6,19,112,9,9,2,0
+6,20,112,9,9,2,0
+6,21,112,9,9,2,0
+6,22,112,9,9,7,0
+6,23,112,9,9,7,0
+6,24,112,9,9,7,0
+6,25,112,9,9,7,0
+6,26,112,9,9,7,0
+6,27,112,9,9,7,0
+6,28,112,9,9,7,0
+6,29,112,9,9,7,0
+6,30,112,9,9,7,0
+6,31,112,9,9,7,0
+6,32,112,9,9,7,0
+6,33,112,9,9,7,0
+7,3,104,4,9,7,1
+7,4,102,4,9,7,1
+7,5,104,4,9,7,0
+7,6,121,4,9,5,1
+7,7,111,6,9,5,0
+7,8,121,4,9,5,1
+7,9,111,6,9,5,0
+7,10,122,4,9,5,1
+7,11,111,9,9,7,1
+7,12,111,9,9,7,1
+7,13,111,9,9,7,1
+7,14,111,3,9,7,1
+7,15,111,4,9,7,1
+7,16,111,4,9,7,1
+7,17,102,4,9,7,1
+7,18,102,4,9,7,1
+7,19,102,4,9,7,1
+7,20,102,4,9,7,1
+7,21,102,4,9,5,0
+7,22,121,4,9,5,0
+7,23,121,9,9,5,0
+7,24,121,9,9,5,0
+7,25,112,9,9,2,0
+7,26,112,9,9,2,0
+7,27,112,9,9,2,0
+7,28,112,9,9,2,0
+7,29,112,9,9,2,0
+7,30,112,9,9,7,0
+7,31,112,9,9,7,0
+7,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
index 83cccde4..2ed29b99 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,109,1,0,7,0
-8,5,127,1,8,7,1
-8,6,124,1,7,0,1
-8,7,127,1,7,7,0
-8,8,127,8,9,0,1
-8,9,127,1,1,0,1
-8,10,124,1,7,0,1
-8,11,113,1,7,0,1
-8,12,113,1,7,7,1
-8,13,113,8,5,7,1
-8,14,113,1,5,8,1
-8,15,113,1,5,8,1
-8,16,113,1,5,7,1
-8,17,113,1,5,7,1
-8,18,108,1,5,7,1
-8,19,108,1,5,7,1
-8,20,108,1,5,7,1
-8,21,124,1,5,7,1
-8,22,103,8,5,7,1
-8,23,103,1,5,7,1
-8,24,110,8,5,7,1
-8,25,110,1,5,8,1
-8,26,110,2,5,8,1
-8,27,110,1,5,8,1
-8,28,110,[other],7,0,1
-8,29,127,1,7,7,1
-8,30,108,5,7,0,1
-8,31,127,5,5,0,1
-8,32,102,1,4,0,1
-8,33,103,[other],4,0,1
+8,4,111,9,9,7,1
+8,5,111,4,9,7,1
+8,6,111,4,9,7,1
+8,7,111,4,9,7,1
+8,8,111,4,9,7,1
+8,9,111,4,9,7,1
+8,10,111,4,9,7,1
+8,11,111,4,9,7,1
+8,12,102,4,9,7,1
+8,13,102,4,9,7,1
+8,14,102,4,9,7,1
+8,15,102,4,9,7,1
+8,16,122,4,9,5,1
+8,17,122,9,9,2,0
+8,18,121,9,9,5,0
+8,19,121,9,9,5,0
+8,20,112,9,9,2,0
+8,21,112,9,9,2,0
+8,22,112,9,9,2,0
+8,23,112,9,9,2,0
+8,24,112,9,9,7,0
+8,25,112,9,9,7,0
+8,26,112,9,9,7,0
+8,27,112,9,9,7,0
+8,28,112,9,9,7,0
+8,29,112,9,9,7,0
+8,30,112,9,9,7,0
+8,31,112,9,9,7,0
+8,32,112,9,9,7,0
+8,33,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
index eebbcfd3..bdde68a7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,108,1,8,7,1
-9,4,108,1,[other],7,1
-9,5,108,8,5,7,1
-9,6,108,1,[other],7,1
-9,7,108,8,[other],7,1
-9,8,108,8,[other],7,1
-9,9,124,8,[other],7,1
-9,10,113,8,[other],2,1
-9,11,113,1,[other],8,1
-9,12,113,1,[other],8,1
-9,13,113,1,[other],8,1
-9,14,113,1,[other],0,1
-9,15,113,1,9,2,1
-9,16,113,1,5,7,1
-9,17,105,1,5,7,1
-9,18,105,1,8,7,1
-9,19,103,1,5,7,1
-9,20,105,1,5,7,1
-9,21,113,1,5,7,1
-9,22,113,1,7,7,1
-9,23,113,8,5,7,1
-9,24,113,1,5,8,1
-9,25,113,1,5,8,1
-9,26,113,1,5,7,1
-9,27,108,1,5,7,1
-9,28,108,1,5,7,1
-9,29,108,1,5,7,1
-9,30,108,1,5,7,1
-9,31,124,1,5,7,1
-9,32,103,8,5,7,1
+9,3,113,4,2,7,1
+9,4,113,4,9,7,1
+9,5,113,4,9,7,1
+9,6,113,4,9,7,1
+9,7,111,4,9,7,1
+9,8,102,4,9,7,1
+9,9,102,4,9,7,0
+9,10,113,4,9,7,1
+9,11,113,4,9,7,1
+9,12,113,4,9,7,1
+9,13,113,9,9,7,1
+9,14,111,9,9,7,1
+9,15,111,9,9,7,1
+9,16,111,9,9,7,1
+9,17,111,4,9,7,1
+9,18,102,4,9,7,1
+9,19,102,4,9,7,1
+9,20,102,4,9,7,1
+9,21,102,4,9,7,1
+9,22,102,4,9,7,1
+9,23,104,6,9,5,0
+9,24,121,9,9,5,1
+9,25,122,9,9,2,0
+9,26,112,9,9,2,0
+9,27,112,9,9,2,0
+9,28,112,9,9,2,0
+9,29,112,9,9,2,0
+9,30,112,9,9,7,0
+9,31,112,9,9,7,0
+9,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index a90c5af9..9a24e80c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,104,7,[mask],[unknown],1
-0,5,104,7,9,2,0
-0,6,104,0,9,3,0
-0,7,103,0,[mask],8,0
-0,8,104,7,1,5,0
-0,9,103,5,1,[unknown],0
-0,10,112,0,1,5,0
-0,11,112,7,1,5,0
-0,12,112,5,1,5,0
-0,13,103,0,1,[unknown],0
-0,14,112,7,1,5,0
-0,15,112,0,1,5,0
-0,16,112,7,1,5,0
-0,17,112,7,1,5,0
-0,18,112,7,[mask],5,1
-0,19,119,7,[mask],5,1
-0,20,119,7,[mask],8,0
-0,21,104,0,1,5,1
-0,22,104,7,[mask],8,1
-0,23,104,7,[mask],[unknown],1
-0,24,104,7,[mask],8,1
-0,25,104,7,[mask],8,1
-0,26,108,9,[mask],3,2
-0,27,104,9,1,5,2
-0,28,112,9,[other],1,0
-0,29,111,8,1,4,2
-0,30,112,9,[other],1,2
-0,31,104,9,[other],5,2
-0,32,104,9,[other],1,[mask]
-0,33,104,9,[other],8,[mask]
+0,4,127,1,1,2,1
+0,5,127,1,8,8,1
+0,6,113,1,1,7,0
+0,7,105,8,9,7,1
+0,8,127,1,1,0,1
+0,9,113,1,7,7,1
+0,10,113,8,9,1,1
+0,11,113,1,1,7,1
+0,12,105,8,5,7,1
+0,13,113,1,5,8,1
+0,14,113,1,[other],8,1
+0,15,113,1,5,7,1
+0,16,105,1,5,7,1
+0,17,108,1,8,7,1
+0,18,108,1,5,7,1
+0,19,124,1,5,7,1
+0,20,124,1,5,7,1
+0,21,105,8,5,7,1
+0,22,105,1,5,7,1
+0,23,110,1,5,7,1
+0,24,110,1,5,8,1
+0,25,110,1,7,8,1
+0,26,110,[other],7,7,1
+0,27,110,5,7,7,1
+0,28,103,5,7,0,1
+0,29,102,1,5,0,1
+0,30,103,1,7,7,1
+0,31,102,5,5,0,1
+0,32,103,1,7,0,1
+0,33,102,[mask],7,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index 61720a4a..12e2155a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,104,7,9,1,1
-1,5,104,7,9,8,1
-1,6,104,0,9,8,1
-1,7,104,7,[mask],8,1
-1,8,104,7,9,8,1
-1,9,104,7,9,[unknown],0
-1,10,118,7,9,[unknown],1
-1,11,129,7,9,[unknown],[other]
-1,12,112,7,9,[unknown],0
-1,13,108,7,9,[unknown],2
-1,14,104,9,9,5,2
-1,15,129,9,8,8,[other]
-1,16,112,9,2,5,0
-1,17,104,9,1,5,[other]
-1,18,104,0,1,8,0
-1,19,112,0,1,8,[other]
-1,20,112,0,[mask],5,0
-1,21,112,0,1,5,[mask]
-1,22,104,7,[mask],5,0
-1,23,129,9,1,8,1
-1,24,104,7,1,8,0
-1,25,112,5,1,5,0
-1,26,112,0,1,5,0
-1,27,112,7,1,5,0
-1,28,112,0,1,5,1
-1,29,112,7,[mask],[unknown],0
-1,30,108,7,1,5,1
-1,31,119,7,[mask],[unknown],0
-1,32,119,0,1,5,1
-1,33,104,7,[mask],5,1
+1,4,121,1,0,7,0
+1,5,108,3,0,8,1
+1,6,103,1,0,7,0
+1,7,103,8,0,7,1
+1,8,108,1,0,2,1
+1,9,113,1,0,7,1
+1,10,108,8,7,7,1
+1,11,124,8,7,7,1
+1,12,105,8,7,7,1
+1,13,113,8,5,7,1
+1,14,113,1,[other],8,1
+1,15,113,1,[other],8,1
+1,16,113,1,[other],7,1
+1,17,108,1,5,7,1
+1,18,124,1,[other],7,1
+1,19,108,1,5,7,1
+1,20,124,1,5,7,1
+1,21,124,1,5,7,1
+1,22,105,8,5,7,1
+1,23,105,1,5,7,1
+1,24,124,8,5,7,1
+1,25,110,8,5,7,1
+1,26,110,1,5,8,1
+1,27,110,1,7,8,1
+1,28,110,6,7,7,1
+1,29,110,1,7,0,1
+1,30,113,1,7,0,1
+1,31,113,1,7,0,1
+1,32,113,1,7,0,1
+1,33,113,[mask],5,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index 32aad96e..5cf7bea0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,112,2,[mask],5,0
-2,4,129,7,1,5,1
-2,5,104,7,[mask],8,0
-2,6,129,7,1,5,1
-2,7,104,7,[mask],8,0
-2,8,112,7,1,5,1
-2,9,107,7,[mask],[unknown],1
-2,10,107,7,[mask],[unknown],1
-2,11,104,7,[mask],[unknown],1
-2,12,108,7,[mask],[unknown],1
-2,13,104,9,1,5,2
-2,14,108,7,[mask],[unknown],0
-2,15,104,9,1,5,2
-2,16,104,7,[mask],[unknown],2
-2,17,104,9,[mask],1,2
-2,18,112,9,[other],1,[mask]
-2,19,104,9,[other],5,[mask]
-2,20,112,9,8,8,[mask]
-2,21,112,9,6,5,[mask]
-2,22,112,9,6,8,[mask]
-2,23,112,9,6,5,[mask]
-2,24,104,9,6,8,[other]
-2,25,112,0,8,8,[other]
-2,26,112,0,[mask],8,0
-2,27,104,4,1,8,[mask]
-2,28,112,0,1,8,[other]
-2,29,112,0,[mask],8,0
-2,30,104,1,1,5,2
-2,31,112,0,1,8,0
-2,32,112,1,1,5,2
+2,3,128,1,8,6,0
+2,4,105,0,8,7,2
+2,5,105,1,8,2,1
+2,6,113,1,8,7,2
+2,7,105,1,9,7,1
+2,8,105,1,7,7,1
+2,9,113,1,7,7,1
+2,10,113,1,5,7,1
+2,11,113,1,[other],8,1
+2,12,113,1,[other],7,1
+2,13,113,8,5,7,1
+2,14,113,1,[other],8,1
+2,15,113,1,5,7,1
+2,16,113,1,5,7,1
+2,17,105,1,5,7,1
+2,18,108,1,5,7,1
+2,19,108,1,5,7,1
+2,20,108,1,5,7,1
+2,21,124,1,5,7,1
+2,22,105,8,5,7,1
+2,23,110,1,5,7,1
+2,24,110,8,5,8,1
+2,25,110,1,5,8,1
+2,26,110,[other],5,8,1
+2,27,127,1,7,8,1
+2,28,103,5,7,0,1
+2,29,103,1,5,7,1
+2,30,103,1,5,0,1
+2,31,103,[other],4,0,1
+2,32,127,1,4,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index f0a3eaa7..4e1d63dd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,104,0,[mask],8,1
-3,4,104,0,[mask],8,0
-3,5,104,0,[mask],8,1
-3,6,104,0,[mask],8,0
-3,7,104,0,[mask],8,1
-3,8,104,0,[mask],6,0
-3,9,112,0,[mask],5,0
-3,10,112,1,1,5,2
-3,11,108,2,1,5,0
-3,12,129,9,1,5,2
-3,13,112,9,1,5,0
-3,14,129,9,1,5,0
-3,15,112,9,1,5,0
-3,16,129,0,1,5,1
-3,17,112,7,[mask],5,1
-3,18,119,7,[mask],[unknown],1
-3,19,104,7,[mask],5,0
-3,20,104,7,1,5,1
-3,21,104,7,[mask],8,0
-3,22,108,7,1,5,1
-3,23,104,7,[mask],8,0
-3,24,108,7,[mask],5,1
-3,25,104,9,[mask],8,2
-3,26,112,9,1,5,0
-3,27,112,9,1,5,2
-3,28,103,7,1,5,0
-3,29,112,9,1,5,0
-3,30,104,7,1,5,0
-3,31,112,0,1,8,0
-3,32,112,0,[mask],5,0
-4,3,108,9,1,5,1
-4,4,104,7,1,4,0
-4,5,108,0,1,4,0
-4,6,108,0,1,5,0
-4,7,112,7,1,5,0
-4,8,108,7,1,5,0
-4,9,129,0,1,5,2
-4,10,112,7,1,5,2
-4,11,112,7,[mask],5,0
-4,12,112,7,1,5,1
-4,13,107,7,[mask],[unknown],1
-4,14,107,7,[mask],[unknown],1
-4,15,107,7,[mask],[unknown],1
-4,16,107,7,[mask],[unknown],1
-4,17,104,7,[mask],[unknown],1
-4,18,104,7,[mask],[unknown],1
-4,19,104,7,[mask],[unknown],1
-4,20,118,7,[mask],[unknown],1
-4,21,104,9,[mask],[unknown],2
-4,22,104,9,[mask],1,2
-4,23,104,9,[mask],1,2
-4,24,104,9,[other],1,2
-4,25,104,9,[other],1,[mask]
-4,26,104,9,[other],[unknown],[mask]
-4,27,109,9,[other],[unknown],[mask]
-4,28,104,9,6,5,[mask]
-4,29,112,9,[other],1,[mask]
-4,30,118,2,6,1,[mask]
-4,31,110,1,[other],8,[other]
-4,32,112,[other],6,[unknown],[mask]
+3,3,127,5,0,2,0
+3,4,127,4,0,8,0
+3,5,127,4,0,8,0
+3,6,127,4,0,8,0
+3,7,127,4,0,8,0
+3,8,127,4,0,8,1
+3,9,108,1,0,8,1
+3,10,103,1,7,7,1
+3,11,103,8,9,7,1
+3,12,103,1,7,7,1
+3,13,113,8,7,7,1
+3,14,113,8,5,7,1
+3,15,113,1,5,7,1
+3,16,113,1,5,7,1
+3,17,113,1,5,7,1
+3,18,113,1,[other],8,1
+3,19,113,1,[other],7,1
+3,20,108,8,5,7,1
+3,21,108,1,2,7,1
+3,22,108,1,5,7,1
+3,23,108,1,5,7,1
+3,24,124,1,5,7,1
+3,25,124,1,5,7,1
+3,26,124,8,5,7,1
+3,27,124,1,5,7,1
+3,28,124,8,5,7,1
+3,29,110,8,5,7,1
+3,30,110,1,5,8,1
+3,31,110,1,7,8,1
+3,32,110,[other],7,7,1
+4,3,127,0,9,[unknown],1
+4,4,127,0,1,8,1
+4,5,102,0,1,8,1
+4,6,127,1,4,8,0
+4,7,127,0,1,7,1
+4,8,127,0,4,0,1
+4,9,127,1,1,7,1
+4,10,124,1,9,0,1
+4,11,113,1,7,7,1
+4,12,113,8,5,0,1
+4,13,113,1,5,8,1
+4,14,113,1,7,8,1
+4,15,113,1,5,7,1
+4,16,113,1,5,7,1
+4,17,113,1,5,7,1
+4,18,113,1,5,7,1
+4,19,108,1,5,7,1
+4,20,108,1,5,7,1
+4,21,103,1,5,7,1
+4,22,103,1,5,7,1
+4,23,110,1,5,7,1
+4,24,110,1,5,8,1
+4,25,110,1,5,8,1
+4,26,110,1,5,0,1
+4,27,110,0,5,7,1
+4,28,110,[other],5,0,1
+4,29,127,0,4,7,1
+4,30,103,0,5,0,1
+4,31,103,[other],4,0,1
+4,32,127,1,4,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index 5a908cad..4f5b8222 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,102,7,9,9,1
-5,4,102,7,9,9,1
-5,5,105,7,9,9,1
-5,6,105,7,9,9,1
-5,7,105,7,9,9,1
-5,8,105,7,9,9,1
-5,9,111,[mask],9,9,1
-5,10,104,7,9,0,1
-5,11,103,7,9,8,0
-5,12,105,7,9,9,1
-5,13,111,7,9,9,0
-5,14,103,7,9,3,1
-5,15,111,[mask],9,0,0
-5,16,104,7,9,8,1
-5,17,104,0,9,8,0
-5,18,103,0,9,8,1
-5,19,104,7,9,8,0
-5,20,103,0,9,8,0
-5,21,104,7,9,5,1
-5,22,104,7,9,8,[other]
-5,23,101,0,9,8,0
-5,24,105,7,9,[unknown],0
-5,25,103,7,9,8,1
-5,26,104,7,9,8,0
-5,27,101,0,8,8,0
-5,28,112,7,9,9,1
-5,29,104,7,9,8,[other]
-5,30,129,0,1,8,0
-5,31,104,7,[mask],5,1
-5,32,104,7,1,8,[other]
+5,3,128,1,5,4,1
+5,4,128,1,5,6,1
+5,5,105,1,5,6,1
+5,6,128,1,5,0,1
+5,7,105,1,5,7,1
+5,8,105,1,5,0,1
+5,9,113,1,7,7,1
+5,10,105,1,9,7,1
+5,11,103,1,9,7,1
+5,12,108,1,9,7,1
+5,13,113,1,5,7,1
+5,14,113,1,5,8,1
+5,15,113,1,5,8,1
+5,16,113,1,5,8,1
+5,17,113,1,5,7,1
+5,18,105,1,5,7,1
+5,19,105,1,5,7,1
+5,20,103,1,5,7,1
+5,21,103,1,5,7,1
+5,22,103,1,5,7,1
+5,23,103,1,5,7,1
+5,24,108,1,5,7,1
+5,25,124,8,5,7,1
+5,26,110,8,5,7,1
+5,27,110,1,5,8,1
+5,28,110,2,7,8,1
+5,29,110,2,5,0,1
+5,30,110,1,5,7,1
+5,31,124,[other],5,0,1
+5,32,127,1,8,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index 58547214..929f2c8c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,129,9,1,5,2
-6,5,129,9,1,5,0
-6,6,129,9,1,5,1
-6,7,112,7,1,5,0
-6,8,112,7,1,5,1
-6,9,112,7,[mask],[unknown],0
-6,10,129,7,1,5,1
-6,11,104,7,[mask],5,0
-6,12,112,7,[mask],5,1
-6,13,119,7,[mask],[unknown],1
-6,14,104,7,[mask],[unknown],1
-6,15,104,7,[mask],[unknown],1
-6,16,104,7,[mask],[unknown],1
-6,17,104,7,1,8,2
-6,18,108,7,[mask],1,2
-6,19,104,9,[other],1,2
-6,20,104,9,[mask],1,2
-6,21,118,9,[other],1,2
-6,22,112,9,[other],1,[mask]
-6,23,118,9,[other],5,[mask]
-6,24,112,9,[other],7,[mask]
-6,25,104,2,6,8,[mask]
-6,26,104,9,6,8,[other]
-6,27,112,9,6,8,[mask]
-6,28,104,1,6,8,[other]
-6,29,112,0,[mask],8,[other]
-6,30,112,1,[mask],8,0
-6,31,112,1,[mask],8,[mask]
-6,32,104,1,[mask],8,[mask]
-6,33,[other],9,[other],8,[mask]
-7,3,129,9,1,5,1
-7,4,112,2,1,5,1
-7,5,112,1,1,5,1
-7,6,112,1,[mask],4,1
-7,7,107,1,[mask],5,1
-7,8,107,2,[mask],4,1
-7,9,104,7,[mask],5,1
-7,10,104,7,[mask],0,1
-7,11,108,7,[mask],0,1
-7,12,104,7,[mask],0,1
-7,13,104,7,9,0,1
-7,14,104,0,9,3,1
-7,15,104,0,[mask],3,1
-7,16,104,[mask],[mask],3,1
-7,17,104,5,9,3,0
-7,18,104,0,9,3,0
-7,19,103,0,9,3,0
-7,20,111,6,9,3,1
-7,21,103,6,9,3,0
-7,22,105,6,[mask],[unknown],1
-7,23,105,7,[mask],[unknown],2
-7,24,108,7,9,[unknown],0
-7,25,104,0,9,5,0
-7,26,104,0,9,5,1
-7,27,104,0,[mask],8,0
-7,28,104,0,[mask],5,1
-7,29,104,7,9,8,0
-7,30,112,0,1,5,0
-7,31,112,7,1,5,0
-7,32,112,7,1,5,0
+6,4,109,4,8,8,1
+6,5,113,1,0,7,0
+6,6,113,1,7,7,1
+6,7,113,1,7,7,1
+6,8,113,1,7,7,1
+6,9,102,8,7,7,1
+6,10,102,5,7,7,1
+6,11,102,1,7,7,1
+6,12,113,1,7,7,1
+6,13,113,8,5,0,1
+6,14,102,1,5,2,1
+6,15,113,1,7,8,1
+6,16,113,1,1,0,1
+6,17,113,1,5,7,1
+6,18,105,1,5,7,1
+6,19,105,1,5,7,1
+6,20,108,1,7,7,1
+6,21,108,8,7,7,1
+6,22,105,5,5,7,1
+6,23,113,1,5,7,1
+6,24,113,1,5,8,1
+6,25,113,1,5,8,1
+6,26,113,1,5,8,1
+6,27,113,1,5,7,1
+6,28,108,1,5,7,1
+6,29,108,1,5,7,1
+6,30,103,1,5,7,1
+6,31,103,1,5,7,1
+6,32,103,1,5,7,1
+6,33,103,8,5,7,1
+7,3,105,0,9,8,0
+7,4,127,0,1,7,0
+7,5,105,0,9,8,1
+7,6,127,0,1,7,0
+7,7,105,0,9,8,1
+7,8,113,0,1,0,0
+7,9,113,1,9,8,1
+7,10,113,1,1,8,1
+7,11,113,1,9,8,1
+7,12,113,1,1,7,1
+7,13,113,1,9,7,1
+7,14,113,1,9,7,1
+7,15,105,1,5,7,1
+7,16,113,1,9,8,1
+7,17,113,1,5,7,1
+7,18,105,1,5,7,1
+7,19,105,1,5,7,1
+7,20,113,1,5,7,1
+7,21,113,1,5,7,1
+7,22,113,1,5,7,1
+7,23,113,1,5,7,1
+7,24,113,1,5,7,1
+7,25,113,1,5,7,1
+7,26,113,1,5,7,1
+7,27,113,1,5,7,1
+7,28,113,1,5,7,1
+7,29,113,1,5,7,1
+7,30,113,1,5,7,1
+7,31,113,1,5,7,1
+7,32,113,1,5,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index d05f5e62..83cccde4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,104,7,1,5,1
-8,5,104,7,1,5,1
-8,6,112,7,[mask],[unknown],1
-8,7,107,7,[mask],5,1
-8,8,104,7,[mask],[unknown],1
-8,9,108,7,[mask],[unknown],1
-8,10,104,9,[mask],5,1
-8,11,108,7,[mask],0,1
-8,12,108,7,9,0,2
-8,13,104,7,9,1,2
-8,14,104,7,9,1,1
-8,15,104,7,9,[unknown],0
-8,16,104,9,8,5,2
-8,17,104,9,8,8,[other]
-8,18,112,9,[mask],3,[mask]
-8,19,104,9,[mask],8,[mask]
-8,20,112,9,[other],8,[mask]
-8,21,112,1,[mask],5,0
-8,22,112,1,1,[unknown],[mask]
-8,23,112,1,1,8,[mask]
-8,24,112,2,1,5,[mask]
-8,25,104,9,1,8,1
-8,26,104,0,[mask],8,1
-8,27,104,0,[mask],8,1
-8,28,104,0,[mask],8,1
-8,29,104,7,[mask],[unknown],1
-8,30,104,7,9,[unknown],1
-8,31,104,7,9,[unknown],1
-8,32,103,5,9,[unknown],0
-8,33,108,5,9,3,2
+8,4,109,1,0,7,0
+8,5,127,1,8,7,1
+8,6,124,1,7,0,1
+8,7,127,1,7,7,0
+8,8,127,8,9,0,1
+8,9,127,1,1,0,1
+8,10,124,1,7,0,1
+8,11,113,1,7,0,1
+8,12,113,1,7,7,1
+8,13,113,8,5,7,1
+8,14,113,1,5,8,1
+8,15,113,1,5,8,1
+8,16,113,1,5,7,1
+8,17,113,1,5,7,1
+8,18,108,1,5,7,1
+8,19,108,1,5,7,1
+8,20,108,1,5,7,1
+8,21,124,1,5,7,1
+8,22,103,8,5,7,1
+8,23,103,1,5,7,1
+8,24,110,8,5,7,1
+8,25,110,1,5,8,1
+8,26,110,2,5,8,1
+8,27,110,1,5,8,1
+8,28,110,[other],7,0,1
+8,29,127,1,7,7,1
+8,30,108,5,7,0,1
+8,31,127,5,5,0,1
+8,32,102,1,4,0,1
+8,33,103,[other],4,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index e30646b9..eebbcfd3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,104,7,9,0,1
-9,4,104,7,9,8,1
-9,5,104,7,[mask],8,1
-9,6,104,7,9,[unknown],1
-9,7,104,7,9,8,0
-9,8,104,7,9,8,1
-9,9,104,7,9,[unknown],0
-9,10,108,7,9,[unknown],0
-9,11,103,6,1,[unknown],2
-9,12,108,5,1,[unknown],2
-9,13,108,9,1,[unknown],0
-9,14,129,9,1,5,2
-9,15,112,9,1,8,0
-9,16,112,0,1,5,2
-9,17,112,7,1,5,0
-9,18,112,0,1,5,1
-9,19,112,7,[mask],[unknown],0
-9,20,102,4,1,5,1
-9,21,112,7,[mask],[unknown],0
-9,22,119,7,[mask],5,1
-9,23,104,7,[mask],8,1
-9,24,104,7,[mask],[unknown],1
-9,25,104,7,[mask],[unknown],1
-9,26,104,7,[mask],8,1
-9,27,104,9,[mask],1,2
-9,28,104,7,9,1,1
-9,29,104,9,8,1,2
-9,30,118,9,[mask],[unknown],2
-9,31,112,9,[other],1,2
-9,32,104,5,[other],1,[mask]
+9,3,108,1,8,7,1
+9,4,108,1,[other],7,1
+9,5,108,8,5,7,1
+9,6,108,1,[other],7,1
+9,7,108,8,[other],7,1
+9,8,108,8,[other],7,1
+9,9,124,8,[other],7,1
+9,10,113,8,[other],2,1
+9,11,113,1,[other],8,1
+9,12,113,1,[other],8,1
+9,13,113,1,[other],8,1
+9,14,113,1,[other],0,1
+9,15,113,1,9,2,1
+9,16,113,1,5,7,1
+9,17,105,1,5,7,1
+9,18,105,1,8,7,1
+9,19,103,1,5,7,1
+9,20,105,1,5,7,1
+9,21,113,1,5,7,1
+9,22,113,1,7,7,1
+9,23,113,8,5,7,1
+9,24,113,1,5,8,1
+9,25,113,1,5,8,1
+9,26,113,1,5,7,1
+9,27,108,1,5,7,1
+9,28,108,1,5,7,1
+9,29,108,1,5,7,1
+9,30,108,1,5,7,1
+9,31,124,1,5,7,1
+9,32,103,8,5,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index aa9878ce..fd49670a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,121,1,1,9,1
-0,5,105,0,[other],8,2
-0,6,113,1,[other],[mask],0
-0,7,105,8,[other],1,1
-0,8,113,1,[other],2,[unknown]
-0,9,113,3,[other],1,1
-0,10,113,1,[other],8,1
-0,11,113,1,[other],[mask],1
-0,12,105,8,9,1,[other]
-0,13,113,1,8,0,1
-0,14,105,0,[other],1,[other]
-0,15,113,[other],8,9,1
-0,16,105,0,[other],1,1
-0,17,113,1,8,8,1
-0,18,113,1,[other],7,1
-0,19,105,8,[other],1,[other]
-0,20,113,1,8,8,[unknown]
-0,21,113,1,[other],[mask],[other]
-0,22,113,[mask],[other],8,1
-0,23,113,1,[other],2,1
-0,24,113,[mask],[other],1,1
-0,25,113,1,[other],2,1
-0,26,113,1,[other],1,1
-0,27,113,1,[other],1,1
-0,28,105,1,[other],1,1
-0,29,108,1,8,0,1
-0,30,103,1,0,7,1
-0,31,103,1,[other],7,1
-0,32,108,5,9,0,1
-0,33,105,0,7,7,1
+0,4,119,9,2,7,[other]
+0,5,111,9,2,7,[other]
+0,6,111,9,2,7,[other]
+0,7,111,9,9,7,[other]
+0,8,111,9,9,7,[other]
+0,9,107,1,9,7,[other]
+0,10,104,1,9,7,[other]
+0,11,102,1,9,7,[other]
+0,12,104,6,9,7,[other]
+0,13,104,1,9,7,[other]
+0,14,104,1,9,[unknown],[other]
+0,15,104,6,9,[unknown],[other]
+0,16,104,6,9,[unknown],[other]
+0,17,104,6,9,7,[other]
+0,18,104,6,9,7,[other]
+0,19,104,6,9,7,[other]
+0,20,104,6,9,7,[other]
+0,21,104,6,9,7,[other]
+0,22,104,6,9,7,[other]
+0,23,104,6,9,7,[other]
+0,24,104,6,9,7,[other]
+0,25,104,6,9,7,[other]
+0,26,104,6,9,7,[other]
+0,27,104,6,9,7,[other]
+0,28,104,6,9,7,[other]
+0,29,104,6,9,7,[other]
+0,30,104,6,9,7,[other]
+0,31,104,6,9,7,[other]
+0,32,104,6,9,7,[other]
+0,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index 5e0f7cd5..c47a6367 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,107,5,[other],[mask],[other]
-1,5,121,3,[other],0,[other]
-1,6,129,3,[other],8,[other]
-1,7,121,[other],[other],8,[other]
-1,8,[unknown],5,[other],[mask],[other]
-1,9,121,6,3,4,1
-1,10,105,1,[other],8,[unknown]
-1,11,121,1,7,[mask],[other]
-1,12,105,0,7,1,[other]
-1,13,105,[other],7,5,0
-1,14,105,1,1,1,1
-1,15,105,1,7,[other],[other]
-1,16,105,0,7,0,1
-1,17,105,1,7,1,1
-1,18,113,1,7,1,1
-1,19,113,1,7,1,1
-1,20,113,1,[other],1,1
-1,21,113,1,1,1,1
-1,22,113,0,[other],1,1
-1,23,113,1,1,0,1
-1,24,108,1,[other],1,1
-1,25,108,0,[other],7,1
-1,26,108,5,[other],0,1
-1,27,113,5,[other],7,1
-1,28,105,8,[other],0,[other]
-1,29,113,[other],8,2,[other]
-1,30,113,[mask],[other],[unknown],1
-1,31,113,1,[other],2,1
-1,32,113,1,[other],8,1
-1,33,113,1,[other],[mask],1
+1,4,111,9,2,7,[other]
+1,5,111,1,9,7,[other]
+1,6,111,1,9,7,[other]
+1,7,111,1,9,7,[other]
+1,8,111,1,9,7,[other]
+1,9,111,1,9,7,[other]
+1,10,102,1,9,7,[other]
+1,11,129,1,9,[unknown],[other]
+1,12,102,1,9,[unknown],[other]
+1,13,104,6,9,[unknown],[other]
+1,14,104,6,9,[unknown],[other]
+1,15,104,6,9,[unknown],[other]
+1,16,104,6,9,[unknown],[other]
+1,17,104,6,9,7,[other]
+1,18,104,9,9,7,[other]
+1,19,104,6,9,7,[other]
+1,20,104,6,9,7,[other]
+1,21,104,6,9,7,[other]
+1,22,104,6,9,7,[other]
+1,23,104,6,9,7,[other]
+1,24,104,6,9,7,[other]
+1,25,104,6,9,7,[other]
+1,26,104,6,9,7,[other]
+1,27,104,6,9,7,[other]
+1,28,104,6,9,7,[other]
+1,29,104,6,9,7,[other]
+1,30,104,6,9,7,[other]
+1,31,104,6,9,7,[other]
+1,32,104,6,9,7,[other]
+1,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index fa4d361a..d60f3445 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,128,0,7,[unknown],2
-2,4,113,8,[other],6,2
-2,5,109,1,[other],[mask],1
-2,6,128,1,[other],7,2
-2,7,105,8,[other],6,[other]
-2,8,121,8,1,[mask],1
-2,9,114,1,[other],8,2
-2,10,113,1,[other],[mask],1
-2,11,104,1,[other],0,1
-2,12,113,1,7,0,1
-2,13,113,1,[other],1,1
-2,14,113,1,[other],1,1
-2,15,113,1,[other],1,1
-2,16,105,1,[other],1,1
-2,17,113,1,[other],0,1
-2,18,105,0,[other],1,1
-2,19,108,0,5,0,1
-2,20,113,0,[other],0,1
-2,21,108,0,9,2,1
-2,22,113,0,[other],0,1
-2,23,113,8,[other],2,1
-2,24,113,3,[other],8,1
-2,25,113,1,[other],8,1
-2,26,113,1,[other],8,1
-2,27,113,1,[other],0,1
-2,28,113,1,[other],7,1
-2,29,105,8,[other],0,1
-2,30,113,1,[other],2,1
-2,31,113,8,[other],1,1
-2,32,113,1,[other],2,1
+2,3,115,9,2,2,[unknown]
+2,4,118,9,2,[mask],[mask]
+2,5,115,9,9,7,[mask]
+2,6,115,9,9,7,[mask]
+2,7,115,9,9,7,[mask]
+2,8,119,9,9,7,[mask]
+2,9,104,9,9,7,[other]
+2,10,104,1,9,7,[other]
+2,11,104,6,9,7,[other]
+2,12,104,6,9,7,[other]
+2,13,104,6,9,7,[other]
+2,14,104,6,9,7,[other]
+2,15,104,6,9,7,[other]
+2,16,104,6,9,7,[other]
+2,17,104,6,9,7,[other]
+2,18,104,6,9,7,[other]
+2,19,104,6,9,7,[other]
+2,20,104,6,9,7,[other]
+2,21,104,6,9,7,[other]
+2,22,104,6,9,7,[other]
+2,23,104,6,9,7,[other]
+2,24,104,6,9,7,[other]
+2,25,104,6,9,7,[other]
+2,26,104,6,9,7,[other]
+2,27,104,6,9,7,[other]
+2,28,104,6,9,7,[other]
+2,29,104,6,9,7,[other]
+2,30,104,6,9,7,[other]
+2,31,104,6,9,7,[other]
+2,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index 865dd214..1923a1cb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,128,5,1,1,2
-3,4,105,2,9,5,[unknown]
-3,5,128,1,1,[mask],1
-3,6,105,8,[other],1,2
-3,7,128,1,9,[mask],1
-3,8,128,0,9,1,1
-3,9,119,1,[other],[mask],2
-3,10,104,1,8,[mask],1
-3,11,104,1,9,0,[other]
-3,12,113,3,[other],1,[other]
-3,13,113,1,[other],[mask],[other]
-3,14,113,3,[other],[unknown],[other]
-3,15,113,3,[other],[mask],[other]
-3,16,105,3,[other],8,[other]
-3,17,105,1,7,8,[other]
-3,18,113,1,7,1,[other]
-3,19,113,1,[other],1,1
-3,20,113,1,[other],1,[other]
-3,21,113,1,[other],[mask],1
-3,22,120,1,[other],1,1
-3,23,113,1,7,7,1
-3,24,105,1,7,1,1
-3,25,113,1,7,1,1
-3,26,113,0,1,1,1
-3,27,105,0,[other],1,[other]
-3,28,108,0,8,0,1
-3,29,113,0,[other],7,1
-3,30,108,8,[other],2,1
-3,31,113,[mask],[other],7,1
-3,32,124,[mask],[other],2,1
-4,3,102,0,[other],[unknown],[unknown]
-4,4,102,8,[other],6,[unknown]
-4,5,105,8,[other],6,[mask]
-4,6,127,1,[other],2,1
-4,7,127,[mask],7,[unknown],1
-4,8,102,[mask],1,2,1
-4,9,127,[mask],[other],8,[unknown]
-4,10,106,[mask],1,0,1
-4,11,113,1,[other],1,1
-4,12,106,8,1,[mask],1
-4,13,105,1,[other],[mask],1
-4,14,106,8,9,0,1
-4,15,113,1,[other],0,1
-4,16,113,1,[other],2,1
-4,17,113,8,[other],1,1
-4,18,113,1,[other],2,1
-4,19,113,3,[other],1,2
-4,20,113,1,[other],[mask],1
-4,21,105,8,[other],1,1
-4,22,113,1,8,[mask],1
-4,23,105,3,[other],1,1
-4,24,113,1,8,0,1
-4,25,105,1,[other],1,1
-4,26,113,1,9,0,1
-4,27,113,1,[other],1,1
-4,28,113,1,[other],1,2
-4,29,113,1,[other],1,[other]
-4,30,113,0,[other],1,[other]
-4,31,113,0,[other],[mask],[other]
-4,32,121,0,[other],2,[other]
+3,3,112,9,2,[mask],[mask]
+3,4,115,9,2,9,[other]
+3,5,112,9,2,2,[mask]
+3,6,112,9,2,2,[mask]
+3,7,112,9,2,2,[mask]
+3,8,112,9,9,2,[other]
+3,9,112,9,2,2,[other]
+3,10,112,9,9,7,[other]
+3,11,112,9,9,7,[other]
+3,12,112,9,9,7,[other]
+3,13,112,9,9,7,[other]
+3,14,112,9,9,7,[other]
+3,15,112,9,9,7,[other]
+3,16,112,9,9,7,[other]
+3,17,112,9,9,7,[other]
+3,18,112,9,9,7,[other]
+3,19,112,9,9,7,[other]
+3,20,112,9,9,7,[other]
+3,21,112,9,9,7,[other]
+3,22,112,9,9,7,[other]
+3,23,112,9,9,7,[other]
+3,24,112,9,9,7,[other]
+3,25,112,9,9,7,[other]
+3,26,112,9,9,7,[other]
+3,27,112,9,9,7,[other]
+3,28,112,9,9,7,[other]
+3,29,112,9,9,7,[other]
+3,30,112,9,9,7,[other]
+3,31,112,9,9,7,[other]
+3,32,112,9,9,7,[other]
+4,3,107,9,2,7,[other]
+4,4,107,9,9,7,[other]
+4,5,107,9,9,7,[other]
+4,6,107,9,9,7,[other]
+4,7,107,9,9,7,[other]
+4,8,104,6,9,7,[other]
+4,9,104,6,9,7,[other]
+4,10,104,6,9,7,[other]
+4,11,104,6,9,7,[other]
+4,12,104,6,9,7,[other]
+4,13,104,6,9,7,[other]
+4,14,104,6,9,7,[other]
+4,15,104,6,9,7,[other]
+4,16,104,6,9,7,[other]
+4,17,104,6,9,7,[other]
+4,18,104,6,9,7,[other]
+4,19,104,6,9,7,[other]
+4,20,104,6,9,7,[other]
+4,21,104,6,9,7,[other]
+4,22,104,6,9,7,[other]
+4,23,104,6,9,7,[other]
+4,24,104,6,9,7,[other]
+4,25,104,6,9,7,[other]
+4,26,104,6,9,7,[other]
+4,27,104,6,9,7,[other]
+4,28,104,6,9,7,[other]
+4,29,104,6,9,7,[other]
+4,30,104,6,9,7,[other]
+4,31,104,6,9,7,[other]
+4,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index 67ddc700..fe484f33 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,128,3,[other],4,1
-5,4,128,1,[other],6,1
-5,5,105,1,[other],7,1
-5,6,105,1,[other],0,1
-5,7,105,1,[other],0,1
-5,8,105,1,[other],0,1
-5,9,105,1,7,2,[other]
-5,10,113,[mask],7,0,1
-5,11,113,1,[other],2,1
-5,12,113,[mask],[other],1,1
-5,13,113,1,[other],2,1
-5,14,113,[mask],[other],1,1
-5,15,113,1,[other],2,1
-5,16,113,1,[other],1,1
-5,17,113,1,1,[mask],1
-5,18,108,1,9,1,1
-5,19,103,1,0,7,1
-5,20,108,1,[other],7,1
-5,21,108,8,[other],7,1
-5,22,108,5,[other],2,1
-5,23,113,[mask],[other],7,1
-5,24,105,8,[other],2,[other]
-5,25,113,8,[other],8,1
-5,26,113,1,[other],[mask],1
-5,27,113,8,[other],2,1
-5,28,113,1,[other],8,1
-5,29,113,1,[other],[mask],1
-5,30,113,8,[other],2,1
-5,31,113,1,[other],2,1
-5,32,113,1,[other],7,1
+5,3,111,3,2,2,[other]
+5,4,111,3,2,8,[other]
+5,5,111,1,2,7,[other]
+5,6,111,1,9,7,[other]
+5,7,111,1,9,7,[other]
+5,8,111,1,9,7,[other]
+5,9,102,1,9,7,[other]
+5,10,129,1,9,[unknown],[other]
+5,11,102,1,9,[unknown],[other]
+5,12,102,6,9,[unknown],[other]
+5,13,104,6,9,[unknown],[other]
+5,14,104,6,9,[unknown],[other]
+5,15,104,6,9,7,[other]
+5,16,104,6,9,7,[other]
+5,17,104,9,9,7,[other]
+5,18,104,9,9,7,[other]
+5,19,104,6,9,7,[other]
+5,20,104,6,9,7,[other]
+5,21,104,6,9,7,[other]
+5,22,104,6,9,7,[other]
+5,23,104,6,9,7,[other]
+5,24,104,6,9,7,[other]
+5,25,104,6,9,7,[other]
+5,26,104,6,9,7,[other]
+5,27,104,6,9,7,[other]
+5,28,104,6,9,7,[other]
+5,29,104,6,9,7,[other]
+5,30,104,6,9,7,[other]
+5,31,104,6,9,7,[other]
+5,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index 62b206ad..22da0edc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,122,2,[other],8,[unknown]
-6,5,113,[mask],[other],8,[other]
-6,6,113,1,[other],[unknown],1
-6,7,113,[mask],[other],2,1
-6,8,113,1,[other],2,1
-6,9,113,3,[other],1,1
-6,10,121,1,[other],[mask],1
-6,11,108,0,7,1,1
-6,12,113,1,1,[mask],1
-6,13,108,8,9,1,[other]
-6,14,113,0,1,[mask],1
-6,15,108,8,[other],1,[other]
-6,16,113,0,[other],7,[other]
-6,17,105,8,[other],2,[other]
-6,18,113,[mask],[other],8,[other]
-6,19,113,1,[other],[mask],1
-6,20,113,1,[other],1,1
-6,21,113,1,[other],[mask],1
-6,22,113,1,[other],[mask],1
-6,23,124,1,[other],2,1
-6,24,113,1,[other],7,[unknown]
-6,25,105,8,[other],1,1
-6,26,113,1,[other],2,1
-6,27,113,8,[other],1,1
-6,28,113,1,[other],2,1
-6,29,113,8,[other],1,1
-6,30,113,1,[other],2,1
-6,31,113,[mask],9,1,1
-6,32,113,1,[other],1,1
-6,33,113,1,9,1,1
-7,3,105,0,[other],8,[mask]
-7,4,105,0,[other],1,[other]
-7,5,105,0,9,6,[unknown]
-7,6,105,0,[other],1,[unknown]
-7,7,113,[other],[other],1,[other]
-7,8,113,0,[other],1,1
-7,9,113,0,[other],1,2
-7,10,113,1,[other],[mask],1
-7,11,113,3,[other],1,1
-7,12,113,1,[other],[mask],1
-7,13,108,3,[other],1,1
-7,14,113,1,[other],[mask],1
-7,15,124,3,9,2,1
-7,16,113,0,8,0,1
-7,17,105,1,[other],[mask],[other]
-7,18,108,0,7,0,[other]
-7,19,113,5,[other],7,[other]
-7,20,105,8,[other],[mask],[other]
-7,21,113,[mask],[other],8,[other]
-7,22,113,1,[other],[mask],1
-7,23,113,1,[other],8,1
-7,24,113,1,[other],8,1
-7,25,113,1,[other],8,1
-7,26,113,1,[other],[mask],1
-7,27,105,8,[other],2,1
-7,28,113,1,8,2,1
-7,29,113,1,[other],1,1
-7,30,113,8,[other],1,1
-7,31,113,1,[other],2,1
-7,32,113,3,[other],1,1
+6,4,111,9,2,7,[other]
+6,5,[unknown],9,2,7,[other]
+6,6,111,9,9,7,[other]
+6,7,111,1,9,7,[other]
+6,8,111,1,9,7,[other]
+6,9,102,1,9,7,[other]
+6,10,129,1,9,[unknown],[other]
+6,11,129,1,9,[unknown],[other]
+6,12,129,1,9,[unknown],[other]
+6,13,104,1,9,[unknown],[other]
+6,14,104,6,9,[unknown],[other]
+6,15,104,6,9,[unknown],[other]
+6,16,104,6,9,[unknown],[other]
+6,17,104,6,9,7,[other]
+6,18,104,6,9,7,[other]
+6,19,104,6,9,7,[other]
+6,20,104,6,9,7,[other]
+6,21,104,6,9,7,[other]
+6,22,104,6,9,7,[other]
+6,23,104,6,9,7,[other]
+6,24,104,6,9,7,[other]
+6,25,104,6,9,7,[other]
+6,26,104,6,9,7,[other]
+6,27,104,6,9,7,[other]
+6,28,104,6,9,7,[other]
+6,29,104,6,9,7,[other]
+6,30,104,6,9,7,[other]
+6,31,104,6,9,7,[other]
+6,32,104,6,9,7,[other]
+6,33,104,6,9,7,[other]
+7,3,117,2,9,7,[other]
+7,4,129,6,9,7,[other]
+7,5,104,9,9,7,[other]
+7,6,129,9,9,7,[other]
+7,7,129,9,9,7,[other]
+7,8,129,9,9,7,[other]
+7,9,129,6,9,2,[other]
+7,10,104,9,9,7,[other]
+7,11,112,9,9,7,[other]
+7,12,112,9,9,7,[other]
+7,13,112,[mask],9,7,[other]
+7,14,112,[mask],9,8,[other]
+7,15,112,[mask],9,7,[other]
+7,16,112,[mask],9,7,[other]
+7,17,112,9,9,7,[other]
+7,18,112,[mask],9,7,[other]
+7,19,112,[mask],9,7,[other]
+7,20,112,[mask],9,7,[other]
+7,21,129,1,9,7,[other]
+7,22,129,1,9,7,[other]
+7,23,129,9,9,7,[other]
+7,24,112,9,9,7,[other]
+7,25,112,9,9,7,[other]
+7,26,112,9,9,7,[other]
+7,27,112,9,9,7,[other]
+7,28,112,9,9,7,[other]
+7,29,112,9,9,8,[other]
+7,30,112,9,9,7,[other]
+7,31,112,[mask],9,7,[unknown]
+7,32,112,[mask],9,8,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index f7a88c52..caac944b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,101,1,[other],8,2
-8,5,104,1,[other],[unknown],2
-8,6,104,3,[other],[other],[other]
-8,7,103,5,[other],0,1
-8,8,128,1,[other],0,[other]
-8,9,105,3,[other],6,[other]
-8,10,128,1,7,[mask],[other]
-8,11,120,[mask],7,[unknown],[other]
-8,12,113,1,[other],[mask],1
-8,13,105,1,[other],2,[other]
-8,14,113,1,7,2,1
-8,15,113,1,[other],1,1
-8,16,113,1,[other],[mask],1
-8,17,113,1,[other],2,1
-8,18,113,1,[other],1,1
-8,19,113,1,[other],1,1
-8,20,105,0,[other],1,1
-8,21,105,1,8,9,1
-8,22,103,0,9,1,1
-8,23,113,0,8,0,1
-8,24,113,0,[other],1,1
-8,25,113,0,[other],1,1
-8,26,113,0,[other],1,1
-8,27,113,0,[other],8,1
-8,28,113,1,[other],8,1
-8,29,113,1,[other],8,1
-8,30,113,1,[other],0,1
-8,31,113,1,[other],0,1
-8,32,108,1,[other],1,1
-8,33,113,1,[other],7,1
+8,4,111,9,9,7,[other]
+8,5,111,9,9,7,[other]
+8,6,111,9,9,7,[other]
+8,7,111,1,9,7,[other]
+8,8,111,1,9,7,[other]
+8,9,111,1,9,7,[other]
+8,10,102,1,9,7,[other]
+8,11,129,1,9,[unknown],[other]
+8,12,102,1,9,[unknown],[other]
+8,13,104,6,9,[unknown],[other]
+8,14,104,1,9,[unknown],[other]
+8,15,104,6,9,[unknown],[other]
+8,16,104,6,9,[unknown],[other]
+8,17,104,6,9,7,[other]
+8,18,104,6,9,7,[other]
+8,19,104,6,9,7,[other]
+8,20,104,6,9,7,[other]
+8,21,104,6,9,7,[other]
+8,22,104,6,9,7,[other]
+8,23,104,6,9,7,[other]
+8,24,104,6,9,7,[other]
+8,25,104,6,9,7,[other]
+8,26,104,6,9,7,[other]
+8,27,104,6,9,7,[other]
+8,28,104,6,9,7,[other]
+8,29,104,6,9,7,[other]
+8,30,104,6,9,7,[other]
+8,31,104,6,9,7,[other]
+8,32,104,6,9,7,[other]
+8,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index 7f5bf564..3f1d5fac 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,108,5,[other],8,2
-9,4,122,5,[other],7,1
-9,5,105,5,[other],7,[other]
-9,6,122,5,7,7,[other]
-9,7,122,5,[other],7,[other]
-9,8,122,5,[other],7,1
-9,9,122,1,[other],2,1
-9,10,113,[mask],7,8,1
-9,11,113,1,[other],2,1
-9,12,113,[mask],7,2,1
-9,13,113,1,[other],2,1
-9,14,113,[mask],[other],2,1
-9,15,113,[mask],1,2,1
-9,16,113,1,[other],7,1
-9,17,108,8,1,7,1
-9,18,108,1,[other],7,1
-9,19,108,1,[other],7,1
-9,20,108,1,[other],7,1
-9,21,108,8,[other],7,1
-9,22,122,1,[other],2,1
-9,23,113,1,[other],2,1
-9,24,113,8,[other],8,1
-9,25,113,1,[other],8,1
-9,26,113,1,[other],0,1
-9,27,113,8,[other],2,1
-9,28,113,1,[other],2,1
-9,29,113,8,[other],1,1
-9,30,113,1,1,2,1
-9,31,113,[mask],[other],1,1
-9,32,113,1,1,2,1
+9,3,113,3,2,[other],[other]
+9,4,119,9,2,7,[mask]
+9,5,129,9,9,[other],[other]
+9,6,119,9,9,7,[mask]
+9,7,104,9,9,7,[other]
+9,8,119,9,9,7,[other]
+9,9,129,9,9,7,[other]
+9,10,112,9,9,7,[other]
+9,11,112,9,9,7,[unknown]
+9,12,112,[mask],9,7,[other]
+9,13,112,[mask],9,7,[other]
+9,14,112,[mask],9,7,[other]
+9,15,112,[mask],9,7,[other]
+9,16,112,[mask],9,7,[other]
+9,17,112,[mask],9,8,[other]
+9,18,112,[mask],9,7,[other]
+9,19,129,1,9,7,[other]
+9,20,129,1,9,7,[other]
+9,21,129,1,9,7,[other]
+9,22,104,1,9,7,[other]
+9,23,104,6,9,7,[other]
+9,24,104,6,9,7,[other]
+9,25,104,6,9,7,[other]
+9,26,104,6,9,7,[other]
+9,27,104,6,9,7,[other]
+9,28,104,6,9,7,[other]
+9,29,104,6,9,7,[other]
+9,30,104,6,9,7,[other]
+9,31,104,6,9,7,[other]
+9,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
new file mode 100644
index 00000000..91865281
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
@@ -0,0 +1,5 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.03451868,0.031921398,0.029532596,0.027744867,0.02760109,0.030242559,0.033749565,0.027132228,0.033893008,0.030211333,0.027323615,0.0273727,0.027940668,0.02913617,0.03199635,0.030564532,0.02930324,0.031335846,0.02836081,0.035665363,0.032468908,0.031003729,0.033122327,0.030269567,0.03076924,0.030623542,0.02832116,0.03149027,0.02777426,0.029747033,0.030456595,0.027495332,0.030911474
+0.027516257,0.028654557,0.032131042,0.031892225,0.024753015,0.034306064,0.031358637,0.02987722,0.03253693,0.030763064,0.027757306,0.029624237,0.029635688,0.028321382,0.029743204,0.027194707,0.02963858,0.02781912,0.030843636,0.029608745,0.027764315,0.02745214,0.029405791,0.029906224,0.033089224,0.034630504,0.02899261,0.03382853,0.038423352,0.030570587,0.031674568,0.03056725,0.029719174
+0.030222073,0.031830907,0.028284168,0.028407857,0.027115861,0.03235872,0.029203441,0.02976575,0.026956126,0.027495082,0.03189054,0.028712502,0.03204519,0.030992731,0.03524834,0.032565366,0.027216628,0.034662586,0.02938502,0.033734377,0.028660439,0.033143435,0.03173607,0.03174802,0.031067362,0.03277405,0.029025575,0.031590212,0.025999047,0.02817474,0.026592813,0.028226981,0.033168018
+0.025530642,0.026080305,0.02652737,0.03402691,0.027508548,0.038157314,0.027138228,0.031741947,0.03190737,0.030657675,0.026315706,0.030096767,0.028837977,0.03074533,0.031430572,0.033588294,0.028155396,0.029817235,0.024138339,0.028173123,0.027259734,0.032578666,0.034752455,0.034232706,0.03252036,0.0357678,0.029168166,0.032422952,0.029197408,0.030990023,0.025777996,0.031730004,0.0330266
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
new file mode 100644
index 00000000..a3d9edc1
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
@@ -0,0 +1,5 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.024938403,0.025853546,0.030925326,0.035836563,0.027472962,0.033917453,0.031773735,0.033111185,0.030894933,0.03310667,0.03077764,0.031968474,0.027649606,0.029903907,0.03066213,0.030050345,0.028726462,0.024805613,0.029500253,0.026645845,0.025433421,0.030492553,0.029441224,0.033256244,0.030633831,0.03083032,0.029116308,0.03357199,0.035229903,0.031910744,0.02843861,0.03163502,0.031488776
+0.027546028,0.024828414,0.029011007,0.033997692,0.028309349,0.034571003,0.028885573,0.030300727,0.033193853,0.03057627,0.030879539,0.030720139,0.028182643,0.03336716,0.031966623,0.029043915,0.030032871,0.026891707,0.028006839,0.028216656,0.030012557,0.03335293,0.03443931,0.034874037,0.031277187,0.02787145,0.027704827,0.033389967,0.027017491,0.03146076,0.027681751,0.030754598,0.031635087
+0.025267702,0.02593611,0.026375175,0.033555325,0.02729746,0.03769069,0.028252527,0.03147091,0.031586744,0.02977779,0.026451198,0.029979987,0.028936028,0.031154884,0.0320948,0.03343466,0.027493577,0.029236242,0.024690645,0.028272472,0.026758416,0.031592447,0.03592271,0.03407506,0.03227401,0.035707034,0.030419732,0.03240453,0.030306093,0.030438218,0.026424794,0.031391647,0.033330467
+0.027293371,0.028949117,0.028703904,0.02919847,0.030929094,0.032399643,0.029274687,0.034145895,0.02837577,0.029229118,0.030326605,0.03315343,0.032393903,0.032588884,0.030739764,0.030741734,0.033645798,0.030104026,0.032112077,0.03169782,0.029715285,0.030085078,0.033176463,0.032097504,0.027618587,0.02773887,0.030057417,0.028018357,0.025789501,0.03361293,0.027174367,0.03010951,0.02880308
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
new file mode 100644
index 00000000..e4570072
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
@@ -0,0 +1,4 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.033277225,0.030706238,0.03168895,0.033311088,0.029628102,0.026783517,0.033642292,0.028419774,0.03004662,0.030877694,0.030766364,0.02872636,0.028279921,0.025215175,0.032209065,0.028634787,0.031974144,0.031396896,0.027670972,0.030388262,0.03137201,0.0331837,0.024129286,0.02784322,0.03336912,0.025571778,0.033213194,0.032749813,0.033496857,0.029778576,0.03258353,0.029908357,0.029157136
+0.025729796,0.029602438,0.029645566,0.029040024,0.031028586,0.030800173,0.033909444,0.031098856,0.028809031,0.027932486,0.029731028,0.031625725,0.033525992,0.029835034,0.029137556,0.028564114,0.03216789,0.027496634,0.034953404,0.028403202,0.026970934,0.027889818,0.031215824,0.029021468,0.030363686,0.03254443,0.03467346,0.028671235,0.034229062,0.029859748,0.031822145,0.030721655,0.028979547
+0.028761078,0.027854176,0.03048727,0.034092404,0.029769653,0.030306583,0.03338378,0.032568734,0.0329339,0.03081607,0.026996588,0.028901242,0.028271142,0.026599998,0.027621023,0.02927734,0.034487005,0.032435104,0.022172494,0.029216152,0.028780058,0.030934772,0.025088819,0.030961733,0.032769565,0.028614981,0.03369568,0.030542089,0.03556081,0.033930458,0.030258873,0.03194647,0.029963972
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
new file mode 100644
index 00000000..fc1bbb92
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
@@ -0,0 +1,7 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.033417575,0.030952973,0.031909764,0.03328892,0.029420735,0.02633423,0.03413355,0.028891983,0.02999831,0.030924743,0.030936386,0.02908393,0.028261695,0.025051711,0.032034438,0.028418737,0.032599755,0.03089184,0.027745165,0.0306719,0.031189425,0.032526094,0.023779722,0.027861075,0.03290723,0.02475645,0.03326867,0.03255766,0.03400532,0.030586168,0.032822493,0.030047486,0.028723868
+0.034453217,0.0305892,0.029416787,0.029322578,0.028612511,0.031616293,0.028566025,0.026076926,0.03032813,0.02860342,0.032680735,0.028623346,0.028884213,0.03250039,0.034757424,0.029118933,0.027070051,0.031802647,0.031775847,0.03249104,0.03419997,0.0347499,0.034769706,0.03168993,0.032166995,0.029576255,0.02648399,0.032632153,0.022518143,0.027031703,0.028449764,0.027058687,0.031383082
+0.03126686,0.028659452,0.026583998,0.02960171,0.029598372,0.0348543,0.028967595,0.027243897,0.03295455,0.028712928,0.027025098,0.02840309,0.029796531,0.030191114,0.032229044,0.030590648,0.030458689,0.030780101,0.028688198,0.030927148,0.03236309,0.035447787,0.03583643,0.031489626,0.034140706,0.032815706,0.028919876,0.032643765,0.022453642,0.028814038,0.028506618,0.028614122,0.030421335
+0.024738476,0.030802721,0.029338533,0.030698724,0.026044017,0.036583103,0.02847107,0.0297974,0.030350022,0.031745814,0.027877493,0.030870043,0.031255092,0.029293576,0.030610705,0.032736078,0.027911829,0.030885635,0.028708616,0.030616442,0.026017355,0.031379726,0.033175353,0.032386277,0.030061183,0.03907589,0.027840383,0.030412367,0.03361048,0.028390406,0.025730342,0.030818133,0.031766724
+0.027777517,0.028804556,0.031672627,0.030206317,0.025076197,0.034134787,0.032481443,0.029717883,0.03154765,0.028994586,0.027599279,0.028896887,0.030154243,0.028881382,0.030242773,0.027787082,0.02839826,0.0288448,0.031094935,0.029876556,0.027011143,0.026954878,0.030751621,0.029737474,0.032815382,0.035880763,0.031499054,0.03416633,0.03777237,0.028893098,0.032805875,0.02953868,0.029983671
+0.027476907,0.029252572,0.029974613,0.03427469,0.03160824,0.029946703,0.026536744,0.03142502,0.029297996,0.033668708,0.030066026,0.031427413,0.031061247,0.026917415,0.030150387,0.033821,0.0326446,0.032380875,0.025550779,0.02765895,0.02903609,0.035553962,0.025652928,0.029973598,0.030050268,0.028396083,0.03117175,0.03078121,0.032512754,0.03008112,0.02760991,0.033881046,0.03015839
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
new file mode 100644
index 00000000..66ac322a
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
@@ -0,0 +1,4 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.02787407,0.027104642,0.030358018,0.034553517,0.028322224,0.031050757,0.034033015,0.031545833,0.033861537,0.030875284,0.027181892,0.028938169,0.027790133,0.027383799,0.028615393,0.028780295,0.03324079,0.030557051,0.022777924,0.029385028,0.028478311,0.030502185,0.026848188,0.03165467,0.032651097,0.028698592,0.032736804,0.031015739,0.037046205,0.033572696,0.030005831,0.03182181,0.030738411
+0.02670878,0.025883488,0.02744727,0.03386368,0.029877376,0.032459594,0.030426193,0.03247706,0.02878972,0.027941974,0.027739787,0.030217001,0.030183967,0.029890094,0.032915857,0.03072404,0.030120345,0.030178469,0.027195016,0.02812506,0.028859448,0.029241446,0.031200247,0.029883275,0.0328588,0.029916031,0.035874527,0.030879542,0.033642642,0.031450473,0.030106302,0.03142932,0.03149319
+0.032150064,0.028221862,0.028330833,0.02810167,0.027819155,0.03160524,0.027474659,0.027735386,0.033368625,0.028838621,0.028449835,0.027917936,0.032627217,0.031592034,0.03244308,0.028486887,0.03201186,0.028425962,0.033350077,0.031887256,0.03447993,0.030445999,0.032843858,0.028395893,0.032929353,0.029940993,0.026082069,0.032996614,0.026771395,0.031200353,0.031379223,0.030597694,0.031098424
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
new file mode 100644
index 00000000..b6233023
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
@@ -0,0 +1,8 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.027693711,0.029121894,0.03263769,0.030477673,0.025150716,0.03287544,0.032045007,0.030850839,0.031303454,0.02993946,0.027595572,0.02886764,0.031017965,0.027735751,0.02935163,0.027287817,0.030652735,0.029687006,0.030908603,0.030391091,0.027535228,0.026636822,0.027238684,0.028523553,0.03267429,0.03431938,0.03061821,0.033247806,0.040079303,0.030726168,0.03266249,0.03049625,0.02965006
+0.025466071,0.029875752,0.029835982,0.029151956,0.028186664,0.03173176,0.03316926,0.030232249,0.031046985,0.02931059,0.029201102,0.03172116,0.033616595,0.029696383,0.029422067,0.027533485,0.034060102,0.026428862,0.035074487,0.030573001,0.027960634,0.02801773,0.03157427,0.029605145,0.0301314,0.03144874,0.030063352,0.028544225,0.03480755,0.032065663,0.030100726,0.031310387,0.029035721
+0.026478115,0.027542712,0.025084658,0.02916108,0.0296517,0.030709622,0.029443054,0.031282466,0.031903658,0.029175015,0.02586855,0.028222973,0.03547591,0.02836005,0.03144234,0.03363753,0.03476335,0.02754813,0.028392714,0.029871454,0.028496739,0.030743234,0.02964027,0.027810814,0.032352984,0.031880755,0.030597484,0.03063106,0.032444194,0.03375546,0.029653978,0.03467788,0.033299997
+0.026364712,0.029757837,0.032045525,0.031597987,0.030919015,0.031508707,0.031495456,0.03135381,0.03211043,0.03435329,0.03110224,0.032581497,0.029810883,0.029403877,0.027003385,0.030518897,0.03186091,0.026397405,0.03112966,0.027613735,0.026846573,0.032183122,0.029105142,0.032442607,0.028408209,0.031621095,0.028020909,0.03026787,0.031812698,0.030703535,0.02858388,0.031764213,0.029310849
+0.03315158,0.030279914,0.029608734,0.029177485,0.027309386,0.032146376,0.0286247,0.026008293,0.032145012,0.029281488,0.03193118,0.028416706,0.029567713,0.032534126,0.033859003,0.02824213,0.02870284,0.031056287,0.0315024,0.033499297,0.03437484,0.034205765,0.034372948,0.031886913,0.032041393,0.029588355,0.024660248,0.03200928,0.023830036,0.028581526,0.027754126,0.027885024,0.03176486
+0.025848068,0.028431289,0.029892355,0.028693581,0.030470718,0.029235307,0.030940453,0.03242136,0.029718446,0.027495537,0.029365137,0.02925045,0.03549993,0.030839637,0.028326172,0.029380446,0.034169305,0.03261548,0.028191667,0.0298291,0.028319623,0.027590321,0.02789852,0.028912824,0.029687619,0.03054694,0.033803966,0.027655903,0.037782732,0.031797566,0.030679887,0.03306369,0.031645942
+0.024057524,0.02544963,0.025607307,0.029881254,0.029157504,0.031960018,0.030122014,0.032780554,0.031288777,0.028194267,0.027988294,0.028752653,0.036150485,0.031077877,0.031034835,0.031049933,0.034480184,0.02570712,0.030488791,0.028481508,0.027200652,0.029486086,0.029952202,0.029214997,0.032385252,0.032020196,0.029338684,0.03019485,0.03324283,0.034965817,0.028807197,0.034836706,0.034644097
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
new file mode 100644
index 00000000..c4d448a9
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
@@ -0,0 +1,5 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.02861608,0.02783555,0.031125112,0.034070816,0.029551925,0.02920194,0.03443248,0.032701004,0.03236621,0.031025173,0.027565764,0.029052038,0.028101817,0.02655173,0.028142825,0.029290026,0.03396453,0.031744126,0.022819739,0.029417034,0.028412648,0.03002513,0.02473677,0.030448401,0.031774003,0.027520519,0.034440182,0.030517727,0.037965015,0.03373207,0.030928032,0.03187111,0.030052474
+0.029571708,0.027468769,0.028687142,0.030722428,0.030333482,0.028803239,0.034810323,0.032661207,0.028865434,0.028517043,0.03039022,0.029620783,0.030924972,0.029667925,0.033478405,0.02773338,0.03547341,0.02907416,0.032520335,0.033110958,0.03192609,0.030285636,0.027565196,0.028671881,0.031343654,0.024217593,0.030943919,0.029093694,0.028387735,0.03578851,0.029546076,0.029263297,0.0305314
+0.03309813,0.030376514,0.031082107,0.032114696,0.0303959,0.026847737,0.033983357,0.028783653,0.029035237,0.02931486,0.030565187,0.0282617,0.028926874,0.025849348,0.032453902,0.029420223,0.030807314,0.032321904,0.027733354,0.029939143,0.030494139,0.032511473,0.024984352,0.027675122,0.033265285,0.026638621,0.03590076,0.032994192,0.033117052,0.028538441,0.033555407,0.02946036,0.029553706
+0.032728393,0.028721752,0.02882793,0.027880307,0.028566426,0.030735191,0.027027326,0.027918302,0.033026785,0.029266033,0.028729456,0.028066441,0.03297658,0.031279273,0.031866837,0.028559575,0.032620937,0.02890799,0.033787943,0.03182001,0.03511367,0.030451855,0.03162444,0.027786268,0.03249884,0.029433765,0.025993556,0.032565445,0.026775395,0.031265218,0.031698216,0.030825824,0.030654097
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
new file mode 100644
index 00000000..4e8c02ca
--- /dev/null
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
@@ -0,0 +1,4 @@
+[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
+0.03475592,0.03205031,0.029370865,0.02778089,0.027614197,0.030346887,0.03223454,0.026713328,0.03431964,0.030279862,0.027550474,0.027702024,0.028271765,0.029716752,0.031870633,0.03032701,0.029443298,0.03097386,0.02868564,0.035414,0.033398263,0.030973615,0.033969566,0.030356236,0.030772602,0.030513762,0.027257016,0.031204006,0.027268384,0.029877514,0.030222507,0.02795559,0.03080913
+0.027402356,0.027945139,0.02530322,0.029322542,0.029918933,0.029952679,0.028517315,0.03166655,0.031823326,0.02963542,0.026160058,0.028808074,0.03524668,0.028342647,0.03120509,0.0340563,0.03451171,0.02670396,0.02878434,0.029324375,0.028796544,0.029988907,0.029430298,0.027544383,0.031944152,0.031181917,0.030243525,0.031055551,0.032528445,0.03417648,0.030487247,0.03531119,0.032680634
+0.032010403,0.03287599,0.029452827,0.027987542,0.027845826,0.03130848,0.027573993,0.030086143,0.02799824,0.028927118,0.031524617,0.02894558,0.032493584,0.030415898,0.03303464,0.03172704,0.029664379,0.035562575,0.029617772,0.034429386,0.031013394,0.032945413,0.029464815,0.030695107,0.03063894,0.031342026,0.026874157,0.030781394,0.025375465,0.029842958,0.026907688,0.028956134,0.031680416

From e504c1443bf3100027d8747d3e452d892386cdbc Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 09:57:13 +0200
Subject: [PATCH 37/81] Make special_token_ids metadata field mandatory

---
 .../config/hyperparameter_search_config.py    | 10 ++-
 src/sequifier/config/infer_config.py          |  6 ++
 src/sequifier/config/train_config.py          | 12 +++-
 src/sequifier/infer.py                        |  5 ++
 src/sequifier/preprocess.py                   |  9 +++
 src/sequifier/special_tokens.py               | 25 +++++++
 tests/unit/test_infer.py                      | 52 +++++++++++++-
 tests/unit/test_preprocess.py                 |  1 +
 tests/unit/test_train.py                      | 71 +++++++++++++++++++
 9 files changed, 185 insertions(+), 6 deletions(-)

diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index cec47b2f..cb3841f8 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -18,6 +18,7 @@
     TrainModel,
 )
 from sequifier.helpers import normalize_path, try_catch_excess_keys
+from sequifier.special_tokens import validate_special_token_ids
 
 
 class FloatDistribution(BaseModel):
@@ -119,8 +120,8 @@ def sample_trial(self, trial: Any) -> BERTSpecModel:
 
         replacement_distribution = self.replacement_distribution[
             replacement_distribution_index
-        ].model_copy(deep=True)
-        span_masking = self.span_masking[span_masking_index].model_copy(deep=True)
+        ].model_copy(deep=True)  # type: ignore
+        span_masking = self.span_masking[span_masking_index].model_copy(deep=True)  # type: ignore
 
         logger.info(
             f"{masking_probability = } - {replacement_distribution = } - {span_masking = }"
@@ -164,6 +165,11 @@ def load_hyperparameter_search_config(
         ) as f:
             metadata_config = json.loads(f.read())
 
+        validate_special_token_ids(
+            metadata_config.get("special_token_ids"),
+            source=f"metadata config '{metadata_config_path}'",
+        )
+
         config_values["column_types"] = config_values.get(
             "column_types", [metadata_config["column_types"]]
         )
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 62277c76..59d1ef75 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -15,6 +15,7 @@
 )
 
 from sequifier.helpers import normalize_path, try_catch_excess_keys
+from sequifier.special_tokens import validate_special_token_ids
 
 
 @beartype
@@ -46,6 +47,11 @@ def load_inferer_config(
         ) as f:
             metadata_config = json.load(f)
 
+        validate_special_token_ids(
+            metadata_config.get("special_token_ids"),
+            source=f"metadata config '{metadata_config_path}'",
+        )
+
         config_values["column_types"] = config_values.get(
             "column_types", metadata_config["column_types"]
         )
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index d387e7cd..5aa6ae29 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -23,7 +23,7 @@
 import sequifier
 from sequifier.config.probabilities import ProbabilityDistribution
 from sequifier.helpers import normalize_path, try_catch_excess_keys
-from sequifier.special_tokens import SPECIAL_TOKEN_IDS
+from sequifier.special_tokens import SPECIAL_TOKEN_IDS, validate_special_token_ids
 
 AnyType = str | int | float
 
@@ -98,8 +98,9 @@ def load_train_config(
         )
 
         config_values["id_maps"] = metadata_config["id_maps"]
-        config_values["special_token_ids"] = metadata_config.get(
-            "special_token_ids", SPECIAL_TOKEN_IDS.ids_by_label
+        config_values["special_token_ids"] = validate_special_token_ids(
+            metadata_config.get("special_token_ids"),
+            source=f"metadata config '{metadata_config_path}'",
         )
 
     return try_catch_excess_keys(config_path, TrainModel, config_values)
@@ -575,6 +576,11 @@ class TrainModel(BaseModel):
     model_spec: ModelSpecModel
     training_spec: TrainingSpecModel
 
+    @field_validator("special_token_ids")
+    @classmethod
+    def validate_special_token_ids_match_runtime(cls, v):
+        return validate_special_token_ids(v, source="TrainModel")
+
     @model_validator(mode="after")
     def validate_bert_prediction_length_matches_seq_length(self):
         if (
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 41b491cb..57902f78 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -24,6 +24,7 @@
     unpack_preprocessed_pt_tuple,
     write_data,
 )
+from sequifier.special_tokens import validate_special_token_ids
 from sequifier.train import (
     infer_with_embedding_model,
     infer_with_generative_model,
@@ -66,6 +67,10 @@ def infer(args: Any, args_config: dict[str, Any]) -> None:
             normalize_path(config.metadata_config_path, config.project_root), "r"
         ) as f:
             metadata_config = json.loads(f.read())
+            validate_special_token_ids(
+                metadata_config.get("special_token_ids"),
+                source=f"metadata config '{config.metadata_config_path}'",
+            )
             id_maps = metadata_config["id_maps"]
             selected_columns_statistics = metadata_config["selected_columns_statistics"]
     else:
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 46747d4a..c67cb9eb 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -26,6 +26,7 @@
     SPECIAL_TOKEN_ID_VALUES,
     SPECIAL_TOKEN_IDS,
     SPECIAL_TOKEN_LABELS,
+    validate_special_token_ids,
 )
 
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
@@ -192,6 +193,10 @@ def __init__(
                 with open(metadata_path, "r") as f:
                     preexisting_metadata = json.load(f)
 
+                validate_special_token_ids(
+                    preexisting_metadata.get("special_token_ids"),
+                    source=f"metadata config '{self.metadata_config_path}'",
+                )
                 id_maps = preexisting_metadata["id_maps"]
                 selected_columns_statistics = preexisting_metadata[
                     "selected_columns_statistics"
@@ -282,6 +287,10 @@ def __init__(
                 with open(metadata_path, "r") as f:
                     preexisting_metadata = json.load(f)
 
+                validate_special_token_ids(
+                    preexisting_metadata.get("special_token_ids"),
+                    source=f"metadata config '{self.metadata_config_path}'",
+                )
                 id_maps = preexisting_metadata["id_maps"]
                 selected_columns_statistics = preexisting_metadata[
                     "selected_columns_statistics"
diff --git a/src/sequifier/special_tokens.py b/src/sequifier/special_tokens.py
index 9f1cd4ba..c92e23d8 100644
--- a/src/sequifier/special_tokens.py
+++ b/src/sequifier/special_tokens.py
@@ -1,4 +1,6 @@
+from collections.abc import Mapping
 from dataclasses import asdict, dataclass
+from typing import Any
 
 
 @dataclass(frozen=True)
@@ -27,3 +29,26 @@ def ids_by_label(self) -> dict[str, int]:
 SPECIAL_TOKEN_IDS = SpecialTokenIds()
 SPECIAL_TOKEN_LABELS = frozenset(SPECIAL_TOKEN_IDS.ids_by_label.keys())
 SPECIAL_TOKEN_ID_VALUES = frozenset(SPECIAL_TOKEN_IDS.labels_by_id.keys())
+
+
+def validate_special_token_ids(
+    special_token_ids: Mapping[str, Any] | None,
+    source: str = "metadata",
+) -> dict[str, int]:
+    """Validate persisted special token IDs against runtime constants."""
+    expected = SPECIAL_TOKEN_IDS.ids_by_label
+    assert special_token_ids is not None
+    try:
+        normalized = {str(label): int(id_) for label, id_ in special_token_ids.items()}
+    except (AttributeError, TypeError, ValueError) as exc:
+        raise ValueError(
+            f"{source} special_token_ids must be a mapping from token labels to integer IDs."
+        ) from exc
+
+    if normalized != expected:
+        raise ValueError(
+            f"{source} special_token_ids must match the sequifier runtime constants. "
+            f"Expected {expected}, found {dict(special_token_ids)}."
+        )
+
+    return normalized
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 5416dc41..9bfdaa63 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -1,11 +1,13 @@
+import json
 from unittest.mock import patch
 
 import numpy as np
 import polars as pl
 import pytest
 import torch
+import yaml
 
-from sequifier.config.infer_config import InfererModel
+from sequifier.config.infer_config import InfererModel, load_inferer_config
 from sequifier.infer import (
     Inferer,
     calculate_item_positions,
@@ -234,6 +236,54 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
         )
 
 
+def test_load_inferer_config_rejects_mismatched_metadata_special_token_ids(tmp_path):
+    data_path = tmp_path / "data.parquet"
+    data_path.touch()
+    metadata_path = tmp_path / "metadata.json"
+    config_path = tmp_path / "infer.yaml"
+
+    metadata_path.write_text(
+        json.dumps(
+            {
+                "split_paths": [str(data_path)],
+                "column_types": {"target_col": "int64"},
+                "id_maps": {"target_col": {"A": 3}},
+                "selected_columns_statistics": {},
+                "special_token_ids": {
+                    "[unknown]": 10,
+                    "[other]": 11,
+                    "[mask]": 12,
+                },
+            }
+        )
+    )
+    config_path.write_text(
+        yaml.safe_dump(
+            {
+                "project_root": str(tmp_path),
+                "metadata_config_path": metadata_path.name,
+                "model_path": "dummy.onnx",
+                "model_type": "generative",
+                "training_objective": "causal",
+                "data_path": str(data_path),
+                "input_columns": ["target_col"],
+                "categorical_columns": ["target_col"],
+                "real_columns": [],
+                "target_columns": ["target_col"],
+                "column_types": {"target_col": "int64"},
+                "target_column_types": {"target_col": "categorical"},
+                "seed": 42,
+                "device": "cpu",
+                "seq_length": 3,
+                "inference_batch_size": 2,
+            }
+        )
+    )
+
+    with pytest.raises(ValueError, match="special_token_ids must match"):
+        load_inferer_config(str(config_path), {}, skip_metadata=False)
+
+
 def test_infer_pure_passes_attention_valid_mask_to_onnx(mock_inferer):
     class Input:
         def __init__(self, name):
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 416d13d9..e18caaf4 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -194,6 +194,7 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
                 "selected_columns_statistics": {
                     "itemValue": {"mean": 60.0, "std": 10.0}
                 },
+                "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
             }
         )
     )
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index fb451533..50d7e106 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -1,7 +1,9 @@
 import copy
+import json
 
 import pytest
 import torch
+import yaml
 from pydantic import ValidationError
 
 from sequifier.config.probabilities import PoissonDistributionFloor
@@ -11,7 +13,9 @@
     ReplacementDistribution,
     TrainingSpecModel,
     TrainModel,
+    load_train_config,
 )
+from sequifier.special_tokens import SPECIAL_TOKEN_IDS
 from sequifier.train import TransformerModel
 
 
@@ -218,6 +222,73 @@ def test_train_model_requires_bert_prediction_length_to_equal_seq_length(model_c
         TrainModel(**config_values)
 
 
+def test_train_model_rejects_mismatched_special_token_ids(model_config):
+    config_values = model_config.model_dump()
+    config_values["special_token_ids"] = {
+        "[unknown]": 10,
+        "[other]": 11,
+        "[mask]": 12,
+    }
+
+    with pytest.raises(ValidationError, match="special_token_ids must match"):
+        TrainModel(**config_values)
+
+
+def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
+    tmp_path, model_config
+):
+    config_path = tmp_path / "train.yaml"
+    metadata_path = tmp_path / "metadata.json"
+    config_values = model_config.model_dump()
+    config_values["project_root"] = str(tmp_path)
+    config_values["metadata_config_path"] = metadata_path.name
+    config_path.write_text(yaml.safe_dump(config_values))
+    metadata_path.write_text(
+        json.dumps(
+            {
+                "split_paths": ["data/train.pt", "data/val.pt"],
+                "column_types": config_values["column_types"],
+                "n_classes": config_values["n_classes"],
+                "id_maps": config_values["id_maps"],
+                "special_token_ids": {
+                    "[unknown]": 10,
+                    "[other]": 11,
+                    "[mask]": 12,
+                },
+            }
+        )
+    )
+
+    with pytest.raises(ValueError, match="special_token_ids must match"):
+        load_train_config(str(config_path), {}, skip_metadata=False)
+
+
+def test_load_train_config_defaults_missing_metadata_special_token_ids(
+    tmp_path, model_config
+):
+    config_path = tmp_path / "train.yaml"
+    metadata_path = tmp_path / "metadata.json"
+    config_values = model_config.model_dump()
+    config_values["project_root"] = str(tmp_path)
+    config_values["metadata_config_path"] = metadata_path.name
+    config_path.write_text(yaml.safe_dump(config_values))
+    metadata_path.write_text(
+        json.dumps(
+            {
+                "split_paths": ["data/train.pt", "data/val.pt"],
+                "column_types": config_values["column_types"],
+                "n_classes": config_values["n_classes"],
+                "id_maps": config_values["id_maps"],
+                "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
+            }
+        )
+    )
+
+    config = load_train_config(str(config_path), {}, skip_metadata=False)
+
+    assert config.special_token_ids == SPECIAL_TOKEN_IDS.ids_by_label
+
+
 def test_forward_train_shapes(model, model_config):
     """Tests the output shapes of the forward_train method."""
     batch_size = model_config.training_spec.batch_size

From c3f03f0a4515b39265775a2bc9333d71002b8d1e Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 11:16:29 +0200
Subject: [PATCH 38/81] Remove tuple backward compatibility

---
 src/sequifier/helpers.py                      |  26 ----
 src/sequifier/infer.py                        |   5 +-
 .../io/sequifier_dataset_from_folder_pt.py    |  10 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  10 +-
 src/sequifier/preprocess.py                   |  11 +-
 tests/integration/conftest.py                 | 124 ++++++++++--------
 ...t_sequifier_dataset_from_folder_pt_lazy.py |   1 +
 7 files changed, 76 insertions(+), 111 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index fd3fdb90..5cb2db62 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -377,32 +377,6 @@ def generate_padding_masks(
     }
 
 
-@beartype
-def unpack_preprocessed_pt_tuple(
-    data: tuple[Any, ...],
-) -> tuple[Any, Any, Any, Any, Optional[Any]]:
-    """Unpacks both v1 four-item and v2 five-item preprocessed PT tuples."""
-    if len(data) == 4:
-        sequences_dict, sequence_ids, subsequence_ids, start_positions = data
-        return sequences_dict, sequence_ids, subsequence_ids, start_positions, None
-    if len(data) == 5:
-        (
-            sequences_dict,
-            sequence_ids,
-            subsequence_ids,
-            start_positions,
-            left_pad_lengths,
-        ) = data
-        return (
-            sequences_dict,
-            sequence_ids,
-            subsequence_ids,
-            start_positions,
-            left_pad_lengths,
-        )
-    raise ValueError(f"Expected a 4- or 5-item preprocessed PT tuple, got {len(data)}")
-
-
 @beartype
 def normalize_path(path: str, project_root: str) -> str:
     """Normalizes a path to be relative to a project path, then joins them.
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 57902f78..f4dbad3f 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -21,7 +21,6 @@
     normalize_path,
     numpy_to_pytorch,
     subset_to_input_columns,
-    unpack_preprocessed_pt_tuple,
     write_data,
 )
 from sequifier.special_tokens import validate_special_token_ids
@@ -491,7 +490,7 @@ def infer_embedding(
                 subsequence_ids_tensor,
                 start_positions_tensor,
                 left_pad_lengths_tensor,
-            ) = unpack_preprocessed_pt_tuple(data)
+            ) = data
             metadata = {}
             if left_pad_lengths_tensor is not None:
                 target_offset = 0 if config.training_objective == "causal" else 1
@@ -738,7 +737,7 @@ def infer_generative(
                 _,
                 start_positions_tensor,
                 left_pad_lengths_tensor,
-            ) = unpack_preprocessed_pt_tuple(data)
+            ) = data
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index f689ac73..988ca94d 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -9,11 +9,7 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import (
-    generate_padding_masks,
-    normalize_path,
-    unpack_preprocessed_pt_tuple,
-)
+from sequifier.helpers import generate_padding_masks, normalize_path
 from sequifier.io.batch import SequifierBatch
 
 
@@ -65,9 +61,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
                 _,
                 _,
                 left_pad_lengths_batch,
-            ) = unpack_preprocessed_pt_tuple(
-                torch.load(file_path, map_location="cpu", weights_only=False)
-            )
+            ) = torch.load(file_path, map_location="cpu", weights_only=False)
             for col in all_sequences.keys():
                 if col in sequences_batch:
                     all_sequences[col].append(sequences_batch[col])
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 37d7940f..dc60aa34 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -10,11 +10,7 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import (
-    generate_padding_masks,
-    normalize_path,
-    unpack_preprocessed_pt_tuple,
-)
+from sequifier.helpers import generate_padding_masks, normalize_path
 from sequifier.io.batch import SequifierBatch
 
 
@@ -233,9 +229,7 @@ def __iter__(
                 _,
                 _,
                 left_pad_lengths_batch,
-            ) = unpack_preprocessed_pt_tuple(
-                torch.load(file_path, map_location="cpu", weights_only=False)
-            )
+            ) = torch.load(file_path, map_location="cpu", weights_only=False)
 
             # Generate indices for the whole file
             indices = torch.arange(file_samples)
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index c67cb9eb..4b5afc75 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -16,12 +16,7 @@
 from loguru import logger
 
 from sequifier.config.preprocess_config import load_preprocessor_config
-from sequifier.helpers import (
-    PANDAS_TO_TORCH_TYPES,
-    read_data,
-    unpack_preprocessed_pt_tuple,
-    write_data,
-)
+from sequifier.helpers import PANDAS_TO_TORCH_TYPES, read_data, write_data
 from sequifier.special_tokens import (
     SPECIAL_TOKEN_ID_VALUES,
     SPECIAL_TOKEN_IDS,
@@ -882,8 +877,8 @@ def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> No
         for file_path in files:
             try:
                 if write_format == "pt":
-                    sequences_dict, _, _, _, _ = unpack_preprocessed_pt_tuple(
-                        torch.load(file_path, weights_only=False)
+                    sequences_dict, _, _, _, _ = torch.load(
+                        file_path, weights_only=False
                     )
                     if sequences_dict:
                         n_samples = sequences_dict[
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index 83208921..30588329 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -1,3 +1,4 @@
+import copy
 import os
 import shutil
 import subprocess
@@ -5,6 +6,7 @@
 
 import polars as pl
 import pytest
+import torch
 import yaml
 
 SELECTED_COLUMNS = {
@@ -318,77 +320,83 @@ def format_configs_locally(
 ):
     from sys import platform
 
-    if platform == "windows":
-        config_paths = [
-            preprocessing_config_path_cat,
-            preprocessing_config_path_cat_multitarget,
-            preprocessing_config_path_real,
-            preprocessing_config_path_multi_file,
-            preprocessing_config_path_interrupted,
-            training_config_path_cat,
-            training_config_path_cat_multitarget,
-            training_config_path_real,
-            training_config_path_cat_inf_size_1,
-            training_config_path_cat_inf_size_3,
-            training_config_path_cat_bert,
-            training_config_path_real_bert,
-            training_config_path_distributed,
-            training_config_path_distributed_lazy_parquet,
-            training_config_path_lazy,
-            training_config_path_resume_epoch,
-            training_config_path_resume_mid_epoch,
-            inference_config_path_cat,
-            inference_config_path_cat_multitarget,
-            training_config_path_cat_multitarget_eager,
-            inference_config_path_real,
-            inference_config_path_real_autoregression,
-            inference_config_path_categorical_autoregression,
-            inference_config_path_cat_inf_size_1,
-            inference_config_path_cat_inf_size_3,
-            inference_config_path_cat_bert,
-            inference_config_path_cat_bert_embedding,
-            inference_config_path_distributed,
-            inference_config_path_distributed_parquet,
-            inference_config_path_lazy,
-            hp_search_configs["grid"],
-            hp_search_configs["sample"],
-            hp_search_configs["bert"],
-            hp_search_configs["bayesian"],
-            hp_search_configs["custom-eval"],
-        ]
-        for config_path in config_paths:
+    config_paths = [
+        preprocessing_config_path_cat,
+        preprocessing_config_path_cat_multitarget,
+        preprocessing_config_path_real,
+        preprocessing_config_path_multi_file,
+        preprocessing_config_path_interrupted,
+        training_config_path_cat,
+        training_config_path_cat_multitarget,
+        training_config_path_real,
+        training_config_path_cat_inf_size_1,
+        training_config_path_cat_inf_size_3,
+        training_config_path_cat_bert,
+        training_config_path_real_bert,
+        training_config_path_distributed,
+        training_config_path_distributed_lazy_parquet,
+        training_config_path_lazy,
+        training_config_path_resume_epoch,
+        training_config_path_resume_mid_epoch,
+        inference_config_path_cat,
+        inference_config_path_cat_multitarget,
+        training_config_path_cat_multitarget_eager,
+        inference_config_path_real,
+        inference_config_path_real_autoregression,
+        inference_config_path_categorical_autoregression,
+        inference_config_path_cat_inf_size_1,
+        inference_config_path_cat_inf_size_3,
+        inference_config_path_cat_bert,
+        inference_config_path_cat_bert_embedding,
+        inference_config_path_distributed,
+        inference_config_path_distributed_parquet,
+        inference_config_path_lazy,
+        hp_search_configs["grid"],
+        hp_search_configs["sample"],
+        hp_search_configs["bert"],
+        hp_search_configs["bayesian"],
+        hp_search_configs["custom-eval"],
+    ]
+    original_configs = {}
+
+    cuda_available = torch.cuda.is_available()
+
+    for config_path in config_paths:
+        if str(platform).startswith("win") or cuda_available:
             with open(config_path, "r") as f:
                 config = yaml.safe_load(f)
 
+            original_config = copy.deepcopy(config)
+            original_configs[config_path] = original_config
+
             assert config is not None, config_path
 
-            config_formatted = {
-                attr: reformat_parameter(attr, param, "linux->local")
-                for attr, param in config.items()
-            }
+            if platform == "windows":
+                config = {
+                    attr: reformat_parameter(attr, param, "linux->local")
+                    for attr, param in config.items()
+                }
+
+            if cuda_available:
+                if "training_spec" in config:
+                    config["training_spec"]["device"] = "cuda"  # type: ignore
+                if "device" in config:
+                    config["device"] = "cuda"
 
             with open(config_path, "w") as f:
-                yaml.dump(
-                    config_formatted, f, default_flow_style=False, sort_keys=False
-                )
+                yaml.dump(config, f, default_flow_style=False, sort_keys=False)
 
-        yield
+    yield
 
+    if str(platform).startswith("win") or cuda_available:
         for config_path in config_paths:
-            with open(config_path, "r") as f:
-                config = yaml.safe_load(f)
-
-            config_formatted = {
-                attr: reformat_parameter(attr, param, "local->linux")
-                for attr, param in config.items()
-            }
-
             with open(config_path, "w") as f:
                 yaml.dump(
-                    config_formatted, f, default_flow_style=False, sort_keys=False
+                    original_configs[config_path],
+                    f,
+                    default_flow_style=False,
+                    sort_keys=False,
                 )
-    else:
-        yield
 
 
 @pytest.fixture(scope="session")
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index f3194651..2c697037 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -65,6 +65,7 @@ def side_effect(path, map_location, weights_only):
                 None,
                 None,
                 None,
+                None,
             )
 
         mock_load.side_effect = side_effect

From ecbb09745a145ca0441b625cf79d484d1043b62f Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 11:43:27 +0200
Subject: [PATCH 39/81] Create dummy metadata

---
 src/sequifier/train.py | 27 ++++++++++++++++++++-------
 1 file changed, 20 insertions(+), 7 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index c05be463..39795814 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -104,7 +104,9 @@ def cleanup():
 
 
 @beartype
-def create_dummy_data(config: TrainModel, local_rank: int) -> dict[str, Tensor]:
+def create_dummy_data_and_metadata(
+    config: TrainModel, local_rank: int
+) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
     dummy_data = {}
     for col in config.input_columns:
         dtype = torch.int64 if col in config.categorical_columns else torch.float32
@@ -114,7 +116,14 @@ def create_dummy_data(config: TrainModel, local_rank: int) -> dict[str, Tensor]:
             device=local_rank,
         )
 
-    return dummy_data
+    dummy_metadata = {
+        "attention_valid_mask": torch.ones(
+            (config.training_spec.batch_size, config.seq_length),
+            dtype=torch.bool,
+            device=local_rank,
+        )
+    }
+    return dummy_data, dummy_metadata
 
 
 @beartype
@@ -345,9 +354,11 @@ def train_worker(
                     model.layers[i] = torch.compile(model.layers[i])
 
         if config.training_spec.device.startswith("cuda"):
-            dummy_data = create_dummy_data(config, local_rank)
+            dummy_data, dummy_metadata = create_dummy_data_and_metadata(
+                config, local_rank
+            )
             with torch.no_grad():
-                _ = model(dummy_data, False)
+                _ = model(dummy_data, dummy_metadata, False)
 
             dist.barrier()
 
@@ -387,16 +398,18 @@ def train_worker(
         ddp_model = DDP(model, device_ids=device_ids, find_unused_parameters=False)
 
         if config.training_spec.device.startswith("cuda"):
-            dummy_data = create_dummy_data(config, local_rank)
+            dummy_data, dummy_metadata = create_dummy_data_and_metadata(
+                config, local_rank
+            )
 
             if config.training_spec.layer_autocast:
                 with torch.no_grad(), torch.autocast(
                     device_type="cuda", dtype=torch.bfloat16
                 ):
-                    _ = ddp_model(dummy_data, False)
+                    _ = ddp_model(dummy_data, dummy_metadata, False)
             else:
                 with torch.no_grad():
-                    _ = ddp_model(dummy_data, False)
+                    _ = ddp_model(dummy_data, dummy_metadata, False)
 
             dist.barrier()
         model.train_model(train_loader, valid_loader, ddp_model=ddp_model)

From 24192e56b95b979e8fa8df262e4c6f110c2d28f6 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 13:29:31 +0200
Subject: [PATCH 40/81] Introduce target_max_offset

---
 README.md                                     |  20 +-
 documentation/configs/preprocess.md           |   1 +
 documentation/consolidated-docs.md            |  19 +-
 .../config/hyperparameter_search_config.py    |  16 +-
 src/sequifier/config/infer_config.py          |  36 +++-
 src/sequifier/config/preprocess_config.py     |   3 +
 src/sequifier/config/train_config.py          |  46 ++++-
 src/sequifier/helpers.py                      | 122 +++++++++++-
 src/sequifier/infer.py                        |  51 +++--
 .../io/sequifier_dataset_from_file.py         |   7 +-
 .../sequifier_dataset_from_folder_parquet.py  |  32 +--
 ...uifier_dataset_from_folder_parquet_lazy.py |  32 +--
 .../io/sequifier_dataset_from_folder_pt.py    |  33 ++--
 .../sequifier_dataset_from_folder_pt_lazy.py  |  32 +--
 src/sequifier/make.py                         |   4 +
 src/sequifier/preprocess.py                   | 182 ++++++++++++------
 ...infer-test-categorical-autoregression.yaml |   1 +
 ...infer-test-categorical-bert-embedding.yaml |   1 +
 .../configs/infer-test-categorical-bert.yaml  |   1 +
 .../infer-test-categorical-embedding.yaml     |   1 +
 .../infer-test-categorical-inf-size-1.yaml    |   1 +
 ...test-categorical-inf-size-3-embedding.yaml |   1 +
 .../infer-test-categorical-inf-size-3.yaml    |   1 +
 .../infer-test-categorical-multitarget.yaml   |   1 +
 tests/configs/infer-test-categorical.yaml     |   1 +
 .../infer-test-distributed-parquet.yaml       |   1 +
 tests/configs/infer-test-distributed.yaml     |   1 +
 tests/configs/infer-test-lazy.yaml            |   1 +
 .../infer-test-real-autoregression.yaml       |   1 +
 tests/configs/infer-test-real.yaml            |   1 +
 .../configs/train-test-categorical-bert.yaml  |   2 +
 .../train-test-categorical-inf-size-1.yaml    |   2 +
 .../train-test-categorical-inf-size-3.yaml    |   2 +
 ...in-test-categorical-multitarget-eager.yaml |   2 +
 .../train-test-categorical-multitarget.yaml   |   2 +
 tests/configs/train-test-categorical.yaml     |   2 +
 .../train-test-distributed-lazy-parquet.yaml  |   2 +
 tests/configs/train-test-distributed.yaml     |   2 +
 tests/configs/train-test-lazy.yaml            |   2 +
 tests/configs/train-test-real-bert.yaml       |   2 +
 tests/configs/train-test-real.yaml            |   2 +
 tests/configs/train-test-resume-epoch.yaml    |   2 +
 .../configs/train-test-resume-mid-epoch.yaml  |   2 +
 tests/integration/conftest.py                 |   4 +-
 tests/integration/test_make.py                |   2 +
 tests/integration/test_onnx_export.py         |   4 +
 tests/integration/test_preprocessing.py       |  27 +--
 .../preprocess-manifest.json                  |  11 ++
 ...uifier_dataset_from_folder_parquet_lazy.py |   1 +
 ...t_sequifier_dataset_from_folder_pt_lazy.py |   4 +-
 tests/unit/test_helpers.py                    |  24 ++-
 .../unit/test_hyperparameter_search_config.py |  14 +-
 tests/unit/test_infer.py                      |   5 +
 tests/unit/test_preprocess.py                 |  43 ++++-
 tests/unit/test_train.py                      |   3 +
 55 files changed, 617 insertions(+), 201 deletions(-)
 create mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json

diff --git a/README.md b/README.md
index 85f64440..30617973 100644
--- a/README.md
+++ b/README.md
@@ -97,16 +97,16 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`, which takes this form:
-
-|sequenceId|subsequenceId|startItemPosition|columnName|[Subsequence Length]|[Subsequence Length - 1]|...|0|
-|----------|-------------|-----------------|----------|--------------------|------------------------| - |-|
-|0|0|0|column1|"high"|"high"|...|"low"|
-|0|0|0|column2|12.3|10.2|...|14.9|
-|...|...|...|...|...|...|...|...|
-|1|0|15|column1|"medium"|"high"|...|"medium"|
-|1|0|15|column2|20.6|18.5|...|21.6|
-|...|...|...|...|...|...|...|...|
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `seq_length + target_max_offset` (`target_max_offset` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
+
+|sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
+|----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
+|0|0|0|0|column1|"high"|"high"|...|"low"|
+|0|0|0|0|column2|12.3|10.2|...|14.9|
+|...|...|...|...|...|...|...|...|...|
+|1|0|15|0|column1|"medium"|"high"|...|"medium"|
+|1|0|15|0|column2|20.6|18.5|...|21.6|
+|...|...|...|...|...|...|...|...|...|
 
 On inference, the output is returned in the library input format, introduced first.
 
diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index 74ef6c5d..29920541 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -46,6 +46,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `seq_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
+| `target_max_offset` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `seq_length + target_max_offset`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
 | `stride_by_split` | `list[int]` | No | `[seq_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 657cf59c..241983bc 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -97,16 +97,16 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`, which takes this form:
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `seq_length + target_max_offset` (`target_max_offset` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
 
-|sequenceId|subsequenceId|startItemPosition|columnName|[Subsequence Length]|[Subsequence Length - 1]|...|0|
-|----------|-------------|-----------------|----------|--------------------|------------------------| - |-|
-|0|0|0|column1|"high"|"high"|...|"low"|
-|0|0|0|column2|12.3|10.2|...|14.9|
-|...|...|...|...|...|...|...|...|
-|1|0|15|column1|"medium"|"high"|...|"medium"|
-|1|0|15|column2|20.6|18.5|...|21.6|
-|...|...|...|...|...|...|...|...|
+|sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
+|----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
+|0|0|0|0|column1|"high"|"high"|...|"low"|
+|0|0|0|0|column2|12.3|10.2|...|14.9|
+|...|...|...|...|...|...|...|...|...|
+|1|0|15|0|column1|"medium"|"high"|...|"medium"|
+|1|0|15|0|column2|20.6|18.5|...|21.6|
+|...|...|...|...|...|...|...|...|...|
 
 On inference, the output is returned in the library input format, introduced first.
 
@@ -254,6 +254,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `seq_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
+| `target_max_offset` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `seq_length + target_max_offset`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
 | `stride_by_split` | `list[int]` | No | `[seq_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index cb3841f8..74080dcf 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -415,7 +415,9 @@ def validate_scheduler_config(cls, v, info_dict):
                     )
         return v
 
-    def sample_trial(self, trial: Any) -> TrainingSpecModel:
+    def sample_trial(
+        self, trial: Any, target_max_offset: int, window_length: int
+    ) -> TrainingSpecModel:
         """Samples training hyperparameters using an Optuna trial.
 
         This method leverages the provided Optuna trial to suggest values for
@@ -498,6 +500,8 @@ def sample_trial(self, trial: Any) -> TrainingSpecModel:
             fsdp_cpu_offload=self.fsdp_cpu_offload,
             torch_compile=self.torch_compile,
             float32_matmul_precision=self.float32_matmul_precision,
+            target_max_offset=target_max_offset,
+            window_length=window_length,
         )
 
 
@@ -699,6 +703,7 @@ class HyperparameterSearchConfig(BaseModel):
     id_maps: dict[str, dict[str | int, int]]
 
     seq_length: list[int]
+    target_max_offset: int = Field(default=1, ge=0)
     n_classes: dict[str, int]
     inference_batch_size: int
 
@@ -816,12 +821,17 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             TrainModel: A fully populated configuration instance for the current trial.
         """
         model_spec = self.model_hyperparameter_sampling.sample_trial(trial)
-        training_spec = self.training_hyperparameter_sampling.sample_trial(trial)
 
         input_columns_index = trial.suggest_categorical(
             "input_columns_index", list(range(len(self.input_columns)))
         )
         seq_length = trial.suggest_categorical("seq_length", self.seq_length)
+        window_length = seq_length + self.target_max_offset
+        training_spec = self.training_hyperparameter_sampling.sample_trial(
+            trial,
+            target_max_offset=self.target_max_offset,
+            window_length=window_length,
+        )
 
         logger.info(f"{input_columns_index = } - {seq_length = }")
 
@@ -840,6 +850,8 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             target_column_types=self.target_column_types,
             id_maps=self.id_maps,
             seq_length=seq_length,
+            target_max_offset=self.target_max_offset,
+            window_length=window_length,
             n_classes=self.n_classes,
             inference_batch_size=self.inference_batch_size,
             seed=101,
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 59d1ef75..a05053bb 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -14,7 +14,11 @@
     model_validator,
 )
 
-from sequifier.helpers import normalize_path, try_catch_excess_keys
+from sequifier.helpers import (
+    normalize_path,
+    sequence_layout_from_metadata,
+    try_catch_excess_keys,
+)
 from sequifier.special_tokens import validate_special_token_ids
 
 
@@ -51,6 +55,14 @@ def load_inferer_config(
             metadata_config.get("special_token_ids"),
             source=f"metadata config '{metadata_config_path}'",
         )
+        sequence_layout = sequence_layout_from_metadata(
+            metadata_config, config_values["seq_length"]
+        )
+        config_values["target_max_offset"] = sequence_layout.target_max_offset
+        config_values["window_length"] = sequence_layout.window_length
+        config_values["sequence_layout_version"] = (
+            sequence_layout.sequence_layout_version
+        )
 
         config_values["column_types"] = config_values.get(
             "column_types", metadata_config["column_types"]
@@ -143,6 +155,9 @@ class InfererModel(BaseModel):
     seed: int
     device: str
     seq_length: int
+    target_max_offset: int = Field(default=1, ge=0)
+    window_length: int
+    sequence_layout_version: int = 1
     prediction_length: Optional[int] = None
     inference_batch_size: int
 
@@ -153,6 +168,11 @@ class InfererModel(BaseModel):
 
     @model_validator(mode="after")
     def normalize_prediction_length(self):
+        if self.window_length != self.seq_length + self.target_max_offset:
+            raise ValueError(
+                "window_length must equal seq_length + target_max_offset "
+                f"({self.window_length} != {self.seq_length} + {self.target_max_offset})."
+            )
         if self.prediction_length is None:
             self.prediction_length = (
                 self.seq_length if self.training_objective == "bert" else 1
@@ -163,8 +183,22 @@ def normalize_prediction_length(self):
                     "For BERT inference, prediction_length must be equal to seq_length "
                     f"(got prediction_length={self.prediction_length}, seq_length={self.seq_length})."
                 )
+        elif self.target_max_offset < 1:
+            raise ValueError(
+                "Causal inference requires data preprocessed with target_max_offset >= 1"
+            )
         return self
 
+    @property
+    def data_offset(self) -> int:
+        return self.target_max_offset
+
+    @property
+    def target_offset(self) -> int:
+        if self.training_objective == "bert":
+            return self.target_max_offset
+        return self.target_max_offset - 1
+
     @field_validator("training_objective")
     @classmethod
     def validate_training_objective(cls, v):
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index 4777d26d..e72a62ed 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -8,6 +8,7 @@
 from pydantic import (
     BaseModel,
     ConfigDict,
+    Field,
     ValidationInfo,
     field_validator,
     model_validator,
@@ -54,6 +55,7 @@ class PreprocessorModel(BaseModel):
         split_ratios: A list of floats that define the relative sizes of data splits (e.g., for train, validation, test).
                            The sum of proportions must be 1.0.
         seq_length: The sequence length for the model inputs.
+        target_max_offset: The maximum retained target offset after the model input window.
         stride_by_split: A list of step sizes for creating subsequences within each data split.
         max_rows: The maximum number of input rows to process. If None, all rows are processed.
         seed: A random seed for reproducibility.
@@ -77,6 +79,7 @@ class PreprocessorModel(BaseModel):
 
     split_ratios: list[float]
     seq_length: int
+    target_max_offset: int = Field(default=1, ge=0)
     stride_by_split: Optional[list[int]] = None
     max_rows: Optional[int] = None
     seed: int
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 5aa6ae29..992c76ee 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -22,7 +22,11 @@
 
 import sequifier
 from sequifier.config.probabilities import ProbabilityDistribution
-from sequifier.helpers import normalize_path, try_catch_excess_keys
+from sequifier.helpers import (
+    normalize_path,
+    sequence_layout_from_metadata,
+    try_catch_excess_keys,
+)
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS, validate_special_token_ids
 
 AnyType = str | int | float
@@ -58,6 +62,20 @@ def load_train_config(
         ) as f:
             metadata_config = json.loads(f.read())
 
+        sequence_layout = sequence_layout_from_metadata(
+            metadata_config, config_values["seq_length"]
+        )
+        config_values["target_max_offset"] = sequence_layout.target_max_offset
+        config_values["window_length"] = sequence_layout.window_length
+        config_values["sequence_layout_version"] = (
+            sequence_layout.sequence_layout_version
+        )
+        config_values.setdefault("training_spec", {})
+        config_values["training_spec"]["target_max_offset"] = (
+            sequence_layout.target_max_offset
+        )
+        config_values["training_spec"]["window_length"] = sequence_layout.window_length
+
         split_paths = metadata_config["split_paths"]
 
         config_values["column_types"] = config_values.get(
@@ -227,6 +245,8 @@ class TrainingSpecModel(BaseModel):
     fsdp_cpu_offload: Optional[bool] = None
     torch_compile: str = "outer"
     float32_matmul_precision: str = "highest"
+    target_max_offset: int = Field(default=1, ge=0)
+    window_length: int
 
     def __init__(self, **kwargs):
         super().__init__(
@@ -244,11 +264,13 @@ def serialize_dotdict(self, value: DotDict) -> dict[str, Any]:
 
     @property
     def data_offset(self) -> int:
-        return 1
+        return self.target_max_offset
 
     @property
     def target_offset(self) -> int:
-        return 0 if self.training_objective == "causal" else 1
+        if self.training_objective == "bert":
+            return self.target_max_offset
+        return self.target_max_offset - 1
 
     @field_validator("layer_type_dtypes")
     @classmethod
@@ -360,6 +382,14 @@ def validate_bert_spec_matches_objective(self):
             )
         return self
 
+    @model_validator(mode="after")
+    def validate_sequence_layout(self):
+        if self.training_objective == "causal" and self.target_max_offset < 1:
+            raise ValueError(
+                "Causal training requires data preprocessed with target_max_offset >= 1"
+            )
+        return self
+
     @field_validator("sampling_strategy")
     @classmethod
     def validate_sampling_strategy(cls, v):
@@ -563,6 +593,9 @@ class TrainModel(BaseModel):
     )
 
     seq_length: int
+    target_max_offset: int = Field(default=1, ge=0)
+    window_length: int
+    sequence_layout_version: int = 1
     n_classes: dict[str, int]
     inference_batch_size: int
     seed: int
@@ -583,6 +616,13 @@ def validate_special_token_ids_match_runtime(cls, v):
 
     @model_validator(mode="after")
     def validate_bert_prediction_length_matches_seq_length(self):
+        if self.window_length != self.seq_length + self.target_max_offset:
+            raise ValueError(
+                "window_length must equal seq_length + target_max_offset "
+                f"({self.window_length} != {self.seq_length} + {self.target_max_offset})."
+            )
+        self.training_spec.target_max_offset = self.target_max_offset
+        self.training_spec.window_length = self.window_length
         if (
             self.training_spec.training_objective == "bert"
             and self.model_spec.prediction_length != self.seq_length
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 5cb2db62..1bac51bd 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -3,6 +3,7 @@
 import random
 import re
 import sys
+from dataclasses import dataclass
 from datetime import datetime
 from typing import Any, Dict, Optional, Union
 
@@ -48,6 +49,100 @@
 )
 
 
+@dataclass(frozen=True)
+class SequenceLayout:
+    seq_length: int
+    target_max_offset: int = 1
+    sequence_layout_version: int = 2
+
+    def __post_init__(self) -> None:
+        if self.seq_length < 1:
+            raise ValueError("seq_length must be a positive integer")
+        if self.target_max_offset < 0:
+            raise ValueError("target_max_offset must be non-negative")
+
+    @property
+    def window_length(self) -> int:
+        return self.seq_length + self.target_max_offset
+
+    @property
+    def input_offset(self) -> int:
+        return self.target_max_offset
+
+    def target_offset(self, horizon: int) -> int:
+        offset = self.target_max_offset - horizon
+        if offset < 0:
+            raise ValueError(
+                f"horizon={horizon} exceeds target_max_offset={self.target_max_offset}"
+            )
+        return offset
+
+
+@beartype
+def sequence_column_names(seq_length: int, offset: int) -> list[str]:
+    if offset < 0:
+        raise ValueError("offset must be non-negative")
+    return [str(i) for i in range(seq_length - 1 + offset, offset - 1, -1)]
+
+
+@beartype
+def slice_window(tensor: Tensor, seq_length: int, offset: int) -> Tensor:
+    if offset < 0:
+        raise ValueError("offset must be non-negative")
+
+    end = -offset if offset else None
+    result = tensor[:, -(seq_length + offset) : end]
+
+    if result.shape[1] != seq_length:
+        raise ValueError(
+            f"Stored window width {tensor.shape[1]} cannot provide "
+            f"seq_length={seq_length} at offset={offset}"
+        )
+    return result
+
+
+@beartype
+def validate_stored_window_width(tensor: Tensor, window_length: int) -> None:
+    if tensor.shape[1] != window_length:
+        raise ValueError(
+            f"Stored window width {tensor.shape[1]} does not match "
+            f"metadata window_length={window_length}."
+        )
+
+
+@beartype
+def sequence_layout_from_metadata(metadata: dict, seq_length: int) -> SequenceLayout:
+    metadata_seq_length = int(metadata.get("seq_length", seq_length))
+    if metadata_seq_length != seq_length:
+        raise ValueError(
+            f"Configured seq_length={seq_length} does not match preprocessed "
+            f"metadata seq_length={metadata_seq_length}."
+        )
+
+    target_max_offset = int(metadata.get("target_max_offset", 1))
+    window_length = int(
+        metadata.get("window_length", metadata_seq_length + target_max_offset)
+    )
+    if window_length != metadata_seq_length + target_max_offset:
+        raise ValueError(
+            "Invalid sequence layout metadata: window_length must equal "
+            "seq_length + target_max_offset "
+            f"({window_length} != {metadata_seq_length} + {target_max_offset})."
+        )
+
+    layout = SequenceLayout(
+        seq_length=metadata_seq_length,
+        target_max_offset=target_max_offset,
+        sequence_layout_version=int(metadata.get("sequence_layout_version", 1)),
+    )
+    if layout.window_length != window_length:
+        raise ValueError(
+            f"Resolved layout window_length={layout.window_length} does not match "
+            f"metadata window_length={window_length}."
+        )
+    return layout
+
+
 # Check an environment variable to see if we are in a testing context
 IS_TESTING = os.environ.get("SEQUIFIER_TESTING", "0") == "1"
 
@@ -245,6 +340,7 @@ def numpy_to_pytorch(
     column_types: dict[str, torch.dtype],
     all_columns: list[str],
     seq_length: int,
+    window_length: int,
     data_offset: int,
     target_offset: int,
 ) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
@@ -282,12 +378,8 @@ def numpy_to_pytorch(
           (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
         - a metadata dictionary containing any explicit masks.
     """
-    input_seq_cols = [
-        str(c) for c in range(seq_length - 1 + data_offset, (-1 + data_offset), -1)
-    ]
-    target_seq_cols = [
-        str(c) for c in range(seq_length - 1 + target_offset, (-1 + target_offset), -1)
-    ]
+    input_seq_cols = sequence_column_names(seq_length, data_offset)
+    target_seq_cols = sequence_column_names(seq_length, target_offset)
 
     # We will create a unified dictionary
     unified_tensors = {}
@@ -315,7 +407,11 @@ def numpy_to_pytorch(
     left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
     if left_pad_lengths is not None:
         metadata = generate_padding_masks(
-            left_pad_lengths, seq_length, data_offset, target_offset
+            left_pad_lengths,
+            seq_length,
+            window_length,
+            data_offset,
+            target_offset,
         )
     else:
         metadata = {}
@@ -362,17 +458,17 @@ def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Optional[
 def generate_padding_masks(
     left_pad_lengths: Tensor,
     seq_length: int,
+    window_length: int,
     data_offset: int,
     target_offset: int,
 ) -> dict[str, Tensor]:
     """Generates explicit attention and target masks as a metadata dictionary."""
-    full_length = seq_length + 1
     return {
         "attention_valid_mask": build_valid_mask(
-            left_pad_lengths, full_length, data_offset, seq_length
+            left_pad_lengths, window_length, data_offset, seq_length
         ),
         "target_valid_mask": build_valid_mask(
-            left_pad_lengths, full_length, target_offset, seq_length
+            left_pad_lengths, window_length, target_offset, seq_length
         ),
     }
 
@@ -603,6 +699,12 @@ def apply_bert_masking(
 
     batch_size, seq_len = valid_mask.shape
     device = valid_mask.device
+    for target_name, target in targets_batch.items():
+        if target.shape != valid_mask.shape:
+            raise ValueError(
+                f"BERT target {target_name!r} has shape {target.shape}; "
+                f"expected {valid_mask.shape}"
+            )
 
     if eval_seed is not None:
         cpu_rng_state = torch.get_rng_state()
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index f4dbad3f..d6c2e864 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -20,7 +20,9 @@
     generate_padding_masks,
     normalize_path,
     numpy_to_pytorch,
+    slice_window,
     subset_to_input_columns,
+    validate_stored_window_width,
     write_data,
 )
 from sequifier.special_tokens import validate_special_token_ids
@@ -362,8 +364,9 @@ def _bert_target_valid_mask_from_preprocessed_data(
         metadata = generate_padding_masks(
             left_pad_lengths,
             config.seq_length,
-            data_offset=1,
-            target_offset=1,
+            config.window_length,
+            data_offset=config.data_offset,
+            target_offset=config.target_offset,
         )
         return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
 
@@ -491,14 +494,16 @@ def infer_embedding(
                 start_positions_tensor,
                 left_pad_lengths_tensor,
             ) = data
+            for tensor in sequences_dict.values():
+                validate_stored_window_width(tensor, config.window_length)
             metadata = {}
             if left_pad_lengths_tensor is not None:
-                target_offset = 0 if config.training_objective == "causal" else 1
                 metadata = generate_padding_masks(
                     left_pad_lengths_tensor,
                     config.seq_length,
-                    data_offset=1,
-                    target_offset=target_offset,
+                    config.window_length,
+                    data_offset=config.data_offset,
+                    target_offset=config.target_offset,
                 )
             embeddings = get_embeddings_pt(
                 config, inferer, sequences_dict, metadata=metadata
@@ -738,6 +743,8 @@ def infer_generative(
                 start_positions_tensor,
                 left_pad_lengths_tensor,
             ) = data
+            for tensor in sequences_dict.values():
+                validate_stored_window_width(tensor, config.window_length)
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
@@ -745,19 +752,19 @@ def infer_generative(
             )
 
             sequences_dict = {
-                key: tensor[:, :-1]
+                key: slice_window(tensor, config.seq_length, config.data_offset)
                 for key, tensor in sequences_dict.items()
                 if key in config.input_columns
             }
 
             metadata = {}
             if left_pad_lengths_tensor is not None:
-                target_offset = 0 if config.training_objective == "causal" else 1
                 metadata = generate_padding_masks(
                     left_pad_lengths_tensor,
                     config.seq_length,
-                    data_offset=1,
-                    target_offset=target_offset,
+                    config.window_length,
+                    data_offset=config.data_offset,
+                    target_offset=config.target_offset,
                 )
 
             probs, preds = get_probs_preds_from_dict(
@@ -936,8 +943,10 @@ def get_embeddings_pt(
     Returns:
         A NumPy array containing the computed embeddings for the batch.
     """
+    for tensor in data.values():
+        validate_stored_window_width(tensor, config.window_length)
     X = {
-        key: val[:, :-1].numpy()
+        key: slice_window(val, config.seq_length, config.data_offset).numpy()
         for key, val in data.items()
         if key in config.input_columns
     }
@@ -1087,9 +1096,14 @@ def get_embeddings(
         A NumPy array containing the computed embeddings for the batch.
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
-    target_offset = 0 if config.training_objective == "causal" else 1
     X, metadata = numpy_to_pytorch(
-        data, column_types, all_columns, config.seq_length, 1, target_offset
+        data,
+        column_types,
+        all_columns,
+        config.seq_length,
+        config.window_length,
+        config.data_offset,
+        config.target_offset,
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     metadata_np = (
@@ -1135,10 +1149,14 @@ def get_probs_preds_from_df(
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
 
-    target_offset = 0 if config.training_objective == "causal" else 1
-
     X, metadata = numpy_to_pytorch(
-        data, column_types, all_columns, config.seq_length, 1, target_offset
+        data,
+        column_types,
+        all_columns,
+        config.seq_length,
+        config.window_length,
+        config.data_offset,
+        config.target_offset,
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     metadata_np = (
@@ -1266,7 +1284,8 @@ def get_probs_preds_autoregression(
         column_types,
         config.input_columns,
         seq_length,
-        data_offset=1,
+        config.window_length,
+        data_offset=config.data_offset,
         target_offset=0,
     )
 
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index 87203d2e..51a778ce 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -41,16 +41,15 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in config.column_types
         }
 
-        target_offset = 0 if config.training_spec.training_objective == "causal" else 1
-
         # self.all_tensors now holds both inputs and targets
         all_tensors, metadata_tensors = numpy_to_pytorch(
             data=data_df,
             column_types=column_types,
             all_columns=all_columns,
             seq_length=config.seq_length,
-            data_offset=1,
-            target_offset=target_offset,
+            window_length=config.window_length,
+            data_offset=config.training_spec.data_offset,
+            target_offset=config.training_spec.target_offset,
         )
         self.n_samples = all_tensors[all_columns[0]].shape[0]
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 2d2cae63..d667f6c5 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -15,6 +15,8 @@
     generate_padding_masks,
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
+    sequence_column_names,
+    sequence_layout_from_metadata,
 )
 from sequifier.io.batch import SequifierBatch
 
@@ -47,6 +49,13 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
+        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
+        if folder_layout.window_length != config.window_length:
+            raise ValueError(
+                f"Preprocessed folder window_length={folder_layout.window_length} "
+                f"does not match config window_length={config.window_length}."
+            )
+
         self.n_samples = metadata["total_samples"]
 
         logger.info(
@@ -60,22 +69,12 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
 
         # Sequence formatting structures matching long-format schema boundaries
         train_seq_len = self.config.seq_length
-        input_seq_cols = [
-            str(c)
-            for c in range(
-                train_seq_len - 1 + self.config.training_spec.data_offset,
-                (-1 + self.config.training_spec.data_offset),
-                -1,
-            )
-        ]
-        target_seq_cols = [
-            str(c)
-            for c in range(
-                train_seq_len - 1 + self.config.training_spec.target_offset,
-                (-1 + self.config.training_spec.target_offset),
-                -1,
-            )
-        ]
+        input_seq_cols = sequence_column_names(
+            train_seq_len, self.config.training_spec.data_offset
+        )
+        target_seq_cols = sequence_column_names(
+            train_seq_len, self.config.training_spec.target_offset
+        )
         all_sequences: Dict[str, list[torch.Tensor]] = {
             col: [] for col in config.input_columns
         }
@@ -232,6 +231,7 @@ def __iter__(
                 metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
+                    self.config.window_length,
                     self.config.training_spec.data_offset,
                     self.config.training_spec.target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index c922d88c..2145539b 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -16,6 +16,8 @@
     generate_padding_masks,
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
+    sequence_column_names,
+    sequence_layout_from_metadata,
 )
 from sequifier.io.batch import SequifierBatch
 
@@ -63,6 +65,13 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
+        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
+        if folder_layout.window_length != config.window_length:
+            raise ValueError(
+                f"Preprocessed folder window_length={folder_layout.window_length} "
+                f"does not match config window_length={config.window_length}."
+            )
+
         self.batch_files_info = metadata["batch_files"]
         self.total_samples = metadata["total_samples"]
         self.sampling_strategy = config.training_spec.sampling_strategy
@@ -213,22 +222,12 @@ def __iter__(
         train_seq_len = self.config.seq_length
         global_file_start_sample = 0
 
-        input_seq_cols = [
-            str(c)
-            for c in range(
-                train_seq_len - 1 + self.config.training_spec.data_offset,
-                (-1 + self.config.training_spec.data_offset),
-                -1,
-            )
-        ]
-        target_seq_cols = [
-            str(c)
-            for c in range(
-                train_seq_len - 1 + self.config.training_spec.target_offset,
-                (-1 + self.config.training_spec.target_offset),
-                -1,
-            )
-        ]
+        input_seq_cols = sequence_column_names(
+            train_seq_len, self.config.training_spec.data_offset
+        )
+        target_seq_cols = sequence_column_names(
+            train_seq_len, self.config.training_spec.target_offset
+        )
 
         # Initialize cross-file buffers
         seq_buffer: Dict[str, torch.Tensor] = {}
@@ -320,6 +319,7 @@ def __iter__(
                 new_meta = generate_padding_masks(
                     left_pad_lengths[worker_indices],
                     train_seq_len,
+                    self.config.window_length,
                     self.config.training_spec.data_offset,
                     self.config.training_spec.target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 988ca94d..b6108fb6 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -9,7 +9,13 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import generate_padding_masks, normalize_path
+from sequifier.helpers import (
+    generate_padding_masks,
+    normalize_path,
+    sequence_layout_from_metadata,
+    slice_window,
+    validate_stored_window_width,
+)
 from sequifier.io.batch import SequifierBatch
 
 
@@ -41,6 +47,13 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
+        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
+        if folder_layout.window_length != config.window_length:
+            raise ValueError(
+                f"Preprocessed folder window_length={folder_layout.window_length} "
+                f"does not match config window_length={config.window_length}."
+            )
+
         self.n_samples = metadata["total_samples"]
 
         logger.info(
@@ -64,6 +77,9 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             ) = torch.load(file_path, map_location="cpu", weights_only=False)
             for col in all_sequences.keys():
                 if col in sequences_batch:
+                    validate_stored_window_width(
+                        sequences_batch[col], config.window_length
+                    )
                     all_sequences[col].append(sequences_batch[col])
             if left_pad_lengths_batch is not None:
                 all_left_pad_lengths.append(left_pad_lengths_batch)
@@ -166,22 +182,12 @@ def __iter__(
             data_offset = self.config.training_spec.data_offset
             target_offset = self.config.training_spec.target_offset
             data_batch = {
-                key: tensor[
-                    batch_indices,
-                    -(train_seq_len + data_offset) : (
-                        -data_offset if data_offset > 0 else None
-                    ),
-                ]
+                key: slice_window(tensor[batch_indices], train_seq_len, data_offset)
                 for key, tensor in self.sequences.items()
                 if key in self.config.input_columns
             }
             targets_batch = {
-                key: tensor[
-                    batch_indices,
-                    -(train_seq_len + target_offset) : (
-                        -target_offset if target_offset > 0 else None
-                    ),
-                ]
+                key: slice_window(tensor[batch_indices], train_seq_len, target_offset)
                 for key, tensor in self.sequences.items()
                 if key in self.config.target_columns
             }
@@ -191,6 +197,7 @@ def __iter__(
                 metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
+                    self.config.window_length,
                     data_offset,
                     target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index dc60aa34..36e43b7b 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -10,7 +10,13 @@
 from torch.utils.data import IterableDataset, get_worker_info
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import generate_padding_masks, normalize_path
+from sequifier.helpers import (
+    generate_padding_masks,
+    normalize_path,
+    sequence_layout_from_metadata,
+    slice_window,
+    validate_stored_window_width,
+)
 from sequifier.io.batch import SequifierBatch
 
 
@@ -57,6 +63,13 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
+        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
+        if folder_layout.window_length != config.window_length:
+            raise ValueError(
+                f"Preprocessed folder window_length={folder_layout.window_length} "
+                f"does not match config window_length={config.window_length}."
+            )
+
         self.batch_files_info = metadata["batch_files"]
         self.total_samples = metadata["total_samples"]
         self.sampling_strategy = config.training_spec.sampling_strategy
@@ -230,6 +243,8 @@ def __iter__(
                 _,
                 left_pad_lengths_batch,
             ) = torch.load(file_path, map_location="cpu", weights_only=False)
+            for tensor in sequences_batch.values():
+                validate_stored_window_width(tensor, self.config.window_length)
 
             # Generate indices for the whole file
             indices = torch.arange(file_samples)
@@ -253,22 +268,12 @@ def __iter__(
             data_offset = self.config.training_spec.data_offset
             target_offset = self.config.training_spec.target_offset
             new_seq = {
-                k: v[
-                    worker_indices,
-                    -(train_seq_len + data_offset) : (
-                        -data_offset if data_offset > 0 else None
-                    ),
-                ]
+                k: slice_window(v[worker_indices], train_seq_len, data_offset)
                 for k, v in sequences_batch.items()
                 if k in self.config.input_columns
             }
             new_tgt = {
-                k: v[
-                    worker_indices,
-                    -(train_seq_len + target_offset) : (
-                        -target_offset if target_offset > 0 else None
-                    ),
-                ]
+                k: slice_window(v[worker_indices], train_seq_len, target_offset)
                 for k, v in sequences_batch.items()
                 if k in self.config.target_columns
             }
@@ -278,6 +283,7 @@ def __iter__(
                 new_meta = generate_padding_masks(
                     left_pad_lengths_batch[worker_indices],
                     train_seq_len,
+                    self.config.window_length,
                     data_offset,
                     target_offset,
                 )
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 8777ac97..f7434b49 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -12,6 +12,7 @@
 - 0.1
 - 0.1
 seq_length: 48
+target_max_offset: 1
 stride_by_split:
 - 1
 - 1
@@ -30,6 +31,7 @@
   EXAMPLE_TARGET_COLUMN_NAME: real
 
 seq_length: 48
+window_length: 49
 inference_batch_size: 10
 
 export_generative_model: PLEASE FILL # true or false
@@ -47,6 +49,7 @@
   num_layers: 3
   prediction_length: 1
 training_spec:
+  window_length: 49
   training_objective: causal
   device: cuda
   epochs: 10
@@ -88,6 +91,7 @@
 map_to_id: true
 device: cpu
 seq_length: 48
+window_length: 49
 inference_batch_size: 10
 
 autoregression: true
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 4b5afc75..01d312b1 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -16,7 +16,12 @@
 from loguru import logger
 
 from sequifier.config.preprocess_config import load_preprocessor_config
-from sequifier.helpers import PANDAS_TO_TORCH_TYPES, read_data, write_data
+from sequifier.helpers import (
+    PANDAS_TO_TORCH_TYPES,
+    SequenceLayout,
+    read_data,
+    write_data,
+)
 from sequifier.special_tokens import (
     SPECIAL_TOKEN_ID_VALUES,
     SPECIAL_TOKEN_IDS,
@@ -90,6 +95,7 @@ def __init__(
         subsequence_start_mode: str,
         use_precomputed_maps: Optional[list[str]],
         metadata_config_path: Optional[str],
+        target_max_offset: int = 1,
         mask_column: Optional[str] = None,
     ):
         """Initializes the Preprocessor with the given parameters.
@@ -103,6 +109,7 @@ def __init__(
             selected_columns: A list of columns to be included in the preprocessing.
             split_ratios: A list of floats that define the relative sizes of data splits.
             seq_length: The sequence length for the model inputs.
+            target_max_offset: The maximum target horizon retained after each input window.
             stride_by_split: A list of step sizes for creating subsequences.
             max_rows: The maximum number of input rows to process.
             seed: A random seed for reproducibility.
@@ -137,6 +144,7 @@ def __init__(
         np.random.seed(seed)
         self.n_cores = n_cores or multiprocessing.cpu_count()
         self.continue_preprocessing = continue_preprocessing
+        self.layout = SequenceLayout(seq_length, target_max_offset)
         self._setup_directories()
 
         if selected_columns is not None:
@@ -232,7 +240,13 @@ def __init__(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
 
-            schema = self._create_schema(col_types, seq_length + 1)
+            self._write_or_validate_resume_manifest(
+                selected_columns,
+                write_format,
+                data_columns,
+            )
+
+            schema = self._create_schema(col_types, self.layout.window_length)
 
             data = data.sort(["sequenceId", "itemPosition"])
             n_batches = _process_batches_single_file(
@@ -241,7 +255,7 @@ def __init__(
                 data,
                 schema,
                 self.n_cores,
-                seq_length,
+                self.layout,
                 stride_by_split,
                 data_columns,
                 col_types,
@@ -323,7 +337,12 @@ def __init__(
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
-            schema = self._create_schema(col_types, seq_length + 1)
+            self._write_or_validate_resume_manifest(
+                selected_columns,
+                write_format,
+                data_columns,
+            )
+            schema = self._create_schema(col_types, self.layout.window_length)
 
             self._process_batches_multiple_files(
                 files_to_process,
@@ -332,7 +351,7 @@ def __init__(
                 max_rows,
                 schema,
                 self.n_cores,
-                seq_length,
+                self.layout,
                 stride_by_split,
                 data_columns,
                 n_classes,
@@ -349,19 +368,19 @@ def __init__(
 
     @beartype
     def _create_schema(
-        self, col_types: dict[str, str], seq_length: int
+        self, col_types: dict[str, str], window_length: int
     ) -> dict[str, Any]:
         """Creates the Polars schema for the intermediate sequence DataFrame.
 
         This schema defines the structure of the DataFrame after sequence
         extraction, which includes sequence identifiers, the start item position
         within the original sequence, the input column name, and columns for
-        each item in the sequence (named '0', '1', ..., 'seq_length-1').
+        each stored item in the serialized window.
 
         Args:
             col_types: A dictionary mapping data column names to their Polars
                 string representations (e.g., "Int64", "Float64").
-            seq_length: The length of the sequences being extracted.
+            window_length: The number of items stored for each extracted window.
 
         Returns:
             A dictionary defining the Polars schema. Keys are column names
@@ -391,7 +410,7 @@ def _create_schema(
             sequence_position_type = pl.Float64
 
         schema.update(
-            {str(i): sequence_position_type for i in range(seq_length - 1, -1, -1)}
+            {str(i): sequence_position_type for i in range(window_length - 1, -1, -1)}
         )
 
         return schema
@@ -594,7 +613,7 @@ def _process_batches_multiple_files(
         max_rows: Optional[int],
         schema: Any,
         n_cores: int,
-        seq_length: int,
+        layout: SequenceLayout,
         stride_by_split: list[int],
         data_columns: list[str],
         n_classes: dict[str, int],
@@ -616,7 +635,7 @@ def _process_batches_multiple_files(
             max_rows: The maximum number of rows to process.
             schema: The schema for the preprocessed data.
             n_cores: The number of cores to use for parallel processing.
-            seq_length: The sequence length for the model inputs.
+            layout: The sequence and serialized-window layout.
             stride_by_split: A list of step sizes for creating subsequences.
             data_columns: A list of data columns.
             n_classes: A dictionary containing the number of classes for each categorical column.
@@ -643,7 +662,7 @@ def _process_batches_multiple_files(
                 max_rows=max_rows,
                 schema=schema,
                 n_cores=n_cores,
-                seq_length=seq_length,
+                layout=layout,
                 stride_by_split=stride_by_split,
                 data_columns=data_columns,
                 n_classes=n_classes,
@@ -688,7 +707,7 @@ def _process_batches_multiple_files(
                 "max_rows": max_rows,
                 "schema": schema,
                 "n_cores": 1,
-                "seq_length": seq_length,
+                "layout": layout,
                 "stride_by_split": stride_by_split,
                 "data_columns": data_columns,
                 "n_classes": n_classes,
@@ -790,6 +809,49 @@ def _cleanup(self, write_format: str) -> None:
         if not os.listdir(directory) or self.target_dir == "temp":
             shutil.rmtree(directory)
 
+    @beartype
+    def _layout_metadata(self) -> dict[str, int]:
+        return {
+            "seq_length": self.layout.seq_length,
+            "target_max_offset": self.layout.target_max_offset,
+            "window_length": self.layout.window_length,
+            "sequence_layout_version": self.layout.sequence_layout_version,
+        }
+
+    @beartype
+    def _write_or_validate_resume_manifest(
+        self,
+        selected_columns: Optional[list[str]],
+        write_format: str,
+        data_columns: list[str],
+    ) -> None:
+        manifest = {
+            **self._layout_metadata(),
+            "selected_columns": selected_columns,
+            "data_columns": data_columns,
+            "write_format": write_format,
+        }
+        manifest_path = os.path.join(
+            self.project_root, "data", self.target_dir, "preprocess-manifest.json"
+        )
+
+        if self.continue_preprocessing:
+            if not os.path.exists(manifest_path):
+                raise ValueError(
+                    "Cannot continue preprocessing because the temp manifest is missing."
+                )
+            with open(manifest_path, "r") as f:
+                previous_manifest = json.load(f)
+            if previous_manifest != manifest:
+                raise ValueError(
+                    "Cannot continue preprocessing with different sequence layout, "
+                    "selected columns, or write format."
+                )
+            return
+
+        with open(manifest_path, "w") as f:
+            json.dump(manifest, f, indent=4)
+
     @beartype
     def _export_metadata(
         self,
@@ -828,6 +890,7 @@ def _export_metadata(
                 col: {"mean": stats["mean"], "std": stats["std"]}
                 for col, stats in selected_columns_statistics.items()
             },
+            **self._layout_metadata(),
         }
         os.makedirs(
             os.path.join(self.project_root, "configs", "metadata_configs"),
@@ -908,6 +971,7 @@ def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> No
         metadata = {
             "total_samples": total_samples,
             "batch_files": batch_files_metadata,
+            **self._layout_metadata(),
         }
 
         metadata_path = directory / "metadata.json"
@@ -1404,7 +1468,7 @@ def _process_batches_multiple_files_inner(
     max_rows: Optional[int],
     schema: Any,
     n_cores: int,
-    seq_length: int,
+    layout: SequenceLayout,
     stride_by_split: list[int],
     data_columns: list[str],
     n_classes: dict[str, int],
@@ -1433,7 +1497,7 @@ def _process_batches_multiple_files_inner(
         max_rows: The maximum number of rows to process.
         schema: The schema for the preprocessed data.
         n_cores: The number of cores to use for parallel processing.
-        seq_length: The sequence length for the model inputs.
+        layout: The sequence and serialized-window layout.
         stride_by_split: A list of step sizes for creating subsequences.
         data_columns: A list of data columns.
         n_classes: A dictionary containing the number of classes for each categorical column.
@@ -1515,7 +1579,7 @@ def _process_batches_multiple_files_inner(
                 data,
                 schema,
                 n_cores,
-                seq_length,
+                layout,
                 stride_by_split,
                 data_columns,
                 col_types,
@@ -1562,7 +1626,7 @@ def _process_batches_single_file(
     data: pl.DataFrame,
     schema: Any,
     n_cores: Optional[int],
-    seq_length: int,
+    layout: SequenceLayout,
     stride_by_split: list[int],
     data_columns: list[str],
     col_types: dict[str, str],
@@ -1582,7 +1646,7 @@ def _process_batches_single_file(
         data: The data to process.
         schema: The schema for the preprocessed data.
         n_cores: The number of cores to use for parallel processing.
-        seq_length: The sequence length for the model inputs.
+        layout: The sequence and serialized-window layout.
         stride_by_split: A list of step sizes for creating subsequences.
         data_columns: A list of data columns.
         col_types: A dictionary containing the column types.
@@ -1608,7 +1672,7 @@ def _process_batches_single_file(
             data.slice(start, end - start),
             schema,
             split_paths,
-            seq_length,
+            layout,
             stride_by_split,
             data_columns,
             col_types,
@@ -1840,7 +1904,10 @@ def get_group_bounds(data_subset: pl.DataFrame, split_ratios: list[float]):
 
 @beartype
 def process_and_write_data_pt(
-    data: pl.DataFrame, seq_length: int, path: str, column_types: dict[str, str]
+    data: pl.DataFrame,
+    window_length: int,
+    path: str,
+    column_types: dict[str, str],
 ):
     """Processes the sequence DataFrame and writes it to a .pt file.
 
@@ -1849,17 +1916,15 @@ def process_and_write_data_pt(
     so that each `inputCol` becomes its own column containing a list
     of sequence items. It also extracts the `startItemPosition`.
 
-    It then converts these lists into NumPy arrays, splits them into
-    `sequences` (all but last item) and `targets` (all but first item),
-    and converts them to PyTorch tensors along with sequence/subsequence IDs
-    and start positions. The final data tuple
-    `(sequences_dict, targets_dict, sequence_ids_tensor, subsequence_ids_tensor, start_item_positions_tensor)`
+    It then converts these lists into NumPy arrays and stores one full
+    sequence tensor per feature. Each tensor has shape
+    `(batch_size, window_length)`. The final five-element tuple
+    `(sequences_dict, sequence_ids_tensor, subsequence_ids_tensor, start_item_positions_tensor, left_pad_lengths_tensor)`
     is saved to a .pt file using `torch.save`.
 
     Args:
         data: The long-format Polars DataFrame of extracted sequences.
-        seq_length: The total sequence length (N). The resulting tensors
-            will have sequence length N-1.
+        window_length: The stored serialized window width.
         path: The output file path (e.g., "data/batch_0.pt").
         column_types: A dictionary mapping column names to their
             string data types, used to determine the correct torch dtype.
@@ -1867,7 +1932,7 @@ def process_and_write_data_pt(
     if data.is_empty():
         return
 
-    sequence_cols = [str(c) for c in range(seq_length, -1, -1)]
+    sequence_cols = [str(c) for c in range(window_length - 1, -1, -1)]
 
     all_feature_cols = data.get_column("inputCol").unique().to_list()
 
@@ -1936,7 +2001,7 @@ def _write_accumulated_sequences(
     process_id: int,
     file_index_str: str,
     target_dir: str,
-    seq_length: int,
+    layout: SequenceLayout,
     col_types: dict[str, str],
 ):
     """Helper to write a batch of accumulated sequences to a single .pt file.
@@ -1953,7 +2018,7 @@ def _write_accumulated_sequences(
         process_id: The ID of the parent multiprocessing process.
         file_index: The index of this file batch for this split and process.
         target_dir: The temporary directory to write the file into.
-        seq_length: The total sequence length.
+        layout: The sequence and serialized-window layout.
         col_types: A dictionary mapping column names to their string types.
     """
 
@@ -1967,7 +2032,9 @@ def _write_accumulated_sequences(
     out_path = insert_top_folder(split_path_batch_seq, target_dir)
 
     if write_format == "pt":
-        process_and_write_data_pt(combined_df, seq_length, out_path, col_types)
+        process_and_write_data_pt(
+            combined_df, layout.window_length, out_path, col_types
+        )
     elif write_format == "parquet":
         combined_df.write_parquet(out_path)
 
@@ -1980,7 +2047,7 @@ def preprocess_batch(
     batch: pl.DataFrame,
     schema: Any,
     split_paths: list[str],
-    seq_length: int,
+    layout: SequenceLayout,
     stride_by_split: list[int],
     data_columns: list[str],
     col_types: dict[str, str],
@@ -2000,7 +2067,7 @@ def preprocess_batch(
         batch: The batch of data to process.
         schema: The schema for the preprocessed data.
         split_paths: The paths to the output split files.
-        seq_length: The sequence length for the model inputs.
+        layout: The sequence and serialized-window layout.
         stride_by_split: A list of step sizes for creating subsequences.
         data_columns: A list of data columns.
         col_types: A dictionary containing the column types.
@@ -2026,7 +2093,7 @@ def preprocess_batch(
                     extract_sequences(
                         data_subset.slice(lb, ub - lb),
                         schema,
-                        seq_length,
+                        layout,
                         stride_by_split[i],
                         data_columns,
                         subsequence_start_mode,
@@ -2048,7 +2115,7 @@ def preprocess_batch(
                         process_id,
                         str(file_indices[group]).zfill(pad_width),
                         target_dir,
-                        seq_length,
+                        layout,
                         col_types,
                     )
                     # Reset the accumulator and increment the file index
@@ -2064,7 +2131,7 @@ def preprocess_batch(
                 process_id,
                 str(file_indices[group]).zfill(pad_width),
                 target_dir,
-                seq_length,
+                layout,
                 col_types,
             )
 
@@ -2078,7 +2145,7 @@ def preprocess_batch(
                     extract_sequences(
                         data_subset.slice(lb, ub - lb),
                         schema,
-                        seq_length,
+                        layout,
                         stride_by_split[j],
                         data_columns,
                         subsequence_start_mode,
@@ -2119,7 +2186,7 @@ def preprocess_batch(
 def extract_sequences(
     data: pl.DataFrame,
     schema: Any,
-    seq_length: int,
+    layout: SequenceLayout,
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
@@ -2128,7 +2195,7 @@ def extract_sequences(
 
     This function takes a DataFrame where each row contains all items
     for a single `sequenceId`. It iterates through each `sequenceId`,
-    extracts all possible subsequences of `seq_length` using the
+    extracts all possible serialized windows using the
     specified `stride_for_split`, calculates the starting position of each
     subsequence within the original sequence, and formats them into a new,
     long-format DataFrame that conforms to the provided `schema`.
@@ -2136,7 +2203,7 @@ def extract_sequences(
     Args:
         data: The input Polars DataFrame, grouped by "sequenceId".
         schema: The schema for the output long-format DataFrame.
-        seq_length: The length of the subsequences to extract.
+        layout: The sequence and serialized-window layout.
         stride_for_split: The step size to use when sliding the window
             to create subsequences.
         columns: A list of the data column names (features) to extract.
@@ -2161,7 +2228,7 @@ def extract_sequences(
 
         subsequences, left_pad_lengths, subsequence_starts = extract_subsequences(
             in_seq_lists_only,
-            seq_length,
+            layout.window_length,
             stride_for_split,
             columns,
             subsequence_start_mode,
@@ -2177,9 +2244,10 @@ def extract_sequences(
                     left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
-                if len(row) != (seq_length + 6):
+                expected_row_length = 5 + layout.window_length
+                if len(row) != expected_row_length:
                     raise RuntimeError(
-                        f"Row length mismatch. Expected {seq_length + 6}, got {len(row)}. Row: {row}"
+                        f"Row length mismatch. Expected {expected_row_length}, got {len(row)}. Row: {row}"
                     )
                 rows.append(row)
 
@@ -2194,21 +2262,21 @@ def extract_sequences(
 @beartype
 def get_subsequence_starts(
     in_seq_length: int,
-    seq_length: int,
+    window_length: int,
     stride_for_split: int,
     subsequence_start_mode: str,
 ) -> np.ndarray:
     """Calculates the start indices for extracting subsequences.
 
     This function determines the starting indices for sliding a window of
-    `seq_length` over an input sequence of `in_seq_length`. It aims to
+    `window_length` over an input sequence of `in_seq_length`. It aims to
     use `stride_for_split`, but adjusts the step size slightly to ensure
     that the windows are distributed as evenly as possible and cover the
     full sequence from the beginning to the end.
 
     Args:
         in_seq_length: The length of the original input sequence.
-        seq_length: The length of the subsequences to extract.
+        window_length: The stored window length to extract.
         stride_for_split: The *desired* step size between subsequences.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
 
@@ -2221,7 +2289,7 @@ def get_subsequence_starts(
         )
 
     if subsequence_start_mode == "distribute":
-        last_available_start = in_seq_length - (seq_length + 1)
+        last_available_start = in_seq_length - window_length
         raw_starts = np.arange(
             0, last_available_start + stride_for_split, stride_for_split
         )
@@ -2232,11 +2300,11 @@ def get_subsequence_starts(
         return np.unique(starts)
 
     if subsequence_start_mode == "exact":
-        if ((in_seq_length - 1) - seq_length) % stride_for_split != 0:
+        if (in_seq_length - window_length) % stride_for_split != 0:
             raise ValueError(
-                f"'exact' mode requires sequence length alignment, i.e. if: ((in_seq_length - 1) - seq_length) % stride_for_split == 0, {(in_seq_length -1) = }, {seq_length = }, {stride_for_split = }"
+                f"'exact' mode requires sequence length alignment, i.e. if: (in_seq_length - window_length) % stride_for_split == 0, {in_seq_length = }, {window_length = }, {stride_for_split = }"
             )
-        last_possible_start = in_seq_length - (seq_length + 1)
+        last_possible_start = in_seq_length - window_length
         return np.arange(0, last_possible_start + 1, stride_for_split)
     return np.array([])
 
@@ -2244,7 +2312,7 @@ def get_subsequence_starts(
 @beartype
 def extract_subsequences(
     in_seq: dict[str, list],
-    seq_length: int,
+    window_length: int,
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
@@ -2254,14 +2322,14 @@ def extract_subsequences(
     This function takes a dictionary `in_seq` where keys are column
     names and values are lists of items for a single full sequence.
     It first pads the sequences with 0s at the beginning if they are
-    shorter than `seq_length`. Then, it calculates the subsequence
+    shorter than `window_length`. Then, it calculates the subsequence
     start indices using `get_subsequence_starts` and extracts all
     subsequences.
 
     Args:
         in_seq: A dictionary mapping column names to lists of items
             (e.g., `{'col_A': [1, 2, 3, 4, 5], 'col_B': [6, 7, 8, 9, 10]}`).
-        seq_length: The length of the subsequences to extract.
+        window_length: The stored window length to extract.
         stride_for_split: The desired step size between subsequences.
         columns: A list of the column names (keys in `in_seq`) to process.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
@@ -2272,13 +2340,13 @@ def extract_subsequences(
     """
     in_seq_len = len(in_seq[columns[0]])
     pad_len = 0
-    if in_seq_len < (seq_length + 1):
-        pad_len = (seq_length + 1) - in_seq_len
+    if in_seq_len < window_length:
+        pad_len = window_length - in_seq_len
         in_seq = {col: ([0] * pad_len) + in_seq[col] for col in columns}
     in_seq_length = len(in_seq[columns[0]])
 
     subsequence_starts = get_subsequence_starts(
-        in_seq_length, seq_length, stride_for_split, subsequence_start_mode
+        in_seq_length, window_length, stride_for_split, subsequence_start_mode
     )
     subsequence_starts_diff = subsequence_starts[1:] - subsequence_starts[:-1]
     if not np.all(subsequence_starts_diff <= stride_for_split):
@@ -2287,7 +2355,7 @@ def extract_subsequences(
         )
 
     result = {
-        col: [list(in_seq[col][i : i + seq_length + 1]) for i in subsequence_starts]
+        col: [list(in_seq[col][i : i + window_length]) for i in subsequence_starts]
         for col in columns
     }
     left_pad_lengths = [pad_len] * len(subsequence_starts)
diff --git a/tests/configs/infer-test-categorical-autoregression.yaml b/tests/configs/infer-test-categorical-autoregression.yaml
index 76d9d06b..f34d4771 100644
--- a/tests/configs/infer-test-categorical-autoregression.yaml
+++ b/tests/configs/infer-test-categorical-autoregression.yaml
@@ -17,6 +17,7 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical-bert-embedding.yaml b/tests/configs/infer-test-categorical-bert-embedding.yaml
index 14e95fea..17b591fd 100644
--- a/tests/configs/infer-test-categorical-bert-embedding.yaml
+++ b/tests/configs/infer-test-categorical-bert-embedding.yaml
@@ -17,6 +17,7 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 prediction_length: null
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-bert.yaml b/tests/configs/infer-test-categorical-bert.yaml
index c6ec607f..c213d865 100644
--- a/tests/configs/infer-test-categorical-bert.yaml
+++ b/tests/configs/infer-test-categorical-bert.yaml
@@ -17,6 +17,7 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 prediction_length: null
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-embedding.yaml b/tests/configs/infer-test-categorical-embedding.yaml
index 9ea82e5a..bbb67fcf 100644
--- a/tests/configs/infer-test-categorical-embedding.yaml
+++ b/tests/configs/infer-test-categorical-embedding.yaml
@@ -17,6 +17,7 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical-inf-size-1.yaml b/tests/configs/infer-test-categorical-inf-size-1.yaml
index 37280c9c..f1632ef5 100644
--- a/tests/configs/infer-test-categorical-inf-size-1.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-1.yaml
@@ -16,6 +16,7 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
index 6f0e3085..6fdbb4e6 100644
--- a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
@@ -19,6 +19,7 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-inf-size-3.yaml b/tests/configs/infer-test-categorical-inf-size-3.yaml
index 2a99c2be..0f8f6bff 100644
--- a/tests/configs/infer-test-categorical-inf-size-3.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3.yaml
@@ -19,6 +19,7 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-multitarget.yaml b/tests/configs/infer-test-categorical-multitarget.yaml
index 29c4d8d8..4b0c13d1 100644
--- a/tests/configs/infer-test-categorical-multitarget.yaml
+++ b/tests/configs/infer-test-categorical-multitarget.yaml
@@ -19,6 +19,7 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical.yaml b/tests/configs/infer-test-categorical.yaml
index eeca3e19..6b8d52cd 100644
--- a/tests/configs/infer-test-categorical.yaml
+++ b/tests/configs/infer-test-categorical.yaml
@@ -17,6 +17,7 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-distributed-parquet.yaml b/tests/configs/infer-test-distributed-parquet.yaml
index 6e01ac6a..c9716b2a 100644
--- a/tests/configs/infer-test-distributed-parquet.yaml
+++ b/tests/configs/infer-test-distributed-parquet.yaml
@@ -18,6 +18,7 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-distributed.yaml b/tests/configs/infer-test-distributed.yaml
index 2e468055..a344c553 100644
--- a/tests/configs/infer-test-distributed.yaml
+++ b/tests/configs/infer-test-distributed.yaml
@@ -20,4 +20,5 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 2
diff --git a/tests/configs/infer-test-lazy.yaml b/tests/configs/infer-test-lazy.yaml
index 472b1dfa..aafbe379 100644
--- a/tests/configs/infer-test-lazy.yaml
+++ b/tests/configs/infer-test-lazy.yaml
@@ -18,4 +18,5 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 2
diff --git a/tests/configs/infer-test-real-autoregression.yaml b/tests/configs/infer-test-real-autoregression.yaml
index 72c4128e..357dc11b 100644
--- a/tests/configs/infer-test-real-autoregression.yaml
+++ b/tests/configs/infer-test-real-autoregression.yaml
@@ -17,6 +17,7 @@ output_probabilities: false
 map_to_id: false
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: false
 
diff --git a/tests/configs/infer-test-real.yaml b/tests/configs/infer-test-real.yaml
index 009823e0..bc35d9e4 100644
--- a/tests/configs/infer-test-real.yaml
+++ b/tests/configs/infer-test-real.yaml
@@ -15,6 +15,7 @@ output_probabilities: false
 map_to_id: false
 device: cpu
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/train-test-categorical-bert.yaml b/tests/configs/train-test-categorical-bert.yaml
index 12adcc5a..abf8e2d4 100644
--- a/tests/configs/train-test-categorical-bert.yaml
+++ b/tests/configs/train-test-categorical-bert.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,6 +33,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
+  window_length: 9
   training_objective: bert
   bert_spec:
     masking_probability: 0.5
diff --git a/tests/configs/train-test-categorical-inf-size-1.yaml b/tests/configs/train-test-categorical-inf-size-1.yaml
index 92cae64c..259252dc 100644
--- a/tests/configs/train-test-categorical-inf-size-1.yaml
+++ b/tests/configs/train-test-categorical-inf-size-1.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -31,6 +32,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-inf-size-3.yaml b/tests/configs/train-test-categorical-inf-size-3.yaml
index c20d9ac0..33644d6d 100644
--- a/tests/configs/train-test-categorical-inf-size-3.yaml
+++ b/tests/configs/train-test-categorical-inf-size-3.yaml
@@ -11,6 +11,7 @@ target_column_types:
   supCat2: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -38,6 +39,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 10000.0
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-multitarget-eager.yaml b/tests/configs/train-test-categorical-multitarget-eager.yaml
index 55d55d83..e2212baf 100644
--- a/tests/configs/train-test-categorical-multitarget-eager.yaml
+++ b/tests/configs/train-test-categorical-multitarget-eager.yaml
@@ -10,6 +10,7 @@ target_column_types:
   supCat1: categorical
   supReal3: real
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -38,6 +39,7 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-multitarget.yaml b/tests/configs/train-test-categorical-multitarget.yaml
index 6066730a..c8d32a20 100644
--- a/tests/configs/train-test-categorical-multitarget.yaml
+++ b/tests/configs/train-test-categorical-multitarget.yaml
@@ -11,6 +11,7 @@ target_column_types:
   supCat1: categorical
   supReal3: real
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -39,6 +40,7 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical.yaml b/tests/configs/train-test-categorical.yaml
index eafdbd8d..36dcff1c 100644
--- a/tests/configs/train-test-categorical.yaml
+++ b/tests/configs/train-test-categorical.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,6 +33,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "outer"
diff --git a/tests/configs/train-test-distributed-lazy-parquet.yaml b/tests/configs/train-test-distributed-lazy-parquet.yaml
index 871fa0c1..576bcee4 100644
--- a/tests/configs/train-test-distributed-lazy-parquet.yaml
+++ b/tests/configs/train-test-distributed-lazy-parquet.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -25,6 +26,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-distributed.yaml b/tests/configs/train-test-distributed.yaml
index 23e92988..923fbf4a 100644
--- a/tests/configs/train-test-distributed.yaml
+++ b/tests/configs/train-test-distributed.yaml
@@ -11,6 +11,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -27,6 +28,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-lazy.yaml b/tests/configs/train-test-lazy.yaml
index e53b9e2f..1d2ce592 100644
--- a/tests/configs/train-test-lazy.yaml
+++ b/tests/configs/train-test-lazy.yaml
@@ -11,6 +11,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -27,6 +28,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-real-bert.yaml b/tests/configs/train-test-real-bert.yaml
index 426a129c..4e915730 100644
--- a/tests/configs/train-test-real-bert.yaml
+++ b/tests/configs/train-test-real-bert.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemValue: real
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,6 +33,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
+  window_length: 9
   training_objective: bert
   bert_spec:
     masking_probability: 0.5
diff --git a/tests/configs/train-test-real.yaml b/tests/configs/train-test-real.yaml
index 0b02444a..2596a365 100644
--- a/tests/configs/train-test-real.yaml
+++ b/tests/configs/train-test-real.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemValue: real
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,6 +33,7 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "inner"
diff --git a/tests/configs/train-test-resume-epoch.yaml b/tests/configs/train-test-resume-epoch.yaml
index ce93f148..da990807 100644
--- a/tests/configs/train-test-resume-epoch.yaml
+++ b/tests/configs/train-test-resume-epoch.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemValue: real
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: false
@@ -32,6 +33,7 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-resume-mid-epoch.yaml b/tests/configs/train-test-resume-mid-epoch.yaml
index 6398cb81..b76a474c 100644
--- a/tests/configs/train-test-resume-mid-epoch.yaml
+++ b/tests/configs/train-test-resume-mid-epoch.yaml
@@ -9,6 +9,7 @@ target_column_types:
   itemId: categorical
 
 seq_length: 8
+window_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,6 +33,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
+  window_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "outer"
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index 30588329..3adcf402 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -410,7 +410,7 @@ def copy_interrupted_data(project_root, remove_project_root_contents):
         project_root, "data", "test-data-categorical-1-interrupted-temp"
     )
 
-    shutil.copytree(source_path, target_path)
+    shutil.copytree(source_path, target_path, dirs_exist_ok=True)
 
 
 @pytest.fixture(scope="session")
@@ -447,7 +447,7 @@ def run_preprocessing(
 
     source_path = os.path.join("tests", "resources", "source_configs", "id_maps")
     target_path = os.path.join(project_root, "configs", "id_maps")
-    shutil.copytree(source_path, target_path)
+    shutil.copytree(source_path, target_path, dirs_exist_ok=True)
     run_and_log(
         f"sequifier preprocess --config-path {preprocessing_config_path_cat_multitarget}"
     )
diff --git a/tests/integration/test_make.py b/tests/integration/test_make.py
index d4b3cd26..649b3903 100644
--- a/tests/integration/test_make.py
+++ b/tests/integration/test_make.py
@@ -88,6 +88,7 @@ def adapt_configs(config_strings):
             .replace("epochs: 10", "epochs: 3")
             .replace("device: cuda", "device: cpu")
             .replace("seq_length: 48", "seq_length: 10")
+            .replace("window_length: 49", "window_length: 11")
             .replace("total_steps: PLEASE FILL", "total_steps: 10000")
         )
 
@@ -118,6 +119,7 @@ def adapt_configs(config_strings):
             .replace("[EXAMPLE_INPUT_COLUMN_NAME]", "[itemId]")
             .replace("[EXAMPLE_TARGET_COLUMN_NAME]", "[itemId]")
             .replace("seq_length: 48", "seq_length: 10")
+            .replace("window_length: 49", "window_length: 11")
             .replace("autoregression: true", "autoregression: false")
             .replace("EXAMPLE_TARGET_COLUMN_NAME: real", "itemId: categorical")
         )
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index e26d3d8e..d90fe0ff 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -28,6 +28,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
         n_classes={"cat_col": 6},
         seq_length=seq_length,
+        window_length=5,
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -71,6 +72,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
             loss_weights={"cat_col": 1.0, "real_col": 1.0},
             torch_compile="none",
             layer_autocast=False,
+            window_length=5,
         ),
     )
     model = TransformerModel(config)
@@ -126,6 +128,7 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
         },
         n_classes={"cat2": 7, "cat10": 6},
         seq_length=seq_length,
+        window_length=5,
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -160,6 +163,7 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             loss_weights={"cat2": 1.0},
             torch_compile="none",
             layer_autocast=False,
+            window_length=5,
         ),
     )
     model = TransformerModel(config)
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 95061cf3..54cb4839 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -24,21 +24,22 @@ def metadata_configs(run_preprocessing, project_root):
 
 
 def test_metadata_config(metadata_configs):
+    expected_metadata_keys = [
+        "n_classes",
+        "id_maps",
+        "special_token_ids",
+        "split_paths",
+        "column_types",
+        "selected_columns_statistics",
+        "seq_length",
+        "target_max_offset",
+        "window_length",
+        "sequence_layout_version",
+    ]
+
     for file_name, metadata_config in metadata_configs.items():
         print(f"Verifying metadata_config for: {file_name}")
-        assert np.all(
-            np.array(list(metadata_config.keys()))
-            == np.array(
-                [
-                    "n_classes",
-                    "id_maps",
-                    "special_token_ids",
-                    "split_paths",
-                    "column_types",
-                    "selected_columns_statistics",
-                ]
-            )
-        ), list(metadata_config.keys())
+        assert list(metadata_config.keys()) == expected_metadata_keys
         assert metadata_config["special_token_ids"] == {
             "[unknown]": 0,
             "[other]": 1,
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
new file mode 100644
index 00000000..d8212897
--- /dev/null
+++ b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
@@ -0,0 +1,11 @@
+{
+    "seq_length": 8,
+    "target_max_offset": 1,
+    "window_length": 9,
+    "sequence_layout_version": 2,
+    "selected_columns": null,
+    "data_columns": [
+        "itemId"
+    ],
+    "write_format": "pt"
+}
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index dff33ff3..0c3feb6b 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -16,6 +16,7 @@ def mock_config():
     config.project_root = "."
     config.seed = 42
     config.seq_length = 2
+    config.window_length = 3
     config.column_types = {"item": "Float64"}
     config.training_spec.batch_size = 5
     config.training_spec.num_workers = 0
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 2c697037..6bea73f4 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -20,6 +20,7 @@ def mock_config(tmp_path):
     config.training_spec.num_workers = 0
     config.seed = 42
     config.seq_length = 5
+    config.window_length = 6
     config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
     config.training_spec.data_offset = 1
@@ -57,8 +58,7 @@ def mock_torch_load():
     with patch("torch.load") as mock_load:
 
         def side_effect(path, map_location, weights_only):
-            # Tensors size: (10 samples per file, sequence length 10 to allow slicing)
-            dummy_seq = {"col1": torch.ones((10, 10)), "tgt1": torch.zeros((10, 10))}
+            dummy_seq = {"col1": torch.ones((10, 6)), "tgt1": torch.zeros((10, 6))}
 
             return (
                 dummy_seq,
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index dd1a23c4..cd111b6d 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -8,6 +8,7 @@
     build_valid_mask,
     construct_index_maps,
     numpy_to_pytorch,
+    slice_window,
 )
 
 # ==========================================
@@ -86,7 +87,13 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     seq_length = 3
 
     tensors, metadata = numpy_to_pytorch(
-        data, column_types, all_columns, seq_length, data_offset=1, target_offset=0
+        data,
+        column_types,
+        all_columns,
+        seq_length,
+        seq_length + 1,
+        data_offset=1,
+        target_offset=0,
     )
 
     # 1. Check Keys
@@ -122,6 +129,7 @@ def test_numpy_to_pytorch_dtypes():
         {"int_col": torch.int64},
         ["int_col"],
         seq_length=1,
+        window_length=2,
         data_offset=1,
         target_offset=0,
     )
@@ -134,6 +142,7 @@ def test_numpy_to_pytorch_dtypes():
         {"float_col": torch.float32},
         ["float_col"],
         seq_length=1,
+        window_length=2,
         data_offset=1,
         target_offset=0,
     )
@@ -172,6 +181,18 @@ def test_build_valid_mask_from_left_pad_lengths():
     )
 
 
+def test_slice_window_validates_stored_width():
+    tensor = torch.arange(8).reshape(2, 4)
+
+    assert torch.equal(
+        slice_window(tensor, seq_length=3, offset=1),
+        torch.tensor([[0, 1, 2], [4, 5, 6]]),
+    )
+    assert torch.equal(
+        slice_window(tensor[:, :3], seq_length=3, offset=0), tensor[:, :3]
+    )
+
+
 def test_numpy_to_pytorch_includes_explicit_padding_masks():
     data = pl.DataFrame(
         {
@@ -192,6 +213,7 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
         {"A": torch.float32},
         ["A"],
         seq_length=3,
+        window_length=4,
         data_offset=1,
         target_offset=0,
     )
diff --git a/tests/unit/test_hyperparameter_search_config.py b/tests/unit/test_hyperparameter_search_config.py
index b09eed26..7b9103d7 100644
--- a/tests/unit/test_hyperparameter_search_config.py
+++ b/tests/unit/test_hyperparameter_search_config.py
@@ -62,14 +62,19 @@ def bert_spec_sampling_config():
     }
 
 
+def sample_training_spec(sampling, trial):
+    return sampling.sample_trial(trial, target_max_offset=1, window_length=9)
+
+
 def test_training_objective_defaults_to_causal_without_bert_spec():
     sampling = make_training_sampling()
     trial = RecordingTrial()
 
-    training_spec = sampling.sample_trial(trial)
+    training_spec = sample_training_spec(sampling, trial)
 
     assert training_spec.training_objective == "causal"
     assert training_spec.bert_spec is None
+    assert training_spec.window_length == 9
     assert trial.params["training_objective"] == "causal"
 
 
@@ -87,9 +92,10 @@ def test_bert_spec_fields_are_sampled_separately_for_bert_objective():
         }
     )
 
-    training_spec = sampling.sample_trial(trial)
+    training_spec = sample_training_spec(sampling, trial)
 
     assert training_spec.training_objective == "bert"
+    assert training_spec.bert_spec is not None
     assert training_spec.bert_spec.masking_probability == 0.30
     assert training_spec.bert_spec.replacement_distribution.masked == 0.6
     assert training_spec.bert_spec.replacement_distribution.random == 0.2
@@ -112,7 +118,7 @@ def test_causal_objective_does_not_sample_bert_fields():
     )
     trial = RecordingTrial({"training_objective": "causal"})
 
-    training_spec = sampling.sample_trial(trial)
+    training_spec = sample_training_spec(sampling, trial)
 
     assert training_spec.training_objective == "causal"
     assert training_spec.bert_spec is None
@@ -131,7 +137,7 @@ def test_bert_training_spec_dumps_to_plain_yaml():
         training_objective=["bert"],
         bert_spec=bert_spec_sampling_config(),
     )
-    training_spec = sampling.sample_trial(RecordingTrial())
+    training_spec = sample_training_spec(sampling, RecordingTrial())
 
     dumped = yaml.dump(training_spec, Dumper=TrainModelDumper, sort_keys=False)
     loaded = yaml.safe_load(dumped)
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 9bfdaa63..ccf5e2ae 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -174,6 +174,7 @@ def test_infer_config_defaults_bert_prediction_length_to_seq_length():
         device="cpu",
         prediction_length=None,
         seq_length=3,
+        window_length=4,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -201,6 +202,7 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
         device="cpu",
         prediction_length=None,
         seq_length=3,
+        window_length=4,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -229,6 +231,7 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             device="cpu",
             prediction_length=1,
             seq_length=3,
+            window_length=4,
             inference_batch_size=2,
             output_probabilities=False,
             map_to_id=False,
@@ -387,6 +390,7 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         device="cpu",
         prediction_length=None,
         seq_length=3,
+        window_length=4,
         inference_batch_size=4,
         output_probabilities=False,
         map_to_id=True,
@@ -594,6 +598,7 @@ def ar_config():
         device="cpu",
         prediction_length=1,
         seq_length=3,
+        window_length=4,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,  # Set to False to bypass ID mapping requirements
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index e18caaf4..6000208e 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -7,6 +7,7 @@
 from pydantic import ValidationError
 
 from sequifier.config.preprocess_config import PreprocessorModel
+from sequifier.helpers import SequenceLayout
 from sequifier.preprocess import (
     Preprocessor,
     _apply_column_statistics,
@@ -40,7 +41,11 @@ def test_extract_subsequences_basic():
     # Windows: [10,11,12,13], [11,12,13,14], [12,13,14,15]
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data, seq_length, stride, columns, subsequence_start_mode="distribute"
+        input_data,
+        seq_length + 1,
+        stride,
+        columns,
+        subsequence_start_mode="distribute",
     )
 
     assert len(result["col1"]) == 3
@@ -48,17 +53,36 @@ def test_extract_subsequences_basic():
     assert result["col1"][2] == [12, 13, 14, 15]
 
 
+def test_extract_subsequences_bert_width():
+    input_data = {"col1": [10, 11, 12, 13, 14, 15]}
+    seq_length = 3
+    stride = 1
+    columns = ["col1"]
+
+    result, left_pad_lengths, subsequence_starts = extract_subsequences(
+        input_data,
+        window_length=seq_length,
+        stride_for_split=stride,
+        columns=columns,
+        subsequence_start_mode="distribute",
+    )
+
+    assert len(result["col1"]) == 4
+    assert result["col1"][0] == [10, 11, 12]
+    assert result["col1"][3] == [13, 14, 15]
+
+
 def test_extract_subsequences_padding():
     """Tests that sequences shorter than seq_length are padded with 0s."""
     input_data = {"col1": [1, 2]}  # Length 2
-    seq_length = 4  # Req length 5 (4+1)
+    window_length = 5
     stride = 1
     columns = ["col1"]
 
     # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data, seq_length, stride, columns, subsequence_start_mode="distribute"
+        input_data, window_length, stride, columns, subsequence_start_mode="distribute"
     )
 
     assert len(result["col1"]) == 1
@@ -69,7 +93,7 @@ def test_extract_subsequences_returns_left_pad_lengths_when_requested():
     input_data = {"col1": [0.0, 1.5]}
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        seq_length=4,
+        window_length=5,
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -103,7 +127,7 @@ def test_extract_sequences_persists_left_pad_length_metadata():
     sequences = extract_sequences(
         data,
         schema,
-        seq_length=4,
+        layout=SequenceLayout(seq_length=4),
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -130,7 +154,7 @@ def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
 
     process_and_write_data_pt(
         data,
-        seq_length=3,
+        window_length=4,
         path=str(out_path),
         column_types={"col1": "Float64"},
     )
@@ -426,13 +450,14 @@ def test_extract_subsequences_modes(mode):
     # exact: strictly adheres to stride, throws error if misalignment.
     input_data = {"col1": list(range(10))}
     seq_length = 2
+    window_length = seq_length + 1
     columns = ["col1"]
 
     if mode == "distribute":
         stride = 4
         # distribute might adjust indices to maximize coverage
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data, seq_length, stride, columns, mode
+            input_data, window_length, stride, columns, mode
         )
         assert len(result["col1"]) > 0
 
@@ -442,13 +467,13 @@ def test_extract_subsequences_modes(mode):
         )
         # Testing a failing exact case
         with pytest.raises(ValueError):
-            extract_subsequences(input_data, seq_length, 4, columns, mode)
+            extract_subsequences(input_data, window_length, 4, columns, mode)
 
         # Testing a passing exact case
         # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
         input_data_exact = {"col1": list(range(11))}
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data_exact, seq_length, 4, columns, mode
+            input_data_exact, window_length, 4, columns, mode
         )
         assert len(result["col1"]) > 0
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 50d7e106..74292545 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -31,6 +31,7 @@ def _training_spec_kwargs(**overrides):
         "optimizer": {"name": "Adam"},
         "scheduler": {"name": "StepLR", "step_size": 1, "gamma": 0.1},
         "loss_weights": {"cat_col": 1.0, "real_col": 1.0},
+        "window_length": 11,
     }
     values.update(overrides)
     return values
@@ -117,6 +118,7 @@ def model_config(tmp_path):
         optimizer={"name": "Adam"},
         scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
         loss_weights={"cat_col": 1.0, "real_col": 1.0},
+        window_length=11,
     )
 
     config = TrainModel(
@@ -135,6 +137,7 @@ def model_config(tmp_path):
         id_maps={"cat_col": {"a": 1, "b": 2, "c": 3, "d": 4}},
         n_classes={"cat_col": 5},  # 0 + 4 classes
         seq_length=10,
+        window_length=11,
         inference_batch_size=4,
         seed=42,
         export_generative_model=True,

From 0265aa370fbd36cdb5064c3fed9457766783bc8c Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 13:42:39 +0200
Subject: [PATCH 41/81] Rename seq_length -> context_length target_max_offset
 -> max_lookahead window_length -> sample_length

---
 README.md                                     |   2 +-
 .../configs/hyperparameter-search.md          |   4 +-
 documentation/configs/infer.md                |   2 +-
 documentation/configs/preprocess.md           |  10 +-
 documentation/configs/train.md                |   2 +-
 documentation/consolidated-docs.md            |  20 ++--
 .../demos/self-contained-example.ipynb        |   6 +-
 .../config/hyperparameter_search_config.py    |  30 ++---
 src/sequifier/config/infer_config.py          |  38 +++----
 src/sequifier/config/preprocess_config.py     |  12 +-
 src/sequifier/config/train_config.py          |  52 +++++----
 src/sequifier/helpers.py                      | 104 +++++++++---------
 src/sequifier/infer.py                        |  64 +++++------
 .../io/sequifier_dataset_from_file.py         |   4 +-
 .../sequifier_dataset_from_folder_parquet.py  |  14 +--
 ...uifier_dataset_from_folder_parquet_lazy.py |  12 +-
 .../io/sequifier_dataset_from_folder_pt.py    |  14 +--
 .../sequifier_dataset_from_folder_pt_lazy.py  |  14 +--
 src/sequifier/make.py                         |  14 +--
 src/sequifier/model/layers.py                 |  10 +-
 src/sequifier/preprocess.py                   |  74 ++++++-------
 src/sequifier/train.py                        |  66 +++++------
 .../hyperparameter-search-bayesian.yaml       |   2 +-
 tests/configs/hyperparameter-search-bert.yaml |   2 +-
 ...arameter-search-custom-eval-inference.yaml |   2 +-
 .../hyperparameter-search-custom-eval.yaml    |   2 +-
 tests/configs/hyperparameter-search-grid.yaml |   2 +-
 .../configs/hyperparameter-search-sample.yaml |   2 +-
 ...infer-test-categorical-autoregression.yaml |   4 +-
 ...infer-test-categorical-bert-embedding.yaml |   4 +-
 .../configs/infer-test-categorical-bert.yaml  |   4 +-
 .../infer-test-categorical-embedding.yaml     |   4 +-
 .../infer-test-categorical-inf-size-1.yaml    |   4 +-
 ...test-categorical-inf-size-3-embedding.yaml |   4 +-
 .../infer-test-categorical-inf-size-3.yaml    |   4 +-
 .../infer-test-categorical-multitarget.yaml   |   4 +-
 tests/configs/infer-test-categorical.yaml     |   4 +-
 .../infer-test-distributed-parquet.yaml       |   4 +-
 tests/configs/infer-test-distributed.yaml     |   4 +-
 tests/configs/infer-test-lazy.yaml            |   4 +-
 .../infer-test-real-autoregression.yaml       |   4 +-
 tests/configs/infer-test-real.yaml            |   4 +-
 .../preprocess-test-categorical-exact-pt.yaml |   2 +-
 .../preprocess-test-categorical-exact.yaml    |   2 +-
 ...eprocess-test-categorical-interrupted.yaml |   2 +-
 ...eprocess-test-categorical-multitarget.yaml |   2 +-
 ...ategorical-precomputed-stats-negative.yaml |   2 +-
 ...ss-test-categorical-precomputed-stats.yaml |   2 +-
 .../configs/preprocess-test-categorical.yaml  |   2 +-
 tests/configs/preprocess-test-multi-file.yaml |   2 +-
 tests/configs/preprocess-test-real.yaml       |   2 +-
 .../configs/train-test-categorical-bert.yaml  |   6 +-
 .../train-test-categorical-inf-size-1.yaml    |   6 +-
 .../train-test-categorical-inf-size-3.yaml    |   6 +-
 ...in-test-categorical-multitarget-eager.yaml |   6 +-
 .../train-test-categorical-multitarget.yaml   |   6 +-
 tests/configs/train-test-categorical.yaml     |   6 +-
 .../train-test-distributed-lazy-parquet.yaml  |   6 +-
 tests/configs/train-test-distributed.yaml     |   6 +-
 tests/configs/train-test-lazy.yaml            |   6 +-
 tests/configs/train-test-real-bert.yaml       |   6 +-
 tests/configs/train-test-real.yaml            |   6 +-
 tests/configs/train-test-resume-epoch.yaml    |   6 +-
 .../configs/train-test-resume-mid-epoch.yaml  |   6 +-
 tests/integration/test_inference.py           |   2 +-
 tests/integration/test_make.py                |  10 +-
 tests/integration/test_onnx_export.py         |  20 ++--
 tests/integration/test_preprocessing.py       |   6 +-
 .../preprocess-manifest.json                  |   6 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |   4 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |   4 +-
 tests/unit/test_helpers.py                    |  30 ++---
 .../unit/test_hyperparameter_search_config.py |   4 +-
 tests/unit/test_infer.py                      |  30 ++---
 tests/unit/test_preprocess.py                 |  40 +++----
 tests/unit/test_train.py                      |  30 ++---
 tools/split_file.py                           |  11 +-
 77 files changed, 471 insertions(+), 458 deletions(-)

diff --git a/README.md b/README.md
index 30617973..8aa1c7f9 100644
--- a/README.md
+++ b/README.md
@@ -97,7 +97,7 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `seq_length + target_max_offset` (`target_max_offset` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `context_length + max_lookahead` (`max_lookahead` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
 
 |sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
 |----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
diff --git a/documentation/configs/hyperparameter-search.md b/documentation/configs/hyperparameter-search.md
index bc890ed1..23402080 100644
--- a/documentation/configs/hyperparameter-search.md
+++ b/documentation/configs/hyperparameter-search.md
@@ -59,7 +59,7 @@ Sequifier allows you to search not just for model parameters, but for the best *
 | --- | --- | --- | --- |
 | `input_columns` | `list[list[str]]` | **Yes** | A list of input sets. E.g., `[['col1'], ['col1', 'col2']]`. |
 | `target_columns` | `list[str]` | **Yes** | The target column(s) to predict. Fixed across all runs. |
-| `seq_length` | `list[int]` | **Yes** | List of sequence lengths to test (e.g., `[24, 48]`). |
+| `context_length` | `list[int]` | **Yes** | List of sequence lengths to test (e.g., `[24, 48]`). |
 | `target_column_types` | `dict` | **Yes** | Map of target columns to `categorical` or `real`. |
 | `column_types` | `list[dict]` | *Conditional* | Required if `input_columns` varies. List of type maps corresponding to the input sets. |
 
@@ -186,7 +186,7 @@ All other parameters are considered **Independent**. Sequifier will test every v
 
   * **Model:** `num_layers`, `dim_feedforward`, `activation_fn`, `normalization`, `norm_first`, `positional_encoding`, `attention_type`, `rope_theta`.
   * **Training:** `batch_size`, `dropout`, `accumulation_steps`, `optimizer`.
-  * **Data:** `seq_length`.
+  * **Data:** `context_length`.
 
 ### 3\. Special Case: `n_kv_heads`
 
diff --git a/documentation/configs/infer.md b/documentation/configs/infer.md
index 0ec62861..6a80934e 100644
--- a/documentation/configs/infer.md
+++ b/documentation/configs/infer.md
@@ -40,7 +40,7 @@ These fields tell the inference engine which columns to extract from the new dat
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `model_type` | `str` | **Yes** | - | `generative` (predict next value) or `embedding` (extract vector representation). |
-| `seq_length` | `int` | **Yes** | - | The context window size. Must match training. |
+| `context_length` | `int` | **Yes** | - | The context window size. Must match training. |
 | `prediction_length` | `int` | No | `1` | Number of steps to predict *simultaneously*. **Must be 1** if `autoregression: true`. |
 | `inference_batch_size`| `int` | **Yes** | - | Number of sequences to process at once. |
 | `autoregression` | `bool` | No | `false` | If `true`, feeds predictions back into the model to predict further into the future. |
diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index 29920541..8185baec 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -45,10 +45,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
-| `seq_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
-| `target_max_offset` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `seq_length + target_max_offset`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
+| `context_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
+| `max_lookahead` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `context_length + max_lookahead`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
-| `stride_by_split` | `list[int]` | No | `[seq_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
+| `stride_by_split` | `list[int]` | No | `[context_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
 
 ### 4\. Performance & System
@@ -73,10 +73,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 This controls data augmentation and redundancy.
 
-  * **Stride = `seq_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
+  * **Stride = `context_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
   * **Stride = 1 (Maximum Overlap):** Maximizes data volume. The model sees every possible sequence. This yields the highest accuracy but significantly increases the size of the preprocessed data and training time.
   * **Hybrid Approach:** It is common practice to set a large stride for the training and validation splits (index 0) to reduce the size on disk of the dataset, and a stride=1 for the test split to evaluate the model on each point in the test set. This supposes that the test split value is low.
-      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `seq_length: 48`).
+      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `context_length: 48`).
 
 ### 3\. `subsequence_start_mode`: `distribute` vs `exact`
 
diff --git a/documentation/configs/train.md b/documentation/configs/train.md
index 1e929b0b..48df255a 100644
--- a/documentation/configs/train.md
+++ b/documentation/configs/train.md
@@ -30,7 +30,7 @@ The configuration is defined in a YAML file (e.g., `train.yaml`). The file is st
 | `target_columns` | `list[str]`| **Yes** | - | The specific column(s) the model should learn to predict. |
 | `target_column_types`| `dict` | **Yes** | - | Map of target columns to their type: `'categorical'` or `'real'`. The key order in target_column_types must exactly match the list order in target_columns |
 | `input_columns` | `list[str]`| No | All | Subset of columns to use as input features. Defaults to all available in metadata. |
-| `seq_length` | `int` | **Yes** | - | Must match the `seq_length` used in preprocessing. |
+| `context_length` | `int` | **Yes** | - | Must match the `context_length` used in preprocessing. |
 
 ### 3\. Model Architecture (`model_spec`)
 
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 241983bc..0baf670c 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -97,7 +97,7 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `seq_length + target_max_offset` (`target_max_offset` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `context_length + max_lookahead` (`max_lookahead` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
 
 |sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
 |----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
@@ -253,10 +253,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
-| `seq_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
-| `target_max_offset` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `seq_length + target_max_offset`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
+| `context_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
+| `max_lookahead` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `context_length + max_lookahead`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
-| `stride_by_split` | `list[int]` | No | `[seq_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
+| `stride_by_split` | `list[int]` | No | `[context_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
 
 ### 4\. Performance & System
@@ -281,10 +281,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 This controls data augmentation and redundancy.
 
-  * **Stride = `seq_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
+  * **Stride = `context_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
   * **Stride = 1 (Maximum Overlap):** Maximizes data volume. The model sees every possible sequence. This yields the highest accuracy but significantly increases the size of the preprocessed data and training time.
   * **Hybrid Approach:** It is common practice to set a large stride for the training and validation splits (index 0) to reduce the size on disk of the dataset, and a stride=1 for the test split to evaluate the model on each point in the test set. This supposes that the test split value is low.
-      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `seq_length: 48`).
+      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `context_length: 48`).
 
 ### 3\. `subsequence_start_mode`: `distribute` vs `exact`
 
@@ -379,7 +379,7 @@ The configuration is defined in a YAML file (e.g., `train.yaml`). The file is st
 | `target_columns` | `list[str]`| **Yes** | - | The specific column(s) the model should learn to predict. |
 | `target_column_types`| `dict` | **Yes** | - | Map of target columns to their type: `'categorical'` or `'real'`. The key order in target_column_types must exactly match the list order in target_columns |
 | `input_columns` | `list[str]`| No | All | Subset of columns to use as input features. Defaults to all available in metadata. |
-| `seq_length` | `int` | **Yes** | - | Must match the `seq_length` used in preprocessing. |
+| `context_length` | `int` | **Yes** | - | Must match the `context_length` used in preprocessing. |
 
 ### 3\. Model Architecture (`model_spec`)
 
@@ -604,7 +604,7 @@ These fields tell the inference engine which columns to extract from the new dat
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `model_type` | `str` | **Yes** | - | `generative` (predict next value) or `embedding` (extract vector representation). |
-| `seq_length` | `int` | **Yes** | - | The context window size. Must match training. |
+| `context_length` | `int` | **Yes** | - | The context window size. Must match training. |
 | `prediction_length` | `int` | No | `1` | Number of steps to predict *simultaneously*. **Must be 1** if `autoregression: true`. |
 | `inference_batch_size`| `int` | **Yes** | - | Number of sequences to process at once. |
 | `autoregression` | `bool` | No | `false` | If `true`, feeds predictions back into the model to predict further into the future. |
@@ -788,7 +788,7 @@ Sequifier allows you to search not just for model parameters, but for the best *
 | --- | --- | --- | --- |
 | `input_columns` | `list[list[str]]` | **Yes** | A list of input sets. E.g., `[['col1'], ['col1', 'col2']]`. |
 | `target_columns` | `list[str]` | **Yes** | The target column(s) to predict. Fixed across all runs. |
-| `seq_length` | `list[int]` | **Yes** | List of sequence lengths to test (e.g., `[24, 48]`). |
+| `context_length` | `list[int]` | **Yes** | List of sequence lengths to test (e.g., `[24, 48]`). |
 | `target_column_types` | `dict` | **Yes** | Map of target columns to `categorical` or `real`. |
 | `column_types` | `list[dict]` | *Conditional* | Required if `input_columns` varies. List of type maps corresponding to the input sets. |
 
@@ -915,7 +915,7 @@ All other parameters are considered **Independent**. Sequifier will test every v
 
   * **Model:** `num_layers`, `dim_feedforward`, `activation_fn`, `normalization`, `norm_first`, `positional_encoding`, `attention_type`, `rope_theta`.
   * **Training:** `batch_size`, `dropout`, `accumulation_steps`, `optimizer`.
-  * **Data:** `seq_length`.
+  * **Data:** `context_length`.
 
 ### 3\. Special Case: `n_kv_heads`
 
diff --git a/documentation/demos/self-contained-example.ipynb b/documentation/demos/self-contained-example.ipynb
index 27cc4893..80c2a59b 100644
--- a/documentation/demos/self-contained-example.ipynb
+++ b/documentation/demos/self-contained-example.ipynb
@@ -173,7 +173,7 @@
     "    \"write_format\": \"parquet\",\n",
     "    \"selected_columns\": [\"targetToken\"],\n",
     "    \"split_ratios\": [0.8, 0.1, 0.1],  # Train, Val, Test\n",
-    "    \"seq_length\": 10,  # Lookback window\n",
+    "    \"context_length\": 10,  # Lookback window\n",
     "    \"stride_by_split\": [1, 1, 1],  # Dense sampling\n",
     "    \"max_rows\": None,\n",
     "    \"seed\": 42,\n",
@@ -199,7 +199,7 @@
     "    \"input_columns\": [\"targetToken\"],\n",
     "    \"target_columns\": [\"targetToken\"],\n",
     "    \"target_column_types\": {\"targetToken\": \"categorical\"},\n",
-    "    \"seq_length\": 10,\n",
+    "    \"context_length\": 10,\n",
     "    \"inference_batch_size\": 10,\n",
     "    \"export_generative_model\": True,\n",
     "    \"export_embedding_model\": False,\n",
@@ -257,7 +257,7 @@
     "    \"output_probabilities\": False,\n",
     "    \"map_to_id\": True,  # Map integer classes back to original values (e.g., \"1\", \"2\")\n",
     "    \"device\": \"cuda\" if torch.cuda.is_available() else \"cpu\",\n",
-    "    \"seq_length\": 10,\n",
+    "    \"context_length\": 10,\n",
     "    \"inference_batch_size\": 10,\n",
     "    \"autoregression\": True,\n",
     "    \"autoregression_total_steps\": 5,  # Predict 5 steps into the future\n",
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 74080dcf..6e0282c0 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -416,7 +416,7 @@ def validate_scheduler_config(cls, v, info_dict):
         return v
 
     def sample_trial(
-        self, trial: Any, target_max_offset: int, window_length: int
+        self, trial: Any, max_lookahead: int, sample_length: int
     ) -> TrainingSpecModel:
         """Samples training hyperparameters using an Optuna trial.
 
@@ -500,8 +500,8 @@ def sample_trial(
             fsdp_cpu_offload=self.fsdp_cpu_offload,
             torch_compile=self.torch_compile,
             float32_matmul_precision=self.float32_matmul_precision,
-            target_max_offset=target_max_offset,
-            window_length=window_length,
+            max_lookahead=max_lookahead,
+            sample_length=sample_length,
         )
 
 
@@ -669,7 +669,7 @@ class HyperparameterSearchConfig(BaseModel):
         target_columns: The list of target columns for model training.
         target_column_types: A dictionary mapping target columns to their types.
         id_maps: A dictionary mapping categorical values to their indexed representation.
-        seq_length: A list of possible sequence lengths.
+        context_length: A list of possible sequence lengths.
         n_classes: The number of classes for each categorical column.
         inference_batch_size: The batch size for inference.
         export_onnx: If True, exports the model in ONNX format.
@@ -702,8 +702,8 @@ class HyperparameterSearchConfig(BaseModel):
     target_column_types: dict[str, str]
     id_maps: dict[str, dict[str | int, int]]
 
-    seq_length: list[int]
-    target_max_offset: int = Field(default=1, ge=0)
+    context_length: list[int]
+    max_lookahead: int = Field(default=1, ge=0)
     n_classes: dict[str, int]
     inference_batch_size: int
 
@@ -825,15 +825,17 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
         input_columns_index = trial.suggest_categorical(
             "input_columns_index", list(range(len(self.input_columns)))
         )
-        seq_length = trial.suggest_categorical("seq_length", self.seq_length)
-        window_length = seq_length + self.target_max_offset
+        context_length = trial.suggest_categorical(
+            "context_length", self.context_length
+        )
+        sample_length = context_length + self.max_lookahead
         training_spec = self.training_hyperparameter_sampling.sample_trial(
             trial,
-            target_max_offset=self.target_max_offset,
-            window_length=window_length,
+            max_lookahead=self.max_lookahead,
+            sample_length=sample_length,
         )
 
-        logger.info(f"{input_columns_index = } - {seq_length = }")
+        logger.info(f"{input_columns_index = } - {context_length = }")
 
         return TrainModel(
             project_root=self.project_root,
@@ -849,9 +851,9 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             target_columns=self.target_columns,
             target_column_types=self.target_column_types,
             id_maps=self.id_maps,
-            seq_length=seq_length,
-            target_max_offset=self.target_max_offset,
-            window_length=window_length,
+            context_length=context_length,
+            max_lookahead=self.max_lookahead,
+            sample_length=sample_length,
             n_classes=self.n_classes,
             inference_batch_size=self.inference_batch_size,
             seed=101,
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index a05053bb..99f1be90 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -56,10 +56,10 @@ def load_inferer_config(
             source=f"metadata config '{metadata_config_path}'",
         )
         sequence_layout = sequence_layout_from_metadata(
-            metadata_config, config_values["seq_length"]
+            metadata_config, config_values["context_length"]
         )
-        config_values["target_max_offset"] = sequence_layout.target_max_offset
-        config_values["window_length"] = sequence_layout.window_length
+        config_values["max_lookahead"] = sequence_layout.max_lookahead
+        config_values["sample_length"] = sequence_layout.sample_length
         config_values["sequence_layout_version"] = (
             sequence_layout.sequence_layout_version
         )
@@ -122,7 +122,7 @@ class InfererModel(BaseModel):
         map_to_id: If True, maps categorical output values back to their original IDs.
         seed: The random seed for reproducibility.
         device: The device to run inference on (e.g., 'cuda', 'cpu', 'mps').
-        seq_length: The sequence length of the model's input.
+        context_length: The sequence length of the model's input.
         inference_batch_size: The batch size for inference.
         sample_from_distribution_columns: A list of columns from which to sample from the distribution.
         infer_with_dropout: If True, applies dropout during inference.
@@ -154,9 +154,9 @@ class InfererModel(BaseModel):
     map_to_id: bool = Field(default=True)
     seed: int
     device: str
-    seq_length: int
-    target_max_offset: int = Field(default=1, ge=0)
-    window_length: int
+    context_length: int
+    max_lookahead: int = Field(default=1, ge=0)
+    sample_length: int
     sequence_layout_version: int = 1
     prediction_length: Optional[int] = None
     inference_batch_size: int
@@ -168,36 +168,36 @@ class InfererModel(BaseModel):
 
     @model_validator(mode="after")
     def normalize_prediction_length(self):
-        if self.window_length != self.seq_length + self.target_max_offset:
+        if self.sample_length != self.context_length + self.max_lookahead:
             raise ValueError(
-                "window_length must equal seq_length + target_max_offset "
-                f"({self.window_length} != {self.seq_length} + {self.target_max_offset})."
+                "sample_length must equal context_length + max_lookahead "
+                f"({self.sample_length} != {self.context_length} + {self.max_lookahead})."
             )
         if self.prediction_length is None:
             self.prediction_length = (
-                self.seq_length if self.training_objective == "bert" else 1
+                self.context_length if self.training_objective == "bert" else 1
             )
         if self.training_objective == "bert":
-            if self.prediction_length != self.seq_length:
+            if self.prediction_length != self.context_length:
                 raise ValueError(
-                    "For BERT inference, prediction_length must be equal to seq_length "
-                    f"(got prediction_length={self.prediction_length}, seq_length={self.seq_length})."
+                    "For BERT inference, prediction_length must be equal to context_length "
+                    f"(got prediction_length={self.prediction_length}, context_length={self.context_length})."
                 )
-        elif self.target_max_offset < 1:
+        elif self.max_lookahead < 1:
             raise ValueError(
-                "Causal inference requires data preprocessed with target_max_offset >= 1"
+                "Causal inference requires data preprocessed with max_lookahead >= 1"
             )
         return self
 
     @property
     def data_offset(self) -> int:
-        return self.target_max_offset
+        return self.max_lookahead
 
     @property
     def target_offset(self) -> int:
         if self.training_objective == "bert":
-            return self.target_max_offset
-        return self.target_max_offset - 1
+            return self.max_lookahead
+        return self.max_lookahead - 1
 
     @field_validator("training_objective")
     @classmethod
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index e72a62ed..a2532345 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -54,8 +54,8 @@ class PreprocessorModel(BaseModel):
         selected_columns: A list of columns to be included in the preprocessing. If None, all columns are used.
         split_ratios: A list of floats that define the relative sizes of data splits (e.g., for train, validation, test).
                            The sum of proportions must be 1.0.
-        seq_length: The sequence length for the model inputs.
-        target_max_offset: The maximum retained target offset after the model input window.
+        context_length: The sequence length for the model inputs.
+        max_lookahead: The maximum retained target offset after the model input window.
         stride_by_split: A list of step sizes for creating subsequences within each data split.
         max_rows: The maximum number of input rows to process. If None, all rows are processed.
         seed: A random seed for reproducibility.
@@ -78,8 +78,8 @@ class PreprocessorModel(BaseModel):
     selected_columns: Optional[list[str]] = None
 
     split_ratios: list[float]
-    seq_length: int
-    target_max_offset: int = Field(default=1, ge=0)
+    context_length: int
+    max_lookahead: int = Field(default=1, ge=0)
     stride_by_split: Optional[list[int]] = None
     max_rows: Optional[int] = None
     seed: int
@@ -195,7 +195,9 @@ def validate_mask_column_requires_metadata(self) -> "PreprocessorModel":
         return self
 
     def __init__(self, **kwargs):
-        default_stride_for_split = [kwargs["seq_length"]] * len(kwargs["split_ratios"])
+        default_stride_for_split = [kwargs["context_length"]] * len(
+            kwargs["split_ratios"]
+        )
         kwargs["stride_by_split"] = kwargs.get(
             "stride_by_split", default_stride_for_split
         )
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 992c76ee..f3621360 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -63,18 +63,16 @@ def load_train_config(
             metadata_config = json.loads(f.read())
 
         sequence_layout = sequence_layout_from_metadata(
-            metadata_config, config_values["seq_length"]
+            metadata_config, config_values["context_length"]
         )
-        config_values["target_max_offset"] = sequence_layout.target_max_offset
-        config_values["window_length"] = sequence_layout.window_length
+        config_values["max_lookahead"] = sequence_layout.max_lookahead
+        config_values["sample_length"] = sequence_layout.sample_length
         config_values["sequence_layout_version"] = (
             sequence_layout.sequence_layout_version
         )
         config_values.setdefault("training_spec", {})
-        config_values["training_spec"]["target_max_offset"] = (
-            sequence_layout.target_max_offset
-        )
-        config_values["training_spec"]["window_length"] = sequence_layout.window_length
+        config_values["training_spec"]["max_lookahead"] = sequence_layout.max_lookahead
+        config_values["training_spec"]["sample_length"] = sequence_layout.sample_length
 
         split_paths = metadata_config["split_paths"]
 
@@ -245,8 +243,8 @@ class TrainingSpecModel(BaseModel):
     fsdp_cpu_offload: Optional[bool] = None
     torch_compile: str = "outer"
     float32_matmul_precision: str = "highest"
-    target_max_offset: int = Field(default=1, ge=0)
-    window_length: int
+    max_lookahead: int = Field(default=1, ge=0)
+    sample_length: int
 
     def __init__(self, **kwargs):
         super().__init__(
@@ -264,13 +262,13 @@ def serialize_dotdict(self, value: DotDict) -> dict[str, Any]:
 
     @property
     def data_offset(self) -> int:
-        return self.target_max_offset
+        return self.max_lookahead
 
     @property
     def target_offset(self) -> int:
         if self.training_objective == "bert":
-            return self.target_max_offset
-        return self.target_max_offset - 1
+            return self.max_lookahead
+        return self.max_lookahead - 1
 
     @field_validator("layer_type_dtypes")
     @classmethod
@@ -384,9 +382,9 @@ def validate_bert_spec_matches_objective(self):
 
     @model_validator(mode="after")
     def validate_sequence_layout(self):
-        if self.training_objective == "causal" and self.target_max_offset < 1:
+        if self.training_objective == "causal" and self.max_lookahead < 1:
             raise ValueError(
-                "Causal training requires data preprocessed with target_max_offset >= 1"
+                "Causal training requires data preprocessed with max_lookahead >= 1"
             )
         return self
 
@@ -558,7 +556,7 @@ class TrainModel(BaseModel):
         target_columns: The list of target columns for model training.
         target_column_types: A dictionary mapping target columns to their types ('categorical' or 'real').
         id_maps: For each categorical column, a map from distinct values to their indexed representation.
-        seq_length: The sequence length of the model's input.
+        context_length: The sequence length of the model's input.
         n_classes: The number of classes for each categorical column.
         inference_batch_size: The batch size to be used for inference after model export.
         seed: The random seed for numpy and PyTorch.
@@ -592,9 +590,9 @@ class TrainModel(BaseModel):
         default_factory=lambda: SPECIAL_TOKEN_IDS.ids_by_label
     )
 
-    seq_length: int
-    target_max_offset: int = Field(default=1, ge=0)
-    window_length: int
+    context_length: int
+    max_lookahead: int = Field(default=1, ge=0)
+    sample_length: int
     sequence_layout_version: int = 1
     n_classes: dict[str, int]
     inference_batch_size: int
@@ -615,21 +613,21 @@ def validate_special_token_ids_match_runtime(cls, v):
         return validate_special_token_ids(v, source="TrainModel")
 
     @model_validator(mode="after")
-    def validate_bert_prediction_length_matches_seq_length(self):
-        if self.window_length != self.seq_length + self.target_max_offset:
+    def validate_bert_prediction_length_matches_context_length(self):
+        if self.sample_length != self.context_length + self.max_lookahead:
             raise ValueError(
-                "window_length must equal seq_length + target_max_offset "
-                f"({self.window_length} != {self.seq_length} + {self.target_max_offset})."
+                "sample_length must equal context_length + max_lookahead "
+                f"({self.sample_length} != {self.context_length} + {self.max_lookahead})."
             )
-        self.training_spec.target_max_offset = self.target_max_offset
-        self.training_spec.window_length = self.window_length
+        self.training_spec.max_lookahead = self.max_lookahead
+        self.training_spec.sample_length = self.sample_length
         if (
             self.training_spec.training_objective == "bert"
-            and self.model_spec.prediction_length != self.seq_length
+            and self.model_spec.prediction_length != self.context_length
         ):
             raise ValueError(
-                "For BERT training, model_spec.prediction_length must be equal to seq_length "
-                f"(got prediction_length={self.model_spec.prediction_length}, seq_length={self.seq_length})."
+                "For BERT training, model_spec.prediction_length must be equal to context_length "
+                f"(got prediction_length={self.model_spec.prediction_length}, context_length={self.context_length})."
             )
         return self
 
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 1bac51bd..335406ce 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -51,94 +51,96 @@
 
 @dataclass(frozen=True)
 class SequenceLayout:
-    seq_length: int
-    target_max_offset: int = 1
+    context_length: int
+    max_lookahead: int = 1
     sequence_layout_version: int = 2
 
     def __post_init__(self) -> None:
-        if self.seq_length < 1:
-            raise ValueError("seq_length must be a positive integer")
-        if self.target_max_offset < 0:
-            raise ValueError("target_max_offset must be non-negative")
+        if self.context_length < 1:
+            raise ValueError("context_length must be a positive integer")
+        if self.max_lookahead < 0:
+            raise ValueError("max_lookahead must be non-negative")
 
     @property
-    def window_length(self) -> int:
-        return self.seq_length + self.target_max_offset
+    def sample_length(self) -> int:
+        return self.context_length + self.max_lookahead
 
     @property
     def input_offset(self) -> int:
-        return self.target_max_offset
+        return self.max_lookahead
 
     def target_offset(self, horizon: int) -> int:
-        offset = self.target_max_offset - horizon
+        offset = self.max_lookahead - horizon
         if offset < 0:
             raise ValueError(
-                f"horizon={horizon} exceeds target_max_offset={self.target_max_offset}"
+                f"horizon={horizon} exceeds max_lookahead={self.max_lookahead}"
             )
         return offset
 
 
 @beartype
-def sequence_column_names(seq_length: int, offset: int) -> list[str]:
+def sequence_column_names(context_length: int, offset: int) -> list[str]:
     if offset < 0:
         raise ValueError("offset must be non-negative")
-    return [str(i) for i in range(seq_length - 1 + offset, offset - 1, -1)]
+    return [str(i) for i in range(context_length - 1 + offset, offset - 1, -1)]
 
 
 @beartype
-def slice_window(tensor: Tensor, seq_length: int, offset: int) -> Tensor:
+def slice_window(tensor: Tensor, context_length: int, offset: int) -> Tensor:
     if offset < 0:
         raise ValueError("offset must be non-negative")
 
     end = -offset if offset else None
-    result = tensor[:, -(seq_length + offset) : end]
+    result = tensor[:, -(context_length + offset) : end]
 
-    if result.shape[1] != seq_length:
+    if result.shape[1] != context_length:
         raise ValueError(
             f"Stored window width {tensor.shape[1]} cannot provide "
-            f"seq_length={seq_length} at offset={offset}"
+            f"context_length={context_length} at offset={offset}"
         )
     return result
 
 
 @beartype
-def validate_stored_window_width(tensor: Tensor, window_length: int) -> None:
-    if tensor.shape[1] != window_length:
+def validate_stored_window_width(tensor: Tensor, sample_length: int) -> None:
+    if tensor.shape[1] != sample_length:
         raise ValueError(
             f"Stored window width {tensor.shape[1]} does not match "
-            f"metadata window_length={window_length}."
+            f"metadata sample_length={sample_length}."
         )
 
 
 @beartype
-def sequence_layout_from_metadata(metadata: dict, seq_length: int) -> SequenceLayout:
-    metadata_seq_length = int(metadata.get("seq_length", seq_length))
-    if metadata_seq_length != seq_length:
+def sequence_layout_from_metadata(
+    metadata: dict, context_length: int
+) -> SequenceLayout:
+    metadata_context_length = int(metadata.get("context_length", context_length))
+    if metadata_context_length != context_length:
         raise ValueError(
-            f"Configured seq_length={seq_length} does not match preprocessed "
-            f"metadata seq_length={metadata_seq_length}."
+            f"Configured context_length={context_length} does not match preprocessed "
+            f"metadata context_length={metadata_context_length}."
         )
 
-    target_max_offset = int(metadata.get("target_max_offset", 1))
-    window_length = int(
-        metadata.get("window_length", metadata_seq_length + target_max_offset)
+    max_lookahead = int(metadata.get("max_lookahead", 1))
+    sample_length = int(
+        metadata.get("sample_length", metadata_context_length + max_lookahead)
     )
-    if window_length != metadata_seq_length + target_max_offset:
+    if sample_length != metadata_context_length + max_lookahead:
         raise ValueError(
-            "Invalid sequence layout metadata: window_length must equal "
-            "seq_length + target_max_offset "
-            f"({window_length} != {metadata_seq_length} + {target_max_offset})."
+            "Invalid sequence layout metadata: sample_length must equal "
+            "context_length + max_lookahead "
+            f"({sample_length} != {metadata_context_length} + {max_lookahead})."
         )
 
     layout = SequenceLayout(
-        seq_length=metadata_seq_length,
-        target_max_offset=target_max_offset,
+        context_length=metadata_context_length,
+        max_lookahead=max_lookahead,
         sequence_layout_version=int(metadata.get("sequence_layout_version", 1)),
     )
-    if layout.window_length != window_length:
+    if layout.sample_length != sample_length:
         raise ValueError(
-            f"Resolved layout window_length={layout.window_length} does not match "
-            f"metadata window_length={window_length}."
+            f"Resolved layout sample_length={layout.sample_length} does not match "
+            f"metadata sample_length={sample_length}."
         )
     return layout
 
@@ -339,8 +341,8 @@ def numpy_to_pytorch(
     data: pl.DataFrame,
     column_types: dict[str, torch.dtype],
     all_columns: list[str],
-    seq_length: int,
-    window_length: int,
+    context_length: int,
+    sample_length: int,
     data_offset: int,
     target_offset: int,
 ) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
@@ -357,7 +359,7 @@ def numpy_to_pytorch(
     2.  A "target" tensor (from time steps L-1 down to 0).
 
     Example:
-        For `seq_length = 3` and `all_columns = ['price']`, it will create:
+        For `context_length = 3` and `all_columns = ['price']`, it will create:
         - 'price': Tensor from columns ["3", "2", "1"]
         - 'price_target': Tensor from columns ["2", "1", "0"]
 
@@ -368,7 +370,7 @@ def numpy_to_pytorch(
             to their desired `torch.dtype`.
         all_columns: A list of all feature names (from "inputCol") to
             be processed and converted into tensors.
-        seq_length: The total sequence length (L). This determines the
+        context_length: The total sequence length (L). This determines the
             column names for time steps (e.g., "0" to "L").
 
     Returns:
@@ -378,8 +380,8 @@ def numpy_to_pytorch(
           (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
         - a metadata dictionary containing any explicit masks.
     """
-    input_seq_cols = sequence_column_names(seq_length, data_offset)
-    target_seq_cols = sequence_column_names(seq_length, target_offset)
+    input_seq_cols = sequence_column_names(context_length, data_offset)
+    target_seq_cols = sequence_column_names(context_length, target_offset)
 
     # We will create a unified dictionary
     unified_tensors = {}
@@ -408,8 +410,8 @@ def numpy_to_pytorch(
     if left_pad_lengths is not None:
         metadata = generate_padding_masks(
             left_pad_lengths,
-            seq_length,
-            window_length,
+            context_length,
+            sample_length,
             data_offset,
             target_offset,
         )
@@ -424,7 +426,7 @@ def build_valid_mask(
     left_pad_lengths: Tensor,
     full_length: int,
     offset: int,
-    seq_length: int,
+    context_length: int,
 ) -> Tensor:
     """Builds a boolean validity mask from explicit left-padding metadata."""
 
@@ -433,7 +435,7 @@ def build_valid_mask(
     )
     full_valid = full_positions[None, :] >= left_pad_lengths[:, None]
 
-    return full_valid[:, -(seq_length + offset) : (-offset if offset > 0 else None)]
+    return full_valid[:, -(context_length + offset) : (-offset if offset > 0 else None)]
 
 
 @beartype
@@ -457,18 +459,18 @@ def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Optional[
 @beartype
 def generate_padding_masks(
     left_pad_lengths: Tensor,
-    seq_length: int,
-    window_length: int,
+    context_length: int,
+    sample_length: int,
     data_offset: int,
     target_offset: int,
 ) -> dict[str, Tensor]:
     """Generates explicit attention and target masks as a metadata dictionary."""
     return {
         "attention_valid_mask": build_valid_mask(
-            left_pad_lengths, window_length, data_offset, seq_length
+            left_pad_lengths, sample_length, data_offset, context_length
         ),
         "target_valid_mask": build_valid_mask(
-            left_pad_lengths, window_length, target_offset, seq_length
+            left_pad_lengths, sample_length, target_offset, context_length
         ),
     }
 
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index d6c2e864..6d481926 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -221,7 +221,7 @@ def infer_worker(
                 f"Unsupported input type or read format: {config.read_format}"
             )
 
-        default_prediction_length = {"causal": 1, "bert": config.seq_length}
+        default_prediction_length = {"causal": 1, "bert": config.context_length}
         prediction_length = (
             config.prediction_length
             if config.prediction_length is not None
@@ -269,7 +269,7 @@ def infer_worker(
 
 def calculate_item_positions(
     start_positions: np.ndarray,
-    seq_length: int,
+    context_length: int,
     prediction_length: int,
     training_objective: str,
 ) -> np.ndarray:
@@ -278,7 +278,7 @@ def calculate_item_positions(
 
     Args:
         start_positions: 1D array of base start positions for each sequence in the batch.
-        seq_length: The length of the input sequence window.
+        context_length: The length of the input sequence window.
         prediction_length: The total number of predicted tokens per sequence.
         training_objective: Either "causal" or "bert".
 
@@ -291,7 +291,7 @@ def calculate_item_positions(
         position_offsets = np.arange(0, prediction_length)
     else:
         # Anchor positions to the future token step and tile backwards
-        base_positions = start_positions + seq_length
+        base_positions = start_positions + context_length
         position_offsets = np.arange(-prediction_length + 1, 1)
 
     # Repeat base anchors to match the number of predictions per sequence window
@@ -363,8 +363,8 @@ def _bert_target_valid_mask_from_preprocessed_data(
         )
         metadata = generate_padding_masks(
             left_pad_lengths,
-            config.seq_length,
-            config.window_length,
+            config.context_length,
+            config.sample_length,
             data_offset=config.data_offset,
             target_offset=config.target_offset,
         )
@@ -495,13 +495,13 @@ def infer_embedding(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.window_length)
+                validate_stored_window_width(tensor, config.sample_length)
             metadata = {}
             if left_pad_lengths_tensor is not None:
                 metadata = generate_padding_masks(
                     left_pad_lengths_tensor,
-                    config.seq_length,
-                    config.window_length,
+                    config.context_length,
+                    config.sample_length,
                     data_offset=config.data_offset,
                     target_offset=config.target_offset,
                 )
@@ -522,7 +522,7 @@ def infer_embedding(
         # Step 2: Calculate absolute positions and repeat IDs
         # (e.g., for seq_len=50, inf_size=5, offsets are [45, 46, 47, 48, 49])
         base_offsets = np.arange(
-            config.seq_length - prediction_length, config.seq_length
+            config.context_length - prediction_length, config.context_length
         )
 
         # Tile these offsets for each sample in the batch
@@ -669,7 +669,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.seq_length,
+                    config.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -682,7 +682,7 @@ def infer_generative(
                 # Unpack the new third return value
                 probs, preds, sequence_ids_for_preds, item_positions_for_preds = (
                     get_probs_preds_autoregression(
-                        config, inferer, data, column_types, config.seq_length
+                        config, inferer, data, column_types, config.context_length
                     )
                 )
         elif config.read_format == "parquet" and is_folder_input:
@@ -720,7 +720,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.seq_length,
+                    config.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -732,7 +732,7 @@ def infer_generative(
 
                 probs, preds, sequence_ids_for_preds, item_positions_for_preds = (
                     get_probs_preds_autoregression(
-                        config, inferer, data, column_types, config.seq_length
+                        config, inferer, data, column_types, config.context_length
                     )
                 )
         elif config.read_format == "pt":
@@ -744,7 +744,7 @@ def infer_generative(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.window_length)
+                validate_stored_window_width(tensor, config.sample_length)
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
@@ -752,7 +752,7 @@ def infer_generative(
             )
 
             sequences_dict = {
-                key: slice_window(tensor, config.seq_length, config.data_offset)
+                key: slice_window(tensor, config.context_length, config.data_offset)
                 for key, tensor in sequences_dict.items()
                 if key in config.input_columns
             }
@@ -761,8 +761,8 @@ def infer_generative(
             if left_pad_lengths_tensor is not None:
                 metadata = generate_padding_masks(
                     left_pad_lengths_tensor,
-                    config.seq_length,
-                    config.window_length,
+                    config.context_length,
+                    config.sample_length,
                     data_offset=config.data_offset,
                     target_offset=config.target_offset,
                 )
@@ -788,7 +788,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.seq_length,
+                    config.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -798,8 +798,8 @@ def infer_generative(
                     sequence_ids_tensor.numpy(), total_steps
                 )
                 item_position_boundaries = zip(
-                    list(start_positions_tensor + config.seq_length),
-                    list(start_positions_tensor + config.seq_length + total_steps),
+                    list(start_positions_tensor + config.context_length),
+                    list(start_positions_tensor + config.context_length + total_steps),
                 )
                 item_positions_for_preds = np.concatenate(
                     [np.arange(start, end) for start, end in item_position_boundaries],
@@ -944,9 +944,9 @@ def get_embeddings_pt(
         A NumPy array containing the computed embeddings for the batch.
     """
     for tensor in data.values():
-        validate_stored_window_width(tensor, config.window_length)
+        validate_stored_window_width(tensor, config.sample_length)
     X = {
-        key: slice_window(val, config.seq_length, config.data_offset).numpy()
+        key: slice_window(val, config.context_length, config.data_offset).numpy()
         for key, val in data.items()
         if key in config.input_columns
     }
@@ -1100,8 +1100,8 @@ def get_embeddings(
         data,
         column_types,
         all_columns,
-        config.seq_length,
-        config.window_length,
+        config.context_length,
+        config.sample_length,
         config.data_offset,
         config.target_offset,
     )
@@ -1153,8 +1153,8 @@ def get_probs_preds_from_df(
         data,
         column_types,
         all_columns,
-        config.seq_length,
-        config.window_length,
+        config.context_length,
+        config.sample_length,
         config.data_offset,
         config.target_offset,
     )
@@ -1239,7 +1239,7 @@ def get_probs_preds_autoregression(
     inferer: "Inferer",
     data: pl.DataFrame,
     column_types: dict[str, torch.dtype],
-    seq_length: int,
+    context_length: int,
 ) -> tuple[
     Optional[dict[str, np.ndarray]], dict[str, np.ndarray], np.ndarray, np.ndarray
 ]:
@@ -1253,7 +1253,7 @@ def get_probs_preds_autoregression(
         inferer: Initialized `Inferer` instance.
         data: Input DataFrame, sorted globally by `sequenceId` and locally by `subsequenceId`.
         column_types: Mapping of input column names to their `torch.dtype`.
-        seq_length: Length of the input sequence context.
+        context_length: Length of the input sequence context.
 
     Returns:
         A tuple containing:
@@ -1276,15 +1276,15 @@ def get_probs_preds_autoregression(
         .agg(pl.col("startItemPosition").max())
         .get_column("startItemPosition")
         .to_numpy()
-        + seq_length
+        + context_length
     )
 
     head_data, metadata = numpy_to_pytorch(
         head_data_df,
         column_types,
         config.input_columns,
-        seq_length,
-        config.window_length,
+        context_length,
+        config.sample_length,
         data_offset=config.data_offset,
         target_offset=0,
     )
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index 51a778ce..fc082183 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -46,8 +46,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             data=data_df,
             column_types=column_types,
             all_columns=all_columns,
-            seq_length=config.seq_length,
-            window_length=config.window_length,
+            context_length=config.context_length,
+            sample_length=config.sample_length,
             data_offset=config.training_spec.data_offset,
             target_offset=config.training_spec.target_offset,
         )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index d667f6c5..dc1965b7 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -49,11 +49,11 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
-        if folder_layout.window_length != config.window_length:
+        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        if folder_layout.sample_length != config.sample_length:
             raise ValueError(
-                f"Preprocessed folder window_length={folder_layout.window_length} "
-                f"does not match config window_length={config.window_length}."
+                f"Preprocessed folder sample_length={folder_layout.sample_length} "
+                f"does not match config sample_length={config.sample_length}."
             )
 
         self.n_samples = metadata["total_samples"]
@@ -68,7 +68,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         }
 
         # Sequence formatting structures matching long-format schema boundaries
-        train_seq_len = self.config.seq_length
+        train_seq_len = self.config.context_length
         input_seq_cols = sequence_column_names(
             train_seq_len, self.config.training_spec.data_offset
         )
@@ -213,7 +213,7 @@ def __iter__(
         indices_for_worker = indices_for_rank[worker_id::num_workers]
 
         # 5. Extract and pass unified data frames
-        train_seq_len = self.config.seq_length
+        train_seq_len = self.config.context_length
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
@@ -231,7 +231,7 @@ def __iter__(
                 metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
-                    self.config.window_length,
+                    self.config.sample_length,
                     self.config.training_spec.data_offset,
                     self.config.training_spec.target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 2145539b..980007ae 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -65,11 +65,11 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
-        if folder_layout.window_length != config.window_length:
+        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        if folder_layout.sample_length != config.sample_length:
             raise ValueError(
-                f"Preprocessed folder window_length={folder_layout.window_length} "
-                f"does not match config window_length={config.window_length}."
+                f"Preprocessed folder sample_length={folder_layout.sample_length} "
+                f"does not match config sample_length={config.sample_length}."
             )
 
         self.batch_files_info = metadata["batch_files"]
@@ -219,7 +219,7 @@ def __iter__(
 
         # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
-        train_seq_len = self.config.seq_length
+        train_seq_len = self.config.context_length
         global_file_start_sample = 0
 
         input_seq_cols = sequence_column_names(
@@ -319,7 +319,7 @@ def __iter__(
                 new_meta = generate_padding_masks(
                     left_pad_lengths[worker_indices],
                     train_seq_len,
-                    self.config.window_length,
+                    self.config.sample_length,
                     self.config.training_spec.data_offset,
                     self.config.training_spec.target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index b6108fb6..de193564 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -47,11 +47,11 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
-        if folder_layout.window_length != config.window_length:
+        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        if folder_layout.sample_length != config.sample_length:
             raise ValueError(
-                f"Preprocessed folder window_length={folder_layout.window_length} "
-                f"does not match config window_length={config.window_length}."
+                f"Preprocessed folder sample_length={folder_layout.sample_length} "
+                f"does not match config sample_length={config.sample_length}."
             )
 
         self.n_samples = metadata["total_samples"]
@@ -78,7 +78,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in all_sequences.keys():
                 if col in sequences_batch:
                     validate_stored_window_width(
-                        sequences_batch[col], config.window_length
+                        sequences_batch[col], config.sample_length
                     )
                     all_sequences[col].append(sequences_batch[col])
             if left_pad_lengths_batch is not None:
@@ -175,7 +175,7 @@ def __iter__(
         indices_for_worker = indices_for_rank[worker_id::num_workers]
 
         # 5. Yield full batches
-        train_seq_len = self.config.seq_length
+        train_seq_len = self.config.context_length
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
@@ -197,7 +197,7 @@ def __iter__(
                 metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
-                    self.config.window_length,
+                    self.config.sample_length,
                     data_offset,
                     target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 36e43b7b..cc629186 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -63,11 +63,11 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.seq_length)
-        if folder_layout.window_length != config.window_length:
+        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        if folder_layout.sample_length != config.sample_length:
             raise ValueError(
-                f"Preprocessed folder window_length={folder_layout.window_length} "
-                f"does not match config window_length={config.window_length}."
+                f"Preprocessed folder sample_length={folder_layout.sample_length} "
+                f"does not match config sample_length={config.sample_length}."
             )
 
         self.batch_files_info = metadata["batch_files"]
@@ -212,7 +212,7 @@ def __iter__(
 
         # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
-        train_seq_len = self.config.seq_length
+        train_seq_len = self.config.context_length
         global_file_start_sample = 0
 
         # Initialize cross-file buffers
@@ -244,7 +244,7 @@ def __iter__(
                 left_pad_lengths_batch,
             ) = torch.load(file_path, map_location="cpu", weights_only=False)
             for tensor in sequences_batch.values():
-                validate_stored_window_width(tensor, self.config.window_length)
+                validate_stored_window_width(tensor, self.config.sample_length)
 
             # Generate indices for the whole file
             indices = torch.arange(file_samples)
@@ -283,7 +283,7 @@ def __iter__(
                 new_meta = generate_padding_masks(
                     left_pad_lengths_batch[worker_indices],
                     train_seq_len,
-                    self.config.window_length,
+                    self.config.sample_length,
                     data_offset,
                     target_offset,
                 )
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index f7434b49..8f27cf17 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -11,8 +11,8 @@
 - 0.8
 - 0.1
 - 0.1
-seq_length: 48
-target_max_offset: 1
+context_length: 48
+max_lookahead: 1
 stride_by_split:
 - 1
 - 1
@@ -30,8 +30,8 @@
 target_column_types: # 'criterion' in training_spec must also be adapted
   EXAMPLE_TARGET_COLUMN_NAME: real
 
-seq_length: 48
-window_length: 49
+context_length: 48
+sample_length: 49
 inference_batch_size: 10
 
 export_generative_model: PLEASE FILL # true or false
@@ -49,7 +49,7 @@
   num_layers: 3
   prediction_length: 1
 training_spec:
-  window_length: 49
+  sample_length: 49
   training_objective: causal
   device: cuda
   epochs: 10
@@ -90,8 +90,8 @@
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 48
-window_length: 49
+context_length: 48
+sample_length: 49
 inference_batch_size: 10
 
 autoregression: true
diff --git a/src/sequifier/model/layers.py b/src/sequifier/model/layers.py
index ec970e8b..4e485d4c 100644
--- a/src/sequifier/model/layers.py
+++ b/src/sequifier/model/layers.py
@@ -106,7 +106,7 @@ def __init__(
         n_kv_heads,
         attention_type,
         dropout,
-        seq_length,
+        context_length,
         use_rope=False,
         rope_theta=10000.0,
     ):
@@ -127,7 +127,7 @@ def __init__(
 
         if use_rope:
             self.rope = RotaryEmbedding(
-                self.head_dim, max_seq_len=seq_length, theta=rope_theta
+                self.head_dim, max_seq_len=context_length, theta=rope_theta
             )
             if self.head_dim % 2 != 0:
                 raise ValueError(f"head_dim ({self.head_dim}) must be even for RoPE")
@@ -176,7 +176,9 @@ def forward(self, x, mask=None):
 
 
 class SequifierEncoderLayer(nn.Module):
-    def __init__(self, config, dim_model, n_head, dim_feedforward, dropout, seq_length):
+    def __init__(
+        self, config, dim_model, n_head, dim_feedforward, dropout, context_length
+    ):
         super().__init__()
         self.norm_first = config.norm_first
 
@@ -192,7 +194,7 @@ def __init__(self, config, dim_model, n_head, dim_feedforward, dropout, seq_leng
             n_kv_heads=config.n_kv_heads,
             attention_type=config.attention_type,
             dropout=dropout,
-            seq_length=seq_length,
+            context_length=context_length,
             use_rope=(config.positional_encoding == "rope"),
             rope_theta=config.rope_theta,
         )
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 01d312b1..962dd204 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -85,7 +85,7 @@ def __init__(
         merge_output: bool,
         selected_columns: Optional[list[str]],
         split_ratios: list[float],
-        seq_length: int,
+        context_length: int,
         stride_by_split: list[int],
         max_rows: Optional[int],
         seed: int,
@@ -95,7 +95,7 @@ def __init__(
         subsequence_start_mode: str,
         use_precomputed_maps: Optional[list[str]],
         metadata_config_path: Optional[str],
-        target_max_offset: int = 1,
+        max_lookahead: int = 1,
         mask_column: Optional[str] = None,
     ):
         """Initializes the Preprocessor with the given parameters.
@@ -108,8 +108,8 @@ def __init__(
             merge_output: Whether to combine the output into a single file.
             selected_columns: A list of columns to be included in the preprocessing.
             split_ratios: A list of floats that define the relative sizes of data splits.
-            seq_length: The sequence length for the model inputs.
-            target_max_offset: The maximum target horizon retained after each input window.
+            context_length: The sequence length for the model inputs.
+            max_lookahead: The maximum target horizon retained after each input window.
             stride_by_split: A list of step sizes for creating subsequences.
             max_rows: The maximum number of input rows to process.
             seed: A random seed for reproducibility.
@@ -144,7 +144,7 @@ def __init__(
         np.random.seed(seed)
         self.n_cores = n_cores or multiprocessing.cpu_count()
         self.continue_preprocessing = continue_preprocessing
-        self.layout = SequenceLayout(seq_length, target_max_offset)
+        self.layout = SequenceLayout(context_length, max_lookahead)
         self._setup_directories()
 
         if selected_columns is not None:
@@ -246,7 +246,7 @@ def __init__(
                 data_columns,
             )
 
-            schema = self._create_schema(col_types, self.layout.window_length)
+            schema = self._create_schema(col_types, self.layout.sample_length)
 
             data = data.sort(["sequenceId", "itemPosition"])
             n_batches = _process_batches_single_file(
@@ -342,7 +342,7 @@ def __init__(
                 write_format,
                 data_columns,
             )
-            schema = self._create_schema(col_types, self.layout.window_length)
+            schema = self._create_schema(col_types, self.layout.sample_length)
 
             self._process_batches_multiple_files(
                 files_to_process,
@@ -368,7 +368,7 @@ def __init__(
 
     @beartype
     def _create_schema(
-        self, col_types: dict[str, str], window_length: int
+        self, col_types: dict[str, str], sample_length: int
     ) -> dict[str, Any]:
         """Creates the Polars schema for the intermediate sequence DataFrame.
 
@@ -380,7 +380,7 @@ def _create_schema(
         Args:
             col_types: A dictionary mapping data column names to their Polars
                 string representations (e.g., "Int64", "Float64").
-            window_length: The number of items stored for each extracted window.
+            sample_length: The number of items stored for each extracted window.
 
         Returns:
             A dictionary defining the Polars schema. Keys are column names
@@ -410,7 +410,7 @@ def _create_schema(
             sequence_position_type = pl.Float64
 
         schema.update(
-            {str(i): sequence_position_type for i in range(window_length - 1, -1, -1)}
+            {str(i): sequence_position_type for i in range(sample_length - 1, -1, -1)}
         )
 
         return schema
@@ -812,9 +812,9 @@ def _cleanup(self, write_format: str) -> None:
     @beartype
     def _layout_metadata(self) -> dict[str, int]:
         return {
-            "seq_length": self.layout.seq_length,
-            "target_max_offset": self.layout.target_max_offset,
-            "window_length": self.layout.window_length,
+            "context_length": self.layout.context_length,
+            "max_lookahead": self.layout.max_lookahead,
+            "sample_length": self.layout.sample_length,
             "sequence_layout_version": self.layout.sequence_layout_version,
         }
 
@@ -1905,7 +1905,7 @@ def get_group_bounds(data_subset: pl.DataFrame, split_ratios: list[float]):
 @beartype
 def process_and_write_data_pt(
     data: pl.DataFrame,
-    window_length: int,
+    sample_length: int,
     path: str,
     column_types: dict[str, str],
 ):
@@ -1918,13 +1918,13 @@ def process_and_write_data_pt(
 
     It then converts these lists into NumPy arrays and stores one full
     sequence tensor per feature. Each tensor has shape
-    `(batch_size, window_length)`. The final five-element tuple
+    `(batch_size, sample_length)`. The final five-element tuple
     `(sequences_dict, sequence_ids_tensor, subsequence_ids_tensor, start_item_positions_tensor, left_pad_lengths_tensor)`
     is saved to a .pt file using `torch.save`.
 
     Args:
         data: The long-format Polars DataFrame of extracted sequences.
-        window_length: The stored serialized window width.
+        sample_length: The stored serialized window width.
         path: The output file path (e.g., "data/batch_0.pt").
         column_types: A dictionary mapping column names to their
             string data types, used to determine the correct torch dtype.
@@ -1932,7 +1932,7 @@ def process_and_write_data_pt(
     if data.is_empty():
         return
 
-    sequence_cols = [str(c) for c in range(window_length - 1, -1, -1)]
+    sequence_cols = [str(c) for c in range(sample_length - 1, -1, -1)]
 
     all_feature_cols = data.get_column("inputCol").unique().to_list()
 
@@ -2033,7 +2033,7 @@ def _write_accumulated_sequences(
 
     if write_format == "pt":
         process_and_write_data_pt(
-            combined_df, layout.window_length, out_path, col_types
+            combined_df, layout.sample_length, out_path, col_types
         )
     elif write_format == "parquet":
         combined_df.write_parquet(out_path)
@@ -2228,7 +2228,7 @@ def extract_sequences(
 
         subsequences, left_pad_lengths, subsequence_starts = extract_subsequences(
             in_seq_lists_only,
-            layout.window_length,
+            layout.sample_length,
             stride_for_split,
             columns,
             subsequence_start_mode,
@@ -2244,7 +2244,7 @@ def extract_sequences(
                     left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
-                expected_row_length = 5 + layout.window_length
+                expected_row_length = 5 + layout.sample_length
                 if len(row) != expected_row_length:
                     raise RuntimeError(
                         f"Row length mismatch. Expected {expected_row_length}, got {len(row)}. Row: {row}"
@@ -2261,22 +2261,22 @@ def extract_sequences(
 
 @beartype
 def get_subsequence_starts(
-    in_seq_length: int,
-    window_length: int,
+    in_context_length: int,
+    sample_length: int,
     stride_for_split: int,
     subsequence_start_mode: str,
 ) -> np.ndarray:
     """Calculates the start indices for extracting subsequences.
 
     This function determines the starting indices for sliding a window of
-    `window_length` over an input sequence of `in_seq_length`. It aims to
+    `sample_length` over an input sequence of `in_context_length`. It aims to
     use `stride_for_split`, but adjusts the step size slightly to ensure
     that the windows are distributed as evenly as possible and cover the
     full sequence from the beginning to the end.
 
     Args:
-        in_seq_length: The length of the original input sequence.
-        window_length: The stored window length to extract.
+        in_context_length: The length of the original input sequence.
+        sample_length: The stored window length to extract.
         stride_for_split: The *desired* step size between subsequences.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
 
@@ -2289,7 +2289,7 @@ def get_subsequence_starts(
         )
 
     if subsequence_start_mode == "distribute":
-        last_available_start = in_seq_length - window_length
+        last_available_start = in_context_length - sample_length
         raw_starts = np.arange(
             0, last_available_start + stride_for_split, stride_for_split
         )
@@ -2300,11 +2300,11 @@ def get_subsequence_starts(
         return np.unique(starts)
 
     if subsequence_start_mode == "exact":
-        if (in_seq_length - window_length) % stride_for_split != 0:
+        if (in_context_length - sample_length) % stride_for_split != 0:
             raise ValueError(
-                f"'exact' mode requires sequence length alignment, i.e. if: (in_seq_length - window_length) % stride_for_split == 0, {in_seq_length = }, {window_length = }, {stride_for_split = }"
+                f"'exact' mode requires sequence length alignment, i.e. if: (in_context_length - sample_length) % stride_for_split == 0, {in_context_length = }, {sample_length = }, {stride_for_split = }"
             )
-        last_possible_start = in_seq_length - window_length
+        last_possible_start = in_context_length - sample_length
         return np.arange(0, last_possible_start + 1, stride_for_split)
     return np.array([])
 
@@ -2312,7 +2312,7 @@ def get_subsequence_starts(
 @beartype
 def extract_subsequences(
     in_seq: dict[str, list],
-    window_length: int,
+    sample_length: int,
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
@@ -2322,14 +2322,14 @@ def extract_subsequences(
     This function takes a dictionary `in_seq` where keys are column
     names and values are lists of items for a single full sequence.
     It first pads the sequences with 0s at the beginning if they are
-    shorter than `window_length`. Then, it calculates the subsequence
+    shorter than `sample_length`. Then, it calculates the subsequence
     start indices using `get_subsequence_starts` and extracts all
     subsequences.
 
     Args:
         in_seq: A dictionary mapping column names to lists of items
             (e.g., `{'col_A': [1, 2, 3, 4, 5], 'col_B': [6, 7, 8, 9, 10]}`).
-        window_length: The stored window length to extract.
+        sample_length: The stored window length to extract.
         stride_for_split: The desired step size between subsequences.
         columns: A list of the column names (keys in `in_seq`) to process.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
@@ -2340,13 +2340,13 @@ def extract_subsequences(
     """
     in_seq_len = len(in_seq[columns[0]])
     pad_len = 0
-    if in_seq_len < window_length:
-        pad_len = window_length - in_seq_len
+    if in_seq_len < sample_length:
+        pad_len = sample_length - in_seq_len
         in_seq = {col: ([0] * pad_len) + in_seq[col] for col in columns}
-    in_seq_length = len(in_seq[columns[0]])
+    in_context_length = len(in_seq[columns[0]])
 
     subsequence_starts = get_subsequence_starts(
-        in_seq_length, window_length, stride_for_split, subsequence_start_mode
+        in_context_length, sample_length, stride_for_split, subsequence_start_mode
     )
     subsequence_starts_diff = subsequence_starts[1:] - subsequence_starts[:-1]
     if not np.all(subsequence_starts_diff <= stride_for_split):
@@ -2355,7 +2355,7 @@ def extract_subsequences(
         )
 
     result = {
-        col: [list(in_seq[col][i : i + window_length]) for i in subsequence_starts]
+        col: [list(in_seq[col][i : i + sample_length]) for i in subsequence_starts]
         for col in columns
     }
     left_pad_lengths = [pad_len] * len(subsequence_starts)
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 39795814..9b37f3f1 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -111,14 +111,14 @@ def create_dummy_data_and_metadata(
     for col in config.input_columns:
         dtype = torch.int64 if col in config.categorical_columns else torch.float32
         dummy_data[col] = torch.ones(
-            (config.training_spec.batch_size, config.seq_length),
+            (config.training_spec.batch_size, config.context_length),
             dtype=dtype,
             device=local_rank,
         )
 
     dummy_metadata = {
         "attention_valid_mask": torch.ones(
-            (config.training_spec.batch_size, config.seq_length),
+            (config.training_spec.batch_size, config.context_length),
             dtype=torch.bool,
             device=local_rank,
         )
@@ -627,7 +627,7 @@ def __init__(
         self.target_columns = hparams.target_columns
         self.target_column_types = hparams.target_column_types
         self.loss_weights = hparams.training_spec.loss_weights
-        self.seq_length = hparams.seq_length
+        self.context_length = hparams.context_length
         self.n_classes = hparams.n_classes
         self.inference_batch_size = hparams.inference_batch_size
         self.log_interval = hparams.training_spec.log_interval
@@ -686,12 +686,12 @@ def __init__(
             self.pos_encoder = ModuleDict()
             for col in self.real_columns:
                 self.pos_encoder[col] = nn.Embedding(
-                    self.seq_length, self.feature_embedding_dims[col]
+                    self.context_length, self.feature_embedding_dims[col]
                 )
             for col, n_classes in self.n_classes.items():
                 if col in self.categorical_columns:
                     self.pos_encoder[col] = nn.Embedding(
-                        self.seq_length, self.feature_embedding_dims[col]
+                        self.context_length, self.feature_embedding_dims[col]
                     )
         else:
             self.pos_encoder = None
@@ -704,7 +704,7 @@ def __init__(
                     hparams.model_spec.n_head,
                     hparams.model_spec.dim_feedforward,
                     hparams.training_spec.dropout,
-                    hparams.seq_length,
+                    hparams.context_length,
                 )
                 for _ in range(hparams.model_spec.num_layers)
             ]
@@ -758,13 +758,13 @@ def __init__(
         if hparams.training_spec.training_objective == "causal":
             self.register_buffer(
                 "src_mask",
-                self._generate_square_subsequent_mask(self.seq_length),
+                self._generate_square_subsequent_mask(self.context_length),
                 persistent=False,  # Optional: prevents the mask from being saved in your checkpoints
             )
         elif hparams.training_spec.training_objective == "bert":
             self.register_buffer(
                 "src_mask",
-                torch.zeros(self.seq_length, self.seq_length),
+                torch.zeros(self.context_length, self.context_length),
                 persistent=False,
             )
         else:
@@ -1020,18 +1020,18 @@ def _recursive_concat(self, srcs: list[Tensor]):
 
     @conditional_beartype
     def _build_attention_mask(self, valid_mask: Tensor, dtype: torch.dtype) -> Tensor:
-        batch_size, seq_length = valid_mask.shape
+        batch_size, context_length = valid_mask.shape
         device = valid_mask.device
 
-        expected_seq_length = self.src_mask.shape[-1]
-        if seq_length != expected_seq_length:
+        expected_context_length = self.src_mask.shape[-1]
+        if context_length != expected_context_length:
             raise ValueError(
-                f"valid_mask sequence length ({seq_length}) must match "
-                f"model sequence length ({expected_seq_length})."
+                f"valid_mask sequence length ({context_length}) must match "
+                f"model sequence length ({expected_context_length})."
             )
 
         base_mask = self.src_mask.to(device=device, dtype=dtype)
-        base_mask = base_mask.view(1, 1, seq_length, seq_length)
+        base_mask = base_mask.view(1, 1, context_length, context_length)
 
         invalid_keys = ~valid_mask.bool()
 
@@ -1039,7 +1039,7 @@ def _build_attention_mask(self, valid_mask: Tensor, dtype: torch.dtype) -> Tenso
             batch_size,
             1,
             1,
-            seq_length,
+            context_length,
             device=device,
             dtype=dtype,
         )
@@ -1055,7 +1055,7 @@ def _build_attention_mask(self, valid_mask: Tensor, dtype: torch.dtype) -> Tenso
     def _zero_padding_positions(self, x: Tensor, valid_mask: Tensor) -> Tensor:
         """
         x shape:
-            (batch_size, seq_length, dim_model)
+            (batch_size, context_length, dim_model)
 
         Zeroes padded query positions so padded rows do not keep evolving.
         """
@@ -1072,11 +1072,11 @@ def forward_inner(
 
         Args:
             src: A dictionary mapping column names to input tensors
-                 (batch_size, seq_length).
+                 (batch_size, context_length).
 
         Returns:
             The raw output tensor from the TransformerEncoder
-            (seq_length, batch_size, dim_model).
+            (context_length, batch_size, dim_model).
         """
         srcs = []
         for col in self.categorical_columns:
@@ -1087,7 +1087,7 @@ def forward_inner(
             if not self.use_rope:
                 pos = (
                     torch.arange(
-                        0, self.seq_length, dtype=torch.long, device=src_t.device
+                        0, self.context_length, dtype=torch.long, device=src_t.device
                     )
                     .repeat(src_t.shape[1], 1)
                     .T
@@ -1115,7 +1115,7 @@ def forward_inner(
             if not self.use_rope:
                 pos = (
                     torch.arange(
-                        0, self.seq_length, dtype=torch.long, device=src_t.device
+                        0, self.context_length, dtype=torch.long, device=src_t.device
                     )
                     .repeat(src_t.shape[1], 1)
                     .T
@@ -1137,7 +1137,7 @@ def forward_inner(
         if valid_mask.shape != src2.shape[:2]:
             raise ValueError(
                 f"Invalid attention mask shape: got {tuple(valid_mask.shape)}, "
-                f"expected {tuple(src2.shape[:2])} = (batch_size, seq_length). "
+                f"expected {tuple(src2.shape[:2])} = (batch_size, context_length). "
                 "Check attention_valid_mask / leftPadLength construction."
             )
         src2 = self._zero_padding_positions(src2, valid_mask)
@@ -1163,7 +1163,7 @@ def forward_embed(
 
         Args:
             src: A dictionary mapping column names to input tensors
-                 (batch_size, seq_length).
+                 (batch_size, context_length).
 
         Returns:
             The embedding tensor for the last token
@@ -1182,11 +1182,11 @@ def forward_train(
 
         Args:
             src: A dictionary mapping column names to input tensors
-                 (batch_size, seq_length).
+                 (batch_size, context_length).
 
         Returns:
             A dictionary mapping target column names to their raw output
-            (logit) tensors (seq_length, batch_size, n_classes/1).
+            (logit) tensors (context_length, batch_size, n_classes/1).
         """
         output = self.forward_inner(src, metadata)
         output = {
@@ -1205,11 +1205,11 @@ def decode(self, target_column: str, output: Tensor) -> Tensor:
         Args:
             target_column: The name of the target column to decode.
             output: The raw output tensor from the TransformerEncoder
-                    (seq_length, batch_size, dim_model).
+                    (context_length, batch_size, dim_model).
 
         Returns:
             The decoded output (logits or real value) for the target column
-            (seq_length, batch_size, n_classes/1).
+            (context_length, batch_size, n_classes/1).
         """
 
         target_dtype = self.decoder[target_column].weight.dtype
@@ -1250,7 +1250,7 @@ def forward(
 
         Args:
             src: A dictionary mapping column names to input tensors
-                 (batch_size, seq_length).
+                 (batch_size, context_length).
             return_logits: Return logits
 
         Returns:
@@ -1724,9 +1724,9 @@ def _calculate_loss(
 
         Args:
             output: A dictionary of output tensors from the model
-                    (seq_length, batch_size, n_classes/1).
+                    (context_length, batch_size, n_classes/1).
             targets: A dictionary of target tensors
-                     (batch_size, seq_length).
+                     (batch_size, context_length).
 
         Returns:
             A tuple containing:
@@ -2160,14 +2160,16 @@ def _export_model(
 
             x_cat = {
                 col: torch.randint(
-                    0, self.n_classes[col], (self.inference_batch_size, self.seq_length)
+                    0,
+                    self.n_classes[col],
+                    (self.inference_batch_size, self.context_length),
                 ).to(export_device, non_blocking=True)
                 for col in self.categorical_columns
             }
 
             dtype_real = torch.float32 if is_different_type else None
             x_real = {
-                col: torch.rand(self.inference_batch_size, self.seq_length).to(
+                col: torch.rand(self.inference_batch_size, self.context_length).to(
                     export_device, non_blocking=True, dtype=dtype_real
                 )
                 for col in self.real_columns
@@ -2176,7 +2178,7 @@ def _export_model(
             input_dict = {**x_cat, **x_real}
             attention_valid_mask = torch.ones(
                 self.inference_batch_size,
-                self.seq_length,
+                self.context_length,
                 dtype=torch.bool,
                 device=export_device,
             )
diff --git a/tests/configs/hyperparameter-search-bayesian.yaml b/tests/configs/hyperparameter-search-bayesian.yaml
index e15a2b84..c5f8d733 100644
--- a/tests/configs/hyperparameter-search-bayesian.yaml
+++ b/tests/configs/hyperparameter-search-bayesian.yaml
@@ -7,7 +7,7 @@ read_format: pt
 target_columns: [itemId]
 target_column_types:
   itemId: categorical
-seq_length: [8]
+context_length: [8]
 inference_batch_size: 10
 
 # Search Strategy
diff --git a/tests/configs/hyperparameter-search-bert.yaml b/tests/configs/hyperparameter-search-bert.yaml
index 84b49cf3..ff053b1e 100644
--- a/tests/configs/hyperparameter-search-bert.yaml
+++ b/tests/configs/hyperparameter-search-bert.yaml
@@ -7,7 +7,7 @@ read_format: pt
 target_columns: [itemId]
 target_column_types:
   itemId: categorical
-seq_length: [8]
+context_length: [8]
 inference_batch_size: 10
 
 search_strategy: sample
diff --git a/tests/configs/hyperparameter-search-custom-eval-inference.yaml b/tests/configs/hyperparameter-search-custom-eval-inference.yaml
index ef23b3a0..aa40c11c 100644
--- a/tests/configs/hyperparameter-search-custom-eval-inference.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval-inference.yaml
@@ -31,7 +31,7 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
+context_length: 8
 prediction_length: 1
 inference_batch_size: 10
 enforce_determinism: true
diff --git a/tests/configs/hyperparameter-search-custom-eval.yaml b/tests/configs/hyperparameter-search-custom-eval.yaml
index 6d870395..2dfc8d7a 100644
--- a/tests/configs/hyperparameter-search-custom-eval.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval.yaml
@@ -16,7 +16,7 @@ target_column_types:
   supCat2: categorical
   supCat3: categorical
   supCat4: categorical
-seq_length: [8]
+context_length: [8]
 inference_batch_size: 10
 
 # Search Strategy
diff --git a/tests/configs/hyperparameter-search-grid.yaml b/tests/configs/hyperparameter-search-grid.yaml
index df79f038..2922a641 100644
--- a/tests/configs/hyperparameter-search-grid.yaml
+++ b/tests/configs/hyperparameter-search-grid.yaml
@@ -7,7 +7,7 @@ read_format: pt
 target_columns: [itemId]
 target_column_types:
   itemId: categorical
-seq_length: [8]
+context_length: [8]
 inference_batch_size: 10
 
 # Search Strategy
diff --git a/tests/configs/hyperparameter-search-sample.yaml b/tests/configs/hyperparameter-search-sample.yaml
index 5a0085ba..5d429e8b 100644
--- a/tests/configs/hyperparameter-search-sample.yaml
+++ b/tests/configs/hyperparameter-search-sample.yaml
@@ -7,7 +7,7 @@ read_format: pt
 target_columns: [itemId]
 target_column_types:
   itemId: categorical
-seq_length: [8]
+context_length: [8]
 inference_batch_size: 10
 
 # Search Strategy
diff --git a/tests/configs/infer-test-categorical-autoregression.yaml b/tests/configs/infer-test-categorical-autoregression.yaml
index f34d4771..842cd2ea 100644
--- a/tests/configs/infer-test-categorical-autoregression.yaml
+++ b/tests/configs/infer-test-categorical-autoregression.yaml
@@ -16,8 +16,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical-bert-embedding.yaml b/tests/configs/infer-test-categorical-bert-embedding.yaml
index 17b591fd..72c14459 100644
--- a/tests/configs/infer-test-categorical-bert-embedding.yaml
+++ b/tests/configs/infer-test-categorical-bert-embedding.yaml
@@ -16,8 +16,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 prediction_length: null
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-bert.yaml b/tests/configs/infer-test-categorical-bert.yaml
index c213d865..d4d2b4b3 100644
--- a/tests/configs/infer-test-categorical-bert.yaml
+++ b/tests/configs/infer-test-categorical-bert.yaml
@@ -16,8 +16,8 @@ target_column_types:
 output_probabilities: true
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 prediction_length: null
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-embedding.yaml b/tests/configs/infer-test-categorical-embedding.yaml
index bbb67fcf..2c2bac57 100644
--- a/tests/configs/infer-test-categorical-embedding.yaml
+++ b/tests/configs/infer-test-categorical-embedding.yaml
@@ -16,8 +16,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical-inf-size-1.yaml b/tests/configs/infer-test-categorical-inf-size-1.yaml
index f1632ef5..89af4da8 100644
--- a/tests/configs/infer-test-categorical-inf-size-1.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-1.yaml
@@ -15,8 +15,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
index 6fdbb4e6..25f9f86e 100644
--- a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
@@ -18,8 +18,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-inf-size-3.yaml b/tests/configs/infer-test-categorical-inf-size-3.yaml
index 0f8f6bff..014e53a4 100644
--- a/tests/configs/infer-test-categorical-inf-size-3.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3.yaml
@@ -18,8 +18,8 @@ target_column_types:
 output_probabilities: true
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-multitarget.yaml b/tests/configs/infer-test-categorical-multitarget.yaml
index 4b0c13d1..ec7064f4 100644
--- a/tests/configs/infer-test-categorical-multitarget.yaml
+++ b/tests/configs/infer-test-categorical-multitarget.yaml
@@ -18,8 +18,8 @@ target_column_types:
 output_probabilities: true
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical.yaml b/tests/configs/infer-test-categorical.yaml
index 6b8d52cd..72214fba 100644
--- a/tests/configs/infer-test-categorical.yaml
+++ b/tests/configs/infer-test-categorical.yaml
@@ -16,8 +16,8 @@ target_column_types:
 output_probabilities: true
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-distributed-parquet.yaml b/tests/configs/infer-test-distributed-parquet.yaml
index c9716b2a..a7f2d423 100644
--- a/tests/configs/infer-test-distributed-parquet.yaml
+++ b/tests/configs/infer-test-distributed-parquet.yaml
@@ -17,8 +17,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-distributed.yaml b/tests/configs/infer-test-distributed.yaml
index a344c553..a89c17c8 100644
--- a/tests/configs/infer-test-distributed.yaml
+++ b/tests/configs/infer-test-distributed.yaml
@@ -19,6 +19,6 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 2
diff --git a/tests/configs/infer-test-lazy.yaml b/tests/configs/infer-test-lazy.yaml
index aafbe379..20bf3ec5 100644
--- a/tests/configs/infer-test-lazy.yaml
+++ b/tests/configs/infer-test-lazy.yaml
@@ -17,6 +17,6 @@ target_column_types:
 output_probabilities: false
 map_to_id: true
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 2
diff --git a/tests/configs/infer-test-real-autoregression.yaml b/tests/configs/infer-test-real-autoregression.yaml
index 357dc11b..79896d4c 100644
--- a/tests/configs/infer-test-real-autoregression.yaml
+++ b/tests/configs/infer-test-real-autoregression.yaml
@@ -16,8 +16,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: false
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: false
 
diff --git a/tests/configs/infer-test-real.yaml b/tests/configs/infer-test-real.yaml
index bc35d9e4..e8610c0c 100644
--- a/tests/configs/infer-test-real.yaml
+++ b/tests/configs/infer-test-real.yaml
@@ -14,8 +14,8 @@ target_column_types:
 output_probabilities: false
 map_to_id: false
 device: cpu
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/preprocess-test-categorical-exact-pt.yaml b/tests/configs/preprocess-test-categorical-exact-pt.yaml
index cea9f758..905eaf7e 100644
--- a/tests/configs/preprocess-test-categorical-exact-pt.yaml
+++ b/tests/configs/preprocess-test-categorical-exact-pt.yaml
@@ -7,7 +7,7 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-seq_length: 5
+context_length: 5
 stride_by_split:
 - 5
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-exact.yaml b/tests/configs/preprocess-test-categorical-exact.yaml
index 09720cde..d0cb1349 100644
--- a/tests/configs/preprocess-test-categorical-exact.yaml
+++ b/tests/configs/preprocess-test-categorical-exact.yaml
@@ -7,7 +7,7 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-seq_length: 5
+context_length: 5
 stride_by_split:
 - 5
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-interrupted.yaml b/tests/configs/preprocess-test-categorical-interrupted.yaml
index 5e9ddbab..16487d15 100644
--- a/tests/configs/preprocess-test-categorical-interrupted.yaml
+++ b/tests/configs/preprocess-test-categorical-interrupted.yaml
@@ -9,7 +9,7 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-multitarget.yaml b/tests/configs/preprocess-test-categorical-multitarget.yaml
index aee30ff6..b90646c8 100644
--- a/tests/configs/preprocess-test-categorical-multitarget.yaml
+++ b/tests/configs/preprocess-test-categorical-multitarget.yaml
@@ -9,7 +9,7 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
index b7e38c7a..3a088619 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
@@ -8,7 +8,7 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
index 11b3853c..4c730a1e 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
@@ -7,7 +7,7 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical.yaml b/tests/configs/preprocess-test-categorical.yaml
index 4571cfcd..c9ff762b 100644
--- a/tests/configs/preprocess-test-categorical.yaml
+++ b/tests/configs/preprocess-test-categorical.yaml
@@ -9,7 +9,7 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-multi-file.yaml b/tests/configs/preprocess-test-multi-file.yaml
index 106e21c8..4ac995c6 100644
--- a/tests/configs/preprocess-test-multi-file.yaml
+++ b/tests/configs/preprocess-test-multi-file.yaml
@@ -12,7 +12,7 @@ split_ratios:
 - 0.8
 - 0.1
 - 0.1
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-real.yaml b/tests/configs/preprocess-test-real.yaml
index b5593b45..3265dccc 100644
--- a/tests/configs/preprocess-test-real.yaml
+++ b/tests/configs/preprocess-test-real.yaml
@@ -8,7 +8,7 @@ selected_columns: null
 split_ratios:
 - 0.5
 - 0.5
-seq_length: 8
+context_length: 8
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/train-test-categorical-bert.yaml b/tests/configs/train-test-categorical-bert.yaml
index abf8e2d4..64a584fb 100644
--- a/tests/configs/train-test-categorical-bert.yaml
+++ b/tests/configs/train-test-categorical-bert.yaml
@@ -8,8 +8,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +33,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: bert
   bert_spec:
     masking_probability: 0.5
diff --git a/tests/configs/train-test-categorical-inf-size-1.yaml b/tests/configs/train-test-categorical-inf-size-1.yaml
index 259252dc..bc3a78f5 100644
--- a/tests/configs/train-test-categorical-inf-size-1.yaml
+++ b/tests/configs/train-test-categorical-inf-size-1.yaml
@@ -8,8 +8,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,7 +32,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-inf-size-3.yaml b/tests/configs/train-test-categorical-inf-size-3.yaml
index 33644d6d..9e0e1ebc 100644
--- a/tests/configs/train-test-categorical-inf-size-3.yaml
+++ b/tests/configs/train-test-categorical-inf-size-3.yaml
@@ -10,8 +10,8 @@ target_column_types:
   supCat1: categorical
   supCat2: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -39,7 +39,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 10000.0
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-multitarget-eager.yaml b/tests/configs/train-test-categorical-multitarget-eager.yaml
index e2212baf..4439f4e4 100644
--- a/tests/configs/train-test-categorical-multitarget-eager.yaml
+++ b/tests/configs/train-test-categorical-multitarget-eager.yaml
@@ -9,8 +9,8 @@ target_column_types:
   itemId: categorical
   supCat1: categorical
   supReal3: real
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -39,7 +39,7 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-multitarget.yaml b/tests/configs/train-test-categorical-multitarget.yaml
index c8d32a20..f7594939 100644
--- a/tests/configs/train-test-categorical-multitarget.yaml
+++ b/tests/configs/train-test-categorical-multitarget.yaml
@@ -10,8 +10,8 @@ target_column_types:
   itemId: categorical
   supCat1: categorical
   supReal3: real
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -40,7 +40,7 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical.yaml b/tests/configs/train-test-categorical.yaml
index 36dcff1c..018226c2 100644
--- a/tests/configs/train-test-categorical.yaml
+++ b/tests/configs/train-test-categorical.yaml
@@ -8,8 +8,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +33,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "outer"
diff --git a/tests/configs/train-test-distributed-lazy-parquet.yaml b/tests/configs/train-test-distributed-lazy-parquet.yaml
index 576bcee4..ba63aaec 100644
--- a/tests/configs/train-test-distributed-lazy-parquet.yaml
+++ b/tests/configs/train-test-distributed-lazy-parquet.yaml
@@ -8,8 +8,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -26,7 +26,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-distributed.yaml b/tests/configs/train-test-distributed.yaml
index 923fbf4a..3c0bb02d 100644
--- a/tests/configs/train-test-distributed.yaml
+++ b/tests/configs/train-test-distributed.yaml
@@ -10,8 +10,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -28,7 +28,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-lazy.yaml b/tests/configs/train-test-lazy.yaml
index 1d2ce592..6367027d 100644
--- a/tests/configs/train-test-lazy.yaml
+++ b/tests/configs/train-test-lazy.yaml
@@ -10,8 +10,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -28,7 +28,7 @@ model_spec:
   prediction_length: 1
 
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-real-bert.yaml b/tests/configs/train-test-real-bert.yaml
index 4e915730..4a8ef116 100644
--- a/tests/configs/train-test-real-bert.yaml
+++ b/tests/configs/train-test-real-bert.yaml
@@ -8,8 +8,8 @@ target_columns: [itemValue]
 target_column_types:
   itemValue: real
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +33,7 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: bert
   bert_spec:
     masking_probability: 0.5
diff --git a/tests/configs/train-test-real.yaml b/tests/configs/train-test-real.yaml
index 2596a365..33a0c67d 100644
--- a/tests/configs/train-test-real.yaml
+++ b/tests/configs/train-test-real.yaml
@@ -8,8 +8,8 @@ target_columns: [itemValue]
 target_column_types:
   itemValue: real
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +33,7 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "inner"
diff --git a/tests/configs/train-test-resume-epoch.yaml b/tests/configs/train-test-resume-epoch.yaml
index da990807..f229fb1d 100644
--- a/tests/configs/train-test-resume-epoch.yaml
+++ b/tests/configs/train-test-resume-epoch.yaml
@@ -8,8 +8,8 @@ target_columns: [itemValue]
 target_column_types:
   itemValue: real
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: false
@@ -33,7 +33,7 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-resume-mid-epoch.yaml b/tests/configs/train-test-resume-mid-epoch.yaml
index b76a474c..b43c8c2b 100644
--- a/tests/configs/train-test-resume-mid-epoch.yaml
+++ b/tests/configs/train-test-resume-mid-epoch.yaml
@@ -8,8 +8,8 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 
-seq_length: 8
-window_length: 9
+context_length: 8
+sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +33,7 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
-  window_length: 9
+  sample_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "outer"
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 36d5bbdd..8136c84e 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -168,7 +168,7 @@ def test_probabilities(probabilities):
         )
 
 
-def test_bert_generative_predictions_default_to_seq_length(
+def test_bert_generative_predictions_default_to_context_length(
     bert_predictions, project_root
 ):
     metadata = _categorical_metadata(project_root)
diff --git a/tests/integration/test_make.py b/tests/integration/test_make.py
index 649b3903..471de7a0 100644
--- a/tests/integration/test_make.py
+++ b/tests/integration/test_make.py
@@ -54,7 +54,7 @@ def adapt_configs(config_strings):
                 "selected_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "selected_columns: "
             )
             .replace("input_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "input_columns: ")
-            .replace("seq_length: 48", "seq_length: 10")
+            .replace("context_length: 48", "context_length: 10")
             .replace("max_rows: null", "max_rows: null\nn_cores: 1")
         )
 
@@ -87,8 +87,8 @@ def adapt_configs(config_strings):
             .replace("EXAMPLE_TARGET_COLUMN_NAME: MSELoss", "itemId: CrossEntropyLoss")
             .replace("epochs: 10", "epochs: 3")
             .replace("device: cuda", "device: cpu")
-            .replace("seq_length: 48", "seq_length: 10")
-            .replace("window_length: 49", "window_length: 11")
+            .replace("context_length: 48", "context_length: 10")
+            .replace("sample_length: 49", "sample_length: 11")
             .replace("total_steps: PLEASE FILL", "total_steps: 10000")
         )
 
@@ -118,8 +118,8 @@ def adapt_configs(config_strings):
             )
             .replace("[EXAMPLE_INPUT_COLUMN_NAME]", "[itemId]")
             .replace("[EXAMPLE_TARGET_COLUMN_NAME]", "[itemId]")
-            .replace("seq_length: 48", "seq_length: 10")
-            .replace("window_length: 49", "window_length: 11")
+            .replace("context_length: 48", "context_length: 10")
+            .replace("sample_length: 49", "sample_length: 11")
             .replace("autoregression: true", "autoregression: false")
             .replace("EXAMPLE_TARGET_COLUMN_NAME: real", "itemId: categorical")
         )
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index d90fe0ff..cb663dc2 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -11,7 +11,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
     project_root = str(tmp_path)
     (tmp_path / "logs").mkdir()
 
-    seq_length = 4
+    context_length = 4
     inference_batch_size = 2
     config = TrainModel(
         project_root=project_root,
@@ -27,8 +27,8 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         real_columns=["real_col"],
         id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
         n_classes={"cat_col": 6},
-        seq_length=seq_length,
-        window_length=5,
+        context_length=context_length,
+        sample_length=5,
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -41,7 +41,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
             n_head=2,
             dim_feedforward=8,
             num_layers=1,
-            prediction_length=seq_length,
+            prediction_length=context_length,
             feature_embedding_dims={"cat_col": 7, "real_col": 1},
             activation_fn="relu",
             normalization="layer_norm",
@@ -72,7 +72,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
             loss_weights={"cat_col": 1.0, "real_col": 1.0},
             torch_compile="none",
             layer_autocast=False,
-            window_length=5,
+            sample_length=5,
         ),
     )
     model = TransformerModel(config)
@@ -108,7 +108,7 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
     project_root = str(tmp_path)
     (tmp_path / "logs").mkdir()
 
-    seq_length = 4
+    context_length = 4
     inference_batch_size = 2
     config = TrainModel(
         project_root=project_root,
@@ -127,8 +127,8 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             "cat10": {"x": 3, "y": 4, "z": 5},
         },
         n_classes={"cat2": 7, "cat10": 6},
-        seq_length=seq_length,
-        window_length=5,
+        context_length=context_length,
+        sample_length=5,
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -163,7 +163,7 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             loss_weights={"cat2": 1.0},
             torch_compile="none",
             layer_autocast=False,
-            window_length=5,
+            sample_length=5,
         ),
     )
     model = TransformerModel(config)
@@ -186,7 +186,7 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             "cat2_in": np.array([[3, 4, 5, 6], [6, 5, 4, 3]], dtype=np.int64),
             "cat10_in": np.array([[3, 4, 5, 3], [5, 4, 3, 5]], dtype=np.int64),
             "attention_valid_mask": np.ones(
-                (inference_batch_size, seq_length), dtype=np.bool_
+                (inference_batch_size, context_length), dtype=np.bool_
             ),
         },
     )
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 54cb4839..999c9cbf 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -31,9 +31,9 @@ def test_metadata_config(metadata_configs):
         "split_paths",
         "column_types",
         "selected_columns_statistics",
-        "seq_length",
-        "target_max_offset",
-        "window_length",
+        "context_length",
+        "max_lookahead",
+        "sample_length",
         "sequence_layout_version",
     ]
 
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
index d8212897..03a72927 100644
--- a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
+++ b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
@@ -1,7 +1,7 @@
 {
-    "seq_length": 8,
-    "target_max_offset": 1,
-    "window_length": 9,
+    "context_length": 8,
+    "max_lookahead": 1,
+    "sample_length": 9,
     "sequence_layout_version": 2,
     "selected_columns": null,
     "data_columns": [
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 0c3feb6b..d51685bf 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -15,8 +15,8 @@ def mock_config():
     config = MagicMock()
     config.project_root = "."
     config.seed = 42
-    config.seq_length = 2
-    config.window_length = 3
+    config.context_length = 2
+    config.sample_length = 3
     config.column_types = {"item": "Float64"}
     config.training_spec.batch_size = 5
     config.training_spec.num_workers = 0
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 6bea73f4..d5688ddc 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -19,8 +19,8 @@ def mock_config(tmp_path):
     config.training_spec.sampling_strategy = "exact"
     config.training_spec.num_workers = 0
     config.seed = 42
-    config.seq_length = 5
-    config.window_length = 6
+    config.context_length = 5
+    config.sample_length = 6
     config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
     config.training_spec.data_offset = 1
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index cd111b6d..b36c1624 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -65,7 +65,7 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     Tests that DataFrames are correctly converted to input and target Tensors.
 
     Logic to test:
-    If seq_length = 3:
+    If context_length = 3:
     - Input cols:  ['3', '2', '1']
     - Target cols: ['2', '1', '0']
     """
@@ -84,14 +84,14 @@ def test_numpy_to_pytorch_shapes_and_shifting():
 
     column_types = {"A": torch.float32}
     all_columns = ["A"]
-    seq_length = 3
+    context_length = 3
 
     tensors, metadata = numpy_to_pytorch(
         data,
         column_types,
         all_columns,
-        seq_length,
-        seq_length + 1,
+        context_length,
+        context_length + 1,
         data_offset=1,
         target_offset=0,
     )
@@ -128,8 +128,8 @@ def test_numpy_to_pytorch_dtypes():
         data_int,
         {"int_col": torch.int64},
         ["int_col"],
-        seq_length=1,
-        window_length=2,
+        context_length=1,
+        sample_length=2,
         data_offset=1,
         target_offset=0,
     )
@@ -141,8 +141,8 @@ def test_numpy_to_pytorch_dtypes():
         data_float,
         {"float_col": torch.float32},
         ["float_col"],
-        seq_length=1,
-        window_length=2,
+        context_length=1,
+        sample_length=2,
         data_offset=1,
         target_offset=0,
     )
@@ -153,10 +153,10 @@ def test_build_valid_mask_from_left_pad_lengths():
     left_pad_lengths = torch.tensor([0, 2, 5], dtype=torch.int64)
 
     input_mask = build_valid_mask(
-        left_pad_lengths, full_length=6, offset=1, seq_length=5
+        left_pad_lengths, full_length=6, offset=1, context_length=5
     )
     target_mask = build_valid_mask(
-        left_pad_lengths, full_length=6, offset=0, seq_length=5
+        left_pad_lengths, full_length=6, offset=0, context_length=5
     )
 
     assert torch.equal(
@@ -185,11 +185,11 @@ def test_slice_window_validates_stored_width():
     tensor = torch.arange(8).reshape(2, 4)
 
     assert torch.equal(
-        slice_window(tensor, seq_length=3, offset=1),
+        slice_window(tensor, context_length=3, offset=1),
         torch.tensor([[0, 1, 2], [4, 5, 6]]),
     )
     assert torch.equal(
-        slice_window(tensor[:, :3], seq_length=3, offset=0), tensor[:, :3]
+        slice_window(tensor[:, :3], context_length=3, offset=0), tensor[:, :3]
     )
 
 
@@ -212,8 +212,8 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
         data,
         {"A": torch.float32},
         ["A"],
-        seq_length=3,
-        window_length=4,
+        context_length=3,
+        sample_length=4,
         data_offset=1,
         target_offset=0,
     )
@@ -238,7 +238,7 @@ def test_apply_bert_masking_uses_explicit_valid_mask_for_zero_values():
         categorical_columns=[],
         real_columns=["real_col"],
         n_classes={},
-        seq_length=4,
+        context_length=4,
         training_spec=SimpleNamespace(
             batch_size=1,
             bert_spec=SimpleNamespace(
diff --git a/tests/unit/test_hyperparameter_search_config.py b/tests/unit/test_hyperparameter_search_config.py
index 7b9103d7..8a50fc34 100644
--- a/tests/unit/test_hyperparameter_search_config.py
+++ b/tests/unit/test_hyperparameter_search_config.py
@@ -63,7 +63,7 @@ def bert_spec_sampling_config():
 
 
 def sample_training_spec(sampling, trial):
-    return sampling.sample_trial(trial, target_max_offset=1, window_length=9)
+    return sampling.sample_trial(trial, max_lookahead=1, sample_length=9)
 
 
 def test_training_objective_defaults_to_causal_without_bert_spec():
@@ -74,7 +74,7 @@ def test_training_objective_defaults_to_causal_without_bert_spec():
 
     assert training_spec.training_objective == "causal"
     assert training_spec.bert_spec is None
-    assert training_spec.window_length == 9
+    assert training_spec.sample_length == 9
     assert trial.params["training_objective"] == "causal"
 
 
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index ccf5e2ae..5cc76580 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -156,7 +156,7 @@ def test_inferer_prepare_inference_batches_split(mock_inferer):
     np.testing.assert_array_equal(batches[2]["cat_col"], [[5]])
 
 
-def test_infer_config_defaults_bert_prediction_length_to_seq_length():
+def test_infer_config_defaults_bert_prediction_length_to_context_length():
     config = InfererModel(
         project_root=".",
         metadata_config_path="dummy.json",
@@ -173,15 +173,15 @@ def test_infer_config_defaults_bert_prediction_length_to_seq_length():
         seed=42,
         device="cpu",
         prediction_length=None,
-        seq_length=3,
-        window_length=4,
+        context_length=3,
+        sample_length=4,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
         autoregression=False,
     )
 
-    assert config.prediction_length == config.seq_length
+    assert config.prediction_length == config.context_length
 
 
 def test_infer_config_defaults_causal_prediction_length_to_one():
@@ -201,8 +201,8 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
         seed=42,
         device="cpu",
         prediction_length=None,
-        seq_length=3,
-        window_length=4,
+        context_length=3,
+        sample_length=4,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -230,8 +230,8 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             seed=42,
             device="cpu",
             prediction_length=1,
-            seq_length=3,
-            window_length=4,
+            context_length=3,
+            sample_length=4,
             inference_batch_size=2,
             output_probabilities=False,
             map_to_id=False,
@@ -277,7 +277,7 @@ def test_load_inferer_config_rejects_mismatched_metadata_special_token_ids(tmp_p
                 "target_column_types": {"target_col": "categorical"},
                 "seed": 42,
                 "device": "cpu",
-                "seq_length": 3,
+                "context_length": 3,
                 "inference_batch_size": 2,
             }
         )
@@ -350,7 +350,7 @@ def get_inputs(self):
 def test_calculate_item_positions_bert_uses_full_input_window():
     positions = calculate_item_positions(
         np.array([10, 20]),
-        seq_length=4,
+        context_length=4,
         prediction_length=4,
         training_objective="bert",
     )
@@ -361,7 +361,7 @@ def test_calculate_item_positions_bert_uses_full_input_window():
 def test_calculate_item_positions_causal_uses_future_window_tail():
     positions = calculate_item_positions(
         np.array([10, 20]),
-        seq_length=4,
+        context_length=4,
         prediction_length=2,
         training_objective="causal",
     )
@@ -389,8 +389,8 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         seed=42,
         device="cpu",
         prediction_length=None,
-        seq_length=3,
-        window_length=4,
+        context_length=3,
+        sample_length=4,
         inference_batch_size=4,
         output_probabilities=False,
         map_to_id=True,
@@ -597,8 +597,8 @@ def ar_config():
         seed=42,
         device="cpu",
         prediction_length=1,
-        seq_length=3,
-        window_length=4,
+        context_length=3,
+        sample_length=4,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,  # Set to False to bypass ID mapping requirements
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 6000208e..f6c13e17 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -33,16 +33,16 @@
 def test_extract_subsequences_basic():
     """Tests basic sliding window extraction with sufficient length."""
     input_data = {"col1": [10, 11, 12, 13, 14, 15]}
-    seq_length = 3
+    context_length = 3
     stride = 1
     columns = ["col1"]
 
-    # Expected behavior: Window size is seq_length + 1 (history + target)
+    # Expected behavior: Window size is context_length + 1 (history + target)
     # Windows: [10,11,12,13], [11,12,13,14], [12,13,14,15]
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        seq_length + 1,
+        context_length + 1,
         stride,
         columns,
         subsequence_start_mode="distribute",
@@ -55,13 +55,13 @@ def test_extract_subsequences_basic():
 
 def test_extract_subsequences_bert_width():
     input_data = {"col1": [10, 11, 12, 13, 14, 15]}
-    seq_length = 3
+    context_length = 3
     stride = 1
     columns = ["col1"]
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        window_length=seq_length,
+        sample_length=context_length,
         stride_for_split=stride,
         columns=columns,
         subsequence_start_mode="distribute",
@@ -73,16 +73,16 @@ def test_extract_subsequences_bert_width():
 
 
 def test_extract_subsequences_padding():
-    """Tests that sequences shorter than seq_length are padded with 0s."""
+    """Tests that sequences shorter than context_length are padded with 0s."""
     input_data = {"col1": [1, 2]}  # Length 2
-    window_length = 5
+    sample_length = 5
     stride = 1
     columns = ["col1"]
 
     # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data, window_length, stride, columns, subsequence_start_mode="distribute"
+        input_data, sample_length, stride, columns, subsequence_start_mode="distribute"
     )
 
     assert len(result["col1"]) == 1
@@ -93,7 +93,7 @@ def test_extract_subsequences_returns_left_pad_lengths_when_requested():
     input_data = {"col1": [0.0, 1.5]}
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        window_length=5,
+        sample_length=5,
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -127,7 +127,7 @@ def test_extract_sequences_persists_left_pad_length_metadata():
     sequences = extract_sequences(
         data,
         schema,
-        layout=SequenceLayout(seq_length=4),
+        layout=SequenceLayout(context_length=4),
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -154,7 +154,7 @@ def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
 
     process_and_write_data_pt(
         data,
-        window_length=4,
+        sample_length=4,
         path=str(out_path),
         column_types={"col1": "Float64"},
     )
@@ -242,7 +242,7 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
         merge_output=True,
         selected_columns=["itemId", "itemValue"],
         split_ratios=[1.0],
-        seq_length=2,
+        context_length=2,
         stride_by_split=[1],
         max_rows=None,
         seed=1010,
@@ -379,7 +379,7 @@ def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
             merge_output=True,
             selected_columns=["itemId"],
             split_ratios=[1.0],
-            seq_length=1,
+            context_length=1,
             stride_by_split=[1],
             max_rows=None,
             seed=1010,
@@ -408,7 +408,7 @@ def test_preprocessor_config_defaults_mask_column_to_none(tmp_path):
         project_root=str(tmp_path),
         data_path=str(data_path),
         split_ratios=[1.0],
-        seq_length=1,
+        context_length=1,
         seed=1010,
     )
 
@@ -436,7 +436,7 @@ def test_preprocessor_config_requires_metadata_config_for_mask_column(
             project_root=str(tmp_path),
             data_path=str(data_path),
             split_ratios=[1.0],
-            seq_length=1,
+            context_length=1,
             seed=1010,
             mask_column=RESERVED_MASK_COLUMN,
         )
@@ -449,15 +449,15 @@ def test_extract_subsequences_modes(mode):
     # distribute: adjusts stride to cover data evenly.
     # exact: strictly adheres to stride, throws error if misalignment.
     input_data = {"col1": list(range(10))}
-    seq_length = 2
-    window_length = seq_length + 1
+    context_length = 2
+    sample_length = context_length + 1
     columns = ["col1"]
 
     if mode == "distribute":
         stride = 4
         # distribute might adjust indices to maximize coverage
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data, window_length, stride, columns, mode
+            input_data, sample_length, stride, columns, mode
         )
         assert len(result["col1"]) > 0
 
@@ -467,13 +467,13 @@ def test_extract_subsequences_modes(mode):
         )
         # Testing a failing exact case
         with pytest.raises(ValueError):
-            extract_subsequences(input_data, window_length, 4, columns, mode)
+            extract_subsequences(input_data, sample_length, 4, columns, mode)
 
         # Testing a passing exact case
         # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
         input_data_exact = {"col1": list(range(11))}
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data_exact, window_length, 4, columns, mode
+            input_data_exact, sample_length, 4, columns, mode
         )
         assert len(result["col1"]) > 0
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 74292545..054d0956 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -31,7 +31,7 @@ def _training_spec_kwargs(**overrides):
         "optimizer": {"name": "Adam"},
         "scheduler": {"name": "StepLR", "step_size": 1, "gamma": 0.1},
         "loss_weights": {"cat_col": 1.0, "real_col": 1.0},
-        "window_length": 11,
+        "sample_length": 11,
     }
     values.update(overrides)
     return values
@@ -118,7 +118,7 @@ def model_config(tmp_path):
         optimizer={"name": "Adam"},
         scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
         loss_weights={"cat_col": 1.0, "real_col": 1.0},
-        window_length=11,
+        sample_length=11,
     )
 
     config = TrainModel(
@@ -136,8 +136,8 @@ def model_config(tmp_path):
         # id_maps is needed for constructing index_maps in model init
         id_maps={"cat_col": {"a": 1, "b": 2, "c": 3, "d": 4}},
         n_classes={"cat_col": 5},  # 0 + 4 classes
-        seq_length=10,
-        window_length=11,
+        context_length=10,
+        sample_length=11,
         inference_batch_size=4,
         seed=42,
         export_generative_model=True,
@@ -164,7 +164,7 @@ def causal_model(model_config):
 @pytest.fixture
 def bert_model(model_config):
     config_values = model_config.model_dump()
-    config_values["model_spec"]["prediction_length"] = model_config.seq_length
+    config_values["model_spec"]["prediction_length"] = model_config.context_length
     config_values["training_spec"] = _training_spec_kwargs(
         training_objective="bert",
         bert_spec=_bert_spec(),
@@ -213,9 +213,11 @@ def test_transformer_model_initialization(model, model_config):
         assert "real_col" in model.encoder
 
 
-def test_train_model_requires_bert_prediction_length_to_equal_seq_length(model_config):
+def test_train_model_requires_bert_prediction_length_to_equal_context_length(
+    model_config,
+):
     config_values = model_config.model_dump()
-    config_values["model_spec"]["prediction_length"] = model_config.seq_length - 1
+    config_values["model_spec"]["prediction_length"] = model_config.context_length - 1
     config_values["training_spec"] = _training_spec_kwargs(
         training_objective="bert",
         bert_spec=_bert_spec(),
@@ -295,7 +297,7 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
 def test_forward_train_shapes(model, model_config):
     """Tests the output shapes of the forward_train method."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.seq_length
+    seq_len = model_config.context_length
 
     # Create dummy inputs
     # Categorical: (batch, seq_len) integers
@@ -326,7 +328,7 @@ def test_forward_train_shapes(model, model_config):
 def test_forward_inference_shapes(model, model_config):
     """Tests the output shapes of the forward (inference) method."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.seq_length
+    seq_len = model_config.context_length
     prediction_length = model_config.model_spec.prediction_length  # 1
 
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
@@ -359,7 +361,7 @@ def test_forward_inference_shapes(model, model_config):
 def test_calculate_loss(model, model_config):
     """Tests that loss calculation returns a scalar tensor."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.seq_length
+    seq_len = model_config.context_length
 
     # Inputs
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
@@ -448,7 +450,7 @@ def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
 
 
 def test_padding_keys_are_masked(bert_model):
-    seq_len = bert_model.seq_length
+    seq_len = bert_model.context_length
 
     valid_mask = torch.ones(
         2,
@@ -477,7 +479,7 @@ def test_padding_keys_are_masked(bert_model):
 
 
 def test_causal_and_padding_masks_are_combined(causal_model):
-    seq_len = causal_model.seq_length
+    seq_len = causal_model.context_length
 
     valid_mask = torch.ones(
         1,
@@ -510,7 +512,7 @@ def test_causal_and_padding_masks_are_combined(causal_model):
 
 @pytest.fixture
 def batch(model):
-    seq_len = model.seq_length
+    seq_len = model.context_length
 
     return {
         "cat_col": torch.tensor(
@@ -534,7 +536,7 @@ def batch(model):
 
 @pytest.fixture
 def batch_metadata(model):
-    seq_len = model.seq_length
+    seq_len = model.context_length
     valid_mask = torch.ones(
         2,
         seq_len,
diff --git a/tools/split_file.py b/tools/split_file.py
index 1cceee9a..aa99ee79 100644
--- a/tools/split_file.py
+++ b/tools/split_file.py
@@ -4,7 +4,7 @@
 def split_file_up(
     data: pd.DataFrame,
     number_of_files: int,
-    seq_length: int,
+    context_length: int,
     prediction_length: int,
     folder_path: str,
 ):
@@ -15,18 +15,21 @@ def split_file_up(
     i_offset = 0
     first_start = True
     for i in range(number_of_files):
-        if (i * rows_per_file) >= (seq_length - prediction_length):
+        if (i * rows_per_file) >= (context_length - prediction_length):
             if first_start:
                 start = 0
                 first_start = False
             else:
                 start = (
-                    leading_rows + (i * rows_per_file) - seq_length + prediction_length
+                    leading_rows
+                    + (i * rows_per_file)
+                    - context_length
+                    + prediction_length
                 )
 
             end = leading_rows + ((i + 1) * rows_per_file) + 1
             print(
-                f"{start = }, {end = }, {(start + (seq_length - prediction_length)) = }, {(end - (start + (seq_length - prediction_length))) = })"
+                f"{start = }, {end = }, {(start + (context_length - prediction_length)) = }, {(end - (start + (context_length - prediction_length))) = })"
             )
             data_subset = data.iloc[start:end, :]
             data_subset.to_parquet(

From 0b0296b4d6ac4026f84243d74b7dfd96aa5bad67 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 16:06:27 +0200
Subject: [PATCH 42/81] set max_lookahead to 0 for bert integration tests

---
 tests/configs/hyperparameter-search-bert.yaml |   3 +-
 ...infer-test-categorical-bert-embedding.yaml |   4 +-
 .../configs/infer-test-categorical-bert.yaml  |   4 +-
 ...eprocess-test-categorical-lookahead-0.yaml |  19 ++
 .../configs/train-test-categorical-bert.yaml  |   2 +-
 tests/integration-test-log.txt                |   1 +
 tests/integration/conftest.py                 |  21 ++
 .../integration/test_hyperparameter_search.py |  11 +
 tests/integration/test_inference.py           |   9 +-
 tests/integration/test_preprocessing.py       |  45 ++++
 .../test-data-categorical-1-lookahead-0.csv   | 201 ++++++++++++++++++
 11 files changed, 311 insertions(+), 9 deletions(-)
 create mode 100644 tests/configs/preprocess-test-categorical-lookahead-0.yaml
 create mode 100644 tests/resources/source_data/test-data-categorical-1-lookahead-0.csv

diff --git a/tests/configs/hyperparameter-search-bert.yaml b/tests/configs/hyperparameter-search-bert.yaml
index ff053b1e..b4490817 100644
--- a/tests/configs/hyperparameter-search-bert.yaml
+++ b/tests/configs/hyperparameter-search-bert.yaml
@@ -1,5 +1,5 @@
 project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1-lookahead-0.json
 hp_search_name: test-hp-search-bert
 model_config_write_path: configs
 
@@ -8,6 +8,7 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 context_length: [8]
+max_lookahead: 0
 inference_batch_size: 10
 
 search_strategy: sample
diff --git a/tests/configs/infer-test-categorical-bert-embedding.yaml b/tests/configs/infer-test-categorical-bert-embedding.yaml
index 72c14459..b74491a0 100644
--- a/tests/configs/infer-test-categorical-bert-embedding.yaml
+++ b/tests/configs/infer-test-categorical-bert-embedding.yaml
@@ -1,10 +1,10 @@
 project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical-bert.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1-lookahead-0.json
 model_type: embedding
 training_objective: bert
 model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-embedding-1.pt
-data_path: tests/project_folder/data/test-data-categorical-1-split2
+data_path: tests/project_folder/data/test-data-categorical-1-lookahead-0-split2
 read_format: pt
 write_format: csv
 
diff --git a/tests/configs/infer-test-categorical-bert.yaml b/tests/configs/infer-test-categorical-bert.yaml
index d4d2b4b3..ca23aba1 100644
--- a/tests/configs/infer-test-categorical-bert.yaml
+++ b/tests/configs/infer-test-categorical-bert.yaml
@@ -1,10 +1,10 @@
 project_root: tests/project_folder
 training_config_path: tests/configs/train-test-categorical-bert.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
+metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1-lookahead-0.json
 model_type: generative
 training_objective: bert
 model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-1.pt
-data_path: tests/project_folder/data/test-data-categorical-1-split2
+data_path: tests/project_folder/data/test-data-categorical-1-lookahead-0-split2
 read_format: pt
 write_format: csv
 
diff --git a/tests/configs/preprocess-test-categorical-lookahead-0.yaml b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
new file mode 100644
index 00000000..520f4b8d
--- /dev/null
+++ b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
@@ -0,0 +1,19 @@
+project_root: tests/project_folder
+data_path: tests/resources/source_data/test-data-categorical-1.csv
+merge_output: false
+read_format: csv
+write_format: pt
+selected_columns: null
+
+split_ratios:
+- 0.6
+- 0.2
+- 0.2
+context_length: 8
+max_lookahead: 0
+stride_by_split:
+- 1
+- 1
+- 1
+max_rows: null
+subsequence_start_mode: distribute
diff --git a/tests/configs/train-test-categorical-bert.yaml b/tests/configs/train-test-categorical-bert.yaml
index 64a584fb..af929462 100644
--- a/tests/configs/train-test-categorical-bert.yaml
+++ b/tests/configs/train-test-categorical-bert.yaml
@@ -1,7 +1,7 @@
 project_root: tests/project_folder
 model_name: model-categorical-bert
 read_format: pt
-metadata_config_path: configs/metadata_configs/test-data-categorical-1.json
+metadata_config_path: configs/metadata_configs/test-data-categorical-1-lookahead-0.json
 
 input_columns: [itemId]
 target_columns: [itemId]
diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
index e44783c3..5d4ec365 100644
--- a/tests/integration-test-log.txt
+++ b/tests/integration-test-log.txt
@@ -6,6 +6,7 @@ sequifier preprocess --config-path tests/configs/preprocess-test-categorical.yam
 sequifier preprocess --config-path tests/configs/preprocess-test-real.yaml --data-path tests/resources/source_data/test-data-real-5.csv --selected-columns itemValue supReal1 supReal2 supReal3 supReal4
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical.yaml --data-path tests/resources/source_data/test-data-categorical-50.csv --selected-columns None
 sequifier preprocess --config-path tests/configs/preprocess-test-real.yaml --data-path tests/resources/source_data/test-data-real-50.csv --selected-columns itemValue supReal1 supReal2 supReal3 supReal4 supReal5 supReal6 supReal7 supReal8 supReal9 supReal10 supReal11 supReal12 supReal13 supReal14 supReal15 supReal16 supReal17 supReal18 supReal19 supReal20 supReal21 supReal22 supReal23 supReal24 supReal25 supReal26 supReal27 supReal28 supReal29 supReal30 supReal31 supReal32 supReal33 supReal34 supReal35 supReal36 supReal37 supReal38 supReal39 supReal40 supReal41 supReal42 supReal43 supReal44 supReal45 supReal46 supReal47 supReal48 supReal49
+sequifier preprocess --config-path tests/configs/preprocess-test-categorical-lookahead-0.yaml --data-path tests/resources/source_data/test-data-categorical-1-lookahead-0.csv --selected-columns None
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-multitarget.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-multi-file.yaml
 sequifier preprocess --config-path tests/configs/preprocess-test-categorical-interrupted.yaml
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index 3adcf402..a8575c49 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -60,6 +60,13 @@ def preprocessing_config_path_cat_multitarget():
     )
 
 
+@pytest.fixture(scope="session")
+def preprocessing_config_path_cat_lookahead_0():
+    return os.path.join(
+        "tests", "configs", "preprocess-test-categorical-lookahead-0.yaml"
+    )
+
+
 @pytest.fixture(scope="session")
 def preprocessing_config_path_multi_file():
     return os.path.join("tests", "configs", "preprocess-test-multi-file.yaml")
@@ -288,6 +295,7 @@ def reformat_parameter(attr, param, type):
 def format_configs_locally(
     preprocessing_config_path_cat,
     preprocessing_config_path_cat_multitarget,
+    preprocessing_config_path_cat_lookahead_0,
     preprocessing_config_path_real,
     preprocessing_config_path_multi_file,
     preprocessing_config_path_interrupted,
@@ -323,6 +331,7 @@ def format_configs_locally(
     config_paths = [
         preprocessing_config_path_cat,
         preprocessing_config_path_cat_multitarget,
+        preprocessing_config_path_cat_lookahead_0,
         preprocessing_config_path_real,
         preprocessing_config_path_multi_file,
         preprocessing_config_path_interrupted,
@@ -418,6 +427,7 @@ def run_preprocessing(
     project_root,
     preprocessing_config_path_cat,
     preprocessing_config_path_cat_multitarget,
+    preprocessing_config_path_cat_lookahead_0,
     preprocessing_config_path_real,
     preprocessing_config_path_multi_file,
     preprocessing_config_path_interrupted,
@@ -445,6 +455,17 @@ def run_preprocessing(
             f"sequifier preprocess --config-path {preprocessing_config_path_real} --data-path {data_path_real} --selected-columns {SELECTED_COLUMNS['real'][data_number]}"
         )
 
+    data_path_cat_lookahead_0 = os.path.join(
+        "tests",
+        "resources",
+        "source_data",
+        "test-data-categorical-1-lookahead-0.csv",
+    )
+    run_and_log(
+        f"sequifier preprocess --config-path {preprocessing_config_path_cat_lookahead_0} "
+        f"--data-path {data_path_cat_lookahead_0} --selected-columns None "
+    )
+
     source_path = os.path.join("tests", "resources", "source_configs", "id_maps")
     target_path = os.path.join(project_root, "configs", "id_maps")
     shutil.copytree(source_path, target_path, dirs_exist_ok=True)
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index 187c8312..6cbfbbbf 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -2,6 +2,8 @@
 import json
 import os
 
+import yaml
+
 
 def test_hp_search_grid_outputs(run_hp_search, project_root):
     hp_name = "test-hp-search-grid"
@@ -32,6 +34,15 @@ def test_hp_search_bert_outputs(run_hp_search, project_root):
         len(generated_configs) == 1
     ), f"Expected 1 BERT sample config, found {len(generated_configs)}"
 
+    with open(generated_configs[0], "r") as f:
+        generated_config = yaml.safe_load(f)
+
+    assert generated_config["metadata_config_path"].endswith(
+        "test-data-categorical-1-lookahead-0.json"
+    )
+    assert generated_config["max_lookahead"] == 0
+    assert generated_config["sample_length"] == generated_config["context_length"]
+
 
 def test_hp_search_bayesian_outputs(run_hp_search, project_root):
     hp_name = "test-hp-search-bayesian"
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 8136c84e..bce3bddd 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -8,18 +8,19 @@
 TARGET_VARIABLE_DICT = {"categorical": "itemId", "real": "itemValue"}
 BERT_SEQ_LENGTH = 8
 BERT_EMBEDDING_DIM = 16
+BERT_DATA_NAME = "test-data-categorical-1-lookahead-0"
 
 
 def _categorical_metadata(project_root):
     metadata_path = os.path.join(
-        project_root, "configs", "metadata_configs", "test-data-categorical-1.json"
+        project_root, "configs", "metadata_configs", f"{BERT_DATA_NAME}.json"
     )
     with open(metadata_path, "r") as f:
         return json.load(f)
 
 
 def _bert_inference_metadata(project_root):
-    data_path = os.path.join(project_root, "data", "test-data-categorical-1-split2")
+    data_path = os.path.join(project_root, "data", f"{BERT_DATA_NAME}-split2")
     contents = []
     for root, _, files in os.walk(data_path):
         for file in sorted(files):
@@ -172,7 +173,9 @@ def test_bert_generative_predictions_default_to_context_length(
     bert_predictions, project_root
 ):
     metadata = _categorical_metadata(project_root)
-    valid_values = {str(v) for v in metadata["id_maps"]["itemId"].keys()}
+    valid_values = {str(v) for v in metadata["id_maps"]["itemId"].keys()}.union(
+        {"[unknown]", "[other]"}
+    )
 
     assert bert_predictions.height > 0
     assert set(bert_predictions["itemId"].to_list()).issubset(valid_values)
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 999c9cbf..3cabc262 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -200,6 +200,8 @@ def read_preprocessing_outputs(path, variant):
         ):
             return load_parquet_folder_outputs(path)
         return load_pt_outputs(path)
+    else:
+        raise ValueError(f"Invalid variant: {variant}")
 
 
 @pytest.fixture()
@@ -260,6 +262,49 @@ def test_preprocessed_data_categorical(data_splits):
                 ), f'{list(group["5"].to_numpy()[:-j]) = } != {list(group["6"].to_numpy()[j:]) = }'
 
 
+def test_preprocessed_data_categorical_lookahead_0(run_preprocessing, project_root):
+    metadata_path = os.path.join(
+        project_root,
+        "configs",
+        "metadata_configs",
+        "test-data-categorical-1-lookahead-0.json",
+    )
+    with open(metadata_path, "r") as f:
+        metadata_config = json.load(f)
+
+    assert metadata_config["max_lookahead"] == 0
+    assert metadata_config["sample_length"] == metadata_config["context_length"]
+
+    for split in range(3):
+        data = read_preprocessing_outputs(
+            os.path.join(
+                project_root,
+                "data",
+                f"test-data-categorical-1-lookahead-0-split{split}",
+            ),
+            "categorical",
+        )
+
+        assert data is not None
+        assert data.shape[1] == 13
+        assert "8" not in data.columns
+        assert set(str(i) for i in range(8)).issubset(set(data.columns))
+
+        for sequence_id, group in data.group_by("sequenceId"):
+            assert np.all(
+                np.abs(group["1"].to_numpy()[:-1] - group["2"].to_numpy()[1:]) < 0.0001
+            ), (
+                f"{sequence_id = }: {list(group['1'].to_numpy()[:-1]) = } "
+                f"!= {list(group['2'].to_numpy()[1:]) = }"
+            )
+            assert np.all(
+                np.abs(group["5"].to_numpy()[:-1] - group["6"].to_numpy()[1:]) < 0.0001
+            ), (
+                f"{sequence_id = }: {list(group['5'].to_numpy()[:-1]) = } "
+                f"!= {list(group['6'].to_numpy()[1:]) = }"
+            )
+
+
 def unnest(list_var):
     return [x for y in list_var for x in y]
 
diff --git a/tests/resources/source_data/test-data-categorical-1-lookahead-0.csv b/tests/resources/source_data/test-data-categorical-1-lookahead-0.csv
new file mode 100644
index 00000000..9c89f83e
--- /dev/null
+++ b/tests/resources/source_data/test-data-categorical-1-lookahead-0.csv
@@ -0,0 +1,201 @@
+sequenceId,itemPosition,itemId
+0,0,112
+0,1,104
+0,2,129
+0,3,121
+0,4,109
+0,5,111
+0,6,121
+0,7,109
+0,8,126
+0,9,112
+0,10,104
+0,11,103
+0,12,119
+0,13,103
+0,14,105
+0,15,121
+0,16,107
+0,17,128
+0,18,128
+0,19,116
+0,20,106
+0,21,103
+0,22,119
+0,23,105
+0,24,100
+1,25,113
+1,26,116
+1,27,105
+1,28,113
+1,29,108
+1,30,118
+1,31,102
+1,32,102
+1,33,121
+1,34,104
+1,35,111
+1,36,110
+1,37,102
+1,38,102
+1,39,120
+1,40,109
+1,41,101
+1,42,111
+1,43,127
+1,44,105
+1,45,115
+1,46,104
+2,47,112
+2,48,124
+2,49,121
+2,50,124
+2,51,129
+2,52,103
+2,53,119
+2,54,118
+2,55,110
+2,56,102
+2,57,121
+2,58,108
+2,59,101
+2,60,102
+2,61,110
+2,62,106
+2,63,102
+2,64,110
+2,65,109
+2,66,115
+3,67,122
+3,68,129
+3,69,101
+3,70,117
+3,71,120
+3,72,127
+3,73,122
+3,74,112
+3,75,127
+3,76,115
+3,77,117
+3,78,102
+3,79,120
+3,80,108
+3,81,126
+3,82,116
+3,83,102
+3,84,121
+3,85,118
+3,86,105
+4,87,113
+4,88,127
+4,89,105
+4,90,104
+4,91,106
+4,92,118
+4,93,112
+4,94,122
+4,95,105
+4,96,121
+4,97,119
+4,98,126
+4,99,120
+4,100,103
+4,101,117
+4,102,114
+5,103,120
+5,104,123
+5,105,127
+5,106,105
+5,107,112
+5,108,108
+5,109,123
+5,110,108
+5,111,129
+5,112,122
+5,113,117
+5,114,122
+5,115,109
+5,116,112
+5,117,126
+5,118,120
+6,119,127
+6,120,119
+6,121,127
+6,122,102
+6,123,113
+6,124,119
+6,125,103
+6,126,126
+6,127,128
+6,128,113
+6,129,111
+6,130,117
+6,131,124
+6,132,100
+6,133,119
+6,134,117
+6,135,103
+6,136,110
+6,137,108
+6,138,125
+6,139,107
+7,140,111
+7,141,112
+7,142,113
+7,143,113
+7,144,113
+7,145,121
+7,146,100
+7,147,129
+7,148,100
+7,149,125
+7,150,118
+7,151,121
+7,152,127
+7,153,121
+7,154,129
+7,155,113
+7,156,110
+8,157,101
+8,158,119
+8,159,103
+8,160,113
+8,161,105
+8,162,111
+8,163,124
+8,164,108
+8,165,118
+8,166,124
+8,167,109
+8,168,121
+8,169,110
+8,170,120
+8,171,114
+8,172,107
+8,173,100
+8,174,111
+8,175,107
+8,176,109
+8,177,114
+8,178,102
+8,179,126
+8,180,115
+9,181,128
+9,182,120
+9,183,103
+9,184,116
+9,185,122
+9,186,100
+9,187,128
+9,188,102
+9,189,122
+9,190,110
+9,191,122
+9,192,107
+9,193,123
+9,194,103
+9,195,128
+9,196,106
+9,197,108
+9,198,119
+9,199,129

From 6a95a7c48f4938fad468814f14eaee00541e1069 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 18:52:52 +0200
Subject: [PATCH 43/81] fix tests, reverse metadata export order

---
 src/sequifier/config/preprocess_config.py     |  2 +
 src/sequifier/helpers.py                      | 13 +--
 src/sequifier/infer.py                        | 95 ++++++++-----------
 src/sequifier/preprocess.py                   | 12 +--
 src/sequifier/train.py                        | 10 +-
 ...uifier_dataset_from_folder_parquet_lazy.py | 32 +++++--
 ...t_sequifier_dataset_from_folder_pt_lazy.py | 51 ++++++----
 tests/unit/test_infer.py                      | 22 ++++-
 tests/unit/test_train.py                      |  8 ++
 9 files changed, 135 insertions(+), 110 deletions(-)

diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index a2532345..ed1aa518 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -192,6 +192,8 @@ def validate_subsequence_start_mode(cls, v: str) -> str:
     def validate_mask_column_requires_metadata(self) -> "PreprocessorModel":
         if self.mask_column is not None and self.metadata_config_path is None:
             raise ValueError("metadata_config_path must be set when mask_column is set")
+        if self.mask_column in ("sequenceId", "itemPosition"):
+            raise ValueError("mask_column cannot be sequenceId or itemPosition")
         return self
 
     def __init__(self, **kwargs):
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 335406ce..d228b456 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -114,17 +114,10 @@ def validate_stored_window_width(tensor: Tensor, sample_length: int) -> None:
 def sequence_layout_from_metadata(
     metadata: dict, context_length: int
 ) -> SequenceLayout:
-    metadata_context_length = int(metadata.get("context_length", context_length))
-    if metadata_context_length != context_length:
-        raise ValueError(
-            f"Configured context_length={context_length} does not match preprocessed "
-            f"metadata context_length={metadata_context_length}."
-        )
+    metadata_context_length = int(metadata["context_length"])
 
-    max_lookahead = int(metadata.get("max_lookahead", 1))
-    sample_length = int(
-        metadata.get("sample_length", metadata_context_length + max_lookahead)
-    )
+    max_lookahead = int(metadata["max_lookahead"])
+    sample_length = int(metadata["sample_length"])
     if sample_length != metadata_context_length + max_lookahead:
         raise ValueError(
             "Invalid sequence layout metadata: sample_length must equal "
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 6d481926..8abdef3d 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -305,7 +305,7 @@ def calculate_item_positions(
 @beartype
 def _flatten_bert_target_valid_mask(
     config: InfererModel,
-    metadata: Optional[dict[str, Any]],
+    metadata: dict[str, Any],
     prediction_length: int,
 ) -> Optional[np.ndarray]:
     if config.training_objective != "bert" or not metadata:
@@ -768,7 +768,7 @@ def infer_generative(
                 )
 
             probs, preds = get_probs_preds_from_dict(
-                config, inferer, sequences_dict, total_steps, metadata=metadata
+                config, inferer, sequences_dict, metadata, total_steps
             )
 
             prediction_length = inferer.prediction_length  # Get prediction_length
@@ -923,7 +923,7 @@ def get_embeddings_pt(
     config: Any,
     inferer: "Inferer",
     data: dict[str, torch.Tensor],
-    metadata: Optional[dict[str, torch.Tensor]] = None,
+    metadata: dict[str, torch.Tensor],
 ) -> np.ndarray:
     """Generates embeddings from a batch of PyTorch tensor data.
 
@@ -950,9 +950,7 @@ def get_embeddings_pt(
         for key, val in data.items()
         if key in config.input_columns
     }
-    metadata_np = (
-        {key: val.numpy() for key, val in metadata.items()} if metadata else None
-    )
+    metadata_np = {key: val.numpy() for key, val in metadata.items()}
     embeddings = inferer.infer_embedding(X, metadata=metadata_np)
     return embeddings
 
@@ -962,8 +960,8 @@ def get_probs_preds_from_dict(
     config: Any,
     inferer: "Inferer",
     data: dict[str, torch.Tensor],
+    metadata: dict[str, torch.Tensor],
     total_steps: int = 1,
-    metadata: Optional[dict[str, torch.Tensor]] = None,
 ) -> tuple[Optional[dict[str, np.ndarray]], dict[str, np.ndarray]]:
     """Generates predictions from PyTorch tensor data, supporting autoregression.
 
@@ -1006,9 +1004,7 @@ def get_probs_preds_from_dict(
         for key, tensor in data.items()
         if key in config.input_columns
     }
-    metadata_np = (
-        {key: tensor.numpy() for key, tensor in metadata.items()} if metadata else None
-    )
+    metadata_np = {key: tensor.numpy() for key, tensor in metadata.items()}
     all_probs_list = {col: [] for col in target_cols}
     all_preds_list = {col: [] for col in target_cols}
 
@@ -1017,14 +1013,18 @@ def get_probs_preds_from_dict(
     for i in range(total_steps):
         if config.output_probabilities:
             probs_for_step = inferer.infer_generative(
-                X, return_probs=True, metadata=metadata_for_step
+                X,
+                metadata_for_step,
+                return_probs=True,
+            )
+            preds_for_step = inferer.infer_generative(
+                None, metadata_for_step, probs_for_step
             )
-            preds_for_step = inferer.infer_generative(None, probs_for_step)
             for col in target_cols:
                 all_probs_list[col].append(probs_for_step[col])
         else:
             preds_for_step = inferer.infer_generative(
-                X, return_probs=False, metadata=metadata_for_step
+                X, metadata_for_step, return_probs=False
             )
 
         for col in target_cols:
@@ -1106,11 +1106,7 @@ def get_embeddings(
         config.target_offset,
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
-    metadata_np = (
-        {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
-        if metadata
-        else None
-    )
+    metadata_np = {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
     del data
 
     embeddings = inferer.infer_embedding(X, metadata=metadata_np)
@@ -1159,19 +1155,15 @@ def get_probs_preds_from_df(
         config.target_offset,
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
-    metadata_np = (
-        {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
-        if metadata
-        else None
-    )
+    metadata_np = {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
     del data
 
     if config.output_probabilities:
-        probs = inferer.infer_generative(X, return_probs=True, metadata=metadata_np)
-        preds = inferer.infer_generative(None, probs)
+        probs = inferer.infer_generative(X, metadata_np, return_probs=True)
+        preds = inferer.infer_generative(None, metadata_np, probs)
     else:
         probs = None
-        preds = inferer.infer_generative(X, metadata=metadata_np)
+        preds = inferer.infer_generative(X, metadata_np)
 
     return (probs, preds)
 
@@ -1294,8 +1286,8 @@ def get_probs_preds_autoregression(
         config,
         inferer,
         head_data,
-        total_steps=config.autoregression_total_steps,
-        metadata=metadata,
+        metadata,
+        config.autoregression_total_steps,
     )
 
     # 4. Generate the final output arrays using the perfectly aligned bases
@@ -1456,7 +1448,7 @@ def invert_normalization(
     def infer_embedding(
         self,
         x: dict[str, np.ndarray],
-        metadata: Optional[dict[str, np.ndarray]] = None,
+        metadata: dict[str, np.ndarray],
     ) -> np.ndarray:
         """Performs inference with an embedding model.
 
@@ -1473,7 +1465,7 @@ def infer_embedding(
         """
         assert x is not None
         size = x[list(x.keys())[0]].shape[0]
-        embedding = self.adjust_and_infer_embedding(x, size, metadata=metadata)
+        embedding = self.adjust_and_infer_embedding(x, size, metadata)
 
         return embedding
 
@@ -1481,9 +1473,9 @@ def infer_embedding(
     def infer_generative(
         self,
         x: Optional[dict[str, np.ndarray]],
+        metadata: dict[str, np.ndarray],
         probs: Optional[dict[str, np.ndarray]] = None,
         return_probs: bool = False,
-        metadata: Optional[dict[str, np.ndarray]] = None,
     ) -> dict[str, np.ndarray]:
         """Performs generative inference, returning probabilities or predictions.
 
@@ -1529,7 +1521,7 @@ def infer_generative(
                     f"Not all keys in x are in probs - {x.keys() = } != {probs.keys() = }. Full inference is executed."
                 )
 
-            outs = self.adjust_and_infer_generative(x, size, metadata=metadata)
+            outs = self.adjust_and_infer_generative(x, size, metadata)
 
             for target_column, target_outs in outs.items():
                 if np.any(target_outs == np.inf):
@@ -1571,7 +1563,7 @@ def adjust_and_infer_embedding(
         self,
         x: dict[str, np.ndarray],
         size: int,
-        metadata: Optional[dict[str, np.ndarray]] = None,
+        metadata: dict[str, np.ndarray],
     ):
         """Handles batching and backend-specific calls for embedding inference.
 
@@ -1590,11 +1582,10 @@ def adjust_and_infer_embedding(
         if self.inference_model_type == "onnx":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=True)
-            metadata_adjusted = (
-                self.prepare_inference_batches(metadata, pad_to_batch_size=True)
-                if metadata
-                else [None] * len(x_adjusted)
+            metadata_adjusted = self.prepare_inference_batches(
+                metadata, pad_to_batch_size=True
             )
+
             inference_batch_embeddings = [
                 self.infer_pure(x_sub, metadata_sub)[0]
                 for x_sub, metadata_sub in zip(x_adjusted, metadata_adjusted)
@@ -1602,11 +1593,10 @@ def adjust_and_infer_embedding(
             embeddings = np.concatenate(inference_batch_embeddings, axis=0)[:size]
         elif self.inference_model_type == "pt":
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=False)
-            metadata_adjusted = (
-                self.prepare_inference_batches(metadata, pad_to_batch_size=False)
-                if metadata
-                else None
+            metadata_adjusted = self.prepare_inference_batches(
+                metadata, pad_to_batch_size=False
             )
+
             embeddings = infer_with_embedding_model(
                 self.inference_model,
                 x_adjusted,
@@ -1624,7 +1614,7 @@ def adjust_and_infer_generative(
         self,
         x: dict[str, np.ndarray],
         size: int,
-        metadata: Optional[dict[str, np.ndarray]] = None,
+        metadata: dict[str, np.ndarray],
     ):
         """Handles batching and backend-specific calls for generative inference.
 
@@ -1645,10 +1635,8 @@ def adjust_and_infer_generative(
         if self.inference_model_type == "onnx":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=True)
-            metadata_adjusted = (
-                self.prepare_inference_batches(metadata, pad_to_batch_size=True)
-                if metadata
-                else [None] * len(x_adjusted)
+            metadata_adjusted = self.prepare_inference_batches(
+                metadata, pad_to_batch_size=True
             )
             out_subs = [
                 dict(zip(self.target_columns, self.infer_pure(x_sub, metadata_sub)))
@@ -1663,10 +1651,8 @@ def adjust_and_infer_generative(
         elif self.inference_model_type == "pt":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=False)
-            metadata_adjusted = (
-                self.prepare_inference_batches(metadata, pad_to_batch_size=False)
-                if metadata
-                else None
+            metadata_adjusted = self.prepare_inference_batches(
+                metadata, pad_to_batch_size=False
             )
             outs = infer_with_generative_model(
                 self.inference_model,
@@ -1731,7 +1717,7 @@ def prepare_inference_batches(
     def infer_pure(
         self,
         x: dict[str, np.ndarray],
-        metadata: Optional[dict[str, np.ndarray]] = None,
+        metadata: dict[str, np.ndarray],
     ) -> list[np.ndarray]:
         """Performs a single inference pass using the ONNX session.
 
@@ -1748,15 +1734,10 @@ def infer_pure(
             ONNX model.
         """
         metadata = metadata or {}
-        reference_shape = next(iter(x.values())).shape
         ort_inputs = {}
         for session_input in self.ort_session.get_inputs():
             input_name = session_input.name
-            if input_name == "attention_valid_mask":
-                value = metadata.get(input_name)
-                if value is None:
-                    value = np.ones(reference_shape[:2], dtype=np.bool_)
-            elif input_name in metadata:
+            if input_name in metadata:
                 value = metadata[input_name]
             elif input_name.endswith("_in") and input_name[:-3] in x:
                 feature_column = input_name[:-3]
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 962dd204..cdbc1740 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -236,15 +236,15 @@ def __init__(
                 col_types=col_types,
             )
             data = _apply_mask_column(data, data_columns, col_types, self.mask_column)
-            self._export_metadata(
-                id_maps, n_classes, col_types, selected_columns_statistics
-            )
 
             self._write_or_validate_resume_manifest(
                 selected_columns,
                 write_format,
                 data_columns,
             )
+            self._export_metadata(
+                id_maps, n_classes, col_types, selected_columns_statistics
+            )
 
             schema = self._create_schema(col_types, self.layout.sample_length)
 
@@ -334,14 +334,14 @@ def __init__(
                 for col in id_maps:
                     col_types[col] = "Int64"
 
-            self._export_metadata(
-                id_maps, n_classes, col_types, selected_columns_statistics
-            )
             self._write_or_validate_resume_manifest(
                 selected_columns,
                 write_format,
                 data_columns,
             )
+            self._export_metadata(
+                id_maps, n_classes, col_types, selected_columns_statistics
+            )
             schema = self._create_schema(col_types, self.layout.sample_length)
 
             self._process_batches_multiple_files(
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 9b37f3f1..66d7ff88 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1714,7 +1714,7 @@ def _calculate_loss(
         self,
         output: dict[str, Tensor],
         targets: dict[str, Tensor],
-        metadata: Optional[dict[str, Tensor]] = None,
+        metadata: dict[str, Tensor],
     ) -> tuple[Tensor, dict[str, Tensor]]:
         """Calculates the loss for the given output and targets.
 
@@ -1738,9 +1738,9 @@ def _calculate_loss(
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
-        valid_mask = metadata["attention_valid_mask"].bool()  # type: ignore
+        valid_mask = metadata["target_valid_mask"].bool()  # type: ignore
 
-        if metadata is not None and "bert_mask" in metadata:
+        if "bert_mask" in metadata:
             valid_mask = valid_mask & metadata["bert_mask"].bool()
 
         mask = valid_mask.T.contiguous().reshape(-1)
@@ -2555,7 +2555,7 @@ def infer_with_embedding_model(
     device: str,
     size: int,
     target_columns: list[str],
-    metadata: Optional[list[dict[str, np.ndarray]]] = None,
+    metadata: list[dict[str, np.ndarray]],
 ) -> np.ndarray:
     """Performs inference with an embedding model.
 
@@ -2615,7 +2615,7 @@ def infer_with_generative_model(
     device: str,
     size: int,
     target_columns: list[str],
-    metadata: Optional[list[dict[str, np.ndarray]]] = None,
+    metadata: list[dict[str, np.ndarray]],
 ) -> dict[str, np.ndarray]:
     """Performs inference with a generative model.
 
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index d51685bf..856b386b 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -9,14 +9,29 @@
     SequifierDatasetFromFolderParquetLazy,
 )
 
+CONTEXT_LENGTH = 2
+MAX_LOOKAHEAD = 1
+SAMPLE_LENGTH = CONTEXT_LENGTH + MAX_LOOKAHEAD
+
+
+def _folder_metadata(total_samples, batch_files):
+    return {
+        "total_samples": total_samples,
+        "batch_files": batch_files,
+        "context_length": CONTEXT_LENGTH,
+        "max_lookahead": MAX_LOOKAHEAD,
+        "sample_length": SAMPLE_LENGTH,
+        "sequence_layout_version": 1,
+    }
+
 
 @pytest.fixture
 def mock_config():
     config = MagicMock()
     config.project_root = "."
     config.seed = 42
-    config.context_length = 2
-    config.sample_length = 3
+    config.context_length = CONTEXT_LENGTH
+    config.sample_length = SAMPLE_LENGTH
     config.column_types = {"item": "Float64"}
     config.training_spec.batch_size = 5
     config.training_spec.num_workers = 0
@@ -57,7 +72,7 @@ def dataset_path(tmp_path):
         df.write_parquet(data_dir / filename)
         batch_files.append({"path": filename, "samples": 10})
 
-    metadata = {"total_samples": 40, "batch_files": batch_files}
+    metadata = _folder_metadata(40, batch_files)
 
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
@@ -119,10 +134,7 @@ def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
     )
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(
-            {
-                "total_samples": 5,
-                "batch_files": [{"path": "file.parquet", "samples": 5}],
-            },
+            _folder_metadata(5, [{"path": "file.parquet", "samples": 5}]),
             f,
         )
 
@@ -186,7 +198,7 @@ def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
         {"path": "file_1.parquet", "samples": 15},
         {"path": "file_2.parquet", "samples": 10},
     ]
-    metadata = {"total_samples": 25, "batch_files": batch_files}
+    metadata = _folder_metadata(25, batch_files)
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
 
@@ -228,7 +240,7 @@ def test_oversampling_strategy(mock_config, tmp_path):
         {"path": "file_1.parquet", "samples": 15},
         {"path": "file_2.parquet", "samples": 10},
     ]
-    metadata = {"total_samples": 25, "batch_files": batch_files}
+    metadata = _folder_metadata(25, batch_files)
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
 
@@ -272,7 +284,7 @@ def test_undersampling_strategy(mock_config, tmp_path):
         {"path": "file_1.parquet", "samples": 15},
         {"path": "file_2.parquet", "samples": 10},
     ]
-    metadata = {"total_samples": 25, "batch_files": batch_files}
+    metadata = _folder_metadata(25, batch_files)
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
 
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index d5688ddc..6cec3c92 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -8,6 +8,21 @@
     SequifierDatasetFromFolderPtLazy,
 )
 
+CONTEXT_LENGTH = 5
+MAX_LOOKAHEAD = 1
+SAMPLE_LENGTH = CONTEXT_LENGTH + MAX_LOOKAHEAD
+
+
+def _folder_metadata(total_samples, batch_files):
+    return {
+        "total_samples": total_samples,
+        "batch_files": batch_files,
+        "context_length": CONTEXT_LENGTH,
+        "max_lookahead": MAX_LOOKAHEAD,
+        "sample_length": SAMPLE_LENGTH,
+        "sequence_layout_version": 1,
+    }
+
 
 # --- Fixtures ---
 @pytest.fixture
@@ -19,8 +34,8 @@ def mock_config(tmp_path):
     config.training_spec.sampling_strategy = "exact"
     config.training_spec.num_workers = 0
     config.seed = 42
-    config.context_length = 5
-    config.sample_length = 6
+    config.context_length = CONTEXT_LENGTH
+    config.sample_length = SAMPLE_LENGTH
     config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
     config.training_spec.data_offset = 1
@@ -36,15 +51,15 @@ def dataset_path(tmp_path):
 
     # Create dummy metadata.json
     # We define 4 files with 10 samples each = 40 total samples
-    metadata = {
-        "total_samples": 40,
-        "batch_files": [
+    metadata = _folder_metadata(
+        40,
+        [
             {"path": "file1.pt", "samples": 10},
             {"path": "file2.pt", "samples": 10},
             {"path": "file3.pt", "samples": 10},
             {"path": "file4.pt", "samples": 10},
         ],
-    }
+    )
 
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
@@ -229,15 +244,15 @@ def test_exact_strategy_uneven_files_exception(
 
     # Rank 0 will get file1 (10) + file3 (10) = 20 samples
     # Rank 1 will get file2 (10) + file4 (5) = 15 samples
-    metadata = {
-        "total_samples": 35,
-        "batch_files": [
+    metadata = _folder_metadata(
+        35,
+        [
             {"path": "file1.pt", "samples": 10},
             {"path": "file2.pt", "samples": 10},
             {"path": "file3.pt", "samples": 10},
             {"path": "file4.pt", "samples": 5},
         ],
-    }
+    )
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
 
@@ -265,14 +280,14 @@ def test_oversampling_strategy(
 
     # Rank 0 gets file1 (10) + file3 (5) = 15 samples
     # Rank 1 gets file2 (10) = 10 samples
-    metadata = {
-        "total_samples": 25,
-        "batch_files": [
+    metadata = _folder_metadata(
+        25,
+        [
             {"path": "file1.pt", "samples": 10},
             {"path": "file2.pt", "samples": 10},
             {"path": "file3.pt", "samples": 5},
         ],
-    }
+    )
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
 
@@ -311,14 +326,14 @@ def test_undersampling_strategy(
 
     # Rank 0 gets file1 (10) + file3 (5) = 15 samples
     # Rank 1 gets file2 (10) = 10 samples
-    metadata = {
-        "total_samples": 25,
-        "batch_files": [
+    metadata = _folder_metadata(
+        25,
+        [
             {"path": "file1.pt", "samples": 10},
             {"path": "file2.pt", "samples": 10},
             {"path": "file3.pt", "samples": 5},
         ],
-    }
+    )
     with open(data_dir / "metadata.json", "w") as f:
         json.dump(metadata, f)
 
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 5cc76580..568a434e 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -337,13 +337,15 @@ def get_inputs(self):
             return [Input("unexpected_input")]
 
     mock_inferer.ort_session = Session()
+    metadata = {}
 
     with pytest.raises(ValueError, match="Could not map ONNX input"):
         mock_inferer.infer_pure(
             {
                 "cat_col": np.array([[1, 2]], dtype=np.int64),
                 "real_col": np.array([[1.0, 2.0]], dtype=np.float32),
-            }
+            },
+            metadata,
         )
 
 
@@ -453,6 +455,14 @@ def _mock_bert_inferer(config):
     )
 
 
+def _dummy_attention_metadata(batch_size, context_length):
+    return {
+        "attention_valid_mask": torch.ones(
+            (batch_size, context_length), dtype=torch.bool
+        )
+    }
+
+
 def test_bert_generative_inference_filters_left_padded_positions(tmp_path):
     config = _bert_inference_config(tmp_path)
     config.output_probabilities = True
@@ -641,6 +651,7 @@ def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
     and accurately shifts the input tensor to append the latest prediction.
     """
     initial_data = {"target_col": torch.tensor([[1.0, 2.0, 3.0], [10.0, 20.0, 30.0]])}
+    metadata = _dummy_attention_metadata(batch_size=2, context_length=3)
 
     # Patch the method on the actual instance
     with patch.object(ar_inferer, "infer_generative") as mock_infer:
@@ -653,7 +664,7 @@ def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
 
         total_steps = 2
         probs, preds = get_probs_preds_from_dict(
-            ar_config, ar_inferer, initial_data, total_steps=total_steps
+            ar_config, ar_inferer, initial_data, metadata, total_steps=total_steps
         )
 
         # 1. Verify Loop Count
@@ -700,12 +711,14 @@ def test_get_probs_preds_from_dict_with_probabilities(
     # What we want our mock sample_with_cumsum to return
     mock_sample.return_value = np.array([1])
 
+    metadata = _dummy_attention_metadata(batch_size=1, context_length=2)
+
     # Mock the *inner* backend call, allowing infer_generative to execute its actual logic
     with patch.object(
         ar_inferer, "adjust_and_infer_generative", return_value=dummy_logits
     ) as mock_adjust:
         probs, preds = get_probs_preds_from_dict(
-            ar_config, ar_inferer, initial_data, total_steps=1
+            ar_config, ar_inferer, initial_data, metadata, total_steps=1
         )
 
         assert mock_adjust.call_count == 1
@@ -735,12 +748,13 @@ def test_get_probs_preds_from_dict_ignores_unselected_columns(ar_config, ar_infe
         "target_col": torch.tensor([[1.0, 2.0]]),
         "ignored_col": torch.tensor([[9.0, 9.0]]),
     }
+    metadata = _dummy_attention_metadata(batch_size=1, context_length=2)
 
     with patch.object(ar_inferer, "infer_generative") as mock_infer:
         mock_infer.return_value = {"target_col": np.array([[3.0]])}
 
         _, _ = get_probs_preds_from_dict(
-            ar_config, ar_inferer, initial_data, total_steps=1
+            ar_config, ar_inferer, initial_data, metadata, total_steps=1
         )
 
         first_call_args = mock_infer.call_args_list[0][0][0]
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 054d0956..875e9109 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -252,6 +252,10 @@ def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
+                "context_length": config_values["context_length"],
+                "max_lookahead": 1,
+                "sample_length": config_values["context_length"] + 1,
+                "sequence_layout_version": 1,
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
@@ -281,6 +285,10 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
+                "context_length": config_values["context_length"],
+                "max_lookahead": 1,
+                "sample_length": config_values["context_length"] + 1,
+                "sequence_layout_version": 1,
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],

From 6bcfe78d39c48cd216b7dc9df56ffca9649cbef4 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 19:03:00 +0200
Subject: [PATCH 44/81] Update outputs

---
 ...cal-bert-best-embedding-1-0-embeddings.csv |   9 +-
 ...cal-bert-best-embedding-1-1-embeddings.csv |   9 +-
 ...cal-bert-best-embedding-1-2-embeddings.csv |   7 +-
 ...cal-bert-best-embedding-1-3-embeddings.csv |  14 +-
 ...cal-bert-best-embedding-1-4-embeddings.csv |   7 +-
 ...cal-bert-best-embedding-1-5-embeddings.csv |  16 +-
 ...cal-bert-best-embedding-1-6-embeddings.csv |   9 +-
 ...cal-bert-best-embedding-1-7-embeddings.csv |   7 +-
 ...-categorical-bert-best-1-0-predictions.csv |   3 +-
 ...-categorical-bert-best-1-1-predictions.csv |   3 +-
 ...-categorical-bert-best-1-2-predictions.csv |   3 +-
 ...-categorical-bert-best-1-3-predictions.csv |   8 +-
 ...-categorical-bert-best-1-4-predictions.csv |   5 +-
 ...-categorical-bert-best-1-5-predictions.csv |  10 +-
 ...-categorical-bert-best-1-6-predictions.csv |   5 +-
 ...-categorical-bert-best-1-7-predictions.csv |   5 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...uifier-model-real-1-best-3-predictions.csv |  48 +--
 ...uifier-model-real-3-best-3-predictions.csv |  48 +--
 ...uifier-model-real-5-best-3-predictions.csv |  48 +--
 ...ifier-model-real-50-best-3-predictions.csv |  48 +--
 ...custom-eval-run-1-best-1-0-predictions.csv |  58 +--
 ...custom-eval-run-1-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-1-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-1-best-1-7-predictions.csv |  60 +--
 ...custom-eval-run-2-best-1-0-predictions.csv |   2 +-
 ...custom-eval-run-2-best-1-1-predictions.csv |  10 +-
 ...custom-eval-run-2-best-1-2-predictions.csv |  26 +-
 ...custom-eval-run-2-best-1-3-predictions.csv |  30 +-
 ...custom-eval-run-2-best-1-4-predictions.csv |   4 +-
 ...custom-eval-run-2-best-1-5-predictions.csv |  70 +--
 ...custom-eval-run-2-best-1-6-predictions.csv |   2 +-
 ...custom-eval-run-2-best-1-7-predictions.csv |  32 +-
 ...custom-eval-run-3-best-1-0-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-1-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-2-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-3-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-4-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-5-predictions.csv | 120 +++---
 ...custom-eval-run-3-best-1-6-predictions.csv |  60 +--
 ...custom-eval-run-3-best-1-7-predictions.csv |  60 +--
 ...ategorical-bert-best-1-0-probabilities.csv |   9 +-
 ...ategorical-bert-best-1-1-probabilities.csv |   9 +-
 ...ategorical-bert-best-1-2-probabilities.csv |   7 +-
 ...ategorical-bert-best-1-3-probabilities.csv |  14 +-
 ...ategorical-bert-best-1-4-probabilities.csv |   7 +-
 ...ategorical-bert-best-1-5-probabilities.csv |  16 +-
 ...ategorical-bert-best-1-6-probabilities.csv |   9 +-
 ...ategorical-bert-best-1-7-probabilities.csv |   7 +-
 53 files changed, 1097 insertions(+), 1067 deletions(-)

diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
index 79067fd4..e3eb7be7 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
@@ -1,5 +1,6 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-0,0,0,-0.46009058,0.45777348,-0.22698353,-0.25634044,0.05071858,-0.7203443,-0.5872563,-0.7930882,1.2241583,-0.81149936,-0.7057305,0.83234805,1.888106,-0.4073704,2.095673,-1.5786989
-0,0,1,1.2350427,-0.07666658,1.6687232,-0.10760293,-0.34233996,-0.6405651,-0.34899667,1.3310788,-1.7401463,0.48453188,-0.39308575,-0.38906077,0.09585311,0.9867051,0.3474074,-2.1432836
-0,0,2,-0.04062133,-1.1088247,-1.103962,-0.7512283,-0.3894962,0.40089244,1.3842655,-0.45180556,1.6632154,0.7094805,-1.2707394,1.2500856,-1.5106493,1.293199,0.25951874,-0.33003083
-0,0,3,0.95941216,0.026796307,0.18043526,-0.27425033,-0.8116876,0.59967464,0.15953179,0.6978008,0.6560124,-2.234662,-0.5438349,-1.4812955,-0.07771553,2.3249497,0.22156315,-0.4244198
+0,0,0,-0.5678054,0.30644566,-0.48223504,-0.37569737,-0.3008679,-0.52658165,-0.32838538,-0.8943048,0.95665604,-0.8637694,-0.3740866,0.95319647,1.8752165,-0.35496402,2.3600826,-1.429543
+0,0,1,0.78683025,-0.48157683,2.1831813,0.00026024866,0.19607927,-1.008873,0.06488745,0.6515402,-1.6177354,1.4758716,-0.756111,-0.30458173,0.38618582,0.34799105,-0.17682126,-1.7913008
+0,0,2,-1.0294193,-1.9363327,-0.10089664,-1.256774,-0.57952416,0.6647573,1.0020547,0.60435945,1.4913826,0.47637454,-0.52362984,1.2796938,-1.1521349,0.8179086,0.8369039,-0.61289996
+0,0,3,1.29303,0.5742223,-0.20042029,0.03720812,-1.1154803,0.26275542,-0.0761054,0.65055084,0.6538049,-1.9854333,-0.64701724,-1.4201233,0.422354,2.221934,-0.06508473,-0.6750595
+0,0,4,-0.14699927,-1.5303152,0.3594103,-0.9477121,-0.4332569,-0.068542324,0.5149603,0.34177327,1.5040935,0.8814393,-0.73049664,-0.66137636,-1.0386573,2.0420604,-1.2320555,1.142538
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
index f1ad3c83..b58246ec 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
@@ -1,5 +1,6 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-1,0,0,0.5496252,-0.29538283,1.4190812,1.0952207,-1.0380151,-0.95464176,0.25532782,2.3793674,0.15770668,-0.4943242,-0.23793282,-0.93078727,-0.4377435,0.48551127,-1.7345339,-0.24836543
-1,0,1,1.1703843,-1.293456,-0.009221493,1.1512711,-1.2613713,-0.14320165,0.21250723,1.3096029,-0.19192691,-1.8777579,1.3773452,0.84090453,0.026053404,0.5442735,-1.1615742,-0.7001723
-1,0,2,1.3515635,0.47997367,-0.15122601,0.20858693,-0.96445537,0.32341665,-0.17822078,0.5217958,0.63256186,-1.831718,-0.63131493,-1.3727977,0.07998669,2.4652927,-0.28591567,-0.66949725
-1,0,3,1.2581677,-1.3701472,1.279228,0.34158385,-0.1695098,-0.12549305,-0.9352961,-0.7049133,0.7994457,-0.7667792,0.9320275,1.0522128,-1.8637831,-0.15688421,-0.91187423,1.3544698
+1,0,0,0.6880688,-0.49868053,1.1768215,0.9832487,-1.6379206,-0.87166226,0.13469009,2.644633,-0.3137418,-0.8415682,0.17771643,-0.33885878,-0.06802454,0.34736037,-1.01832,-0.62350124
+1,0,1,1.3108207,-1.3080025,0.34822875,1.4596325,-1.0497857,-0.16749367,0.485262,0.97075754,0.034637336,-1.7957212,1.4068959,0.78538895,-0.31832948,-0.30980903,-1.3081034,-0.5883961
+1,0,2,1.0650882,-0.09216967,0.36322507,0.18116008,-1.3025135,0.54174715,-0.482454,0.9420967,0.39991647,-2.0071194,-0.3020647,-1.1005768,0.43759206,2.1584413,0.28633898,-1.1556921
+1,0,3,1.8563068,-0.7071074,0.1579034,0.8939158,-0.0670585,-0.49880558,-1.4990184,-0.9207246,0.73329836,-0.34481865,0.627262,1.1733168,-1.3812757,-0.22482446,-1.1992402,1.4072857
+1,0,4,0.49123776,-0.65193045,2.2564845,-0.17658377,0.42154688,1.2235798,-0.052261785,-1.0662595,0.90595466,-0.9850903,-1.1610376,0.45742363,-1.7801737,-0.12341051,0.7481433,-0.5458406
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
index d9c4ea92..50c647c9 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
@@ -1,4 +1,5 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-2,0,0,-0.7215402,-0.9107281,-1.3260528,-1.056419,0.45294917,-1.4000003,1.2015996,0.9110033,0.082641706,1.9949383,0.7476521,-0.21124426,1.3691758,-0.049398102,-0.21682854,-0.8714193
-2,0,1,1.7914528,0.28032902,1.0305467,0.54616165,-1.1185598,-0.72510433,-0.10695522,-0.7386186,-0.97949564,1.8118858,-0.005990093,-0.68710846,-0.26387492,0.34162205,-1.9024037,0.6979735
-2,0,2,-0.30499086,-0.8624588,-0.21858704,0.46042088,1.684188,-0.8045112,0.051364895,0.14763248,-0.15275371,-0.62137896,0.52078897,-2.5685115,1.6713971,0.9936811,0.5037575,-0.5022072
+2,0,0,-0.36413205,-0.6412734,-1.509915,-0.64748913,1.2593658,-1.3857695,0.8524324,0.52342945,-0.109673396,2.4144866,0.8789581,-0.38112083,0.7561693,-0.6839269,-0.28703287,-0.67817223
+2,0,1,1.2825032,-0.4427156,1.9214746,0.46229368,-1.4471735,-0.5868213,-0.8091576,-0.24864331,-1.2884412,1.7199851,0.23889081,0.040785108,-0.05055482,0.22968066,-1.4788846,0.43781894
+2,0,2,0.094167404,0.0859257,-0.5042325,0.7252808,1.3083975,-1.3224137,-0.33199427,0.13993336,0.08184377,-0.40506566,0.29009566,-2.2843225,1.9552679,1.2503077,-0.18841878,-0.93038386
+2,0,3,1.0551077,-1.1094918,0.8473594,-0.12514752,0.4117495,0.18166554,-1.1333376,-0.8610408,0.65473765,-0.5237611,0.96090883,0.9585637,-1.4609623,-0.9971072,-0.89406204,2.0541434
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
index 41d3d3c6..429d9980 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
@@ -1,7 +1,9 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-3,0,0,-0.7998329,-1.009516,-1.2457482,-0.76004726,0.5837586,-1.667505,0.9730014,0.9463384,0.092550665,2.0262837,0.91990626,-0.16828287,1.2223057,-0.15285197,-0.15041813,-0.8101929
-3,0,1,-0.44488773,-0.768482,-0.96725094,-1.0061802,-1.1290205,0.5468874,1.4315809,-0.18544558,0.2409695,-0.26420698,0.2942724,2.9089172,0.037383337,0.20568404,0.09013668,-0.9878235
-3,0,2,0.9197474,-0.54972196,0.19757852,-1.6962574,-1.6389741,0.1349557,1.9117059,-0.8651552,0.5093372,-1.4415438,0.5049666,0.5232043,0.6729788,0.5089508,0.9317495,-0.64518726
-4,0,0,1.0214102,-0.8599426,1.1313342,-0.9132186,-1.1094761,0.36796033,-0.3840757,0.69376415,0.1755591,-0.18347333,-2.1679997,-0.50493,-0.43210468,2.0920951,0.5312533,0.5269235
-4,0,1,1.975615,0.48146883,1.191331,0.069339134,-0.15332681,-0.8564395,0.082964964,0.59752154,-1.4309397,0.55613095,-0.887867,-0.40480894,0.031874895,1.017457,-0.07405214,-2.231385
-4,0,2,0.24375618,-1.9693801,-0.9987763,-1.4291574,0.28681424,0.22939153,0.74847555,1.6184663,0.56282,0.12442069,0.036241677,-1.4214271,-0.19361049,0.4269336,-0.029789643,1.7619773
+3,0,0,-0.40126666,-0.78331727,-1.5756154,-0.31062096,1.1928762,-1.6843582,0.7851598,0.5358782,0.060748313,2.3220642,0.983614,-0.29303667,0.58234143,-0.36522076,-0.30240387,-0.7496143
+3,0,1,-0.7336195,-1.2810793,-0.38829067,-0.4601508,-1.50845,0.5241958,0.8025443,0.6143513,0.056892652,-0.31004176,0.44623825,2.7555754,-0.0637165,0.22621281,0.6022362,-1.314328
+3,0,2,1.3642316,0.12788779,-0.20111935,-1.036736,-1.948152,-0.38286552,1.4817194,-0.9606339,0.7927826,-1.2917897,-0.020328995,0.9448982,1.0943421,0.79177094,0.19575107,-1.0141716
+3,0,3,0.86852545,-0.017761663,0.41102606,-0.18726309,-0.87506944,0.5694903,0.15854332,0.8119648,0.5742181,-2.3636305,-0.37842566,-1.5641385,0.12360405,2.1495688,0.17902005,-0.5240532
+4,0,0,1.1631641,-0.55724025,1.1540875,-1.288715,-0.25030568,0.29519114,-0.26156953,0.73843837,0.3341611,0.36159894,-2.2564068,-1.12792,-0.6742551,1.7842551,-0.06704759,0.6194069
+4,0,1,0.8389171,-0.6922904,2.0705483,0.090488724,-0.24284306,-0.52842766,-0.25121197,0.6185058,-1.8862301,1.0848103,-0.5888317,0.16298746,0.6342511,0.5813959,0.052398324,-1.9904659
+4,0,2,1.2333139,-0.8917692,-1.4689931,-0.6857654,0.11552239,0.31441647,0.30494234,1.6532247,0.5431021,0.16783965,-0.89129585,-1.6974555,0.21965057,-0.051815066,-0.7658714,1.8788624
+4,0,3,-0.026684264,-0.90918845,0.74477476,-0.18694443,-0.103529036,-0.61683804,-0.64630914,-0.77517974,1.4970412,1.2110367,2.242393,0.84123915,-0.07830191,-1.5589653,-0.5940237,-1.0396498
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
index bc0d81ae..92904b29 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
@@ -1,4 +1,5 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-5,0,0,0.050885476,-0.68763536,-0.32008207,1.0336747,1.0007741,-0.9633068,-0.68598473,0.54518497,-0.34585357,-0.47822043,0.52911806,-2.0803304,1.9946527,1.3150282,0.1371507,-1.0467436
-5,0,1,1.5032598,0.98465127,-0.9370364,0.1338618,-0.023070391,-0.055669162,-0.92458737,0.0777592,-0.114300214,0.7075389,0.89388293,-1.1466067,0.17746806,1.8015935,-1.9533355,-1.140061
-5,0,2,0.92082304,-0.24809152,-0.097492896,-0.5940675,-1.1684945,0.82807684,-0.26342613,0.24813169,-0.61301243,-0.017792376,1.3554065,1.2064362,0.016166179,-0.95746326,1.6580712,-2.2975802
+5,0,0,-0.0975191,-0.4876295,-0.2364892,0.8899324,1.3293315,-1.4303449,-0.38584486,0.4049477,0.2341479,-0.24590422,0.4131558,-2.2018154,1.7526464,1.2278333,-0.1008542,-1.0976585
+5,0,1,0.96228456,0.5013961,-0.46397877,0.18880926,-0.21479401,0.068225846,-1.6835177,0.3138923,-0.215686,0.8992038,1.2001247,-0.7830776,0.29820076,1.9573467,-1.7968755,-1.2469637
+5,0,2,1.1223437,0.35189787,-0.4220163,-0.08106533,-1.5475112,0.14656012,-0.7290129,0.048741527,0.048301227,0.31566086,1.0508822,1.2633296,0.41796586,-1.0627791,1.3920348,-2.369732
+5,0,3,0.997344,-0.3670925,1.1623502,-1.5849966,-0.5812203,0.69678265,0.33674517,0.5580875,0.17368835,-0.32163247,-1.9439414,-1.2147049,-0.6150653,1.896507,0.09835436,0.66708636
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
index 32b2488a..4552eb37 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
@@ -1,8 +1,10 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-6,0,0,1.4568412,-0.14638022,1.5296284,-0.23393637,0.47851744,-0.61866605,-0.3450205,0.9253853,-1.4923376,1.1160238,-0.5999185,-0.7552432,-0.13930646,0.5775359,0.43673426,-2.2207265
-6,0,1,1.5268029,-0.8329019,1.6105913,0.68216175,-1.6239593,-0.7086337,-1.3626277,-0.081963986,-1.1570084,1.6663798,0.41809285,0.0707222,-0.10685151,0.3952702,-0.6428383,0.13019788
-6,0,2,1.0811658,0.20269495,-0.8124527,-0.00006336486,-1.4045546,0.38837618,-0.56309855,0.12770331,1.2022637,0.8725209,0.38372105,-2.5436192,0.303661,-0.14655858,1.5541319,-0.6847054
-6,0,3,0.03786793,-1.276104,1.9573607,0.4641471,-0.90920424,-0.64407074,0.8226515,1.1558217,-0.020683609,-0.8913047,-0.7226361,-0.8886756,-0.007999741,-0.83854043,-0.30502415,2.0437944
-7,0,0,-0.4584146,-1.3955376,-0.52467126,-0.618622,-1.2942654,0.7003769,0.48657182,0.09509239,-0.07182534,-0.2584199,0.48787862,2.825618,0.3362462,0.20020695,0.8225655,-1.3267877
-7,0,1,1.4503253,-0.6273112,-0.99216473,1.4233189,1.0060959,0.67193794,-1.6410915,-0.6428859,-0.934798,1.4429418,0.1590132,-1.4877084,0.1463788,0.6280047,-0.5242633,-0.079930626
-7,0,2,1.1836636,-0.09071134,-0.14702518,1.2998279,-1.8555058,0.77598166,-0.72920847,0.43935195,0.4997686,1.223445,0.761577,-2.2151723,-0.3662264,0.030556075,0.093911,-0.94706964
+6,0,0,0.77065754,-0.76632786,1.8472377,-0.36171773,-0.12794422,-0.51205856,0.23821487,0.19354652,-2.007292,1.1207803,-0.47570035,0.12879921,0.6755318,0.19856073,1.0114113,-1.9820988
+6,0,1,1.5402608,-0.6269262,1.811596,0.5628006,-1.0015095,-1.0100083,-0.8554023,-0.5945589,-0.7188355,2.0016418,0.036225002,-0.0963106,-0.34438652,-0.04934737,-1.2410477,0.5716474
+6,0,2,0.099329084,-0.9905449,0.37024456,-0.092512354,-1.0474453,0.5579103,-1.3993762,0.9833203,0.59650284,1.0684012,0.9790155,-2.0937629,0.30437842,-0.46602613,1.7425917,-0.64842325
+6,0,3,0.39516813,-0.6314403,1.4948617,1.0518291,-1.0769643,-1.1019645,0.72841096,0.8454386,-0.1329276,-0.27656975,-1.1176367,-0.6240037,0.08811874,-0.8709323,-1.0424569,2.2322922
+6,0,4,1.5397077,-1.1038709,0.17057961,-0.70247877,0.91511214,-0.19965753,0.10995354,-0.09201375,-0.5179603,-1.2657489,-0.9364169,0.99344826,-0.5904373,2.378038,-1.0655441,0.35856313
+7,0,0,-0.38164693,-1.3991219,-0.57864404,-0.64194506,-1.0146283,0.77735037,1.0805509,0.20559101,0.14912546,0.24122998,0.23987257,2.7745647,-0.1806378,-0.38417065,0.53943276,-1.4561769
+7,0,1,0.7505255,-1.2829787,-0.04241756,1.1337081,0.5512444,1.0470235,-2.240779,0.38265908,-0.9141051,0.9473515,1.0557507,-1.5054818,-0.0026612426,0.71467644,-0.11884089,-0.48391303
+7,0,2,1.4440962,0.640023,-1.0184404,1.4597166,-1.5533735,0.19580375,-0.73750097,0.17983381,0.6347194,1.3996398,0.19554296,-1.8371892,0.27592623,0.28698903,-0.50887656,-1.1177573
+7,0,3,1.6090622,-0.552633,2.0960214,0.36178717,-1.3726065,-0.61812073,-0.5444809,-0.8581807,-1.0325772,1.5600145,0.2399627,-0.15540515,-0.32274023,0.06386035,-1.1094363,0.61451924
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
index c949fdbf..9c9c4f81 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
@@ -1,5 +1,6 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-8,0,0,-0.15959328,-0.6185797,-0.4608494,0.8537297,1.7962078,-1.1127523,-0.532915,0.41857973,-0.025692305,-0.11058396,0.43107894,-2.3876948,1.7336528,0.75202924,0.051930904,-0.62754744
-8,0,1,-0.0847378,-0.80747044,0.54135734,0.21987116,-0.27892733,-0.6947288,-1.0170838,-0.3269096,1.2616061,1.8441427,2.0002346,0.7380944,-0.06132242,-0.8849537,-1.2508907,-1.1942315
-8,0,2,-0.060736626,-0.2758699,-1.7476931,-0.90112156,0.66965175,-1.3700024,1.6178493,0.32909736,0.27642158,1.9732589,0.37533468,-0.48581138,1.1548799,0.011515356,-0.70278364,-0.87338156
-8,0,3,0.7022285,-0.33732405,-0.057168916,-0.8344231,-0.868286,0.98070174,-0.23612636,0.26457375,-0.65804225,0.08511807,1.3956434,1.1746727,-0.08018996,-1.348229,1.8052337,-2.011606
+8,0,0,-0.32614353,-0.64439976,-0.4606882,0.65944755,1.3359828,-1.0852644,-0.16756222,0.14784002,-0.08113572,-0.6550381,0.7323079,-2.066391,1.9876616,1.0655345,0.63813996,-1.1135746
+8,0,1,0.12641025,-0.56098837,0.40993986,0.107800655,0.32741314,-0.5968289,-0.84562564,-0.6825451,1.2423956,1.8815519,1.8854672,0.5606451,-0.055782493,-1.6092077,-1.2014889,-0.98526716
+8,0,2,-0.41545722,-0.6968768,-1.6478468,-0.6066328,0.9194668,-1.1579491,0.6114348,0.7974771,-0.52014804,2.3043096,1.1645346,-0.0551654,0.95790863,-0.3786852,-0.23132549,-1.0501995
+8,0,3,1.0761181,0.35165632,-0.40121642,-0.17262518,-1.2321544,0.42185432,-0.7928826,-0.034794606,-0.17338482,0.422188,1.1260121,1.2503253,0.30318612,-1.3496894,1.4538991,-2.2986958
+8,0,4,1.1404579,-1.0539529,0.6959456,0.0002799956,0.50083333,0.25582743,-1.1861986,-0.89913344,0.5945992,-0.5446149,0.99352646,1.0013199,-1.5414313,-0.9312056,-0.95023686,1.9430875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
index 13a6a356..8e68ce40 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
@@ -1,4 +1,5 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-9,0,0,-0.5791086,0.37848184,-0.3573193,-0.2355006,-0.14353162,-0.3610847,-0.8467316,-0.7029914,0.79546255,-0.9406641,-0.41330278,1.1095508,1.7990614,-0.31147677,2.3270056,-1.514815
-9,0,1,0.80497956,0.5582172,-0.92235243,0.12620161,-1.2371104,0.36691037,-0.58676964,0.394082,1.040764,0.61476994,0.54693735,-2.5416214,-0.021673568,-0.45668596,1.8399619,-0.5678765
-9,0,2,-0.72852725,-1.68056,-0.4684366,-1.2692064,0.31585234,0.8645183,1.0582336,-0.34532753,1.2095294,0.25657162,-0.7247522,1.3270385,-1.5651994,0.36106008,1.513459,-0.11602255
+9,0,0,-0.47500592,0.4115684,-0.3617759,-0.28399315,0.3019857,-0.5060671,-0.6323188,-0.9995569,0.9944466,-0.5976193,-0.70640665,0.9123807,1.9063975,-0.45425025,2.1027493,-1.6554946
+9,0,1,-0.086714655,-0.97823036,0.05636277,-0.01605165,-0.9614703,0.6308942,-1.2854741,1.0908369,0.45563138,0.926207,1.0300418,-2.0144224,0.2822436,-0.24596272,1.959498,-0.8820346
+9,0,2,-0.4039917,-0.86205673,-1.1582229,-1.341583,-0.120875224,0.73452944,1.7072109,-0.28952715,1.6449419,0.5135223,-0.9904969,1.0184028,-1.2944711,1.0233638,0.47282502,-0.68130696
+9,0,3,0.4459618,-1.5170839,-0.057393685,0.48917928,1.573972,1.3741968,-1.1912827,-0.21140826,-0.8198331,0.22414537,1.1317799,-2.0740838,-0.63661385,0.7234822,0.49297914,0.047913294
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
index a6555801..8cccbef6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
@@ -1,5 +1,6 @@
 sequenceId,itemPosition,itemId
-0,0,116
+0,0,[unknown]
 0,1,125
 0,2,111
 0,3,102
+0,4,118
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
index 63cca8be..4bd96488 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
@@ -2,4 +2,5 @@ sequenceId,itemPosition,itemId
 1,0,100
 1,1,120
 1,2,102
-1,3,104
+1,3,119
+1,4,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
index 78e69770..7cd4f7cd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
@@ -1,4 +1,5 @@
 sequenceId,itemPosition,itemId
-2,0,103
+2,0,125
 2,1,115
 2,2,125
+2,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
index aa1d014b..ae179237 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
@@ -1,7 +1,9 @@
 sequenceId,itemPosition,itemId
-3,0,103
-3,1,119
+3,0,123
+3,1,111
 3,2,119
+3,3,102
 4,0,122
 4,1,125
-4,2,118
+4,2,112
+4,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
index d0661d18..c8b395dc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
@@ -1,4 +1,5 @@
 sequenceId,itemPosition,itemId
 5,0,125
-5,1,123
-5,2,117
+5,1,100
+5,2,111
+5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
index 9b747406..d04d1d0e 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
@@ -1,8 +1,10 @@
 sequenceId,itemPosition,itemId
 6,0,125
 6,1,115
-6,2,109
-6,3,106
-7,0,117
+6,2,113
+6,3,115
+6,4,114
+7,0,111
 7,1,125
-7,2,109
+7,2,129
+7,3,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
index 69194925..7143ade0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
@@ -1,5 +1,6 @@
 sequenceId,itemPosition,itemId
 8,0,125
-8,1,126
-8,2,123
+8,1,113
+8,2,125
 8,3,117
+8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
index 5be21105..d379af64 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
@@ -1,4 +1,5 @@
 sequenceId,itemPosition,itemId
 9,0,116
-9,1,128
-9,2,114
+9,1,113
+9,2,111
+9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index 05a3e9a6..4305a322 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.24294445
-0,22,0.2182963
-0,23,0.24425358
-0,24,0.276201
-0,25,0.0940009
-0,26,0.51380986
-0,27,0.13742942
-0,28,0.19333738
-0,29,0.14042449
-0,30,-0.110163026
-0,31,-0.12064577
-0,32,-0.17106278
-0,33,-0.007956264
-0,34,0.05157075
-0,35,-0.14410715
-0,36,-0.038530935
-0,37,0.07652966
-0,38,-0.1730595
-0,39,-0.068232045
-0,40,-0.04052765
-1,19,0.2532388
-1,20,0.2412585
-1,21,0.1653834
-1,22,0.23926179
-1,23,0.33210897
-1,24,0.11496639
-1,25,-0.025427505
-1,26,-0.054255053
-1,27,-0.20999868
-1,28,-0.045269843
-1,29,-0.11365727
-1,30,-0.116153166
-1,31,0.18535054
-1,32,0.15639819
-1,33,0.025363889
-1,34,0.17836204
-1,35,-0.06773287
-1,36,-0.096186034
-1,37,-0.16507263
-1,38,0.009327785
-2,17,0.40199393
-2,18,-0.024678737
-2,19,0.033101153
-2,20,0.15040804
-2,21,0.23327164
-2,22,0.08850994
-2,23,0.16138998
-2,24,0.013258814
-2,25,0.0075182635
-2,26,-0.09369014
-2,27,0.04258554
-2,28,-0.0762189
-2,29,-0.10716796
-2,30,-0.14510551
-2,31,0.12744585
-2,32,-0.072724655
-2,33,0.016253883
-2,34,-0.086202465
-2,35,0.018500187
-2,36,0.13643105
-3,17,0.0030256584
-3,18,0.49384275
-3,19,0.3081484
-3,20,0.16238832
-3,21,0.117462285
-3,22,0.11546557
-3,23,0.045830198
-3,24,0.01562991
-3,25,0.09699597
-3,26,0.033849917
-3,27,-0.032291207
-3,28,0.12744585
-3,29,-0.122143306
-3,30,-0.0832074
-3,31,-0.11515481
-3,32,-0.021808462
-3,33,-0.12314166
-3,34,0.06355102
-3,35,-0.020810105
-3,36,-0.17605457
-4,14,0.0077678524
-4,15,0.050821982
-4,16,0.0075182635
-4,17,-0.116153166
-4,18,0.24325521
-4,19,0.13942613
-4,20,0.12594831
-4,21,0.0665461
-4,22,-0.04252436
-4,23,0.06754445
-4,24,-0.021059694
-4,25,0.13243763
-4,26,0.07652966
-4,27,-0.24993296
-4,28,-0.1361203
-4,29,0.06030637
-4,30,0.070040345
-4,31,-0.002878685
-4,32,-0.06573616
-4,33,-0.16107921
-5,14,0.26821414
-5,15,0.25623387
-5,16,0.36405638
-5,17,0.22129136
-5,18,0.3061517
-5,19,0.077528015
-5,20,0.07503212
-5,21,0.09499926
-5,22,-0.08121069
-5,23,-0.18404141
-5,24,-0.08270822
-5,25,-0.10117782
-5,26,-0.10117782
-5,27,-0.018064626
-5,28,-0.1910299
-5,29,0.33210897
-5,30,-0.16207758
-5,31,0.025987862
-5,32,-0.035535865
-5,33,-0.049762446
-6,18,0.20631602
-6,19,0.27420428
-6,20,0.082020625
-6,21,0.28219113
-6,22,0.3940071
-6,23,0.16638176
-6,24,-0.08370657
-6,25,-0.12613674
-6,26,0.14641462
-6,27,-0.012511266
-6,28,-0.08520411
-6,29,-0.15708579
-6,30,-0.044521075
-6,31,0.069541164
-6,32,-0.0036664505
-6,33,-0.09269179
-6,34,-0.13711865
-6,35,-0.027549013
-6,36,-0.15309235
-6,37,0.07403377
-7,15,0.19832917
-7,16,0.28019443
-7,17,0.052569106
-7,18,-0.13412358
-7,19,-0.039279703
-7,20,0.06704527
-7,21,0.10049022
-7,22,0.055564176
-7,23,0.024365531
-7,24,0.018375391
-7,25,-0.11365727
-7,26,0.003056857
-7,27,0.027360601
-7,28,0.08401734
-7,29,0.07253624
-7,30,-0.12763427
-7,31,-0.1910299
-7,32,-0.012760855
-7,33,0.058559246
-7,34,0.026611833
-8,20,0.34808266
-8,21,-0.02392997
-8,22,-0.16806771
-8,23,0.09499926
-8,24,0.3361024
-8,25,0.017127445
-8,26,-0.15608743
-8,27,-0.10067864
-8,28,-0.12414002
-8,29,-0.013821609
-8,30,0.17137355
-8,31,-0.114156455
-8,32,0.03859211
-8,33,-0.009578593
-8,34,0.06454938
-8,35,-0.11116138
-8,36,0.05132116
-8,37,-0.12713508
-8,38,0.06305185
-8,39,-0.033539154
-9,16,0.19134067
-9,17,0.32811555
-9,18,0.09000748
-9,19,0.09000748
-9,20,-0.13212687
-9,21,0.055564176
-9,22,0.0665461
-9,23,-0.23395924
-9,24,0.15240477
-9,25,-0.021683667
-9,26,-0.090195894
-9,27,-0.06299067
-9,28,0.009639772
-9,29,-0.061992317
-9,30,-0.27688858
-9,31,0.118959814
-9,32,0.18934396
-9,33,-0.066734515
-9,34,0.06704527
-9,35,0.13742942
+0,21,-0.16518183
+0,22,0.28807208
+0,23,0.26411152
+0,24,0.070929505
+0,25,0.106371164
+0,26,0.08790157
+0,27,0.28008524
+0,28,0.18923476
+0,29,-0.0023990057
+0,30,-0.007035904
+0,31,-0.20611446
+0,32,-0.07283385
+0,33,0.10886706
+0,34,0.016643867
+0,35,-0.0878092
+0,36,0.06843361
+0,37,0.032991957
+0,38,-0.0878092
+0,39,-0.12524757
+0,40,-0.18714568
+1,19,0.18324462
+1,20,-0.041136023
+1,21,0.41785845
+1,22,0.41386503
+1,23,0.3938979
+1,24,0.07841718
+1,25,0.0021052985
+1,26,0.06344183
+1,27,-0.04138561
+1,28,-0.045878217
+1,29,-0.03289958
+1,30,-0.06684371
+1,31,0.10537281
+1,32,-0.058856852
+1,33,-0.14820977
+1,34,-0.13023935
+1,35,0.035737436
+1,36,-0.11326729
+1,37,-0.039139308
+1,38,0.11785226
+2,17,0.34198335
+2,18,0.20121504
+2,19,0.07142868
+2,20,0.03773415
+2,21,0.3060425
+2,22,-0.11426565
+2,23,0.07192786
+2,24,0.013773591
+2,25,-0.036643416
+2,26,-0.027034234
+2,27,-0.021293685
+2,28,-0.034397114
+2,29,-0.09030509
+2,30,-0.15120484
+2,31,-0.18914239
+2,32,-0.18914239
+2,33,0.07991471
+2,34,-0.120754965
+2,35,0.011652084
+2,36,-0.14421634
+3,17,0.11385884
+3,18,0.3240129
+3,19,0.29406223
+3,20,0.094390884
+3,21,0.2251756
+3,22,0.36993733
+3,23,0.057202104
+3,24,0.22417724
+3,25,-0.07283385
+3,26,0.16427584
+3,27,0.05545498
+3,28,-0.22108981
+3,29,-0.06834124
+3,30,-0.083316594
+3,31,0.05021361
+3,32,-0.052117944
+3,33,-0.1222525
+3,34,0.1148572
+3,35,-0.011247721
+3,36,-0.0209193
+4,14,0.25612468
+4,15,0.09139582
+4,16,0.2092019
+4,17,0.17126435
+4,18,0.12234487
+4,19,0.22817066
+4,20,0.04047963
+4,21,-0.14022292
+4,22,0.041727576
+4,23,-0.07033796
+4,24,-0.058856852
+4,25,-0.061352745
+4,26,-0.0067239176
+4,27,-0.012994845
+4,28,-0.12874182
+4,29,0.16128078
+4,30,-0.11925743
+4,31,-0.071336314
+4,32,-0.11726072
+4,33,0.06843361
+5,14,-0.2260816
+5,15,0.026253048
+5,16,0.13232844
+5,17,0.17825285
+5,18,0.35196692
+5,19,0.31203264
+5,20,0.06943197
+5,21,-0.10428208
+5,22,-0.10178619
+5,23,0.019638937
+5,24,0.017018251
+5,25,-0.10577962
+5,26,-0.073333025
+5,27,-0.17716211
+5,28,-0.16917527
+5,29,-0.17217033
+5,30,0.23316245
+5,31,0.04047963
+5,32,-0.06784207
+5,33,0.13432515
+6,18,0.2531296
+6,19,0.19222984
+6,20,0.455796
+6,21,0.3958946
+6,22,-0.015490737
+6,23,-0.17316869
+6,24,-0.109273866
+6,25,-0.16717854
+6,26,0.1333268
+6,27,-0.35486957
+6,28,-0.07283385
+6,29,-0.15320155
+6,30,0.12833501
+6,31,-0.04512945
+6,32,-0.12774347
+6,33,0.142312
+6,34,0.018765375
+6,35,0.14630543
+6,36,-0.10428208
+6,37,-0.062351096
+7,15,-0.022167247
+7,16,0.2601181
+7,17,0.08041389
+7,18,-0.019172177
+7,19,-0.1841506
+7,20,0.068932794
+7,21,0.053707857
+7,22,0.028499352
+7,23,-0.20811117
+7,24,-0.039139308
+7,25,-0.033897936
+7,26,-0.07582892
+7,27,0.06593772
+7,28,-0.067342885
+7,29,-0.21609803
+7,30,-0.19712925
+7,31,0.19222984
+7,32,-0.11825907
+7,33,-0.022167247
+7,34,0.065438546
+8,20,0.18324462
+8,21,-0.14122127
+8,22,-0.018922588
+8,23,0.32001948
+8,24,0.060197175
+8,25,0.070929505
+8,26,-0.03639383
+8,27,0.16727091
+8,28,-0.018049026
+8,29,-0.06634453
+8,30,-0.056111373
+8,31,-0.07383221
+8,32,-0.14920813
+8,33,-0.05361548
+8,34,0.10736952
+8,35,0.06793443
+8,36,0.043973878
+8,37,0.3040458
+8,38,-0.077326454
+8,39,-0.1522032
+9,16,0.3060425
+9,17,0.28807208
+9,18,0.3180228
+9,19,0.0068474924
+9,20,0.25612468
+9,21,0.14331035
+9,22,0.22118217
+9,23,-0.16118841
+9,24,-0.06384864
+9,25,-0.04537904
+9,26,0.18923476
+9,27,0.0664369
+9,28,-0.20611446
+9,29,0.07991471
+9,30,-0.19613089
+9,31,-0.109273866
+9,32,0.07642046
+9,33,-0.11975661
+9,34,-0.018548204
+9,35,-0.2350668
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
index 8cc53618..1cf52011 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,0.30016157
-0,9,-0.1820447
-0,10,0.39201036
-0,11,0.27420428
-0,12,-0.30883598
-1,8,0.290178
-1,9,-0.1890332
-1,10,0.3500794
-2,8,-0.15508908
-2,9,0.12445078
-3,8,0.3800301
-3,9,0.3061517
-4,7,-0.14011373
-5,7,0.14641462
-6,8,0.34808266
-6,9,-0.13911536
-6,10,0.3221254
-7,8,-0.16307592
-8,8,0.1743686
-8,9,-0.13711865
-8,10,0.43593806
-8,11,0.4259545
-9,8,0.33011225
-9,9,-0.12813345
+0,8,0.2900688
+0,9,-0.17616376
+0,10,0.3799209
+0,11,0.26411152
+0,12,-0.3009583
+1,8,0.28208193
+1,9,-0.0039628376
+1,10,0.33798993
+2,8,-0.14820977
+2,9,0.11735309
+3,8,0.36993733
+3,9,0.29805565
+4,7,-0.13523114
+5,7,0.14131364
+6,8,0.3359932
+6,9,0.26810494
+6,10,0.31402937
+7,8,-0.15419991
+8,8,0.16826928
+8,9,-0.25703064
+8,10,0.41586173
+8,11,0.40787488
+9,8,0.31402937
+9,9,0.25812137
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
index 405ee2f4..68074514 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.19243994
-0,9,-0.010234638
-0,10,0.020318083
-0,11,0.18382144
-0,12,0.47416487
-1,8,0.09296566
-1,9,-0.059118994
-1,10,-0.2941589
-2,8,0.047784667
-2,9,-0.19342752
-3,8,0.09247188
-3,9,0.17888363
-4,7,0.37935886
-5,7,-0.14503694
-6,8,0.30035383
-6,9,0.33393097
-6,10,0.00545835
-7,8,0.6716774
-8,8,0.21542345
-8,9,0.19073437
-8,10,-0.041836645
-8,11,-0.02529497
-9,8,0.0025419537
-9,9,-0.2546564
+0,8,-0.19088916
+0,9,-0.078307025
+0,10,0.020881303
+0,11,0.13500652
+0,12,0.30980512
+1,8,0.0876035
+1,9,-0.026953768
+1,10,-0.2353295
+2,8,0.046619654
+2,9,-0.18891405
+3,8,0.08957863
+3,9,0.17450903
+4,7,0.37695938
+5,7,-0.18002598
+6,8,0.2999295
+6,9,0.33745688
+6,10,0.007867463
+7,8,0.66730285
+8,8,0.20117323
+8,9,0.19031003
+8,10,-0.03213847
+8,11,-0.012880999
+9,8,0.0023336397
+9,9,-0.2531056
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
index 5e697aea..7c11814b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.083931625
-0,9,-0.35219386
-0,10,-0.061499946
-0,11,-0.23102838
-0,12,-0.24387926
-1,8,-0.27600646
-1,9,-0.108486
-1,10,-0.01715292
-2,8,-0.36320892
-2,9,-0.30262616
-3,8,-0.13740048
-3,9,0.2316035
-4,7,0.106766336
-5,7,-0.35219386
-6,8,0.06591888
-6,9,-0.11628832
-6,10,0.15358028
-7,8,0.11869929
-8,8,-0.17595315
-8,9,-0.14336696
-8,10,-0.3081337
-8,11,-0.3007903
-9,8,-0.1419901
-9,9,-0.01715292
+0,8,-0.07564883
+0,9,-0.34528795
+0,10,-0.06342901
+0,11,-0.22871205
+0,12,-0.23238373
+1,8,-0.2268762
+1,9,-0.10984136
+1,10,-0.016901925
+2,8,-0.3618105
+2,9,-0.29939193
+3,8,-0.1360021
+3,9,0.23850942
+4,7,0.104493044
+5,7,-0.3471238
+6,8,0.09990344
+6,9,-0.11305408
+6,10,0.15406075
+7,8,0.12652314
+8,8,-0.17822644
+8,9,-0.11075929
+8,10,-0.29755607
+8,11,-0.29204854
+9,8,-0.14472234
+9,9,-0.019196726
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
index d6da3019..22ba483f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.08198105
-0,9,-0.215309
-0,10,-0.075096644
-0,11,-0.0692449
-0,12,-0.101486854
-1,8,-0.049674552
-1,9,-0.034134448
-1,10,-0.122140065
-2,8,-0.2749738
-2,9,-0.2575333
-3,8,-0.0012040511
-3,9,0.36940628
-4,7,0.35104787
-5,7,-0.06190154
-6,8,0.34003285
-6,9,0.2757784
-6,10,0.33085364
-7,8,0.35471958
-8,8,0.011417352
-8,9,0.100455634
-8,10,-0.057369307
-8,11,-0.12718862
-9,8,-0.0424531
-9,9,0.18031469
+0,8,-0.08269817
+0,9,-0.22038624
+0,10,-0.060151752
+0,11,-0.066634566
+0,12,-0.0973849
+1,8,-0.055131875
+1,9,-0.02945879
+1,10,-0.12446355
+2,8,-0.25710306
+2,9,-0.24058047
+3,8,0.01506035
+3,9,0.37075448
+4,7,0.3579036
+5,7,-0.12813523
+6,8,0.3395452
+6,9,0.2771266
+6,10,0.31567925
+7,8,0.35423192
+8,8,-0.039154325
+8,9,-0.015919466
+8,10,-0.053353403
+8,11,-0.12629938
+9,8,-0.0007737763
+9,9,0.19084209
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
index 9a24e80c..2132af3a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 0,4,127,1,1,2,1
-0,5,127,1,8,8,1
-0,6,113,1,1,7,0
-0,7,105,8,9,7,1
-0,8,127,1,1,0,1
-0,9,113,1,7,7,1
-0,10,113,8,9,1,1
-0,11,113,1,1,7,1
-0,12,105,8,5,7,1
-0,13,113,1,5,8,1
-0,14,113,1,[other],8,1
-0,15,113,1,5,7,1
-0,16,105,1,5,7,1
-0,17,108,1,8,7,1
-0,18,108,1,5,7,1
-0,19,124,1,5,7,1
-0,20,124,1,5,7,1
-0,21,105,8,5,7,1
-0,22,105,1,5,7,1
-0,23,110,1,5,7,1
-0,24,110,1,5,8,1
-0,25,110,1,7,8,1
-0,26,110,[other],7,7,1
-0,27,110,5,7,7,1
-0,28,103,5,7,0,1
-0,29,102,1,5,0,1
-0,30,103,1,7,7,1
-0,31,102,5,5,0,1
-0,32,103,1,7,0,1
-0,33,102,[mask],7,0,1
+0,5,113,4,9,8,1
+0,6,113,1,1,8,1
+0,7,113,1,9,8,1
+0,8,113,1,9,8,1
+0,9,113,1,9,8,1
+0,10,113,1,8,8,1
+0,11,113,1,0,8,1
+0,12,113,1,9,7,1
+0,13,121,1,8,7,1
+0,14,105,1,8,7,1
+0,15,105,1,8,7,1
+0,16,105,1,8,7,1
+0,17,113,1,7,7,1
+0,18,113,6,7,7,1
+0,19,113,6,7,2,1
+0,20,113,1,7,8,1
+0,21,113,1,7,8,1
+0,22,113,1,7,8,1
+0,23,113,1,0,8,1
+0,24,113,1,0,8,1
+0,25,113,1,0,8,1
+0,26,113,1,0,8,1
+0,27,113,1,0,8,1
+0,28,113,1,0,7,1
+0,29,108,1,8,7,1
+0,30,108,1,0,7,1
+0,31,108,6,0,7,1
+0,32,108,1,0,7,1
+0,33,108,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
index 12e2155a..c473bc6b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,121,1,0,7,0
-1,5,108,3,0,8,1
-1,6,103,1,0,7,0
-1,7,103,8,0,7,1
-1,8,108,1,0,2,1
-1,9,113,1,0,7,1
-1,10,108,8,7,7,1
-1,11,124,8,7,7,1
-1,12,105,8,7,7,1
-1,13,113,8,5,7,1
-1,14,113,1,[other],8,1
-1,15,113,1,[other],8,1
-1,16,113,1,[other],7,1
-1,17,108,1,5,7,1
-1,18,124,1,[other],7,1
-1,19,108,1,5,7,1
-1,20,124,1,5,7,1
-1,21,124,1,5,7,1
-1,22,105,8,5,7,1
-1,23,105,1,5,7,1
-1,24,124,8,5,7,1
-1,25,110,8,5,7,1
-1,26,110,1,5,8,1
-1,27,110,1,7,8,1
-1,28,110,6,7,7,1
-1,29,110,1,7,0,1
-1,30,113,1,7,0,1
-1,31,113,1,7,0,1
-1,32,113,1,7,0,1
-1,33,113,[mask],5,0,1
+1,4,121,6,0,8,0
+1,5,121,1,0,8,0
+1,6,121,1,0,8,0
+1,7,121,1,0,8,0
+1,8,121,1,0,8,0
+1,9,126,1,0,8,0
+1,10,127,1,0,8,0
+1,11,127,0,0,8,0
+1,12,127,0,9,8,0
+1,13,127,0,7,7,0
+1,14,105,8,9,7,1
+1,15,127,0,7,2,1
+1,16,113,4,7,8,1
+1,17,113,1,9,8,1
+1,18,113,1,7,8,1
+1,19,113,1,9,8,1
+1,20,113,1,0,8,1
+1,21,113,1,0,8,1
+1,22,113,1,0,8,1
+1,23,113,1,0,8,1
+1,24,113,1,0,7,1
+1,25,108,1,8,7,1
+1,26,108,1,8,7,1
+1,27,105,6,8,7,1
+1,28,113,1,8,7,1
+1,29,113,6,7,7,1
+1,30,113,6,7,7,1
+1,31,113,6,7,2,1
+1,32,113,1,7,8,1
+1,33,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
index 5cf7bea0..4f612da8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,128,1,8,6,0
-2,4,105,0,8,7,2
-2,5,105,1,8,2,1
-2,6,113,1,8,7,2
-2,7,105,1,9,7,1
-2,8,105,1,7,7,1
-2,9,113,1,7,7,1
-2,10,113,1,5,7,1
-2,11,113,1,[other],8,1
-2,12,113,1,[other],7,1
-2,13,113,8,5,7,1
-2,14,113,1,[other],8,1
-2,15,113,1,5,7,1
-2,16,113,1,5,7,1
-2,17,105,1,5,7,1
-2,18,108,1,5,7,1
-2,19,108,1,5,7,1
-2,20,108,1,5,7,1
-2,21,124,1,5,7,1
-2,22,105,8,5,7,1
-2,23,110,1,5,7,1
-2,24,110,8,5,8,1
-2,25,110,1,5,8,1
-2,26,110,[other],5,8,1
-2,27,127,1,7,8,1
-2,28,103,5,7,0,1
-2,29,103,1,5,7,1
-2,30,103,1,5,0,1
-2,31,103,[other],4,0,1
-2,32,127,1,4,7,1
+2,3,121,1,8,0,0
+2,4,105,0,8,7,0
+2,5,121,1,8,2,0
+2,6,113,4,7,8,0
+2,7,121,1,9,8,0
+2,8,127,1,0,8,0
+2,9,127,4,0,8,1
+2,10,127,1,0,8,0
+2,11,127,4,7,8,1
+2,12,127,1,0,8,1
+2,13,103,1,0,8,1
+2,14,113,1,0,8,1
+2,15,103,1,9,7,1
+2,16,105,1,7,7,1
+2,17,113,1,9,7,1
+2,18,105,1,9,7,1
+2,19,113,1,7,7,1
+2,20,113,6,9,7,1
+2,21,113,1,7,7,1
+2,22,113,6,5,7,1
+2,23,113,1,8,8,1
+2,24,113,1,0,8,1
+2,25,113,1,0,8,1
+2,26,113,1,0,8,1
+2,27,113,1,0,8,1
+2,28,113,1,0,8,1
+2,29,113,1,0,7,1
+2,30,108,1,8,7,1
+2,31,105,1,8,7,1
+2,32,105,1,9,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
index 4e1d63dd..52804bbb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,127,5,0,2,0
-3,4,127,4,0,8,0
-3,5,127,4,0,8,0
-3,6,127,4,0,8,0
-3,7,127,4,0,8,0
-3,8,127,4,0,8,1
-3,9,108,1,0,8,1
-3,10,103,1,7,7,1
-3,11,103,8,9,7,1
-3,12,103,1,7,7,1
-3,13,113,8,7,7,1
-3,14,113,8,5,7,1
-3,15,113,1,5,7,1
-3,16,113,1,5,7,1
-3,17,113,1,5,7,1
-3,18,113,1,[other],8,1
-3,19,113,1,[other],7,1
-3,20,108,8,5,7,1
-3,21,108,1,2,7,1
-3,22,108,1,5,7,1
-3,23,108,1,5,7,1
-3,24,124,1,5,7,1
-3,25,124,1,5,7,1
-3,26,124,8,5,7,1
-3,27,124,1,5,7,1
-3,28,124,8,5,7,1
-3,29,110,8,5,7,1
-3,30,110,1,5,8,1
-3,31,110,1,7,8,1
-3,32,110,[other],7,7,1
-4,3,127,0,9,[unknown],1
-4,4,127,0,1,8,1
-4,5,102,0,1,8,1
-4,6,127,1,4,8,0
-4,7,127,0,1,7,1
-4,8,127,0,4,0,1
-4,9,127,1,1,7,1
-4,10,124,1,9,0,1
-4,11,113,1,7,7,1
-4,12,113,8,5,0,1
-4,13,113,1,5,8,1
-4,14,113,1,7,8,1
-4,15,113,1,5,7,1
-4,16,113,1,5,7,1
-4,17,113,1,5,7,1
-4,18,113,1,5,7,1
-4,19,108,1,5,7,1
-4,20,108,1,5,7,1
-4,21,103,1,5,7,1
-4,22,103,1,5,7,1
-4,23,110,1,5,7,1
-4,24,110,1,5,8,1
-4,25,110,1,5,8,1
-4,26,110,1,5,0,1
-4,27,110,0,5,7,1
-4,28,110,[other],5,0,1
-4,29,127,0,4,7,1
-4,30,103,0,5,0,1
-4,31,103,[other],4,0,1
-4,32,127,1,4,7,1
+3,3,127,4,0,2,0
+3,4,127,4,0,8,1
+3,5,127,4,0,8,1
+3,6,127,1,0,2,1
+3,7,113,4,9,8,1
+3,8,121,1,1,2,1
+3,9,113,0,9,8,1
+3,10,113,1,1,8,1
+3,11,113,1,9,7,1
+3,12,105,1,9,7,1
+3,13,105,1,8,7,1
+3,14,105,1,9,7,1
+3,15,113,1,8,7,1
+3,16,105,8,9,7,1
+3,17,113,1,8,8,1
+3,18,113,1,9,8,1
+3,19,113,1,7,8,1
+3,20,113,1,9,7,1
+3,21,113,1,8,7,1
+3,22,113,1,8,7,1
+3,23,113,1,8,7,1
+3,24,113,1,8,7,1
+3,25,113,1,5,7,1
+3,26,113,1,5,8,1
+3,27,113,1,0,8,1
+3,28,113,1,0,8,1
+3,29,113,1,0,8,1
+3,30,113,1,0,8,1
+3,31,113,1,0,7,1
+3,32,108,1,8,7,1
+4,3,127,4,0,8,0
+4,4,127,4,0,8,1
+4,5,127,1,0,2,1
+4,6,127,4,7,8,1
+4,7,127,1,1,8,1
+4,8,127,1,0,8,1
+4,9,113,1,0,8,1
+4,10,113,1,9,8,1
+4,11,113,1,9,7,1
+4,12,105,1,9,7,1
+4,13,105,1,9,7,1
+4,14,105,1,9,7,1
+4,15,105,1,9,7,1
+4,16,113,1,9,7,1
+4,17,113,1,9,7,1
+4,18,113,1,9,7,1
+4,19,113,1,9,8,1
+4,20,113,1,9,8,1
+4,21,113,1,9,8,1
+4,22,113,1,9,8,1
+4,23,113,1,9,7,1
+4,24,113,1,8,7,1
+4,25,105,1,8,7,1
+4,26,113,1,8,7,1
+4,27,113,1,8,7,1
+4,28,113,1,8,7,1
+4,29,113,1,9,7,1
+4,30,113,1,7,8,1
+4,31,113,1,7,8,1
+4,32,113,1,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
index 4f5b8222..75c58ebf 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,128,1,5,4,1
-5,4,128,1,5,6,1
-5,5,105,1,5,6,1
-5,6,128,1,5,0,1
-5,7,105,1,5,7,1
-5,8,105,1,5,0,1
-5,9,113,1,7,7,1
-5,10,105,1,9,7,1
-5,11,103,1,9,7,1
-5,12,108,1,9,7,1
-5,13,113,1,5,7,1
-5,14,113,1,5,8,1
-5,15,113,1,5,8,1
-5,16,113,1,5,8,1
-5,17,113,1,5,7,1
-5,18,105,1,5,7,1
-5,19,105,1,5,7,1
-5,20,103,1,5,7,1
-5,21,103,1,5,7,1
-5,22,103,1,5,7,1
-5,23,103,1,5,7,1
-5,24,108,1,5,7,1
-5,25,124,8,5,7,1
-5,26,110,8,5,7,1
-5,27,110,1,5,8,1
-5,28,110,2,7,8,1
-5,29,110,2,5,0,1
-5,30,110,1,5,7,1
-5,31,124,[other],5,0,1
-5,32,127,1,8,0,1
+5,3,121,1,5,4,1
+5,4,127,1,0,8,1
+5,5,127,1,0,2,1
+5,6,127,1,7,2,1
+5,7,127,1,1,0,1
+5,8,105,1,9,2,1
+5,9,113,1,7,0,1
+5,10,113,1,9,7,1
+5,11,105,8,9,7,1
+5,12,105,1,8,2,1
+5,13,113,0,7,8,1
+5,14,113,1,9,8,1
+5,15,113,1,9,8,1
+5,16,113,1,9,8,1
+5,17,113,1,0,8,1
+5,18,113,1,0,8,1
+5,19,113,1,0,8,1
+5,20,113,1,0,7,1
+5,21,108,1,8,7,1
+5,22,105,1,8,7,1
+5,23,105,1,8,7,1
+5,24,105,1,9,7,1
+5,25,113,1,7,7,1
+5,26,113,6,7,7,1
+5,27,113,6,7,2,1
+5,28,113,1,7,8,1
+5,29,113,1,7,8,1
+5,30,113,1,9,8,1
+5,31,113,1,0,8,1
+5,32,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
index 929f2c8c..d4edc9b2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,109,4,8,8,1
-6,5,113,1,0,7,0
-6,6,113,1,7,7,1
-6,7,113,1,7,7,1
-6,8,113,1,7,7,1
-6,9,102,8,7,7,1
-6,10,102,5,7,7,1
-6,11,102,1,7,7,1
-6,12,113,1,7,7,1
-6,13,113,8,5,0,1
-6,14,102,1,5,2,1
-6,15,113,1,7,8,1
-6,16,113,1,1,0,1
-6,17,113,1,5,7,1
-6,18,105,1,5,7,1
-6,19,105,1,5,7,1
-6,20,108,1,7,7,1
-6,21,108,8,7,7,1
-6,22,105,5,5,7,1
-6,23,113,1,5,7,1
-6,24,113,1,5,8,1
-6,25,113,1,5,8,1
-6,26,113,1,5,8,1
-6,27,113,1,5,7,1
-6,28,108,1,5,7,1
-6,29,108,1,5,7,1
-6,30,103,1,5,7,1
-6,31,103,1,5,7,1
-6,32,103,1,5,7,1
-6,33,103,8,5,7,1
-7,3,105,0,9,8,0
-7,4,127,0,1,7,0
-7,5,105,0,9,8,1
-7,6,127,0,1,7,0
-7,7,105,0,9,8,1
-7,8,113,0,1,0,0
-7,9,113,1,9,8,1
-7,10,113,1,1,8,1
-7,11,113,1,9,8,1
-7,12,113,1,1,7,1
-7,13,113,1,9,7,1
-7,14,113,1,9,7,1
-7,15,105,1,5,7,1
-7,16,113,1,9,8,1
-7,17,113,1,5,7,1
-7,18,105,1,5,7,1
-7,19,105,1,5,7,1
-7,20,113,1,5,7,1
-7,21,113,1,5,7,1
-7,22,113,1,5,7,1
-7,23,113,1,5,7,1
-7,24,113,1,5,7,1
-7,25,113,1,5,7,1
-7,26,113,1,5,7,1
-7,27,113,1,5,7,1
-7,28,113,1,5,7,1
-7,29,113,1,5,7,1
-7,30,113,1,5,7,1
-7,31,113,1,5,7,1
-7,32,113,1,5,7,1
+6,4,121,4,0,8,0
+6,5,113,1,0,8,0
+6,6,113,1,0,8,1
+6,7,113,1,0,8,1
+6,8,113,1,0,8,1
+6,9,113,1,0,8,1
+6,10,113,1,0,8,1
+6,11,113,1,9,7,1
+6,12,105,8,9,7,1
+6,13,105,1,8,7,1
+6,14,105,1,9,7,1
+6,15,105,1,9,7,1
+6,16,113,1,7,7,1
+6,17,113,6,9,7,1
+6,18,113,1,3,2,1
+6,19,113,6,7,8,1
+6,20,113,1,9,8,1
+6,21,113,1,0,8,1
+6,22,113,1,0,8,1
+6,23,113,1,0,8,1
+6,24,113,1,0,8,1
+6,25,113,1,0,8,1
+6,26,113,1,0,7,1
+6,27,108,1,8,7,1
+6,28,105,1,8,7,1
+6,29,105,1,9,7,1
+6,30,105,1,7,7,1
+6,31,113,6,7,7,1
+6,32,113,6,7,7,1
+6,33,113,6,7,2,1
+7,3,105,4,9,8,0
+7,4,121,0,8,8,0
+7,5,113,0,8,8,0
+7,6,121,4,8,8,0
+7,7,127,0,0,8,0
+7,8,121,4,8,8,0
+7,9,127,4,0,8,0
+7,10,127,1,0,8,0
+7,11,127,4,0,8,1
+7,12,127,1,0,8,0
+7,13,127,1,0,8,1
+7,14,127,1,7,8,1
+7,15,113,1,7,8,1
+7,16,113,1,9,7,1
+7,17,103,1,7,7,1
+7,18,113,8,7,7,1
+7,19,113,1,9,7,1
+7,20,113,1,7,7,1
+7,21,113,6,5,7,1
+7,22,113,1,5,8,1
+7,23,113,1,0,8,1
+7,24,113,1,0,8,1
+7,25,113,1,0,8,1
+7,26,113,1,0,8,1
+7,27,113,1,0,7,1
+7,28,108,1,8,7,1
+7,29,108,1,8,7,1
+7,30,108,1,8,7,1
+7,31,122,6,7,7,1
+7,32,113,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
index 83cccde4..42ff3af8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,109,1,0,7,0
-8,5,127,1,8,7,1
-8,6,124,1,7,0,1
-8,7,127,1,7,7,0
-8,8,127,8,9,0,1
-8,9,127,1,1,0,1
-8,10,124,1,7,0,1
-8,11,113,1,7,0,1
-8,12,113,1,7,7,1
-8,13,113,8,5,7,1
-8,14,113,1,5,8,1
-8,15,113,1,5,8,1
-8,16,113,1,5,7,1
-8,17,113,1,5,7,1
-8,18,108,1,5,7,1
-8,19,108,1,5,7,1
-8,20,108,1,5,7,1
-8,21,124,1,5,7,1
-8,22,103,8,5,7,1
-8,23,103,1,5,7,1
-8,24,110,8,5,7,1
-8,25,110,1,5,8,1
-8,26,110,2,5,8,1
-8,27,110,1,5,8,1
-8,28,110,[other],7,0,1
-8,29,127,1,7,7,1
-8,30,108,5,7,0,1
-8,31,127,5,5,0,1
-8,32,102,1,4,0,1
-8,33,103,[other],4,0,1
+8,4,113,1,0,8,0
+8,5,113,1,0,8,1
+8,6,113,1,0,2,1
+8,7,113,1,8,2,1
+8,8,113,1,0,8,1
+8,9,113,1,0,8,1
+8,10,113,1,0,2,1
+8,11,113,1,0,2,1
+8,12,113,1,8,7,1
+8,13,105,8,9,7,1
+8,14,105,1,8,2,1
+8,15,113,1,9,7,1
+8,16,105,1,9,7,1
+8,17,113,1,7,7,1
+8,18,113,1,9,7,1
+8,19,113,1,7,8,1
+8,20,113,1,9,8,1
+8,21,113,1,0,8,1
+8,22,113,1,9,8,1
+8,23,113,1,0,7,1
+8,24,113,1,8,7,1
+8,25,105,1,8,7,1
+8,26,113,1,8,7,1
+8,27,105,1,9,7,1
+8,28,113,1,8,7,1
+8,29,113,6,9,7,1
+8,30,113,1,8,2,1
+8,31,113,1,7,8,1
+8,32,113,1,7,8,1
+8,33,113,1,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
index eebbcfd3..0c5df9b9 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,108,1,8,7,1
-9,4,108,1,[other],7,1
-9,5,108,8,5,7,1
-9,6,108,1,[other],7,1
-9,7,108,8,[other],7,1
-9,8,108,8,[other],7,1
-9,9,124,8,[other],7,1
-9,10,113,8,[other],2,1
-9,11,113,1,[other],8,1
-9,12,113,1,[other],8,1
-9,13,113,1,[other],8,1
-9,14,113,1,[other],0,1
-9,15,113,1,9,2,1
-9,16,113,1,5,7,1
-9,17,105,1,5,7,1
-9,18,105,1,8,7,1
-9,19,103,1,5,7,1
-9,20,105,1,5,7,1
-9,21,113,1,5,7,1
-9,22,113,1,7,7,1
-9,23,113,8,5,7,1
-9,24,113,1,5,8,1
-9,25,113,1,5,8,1
-9,26,113,1,5,7,1
-9,27,108,1,5,7,1
-9,28,108,1,5,7,1
-9,29,108,1,5,7,1
-9,30,108,1,5,7,1
-9,31,124,1,5,7,1
-9,32,103,8,5,7,1
+9,3,121,1,8,8,1
+9,4,113,1,0,8,1
+9,5,113,1,8,8,1
+9,6,113,1,0,7,1
+9,7,105,1,8,7,1
+9,8,105,1,8,7,1
+9,9,105,1,7,7,1
+9,10,113,1,7,7,1
+9,11,113,1,7,7,1
+9,12,113,6,7,7,1
+9,13,113,6,5,2,1
+9,14,113,1,0,8,1
+9,15,113,1,0,8,1
+9,16,113,1,0,8,1
+9,17,113,1,0,8,1
+9,18,113,1,0,8,1
+9,19,113,1,0,8,1
+9,20,113,1,0,8,1
+9,21,113,1,0,7,1
+9,22,105,1,8,7,1
+9,23,105,1,8,7,1
+9,24,105,1,9,7,1
+9,25,105,1,7,7,1
+9,26,113,6,7,7,1
+9,27,113,6,7,2,1
+9,28,113,1,7,8,1
+9,29,113,1,7,8,1
+9,30,113,1,9,8,1
+9,31,113,1,0,8,1
+9,32,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
index fd49670a..29653eb6 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
@@ -6,7 +6,7 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 0,8,111,9,9,7,[other]
 0,9,107,1,9,7,[other]
 0,10,104,1,9,7,[other]
-0,11,102,1,9,7,[other]
+0,11,116,1,9,7,[other]
 0,12,104,6,9,7,[other]
 0,13,104,1,9,7,[other]
 0,14,104,1,9,[unknown],[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
index c47a6367..ec1ff900 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,111,9,2,7,[other]
-1,5,111,1,9,7,[other]
+1,4,111,9,2,9,[other]
+1,5,111,1,2,7,[other]
 1,6,111,1,9,7,[other]
 1,7,111,1,9,7,[other]
 1,8,111,1,9,7,[other]
@@ -9,11 +9,11 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 1,11,129,1,9,[unknown],[other]
 1,12,102,1,9,[unknown],[other]
 1,13,104,6,9,[unknown],[other]
-1,14,104,6,9,[unknown],[other]
+1,14,104,1,9,[unknown],[other]
 1,15,104,6,9,[unknown],[other]
 1,16,104,6,9,[unknown],[other]
-1,17,104,6,9,7,[other]
-1,18,104,9,9,7,[other]
+1,17,104,6,9,[unknown],[other]
+1,18,104,6,9,7,[other]
 1,19,104,6,9,7,[other]
 1,20,104,6,9,7,[other]
 1,21,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
index d60f3445..cffacbd7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
@@ -1,20 +1,20 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 2,3,115,9,2,2,[unknown]
-2,4,118,9,2,[mask],[mask]
-2,5,115,9,9,7,[mask]
+2,4,118,[mask],2,[mask],[mask]
+2,5,[unknown],9,2,7,[mask]
 2,6,115,9,9,7,[mask]
 2,7,115,9,9,7,[mask]
-2,8,119,9,9,7,[mask]
-2,9,104,9,9,7,[other]
-2,10,104,1,9,7,[other]
-2,11,104,6,9,7,[other]
-2,12,104,6,9,7,[other]
-2,13,104,6,9,7,[other]
-2,14,104,6,9,7,[other]
-2,15,104,6,9,7,[other]
-2,16,104,6,9,7,[other]
-2,17,104,6,9,7,[other]
-2,18,104,6,9,7,[other]
+2,8,115,9,9,7,[mask]
+2,9,119,9,9,7,[other]
+2,10,104,[mask],9,7,[other]
+2,11,112,1,9,7,[other]
+2,12,104,[mask],9,7,[other]
+2,13,112,1,9,7,[other]
+2,14,104,9,9,7,[other]
+2,15,112,6,9,7,[other]
+2,16,104,9,9,7,[other]
+2,17,104,1,9,7,[other]
+2,18,104,1,9,7,[other]
 2,19,104,6,9,7,[other]
 2,20,104,6,9,7,[other]
 2,21,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
index 1923a1cb..9bf3ad90 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
@@ -5,7 +5,7 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 3,6,112,9,2,2,[mask]
 3,7,112,9,2,2,[mask]
 3,8,112,9,9,2,[other]
-3,9,112,9,2,2,[other]
+3,9,112,9,2,2,[mask]
 3,10,112,9,9,7,[other]
 3,11,112,9,9,7,[other]
 3,12,112,9,9,7,[other]
@@ -14,27 +14,27 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 3,15,112,9,9,7,[other]
 3,16,112,9,9,7,[other]
 3,17,112,9,9,7,[other]
-3,18,112,9,9,7,[other]
+3,18,112,9,9,8,[other]
 3,19,112,9,9,7,[other]
-3,20,112,9,9,7,[other]
-3,21,112,9,9,7,[other]
-3,22,112,9,9,7,[other]
-3,23,112,9,9,7,[other]
-3,24,112,9,9,7,[other]
-3,25,112,9,9,7,[other]
-3,26,112,9,9,7,[other]
-3,27,112,9,9,7,[other]
-3,28,112,9,9,7,[other]
-3,29,112,9,9,7,[other]
-3,30,112,9,9,7,[other]
-3,31,112,9,9,7,[other]
+3,20,112,[mask],9,7,[other]
+3,21,112,[mask],9,8,[other]
+3,22,112,[mask],9,7,[other]
+3,23,112,[mask],9,7,[other]
+3,24,112,[mask],9,7,[other]
+3,25,129,[mask],9,7,[other]
+3,26,129,1,9,7,[other]
+3,27,129,1,9,7,[other]
+3,28,129,1,9,7,[other]
+3,29,129,1,9,[unknown],[other]
+3,30,112,6,9,7,[other]
+3,31,104,9,9,7,[other]
 3,32,112,9,9,7,[other]
 4,3,107,9,2,7,[other]
 4,4,107,9,9,7,[other]
 4,5,107,9,9,7,[other]
 4,6,107,9,9,7,[other]
 4,7,107,9,9,7,[other]
-4,8,104,6,9,7,[other]
+4,8,107,9,9,7,[other]
 4,9,104,6,9,7,[other]
 4,10,104,6,9,7,[other]
 4,11,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
index fe484f33..cf194871 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
@@ -2,7 +2,7 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 5,3,111,3,2,2,[other]
 5,4,111,3,2,8,[other]
 5,5,111,1,2,7,[other]
-5,6,111,1,9,7,[other]
+5,6,111,1,2,7,[other]
 5,7,111,1,9,7,[other]
 5,8,111,1,9,7,[other]
 5,9,102,1,9,7,[other]
@@ -11,7 +11,7 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 5,12,102,6,9,[unknown],[other]
 5,13,104,6,9,[unknown],[other]
 5,14,104,6,9,[unknown],[other]
-5,15,104,6,9,7,[other]
+5,15,104,6,9,[unknown],[other]
 5,16,104,6,9,7,[other]
 5,17,104,9,9,7,[other]
 5,18,104,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
index 22da0edc..4cae85f8 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
@@ -4,15 +4,15 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,6,111,9,9,7,[other]
 6,7,111,1,9,7,[other]
 6,8,111,1,9,7,[other]
-6,9,102,1,9,7,[other]
-6,10,129,1,9,[unknown],[other]
+6,9,111,1,9,7,[other]
+6,10,102,1,9,[unknown],[other]
 6,11,129,1,9,[unknown],[other]
-6,12,129,1,9,[unknown],[other]
-6,13,104,1,9,[unknown],[other]
-6,14,104,6,9,[unknown],[other]
+6,12,102,1,9,[unknown],[other]
+6,13,104,6,9,[unknown],[other]
+6,14,104,1,9,[unknown],[other]
 6,15,104,6,9,[unknown],[other]
 6,16,104,6,9,[unknown],[other]
-6,17,104,6,9,7,[other]
+6,17,104,6,9,[unknown],[other]
 6,18,104,6,9,7,[other]
 6,19,104,6,9,7,[other]
 6,20,104,6,9,7,[other]
@@ -29,33 +29,33 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,31,104,6,9,7,[other]
 6,32,104,6,9,7,[other]
 6,33,104,6,9,7,[other]
-7,3,117,2,9,7,[other]
-7,4,129,6,9,7,[other]
-7,5,104,9,9,7,[other]
-7,6,129,9,9,7,[other]
-7,7,129,9,9,7,[other]
-7,8,129,9,9,7,[other]
-7,9,129,6,9,2,[other]
+7,3,111,2,9,7,[other]
+7,4,111,5,9,7,[other]
+7,5,117,1,9,7,[other]
+7,6,129,6,9,[unknown],[other]
+7,7,129,6,9,7,[other]
+7,8,104,9,9,7,[other]
+7,9,129,6,9,[unknown],[other]
 7,10,104,9,9,7,[other]
-7,11,112,9,9,7,[other]
-7,12,112,9,9,7,[other]
-7,13,112,[mask],9,7,[other]
-7,14,112,[mask],9,8,[other]
-7,15,112,[mask],9,7,[other]
-7,16,112,[mask],9,7,[other]
-7,17,112,9,9,7,[other]
-7,18,112,[mask],9,7,[other]
-7,19,112,[mask],9,7,[other]
-7,20,112,[mask],9,7,[other]
-7,21,129,1,9,7,[other]
-7,22,129,1,9,7,[other]
-7,23,129,9,9,7,[other]
-7,24,112,9,9,7,[other]
-7,25,112,9,9,7,[other]
-7,26,112,9,9,7,[other]
-7,27,112,9,9,7,[other]
-7,28,112,9,9,7,[other]
-7,29,112,9,9,8,[other]
-7,30,112,9,9,7,[other]
-7,31,112,[mask],9,7,[unknown]
-7,32,112,[mask],9,8,[other]
+7,11,104,[mask],9,7,[other]
+7,12,104,[mask],9,7,[other]
+7,13,104,[mask],9,7,[other]
+7,14,107,[mask],9,7,[other]
+7,15,107,1,9,7,[other]
+7,16,104,6,9,7,[other]
+7,17,104,6,9,7,[other]
+7,18,104,1,9,7,[other]
+7,19,104,1,9,7,[other]
+7,20,104,6,9,7,[other]
+7,21,104,6,9,7,[other]
+7,22,104,6,9,7,[other]
+7,23,104,6,9,7,[other]
+7,24,104,6,9,7,[other]
+7,25,104,6,9,7,[other]
+7,26,104,6,9,7,[other]
+7,27,104,6,9,7,[other]
+7,28,104,6,9,7,[other]
+7,29,104,6,9,7,[other]
+7,30,104,6,9,7,[other]
+7,31,104,6,9,7,[other]
+7,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
index caac944b..b1dc8274 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
@@ -12,7 +12,7 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 8,14,104,1,9,[unknown],[other]
 8,15,104,6,9,[unknown],[other]
 8,16,104,6,9,[unknown],[other]
-8,17,104,6,9,7,[other]
+8,17,104,6,9,[unknown],[other]
 8,18,104,6,9,7,[other]
 8,19,104,6,9,7,[other]
 8,20,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
index 3f1d5fac..12d83f71 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
@@ -2,24 +2,24 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 9,3,113,3,2,[other],[other]
 9,4,119,9,2,7,[mask]
 9,5,129,9,9,[other],[other]
-9,6,119,9,9,7,[mask]
-9,7,104,9,9,7,[other]
-9,8,119,9,9,7,[other]
-9,9,129,9,9,7,[other]
+9,6,119,9,9,[other],[mask]
+9,7,119,9,9,9,[mask]
+9,8,129,9,9,8,[other]
+9,9,111,9,9,7,[other]
 9,10,112,9,9,7,[other]
-9,11,112,9,9,7,[unknown]
+9,11,112,[mask],9,7,[unknown]
 9,12,112,[mask],9,7,[other]
-9,13,112,[mask],9,7,[other]
-9,14,112,[mask],9,7,[other]
-9,15,112,[mask],9,7,[other]
-9,16,112,[mask],9,7,[other]
-9,17,112,[mask],9,8,[other]
-9,18,112,[mask],9,7,[other]
-9,19,129,1,9,7,[other]
-9,20,129,1,9,7,[other]
-9,21,129,1,9,7,[other]
-9,22,104,1,9,7,[other]
-9,23,104,6,9,7,[other]
+9,13,112,[mask],9,8,[unknown]
+9,14,104,[mask],9,7,[other]
+9,15,129,1,9,7,[other]
+9,16,129,1,9,7,[other]
+9,17,129,1,9,7,[other]
+9,18,104,1,9,7,[other]
+9,19,104,1,9,7,[other]
+9,20,104,1,9,[unknown],[other]
+9,21,104,6,9,[unknown],[other]
+9,22,104,6,9,[unknown],[other]
+9,23,104,6,9,[unknown],[other]
 9,24,104,6,9,7,[other]
 9,25,104,6,9,7,[other]
 9,26,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
index 8c8e70ba..54189e82 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,102,3,9,5,1
-0,5,102,4,3,0,1
-0,6,102,6,9,2,1
-0,7,102,4,3,0,1
-0,8,112,6,9,2,1
-0,9,102,4,3,0,1
-0,10,112,6,9,0,0
-0,11,102,4,3,0,1
-0,12,112,6,9,0,0
-0,13,102,4,3,0,2
-0,14,112,6,3,0,0
-0,15,102,4,9,0,0
-0,16,112,4,3,0,0
-0,17,112,4,3,0,0
-0,18,112,4,3,0,0
-0,19,112,4,3,0,0
-0,20,112,4,3,0,0
-0,21,112,4,3,0,0
-0,22,112,4,3,0,0
-0,23,112,4,3,0,0
-0,24,112,4,3,0,0
-0,25,112,4,3,0,0
-0,26,112,4,3,0,0
-0,27,112,4,3,0,0
-0,28,112,4,3,0,0
-0,29,112,4,3,0,0
-0,30,112,4,3,0,0
-0,31,112,4,3,0,0
-0,32,112,4,3,0,0
-0,33,112,4,3,0,0
+0,4,121,1,1,9,1
+0,5,105,0,[other],8,2
+0,6,113,1,[other],[mask],0
+0,7,105,8,[other],8,1
+0,8,113,1,1,8,0
+0,9,105,3,[other],1,1
+0,10,113,1,8,8,1
+0,11,113,1,[other],1,1
+0,12,113,1,9,1,1
+0,13,113,1,[other],1,1
+0,14,113,0,9,1,1
+0,15,113,1,8,1,1
+0,16,103,0,9,1,1
+0,17,113,0,8,1,1
+0,18,113,0,[other],8,1
+0,19,113,0,[other],0,[mask]
+0,20,108,0,[other],1,[other]
+0,21,108,0,[other],7,1
+0,22,108,5,[other],[mask],1
+0,23,124,[mask],[other],7,1
+0,24,124,1,[other],2,1
+0,25,113,1,[other],8,1
+0,26,113,1,[other],[mask],1
+0,27,122,8,[other],8,1
+0,28,113,1,[other],8,1
+0,29,113,1,[other],0,1
+0,30,113,8,[other],2,1
+0,31,113,1,[other],2,1
+0,32,113,8,9,1,1
+0,33,113,1,[other],[mask],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
index 60c86728..6342f656 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,102,4,0,5,0
-1,5,102,6,0,2,1
-1,6,111,4,2,2,1
-1,7,112,4,9,5,1
-1,8,112,4,9,8,1
-1,9,112,4,9,0,0
-1,10,112,7,9,0,1
-1,11,112,4,3,0,0
-1,12,112,4,9,0,0
-1,13,112,4,9,0,1
-1,14,112,4,9,0,0
-1,15,112,4,3,0,0
-1,16,112,4,9,0,0
-1,17,112,4,3,0,0
-1,18,112,4,9,0,0
-1,19,112,4,3,0,0
-1,20,112,4,3,0,0
-1,21,112,4,3,0,0
-1,22,112,4,3,0,0
-1,23,112,4,3,0,0
-1,24,112,4,3,0,0
-1,25,112,4,3,0,0
-1,26,112,4,3,0,0
-1,27,112,4,3,0,0
-1,28,112,4,3,0,0
-1,29,112,4,3,0,0
-1,30,112,4,3,0,0
-1,31,112,4,3,0,0
-1,32,112,4,3,0,0
-1,33,112,4,3,0,0
+1,4,107,5,[other],[mask],[other]
+1,5,121,3,[other],7,[other]
+1,6,121,5,[other],8,[other]
+1,7,121,5,[other],7,[other]
+1,8,121,5,[other],7,[other]
+1,9,121,1,[other],8,1
+1,10,122,1,7,8,0
+1,11,105,1,7,8,[other]
+1,12,122,1,7,[unknown],0
+1,13,105,0,7,1,1
+1,14,113,[other],7,5,0
+1,15,113,[mask],1,5,1
+1,16,113,1,1,2,1
+1,17,120,[mask],4,1,1
+1,18,113,[mask],1,7,1
+1,19,102,1,[other],2,1
+1,20,113,1,7,8,[unknown]
+1,21,113,1,1,0,1
+1,22,128,1,0,1,1
+1,23,113,1,[other],7,1
+1,24,105,8,[other],2,1
+1,25,113,1,1,2,1
+1,26,113,8,[other],1,1
+1,27,105,1,1,2,1
+1,28,113,0,[other],1,1
+1,29,113,1,[other],2,1
+1,30,113,[mask],9,1,1
+1,31,113,1,[other],1,1
+1,32,113,1,9,1,1
+1,33,113,1,9,1,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
index 9c10138e..dba66fc2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,112,4,0,5,1
-2,4,112,4,3,0,0
-2,5,112,4,9,0,1
-2,6,112,4,9,0,0
-2,7,112,4,3,0,0
-2,8,112,4,9,0,0
-2,9,112,4,3,0,0
-2,10,112,4,9,0,0
-2,11,112,4,3,0,0
-2,12,112,4,3,0,0
-2,13,112,4,3,0,0
-2,14,112,4,3,0,0
-2,15,112,4,3,0,0
-2,16,112,4,3,0,0
-2,17,112,4,3,0,0
-2,18,112,4,3,0,0
-2,19,112,4,3,0,0
-2,20,112,4,3,0,0
-2,21,112,4,3,0,0
-2,22,112,4,3,0,0
-2,23,112,4,3,0,0
-2,24,112,4,3,0,0
-2,25,112,4,3,0,0
-2,26,112,4,3,0,0
-2,27,112,4,3,0,0
-2,28,112,4,3,0,0
-2,29,112,4,3,0,0
-2,30,112,4,3,0,0
-2,31,112,4,3,0,0
-2,32,112,4,3,0,0
+2,3,128,0,7,[unknown],2
+2,4,113,8,[other],6,2
+2,5,109,1,[other],6,1
+2,6,128,1,[other],[mask],2
+2,7,104,[mask],[other],6,1
+2,8,104,1,1,2,1
+2,9,113,1,8,0,2
+2,10,113,1,[other],[mask],1
+2,11,121,1,[other],0,1
+2,12,101,1,[other],7,1
+2,13,105,8,[other],0,1
+2,14,122,1,1,2,1
+2,15,113,[mask],[other],0,1
+2,16,113,1,[other],2,1
+2,17,113,[mask],[other],2,1
+2,18,113,1,[other],2,1
+2,19,113,[mask],[other],1,1
+2,20,113,1,[other],2,1
+2,21,113,1,[other],1,1
+2,22,113,1,9,1,1
+2,23,113,1,[other],1,1
+2,24,103,1,9,1,1
+2,25,103,1,8,1,1
+2,26,103,0,9,7,1
+2,27,105,5,8,0,1
+2,28,105,0,8,1,1
+2,29,113,0,8,9,2
+2,30,105,3,[other],1,[other]
+2,31,113,0,[other],8,[other]
+2,32,113,0,[other],8,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
index 8295c2ce..20206ce0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,112,3,3,5,0
-3,4,102,4,3,5,0
-3,5,112,3,3,0,0
-3,6,112,4,3,5,0
-3,7,112,4,3,0,0
-3,8,112,4,3,0,0
-3,9,112,4,9,0,0
-3,10,112,4,3,0,0
-3,11,112,4,9,0,0
-3,12,112,4,3,0,0
-3,13,112,4,9,0,0
-3,14,112,4,3,0,0
-3,15,112,4,3,0,0
-3,16,112,4,3,0,0
-3,17,112,4,3,0,0
-3,18,112,4,3,0,0
-3,19,112,4,3,0,0
-3,20,112,4,3,0,0
-3,21,112,4,3,0,0
-3,22,112,4,3,0,0
-3,23,112,4,3,0,0
-3,24,112,4,3,0,0
-3,25,112,4,3,0,0
-3,26,112,4,3,0,0
-3,27,112,4,3,0,0
-3,28,112,4,3,0,0
-3,29,112,4,3,0,0
-3,30,112,4,3,0,0
-3,31,112,4,3,0,0
-3,32,112,4,3,0,0
-4,3,113,4,9,5,1
-4,4,112,4,9,8,1
-4,5,118,4,9,8,1
-4,6,102,4,9,0,1
-4,7,112,4,9,0,1
-4,8,112,4,9,0,1
-4,9,112,4,9,0,1
-4,10,112,4,9,0,1
-4,11,112,4,3,0,0
-4,12,112,4,9,0,1
-4,13,112,4,3,0,0
-4,14,112,4,3,0,1
-4,15,112,4,9,0,0
-4,16,112,4,3,0,0
-4,17,112,4,9,0,0
-4,18,112,4,3,0,0
-4,19,112,4,9,0,0
-4,20,112,4,3,0,0
-4,21,112,4,3,0,0
-4,22,112,4,9,0,0
-4,23,112,4,3,0,0
-4,24,112,4,3,0,0
-4,25,112,4,3,0,0
-4,26,112,4,3,0,0
-4,27,112,4,3,0,0
-4,28,112,4,3,0,0
-4,29,112,4,3,0,0
-4,30,112,4,3,0,0
-4,31,112,4,3,0,0
-4,32,112,4,3,0,0
+3,3,102,5,1,1,2
+3,4,128,0,9,5,[unknown]
+3,5,105,0,1,5,2
+3,6,105,[other],9,[mask],[unknown]
+3,7,128,0,1,[mask],1
+3,8,105,0,[other],8,[unknown]
+3,9,121,1,8,[mask],[other]
+3,10,104,0,[other],8,1
+3,11,113,1,[other],[mask],0
+3,12,104,3,[other],1,1
+3,13,113,1,[other],8,2
+3,14,113,1,[other],1,1
+3,15,113,1,[other],1,1
+3,16,113,1,[other],1,1
+3,17,105,0,[other],1,1
+3,18,120,0,8,0,1
+3,19,105,0,8,2,1
+3,20,113,0,8,0,1
+3,21,113,1,[other],2,1
+3,22,113,0,[other],1,1
+3,23,113,0,[other],8,1
+3,24,113,1,[other],8,1
+3,25,113,1,[other],8,1
+3,26,113,1,[other],8,1
+3,27,113,1,[other],7,1
+3,28,105,8,[other],1,1
+3,29,113,1,1,0,1
+3,30,105,1,[other],1,1
+3,31,113,1,9,0,1
+3,32,105,1,[other],1,1
+4,3,102,0,[other],[unknown],[unknown]
+4,4,102,8,[other],8,[unknown]
+4,5,102,0,[other],6,[unknown]
+4,6,105,8,[other],6,[unknown]
+4,7,115,8,[other],2,1
+4,8,127,[mask],7,8,[unknown]
+4,9,113,2,1,0,1
+4,10,122,1,[other],2,1
+4,11,113,[mask],7,2,1
+4,12,113,1,1,2,1
+4,13,113,[mask],[other],1,1
+4,14,113,1,1,2,1
+4,15,113,1,[other],1,1
+4,16,113,1,1,1,1
+4,17,105,1,[other],1,1
+4,18,105,1,9,0,1
+4,19,105,1,8,0,1
+4,20,105,1,9,0,1
+4,21,105,1,8,0,1
+4,22,105,0,9,1,[other]
+4,23,113,0,8,0,[other]
+4,24,113,0,[other],1,[other]
+4,25,113,0,[other],8,2
+4,26,113,1,[other],8,1
+4,27,113,1,[other],8,1
+4,28,113,1,[other],8,1
+4,29,113,1,[other],8,1
+4,30,113,1,[other],[mask],1
+4,31,121,1,[other],1,1
+4,32,126,1,0,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
index cdb9d0fa..051cc930 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,108,4,9,8,1
-5,4,104,4,9,8,1
-5,5,119,4,9,8,1
-5,6,104,4,9,8,1
-5,7,119,4,9,8,1
-5,8,104,4,9,8,0
-5,9,102,4,9,5,1
-5,10,119,4,3,0,0
-5,11,112,4,3,6,0
-5,12,112,4,3,5,0
-5,13,112,4,3,5,0
-5,14,112,4,3,0,0
-5,15,112,4,3,0,0
-5,16,112,4,3,6,0
-5,17,112,4,3,0,0
-5,18,112,4,9,0,0
-5,19,112,4,3,6,0
-5,20,112,4,3,0,0
-5,21,112,4,9,0,0
-5,22,112,4,3,0,0
-5,23,112,4,3,0,0
-5,24,112,4,3,6,0
-5,25,112,4,3,0,0
-5,26,112,4,3,0,0
-5,27,112,4,3,6,0
-5,28,112,4,3,0,0
-5,29,112,4,3,0,0
-5,30,112,4,3,0,0
-5,31,112,4,3,6,0
-5,32,112,4,3,0,0
+5,3,128,3,[other],4,1
+5,4,128,1,[other],6,1
+5,5,105,1,[other],7,1
+5,6,105,1,[other],0,1
+5,7,105,1,[other],0,1
+5,8,105,1,[other],0,1
+5,9,113,1,7,2,1
+5,10,113,1,[other],1,1
+5,11,113,1,[other],1,1
+5,12,113,1,[other],1,1
+5,13,113,1,[other],1,1
+5,14,113,0,[other],1,1
+5,15,113,0,[other],2,1
+5,16,113,0,[other],2,1
+5,17,113,0,1,1,1
+5,18,108,0,9,1,1
+5,19,113,0,8,0,1
+5,20,113,0,[other],[mask],1
+5,21,108,3,[other],2,1
+5,22,113,1,[other],8,1
+5,23,113,1,[other],8,1
+5,24,113,1,[other],8,1
+5,25,113,1,[other],0,1
+5,26,113,1,[other],7,1
+5,27,105,8,[other],0,1
+5,28,113,1,[other],2,1
+5,29,113,8,[other],1,1
+5,30,113,1,[other],2,1
+5,31,113,[mask],[other],1,1
+5,32,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
index 181f8b8a..d92da9a7 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,102,4,9,5,1
-6,5,112,4,3,0,1
-6,6,112,4,9,5,1
-6,7,112,4,3,0,1
-6,8,112,4,9,0,1
-6,9,112,4,3,0,0
-6,10,112,4,9,0,1
-6,11,112,4,3,0,0
-6,12,112,4,9,0,0
-6,13,112,4,3,0,0
-6,14,112,4,9,0,0
-6,15,112,4,3,0,0
-6,16,112,4,9,0,0
-6,17,112,4,3,0,0
-6,18,112,4,9,0,0
-6,19,112,4,3,0,0
-6,20,112,4,3,0,0
-6,21,112,4,3,0,0
-6,22,112,4,3,0,0
-6,23,112,4,3,0,0
-6,24,112,4,3,0,0
-6,25,112,4,3,0,0
-6,26,112,4,3,0,0
-6,27,112,4,3,0,0
-6,28,112,4,3,0,0
-6,29,112,4,3,0,0
-6,30,112,4,3,0,0
-6,31,112,4,3,0,0
-6,32,112,4,3,0,0
-6,33,112,4,3,0,0
-7,3,102,4,0,5,1
-7,4,112,4,0,8,1
-7,5,118,4,9,8,1
-7,6,112,4,9,0,1
-7,7,112,4,9,0,1
-7,8,112,4,9,0,1
-7,9,112,4,9,0,1
-7,10,112,4,9,0,0
-7,11,112,4,3,0,0
-7,12,112,4,9,0,1
-7,13,112,4,9,0,0
-7,14,112,4,3,0,0
-7,15,112,4,9,0,0
-7,16,112,4,3,0,0
-7,17,112,4,9,0,0
-7,18,112,4,3,0,0
-7,19,112,4,3,0,0
-7,20,112,4,3,0,0
-7,21,112,4,3,0,0
-7,22,112,4,3,0,0
-7,23,112,4,3,0,0
-7,24,112,4,3,0,0
-7,25,112,4,3,0,0
-7,26,112,4,3,0,0
-7,27,112,4,3,0,0
-7,28,112,4,3,0,0
-7,29,112,4,3,0,0
-7,30,112,4,3,0,0
-7,31,112,4,3,0,0
-7,32,112,4,3,0,0
+6,4,122,2,[other],8,[unknown]
+6,5,113,[mask],[other],8,[other]
+6,6,113,1,[other],[unknown],1
+6,7,113,[mask],[other],2,1
+6,8,113,1,[other],2,1
+6,9,113,3,[other],1,1
+6,10,121,1,[other],[mask],1
+6,11,108,0,7,1,1
+6,12,113,1,1,[mask],1
+6,13,108,8,9,1,1
+6,14,113,1,1,0,1
+6,15,105,1,9,1,[other]
+6,16,113,0,7,0,[other]
+6,17,113,0,[other],0,[other]
+6,18,113,8,[other],2,[other]
+6,19,113,[mask],[other],8,1
+6,20,113,1,[other],[mask],1
+6,21,113,1,[other],1,1
+6,22,113,1,[other],[mask],1
+6,23,113,1,[other],[mask],1
+6,24,121,1,[other],2,1
+6,25,113,1,[other],8,1
+6,26,105,1,1,[mask],1
+6,27,105,[other],9,1,[other]
+6,28,113,0,7,[unknown],1
+6,29,105,0,[other],1,1
+6,30,113,1,[other],0,1
+6,31,105,1,9,1,1
+6,32,113,0,7,1,1
+6,33,113,1,[other],1,1
+7,3,105,0,[other],8,[mask]
+7,4,105,0,[other],1,[other]
+7,5,105,0,9,6,[unknown]
+7,6,105,0,[other],1,1
+7,7,113,0,[other],1,2
+7,8,113,0,[other],1,[other]
+7,9,113,0,[other],[mask],1
+7,10,113,1,[other],[mask],1
+7,11,113,[mask],[other],8,1
+7,12,113,1,[other],[mask],1
+7,13,113,3,[other],1,1
+7,14,113,1,[other],[mask],1
+7,15,108,1,[other],0,1
+7,16,113,1,[other],7,1
+7,17,105,8,[other],2,1
+7,18,113,1,[other],2,1
+7,19,113,8,[other],1,1
+7,20,113,1,[other],2,1
+7,21,113,3,[other],1,1
+7,22,113,1,[other],2,1
+7,23,113,8,[other],1,1
+7,24,113,1,[other],2,1
+7,25,113,1,9,1,1
+7,26,103,1,9,1,1
+7,27,103,1,8,1,1
+7,28,103,0,9,7,1
+7,29,105,5,8,0,1
+7,30,105,1,8,1,1
+7,31,113,0,9,1,2
+7,32,113,0,[other],1,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
index a4fd8639..acc3eb04 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,112,4,9,5,0
-8,5,112,4,0,5,0
-8,6,112,7,8,5,0
-8,7,112,4,9,5,0
-8,8,112,4,9,5,0
-8,9,112,4,9,5,1
-8,10,112,4,9,0,0
-8,11,112,4,9,0,0
-8,12,112,4,3,0,0
-8,13,112,4,9,0,0
-8,14,112,4,3,0,0
-8,15,112,4,3,0,0
-8,16,112,4,3,0,0
-8,17,112,4,9,0,0
-8,18,112,4,3,0,0
-8,19,112,4,3,0,0
-8,20,112,4,3,0,0
-8,21,112,4,3,0,0
-8,22,112,4,3,0,0
-8,23,112,4,3,0,0
-8,24,112,4,3,0,0
-8,25,112,4,3,0,0
-8,26,112,4,3,0,0
-8,27,112,4,3,0,0
-8,28,112,4,3,0,0
-8,29,112,4,3,0,0
-8,30,112,4,3,0,0
-8,31,112,4,3,0,0
-8,32,112,4,3,0,0
-8,33,112,4,3,0,0
+8,4,101,1,[other],8,2
+8,5,104,1,[other],[unknown],2
+8,6,104,3,[other],[other],1
+8,7,120,5,1,0,1
+8,8,128,1,[other],0,1
+8,9,105,1,[other],7,[other]
+8,10,120,8,7,0,[other]
+8,11,120,5,[other],0,1
+8,12,120,1,7,2,1
+8,13,113,[mask],7,1,1
+8,14,113,1,[other],2,1
+8,15,113,1,[other],2,1
+8,16,113,1,[other],8,1
+8,17,113,1,[other],2,1
+8,18,113,[mask],1,8,1
+8,19,102,1,1,2,1
+8,20,128,1,4,7,1
+8,21,105,8,[other],7,1
+8,22,105,5,5,2,1
+8,23,113,1,8,0,1
+8,24,105,1,[other],7,1
+8,25,105,8,9,2,1
+8,26,113,1,[other],0,1
+8,27,113,8,[other],2,1
+8,28,113,1,[other],8,1
+8,29,113,1,[other],7,1
+8,30,105,8,[other],1,1
+8,31,113,1,[other],2,1
+8,32,113,[mask],[other],1,1
+8,33,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
index 42cdbf62..ff465a4a 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
@@ -1,31 +1,31 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,119,4,9,8,1
-9,4,102,4,9,8,1
-9,5,102,4,9,8,1
-9,6,102,4,9,0,1
-9,7,102,4,9,0,1
-9,8,102,6,9,0,1
-9,9,112,4,9,0,1
-9,10,112,4,9,0,0
-9,11,112,4,3,0,0
-9,12,112,4,3,0,0
-9,13,112,4,3,0,0
-9,14,112,4,3,0,0
-9,15,112,4,3,0,0
-9,16,112,4,9,0,0
-9,17,112,4,3,0,0
-9,18,112,4,3,0,0
-9,19,112,4,3,0,0
-9,20,112,4,3,0,0
-9,21,112,4,3,0,0
-9,22,112,4,3,0,0
-9,23,112,4,3,0,0
-9,24,112,4,3,0,0
-9,25,112,4,3,0,0
-9,26,112,4,3,0,0
-9,27,112,4,3,0,0
-9,28,112,4,3,0,0
-9,29,112,4,3,0,0
-9,30,112,4,3,0,0
-9,31,112,4,3,0,0
-9,32,112,4,3,0,0
+9,3,108,5,[other],8,2
+9,4,122,5,[other],7,1
+9,5,105,5,[other],7,[other]
+9,6,122,5,7,7,[other]
+9,7,122,5,[other],7,1
+9,8,122,5,[other],2,1
+9,9,113,[mask],7,2,1
+9,10,113,1,[other],2,1
+9,11,113,[mask],[other],2,1
+9,12,113,1,[other],2,1
+9,13,113,[mask],[other],2,1
+9,14,113,1,1,2,1
+9,15,113,[mask],[other],1,1
+9,16,113,1,1,2,1
+9,17,129,1,0,7,1
+9,18,108,1,8,7,1
+9,19,108,1,0,7,1
+9,20,108,1,[other],7,1
+9,21,124,8,[other],7,1
+9,22,105,8,[other],2,1
+9,23,113,1,[other],2,1
+9,24,113,8,[other],8,2
+9,25,113,[mask],[other],8,1
+9,26,113,1,[other],2,1
+9,27,113,8,[other],2,1
+9,28,113,1,[other],2,1
+9,29,113,1,[other],7,1
+9,30,108,8,9,2,1
+9,31,113,1,1,7,1
+9,32,108,8,[other],2,1
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
index 91865281..7cec3722 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
@@ -1,5 +1,6 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03451868,0.031921398,0.029532596,0.027744867,0.02760109,0.030242559,0.033749565,0.027132228,0.033893008,0.030211333,0.027323615,0.0273727,0.027940668,0.02913617,0.03199635,0.030564532,0.02930324,0.031335846,0.02836081,0.035665363,0.032468908,0.031003729,0.033122327,0.030269567,0.03076924,0.030623542,0.02832116,0.03149027,0.02777426,0.029747033,0.030456595,0.027495332,0.030911474
-0.027516257,0.028654557,0.032131042,0.031892225,0.024753015,0.034306064,0.031358637,0.02987722,0.03253693,0.030763064,0.027757306,0.029624237,0.029635688,0.028321382,0.029743204,0.027194707,0.02963858,0.02781912,0.030843636,0.029608745,0.027764315,0.02745214,0.029405791,0.029906224,0.033089224,0.034630504,0.02899261,0.03382853,0.038423352,0.030570587,0.031674568,0.03056725,0.029719174
-0.030222073,0.031830907,0.028284168,0.028407857,0.027115861,0.03235872,0.029203441,0.02976575,0.026956126,0.027495082,0.03189054,0.028712502,0.03204519,0.030992731,0.03524834,0.032565366,0.027216628,0.034662586,0.02938502,0.033734377,0.028660439,0.033143435,0.03173607,0.03174802,0.031067362,0.03277405,0.029025575,0.031590212,0.025999047,0.02817474,0.026592813,0.028226981,0.033168018
-0.025530642,0.026080305,0.02652737,0.03402691,0.027508548,0.038157314,0.027138228,0.031741947,0.03190737,0.030657675,0.026315706,0.030096767,0.028837977,0.03074533,0.031430572,0.033588294,0.028155396,0.029817235,0.024138339,0.028173123,0.027259734,0.032578666,0.034752455,0.034232706,0.03252036,0.0357678,0.029168166,0.032422952,0.029197408,0.030990023,0.025777996,0.031730004,0.0330266
+0.03490532,0.031998005,0.028772216,0.027195405,0.027683252,0.029920967,0.03279077,0.02628576,0.034324087,0.030802563,0.027264664,0.027657194,0.027639514,0.029530054,0.03239934,0.030662043,0.029952087,0.030814199,0.028125698,0.0344104,0.033223245,0.032099433,0.032986548,0.029886737,0.03260546,0.030854948,0.027623257,0.031297766,0.026375476,0.030162385,0.030712212,0.027799197,0.031239802
+0.027808135,0.029926706,0.03400922,0.030426748,0.025627963,0.031928588,0.03561634,0.030228743,0.031077368,0.029948497,0.028980114,0.028951673,0.029900176,0.027818188,0.028425556,0.025779901,0.03055118,0.029142106,0.03244458,0.030813629,0.027564976,0.02764529,0.02720103,0.029113112,0.032587785,0.03428341,0.029919686,0.031955376,0.03911367,0.030389592,0.03261028,0.028901618,0.02930875
+0.029915085,0.030543437,0.028834613,0.03042259,0.026555737,0.03297875,0.028495284,0.03012273,0.02954228,0.030395597,0.031444997,0.02910545,0.030163703,0.030961595,0.03490318,0.031798262,0.029193621,0.03289751,0.028760593,0.033082496,0.029938387,0.03467049,0.02820007,0.031503487,0.032639142,0.031294454,0.025620796,0.031352498,0.025782082,0.030491428,0.025737967,0.028993692,0.033657994
+0.025365606,0.025664778,0.026062079,0.032890774,0.026538692,0.03823054,0.028495807,0.029882822,0.033322647,0.031037634,0.025865804,0.029255202,0.02681062,0.03221474,0.03261645,0.034453724,0.027962716,0.028476784,0.023960778,0.027055632,0.026977597,0.03259397,0.03620625,0.032044176,0.032762516,0.037340216,0.030151956,0.033774957,0.029703537,0.0278705,0.028950956,0.030582692,0.03487683
+0.025282321,0.029473947,0.028269365,0.03373335,0.028858237,0.033462204,0.030326918,0.033530205,0.027763087,0.030093702,0.030169412,0.03151387,0.03000025,0.029444912,0.032865997,0.03329371,0.030292938,0.032526247,0.026917703,0.03018704,0.026444286,0.033763483,0.028074782,0.0319315,0.030718518,0.032087997,0.031459063,0.02930742,0.030223848,0.030085208,0.025420504,0.030391077,0.032086886
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
index a3d9edc1..25701dea 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
@@ -1,5 +1,6 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.024938403,0.025853546,0.030925326,0.035836563,0.027472962,0.033917453,0.031773735,0.033111185,0.030894933,0.03310667,0.03077764,0.031968474,0.027649606,0.029903907,0.03066213,0.030050345,0.028726462,0.024805613,0.029500253,0.026645845,0.025433421,0.030492553,0.029441224,0.033256244,0.030633831,0.03083032,0.029116308,0.03357199,0.035229903,0.031910744,0.02843861,0.03163502,0.031488776
-0.027546028,0.024828414,0.029011007,0.033997692,0.028309349,0.034571003,0.028885573,0.030300727,0.033193853,0.03057627,0.030879539,0.030720139,0.028182643,0.03336716,0.031966623,0.029043915,0.030032871,0.026891707,0.028006839,0.028216656,0.030012557,0.03335293,0.03443931,0.034874037,0.031277187,0.02787145,0.027704827,0.033389967,0.027017491,0.03146076,0.027681751,0.030754598,0.031635087
-0.025267702,0.02593611,0.026375175,0.033555325,0.02729746,0.03769069,0.028252527,0.03147091,0.031586744,0.02977779,0.026451198,0.029979987,0.028936028,0.031154884,0.0320948,0.03343466,0.027493577,0.029236242,0.024690645,0.028272472,0.026758416,0.031592447,0.03592271,0.03407506,0.03227401,0.035707034,0.030419732,0.03240453,0.030306093,0.030438218,0.026424794,0.031391647,0.033330467
-0.027293371,0.028949117,0.028703904,0.02919847,0.030929094,0.032399643,0.029274687,0.034145895,0.02837577,0.029229118,0.030326605,0.03315343,0.032393903,0.032588884,0.030739764,0.030741734,0.033645798,0.030104026,0.032112077,0.03169782,0.029715285,0.030085078,0.033176463,0.032097504,0.027618587,0.02773887,0.030057417,0.028018357,0.025789501,0.03361293,0.027174367,0.03010951,0.02880308
+0.025796872,0.025821052,0.030540183,0.03540621,0.026608227,0.03478382,0.03067776,0.031003399,0.033146046,0.03325188,0.030365815,0.0308043,0.027449563,0.030018117,0.03153884,0.029352657,0.029092524,0.02384649,0.029970763,0.026764145,0.02692843,0.03091666,0.031293344,0.03301509,0.030675245,0.031148765,0.027454989,0.03464577,0.03371407,0.03039152,0.029442204,0.031400308,0.032734923
+0.027645376,0.024716172,0.029397551,0.032479983,0.029057788,0.034062088,0.029561806,0.030762905,0.03275578,0.029999051,0.03151822,0.029582402,0.029124012,0.033837236,0.03152767,0.02815874,0.031342242,0.02682736,0.029414995,0.028265033,0.030185502,0.03224051,0.033708274,0.034261037,0.030121397,0.027603159,0.02855955,0.03343462,0.02645024,0.031482406,0.028913507,0.030383296,0.03262006
+0.024845945,0.025052948,0.0263014,0.032901563,0.025363559,0.03965492,0.028409896,0.029733066,0.03424027,0.03162598,0.025910662,0.028759446,0.026962476,0.032856528,0.03243185,0.032826383,0.029125627,0.02801517,0.024131,0.027621036,0.027519526,0.032559667,0.035635147,0.03229265,0.03338464,0.037081745,0.027807528,0.03353886,0.02957665,0.029155599,0.02823219,0.030683924,0.03576222
+0.02797149,0.029818716,0.028976986,0.028290737,0.03108153,0.03052239,0.02963015,0.03228555,0.028276766,0.028560005,0.03070743,0.032421984,0.03245931,0.03343161,0.031278666,0.030921867,0.03263645,0.03018529,0.032240782,0.0317139,0.03008393,0.02913051,0.034749724,0.031110376,0.02584681,0.027162941,0.032870043,0.028430955,0.02776051,0.031122573,0.029412452,0.030131536,0.028776024
+0.027345048,0.029193792,0.029317684,0.026135022,0.02694352,0.03692952,0.02964658,0.031842634,0.02922959,0.029339928,0.02875359,0.02905978,0.030834356,0.036634963,0.029727926,0.030966414,0.030507686,0.032904536,0.030803429,0.032436226,0.028633079,0.029277464,0.032264367,0.030212997,0.030541789,0.037181064,0.026679141,0.030335572,0.026149832,0.030504115,0.028901305,0.028029755,0.032737307
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
index e4570072..91c21c03 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
@@ -1,4 +1,5 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.033277225,0.030706238,0.03168895,0.033311088,0.029628102,0.026783517,0.033642292,0.028419774,0.03004662,0.030877694,0.030766364,0.02872636,0.028279921,0.025215175,0.032209065,0.028634787,0.031974144,0.031396896,0.027670972,0.030388262,0.03137201,0.0331837,0.024129286,0.02784322,0.03336912,0.025571778,0.033213194,0.032749813,0.033496857,0.029778576,0.03258353,0.029908357,0.029157136
-0.025729796,0.029602438,0.029645566,0.029040024,0.031028586,0.030800173,0.033909444,0.031098856,0.028809031,0.027932486,0.029731028,0.031625725,0.033525992,0.029835034,0.029137556,0.028564114,0.03216789,0.027496634,0.034953404,0.028403202,0.026970934,0.027889818,0.031215824,0.029021468,0.030363686,0.03254443,0.03467346,0.028671235,0.034229062,0.029859748,0.031822145,0.030721655,0.028979547
-0.028761078,0.027854176,0.03048727,0.034092404,0.029769653,0.030306583,0.03338378,0.032568734,0.0329339,0.03081607,0.026996588,0.028901242,0.028271142,0.026599998,0.027621023,0.02927734,0.034487005,0.032435104,0.022172494,0.029216152,0.028780058,0.030934772,0.025088819,0.030961733,0.032769565,0.028614981,0.03369568,0.030542089,0.03556081,0.033930458,0.030258873,0.03194647,0.029963972
+0.033804633,0.03133798,0.03240842,0.032249976,0.031567764,0.023804748,0.033451684,0.031181484,0.02904705,0.028814662,0.031253017,0.029255506,0.03155321,0.024031805,0.03100783,0.02836353,0.032650135,0.032079875,0.029468795,0.03124199,0.031478617,0.030717542,0.022784976,0.027685331,0.031394366,0.023952689,0.03493583,0.030857133,0.035084616,0.03195621,0.032159757,0.030788282,0.02763058
+0.025631752,0.029802956,0.030698422,0.029735252,0.029497724,0.031349394,0.033659987,0.030621238,0.02941033,0.029097665,0.030105164,0.031282384,0.032693755,0.030077938,0.02911067,0.026763074,0.03305479,0.026865568,0.036262047,0.029614927,0.028123945,0.028390907,0.031433348,0.029613422,0.029825589,0.03213038,0.0313533,0.028229501,0.03429874,0.030838396,0.030776728,0.03051359,0.02913713
+0.028800884,0.027588563,0.0300563,0.03443287,0.02912721,0.029885918,0.03512253,0.03208439,0.032546118,0.030156115,0.026779838,0.028835671,0.027242702,0.026432548,0.029021429,0.029896194,0.032417633,0.031113513,0.022803584,0.029385664,0.028680721,0.030133814,0.027381856,0.030590741,0.03248337,0.028955875,0.03544596,0.03154471,0.036519587,0.03206369,0.031347316,0.03071588,0.030406838
+0.02862093,0.03000527,0.029652234,0.02862441,0.03345083,0.03031905,0.028309684,0.032872725,0.028253555,0.03009939,0.0307454,0.033102486,0.032555282,0.033184092,0.02956358,0.03107351,0.03461472,0.031375494,0.032400206,0.03102256,0.031462118,0.030181756,0.03170402,0.030066047,0.026040724,0.02641749,0.0313496,0.027319029,0.026150461,0.0320657,0.028823854,0.030786902,0.027786825
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
index fc1bbb92..4b357202 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
@@ -1,7 +1,9 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.033417575,0.030952973,0.031909764,0.03328892,0.029420735,0.02633423,0.03413355,0.028891983,0.02999831,0.030924743,0.030936386,0.02908393,0.028261695,0.025051711,0.032034438,0.028418737,0.032599755,0.03089184,0.027745165,0.0306719,0.031189425,0.032526094,0.023779722,0.027861075,0.03290723,0.02475645,0.03326867,0.03255766,0.03400532,0.030586168,0.032822493,0.030047486,0.028723868
-0.034453217,0.0305892,0.029416787,0.029322578,0.028612511,0.031616293,0.028566025,0.026076926,0.03032813,0.02860342,0.032680735,0.028623346,0.028884213,0.03250039,0.034757424,0.029118933,0.027070051,0.031802647,0.031775847,0.03249104,0.03419997,0.0347499,0.034769706,0.03168993,0.032166995,0.029576255,0.02648399,0.032632153,0.022518143,0.027031703,0.028449764,0.027058687,0.031383082
-0.03126686,0.028659452,0.026583998,0.02960171,0.029598372,0.0348543,0.028967595,0.027243897,0.03295455,0.028712928,0.027025098,0.02840309,0.029796531,0.030191114,0.032229044,0.030590648,0.030458689,0.030780101,0.028688198,0.030927148,0.03236309,0.035447787,0.03583643,0.031489626,0.034140706,0.032815706,0.028919876,0.032643765,0.022453642,0.028814038,0.028506618,0.028614122,0.030421335
-0.024738476,0.030802721,0.029338533,0.030698724,0.026044017,0.036583103,0.02847107,0.0297974,0.030350022,0.031745814,0.027877493,0.030870043,0.031255092,0.029293576,0.030610705,0.032736078,0.027911829,0.030885635,0.028708616,0.030616442,0.026017355,0.031379726,0.033175353,0.032386277,0.030061183,0.03907589,0.027840383,0.030412367,0.03361048,0.028390406,0.025730342,0.030818133,0.031766724
-0.027777517,0.028804556,0.031672627,0.030206317,0.025076197,0.034134787,0.032481443,0.029717883,0.03154765,0.028994586,0.027599279,0.028896887,0.030154243,0.028881382,0.030242773,0.027787082,0.02839826,0.0288448,0.031094935,0.029876556,0.027011143,0.026954878,0.030751621,0.029737474,0.032815382,0.035880763,0.031499054,0.03416633,0.03777237,0.028893098,0.032805875,0.02953868,0.029983671
-0.027476907,0.029252572,0.029974613,0.03427469,0.03160824,0.029946703,0.026536744,0.03142502,0.029297996,0.033668708,0.030066026,0.031427413,0.031061247,0.026917415,0.030150387,0.033821,0.0326446,0.032380875,0.025550779,0.02765895,0.02903609,0.035553962,0.025652928,0.029973598,0.030050268,0.028396083,0.03117175,0.03078121,0.032512754,0.03008112,0.02760991,0.033881046,0.03015839
+0.03365363,0.03131194,0.032062005,0.032612655,0.030762598,0.023996722,0.033894006,0.03161808,0.028858617,0.028537845,0.031369943,0.029386261,0.031269576,0.0240121,0.03156647,0.028297028,0.03270311,0.03169595,0.028860392,0.031636294,0.03105152,0.03050513,0.023036947,0.028302876,0.031403825,0.0235559,0.03512947,0.031095052,0.034982767,0.032558214,0.031808164,0.030668166,0.027796706
+0.032761578,0.029872429,0.029597605,0.029935349,0.026900567,0.032399144,0.02855435,0.026653757,0.03182153,0.03016848,0.032555472,0.028656261,0.029006623,0.03235095,0.03480727,0.028403722,0.028092936,0.029561672,0.03174143,0.03233374,0.033174057,0.034100316,0.03289924,0.032593768,0.032962408,0.029540613,0.02438331,0.032330763,0.023944201,0.029516606,0.027441561,0.028192038,0.0327462
+0.030725045,0.028272143,0.026365971,0.028971849,0.028257614,0.034422457,0.03088777,0.026187155,0.03345275,0.028805315,0.027079374,0.02812067,0.027472235,0.03160347,0.034400526,0.031103302,0.028490582,0.02971379,0.028624378,0.030313667,0.030894523,0.03547529,0.037652876,0.031707373,0.034248933,0.03381967,0.030158943,0.03319742,0.023066722,0.02731636,0.029626993,0.02753793,0.032026935
+0.025067946,0.025263725,0.026296385,0.033531774,0.026559269,0.03958333,0.027857842,0.030703193,0.033559784,0.03183785,0.025875572,0.029518714,0.026663154,0.032539647,0.03160153,0.034056365,0.029149989,0.02885333,0.023305051,0.02679659,0.027072044,0.033060253,0.034554478,0.032332253,0.03327132,0.037098464,0.028822728,0.033504702,0.02874793,0.029003797,0.028169123,0.03096053,0.03478138
+0.024328558,0.03019246,0.02978609,0.030933373,0.02664791,0.03585672,0.029219525,0.03113439,0.030371258,0.031899795,0.027526228,0.031095784,0.02991037,0.029948168,0.030375885,0.03407043,0.028136484,0.03192227,0.028028654,0.029312856,0.025214395,0.031401776,0.030239172,0.030062895,0.029838474,0.03965154,0.029726071,0.030511681,0.035435908,0.027180875,0.027923727,0.030501928,0.03161439
+0.027642481,0.029439252,0.033424422,0.030074678,0.02514382,0.033257872,0.03443825,0.02897289,0.032224122,0.03009342,0.028927157,0.028554885,0.029879149,0.029093452,0.028734183,0.025436884,0.030429183,0.029044895,0.032307327,0.0309847,0.028431837,0.028267644,0.029025367,0.029751183,0.03299192,0.035030864,0.02826856,0.031771418,0.037300482,0.030140122,0.03184548,0.028867587,0.030204555
+0.026242485,0.02856462,0.030316794,0.033157557,0.03299111,0.029088097,0.027124817,0.031284798,0.03198276,0.032629397,0.030249553,0.031380426,0.03177847,0.027848225,0.029764337,0.03523869,0.029829431,0.031178579,0.027072916,0.026859952,0.02733578,0.034253642,0.027440054,0.030315153,0.027151577,0.03023951,0.03223471,0.030502116,0.035234425,0.027642915,0.028225683,0.034010116,0.030831262
+0.031010924,0.027717909,0.028334908,0.029572029,0.031026728,0.028185181,0.03371703,0.03252502,0.028359665,0.028503757,0.029602813,0.028722132,0.03139693,0.029956581,0.033280496,0.027820814,0.037417598,0.029648174,0.03241054,0.03330766,0.034037765,0.030647116,0.027304817,0.027959816,0.031536154,0.023645312,0.031318452,0.029235678,0.025905538,0.035960104,0.030277228,0.029216642,0.030438429
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
index 66ac322a..b58e1cad 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
@@ -1,4 +1,5 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02787407,0.027104642,0.030358018,0.034553517,0.028322224,0.031050757,0.034033015,0.031545833,0.033861537,0.030875284,0.027181892,0.028938169,0.027790133,0.027383799,0.028615393,0.028780295,0.03324079,0.030557051,0.022777924,0.029385028,0.028478311,0.030502185,0.026848188,0.03165467,0.032651097,0.028698592,0.032736804,0.031015739,0.037046205,0.033572696,0.030005831,0.03182181,0.030738411
-0.02670878,0.025883488,0.02744727,0.03386368,0.029877376,0.032459594,0.030426193,0.03247706,0.02878972,0.027941974,0.027739787,0.030217001,0.030183967,0.029890094,0.032915857,0.03072404,0.030120345,0.030178469,0.027195016,0.02812506,0.028859448,0.029241446,0.031200247,0.029883275,0.0328588,0.029916031,0.035874527,0.030879542,0.033642642,0.031450473,0.030106302,0.03142932,0.03149319
-0.032150064,0.028221862,0.028330833,0.02810167,0.027819155,0.03160524,0.027474659,0.027735386,0.033368625,0.028838621,0.028449835,0.027917936,0.032627217,0.031592034,0.03244308,0.028486887,0.03201186,0.028425962,0.033350077,0.031887256,0.03447993,0.030445999,0.032843858,0.028395893,0.032929353,0.029940993,0.026082069,0.032996614,0.026771395,0.031200353,0.031379223,0.030597694,0.031098424
+0.028346347,0.027353464,0.030344704,0.034818895,0.0283053,0.03005997,0.0355836,0.03275808,0.032522377,0.030344306,0.027134521,0.028696552,0.027392156,0.026236838,0.029147837,0.02928847,0.033233523,0.03095584,0.022680035,0.029963996,0.028415956,0.030157642,0.026274486,0.030956142,0.032518826,0.02823263,0.034416623,0.031546608,0.0369284,0.033256564,0.03058284,0.03078088,0.030765546
+0.02614715,0.0258215,0.027550466,0.034463495,0.028367938,0.032731164,0.030727468,0.032352015,0.028698286,0.028598435,0.027810419,0.029569183,0.029639808,0.030452486,0.033374913,0.030064132,0.03173377,0.029634332,0.027205918,0.028680194,0.030136917,0.029390203,0.031379256,0.029033292,0.032703232,0.029795818,0.034223244,0.030701801,0.03382241,0.031033205,0.03053796,0.030996071,0.03262359
+0.031898044,0.028340638,0.02751846,0.026926287,0.027034767,0.030630607,0.029686758,0.027121056,0.03294052,0.029303096,0.027993035,0.02730981,0.03156264,0.03152772,0.034503076,0.02913811,0.031860307,0.026993573,0.033478737,0.031741023,0.03363095,0.030359568,0.033719897,0.027680459,0.03301379,0.030039832,0.02723801,0.032895096,0.027244497,0.031041017,0.032970347,0.029797535,0.032860763
+0.024685703,0.029492697,0.028752888,0.031134205,0.0272171,0.037294548,0.028083863,0.03050206,0.030993259,0.031962592,0.027071144,0.031027092,0.0293045,0.030799314,0.030283516,0.034436323,0.028116086,0.03160242,0.02737299,0.02817356,0.025712473,0.032479364,0.031470202,0.030321851,0.031176731,0.040771965,0.02915091,0.030929118,0.03230334,0.026936619,0.027898025,0.03060591,0.031937685
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
index b6233023..c9e83edb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
@@ -1,8 +1,10 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027693711,0.029121894,0.03263769,0.030477673,0.025150716,0.03287544,0.032045007,0.030850839,0.031303454,0.02993946,0.027595572,0.02886764,0.031017965,0.027735751,0.02935163,0.027287817,0.030652735,0.029687006,0.030908603,0.030391091,0.027535228,0.026636822,0.027238684,0.028523553,0.03267429,0.03431938,0.03061821,0.033247806,0.040079303,0.030726168,0.03266249,0.03049625,0.02965006
-0.025466071,0.029875752,0.029835982,0.029151956,0.028186664,0.03173176,0.03316926,0.030232249,0.031046985,0.02931059,0.029201102,0.03172116,0.033616595,0.029696383,0.029422067,0.027533485,0.034060102,0.026428862,0.035074487,0.030573001,0.027960634,0.02801773,0.03157427,0.029605145,0.0301314,0.03144874,0.030063352,0.028544225,0.03480755,0.032065663,0.030100726,0.031310387,0.029035721
-0.026478115,0.027542712,0.025084658,0.02916108,0.0296517,0.030709622,0.029443054,0.031282466,0.031903658,0.029175015,0.02586855,0.028222973,0.03547591,0.02836005,0.03144234,0.03363753,0.03476335,0.02754813,0.028392714,0.029871454,0.028496739,0.030743234,0.02964027,0.027810814,0.032352984,0.031880755,0.030597484,0.03063106,0.032444194,0.03375546,0.029653978,0.03467788,0.033299997
-0.026364712,0.029757837,0.032045525,0.031597987,0.030919015,0.031508707,0.031495456,0.03135381,0.03211043,0.03435329,0.03110224,0.032581497,0.029810883,0.029403877,0.027003385,0.030518897,0.03186091,0.026397405,0.03112966,0.027613735,0.026846573,0.032183122,0.029105142,0.032442607,0.028408209,0.031621095,0.028020909,0.03026787,0.031812698,0.030703535,0.02858388,0.031764213,0.029310849
-0.03315158,0.030279914,0.029608734,0.029177485,0.027309386,0.032146376,0.0286247,0.026008293,0.032145012,0.029281488,0.03193118,0.028416706,0.029567713,0.032534126,0.033859003,0.02824213,0.02870284,0.031056287,0.0315024,0.033499297,0.03437484,0.034205765,0.034372948,0.031886913,0.032041393,0.029588355,0.024660248,0.03200928,0.023830036,0.028581526,0.027754126,0.027885024,0.03176486
-0.025848068,0.028431289,0.029892355,0.028693581,0.030470718,0.029235307,0.030940453,0.03242136,0.029718446,0.027495537,0.029365137,0.02925045,0.03549993,0.030839637,0.028326172,0.029380446,0.034169305,0.03261548,0.028191667,0.0298291,0.028319623,0.027590321,0.02789852,0.028912824,0.029687619,0.03054694,0.033803966,0.027655903,0.037782732,0.031797566,0.030679887,0.03306369,0.031645942
-0.024057524,0.02544963,0.025607307,0.029881254,0.029157504,0.031960018,0.030122014,0.032780554,0.031288777,0.028194267,0.027988294,0.028752653,0.036150485,0.031077877,0.031034835,0.031049933,0.034480184,0.02570712,0.030488791,0.028481508,0.027200652,0.029486086,0.029952202,0.029214997,0.032385252,0.032020196,0.029338684,0.03019485,0.03324283,0.034965817,0.028807197,0.034836706,0.034644097
+0.02937184,0.030781249,0.033119388,0.028550683,0.025509233,0.03206324,0.03396557,0.028235912,0.032892834,0.029829921,0.028320564,0.02828424,0.030789617,0.02829313,0.028306503,0.025561279,0.031485945,0.029745061,0.0323852,0.031503476,0.02918103,0.028602777,0.028324347,0.028798053,0.033785295,0.0349804,0.028154539,0.03169164,0.035794634,0.030512646,0.032773368,0.02908114,0.029325323
+0.025854453,0.030651193,0.030702325,0.02862513,0.029471286,0.030098798,0.034639407,0.031529788,0.028793877,0.028493036,0.029873291,0.031217096,0.03360634,0.029397383,0.029140493,0.02725684,0.033899527,0.027715212,0.036060736,0.030847853,0.027600395,0.027915468,0.030293047,0.02915647,0.028903313,0.031247552,0.032694653,0.02797646,0.034875035,0.031475507,0.031034105,0.030351939,0.028601935
+0.025826376,0.027251828,0.02673951,0.030150859,0.028219849,0.03097317,0.029552381,0.032052875,0.033098787,0.032143876,0.026451845,0.02858729,0.033598788,0.028336221,0.030879049,0.031548064,0.03830193,0.026910027,0.028226435,0.029459467,0.02940237,0.03086073,0.026327921,0.026920898,0.032483302,0.030522237,0.027463144,0.029610101,0.033719517,0.035605237,0.029845843,0.034718696,0.03421142
+0.02661512,0.030802475,0.032640494,0.031011712,0.03189831,0.030164946,0.03295102,0.030942937,0.031764835,0.032526337,0.032349735,0.03283627,0.03008033,0.029224372,0.02729727,0.029901998,0.029873366,0.025909916,0.03347985,0.027992664,0.02590215,0.030691914,0.030372458,0.033140227,0.026693273,0.03153677,0.029790008,0.029715454,0.033401065,0.02932462,0.029047104,0.03119655,0.028924473
+0.02814665,0.029950805,0.03147324,0.032071218,0.027423255,0.035451543,0.028824573,0.029990712,0.030143596,0.029638186,0.029570555,0.03163136,0.027836546,0.031823345,0.030531272,0.03083544,0.027090862,0.03556963,0.026167938,0.030581702,0.0284657,0.03182817,0.03386525,0.033583216,0.029266147,0.033167303,0.031999893,0.031316996,0.029996168,0.026834186,0.027982738,0.028263362,0.028678378
+0.033281386,0.030130476,0.030221825,0.02840587,0.027716415,0.030989772,0.02878914,0.026732758,0.031153036,0.02907955,0.033042736,0.027861213,0.030571226,0.03249253,0.034299064,0.028136875,0.02837106,0.03143917,0.03291027,0.033115998,0.03358187,0.03396475,0.031448998,0.031539276,0.032543357,0.029416593,0.025101459,0.031998575,0.024063027,0.029163437,0.027996125,0.027905693,0.03253653
+0.024855413,0.026666297,0.02936113,0.031087212,0.02884285,0.0313365,0.029449705,0.033253677,0.030825984,0.029471586,0.028742261,0.02912062,0.034331124,0.030908402,0.028541783,0.028688319,0.03659657,0.030675147,0.027317889,0.029583992,0.029643653,0.028368162,0.027667206,0.028800705,0.030276459,0.029693466,0.030268844,0.028184077,0.036950678,0.034286764,0.029671533,0.033815984,0.032715973
+0.025155025,0.026130434,0.026154922,0.02954241,0.028716099,0.030518187,0.031584375,0.031437334,0.030625204,0.027734924,0.028301684,0.027579084,0.033700705,0.030727208,0.03288186,0.03139454,0.032228857,0.026481455,0.030001791,0.028490013,0.027577322,0.029461626,0.031345062,0.028545208,0.031811137,0.032734197,0.032353926,0.031332225,0.03433076,0.031055674,0.0314804,0.03250627,0.036080144
+0.025650796,0.030116925,0.029914461,0.028343417,0.029756019,0.031495847,0.033737093,0.030940576,0.0293521,0.028435258,0.029297698,0.031161314,0.033550072,0.030386096,0.028744891,0.027141783,0.034294646,0.027434524,0.036070127,0.03000383,0.027885038,0.028399905,0.03164065,0.029249499,0.029878315,0.032681726,0.0320276,0.02796962,0.03277645,0.031302802,0.03109174,0.03045507,0.028814139
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
index c4d448a9..87c707bb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
@@ -1,5 +1,6 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02861608,0.02783555,0.031125112,0.034070816,0.029551925,0.02920194,0.03443248,0.032701004,0.03236621,0.031025173,0.027565764,0.029052038,0.028101817,0.02655173,0.028142825,0.029290026,0.03396453,0.031744126,0.022819739,0.029417034,0.028412648,0.03002513,0.02473677,0.030448401,0.031774003,0.027520519,0.034440182,0.030517727,0.037965015,0.03373207,0.030928032,0.03187111,0.030052474
-0.029571708,0.027468769,0.028687142,0.030722428,0.030333482,0.028803239,0.034810323,0.032661207,0.028865434,0.028517043,0.03039022,0.029620783,0.030924972,0.029667925,0.033478405,0.02773338,0.03547341,0.02907416,0.032520335,0.033110958,0.03192609,0.030285636,0.027565196,0.028671881,0.031343654,0.024217593,0.030943919,0.029093694,0.028387735,0.03578851,0.029546076,0.029263297,0.0305314
-0.03309813,0.030376514,0.031082107,0.032114696,0.0303959,0.026847737,0.033983357,0.028783653,0.029035237,0.02931486,0.030565187,0.0282617,0.028926874,0.025849348,0.032453902,0.029420223,0.030807314,0.032321904,0.027733354,0.029939143,0.030494139,0.032511473,0.024984352,0.027675122,0.033265285,0.026638621,0.03590076,0.032994192,0.033117052,0.028538441,0.033555407,0.02946036,0.029553706
-0.032728393,0.028721752,0.02882793,0.027880307,0.028566426,0.030735191,0.027027326,0.027918302,0.033026785,0.029266033,0.028729456,0.028066441,0.03297658,0.031279273,0.031866837,0.028559575,0.032620937,0.02890799,0.033787943,0.03182001,0.03511367,0.030451855,0.03162444,0.027786268,0.03249884,0.029433765,0.025993556,0.032565445,0.026775395,0.031265218,0.031698216,0.030825824,0.030654097
+0.029685352,0.027744511,0.030044284,0.033951394,0.028732833,0.02994491,0.034402113,0.031651333,0.03374663,0.03049073,0.026828067,0.028434549,0.027637096,0.026393222,0.028637927,0.029101957,0.034120213,0.031485118,0.02227293,0.030039264,0.029686691,0.030844377,0.026529493,0.030811347,0.03335207,0.028110413,0.033255078,0.031356674,0.034736305,0.033637084,0.030754102,0.031082017,0.030499963
+0.030537475,0.027899254,0.029378591,0.0297096,0.031716786,0.02693288,0.03433145,0.033088665,0.02740053,0.02792913,0.030472768,0.028694458,0.03220934,0.029565549,0.032662544,0.027634623,0.03633886,0.030167157,0.033250082,0.03298737,0.033357035,0.029501474,0.026456432,0.027505033,0.030585667,0.023674536,0.03291113,0.028946836,0.028558468,0.035006773,0.030973727,0.02936918,0.03024652
+0.033833686,0.030788528,0.032240216,0.032861616,0.030781893,0.024426019,0.03269575,0.030321872,0.029727243,0.029050065,0.03131785,0.029049996,0.0312546,0.024382828,0.031556305,0.027793216,0.03238788,0.031475447,0.02939823,0.031028157,0.032263253,0.03098234,0.023543414,0.02798694,0.03215333,0.024038076,0.03364459,0.031184737,0.03487223,0.03188977,0.031967424,0.030930823,0.028171666
+0.032235544,0.028513635,0.027943423,0.026533665,0.027607339,0.030138258,0.029216288,0.027271325,0.03264726,0.02924624,0.028209869,0.027331444,0.032256123,0.03174167,0.03372429,0.028924001,0.032310896,0.027531,0.034024764,0.031696826,0.034188505,0.029953811,0.03298522,0.027146649,0.032747537,0.029807689,0.027179489,0.032454353,0.027391035,0.031187288,0.03338235,0.03002738,0.032444846
+0.028649943,0.02983568,0.029618803,0.028534783,0.033335116,0.030336117,0.028125117,0.032931272,0.028101504,0.029766463,0.030861683,0.032960948,0.032717913,0.033539113,0.029642688,0.031016078,0.034416024,0.03157327,0.03229032,0.0310232,0.03154139,0.029931167,0.03196622,0.030088337,0.0260188,0.026400523,0.031526826,0.027428826,0.026188847,0.03198211,0.028969765,0.030762099,0.027919022
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
index 4e8c02ca..08bba466 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
@@ -1,4 +1,5 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03475592,0.03205031,0.029370865,0.02778089,0.027614197,0.030346887,0.03223454,0.026713328,0.03431964,0.030279862,0.027550474,0.027702024,0.028271765,0.029716752,0.031870633,0.03032701,0.029443298,0.03097386,0.02868564,0.035414,0.033398263,0.030973615,0.033969566,0.030356236,0.030772602,0.030513762,0.027257016,0.031204006,0.027268384,0.029877514,0.030222507,0.02795559,0.03080913
-0.027402356,0.027945139,0.02530322,0.029322542,0.029918933,0.029952679,0.028517315,0.03166655,0.031823326,0.02963542,0.026160058,0.028808074,0.03524668,0.028342647,0.03120509,0.0340563,0.03451171,0.02670396,0.02878434,0.029324375,0.028796544,0.029988907,0.029430298,0.027544383,0.031944152,0.031181917,0.030243525,0.031055551,0.032528445,0.03417648,0.030487247,0.03531119,0.032680634
-0.032010403,0.03287599,0.029452827,0.027987542,0.027845826,0.03130848,0.027573993,0.030086143,0.02799824,0.028927118,0.031524617,0.02894558,0.032493584,0.030415898,0.03303464,0.03172704,0.029664379,0.035562575,0.029617772,0.034429386,0.031013394,0.032945413,0.029464815,0.030695107,0.03063894,0.031342026,0.026874157,0.030781394,0.025375465,0.029842958,0.026907688,0.028956134,0.031680416
+0.0347061,0.032037724,0.029720293,0.026950317,0.027587326,0.029580832,0.033706103,0.02672045,0.0336072,0.0304948,0.027389366,0.027341463,0.027687615,0.029654475,0.03191975,0.030349722,0.029564764,0.03208654,0.028236479,0.035157055,0.033096552,0.031182028,0.03215033,0.02952733,0.0320103,0.031021614,0.028249703,0.031084804,0.02777836,0.02990182,0.031047527,0.027308928,0.031142361
+0.026352162,0.027308615,0.026728509,0.030381735,0.027763495,0.031036979,0.029111193,0.031803854,0.033297487,0.032028094,0.026604233,0.028412612,0.03335847,0.028347764,0.031116711,0.031521976,0.037724625,0.027063178,0.02756419,0.02944993,0.029611493,0.030874036,0.026384085,0.027145538,0.033017457,0.03050442,0.027128858,0.030084612,0.03357262,0.03557978,0.029888283,0.03474908,0.034483936
+0.031161968,0.030985577,0.028069872,0.029070867,0.02766797,0.031849217,0.028295552,0.029846,0.028211916,0.028096769,0.031157041,0.028553555,0.030611211,0.031491056,0.03540855,0.033342153,0.027244546,0.03506385,0.028400969,0.032600712,0.029665994,0.034185328,0.029610706,0.030778438,0.03282092,0.032744713,0.02873029,0.032103863,0.02512524,0.028251864,0.02723303,0.028246962,0.033373386
+0.025955163,0.027249493,0.029127032,0.030652158,0.03015496,0.03147695,0.028033996,0.034602076,0.030288333,0.029120373,0.028158711,0.029552525,0.03452732,0.030882543,0.02711481,0.029857256,0.03783241,0.03359149,0.025146987,0.029117852,0.029665263,0.028882185,0.026112327,0.028500061,0.030931538,0.02985717,0.031220669,0.028128866,0.03408741,0.03489066,0.02980184,0.03413182,0.03134773

From 14929b72a8dd55de337d093d2b0bd657985dca60 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 19:24:00 +0200
Subject: [PATCH 45/81] Enforce v2

---
 src/sequifier/helpers.py                      |  6 +--
 ...uifier_dataset_from_folder_parquet_lazy.py |  6 +--
 src/sequifier/preprocess.py                   | 19 +---------
 src/sequifier/train.py                        |  2 +-
 tests/integration/test_preprocessing.py       |  1 +
 ...uifier_dataset_from_folder_parquet_lazy.py |  5 ++-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  2 +-
 tests/unit/test_helpers.py                    | 37 +++++++++++++++++--
 tests/unit/test_infer.py                      |  4 ++
 tests/unit/test_preprocess.py                 | 16 ++------
 tests/unit/test_train.py                      |  4 +-
 11 files changed, 55 insertions(+), 47 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index d228b456..a236b3b5 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -128,7 +128,7 @@ def sequence_layout_from_metadata(
     layout = SequenceLayout(
         context_length=metadata_context_length,
         max_lookahead=max_lookahead,
-        sequence_layout_version=int(metadata.get("sequence_layout_version", 1)),
+        sequence_layout_version=int(metadata["sequence_layout_version"]),
     )
     if layout.sample_length != sample_length:
         raise ValueError(
@@ -432,10 +432,8 @@ def build_valid_mask(
 
 
 @beartype
-def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Optional[Tensor]:
+def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Tensor:
     """Extracts one leftPadLength value per long-format subsequence."""
-    if "leftPadLength" not in data.columns:
-        return None
 
     assert {"sequenceId", "subsequenceId"}.issubset(data.columns)
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 980007ae..f716ae1c 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -284,7 +284,6 @@ def __iter__(
 
             # Process Long format data structures into PyTorch Tensors
             new_seq, new_tgt = {}, {}
-            expected_samples = len(worker_indices_np)
 
             # 2. Iterate over the expected config columns, not the dynamically found ones
             # Process inputs
@@ -296,10 +295,7 @@ def __iter__(
                         dtype=self.column_torch_types[col_name],
                     )
                 else:
-                    new_seq[col_name] = torch.zeros(
-                        (expected_samples, train_seq_len),
-                        dtype=self.column_torch_types[col_name],
-                    )
+                    raise ValueError(f"Column not found in input data: {col_name}")
 
             # Process targets
             for col_name in self.config.target_columns:
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index cdbc1740..9bcb4253 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -1202,24 +1202,9 @@ def load_precomputed_id_maps(
 
                     min_val = min(user_values)
                     if min_val == 2:
-                        warnings.warn(
-                            f"Precomputed map {file} uses legacy user IDs starting at 2; shifting user IDs by 1 to reserve the BERT mask ID.",
-                            stacklevel=2,
+                        raise ValueError(
+                            f"Precomputed map {file} uses legacy user IDs starting at 2; shifting user IDs by 1 to reserve the BERT mask ID."
                         )
-                        m = {
-                            key: value + 1
-                            if key not in SPECIAL_TOKEN_LABELS
-                            and value >= SPECIAL_TOKEN_IDS.mask
-                            else value
-                            for key, value in m.items()
-                        }
-                        user_values = [
-                            value
-                            for key, value in m.items()
-                            if key not in SPECIAL_TOKEN_LABELS
-                        ]
-                        min_val = min(user_values)
-
                     if min_val != SPECIAL_TOKEN_IDS.user_start:
                         raise ValueError(
                             f"minimum non-reserved value in map {file} is {min_val}, must be {SPECIAL_TOKEN_IDS.user_start}."
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 66d7ff88..794a896a 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -92,7 +92,7 @@ def _as_sequifier_batch(batch: Any) -> SequifierBatch:
     return SequifierBatch(
         inputs=data,
         targets=targets,
-        metadata=metadata or {},
+        metadata=metadata,
         sequence_ids=sequence_ids,
         subsequence_ids=subsequence_ids,
     )
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 3cabc262..9eeb3069 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -40,6 +40,7 @@ def test_metadata_config(metadata_configs):
     for file_name, metadata_config in metadata_configs.items():
         print(f"Verifying metadata_config for: {file_name}")
         assert list(metadata_config.keys()) == expected_metadata_keys
+        assert metadata_config["sequence_layout_version"] == 2
         assert metadata_config["special_token_ids"] == {
             "[unknown]": 0,
             "[other]": 1,
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 856b386b..24646c20 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -21,7 +21,7 @@ def _folder_metadata(total_samples, batch_files):
         "context_length": CONTEXT_LENGTH,
         "max_lookahead": MAX_LOOKAHEAD,
         "sample_length": SAMPLE_LENGTH,
-        "sequence_layout_version": 1,
+        "sequence_layout_version": 2,
     }
 
 
@@ -54,6 +54,7 @@ def dataset_path(tmp_path):
         "sequenceId": pl.Int64,
         "subsequenceId": pl.Int64,
         "startItemPosition": pl.Int64,
+        "leftPadLength": pl.Int64,
         "inputCol": pl.String,
         "2": pl.Float64,
         "1": pl.Float64,
@@ -66,7 +67,7 @@ def dataset_path(tmp_path):
         filename = f"file_{i}.parquet"
         rows = []
         for s in range(10):
-            rows.append((s, 0, s * 2, "item", float(s), float(s + 1), float(s + 2)))
+            rows.append((s, 0, s * 2, 0, "item", float(s), float(s + 1), float(s + 2)))
 
         df = pl.DataFrame(rows, schema=schema, orient="row")
         df.write_parquet(data_dir / filename)
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 6cec3c92..879d855f 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -20,7 +20,7 @@ def _folder_metadata(total_samples, batch_files):
         "context_length": CONTEXT_LENGTH,
         "max_lookahead": MAX_LOOKAHEAD,
         "sample_length": SAMPLE_LENGTH,
-        "sequence_layout_version": 1,
+        "sequence_layout_version": 2,
     }
 
 
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index b36c1624..2b9f6b0e 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -74,6 +74,10 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     # Seq 1: [10, 20, 30, 40] (where 40 is t=0, 30 is t=1, etc.)
     data = pl.DataFrame(
         {
+            "sequenceId": [1, 2],
+            "subsequenceId": [0, 0],
+            "startItemPosition": [0, 0],
+            "leftPadLength": [0, 0],
             "inputCol": ["A", "A"],
             "3": [10, 11],
             "2": [20, 21],
@@ -99,7 +103,14 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     # 1. Check Keys
     assert "A" in tensors
     assert "A_target" in tensors
-    assert metadata == {}
+    assert torch.equal(
+        metadata["attention_valid_mask"],
+        torch.tensor([[True, True, True], [True, True, True]]),
+    )
+    assert torch.equal(
+        metadata["target_valid_mask"],
+        torch.tensor([[True, True, True], [True, True, True]]),
+    )
 
     # 2. Check Input Tensor (Cols 3, 2, 1)
     # Row 0: [10, 20, 30]
@@ -123,7 +134,17 @@ def test_numpy_to_pytorch_dtypes():
     # (Polars columns usually have a single type).
 
     # Case 1: Integer
-    data_int = pl.DataFrame({"inputCol": ["int_col"], "1": [10], "0": [20]})
+    data_int = pl.DataFrame(
+        {
+            "sequenceId": [1],
+            "subsequenceId": [0],
+            "startItemPosition": [0],
+            "leftPadLength": [0],
+            "inputCol": ["int_col"],
+            "1": [10],
+            "0": [20],
+        }
+    )
     tensors_int, _ = numpy_to_pytorch(
         data_int,
         {"int_col": torch.int64},
@@ -136,7 +157,17 @@ def test_numpy_to_pytorch_dtypes():
     assert tensors_int["int_col"].dtype == torch.int64
 
     # Case 2: Float
-    data_float = pl.DataFrame({"inputCol": ["float_col"], "1": [10.5], "0": [20.5]})
+    data_float = pl.DataFrame(
+        {
+            "sequenceId": [1],
+            "subsequenceId": [0],
+            "startItemPosition": [0],
+            "leftPadLength": [0],
+            "inputCol": ["float_col"],
+            "1": [10.5],
+            "0": [20.5],
+        }
+    )
     tensors_float, _ = numpy_to_pytorch(
         data_float,
         {"float_col": torch.float32},
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 568a434e..2d16fc81 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -249,6 +249,10 @@ def test_load_inferer_config_rejects_mismatched_metadata_special_token_ids(tmp_p
         json.dumps(
             {
                 "split_paths": [str(data_path)],
+                "context_length": 3,
+                "max_lookahead": 1,
+                "sample_length": 4,
+                "sequence_layout_version": 2,
                 "column_types": {"target_col": "int64"},
                 "id_maps": {"target_col": {"A": 3}},
                 "selected_columns_statistics": {},
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index f6c13e17..bbafeda5 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -218,6 +218,10 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
                 "selected_columns_statistics": {
                     "itemValue": {"mean": 60.0, "std": 10.0}
                 },
+                "context_length": 2,
+                "max_lookahead": 1,
+                "sample_length": 3,
+                "sequence_layout_version": 2,
                 "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
             }
         )
@@ -656,15 +660,3 @@ def test_load_precomputed_id_maps_allows_reserved_entries(tmp_path):
 
     assert id_maps["itemId"]["[mask]"] == 2
     assert id_maps["itemId"]["a"] == 3
-
-
-def test_load_precomputed_id_maps_migrates_legacy_user_ids(tmp_path):
-    project_root = tmp_path / "project"
-    id_map_dir = project_root / "configs" / "id_maps"
-    id_map_dir.mkdir(parents=True)
-    (id_map_dir / "itemId.json").write_text(json.dumps({"a": 2, "b": 3}))
-
-    with pytest.warns(UserWarning, match="legacy user IDs"):
-        id_maps = load_precomputed_id_maps(str(project_root), ["itemId"])
-
-    assert id_maps["itemId"] == {"a": 3, "b": 4}
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 875e9109..ea5e3863 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -255,7 +255,7 @@ def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
                 "context_length": config_values["context_length"],
                 "max_lookahead": 1,
                 "sample_length": config_values["context_length"] + 1,
-                "sequence_layout_version": 1,
+                "sequence_layout_version": 2,
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
@@ -288,7 +288,7 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
                 "context_length": config_values["context_length"],
                 "max_lookahead": 1,
                 "sample_length": config_values["context_length"] + 1,
-                "sequence_layout_version": 1,
+                "sequence_layout_version": 2,
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],

From ead716e3213ac8749840d891b828a2907da3f6cc Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 23:31:28 +0200
Subject: [PATCH 46/81] Make sequence_layout_version required

---
 .../config/hyperparameter_search_config.py    |  5 ++
 src/sequifier/config/infer_config.py          |  2 +-
 src/sequifier/config/train_config.py          |  2 +-
 src/sequifier/helpers.py                      |  4 +-
 src/sequifier/io/batch.py                     | 18 ++++---
 src/sequifier/preprocess.py                   | 28 +++++++----
 src/sequifier/train.py                        | 47 +++++++++----------
 .../integration/test_hyperparameter_search.py |  1 +
 tests/integration/test_onnx_export.py         |  2 +
 tests/unit/io/test_batch.py                   | 32 +++++++++++++
 ...uifier_dataset_from_folder_parquet_lazy.py | 11 +++--
 ...t_sequifier_dataset_from_folder_pt_lazy.py | 10 ++--
 tests/unit/test_infer.py                      |  5 ++
 tests/unit/test_preprocess.py                 |  4 +-
 tests/unit/test_train.py                      |  1 +
 15 files changed, 118 insertions(+), 54 deletions(-)
 create mode 100644 tests/unit/io/test_batch.py

diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 6e0282c0..8f1c5145 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -18,6 +18,7 @@
     TrainModel,
 )
 from sequifier.helpers import normalize_path, try_catch_excess_keys
+from sequifier.preprocess import CURRENT_SEQUENCE_LAYOUT_VERSION
 from sequifier.special_tokens import validate_special_token_ids
 
 
@@ -201,6 +202,9 @@ def load_hyperparameter_search_config(
         config_values["n_classes"] = config_values.get(
             "n_classes", metadata_config["n_classes"]
         )
+        config_values["sequence_layout_version"] = int(
+            metadata_config["sequence_layout_version"]
+        )
         config_values["training_data_path"] = normalize_path(
             config_values.get("training_data_path", metadata_config["split_paths"][0]),
             config_values["project_root"],
@@ -854,6 +858,7 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             context_length=context_length,
             max_lookahead=self.max_lookahead,
             sample_length=sample_length,
+            sequence_layout_version=CURRENT_SEQUENCE_LAYOUT_VERSION,
             n_classes=self.n_classes,
             inference_batch_size=self.inference_batch_size,
             seed=101,
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 99f1be90..174cdc10 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -157,7 +157,7 @@ class InfererModel(BaseModel):
     context_length: int
     max_lookahead: int = Field(default=1, ge=0)
     sample_length: int
-    sequence_layout_version: int = 1
+    sequence_layout_version: int
     prediction_length: Optional[int] = None
     inference_batch_size: int
 
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index f3621360..e1c55b27 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -593,7 +593,7 @@ class TrainModel(BaseModel):
     context_length: int
     max_lookahead: int = Field(default=1, ge=0)
     sample_length: int
-    sequence_layout_version: int = 1
+    sequence_layout_version: int
     n_classes: dict[str, int]
     inference_batch_size: int
     seed: int
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index a236b3b5..6d428721 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -52,8 +52,8 @@
 @dataclass(frozen=True)
 class SequenceLayout:
     context_length: int
-    max_lookahead: int = 1
-    sequence_layout_version: int = 2
+    max_lookahead: int
+    sequence_layout_version: int
 
     def __post_init__(self) -> None:
         if self.context_length < 1:
diff --git a/src/sequifier/io/batch.py b/src/sequifier/io/batch.py
index 36906103..597a4db7 100644
--- a/src/sequifier/io/batch.py
+++ b/src/sequifier/io/batch.py
@@ -1,8 +1,10 @@
 from dataclasses import dataclass
-from typing import Iterator, Optional
+from typing import Optional
 
 import torch
 
+REQUIRED_METADATA_KEYS = frozenset({"attention_valid_mask", "target_valid_mask"})
+
 
 @dataclass(frozen=True)
 class SequifierBatch:
@@ -12,9 +14,11 @@ class SequifierBatch:
     sequence_ids: Optional[torch.Tensor] = None
     subsequence_ids: Optional[torch.Tensor] = None
 
-    def __iter__(self) -> Iterator[object]:
-        yield self.inputs
-        yield self.targets
-        yield self.metadata
-        yield self.sequence_ids
-        yield self.subsequence_ids
+    def __post_init__(self) -> None:
+        metadata_keys = self.metadata.keys() if self.metadata is not None else set()
+        missing_keys = REQUIRED_METADATA_KEYS - metadata_keys
+        if missing_keys:
+            missing = ", ".join(sorted(missing_keys))
+            raise ValueError(
+                f"SequifierBatch metadata is missing required keys: {missing}"
+            )
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 9bcb4253..0e65b0a7 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -31,6 +31,7 @@
 
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
 REAL_MASK_VALUE = 0.0
+CURRENT_SEQUENCE_LAYOUT_VERSION = 2
 
 
 @beartype
@@ -144,7 +145,11 @@ def __init__(
         np.random.seed(seed)
         self.n_cores = n_cores or multiprocessing.cpu_count()
         self.continue_preprocessing = continue_preprocessing
-        self.layout = SequenceLayout(context_length, max_lookahead)
+        self.layout = SequenceLayout(
+            context_length,
+            max_lookahead,
+            sequence_layout_version=CURRENT_SEQUENCE_LAYOUT_VERSION,
+        )
         self._setup_directories()
 
         if selected_columns is not None:
@@ -1020,20 +1025,23 @@ def _selected_columns_with_optional_mask(
 
 
 @beartype
-def _mask_column_expr(mask_dtype: Any, mask_column: str) -> pl.Expr:
+def _validate_and_create_mask_column_expr(
+    series: Any, mask_dtype: Any, mask_column: str
+) -> pl.Expr:
     mask_col = pl.col(mask_column)
 
     if mask_dtype == pl.Boolean:
         return mask_col
 
     if mask_dtype.is_numeric():
+        if not series.drop_nulls().is_in([0, 1]).all():
+            raise ValueError(
+                f"Mask column {mask_column} contains inadmissible values not in (0, 1)"
+            )
         return mask_col == 1
 
-    if mask_dtype in (pl.String, pl.Utf8):
-        return mask_col.str.to_lowercase().is_in(["1"])
-
     raise ValueError(
-        f"Column {mask_column} must be boolean, numeric, or string, got {mask_dtype}"
+        f"Column {mask_column} must be boolean or numeric, got {mask_dtype}"
     )
 
 
@@ -1049,7 +1057,9 @@ def _apply_mask_column(
     if mask_column not in data.columns:
         raise ValueError(f"mask_column '{mask_column}' not found in input data")
 
-    mask_expr = _mask_column_expr(data.schema[mask_column], mask_column)
+    mask_expr = _validate_and_create_mask_column_expr(
+        data[mask_column], data.schema[mask_column], mask_column
+    )
     updates = []
     for col in data_columns:
         mask_value = (
@@ -1273,7 +1283,9 @@ def _get_column_statistics(
             string, integer, nor float).
     """
     if mask_column is not None and mask_column in data.columns:
-        mask_expr = _mask_column_expr(data.schema[mask_column], mask_column)
+        mask_expr = _validate_and_create_mask_column_expr(
+            data[mask_column], data.schema[mask_column], mask_column
+        )
         data = data.filter(~mask_expr)
 
     if data.is_empty():
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 794a896a..efa78380 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -84,20 +84,6 @@
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS  # noqa: E402
 
 
-def _as_sequifier_batch(batch: Any) -> SequifierBatch:
-    if isinstance(batch, SequifierBatch):
-        return batch
-
-    data, targets, metadata, sequence_ids, subsequence_ids = batch
-    return SequifierBatch(
-        inputs=data,
-        targets=targets,
-        metadata=metadata,
-        sequence_ids=sequence_ids,
-        subsequence_ids=subsequence_ids,
-    )
-
-
 def cleanup():
     """Cleans up the distributed training environment."""
     dist.destroy_process_group()
@@ -1538,8 +1524,12 @@ def _train_epoch(
 
         model_to_call.train()
 
-        for batch_count, raw_batch in enumerate(train_loader):
-            batch = _as_sequifier_batch(raw_batch)
+        for batch_count, batch in enumerate(train_loader):
+            if not isinstance(batch, SequifierBatch):
+                raise TypeError(
+                    "Training DataLoader must yield SequifierBatch objects, "
+                    f"got {type(batch).__name__}."
+                )
             if batch_count >= start_batch:
                 data = batch.inputs
                 targets = batch.targets
@@ -1555,8 +1545,7 @@ def _train_epoch(
                     if k in self.target_column_types
                 }
                 metadata = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in (metadata or {}).items()
+                    k: v.to(self.device, non_blocking=True) for k, v in metadata.items()
                 }
                 if self.hparams.training_spec.training_objective == "bert":
                     data, targets, metadata = apply_bert_masking(
@@ -1872,8 +1861,12 @@ def _evaluate(
         model_to_call.eval()
 
         with torch.no_grad():
-            for batch_idx, raw_batch in enumerate(valid_loader):
-                batch = _as_sequifier_batch(raw_batch)
+            for batch_idx, batch in enumerate(valid_loader):
+                if not isinstance(batch, SequifierBatch):
+                    raise TypeError(
+                        "Validation DataLoader must yield SequifierBatch objects, "
+                        f"got {type(batch).__name__}."
+                    )
                 data = batch.inputs
                 targets = batch.targets
                 metadata = batch.metadata
@@ -1889,8 +1882,7 @@ def _evaluate(
                     if k in self.target_column_types
                 }
                 metadata = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in (metadata or {}).items()
+                    k: v.to(self.device, non_blocking=True) for k, v in metadata.items()
                 }
                 if self.hparams.training_spec.training_objective == "bert":
                     data, targets, metadata = apply_bert_masking(
@@ -1980,8 +1972,12 @@ def _evaluate(
             baseline_losses_local_collect = {col: [] for col in self.target_columns}
 
             # Iterate over the sharded validation loader
-            for batch_idx, raw_batch in enumerate(valid_loader):
-                batch = _as_sequifier_batch(raw_batch)
+            for batch_idx, batch in enumerate(valid_loader):
+                if not isinstance(batch, SequifierBatch):
+                    raise TypeError(
+                        "Validation DataLoader must yield SequifierBatch objects, "
+                        f"got {type(batch).__name__}."
+                    )
                 data = batch.inputs
                 targets = batch.targets
                 metadata = batch.metadata
@@ -1996,8 +1992,7 @@ def _evaluate(
                     if k in self.target_column_types
                 }
                 metadata = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in (metadata or {}).items()
+                    k: v.to(self.device, non_blocking=True) for k, v in metadata.items()
                 }
 
                 if self.hparams.training_spec.training_objective == "bert":
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index 6cbfbbbf..0869ac35 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -41,6 +41,7 @@ def test_hp_search_bert_outputs(run_hp_search, project_root):
         "test-data-categorical-1-lookahead-0.json"
     )
     assert generated_config["max_lookahead"] == 0
+    assert generated_config["sequence_layout_version"] == 2
     assert generated_config["sample_length"] == generated_config["context_length"]
 
 
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index cb663dc2..25e8639d 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -29,6 +29,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         n_classes={"cat_col": 6},
         context_length=context_length,
         sample_length=5,
+        sequence_layout_version=2,
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -129,6 +130,7 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
         n_classes={"cat2": 7, "cat10": 6},
         context_length=context_length,
         sample_length=5,
+        sequence_layout_version=2,
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
diff --git a/tests/unit/io/test_batch.py b/tests/unit/io/test_batch.py
new file mode 100644
index 00000000..e8b24ff9
--- /dev/null
+++ b/tests/unit/io/test_batch.py
@@ -0,0 +1,32 @@
+import pytest
+import torch
+
+from sequifier.io.batch import SequifierBatch
+
+
+def _metadata():
+    valid_mask = torch.ones(1, 2, dtype=torch.bool)
+    return {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+
+def test_sequifier_batch_requires_metadata_masks():
+    with pytest.raises(ValueError, match="target_valid_mask"):
+        SequifierBatch(
+            inputs={"item": torch.ones(1, 2)},
+            targets={"item": torch.ones(1, 2)},
+            metadata={"attention_valid_mask": torch.ones(1, 2, dtype=torch.bool)},
+        )
+
+
+def test_sequifier_batch_does_not_support_tuple_unpacking():
+    batch = SequifierBatch(
+        inputs={"item": torch.ones(1, 2)},
+        targets={"item": torch.ones(1, 2)},
+        metadata=_metadata(),
+    )
+
+    with pytest.raises(TypeError):
+        tuple(batch)  # type: ignore
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 24646c20..f4696079 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -101,7 +101,9 @@ def test_iteration_yields_correct_batches(mock_config, dataset_path):
     assert len(batches) == 8  # 40 samples / batch_size 5
 
     # Check structural integrity of first batch
-    seq_batch, tgt_batch, _, _, _ = batches[0]
+    batch = batches[0]
+    seq_batch = batch.inputs
+    tgt_batch = batch.targets
     assert "item" in seq_batch
     assert "item" in tgt_batch
     assert isinstance(seq_batch["item"], torch.Tensor)
@@ -142,7 +144,8 @@ def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
     dataset = SequifierDatasetFromFolderParquetLazy(
         str(data_dir), mock_config, shuffle=False
     )
-    seq_batch, tgt_batch, metadata_batch, _, _ = next(iter(dataset))
+    batch = next(iter(dataset))
+    metadata_batch = batch.metadata
 
     assert torch.equal(
         metadata_batch["attention_valid_mask"],
@@ -185,8 +188,8 @@ def test_distributed_sharding(mock_ws, mock_rank, mock_init, mock_config, datase
     assert len(batches) == 4
 
     # Verify input mapping structures
-    for seq_batch, _, _, _, _ in batches:
-        assert seq_batch["item"].shape[0] == 5
+    for batch in batches:
+        assert batch.inputs["item"].shape[0] == 5
 
 
 def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 879d855f..4fb599cc 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -74,13 +74,14 @@ def mock_torch_load():
 
         def side_effect(path, map_location, weights_only):
             dummy_seq = {"col1": torch.ones((10, 6)), "tgt1": torch.zeros((10, 6))}
+            left_pad_lengths = torch.zeros(10, dtype=torch.int64)
 
             return (
                 dummy_seq,
                 None,
                 None,
                 None,
-                None,
+                left_pad_lengths,
             )
 
         mock_load.side_effect = side_effect
@@ -114,7 +115,9 @@ def test_iteration_yields_correct_batches(mock_config, dataset_path, mock_torch_
     assert mock_torch_load.call_count == 4
 
     # Verify the structure of a yielded batch
-    seq_dict, tgt_dict, _, _, _ = batches[0]
+    batch = batches[0]
+    seq_dict = batch.inputs
+    tgt_dict = batch.targets
 
     assert "col1" in seq_dict, f"{seq_dict = }"
 
@@ -145,7 +148,8 @@ def side_effect(path, map_location, weights_only):
         dataset = SequifierDatasetFromFolderPtLazy(
             dataset_path, mock_config, shuffle=False
         )
-        seq_dict, tgt_dict, metadata_dict, _, _ = next(iter(dataset))
+        batch = next(iter(dataset))
+        metadata_dict = batch.metadata
 
     assert torch.equal(
         metadata_dict["attention_valid_mask"],
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 2d16fc81..321446a5 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -175,6 +175,7 @@ def test_infer_config_defaults_bert_prediction_length_to_context_length():
         prediction_length=None,
         context_length=3,
         sample_length=4,
+        sequence_layout_version=2,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -203,6 +204,7 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
         prediction_length=None,
         context_length=3,
         sample_length=4,
+        sequence_layout_version=2,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -232,6 +234,7 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             prediction_length=1,
             context_length=3,
             sample_length=4,
+            sequence_layout_version=2,
             inference_batch_size=2,
             output_probabilities=False,
             map_to_id=False,
@@ -397,6 +400,7 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         prediction_length=None,
         context_length=3,
         sample_length=4,
+        sequence_layout_version=2,
         inference_batch_size=4,
         output_probabilities=False,
         map_to_id=True,
@@ -613,6 +617,7 @@ def ar_config():
         prediction_length=1,
         context_length=3,
         sample_length=4,
+        sequence_layout_version=2,
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,  # Set to False to bypass ID mapping requirements
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index bbafeda5..a90ae1c8 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -127,7 +127,7 @@ def test_extract_sequences_persists_left_pad_length_metadata():
     sequences = extract_sequences(
         data,
         schema,
-        layout=SequenceLayout(context_length=4),
+        layout=SequenceLayout(4, 1, 2),
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -186,7 +186,7 @@ def test_apply_mask_column_replaces_values_and_drops_column():
         {
             "cat_col": [3, 4, 5, 6],
             "num_col": [-1.0, 2.5, 3.5, 4.5],
-            RESERVED_MASK_COLUMN: ["0", "1", "0", "1"],
+            RESERVED_MASK_COLUMN: [0, 1, 0, 1],
         }
     )
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index ea5e3863..bcc8b00e 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -138,6 +138,7 @@ def model_config(tmp_path):
         n_classes={"cat_col": 5},  # 0 + 4 classes
         context_length=10,
         sample_length=11,
+        sequence_layout_version=2,
         inference_batch_size=4,
         seed=42,
         export_generative_model=True,

From ac1bd3d0d40266d8b3ce1be2c92e4ab20018304f Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Fri, 12 Jun 2026 23:57:22 +0200
Subject: [PATCH 47/81] Small changes

---
 src/sequifier/config/hyperparameter_search_config.py   |  2 +-
 src/sequifier/config/infer_config.py                   |  6 ++----
 src/sequifier/config/train_config.py                   |  6 ++----
 src/sequifier/helpers.py                               | 10 +---------
 src/sequifier/infer.py                                 |  2 +-
 .../io/sequifier_dataset_from_folder_parquet.py        |  2 +-
 .../io/sequifier_dataset_from_folder_parquet_lazy.py   |  2 +-
 src/sequifier/io/sequifier_dataset_from_folder_pt.py   |  2 +-
 .../io/sequifier_dataset_from_folder_pt_lazy.py        |  2 +-
 src/sequifier/preprocess.py                            |  4 ++--
 src/sequifier/special_tokens.py                        |  4 ++--
 11 files changed, 15 insertions(+), 27 deletions(-)

diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 8f1c5145..411f6116 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -167,7 +167,7 @@ def load_hyperparameter_search_config(
             metadata_config = json.loads(f.read())
 
         validate_special_token_ids(
-            metadata_config.get("special_token_ids"),
+            metadata_config["special_token_ids"],
             source=f"metadata config '{metadata_config_path}'",
         )
 
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 174cdc10..7b7e8c24 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -52,12 +52,10 @@ def load_inferer_config(
             metadata_config = json.load(f)
 
         validate_special_token_ids(
-            metadata_config.get("special_token_ids"),
+            metadata_config["special_token_ids"],
             source=f"metadata config '{metadata_config_path}'",
         )
-        sequence_layout = sequence_layout_from_metadata(
-            metadata_config, config_values["context_length"]
-        )
+        sequence_layout = sequence_layout_from_metadata(metadata_config)
         config_values["max_lookahead"] = sequence_layout.max_lookahead
         config_values["sample_length"] = sequence_layout.sample_length
         config_values["sequence_layout_version"] = (
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index e1c55b27..d079002d 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -62,9 +62,7 @@ def load_train_config(
         ) as f:
             metadata_config = json.loads(f.read())
 
-        sequence_layout = sequence_layout_from_metadata(
-            metadata_config, config_values["context_length"]
-        )
+        sequence_layout = sequence_layout_from_metadata(metadata_config)
         config_values["max_lookahead"] = sequence_layout.max_lookahead
         config_values["sample_length"] = sequence_layout.sample_length
         config_values["sequence_layout_version"] = (
@@ -115,7 +113,7 @@ def load_train_config(
 
         config_values["id_maps"] = metadata_config["id_maps"]
         config_values["special_token_ids"] = validate_special_token_ids(
-            metadata_config.get("special_token_ids"),
+            metadata_config["special_token_ids"],
             source=f"metadata config '{metadata_config_path}'",
         )
 
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 6d428721..792bbdc9 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -42,12 +42,6 @@
     "uint8": torch.int16,
 }
 
-EXPLICIT_PADDING_MASK_FALLBACK_WARNING = (
-    "Explicit padding mask not found. Falling back to value-based padding "
-    "inference for real-valued data; leading 0.0 values may be treated as "
-    "padding. Re-run preprocessing to generate explicit masks."
-)
-
 
 @dataclass(frozen=True)
 class SequenceLayout:
@@ -111,9 +105,7 @@ def validate_stored_window_width(tensor: Tensor, sample_length: int) -> None:
 
 
 @beartype
-def sequence_layout_from_metadata(
-    metadata: dict, context_length: int
-) -> SequenceLayout:
+def sequence_layout_from_metadata(metadata: dict) -> SequenceLayout:
     metadata_context_length = int(metadata["context_length"])
 
     max_lookahead = int(metadata["max_lookahead"])
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 8abdef3d..eae7cb56 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -69,7 +69,7 @@ def infer(args: Any, args_config: dict[str, Any]) -> None:
         ) as f:
             metadata_config = json.loads(f.read())
             validate_special_token_ids(
-                metadata_config.get("special_token_ids"),
+                metadata_config["special_token_ids"],
                 source=f"metadata config '{config.metadata_config_path}'",
             )
             id_maps = metadata_config["id_maps"]
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index dc1965b7..27dc8caa 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -49,7 +49,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        folder_layout = sequence_layout_from_metadata(metadata)
         if folder_layout.sample_length != config.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index f716ae1c..3628903f 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -65,7 +65,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        folder_layout = sequence_layout_from_metadata(metadata)
         if folder_layout.sample_length != config.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index de193564..17195b4f 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -47,7 +47,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        folder_layout = sequence_layout_from_metadata(metadata)
         if folder_layout.sample_length != config.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index cc629186..7b36b912 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -63,7 +63,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata, config.context_length)
+        folder_layout = sequence_layout_from_metadata(metadata)
         if folder_layout.sample_length != config.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 0e65b0a7..8232688a 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -202,7 +202,7 @@ def __init__(
                     preexisting_metadata = json.load(f)
 
                 validate_special_token_ids(
-                    preexisting_metadata.get("special_token_ids"),
+                    preexisting_metadata["special_token_ids"],
                     source=f"metadata config '{self.metadata_config_path}'",
                 )
                 id_maps = preexisting_metadata["id_maps"]
@@ -302,7 +302,7 @@ def __init__(
                     preexisting_metadata = json.load(f)
 
                 validate_special_token_ids(
-                    preexisting_metadata.get("special_token_ids"),
+                    preexisting_metadata["special_token_ids"],
                     source=f"metadata config '{self.metadata_config_path}'",
                 )
                 id_maps = preexisting_metadata["id_maps"]
diff --git a/src/sequifier/special_tokens.py b/src/sequifier/special_tokens.py
index c92e23d8..d0224d1f 100644
--- a/src/sequifier/special_tokens.py
+++ b/src/sequifier/special_tokens.py
@@ -32,12 +32,12 @@ def ids_by_label(self) -> dict[str, int]:
 
 
 def validate_special_token_ids(
-    special_token_ids: Mapping[str, Any] | None,
+    special_token_ids: Mapping[str, Any],
     source: str = "metadata",
 ) -> dict[str, int]:
     """Validate persisted special token IDs against runtime constants."""
     expected = SPECIAL_TOKEN_IDS.ids_by_label
-    assert special_token_ids is not None
+
     try:
         normalized = {str(label): int(id_) for label, id_ in special_token_ids.items()}
     except (AttributeError, TypeError, ValueError) as exc:

From 20fbb297eb4eebfd060c851c18c56229badfa7d2 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 00:00:43 +0200
Subject: [PATCH 48/81] Fail on wrong sequence_layout_version

---
 src/sequifier/config/infer_config.py | 5 +++++
 src/sequifier/config/train_config.py | 5 +++++
 2 files changed, 10 insertions(+)

diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 7b7e8c24..230fbd73 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -56,6 +56,11 @@ def load_inferer_config(
             source=f"metadata config '{metadata_config_path}'",
         )
         sequence_layout = sequence_layout_from_metadata(metadata_config)
+        if sequence_layout.sequence_layout_version != 2:
+            raise ValueError(
+                "Inference requires metadata sequence_layout_version=2, "
+                f"got {sequence_layout.sequence_layout_version}."
+            )
         config_values["max_lookahead"] = sequence_layout.max_lookahead
         config_values["sample_length"] = sequence_layout.sample_length
         config_values["sequence_layout_version"] = (
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index d079002d..a01e7635 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -63,6 +63,11 @@ def load_train_config(
             metadata_config = json.loads(f.read())
 
         sequence_layout = sequence_layout_from_metadata(metadata_config)
+        if sequence_layout.sequence_layout_version != 2:
+            raise ValueError(
+                "Training requires metadata sequence_layout_version=2, "
+                f"got {sequence_layout.sequence_layout_version}."
+            )
         config_values["max_lookahead"] = sequence_layout.max_lookahead
         config_values["sample_length"] = sequence_layout.sample_length
         config_values["sequence_layout_version"] = (

From 5b57073dac12adbb47ef409a3b274a888b1e9a71 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 10:51:09 +0200
Subject: [PATCH 49/81] Remove vestiges of v1 processing

---
 .../config/hyperparameter_search_config.py    | 37 ++++++++++++++++---
 src/sequifier/helpers.py                      | 23 ++++++------
 src/sequifier/infer.py                        | 33 ++++++++---------
 .../sequifier_dataset_from_folder_parquet.py  |  7 +---
 ...uifier_dataset_from_folder_parquet_lazy.py | 16 ++++----
 .../io/sequifier_dataset_from_folder_pt.py    | 10 +----
 .../sequifier_dataset_from_folder_pt_lazy.py  | 16 ++++----
 7 files changed, 76 insertions(+), 66 deletions(-)

diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 411f6116..d0496cac 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -17,8 +17,11 @@
     TrainingSpecModel,
     TrainModel,
 )
-from sequifier.helpers import normalize_path, try_catch_excess_keys
-from sequifier.preprocess import CURRENT_SEQUENCE_LAYOUT_VERSION
+from sequifier.helpers import (
+    normalize_path,
+    sequence_layout_from_metadata,
+    try_catch_excess_keys,
+)
 from sequifier.special_tokens import validate_special_token_ids
 
 
@@ -202,9 +205,20 @@ def load_hyperparameter_search_config(
         config_values["n_classes"] = config_values.get(
             "n_classes", metadata_config["n_classes"]
         )
-        config_values["sequence_layout_version"] = int(
-            metadata_config["sequence_layout_version"]
+
+        sequence_layout = sequence_layout_from_metadata(metadata_config)
+        if sequence_layout.sequence_layout_version != 2:
+            raise ValueError(
+                "Hyperparameter search requires metadata sequence_layout_version=2, "
+                f"got {sequence_layout.sequence_layout_version}."
+            )
+
+        config_values["max_lookahead"] = sequence_layout.max_lookahead
+        config_values["sample_length"] = sequence_layout.sample_length
+        config_values["sequence_layout_version"] = (
+            sequence_layout.sequence_layout_version
         )
+
         config_values["training_data_path"] = normalize_path(
             config_values.get("training_data_path", metadata_config["split_paths"][0]),
             config_values["project_root"],
@@ -708,6 +722,8 @@ class HyperparameterSearchConfig(BaseModel):
 
     context_length: list[int]
     max_lookahead: int = Field(default=1, ge=0)
+    sample_length: int
+    sequence_layout_version: int
     n_classes: dict[str, int]
     inference_batch_size: int
 
@@ -727,6 +743,17 @@ class HyperparameterSearchConfig(BaseModel):
 
     override_input: bool = False
 
+    @model_validator(mode="after")
+    def validate_sequence_layout(self):
+        for cl in self.context_length:
+            if cl + self.max_lookahead > self.sample_length:
+                raise ValueError(
+                    f"Sample length mismatch: context_length ({cl}) + max_lookahead "
+                    f"({self.max_lookahead}) > stored sample_length ({self.sample_length}). "
+                    "Model inputs cannot exceed the preprocessed sequence length."
+                )
+        return self
+
     @model_validator(mode="after")
     def validate_prune_trials(self):
         if self.prune_trials and self.training_hyperparameter_sampling.distributed:
@@ -858,7 +885,7 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             context_length=context_length,
             max_lookahead=self.max_lookahead,
             sample_length=sample_length,
-            sequence_layout_version=CURRENT_SEQUENCE_LAYOUT_VERSION,
+            sequence_layout_version=self.sequence_layout_version,
             n_classes=self.n_classes,
             inference_batch_size=self.inference_batch_size,
             seed=101,
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 792bbdc9..c9bf4bce 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -392,16 +392,13 @@ def numpy_to_pytorch(
         unified_tensors[f"{col_name}_target"] = target_tensor
 
     left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
-    if left_pad_lengths is not None:
-        metadata = generate_padding_masks(
-            left_pad_lengths,
-            context_length,
-            sample_length,
-            data_offset,
-            target_offset,
-        )
-    else:
-        metadata = {}
+    metadata = generate_padding_masks(
+        left_pad_lengths,
+        context_length,
+        sample_length,
+        data_offset,
+        target_offset,
+    )
 
     return unified_tensors, metadata
 
@@ -426,7 +423,11 @@ def build_valid_mask(
 @beartype
 def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Tensor:
     """Extracts one leftPadLength value per long-format subsequence."""
-
+    if "leftPadLength" not in data.columns:
+        raise ValueError(
+            "Dataset layout v1 does not contain explicit padding metadata. "
+            "Please re-run preprocessing."
+        )
     assert {"sequenceId", "subsequenceId"}.issubset(data.columns)
 
     lengths = (
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index eae7cb56..b861dabd 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -496,15 +496,14 @@ def infer_embedding(
             ) = data
             for tensor in sequences_dict.values():
                 validate_stored_window_width(tensor, config.sample_length)
-            metadata = {}
-            if left_pad_lengths_tensor is not None:
-                metadata = generate_padding_masks(
-                    left_pad_lengths_tensor,
-                    config.context_length,
-                    config.sample_length,
-                    data_offset=config.data_offset,
-                    target_offset=config.target_offset,
-                )
+
+            metadata = generate_padding_masks(
+                left_pad_lengths_tensor,
+                config.context_length,
+                config.sample_length,
+                data_offset=config.data_offset,
+                target_offset=config.target_offset,
+            )
             embeddings = get_embeddings_pt(
                 config, inferer, sequences_dict, metadata=metadata
             )
@@ -757,15 +756,13 @@ def infer_generative(
                 if key in config.input_columns
             }
 
-            metadata = {}
-            if left_pad_lengths_tensor is not None:
-                metadata = generate_padding_masks(
-                    left_pad_lengths_tensor,
-                    config.context_length,
-                    config.sample_length,
-                    data_offset=config.data_offset,
-                    target_offset=config.target_offset,
-                )
+            metadata = generate_padding_masks(
+                left_pad_lengths_tensor,
+                config.context_length,
+                config.sample_length,
+                data_offset=config.data_offset,
+                target_offset=config.target_offset,
+            )
 
             probs, preds = get_probs_preds_from_dict(
                 config, inferer, sequences_dict, metadata, total_steps
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 27dc8caa..373e0512 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -122,12 +122,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col, tensors in all_targets.items()
             if tensors
         }
-        self.left_pad_lengths = (
-            torch.cat(all_left_pad_lengths)
-            if all_left_pad_lengths
-            and len(all_left_pad_lengths) == len(metadata["batch_files"])
-            else None
-        )
+        self.left_pad_lengths = torch.cat(all_left_pad_lengths)
 
         # Step 3: Prevent serialization duplications across worker forks via shared memory flags
         for tensor in self.sequences.values():
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 3628903f..0546c1d7 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -310,15 +310,13 @@ def __iter__(
                         f"Missing required column {col_name} in Parquet partition"
                     )
 
-            new_meta = {}
-            if left_pad_lengths is not None:
-                new_meta = generate_padding_masks(
-                    left_pad_lengths[worker_indices],
-                    train_seq_len,
-                    self.config.sample_length,
-                    self.config.training_spec.data_offset,
-                    self.config.training_spec.target_offset,
-                )
+            new_meta = generate_padding_masks(
+                left_pad_lengths[worker_indices],
+                train_seq_len,
+                self.config.sample_length,
+                self.config.training_spec.data_offset,
+                self.config.training_spec.target_offset,
+            )
 
             del df
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 17195b4f..9605020e 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -81,18 +81,12 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
                         sequences_batch[col], config.sample_length
                     )
                     all_sequences[col].append(sequences_batch[col])
-            if left_pad_lengths_batch is not None:
-                all_left_pad_lengths.append(left_pad_lengths_batch)
+            all_left_pad_lengths.append(left_pad_lengths_batch)
 
         self.sequences: Dict[str, torch.Tensor] = {
             col: torch.cat(tensors) for col, tensors in all_sequences.items() if tensors
         }
-        self.left_pad_lengths = (
-            torch.cat(all_left_pad_lengths)
-            if all_left_pad_lengths
-            and len(all_left_pad_lengths) == len(metadata["batch_files"])
-            else None
-        )
+        self.left_pad_lengths = torch.cat(all_left_pad_lengths)
         for tensor in self.sequences.values():
             tensor.share_memory_()
         if self.left_pad_lengths is not None:
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 7b36b912..c3b787d2 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -278,15 +278,13 @@ def __iter__(
                 if k in self.config.target_columns
             }
 
-            new_meta = {}
-            if left_pad_lengths_batch is not None:
-                new_meta = generate_padding_masks(
-                    left_pad_lengths_batch[worker_indices],
-                    train_seq_len,
-                    self.config.sample_length,
-                    data_offset,
-                    target_offset,
-                )
+            new_meta = generate_padding_masks(
+                left_pad_lengths_batch[worker_indices],
+                train_seq_len,
+                self.config.sample_length,
+                data_offset,
+                target_offset,
+            )
 
             # Free the large file immediately to keep RAM down
             del sequences_batch, left_pad_lengths_batch

From 30129880b69fc7f7ad65d07ca5e861918b38a9e5 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 12:26:07 +0200
Subject: [PATCH 50/81] Access sequence layout directly

---
 .vscode/settings.json                         |    4 +
 .../config/hyperparameter_search_config.py    |   20 +-
 src/sequifier/config/infer_config.py          |   46 +-
 src/sequifier/config/train_config.py          |   56 +-
 src/sequifier/helpers.py                      |   19 +-
 src/sequifier/infer.py                        |   93 +-
 .../io/sequifier_dataset_from_file.py         |   10 +-
 .../sequifier_dataset_from_folder_parquet.py  |   23 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |   21 +-
 .../io/sequifier_dataset_from_folder_pt.py    |   16 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |   16 +-
 src/sequifier/io/yaml.py                      |    2 +
 src/sequifier/train.py                        |    9 +-
 .../integration/test_hyperparameter_search.py |    5 +-
 tests/integration/test_onnx_export.py         |   19 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |   11 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |   11 +-
 .../unit/test_hyperparameter_search_config.py |    3 +-
 tests/unit/test_infer.py                      |   33 +-
 tests/unit/test_train.py                      |   48 +-
 uv.lock                                       | 2269 +++++++++++++++++
 21 files changed, 2509 insertions(+), 225 deletions(-)
 create mode 100644 .vscode/settings.json
 create mode 100644 uv.lock

diff --git a/.vscode/settings.json b/.vscode/settings.json
new file mode 100644
index 00000000..c820b9a3
--- /dev/null
+++ b/.vscode/settings.json
@@ -0,0 +1,4 @@
+{
+    "python-envs.defaultEnvManager": "ms-python.python:conda",
+    "python-envs.defaultPackageManager": "ms-python.python:conda"
+}
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index d0496cac..c2b6c5ff 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -18,6 +18,7 @@
     TrainModel,
 )
 from sequifier.helpers import (
+    SequenceLayout,
     normalize_path,
     sequence_layout_from_metadata,
     try_catch_excess_keys,
@@ -433,9 +434,7 @@ def validate_scheduler_config(cls, v, info_dict):
                     )
         return v
 
-    def sample_trial(
-        self, trial: Any, max_lookahead: int, sample_length: int
-    ) -> TrainingSpecModel:
+    def sample_trial(self, trial: Any) -> TrainingSpecModel:
         """Samples training hyperparameters using an Optuna trial.
 
         This method leverages the provided Optuna trial to suggest values for
@@ -518,8 +517,6 @@ def sample_trial(
             fsdp_cpu_offload=self.fsdp_cpu_offload,
             torch_compile=self.torch_compile,
             float32_matmul_precision=self.float32_matmul_precision,
-            max_lookahead=max_lookahead,
-            sample_length=sample_length,
         )
 
 
@@ -859,12 +856,12 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
         context_length = trial.suggest_categorical(
             "context_length", self.context_length
         )
-        sample_length = context_length + self.max_lookahead
-        training_spec = self.training_hyperparameter_sampling.sample_trial(
-            trial,
+        layout = SequenceLayout(
+            context_length=context_length,
             max_lookahead=self.max_lookahead,
-            sample_length=sample_length,
+            sequence_layout_version=self.sequence_layout_version,
         )
+        training_spec = self.training_hyperparameter_sampling.sample_trial(trial)
 
         logger.info(f"{input_columns_index = } - {context_length = }")
 
@@ -882,10 +879,7 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             target_columns=self.target_columns,
             target_column_types=self.target_column_types,
             id_maps=self.id_maps,
-            context_length=context_length,
-            max_lookahead=self.max_lookahead,
-            sample_length=sample_length,
-            sequence_layout_version=self.sequence_layout_version,
+            layout=layout,
             n_classes=self.n_classes,
             inference_batch_size=self.inference_batch_size,
             seed=101,
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 230fbd73..92a63b9e 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -15,6 +15,7 @@
 )
 
 from sequifier.helpers import (
+    SequenceLayout,
     normalize_path,
     sequence_layout_from_metadata,
     try_catch_excess_keys,
@@ -61,11 +62,14 @@ def load_inferer_config(
                 "Inference requires metadata sequence_layout_version=2, "
                 f"got {sequence_layout.sequence_layout_version}."
             )
-        config_values["max_lookahead"] = sequence_layout.max_lookahead
-        config_values["sample_length"] = sequence_layout.sample_length
-        config_values["sequence_layout_version"] = (
-            sequence_layout.sequence_layout_version
-        )
+        config_values["layout"] = sequence_layout
+        for key in (
+            "context_length",
+            "max_lookahead",
+            "sample_length",
+            "sequence_layout_version",
+        ):
+            config_values.pop(key, None)
 
         config_values["column_types"] = config_values.get(
             "column_types", metadata_config["column_types"]
@@ -157,10 +161,7 @@ class InfererModel(BaseModel):
     map_to_id: bool = Field(default=True)
     seed: int
     device: str
-    context_length: int
-    max_lookahead: int = Field(default=1, ge=0)
-    sample_length: int
-    sequence_layout_version: int
+    layout: SequenceLayout
     prediction_length: Optional[int] = None
     inference_batch_size: int
 
@@ -171,37 +172,20 @@ class InfererModel(BaseModel):
 
     @model_validator(mode="after")
     def normalize_prediction_length(self):
-        if self.sample_length != self.context_length + self.max_lookahead:
-            raise ValueError(
-                "sample_length must equal context_length + max_lookahead "
-                f"({self.sample_length} != {self.context_length} + {self.max_lookahead})."
-            )
         if self.prediction_length is None:
             self.prediction_length = (
-                self.context_length if self.training_objective == "bert" else 1
+                self.layout.context_length if self.training_objective == "bert" else 1
             )
         if self.training_objective == "bert":
-            if self.prediction_length != self.context_length:
+            if self.prediction_length != self.layout.context_length:
                 raise ValueError(
                     "For BERT inference, prediction_length must be equal to context_length "
-                    f"(got prediction_length={self.prediction_length}, context_length={self.context_length})."
+                    f"(got prediction_length={self.prediction_length}, context_length={self.layout.context_length})."
                 )
-        elif self.max_lookahead < 1:
-            raise ValueError(
-                "Causal inference requires data preprocessed with max_lookahead >= 1"
-            )
+        else:
+            self.layout.get_target_offset(self.training_objective)
         return self
 
-    @property
-    def data_offset(self) -> int:
-        return self.max_lookahead
-
-    @property
-    def target_offset(self) -> int:
-        if self.training_objective == "bert":
-            return self.max_lookahead
-        return self.max_lookahead - 1
-
     @field_validator("training_objective")
     @classmethod
     def validate_training_objective(cls, v):
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index a01e7635..eabfaa2f 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -23,6 +23,7 @@
 import sequifier
 from sequifier.config.probabilities import ProbabilityDistribution
 from sequifier.helpers import (
+    SequenceLayout,
     normalize_path,
     sequence_layout_from_metadata,
     try_catch_excess_keys,
@@ -68,14 +69,16 @@ def load_train_config(
                 "Training requires metadata sequence_layout_version=2, "
                 f"got {sequence_layout.sequence_layout_version}."
             )
-        config_values["max_lookahead"] = sequence_layout.max_lookahead
-        config_values["sample_length"] = sequence_layout.sample_length
-        config_values["sequence_layout_version"] = (
-            sequence_layout.sequence_layout_version
-        )
-        config_values.setdefault("training_spec", {})
-        config_values["training_spec"]["max_lookahead"] = sequence_layout.max_lookahead
-        config_values["training_spec"]["sample_length"] = sequence_layout.sample_length
+        config_values["layout"] = sequence_layout
+        for key in (
+            "context_length",
+            "max_lookahead",
+            "sample_length",
+            "sequence_layout_version",
+        ):
+            config_values.pop(key, None)
+        for key in ("max_lookahead", "sample_length"):
+            config_values.get("training_spec", {}).pop(key, None)
 
         split_paths = metadata_config["split_paths"]
 
@@ -246,8 +249,6 @@ class TrainingSpecModel(BaseModel):
     fsdp_cpu_offload: Optional[bool] = None
     torch_compile: str = "outer"
     float32_matmul_precision: str = "highest"
-    max_lookahead: int = Field(default=1, ge=0)
-    sample_length: int
 
     def __init__(self, **kwargs):
         super().__init__(
@@ -263,16 +264,6 @@ def __init__(self, **kwargs):
     def serialize_dotdict(self, value: DotDict) -> dict[str, Any]:
         return dict(value)
 
-    @property
-    def data_offset(self) -> int:
-        return self.max_lookahead
-
-    @property
-    def target_offset(self) -> int:
-        if self.training_objective == "bert":
-            return self.max_lookahead
-        return self.max_lookahead - 1
-
     @field_validator("layer_type_dtypes")
     @classmethod
     def validate_layer_type_dtypes(cls, v):
@@ -383,14 +374,6 @@ def validate_bert_spec_matches_objective(self):
             )
         return self
 
-    @model_validator(mode="after")
-    def validate_sequence_layout(self):
-        if self.training_objective == "causal" and self.max_lookahead < 1:
-            raise ValueError(
-                "Causal training requires data preprocessed with max_lookahead >= 1"
-            )
-        return self
-
     @field_validator("sampling_strategy")
     @classmethod
     def validate_sampling_strategy(cls, v):
@@ -593,10 +576,7 @@ class TrainModel(BaseModel):
         default_factory=lambda: SPECIAL_TOKEN_IDS.ids_by_label
     )
 
-    context_length: int
-    max_lookahead: int = Field(default=1, ge=0)
-    sample_length: int
-    sequence_layout_version: int
+    layout: SequenceLayout
     n_classes: dict[str, int]
     inference_batch_size: int
     seed: int
@@ -617,20 +597,14 @@ def validate_special_token_ids_match_runtime(cls, v):
 
     @model_validator(mode="after")
     def validate_bert_prediction_length_matches_context_length(self):
-        if self.sample_length != self.context_length + self.max_lookahead:
-            raise ValueError(
-                "sample_length must equal context_length + max_lookahead "
-                f"({self.sample_length} != {self.context_length} + {self.max_lookahead})."
-            )
-        self.training_spec.max_lookahead = self.max_lookahead
-        self.training_spec.sample_length = self.sample_length
+        self.layout.get_target_offset(self.training_spec.training_objective)
         if (
             self.training_spec.training_objective == "bert"
-            and self.model_spec.prediction_length != self.context_length
+            and self.model_spec.prediction_length != self.layout.context_length
         ):
             raise ValueError(
                 "For BERT training, model_spec.prediction_length must be equal to context_length "
-                f"(got prediction_length={self.model_spec.prediction_length}, context_length={self.context_length})."
+                f"(got prediction_length={self.model_spec.prediction_length}, context_length={self.layout.context_length})."
             )
         return self
 
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index c9bf4bce..ae7f4022 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -63,13 +63,18 @@ def sample_length(self) -> int:
     def input_offset(self) -> int:
         return self.max_lookahead
 
-    def target_offset(self, horizon: int) -> int:
-        offset = self.max_lookahead - horizon
-        if offset < 0:
-            raise ValueError(
-                f"horizon={horizon} exceeds max_lookahead={self.max_lookahead}"
-            )
-        return offset
+    def get_target_offset(self, training_objective: str) -> int:
+        if training_objective == "bert":
+            return self.max_lookahead
+        if training_objective == "causal":
+            if self.max_lookahead < 1:
+                raise ValueError(
+                    "Causal training requires data preprocessed with max_lookahead >= 1"
+                )
+            return self.max_lookahead - 1
+        raise ValueError(
+            f"Only 'causal' and 'bert' are allowed, found {training_objective}"
+        )
 
 
 @beartype
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index b861dabd..4514ab76 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -221,7 +221,7 @@ def infer_worker(
                 f"Unsupported input type or read format: {config.read_format}"
             )
 
-        default_prediction_length = {"causal": 1, "bert": config.context_length}
+        default_prediction_length = {"causal": 1, "bert": config.layout.context_length}
         prediction_length = (
             config.prediction_length
             if config.prediction_length is not None
@@ -363,10 +363,10 @@ def _bert_target_valid_mask_from_preprocessed_data(
         )
         metadata = generate_padding_masks(
             left_pad_lengths,
-            config.context_length,
-            config.sample_length,
-            data_offset=config.data_offset,
-            target_offset=config.target_offset,
+            config.layout.context_length,
+            config.layout.sample_length,
+            data_offset=config.layout.input_offset,
+            target_offset=config.layout.get_target_offset(config.training_objective),
         )
         return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
 
@@ -495,14 +495,16 @@ def infer_embedding(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.sample_length)
+                validate_stored_window_width(tensor, config.layout.sample_length)
 
             metadata = generate_padding_masks(
                 left_pad_lengths_tensor,
-                config.context_length,
-                config.sample_length,
-                data_offset=config.data_offset,
-                target_offset=config.target_offset,
+                config.layout.context_length,
+                config.layout.sample_length,
+                data_offset=config.layout.input_offset,
+                target_offset=config.layout.get_target_offset(
+                    config.training_objective
+                ),
             )
             embeddings = get_embeddings_pt(
                 config, inferer, sequences_dict, metadata=metadata
@@ -521,7 +523,8 @@ def infer_embedding(
         # Step 2: Calculate absolute positions and repeat IDs
         # (e.g., for seq_len=50, inf_size=5, offsets are [45, 46, 47, 48, 49])
         base_offsets = np.arange(
-            config.context_length - prediction_length, config.context_length
+            config.layout.context_length - prediction_length,
+            config.layout.context_length,
         )
 
         # Tile these offsets for each sample in the batch
@@ -668,7 +671,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.context_length,
+                    config.layout.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -681,7 +684,11 @@ def infer_generative(
                 # Unpack the new third return value
                 probs, preds, sequence_ids_for_preds, item_positions_for_preds = (
                     get_probs_preds_autoregression(
-                        config, inferer, data, column_types, config.context_length
+                        config,
+                        inferer,
+                        data,
+                        column_types,
+                        config.layout.context_length,
                     )
                 )
         elif config.read_format == "parquet" and is_folder_input:
@@ -719,7 +726,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.context_length,
+                    config.layout.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -731,7 +738,11 @@ def infer_generative(
 
                 probs, preds, sequence_ids_for_preds, item_positions_for_preds = (
                     get_probs_preds_autoregression(
-                        config, inferer, data, column_types, config.context_length
+                        config,
+                        inferer,
+                        data,
+                        column_types,
+                        config.layout.context_length,
                     )
                 )
         elif config.read_format == "pt":
@@ -743,7 +754,7 @@ def infer_generative(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.sample_length)
+                validate_stored_window_width(tensor, config.layout.sample_length)
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
@@ -751,17 +762,21 @@ def infer_generative(
             )
 
             sequences_dict = {
-                key: slice_window(tensor, config.context_length, config.data_offset)
+                key: slice_window(
+                    tensor, config.layout.context_length, config.layout.input_offset
+                )
                 for key, tensor in sequences_dict.items()
                 if key in config.input_columns
             }
 
             metadata = generate_padding_masks(
                 left_pad_lengths_tensor,
-                config.context_length,
-                config.sample_length,
-                data_offset=config.data_offset,
-                target_offset=config.target_offset,
+                config.layout.context_length,
+                config.layout.sample_length,
+                data_offset=config.layout.input_offset,
+                target_offset=config.layout.get_target_offset(
+                    config.training_objective
+                ),
             )
 
             probs, preds = get_probs_preds_from_dict(
@@ -785,7 +800,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.context_length,
+                    config.layout.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -795,8 +810,12 @@ def infer_generative(
                     sequence_ids_tensor.numpy(), total_steps
                 )
                 item_position_boundaries = zip(
-                    list(start_positions_tensor + config.context_length),
-                    list(start_positions_tensor + config.context_length + total_steps),
+                    list(start_positions_tensor + config.layout.context_length),
+                    list(
+                        start_positions_tensor
+                        + config.layout.context_length
+                        + total_steps
+                    ),
                 )
                 item_positions_for_preds = np.concatenate(
                     [np.arange(start, end) for start, end in item_position_boundaries],
@@ -941,9 +960,11 @@ def get_embeddings_pt(
         A NumPy array containing the computed embeddings for the batch.
     """
     for tensor in data.values():
-        validate_stored_window_width(tensor, config.sample_length)
+        validate_stored_window_width(tensor, config.layout.sample_length)
     X = {
-        key: slice_window(val, config.context_length, config.data_offset).numpy()
+        key: slice_window(
+            val, config.layout.context_length, config.layout.input_offset
+        ).numpy()
         for key, val in data.items()
         if key in config.input_columns
     }
@@ -1097,10 +1118,10 @@ def get_embeddings(
         data,
         column_types,
         all_columns,
-        config.context_length,
-        config.sample_length,
-        config.data_offset,
-        config.target_offset,
+        config.layout.context_length,
+        config.layout.sample_length,
+        config.layout.input_offset,
+        config.layout.get_target_offset(config.training_objective),
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     metadata_np = {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
@@ -1146,10 +1167,10 @@ def get_probs_preds_from_df(
         data,
         column_types,
         all_columns,
-        config.context_length,
-        config.sample_length,
-        config.data_offset,
-        config.target_offset,
+        config.layout.context_length,
+        config.layout.sample_length,
+        config.layout.input_offset,
+        config.layout.get_target_offset(config.training_objective),
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     metadata_np = {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
@@ -1273,8 +1294,8 @@ def get_probs_preds_autoregression(
         column_types,
         config.input_columns,
         context_length,
-        config.sample_length,
-        data_offset=config.data_offset,
+        config.layout.sample_length,
+        data_offset=config.layout.input_offset,
         target_offset=0,
     )
 
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index fc082183..71ce10ac 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -46,10 +46,12 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             data=data_df,
             column_types=column_types,
             all_columns=all_columns,
-            context_length=config.context_length,
-            sample_length=config.sample_length,
-            data_offset=config.training_spec.data_offset,
-            target_offset=config.training_spec.target_offset,
+            context_length=config.layout.context_length,
+            sample_length=config.layout.sample_length,
+            data_offset=config.layout.input_offset,
+            target_offset=config.layout.get_target_offset(
+                config.training_spec.training_objective
+            ),
         )
         self.n_samples = all_tensors[all_columns[0]].shape[0]
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 373e0512..6eebda70 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -50,10 +50,10 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             metadata = json.load(f)
 
         folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.sample_length:
+        if folder_layout.sample_length != config.layout.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.sample_length}."
+                f"does not match config sample_length={config.layout.sample_length}."
             )
 
         self.n_samples = metadata["total_samples"]
@@ -68,12 +68,15 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         }
 
         # Sequence formatting structures matching long-format schema boundaries
-        train_seq_len = self.config.context_length
+        train_seq_len = self.config.layout.context_length
         input_seq_cols = sequence_column_names(
-            train_seq_len, self.config.training_spec.data_offset
+            train_seq_len, self.config.layout.input_offset
         )
         target_seq_cols = sequence_column_names(
-            train_seq_len, self.config.training_spec.target_offset
+            train_seq_len,
+            self.config.layout.get_target_offset(
+                self.config.training_spec.training_objective
+            ),
         )
         all_sequences: Dict[str, list[torch.Tensor]] = {
             col: [] for col in config.input_columns
@@ -208,7 +211,7 @@ def __iter__(
         indices_for_worker = indices_for_rank[worker_id::num_workers]
 
         # 5. Extract and pass unified data frames
-        train_seq_len = self.config.context_length
+        train_seq_len = self.config.layout.context_length
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
@@ -226,9 +229,11 @@ def __iter__(
                 metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
-                    self.config.sample_length,
-                    self.config.training_spec.data_offset,
-                    self.config.training_spec.target_offset,
+                    self.config.layout.sample_length,
+                    self.config.layout.input_offset,
+                    self.config.layout.get_target_offset(
+                        self.config.training_spec.training_objective
+                    ),
                 )
 
             yield SequifierBatch(
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 0546c1d7..496fbf6c 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -66,10 +66,10 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             metadata = json.load(f)
 
         folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.sample_length:
+        if folder_layout.sample_length != config.layout.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.sample_length}."
+                f"does not match config sample_length={config.layout.sample_length}."
             )
 
         self.batch_files_info = metadata["batch_files"]
@@ -219,14 +219,17 @@ def __iter__(
 
         # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
-        train_seq_len = self.config.context_length
+        train_seq_len = self.config.layout.context_length
         global_file_start_sample = 0
 
         input_seq_cols = sequence_column_names(
-            train_seq_len, self.config.training_spec.data_offset
+            train_seq_len, self.config.layout.input_offset
         )
         target_seq_cols = sequence_column_names(
-            train_seq_len, self.config.training_spec.target_offset
+            train_seq_len,
+            self.config.layout.get_target_offset(
+                self.config.training_spec.training_objective
+            ),
         )
 
         # Initialize cross-file buffers
@@ -313,9 +316,11 @@ def __iter__(
             new_meta = generate_padding_masks(
                 left_pad_lengths[worker_indices],
                 train_seq_len,
-                self.config.sample_length,
-                self.config.training_spec.data_offset,
-                self.config.training_spec.target_offset,
+                self.config.layout.sample_length,
+                self.config.layout.input_offset,
+                self.config.layout.get_target_offset(
+                    self.config.training_spec.training_objective
+                ),
             )
 
             del df
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 9605020e..a1647fea 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -48,10 +48,10 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             metadata = json.load(f)
 
         folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.sample_length:
+        if folder_layout.sample_length != config.layout.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.sample_length}."
+                f"does not match config sample_length={config.layout.sample_length}."
             )
 
         self.n_samples = metadata["total_samples"]
@@ -78,7 +78,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in all_sequences.keys():
                 if col in sequences_batch:
                     validate_stored_window_width(
-                        sequences_batch[col], config.sample_length
+                        sequences_batch[col], config.layout.sample_length
                     )
                     all_sequences[col].append(sequences_batch[col])
             all_left_pad_lengths.append(left_pad_lengths_batch)
@@ -169,12 +169,14 @@ def __iter__(
         indices_for_worker = indices_for_rank[worker_id::num_workers]
 
         # 5. Yield full batches
-        train_seq_len = self.config.context_length
+        train_seq_len = self.config.layout.context_length
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
-            data_offset = self.config.training_spec.data_offset
-            target_offset = self.config.training_spec.target_offset
+            data_offset = self.config.layout.input_offset
+            target_offset = self.config.layout.get_target_offset(
+                self.config.training_spec.training_objective
+            )
             data_batch = {
                 key: slice_window(tensor[batch_indices], train_seq_len, data_offset)
                 for key, tensor in self.sequences.items()
@@ -191,7 +193,7 @@ def __iter__(
                 metadata_batch = generate_padding_masks(
                     self.left_pad_lengths[batch_indices],
                     train_seq_len,
-                    self.config.sample_length,
+                    self.config.layout.sample_length,
                     data_offset,
                     target_offset,
                 )
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index c3b787d2..4c50e903 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -64,10 +64,10 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             metadata = json.load(f)
 
         folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.sample_length:
+        if folder_layout.sample_length != config.layout.sample_length:
             raise ValueError(
                 f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.sample_length}."
+                f"does not match config sample_length={config.layout.sample_length}."
             )
 
         self.batch_files_info = metadata["batch_files"]
@@ -212,7 +212,7 @@ def __iter__(
 
         # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
-        train_seq_len = self.config.context_length
+        train_seq_len = self.config.layout.context_length
         global_file_start_sample = 0
 
         # Initialize cross-file buffers
@@ -244,7 +244,7 @@ def __iter__(
                 left_pad_lengths_batch,
             ) = torch.load(file_path, map_location="cpu", weights_only=False)
             for tensor in sequences_batch.values():
-                validate_stored_window_width(tensor, self.config.sample_length)
+                validate_stored_window_width(tensor, self.config.layout.sample_length)
 
             # Generate indices for the whole file
             indices = torch.arange(file_samples)
@@ -265,8 +265,10 @@ def __iter__(
                 continue
 
             # Extract the data subset for this worker (Advanced indexing copies the data)
-            data_offset = self.config.training_spec.data_offset
-            target_offset = self.config.training_spec.target_offset
+            data_offset = self.config.layout.input_offset
+            target_offset = self.config.layout.get_target_offset(
+                self.config.training_spec.training_objective
+            )
             new_seq = {
                 k: slice_window(v[worker_indices], train_seq_len, data_offset)
                 for k, v in sequences_batch.items()
@@ -281,7 +283,7 @@ def __iter__(
             new_meta = generate_padding_masks(
                 left_pad_lengths_batch[worker_indices],
                 train_seq_len,
-                self.config.sample_length,
+                self.config.layout.sample_length,
                 data_offset,
                 target_offset,
             )
diff --git a/src/sequifier/io/yaml.py b/src/sequifier/io/yaml.py
index f8cdcd63..b35497b4 100644
--- a/src/sequifier/io/yaml.py
+++ b/src/sequifier/io/yaml.py
@@ -8,6 +8,7 @@
     TrainingSpecModel,
     TrainModel,
 )
+from sequifier.helpers import SequenceLayout
 
 
 def represent_sequifier_object(dumper, data):
@@ -78,6 +79,7 @@ def increase_indent(self, flow=False, indentless=False):
 TrainModelDumper.add_representer(TrainModel, represent_sequifier_object)
 TrainModelDumper.add_representer(ModelSpecModel, represent_sequifier_object)
 TrainModelDumper.add_representer(TrainingSpecModel, represent_sequifier_object)
+TrainModelDumper.add_representer(SequenceLayout, represent_sequifier_object)
 TrainModelDumper.add_multi_representer(BaseModel, represent_sequifier_object)
 TrainModelDumper.add_representer(DotDict, represent_dot_dict)
 TrainModelDumper.add_representer(numpy.float64, represent_numpy_float)
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index efa78380..f6b58150 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -97,14 +97,14 @@ def create_dummy_data_and_metadata(
     for col in config.input_columns:
         dtype = torch.int64 if col in config.categorical_columns else torch.float32
         dummy_data[col] = torch.ones(
-            (config.training_spec.batch_size, config.context_length),
+            (config.training_spec.batch_size, config.layout.context_length),
             dtype=dtype,
             device=local_rank,
         )
 
     dummy_metadata = {
         "attention_valid_mask": torch.ones(
-            (config.training_spec.batch_size, config.context_length),
+            (config.training_spec.batch_size, config.layout.context_length),
             dtype=torch.bool,
             device=local_rank,
         )
@@ -613,7 +613,8 @@ def __init__(
         self.target_columns = hparams.target_columns
         self.target_column_types = hparams.target_column_types
         self.loss_weights = hparams.training_spec.loss_weights
-        self.context_length = hparams.context_length
+        self.layout = hparams.layout
+        self.context_length = hparams.layout.context_length
         self.n_classes = hparams.n_classes
         self.inference_batch_size = hparams.inference_batch_size
         self.log_interval = hparams.training_spec.log_interval
@@ -690,7 +691,7 @@ def __init__(
                     hparams.model_spec.n_head,
                     hparams.model_spec.dim_feedforward,
                     hparams.training_spec.dropout,
-                    hparams.context_length,
+                    hparams.layout.context_length,
                 )
                 for _ in range(hparams.model_spec.num_layers)
             ]
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index 0869ac35..991205fd 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -40,9 +40,8 @@ def test_hp_search_bert_outputs(run_hp_search, project_root):
     assert generated_config["metadata_config_path"].endswith(
         "test-data-categorical-1-lookahead-0.json"
     )
-    assert generated_config["max_lookahead"] == 0
-    assert generated_config["sequence_layout_version"] == 2
-    assert generated_config["sample_length"] == generated_config["context_length"]
+    assert generated_config["layout"]["max_lookahead"] == 0
+    assert generated_config["layout"]["sequence_layout_version"] == 2
 
 
 def test_hp_search_bayesian_outputs(run_hp_search, project_root):
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index 25e8639d..d3c0839b 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -4,6 +4,7 @@
 import onnxruntime
 
 from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
+from sequifier.helpers import SequenceLayout
 from sequifier.train import TransformerModel
 
 
@@ -27,9 +28,11 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         real_columns=["real_col"],
         id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
         n_classes={"cat_col": 6},
-        context_length=context_length,
-        sample_length=5,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=context_length,
+            max_lookahead=1,
+            sequence_layout_version=2,
+        ),
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -73,7 +76,6 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
             loss_weights={"cat_col": 1.0, "real_col": 1.0},
             torch_compile="none",
             layer_autocast=False,
-            sample_length=5,
         ),
     )
     model = TransformerModel(config)
@@ -128,9 +130,11 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             "cat10": {"x": 3, "y": 4, "z": 5},
         },
         n_classes={"cat2": 7, "cat10": 6},
-        context_length=context_length,
-        sample_length=5,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=context_length,
+            max_lookahead=1,
+            sequence_layout_version=2,
+        ),
         inference_batch_size=inference_batch_size,
         seed=42,
         export_generative_model=True,
@@ -165,7 +169,6 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             loss_weights={"cat2": 1.0},
             torch_compile="none",
             layer_autocast=False,
-            sample_length=5,
         ),
     )
     model = TransformerModel(config)
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index f4696079..3cbe5f33 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -5,6 +5,7 @@
 import pytest
 import torch
 
+from sequifier.helpers import SequenceLayout
 from sequifier.io.sequifier_dataset_from_folder_parquet_lazy import (
     SequifierDatasetFromFolderParquetLazy,
 )
@@ -30,14 +31,16 @@ def mock_config():
     config = MagicMock()
     config.project_root = "."
     config.seed = 42
-    config.context_length = CONTEXT_LENGTH
-    config.sample_length = SAMPLE_LENGTH
+    config.layout = SequenceLayout(
+        context_length=CONTEXT_LENGTH,
+        max_lookahead=MAX_LOOKAHEAD,
+        sequence_layout_version=2,
+    )
     config.column_types = {"item": "Float64"}
     config.training_spec.batch_size = 5
     config.training_spec.num_workers = 0
     config.training_spec.sampling_strategy = "exact"
-    config.training_spec.data_offset = 1
-    config.training_spec.target_offset = 0
+    config.training_spec.training_objective = "causal"
     config.input_columns = ["item"]
     config.target_columns = ["item"]
 
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 4fb599cc..3eed7bcd 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -4,6 +4,7 @@
 import pytest
 import torch
 
+from sequifier.helpers import SequenceLayout
 from sequifier.io.sequifier_dataset_from_folder_pt_lazy import (
     SequifierDatasetFromFolderPtLazy,
 )
@@ -34,12 +35,14 @@ def mock_config(tmp_path):
     config.training_spec.sampling_strategy = "exact"
     config.training_spec.num_workers = 0
     config.seed = 42
-    config.context_length = CONTEXT_LENGTH
-    config.sample_length = SAMPLE_LENGTH
+    config.layout = SequenceLayout(
+        context_length=CONTEXT_LENGTH,
+        max_lookahead=MAX_LOOKAHEAD,
+        sequence_layout_version=2,
+    )
     config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
-    config.training_spec.data_offset = 1
-    config.training_spec.target_offset = 0
+    config.training_spec.training_objective = "causal"
     return config
 
 
diff --git a/tests/unit/test_hyperparameter_search_config.py b/tests/unit/test_hyperparameter_search_config.py
index 8a50fc34..77ea8639 100644
--- a/tests/unit/test_hyperparameter_search_config.py
+++ b/tests/unit/test_hyperparameter_search_config.py
@@ -63,7 +63,7 @@ def bert_spec_sampling_config():
 
 
 def sample_training_spec(sampling, trial):
-    return sampling.sample_trial(trial, max_lookahead=1, sample_length=9)
+    return sampling.sample_trial(trial)
 
 
 def test_training_objective_defaults_to_causal_without_bert_spec():
@@ -74,7 +74,6 @@ def test_training_objective_defaults_to_causal_without_bert_spec():
 
     assert training_spec.training_objective == "causal"
     assert training_spec.bert_spec is None
-    assert training_spec.sample_length == 9
     assert trial.params["training_objective"] == "causal"
 
 
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 321446a5..97ba2fb8 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -8,6 +8,7 @@
 import yaml
 
 from sequifier.config.infer_config import InfererModel, load_inferer_config
+from sequifier.helpers import SequenceLayout
 from sequifier.infer import (
     Inferer,
     calculate_item_positions,
@@ -173,16 +174,16 @@ def test_infer_config_defaults_bert_prediction_length_to_context_length():
         seed=42,
         device="cpu",
         prediction_length=None,
-        context_length=3,
-        sample_length=4,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=3, max_lookahead=1, sequence_layout_version=2
+        ),
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
         autoregression=False,
     )
 
-    assert config.prediction_length == config.context_length
+    assert config.prediction_length == config.layout.context_length
 
 
 def test_infer_config_defaults_causal_prediction_length_to_one():
@@ -202,9 +203,9 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
         seed=42,
         device="cpu",
         prediction_length=None,
-        context_length=3,
-        sample_length=4,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=3, max_lookahead=1, sequence_layout_version=2
+        ),
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -232,9 +233,9 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             seed=42,
             device="cpu",
             prediction_length=1,
-            context_length=3,
-            sample_length=4,
-            sequence_layout_version=2,
+            layout=SequenceLayout(
+                context_length=3, max_lookahead=1, sequence_layout_version=2
+            ),
             inference_batch_size=2,
             output_probabilities=False,
             map_to_id=False,
@@ -398,9 +399,9 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         seed=42,
         device="cpu",
         prediction_length=None,
-        context_length=3,
-        sample_length=4,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=3, max_lookahead=1, sequence_layout_version=2
+        ),
         inference_batch_size=4,
         output_probabilities=False,
         map_to_id=True,
@@ -615,9 +616,9 @@ def ar_config():
         seed=42,
         device="cpu",
         prediction_length=1,
-        context_length=3,
-        sample_length=4,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=3, max_lookahead=1, sequence_layout_version=2
+        ),
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,  # Set to False to bypass ID mapping requirements
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index bcc8b00e..f6bedbc4 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -15,6 +15,7 @@
     TrainModel,
     load_train_config,
 )
+from sequifier.helpers import SequenceLayout
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS
 from sequifier.train import TransformerModel
 
@@ -31,7 +32,6 @@ def _training_spec_kwargs(**overrides):
         "optimizer": {"name": "Adam"},
         "scheduler": {"name": "StepLR", "step_size": 1, "gamma": 0.1},
         "loss_weights": {"cat_col": 1.0, "real_col": 1.0},
-        "sample_length": 11,
     }
     values.update(overrides)
     return values
@@ -77,7 +77,8 @@ def test_training_spec_model_dump_excludes_runtime_offsets():
 
     assert "data_offset" not in dumped
     assert "target_offset" not in dumped
-    assert TrainingSpecModel(**dumped).target_offset == 0
+    assert "sample_length" not in dumped
+    assert "max_lookahead" not in dumped
 
 
 def test_poisson_span_masking_samples_at_least_one_token():
@@ -118,7 +119,6 @@ def model_config(tmp_path):
         optimizer={"name": "Adam"},
         scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
         loss_weights={"cat_col": 1.0, "real_col": 1.0},
-        sample_length=11,
     )
 
     config = TrainModel(
@@ -136,9 +136,9 @@ def model_config(tmp_path):
         # id_maps is needed for constructing index_maps in model init
         id_maps={"cat_col": {"a": 1, "b": 2, "c": 3, "d": 4}},
         n_classes={"cat_col": 5},  # 0 + 4 classes
-        context_length=10,
-        sample_length=11,
-        sequence_layout_version=2,
+        layout=SequenceLayout(
+            context_length=10, max_lookahead=1, sequence_layout_version=2
+        ),
         inference_batch_size=4,
         seed=42,
         export_generative_model=True,
@@ -165,7 +165,9 @@ def causal_model(model_config):
 @pytest.fixture
 def bert_model(model_config):
     config_values = model_config.model_dump()
-    config_values["model_spec"]["prediction_length"] = model_config.context_length
+    config_values["model_spec"]["prediction_length"] = (
+        model_config.layout.context_length
+    )
     config_values["training_spec"] = _training_spec_kwargs(
         training_objective="bert",
         bert_spec=_bert_spec(),
@@ -218,7 +220,9 @@ def test_train_model_requires_bert_prediction_length_to_equal_context_length(
     model_config,
 ):
     config_values = model_config.model_dump()
-    config_values["model_spec"]["prediction_length"] = model_config.context_length - 1
+    config_values["model_spec"]["prediction_length"] = (
+        model_config.layout.context_length - 1
+    )
     config_values["training_spec"] = _training_spec_kwargs(
         training_objective="bert",
         bert_spec=_bert_spec(),
@@ -248,15 +252,16 @@ def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
     config_values = model_config.model_dump()
     config_values["project_root"] = str(tmp_path)
     config_values["metadata_config_path"] = metadata_path.name
+    layout = config_values["layout"]
     config_path.write_text(yaml.safe_dump(config_values))
     metadata_path.write_text(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
-                "context_length": config_values["context_length"],
-                "max_lookahead": 1,
-                "sample_length": config_values["context_length"] + 1,
-                "sequence_layout_version": 2,
+                "context_length": layout["context_length"],
+                "max_lookahead": layout["max_lookahead"],
+                "sample_length": layout["context_length"] + layout["max_lookahead"],
+                "sequence_layout_version": layout["sequence_layout_version"],
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
@@ -281,15 +286,16 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
     config_values = model_config.model_dump()
     config_values["project_root"] = str(tmp_path)
     config_values["metadata_config_path"] = metadata_path.name
+    layout = config_values["layout"]
     config_path.write_text(yaml.safe_dump(config_values))
     metadata_path.write_text(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
-                "context_length": config_values["context_length"],
-                "max_lookahead": 1,
-                "sample_length": config_values["context_length"] + 1,
-                "sequence_layout_version": 2,
+                "context_length": layout["context_length"],
+                "max_lookahead": layout["max_lookahead"],
+                "sample_length": layout["context_length"] + layout["max_lookahead"],
+                "sequence_layout_version": layout["sequence_layout_version"],
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
@@ -306,7 +312,7 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
 def test_forward_train_shapes(model, model_config):
     """Tests the output shapes of the forward_train method."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.context_length
+    seq_len = model_config.layout.context_length
 
     # Create dummy inputs
     # Categorical: (batch, seq_len) integers
@@ -337,7 +343,7 @@ def test_forward_train_shapes(model, model_config):
 def test_forward_inference_shapes(model, model_config):
     """Tests the output shapes of the forward (inference) method."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.context_length
+    seq_len = model_config.layout.context_length
     prediction_length = model_config.model_spec.prediction_length  # 1
 
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
@@ -370,7 +376,7 @@ def test_forward_inference_shapes(model, model_config):
 def test_calculate_loss(model, model_config):
     """Tests that loss calculation returns a scalar tensor."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.context_length
+    seq_len = model_config.layout.context_length
 
     # Inputs
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
@@ -459,7 +465,7 @@ def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
 
 
 def test_padding_keys_are_masked(bert_model):
-    seq_len = bert_model.context_length
+    seq_len = bert_model.layout.context_length
 
     valid_mask = torch.ones(
         2,
@@ -488,7 +494,7 @@ def test_padding_keys_are_masked(bert_model):
 
 
 def test_causal_and_padding_masks_are_combined(causal_model):
-    seq_len = causal_model.context_length
+    seq_len = causal_model.layout.context_length
 
     valid_mask = torch.ones(
         1,
diff --git a/uv.lock b/uv.lock
new file mode 100644
index 00000000..f078af13
--- /dev/null
+++ b/uv.lock
@@ -0,0 +1,2269 @@
+version = 1
+revision = 3
+requires-python = ">=3.10"
+resolution-markers = [
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version < '3.11' and platform_machine != 's390x'",
+    "python_full_version < '3.11' and platform_machine == 's390x'",
+]
+
+[[package]]
+name = "agate"
+version = "1.14.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "babel" },
+    { name = "isodate" },
+    { name = "leather" },
+    { name = "parsedatetime" },
+    { name = "python-slugify" },
+    { name = "pytimeparse" },
+    { name = "tzdata", marker = "sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/16/48/dc4d02dba00fbe62e966ed1a7d991e51654668ab343a2738bb816aa82256/agate-1.14.2.tar.gz", hash = "sha256:7f29841c39d84b1de7fde762b8d792085371515324f3a01413b20f810398225b", size = 204186, upload-time = "2026-02-27T16:35:55.432Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/70/9977183cff3420eb1a49a31bb709e90b76ea86b6e7a090de1ea41fac14c5/agate-1.14.2-py3-none-any.whl", hash = "sha256:e11018371d344dda5aca16a68830b6f8e86c84c531a7bb8f81456bf5014937d4", size = 95981, upload-time = "2026-02-27T16:35:53.863Z" },
+]
+
+[[package]]
+name = "agate-dbf"
+version = "0.2.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "agate" },
+    { name = "dbfread" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ad/d8/abf6f39bd8c5767cc367472ea59f7d7cc4d5728388974a1b26a9472a971f/agate_dbf-0.2.4.tar.gz", hash = "sha256:6554828b10048a76dbb5bc4eff8911e059ea2b47155b7a89351e382915ca16fc", size = 7547, upload-time = "2025-12-15T18:47:46.682Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1e/2b/f9661a20b2812e7b3572c082b1cdf190a3c12e5cbfb2ef588c04ac47cbce/agate_dbf-0.2.4-py3-none-any.whl", hash = "sha256:65ef9df31cf25dcf829123fde63044ee7242bca6c47cbd90cedf318e1be0ad3e", size = 3696, upload-time = "2025-12-15T18:47:45.251Z" },
+]
+
+[[package]]
+name = "agate-excel"
+version = "0.4.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "agate" },
+    { name = "olefile" },
+    { name = "openpyxl" },
+    { name = "xlrd" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/83/e5/b2d1bc555fd91145de5d11a7b31241076586713d222881c6d7eac9e4fda9/agate_excel-0.4.2.tar.gz", hash = "sha256:eed1dc6239f0e96720d962dc1bdfb4496e19687332c827fd8b1e587a917ea202", size = 271423, upload-time = "2025-12-15T18:48:59.689Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f9/cf/d34f66780f47c6eed299ee522eaee7fbe56014f3adfdb40b5e64fbec548b/agate_excel-0.4.2-py3-none-any.whl", hash = "sha256:f93f09b3ac02489ed2ed8669c7b3ce74eab30ebf9a03038c0283211dbc3b9c1a", size = 7227, upload-time = "2025-12-15T18:48:58.442Z" },
+]
+
+[[package]]
+name = "agate-sql"
+version = "0.7.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "agate" },
+    { name = "sqlalchemy" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fe/fe/fc7662f1ec3c0917c377f74f143a479eb13c9ae5fe14d77ce28eb165564f/agate_sql-0.7.3.tar.gz", hash = "sha256:4c588a28e80bc625c7d5f915e8f8dff4900140a8a6d8a350a098a2ba9adf9d33", size = 13936, upload-time = "2025-12-15T18:47:59.675Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a7/be/8f425d3e5846fb10a627a572e671a1ba2c2a51df4d398221f40f7b666310/agate_sql-0.7.3-py3-none-any.whl", hash = "sha256:57046f628dbffb125c36c94675716b441b40193bc18461f04e32a3153a8a0971", size = 7464, upload-time = "2025-12-15T18:47:58.295Z" },
+]
+
+[[package]]
+name = "alembic"
+version = "1.18.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "mako" },
+    { name = "sqlalchemy" },
+    { name = "tomli", marker = "python_full_version < '3.11'" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/94/13/8b084e0f2efb0275a1d534838844926f798bd766566b1375174e2448cd31/alembic-1.18.4.tar.gz", hash = "sha256:cb6e1fd84b6174ab8dbb2329f86d631ba9559dd78df550b57804d607672cedbc", size = 2056725, upload-time = "2026-02-10T16:00:47.195Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d2/29/6533c317b74f707ea28f8d633734dbda2119bbadfc61b2f3640ba835d0f7/alembic-1.18.4-py3-none-any.whl", hash = "sha256:a5ed4adcf6d8a4cb575f3d759f071b03cd6e5c7618eb796cb52497be25bfe19a", size = 263893, upload-time = "2026-02-10T16:00:49.997Z" },
+]
+
+[[package]]
+name = "annotated-types"
+version = "0.7.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ee/67/531ea369ba64dcff5ec9c3402f9f51bf748cec26dde048a2f973a4eea7f5/annotated_types-0.7.0.tar.gz", hash = "sha256:aff07c09a53a08bc8cfccb9c85b05f1aa9a2a6f23728d790723543408344ce89", size = 16081, upload-time = "2024-05-20T21:33:25.928Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/78/b6/6307fbef88d9b5ee7421e68d78a9f162e0da4900bc5f5793f6d3d0e34fb8/annotated_types-0.7.0-py3-none-any.whl", hash = "sha256:1f02e8b43a8fbbc3f3e0d4f0f4bfc8131bcb4eebe8849b8e5c773f3a1c582a53", size = 13643, upload-time = "2024-05-20T21:33:24.1Z" },
+]
+
+[[package]]
+name = "babel"
+version = "2.18.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7d/b2/51899539b6ceeeb420d40ed3cd4b7a40519404f9baf3d4ac99dc413a834b/babel-2.18.0.tar.gz", hash = "sha256:b80b99a14bd085fcacfa15c9165f651fbb3406e66cc603abf11c5750937c992d", size = 9959554, upload-time = "2026-02-01T12:30:56.078Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/77/f5/21d2de20e8b8b0408f0681956ca2c69f1320a3848ac50e6e7f39c6159675/babel-2.18.0-py3-none-any.whl", hash = "sha256:e2b422b277c2b9a9630c1d7903c2a00d0830c409c59ac8cae9081c92f1aeba35", size = 10196845, upload-time = "2026-02-01T12:30:53.445Z" },
+]
+
+[[package]]
+name = "beartype"
+version = "0.18.5"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/96/15/4e623478a9628ad4cee2391f19aba0b16c1dd6fedcb2a399f0928097b597/beartype-0.18.5.tar.gz", hash = "sha256:264ddc2f1da9ec94ff639141fbe33d22e12a9f75aa863b83b7046ffff1381927", size = 1193506, upload-time = "2024-04-21T07:25:58.64Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/64/43/7a1259741bd989723272ac7d381a43be932422abcff09a1d9f7ba212cb74/beartype-0.18.5-py3-none-any.whl", hash = "sha256:5301a14f2a9a5540fe47ec6d34d758e9cd8331d36c4760fc7a5499ab86310089", size = 917762, upload-time = "2024-04-21T07:25:55.758Z" },
+]
+
+[[package]]
+name = "colorama"
+version = "0.4.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44", size = 27697, upload-time = "2022-10-25T02:36:22.414Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335, upload-time = "2022-10-25T02:36:20.889Z" },
+]
+
+[[package]]
+name = "colorlog"
+version = "6.10.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a2/61/f083b5ac52e505dfc1c624eafbf8c7589a0d7f32daa398d2e7590efa5fda/colorlog-6.10.1.tar.gz", hash = "sha256:eb4ae5cb65fe7fec7773c2306061a8e63e02efc2c72eba9d27b0fa23c94f1321", size = 17162, upload-time = "2025-10-16T16:14:11.978Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6d/c1/e419ef3723a074172b68aaa89c9f3de486ed4c2399e2dbd8113a4fdcaf9e/colorlog-6.10.1-py3-none-any.whl", hash = "sha256:2d7e8348291948af66122cff006c9f8da6255d224e7cf8e37d8de2df3bad8c9c", size = 11743, upload-time = "2025-10-16T16:14:10.512Z" },
+]
+
+[[package]]
+name = "cramjam"
+version = "2.11.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/14/12/34bf6e840a79130dfd0da7badfb6f7810b8fcfd60e75b0539372667b41b6/cramjam-2.11.0.tar.gz", hash = "sha256:5c82500ed91605c2d9781380b378397012e25127e89d64f460fea6aeac4389b4", size = 99100, upload-time = "2025-07-27T21:25:07.559Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/53/d3/20d0402e4e983b66603117ad3dd3b864a05d7997a830206d3ff9cacef9a2/cramjam-2.11.0-cp310-cp310-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:d0859c65775e8ebf2cbc084bfd51bd0ffda10266da6f9306451123b89f8e5a63", size = 3558999, upload-time = "2025-07-27T21:21:34.105Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/a8/a6e2744288938ccd320a5c6f6f3653faa790f933f5edd088c6e5782a2354/cramjam-2.11.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:1d77b9b0aca02a3f6eeeff27fcd315ca5972616c0919ee38e522cce257bcd349", size = 1861558, upload-time = "2025-07-27T21:21:36.624Z" },
+    { url = "https://files.pythonhosted.org/packages/96/29/7961e09a849eea7d8302e7baa6f829dd3ef3faf199cb25ed29b318ae799b/cramjam-2.11.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:66425bc25b5481359b12a6719b6e7c90ffe76d85d0691f1da7df304bfb8ce45c", size = 1699431, upload-time = "2025-07-27T21:21:38.396Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/60/6665e52f01a8919bf37c43dcf0e03b6dd3866f5c4e95440b357d508ee14e/cramjam-2.11.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:bd748d3407ec63e049b3aea1595e218814fccab329b7fb10bb51120a30e9fb7e", size = 2025262, upload-time = "2025-07-27T21:21:40.417Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/80/79bd84dbeb109e2c6efb74e661b7bd4c3ba393208ebcf69e2ae9454ae80c/cramjam-2.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a6d9a23a35b3a105c42a8de60fc2e80281ae6e758f05a3baea0b68eb1ddcb679", size = 1766177, upload-time = "2025-07-27T21:21:42.224Z" },
+    { url = "https://files.pythonhosted.org/packages/28/ef/b43280767ebcde022ba31f1e9902137655a956ae30e920d75630fa67e36e/cramjam-2.11.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:40a75b95e05e38a2a055b2446f09994ce1139151721659315151d4ad6289bbff", size = 1854031, upload-time = "2025-07-27T21:21:43.651Z" },
+    { url = "https://files.pythonhosted.org/packages/60/1c/79d522757c494dfd9e9b208b0604cc7e97b481483cc477144f5705a06ab7/cramjam-2.11.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e5d042c376d2025300da37d65192d06a457918b63b31140f697f85fd8e310b29", size = 2035812, upload-time = "2025-07-27T21:21:45.473Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/70/3bf0670380069b3abd4c6b53f61d3148f4e08935569c08efbeaf7550e87d/cramjam-2.11.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cb148b35ab20c75b19a06c27f05732e2a321adbd86fadc93f9466dbd7b1154a7", size = 2067661, upload-time = "2025-07-27T21:21:47.901Z" },
+    { url = "https://files.pythonhosted.org/packages/db/7e/4f6ca98a4b474348e965a529b359184785d1119ab7c4c9ec1280b8bea50a/cramjam-2.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0ee47c220f0f5179ddc923ab91fc9e282c27b29fabc60c433dfe06f08084f798", size = 1981523, upload-time = "2025-07-27T21:21:49.704Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/6c/b241511c7ffd5f1da29641429bb0e19b5fbcffafde5ba1bbcbf9394ea456/cramjam-2.11.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:0cf1b5a81b21ea175c976c3ab09e00494258f4b49b7995efc86060cced3f0b2e", size = 2034251, upload-time = "2025-07-27T21:21:51.252Z" },
+    { url = "https://files.pythonhosted.org/packages/14/5c/4ef926c8c3c1bf6da96f9c53450ff334cdb6d0fc1efced0aea97e2090803/cramjam-2.11.0-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:360c00338ecf48921492455007f904be607fc7818de3d681acbcc542aae2fb36", size = 2155322, upload-time = "2025-07-27T21:21:53.348Z" },
+    { url = "https://files.pythonhosted.org/packages/be/fb/eb2aef7fb2730e56c5a2c9000817ee8fb4a95c92f19cc6e441afed42ec29/cramjam-2.11.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:f31fcc0d30dc3f3e94ea6b4d8e1a855071757c6abf6a7b1e284050ab7d4c299c", size = 2169094, upload-time = "2025-07-27T21:21:55.187Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/80/925a5c668dcee1c6f61775067185c5dc9a63c766d5393e5c60d2af4217a7/cramjam-2.11.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:033be66fdceb3d63b2c99b257a98380c4ec22c9e4dca54a2bfec3718cd24e184", size = 2159089, upload-time = "2025-07-27T21:21:57.118Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/ac/b2819640eef0592a6de7ca832c0d23c69bd1620f765ce88b60dbc8da9ba2/cramjam-2.11.0-cp310-cp310-win32.whl", hash = "sha256:1c6cea67f6000b81f6bd27d14c8a6f62d00336ca7252fd03ee16f6b70eb5c0d2", size = 1605046, upload-time = "2025-07-27T21:21:58.617Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/f4/06af04727b9556721049e2127656d727306d275c518e3d97f9ed4cffd0d8/cramjam-2.11.0-cp310-cp310-win_amd64.whl", hash = "sha256:98aa4a351b047b0f7f9e971585982065028adc2c162c5c23c5d5734c5ccc1077", size = 1710647, upload-time = "2025-07-27T21:22:00.279Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/89/8001f6a9b6b6e9fa69bec5319789083475d6f26d52aaea209d3ebf939284/cramjam-2.11.0-cp311-cp311-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:04cfa39118570e70e920a9b75c733299784b6d269733dbc791d9aaed6edd2615", size = 3559272, upload-time = "2025-07-27T21:22:01.988Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/f3/001d00070ca92e5fbe6aacc768e455568b0cde46b0eb944561a4ea132300/cramjam-2.11.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:66a18f68506290349a256375d7aa2f645b9f7993c10fc4cc211db214e4e61d2b", size = 1861743, upload-time = "2025-07-27T21:22:03.754Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/35/041a3af01bf3f6158f120070f798546d4383b962b63c35cd91dcbf193e17/cramjam-2.11.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:50e7d65533857736cd56f6509cf2c4866f28ad84dd15b5bdbf2f8a81e77fa28a", size = 1699631, upload-time = "2025-07-27T21:22:05.192Z" },
+    { url = "https://files.pythonhosted.org/packages/17/eb/5358b238808abebd0c949c42635c3751204ca7cf82b29b984abe9f5e33c8/cramjam-2.11.0-cp311-cp311-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:1f71989668458fc327ac15396db28d92df22f8024bb12963929798b2729d2df5", size = 2025603, upload-time = "2025-07-27T21:22:06.726Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/79/19dba7c03a27408d8d11b5a7a4a7908459cfd4e6f375b73264dc66517bf6/cramjam-2.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ee77ac543f1e2b22af1e8be3ae589f729491b6090582340aacd77d1d757d9569", size = 1766283, upload-time = "2025-07-27T21:22:08.568Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/ad/40e4b3408501d886d082db465c33971655fe82573c535428e52ab905f4d0/cramjam-2.11.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ad52784120e7e4d8a0b5b0517d185b8bf7f74f5e17272857ddc8951a628d9be1", size = 1854407, upload-time = "2025-07-27T21:22:10.518Z" },
+    { url = "https://files.pythonhosted.org/packages/36/6e/c1b60ceb6d7ea6ff8b0bf197520aefe23f878bf2bfb0de65f2b0c2f82cd1/cramjam-2.11.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4b86f8e6d9c1b3f9a75b2af870c93ceee0f1b827cd2507387540e053b35d7459", size = 2035793, upload-time = "2025-07-27T21:22:12.504Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/ad/32a8d5f4b1e3717787945ec6d71bd1c6e6bccba4b7e903fc0d9d4e4b08c3/cramjam-2.11.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:320d61938950d95da2371b46c406ec433e7955fae9f396c8e1bf148ffc187d11", size = 2067499, upload-time = "2025-07-27T21:22:14.067Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/cd/3b5a662736ea62ff7fa4c4a10a85e050bfdaad375cc53dc80427e8afe41c/cramjam-2.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41eafc8c1653a35a5c7e75ad48138f9f60085cc05cd99d592e5298552d944e9f", size = 1981853, upload-time = "2025-07-27T21:22:15.908Z" },
+    { url = "https://files.pythonhosted.org/packages/26/8e/1dbcfaaa7a702ee82ee683ec3a81656934dd7e04a7bc4ee854033686f98a/cramjam-2.11.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:03a7316c6bf763dfa34279335b27702321da44c455a64de58112968c0818ec4a", size = 2034514, upload-time = "2025-07-27T21:22:17.352Z" },
+    { url = "https://files.pythonhosted.org/packages/50/62/f11709bfdce74af79a88b410dcb76dedc97612166e759136931bf63cfd7b/cramjam-2.11.0-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:244c2ed8bd7ccbb294a2abe7ca6498db7e89d7eb5e744691dc511a7dc82e65ca", size = 2155343, upload-time = "2025-07-27T21:22:18.854Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/6d/3b98b61841a5376d9a9b8468ae58753a8e6cf22be9534a0fa5af4d8621cc/cramjam-2.11.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:405f8790bad36ce0b4bbdb964ad51507bfc7942c78447f25cb828b870a1d86a0", size = 2169367, upload-time = "2025-07-27T21:22:20.389Z" },
+    { url = "https://files.pythonhosted.org/packages/11/72/bd5db5c49dbebc8b002f1c4983101b28d2e7fc9419753db1c31ec22b03ef/cramjam-2.11.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:6b1b751a5411032b08fb3ac556160229ca01c6bbe4757bb3a9a40b951ebaac23", size = 2159334, upload-time = "2025-07-27T21:22:22.254Z" },
+    { url = "https://files.pythonhosted.org/packages/34/32/203c57acdb6eea727e7078b2219984e64ed4ad043c996ed56321301ba167/cramjam-2.11.0-cp311-cp311-win32.whl", hash = "sha256:5251585608778b9ac8effed544933df7ad85b4ba21ee9738b551f17798b215ac", size = 1605313, upload-time = "2025-07-27T21:22:24.126Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/bd/102d6deb87a8524ac11cddcd31a7612b8f20bf9b473c3c645045e3b957c7/cramjam-2.11.0-cp311-cp311-win_amd64.whl", hash = "sha256:dca88bc8b68ce6d35dafd8c4d5d59a238a56c43fa02b74c2ce5f9dfb0d1ccb46", size = 1710991, upload-time = "2025-07-27T21:22:25.661Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/0d/7c84c913a5fae85b773a9dcf8874390f9d68ba0fcc6630efa7ff1541b950/cramjam-2.11.0-cp312-cp312-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:dba5c14b8b4f73ea1e65720f5a3fe4280c1d27761238378be8274135c60bbc6e", size = 3553368, upload-time = "2025-07-27T21:22:27.162Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/cc/4f6d185d8a744776f53035e72831ff8eefc2354f46ab836f4bd3c4f6c138/cramjam-2.11.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:11eb40722b3fcf3e6890fba46c711bf60f8dc26360a24876c85e52d76c33b25b", size = 1860014, upload-time = "2025-07-27T21:22:28.738Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/a8/626c76263085c6d5ded0e71823b411e9522bfc93ba6cc59855a5869296e7/cramjam-2.11.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:aeb26e2898994b6e8319f19a4d37c481512acdcc6d30e1b5ecc9d8ec57e835cb", size = 1693512, upload-time = "2025-07-27T21:22:30.999Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/52/0851a16a62447532e30ba95a80e638926fdea869a34b4b5b9d0a020083ba/cramjam-2.11.0-cp312-cp312-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:4f8d82081ed7d8fe52c982bd1f06e4c7631a73fe1fb6d4b3b3f2404f87dc40fe", size = 2025285, upload-time = "2025-07-27T21:22:32.954Z" },
+    { url = "https://files.pythonhosted.org/packages/98/76/122e444f59dbc216451d8e3d8282c9665dc79eaf822f5f1470066be1b695/cramjam-2.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:092a3ec26e0a679305018380e4f652eae1b6dfe3fc3b154ee76aa6b92221a17c", size = 1761327, upload-time = "2025-07-27T21:22:34.484Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/bc/3a0189aef1af2b29632c039c19a7a1b752bc21a4053582a5464183a0ad3d/cramjam-2.11.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:529d6d667c65fd105d10bd83d1cd3f9869f8fd6c66efac9415c1812281196a92", size = 1854075, upload-time = "2025-07-27T21:22:36.157Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/80/8a6343b13778ce52d94bb8d5365a30c3aa951276b1857201fe79d7e2ad25/cramjam-2.11.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:555eb9c90c450e0f76e27d9ff064e64a8b8c6478ab1a5594c91b7bc5c82fd9f0", size = 2032710, upload-time = "2025-07-27T21:22:38.17Z" },
+    { url = "https://files.pythonhosted.org/packages/df/6b/cd1778a207c29eda10791e3dfa018b588001928086e179fc71254793c625/cramjam-2.11.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5edf4c9e32493035b514cf2ba0c969d81ccb31de63bd05490cc8bfe3b431674e", size = 2068353, upload-time = "2025-07-27T21:22:39.615Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/f0/5c2a5cd5711032f3b191ca50cb786c17689b4a9255f9f768866e6c9f04d9/cramjam-2.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2fa2fe41f48c4d58d923803383b0737f048918b5a0d10390de9628bb6272b107", size = 1978104, upload-time = "2025-07-27T21:22:41.106Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/8b/b363a5fb2c3347504fe9a64f8d0f1e276844f0e532aa7162c061cd1ffee4/cramjam-2.11.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:9ca14cf1cabdb0b77d606db1bb9e9ca593b1dbd421fcaf251ec9a5431ec449f3", size = 2030779, upload-time = "2025-07-27T21:22:42.969Z" },
+    { url = "https://files.pythonhosted.org/packages/78/7b/d83dad46adb6c988a74361f81ad9c5c22642be53ad88616a19baedd06243/cramjam-2.11.0-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:309e95bf898829476bccf4fd2c358ec00e7ff73a12f95a3cdeeba4bb1d3683d5", size = 2155297, upload-time = "2025-07-27T21:22:44.6Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/be/60d9be4cb33d8740a4aa94c7513f2ef3c4eba4fd13536f086facbafade71/cramjam-2.11.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:86dca35d2f15ef22922411496c220f3c9e315d5512f316fe417461971cc1648d", size = 2169255, upload-time = "2025-07-27T21:22:46.534Z" },
+    { url = "https://files.pythonhosted.org/packages/11/b0/4a595f01a243aec8ad272b160b161c44351190c35d98d7787919d962e9e5/cramjam-2.11.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:193c6488bd2f514cbc0bef5c18fad61a5f9c8d059dd56edf773b3b37f0e85496", size = 2155651, upload-time = "2025-07-27T21:22:48.46Z" },
+    { url = "https://files.pythonhosted.org/packages/38/47/7776659aaa677046b77f527106e53ddd47373416d8fcdb1e1a881ec5dc06/cramjam-2.11.0-cp312-cp312-win32.whl", hash = "sha256:514e2c008a8b4fa823122ca3ecab896eac41d9aa0f5fc881bd6264486c204e32", size = 1603568, upload-time = "2025-07-27T21:22:50.084Z" },
+    { url = "https://files.pythonhosted.org/packages/75/b1/d53002729cfd94c5844ddfaf1233c86d29f2dbfc1b764a6562c41c044199/cramjam-2.11.0-cp312-cp312-win_amd64.whl", hash = "sha256:53fed080476d5f6ad7505883ec5d1ec28ba36c2273db3b3e92d7224fe5e463db", size = 1709287, upload-time = "2025-07-27T21:22:51.534Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/8b/406c5dc0f8e82385519d8c299c40fd6a56d97eca3fcd6f5da8dad48de75b/cramjam-2.11.0-cp313-cp313-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:2c289729cc1c04e88bafa48b51082fb462b0a57dbc96494eab2be9b14dca62af", size = 3553330, upload-time = "2025-07-27T21:22:53.124Z" },
+    { url = "https://files.pythonhosted.org/packages/00/ad/4186884083d6e4125b285903e17841827ab0d6d0cffc86216d27ed91e91d/cramjam-2.11.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:045201ee17147e36cf43d8ae2fa4b4836944ac672df5874579b81cf6d40f1a1f", size = 1859756, upload-time = "2025-07-27T21:22:54.821Z" },
+    { url = "https://files.pythonhosted.org/packages/54/01/91b485cf76a7efef638151e8a7d35784dae2c4ff221b1aec2c083e4b106d/cramjam-2.11.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:619cd195d74c9e1d2a3ad78d63451d35379c84bd851aec552811e30842e1c67a", size = 1693609, upload-time = "2025-07-27T21:22:56.331Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/84/d0c80d279b2976870fc7d10f15dcb90a3c10c06566c6964b37c152694974/cramjam-2.11.0-cp313-cp313-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6eb3ae5ab72edb2ed68bdc0f5710f0a6cad7fd778a610ec2c31ee15e32d3921e", size = 2024912, upload-time = "2025-07-27T21:22:57.915Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/70/88f2a5cb904281ed5d3c111b8f7d5366639817a5470f059bcd26833fc870/cramjam-2.11.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:df7da3f4b19e3078f9635f132d31b0a8196accb2576e3213ddd7a77f93317c20", size = 1760715, upload-time = "2025-07-27T21:22:59.528Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/06/cf5b02081132537d28964fb385fcef9ed9f8a017dd7d8c59d317e53ba50d/cramjam-2.11.0-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:57286b289cd557ac76c24479d8ecfb6c3d5b854cce54ccc7671f9a2f5e2a2708", size = 1853782, upload-time = "2025-07-27T21:23:01.07Z" },
+    { url = "https://files.pythonhosted.org/packages/57/27/63525087ed40a53d1867021b9c4858b80cc86274ffe7225deed067d88d92/cramjam-2.11.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:28952fbbf8b32c0cb7fa4be9bcccfca734bf0d0989f4b509dc7f2f70ba79ae06", size = 2032354, upload-time = "2025-07-27T21:23:03.021Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/ef/dbba082c6ebfb6410da4dd39a64e654d7194fcfd4567f85991a83fa4ec32/cramjam-2.11.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:78ed2e4099812a438b545dfbca1928ec825e743cd253bc820372d6ef8c3adff4", size = 2068007, upload-time = "2025-07-27T21:23:04.526Z" },
+    { url = "https://files.pythonhosted.org/packages/35/ce/d902b9358a46a086938feae83b2251720e030f06e46006f4c1fc0ac9da20/cramjam-2.11.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7d9aecd5c3845d415bd6c9957c93de8d93097e269137c2ecb0e5a5256374bdc8", size = 1977485, upload-time = "2025-07-27T21:23:06.058Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/03/982f54553244b0afcbdb2ad2065d460f0ab05a72a96896a969a1ca136a1e/cramjam-2.11.0-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:362fcf4d6f5e1242a4540812455f5a594949190f6fbc04f2ffbfd7ae0266d788", size = 2030447, upload-time = "2025-07-27T21:23:07.679Z" },
+    { url = "https://files.pythonhosted.org/packages/74/5f/748e54cdb665ec098ec519e23caacc65fc5ae58718183b071e33fc1c45b4/cramjam-2.11.0-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:13240b3dea41b1174456cb9426843b085dc1a2bdcecd9ee2d8f65ac5703374b0", size = 2154949, upload-time = "2025-07-27T21:23:09.366Z" },
+    { url = "https://files.pythonhosted.org/packages/69/81/c4e6cb06ed69db0dc81f9a8b1dc74995ebd4351e7a1877143f7031ff2700/cramjam-2.11.0-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:c54eed83726269594b9086d827decc7d2015696e31b99bf9b69b12d9063584fe", size = 2168925, upload-time = "2025-07-27T21:23:10.976Z" },
+    { url = "https://files.pythonhosted.org/packages/13/5b/966365523ce8290a08e163e3b489626c5adacdff2b3da9da1b0823dfb14e/cramjam-2.11.0-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:f8195006fdd0fc0a85b19df3d64a3ef8a240e483ae1dfc7ac6a4316019eb5df2", size = 2154950, upload-time = "2025-07-27T21:23:12.514Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/7d/7f8eb5c534b72b32c6eb79d74585bfee44a9a5647a14040bb65c31c2572d/cramjam-2.11.0-cp313-cp313-win32.whl", hash = "sha256:ccf30e3fe6d770a803dcdf3bb863fa44ba5dc2664d4610ba2746a3c73599f2e4", size = 1603199, upload-time = "2025-07-27T21:23:14.38Z" },
+    { url = "https://files.pythonhosted.org/packages/37/05/47b5e0bf7c41a3b1cdd3b7c2147f880c93226a6bef1f5d85183040cbdece/cramjam-2.11.0-cp313-cp313-win_amd64.whl", hash = "sha256:ee36348a204f0a68b03400f4736224e9f61d1c6a1582d7f875c1ca56f0254268", size = 1708924, upload-time = "2025-07-27T21:23:16.332Z" },
+    { url = "https://files.pythonhosted.org/packages/de/07/a1051cdbbe6d723df16d756b97f09da7c1adb69e29695c58f0392bc12515/cramjam-2.11.0-cp314-cp314-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:7ba5e38c9fbd06f086f4a5a64a1a5b7b417cd3f8fc07a20e5c03651f72f36100", size = 3554141, upload-time = "2025-07-27T21:23:17.938Z" },
+    { url = "https://files.pythonhosted.org/packages/74/66/58487d2e16ef3d04f51a7c7f0e69823e806744b4c21101e89da4873074bc/cramjam-2.11.0-cp314-cp314-macosx_10_12_x86_64.whl", hash = "sha256:b8adeee57b41fe08e4520698a4b0bd3cc76dbd81f99424b806d70a5256a391d3", size = 1860353, upload-time = "2025-07-27T21:23:19.593Z" },
+    { url = "https://files.pythonhosted.org/packages/67/b4/67f6254d166ffbcc9d5fa1b56876eaa920c32ebc8e9d3d525b27296b693b/cramjam-2.11.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:b96a74fa03a636c8a7d76f700d50e9a8bc17a516d6a72d28711225d641e30968", size = 1693832, upload-time = "2025-07-27T21:23:21.185Z" },
+    { url = "https://files.pythonhosted.org/packages/55/a3/4e0b31c0d454ae70c04684ed7c13d3c67b4c31790c278c1e788cb804fa4a/cramjam-2.11.0-cp314-cp314-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:c3811a56fa32e00b377ef79121c0193311fd7501f0fb378f254c7f083cc1fbe0", size = 2027080, upload-time = "2025-07-27T21:23:23.303Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/c7/5e8eed361d1d3b8be14f38a54852c5370cc0ceb2c2d543b8ba590c34f080/cramjam-2.11.0-cp314-cp314-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c5d927e87461f8a0d448e4ab5eb2bca9f31ca5d8ea86d70c6f470bb5bc666d7e", size = 1761543, upload-time = "2025-07-27T21:23:24.991Z" },
+    { url = "https://files.pythonhosted.org/packages/09/0c/06b7f8b0ce9fde89470505116a01fc0b6cb92d406c4fb1e46f168b5d3fa5/cramjam-2.11.0-cp314-cp314-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:f1f5c450121430fd89cb5767e0a9728ecc65997768fd4027d069cb0368af62f9", size = 1854636, upload-time = "2025-07-27T21:23:26.987Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/c6/6ebc02c9d5acdf4e5f2b1ec6e1252bd5feee25762246798ae823b3347457/cramjam-2.11.0-cp314-cp314-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:724aa7490be50235d97f07e2ca10067927c5d7f336b786ddbc868470e822aa25", size = 2032715, upload-time = "2025-07-27T21:23:28.603Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/77/a122971c23f5ca4b53e4322c647ac7554626c95978f92d19419315dddd05/cramjam-2.11.0-cp314-cp314-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:54c4637122e7cfd7aac5c1d3d4c02364f446d6923ea34cf9d0e8816d6e7a4936", size = 2069039, upload-time = "2025-07-27T21:23:30.319Z" },
+    { url = "https://files.pythonhosted.org/packages/19/0f/f6121b90b86b9093c066889274d26a1de3f29969d45c2ed1ecbe2033cb78/cramjam-2.11.0-cp314-cp314-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:17eb39b1696179fb471eea2de958fa21f40a2cd8bf6b40d428312d5541e19dc4", size = 1979566, upload-time = "2025-07-27T21:23:32.002Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/a3/f95bc57fd7f4166ce6da816cfa917fb7df4bb80e669eb459d85586498414/cramjam-2.11.0-cp314-cp314-musllinux_1_1_aarch64.whl", hash = "sha256:36aa5a798aa34e11813a80425a30d8e052d8de4a28f27bfc0368cfc454d1b403", size = 2030905, upload-time = "2025-07-27T21:23:33.696Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/52/e429de4e8bc86ee65e090dae0f87f45abd271742c63fb2d03c522ffde28a/cramjam-2.11.0-cp314-cp314-musllinux_1_1_armv7l.whl", hash = "sha256:449fca52774dc0199545fbf11f5128933e5a6833946707885cf7be8018017839", size = 2155592, upload-time = "2025-07-27T21:23:35.375Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/6c/65a7a0207787ad39ad804af4da7f06a60149de19481d73d270b540657234/cramjam-2.11.0-cp314-cp314-musllinux_1_1_i686.whl", hash = "sha256:d87d37b3d476f4f7623c56a232045d25bd9b988314702ea01bd9b4a94948a778", size = 2170839, upload-time = "2025-07-27T21:23:37.197Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/c5/5c5db505ba692bc844246b066e23901d5905a32baf2f33719c620e65887f/cramjam-2.11.0-cp314-cp314-musllinux_1_1_x86_64.whl", hash = "sha256:26cb45c47d71982d76282e303931c6dd4baee1753e5d48f9a89b3a63e690b3a3", size = 2157236, upload-time = "2025-07-27T21:23:38.854Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/22/88e6693e60afe98901e5bbe91b8dea193e3aa7f42e2770f9c3339f5c1065/cramjam-2.11.0-cp314-cp314-win32.whl", hash = "sha256:4efe919d443c2fd112fe25fe636a52f9628250c9a50d9bddb0488d8a6c09acc6", size = 1604136, upload-time = "2025-07-27T21:23:40.56Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/f8/01618801cd59ccedcc99f0f96d20be67d8cfc3497da9ccaaad6b481781dd/cramjam-2.11.0-cp314-cp314-win_amd64.whl", hash = "sha256:ccec3524ea41b9abd5600e3e27001fd774199dbb4f7b9cb248fcee37d4bda84c", size = 1710272, upload-time = "2025-07-27T21:23:42.236Z" },
+    { url = "https://files.pythonhosted.org/packages/40/81/6cdb3ed222d13ae86bda77aafe8d50566e81a1169d49ed195b6263610704/cramjam-2.11.0-cp314-cp314t-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:966ac9358b23d21ecd895c418c048e806fd254e46d09b1ff0cdad2eba195ea3e", size = 3559671, upload-time = "2025-07-27T21:23:44.504Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/43/52b7e54fe5ba1ef0270d9fdc43dabd7971f70ea2d7179be918c997820247/cramjam-2.11.0-cp314-cp314t-macosx_10_12_x86_64.whl", hash = "sha256:387f09d647a0d38dcb4539f8a14281f8eb6bb1d3e023471eb18a5974b2121c86", size = 1867876, upload-time = "2025-07-27T21:23:46.987Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/28/30d5b8d10acd30db3193bc562a313bff722888eaa45cfe32aa09389f2b24/cramjam-2.11.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:665b0d8fbbb1a7f300265b43926457ec78385200133e41fef19d85790fc1e800", size = 1695562, upload-time = "2025-07-27T21:23:48.644Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/86/ec806f986e01b896a650655024ea52a13e25c3ac8a3a382f493089483cdc/cramjam-2.11.0-cp314-cp314t-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:ca905387c7a371531b9622d93471be4d745ef715f2890c3702479cd4fc85aa51", size = 2025056, upload-time = "2025-07-27T21:23:50.404Z" },
+    { url = "https://files.pythonhosted.org/packages/09/43/c2c17586b90848d29d63181f7d14b8bd3a7d00975ad46e3edf2af8af7e1f/cramjam-2.11.0-cp314-cp314t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3c1aa56aef2c8af55a21ed39040a94a12b53fb23beea290f94d19a76027e2ffb", size = 1764084, upload-time = "2025-07-27T21:23:52.265Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/a9/68bc334fadb434a61df10071dc8606702aa4f5b6cdb2df62474fc21d2845/cramjam-2.11.0-cp314-cp314t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e5db59c1cdfaa2ab85cc988e602d6919495f735ca8a5fd7603608eb1e23c26d5", size = 1854859, upload-time = "2025-07-27T21:23:54.085Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/4e/b48e67835b5811ec5e9cb2e2bcba9c3fd76dab3e732569fe801b542c6ca9/cramjam-2.11.0-cp314-cp314t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b1f893014f00fe5e89a660a032e813bf9f6d91de74cd1490cdb13b2b59d0c9a3", size = 2035970, upload-time = "2025-07-27T21:23:55.758Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/70/d2ac33d572b4d90f7f0f2c8a1d60fb48f06b128fdc2c05f9b49891bb0279/cramjam-2.11.0-cp314-cp314t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c26a1eb487947010f5de24943bd7c422dad955b2b0f8650762539778c380ca89", size = 2069320, upload-time = "2025-07-27T21:23:57.494Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/4c/85cec77af4a74308ba5fca8e296c4e2f80ec465c537afc7ab1e0ca2f9a00/cramjam-2.11.0-cp314-cp314t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7d5c8bfb438d94e7b892d1426da5fc4b4a5370cc360df9b8d9d77c33b896c37e", size = 1982668, upload-time = "2025-07-27T21:23:59.126Z" },
+    { url = "https://files.pythonhosted.org/packages/55/45/938546d1629e008cc3138df7c424ef892719b1796ff408a2ab8550032e5e/cramjam-2.11.0-cp314-cp314t-musllinux_1_1_aarch64.whl", hash = "sha256:cb1fb8c9337ab0da25a01c05d69a0463209c347f16512ac43be5986f3d1ebaf4", size = 2034028, upload-time = "2025-07-27T21:24:00.865Z" },
+    { url = "https://files.pythonhosted.org/packages/01/76/b5a53e20505555f1640e66dcf70394bcf51a1a3a072aa18ea35135a0f9ed/cramjam-2.11.0-cp314-cp314t-musllinux_1_1_armv7l.whl", hash = "sha256:1f6449f6de52dde3e2f1038284910c8765a397a25e2d05083870f3f5e7fc682c", size = 2155513, upload-time = "2025-07-27T21:24:02.92Z" },
+    { url = "https://files.pythonhosted.org/packages/84/12/8d3f6ceefae81bbe45a347fdfa2219d9f3ac75ebc304f92cd5fcb4fbddc5/cramjam-2.11.0-cp314-cp314t-musllinux_1_1_i686.whl", hash = "sha256:382dec4f996be48ed9c6958d4e30c2b89435d7c2c4dbf32480b3b8886293dd65", size = 2170035, upload-time = "2025-07-27T21:24:04.558Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/85/3be6f0a1398f976070672be64f61895f8839857618a2d8cc0d3ab529d3dc/cramjam-2.11.0-cp314-cp314t-musllinux_1_1_x86_64.whl", hash = "sha256:d388bd5723732c3afe1dd1d181e4213cc4e1be210b080572e7d5749f6e955656", size = 2160229, upload-time = "2025-07-27T21:24:06.729Z" },
+    { url = "https://files.pythonhosted.org/packages/57/5e/66cfc3635511b20014bbb3f2ecf0095efb3049e9e96a4a9e478e4f3d7b78/cramjam-2.11.0-cp314-cp314t-win32.whl", hash = "sha256:0a70ff17f8e1d13f322df616505550f0f4c39eda62290acb56f069d4857037c8", size = 1610267, upload-time = "2025-07-27T21:24:08.428Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/c6/c71e82e041c95ffe6a92ac707785500aa2a515a4339c2c7dd67e3c449249/cramjam-2.11.0-cp314-cp314t-win_amd64.whl", hash = "sha256:028400d699442d40dbda02f74158c73d05cb76587a12490d0bfedd958fd49188", size = 1713108, upload-time = "2025-07-27T21:24:10.147Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/8f/82e35ec3c5387f1864f46b3c24bce89a07af8bb3ef242ae47281db2c1848/cramjam-2.11.0-pp310-pypy310_pp73-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:37bed927abc4a7ae2d2669baa3675e21904d8a038ed8e4313326ea7b3be62b2b", size = 3573104, upload-time = "2025-07-27T21:24:40.069Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/4e/0c821918080a32ba1e52c040e12dd02dada67728f07305c5f778b808a807/cramjam-2.11.0-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:50e4a58635fa8c6897d84847d6e065eb69f92811670fc5e9f2d9e3b6279a02b6", size = 1873441, upload-time = "2025-07-27T21:24:42.333Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/fd/848d077bf6abc4ce84273d8e3f3a70d61a2240519a339462f699d8acf829/cramjam-2.11.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:3d1ba626dd5f81f7f09bbf59f70b534e2b75e0d6582b056b7bd31b397f1c13e9", size = 1702589, upload-time = "2025-07-27T21:24:44.305Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/1c/899818999bbdb59c601756b413e87d37fd65875d1315346c10e367bb3505/cramjam-2.11.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c71e140d5eb3145d61d59d0be0bf72f07cc4cf4b32cb136b09f712a3b1040f5f", size = 1773646, upload-time = "2025-07-27T21:24:46.495Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/26/c2813c5422c43b3dcd8b6645bc359f08870737c44325ee4accc18f24eee0/cramjam-2.11.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a6ed7926a5cca28edebad7d0fedd2ad492710ae3524d25fc59a2b20546d9ce1", size = 1994179, upload-time = "2025-07-27T21:24:49.131Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/4f/af984f8d7f963f0301812cdd620ddcfd8276461ed7a786c0f89e82b14739/cramjam-2.11.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:5eb4ed3cea945b164b0513fd491884993acac2153a27b93a84019c522e8eda82", size = 1714790, upload-time = "2025-07-27T21:24:51.045Z" },
+    { url = "https://files.pythonhosted.org/packages/81/da/b3301962ccd6fce9fefa1ecd8ea479edaeaa38fadb1f34d5391d2587216a/cramjam-2.11.0-pp311-pypy311_pp73-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:52d5db3369f95b27b9f3c14d067acb0b183333613363ed34268c9e04560f997f", size = 3573546, upload-time = "2025-07-27T21:24:52.944Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/c2/410ddb8ad4b9dfb129284666293cb6559479645da560f7077dc19d6bee9e/cramjam-2.11.0-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:4820516366d455b549a44d0e2210ee7c4575882dda677564ce79092588321d54", size = 1873654, upload-time = "2025-07-27T21:24:54.958Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/99/f68a443c64f7ce7aff5bed369b0aa5b2fac668fa3dfd441837e316e97a1f/cramjam-2.11.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:d9e5db525dc0a950a825202f84ee68d89a072479e07da98795a3469df942d301", size = 1702846, upload-time = "2025-07-27T21:24:57.124Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/02/0ff358ab773def1ee3383587906c453d289953171e9c92db84fdd01bf172/cramjam-2.11.0-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:62ab4971199b2270005359cdc379bc5736071dc7c9a228581c5122d9ffaac50c", size = 1773683, upload-time = "2025-07-27T21:24:59.28Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/31/3298e15f87c9cf2aabdbdd90b153d8644cf989cb42a45d68a1b71e1f7aaf/cramjam-2.11.0-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:24758375cc5414d3035ca967ebb800e8f24604ececcba3c67d6f0218201ebf2d", size = 1994136, upload-time = "2025-07-27T21:25:01.565Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/90/20d1747255f1ee69a412e319da51ea594c18cca195e7a4d4c713f045eff5/cramjam-2.11.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:6c2eea545fef1065c7dd4eda991666fd9c783fbc1d226592ccca8d8891c02f23", size = 1714982, upload-time = "2025-07-27T21:25:05.79Z" },
+]
+
+[[package]]
+name = "csvkit"
+version = "1.5.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "agate" },
+    { name = "agate-dbf" },
+    { name = "agate-excel" },
+    { name = "agate-sql" },
+    { name = "openpyxl" },
+    { name = "sqlalchemy" },
+    { name = "xlrd" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/13/ae/df3f0a1d55a93eeae5be6e1537ecebc4e2c2ca038311e60e8b2e358a1221/csvkit-1.5.0.tar.gz", hash = "sha256:967a8be8fc58edf5621225b5a6a697b0e8730b962ea68085916a91860b22211c", size = 3811034, upload-time = "2024-03-28T15:27:22.731Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/aa/6d15ccff35e6cbfbfca0a4bbb8b6fa266a667561a9c9571bda8c6247ad5d/csvkit-1.5.0-py2.py3-none-any.whl", hash = "sha256:3f81e6f9af1774688ee30f4bdf8abd7f927995c85cad6491cf7478c658cfe7be", size = 44158, upload-time = "2024-03-28T15:27:20.926Z" },
+]
+
+[[package]]
+name = "cuda-bindings"
+version = "13.3.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cuda-pathfinder" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a9/21/8464d133752951c154feafb3b65c297e7d80f301183d220bec4c830f1441/cuda_bindings-13.3.1-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:120fcc53d57903df529c3486962c56528cba5b7d6c57c99537320ed9922c8b86", size = 6073403, upload-time = "2026-05-29T23:11:36.22Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/1f/5ef51f5fbaa5d4d3201bb3d7555af028ec1aa4416275ccbf73c9e34e3d2d/cuda_bindings-13.3.1-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9851b0caa8bfd3bc6fa054eaf57bea7c8e9c3a62db2d2621224677f49f3c53d0", size = 6675244, upload-time = "2026-05-29T23:11:38.664Z" },
+    { url = "https://files.pythonhosted.org/packages/51/6b/457ca12dad3ee9bfcc9a545cfd6b64b359ba49de40f776f6e028e678f262/cuda_bindings-13.3.1-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c5879712accf6e14bb01aa5e67440eb84998b8d104b509cc7a6dc0b8f656a474", size = 6053539, upload-time = "2026-05-29T23:11:43.19Z" },
+    { url = "https://files.pythonhosted.org/packages/95/7a/c5e3c34a409b148f5c0f5a4ea374158f95d488862c1dffedf9aa5c639df9/cuda_bindings-13.3.1-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:04436a9364059c84b8f9636f359eccda1cf814341f5b670c71d80d2f79dbc708", size = 6674166, upload-time = "2026-05-29T23:11:45.478Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/67/5e7dba1ba576dd73da5dee894ca076ca5e959450dfff66d6d510a255d1f7/cuda_bindings-13.3.1-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c7855c4868aabc0cfae28abbe83d56734bdfbd08f08fc234ac1912a12858bf49", size = 6025351, upload-time = "2026-05-29T23:11:49.685Z" },
+    { url = "https://files.pythonhosted.org/packages/39/2a/6d2e9047d1fb243dbaa364b01e0297534b9ed7fd27dba1c9f361519cf69b/cuda_bindings-13.3.1-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e32d08f71ebcdf00f0f41eab2eb37e8da94c8ed411cc9f7f7a019ce6b34abe3a", size = 6657965, upload-time = "2026-05-29T23:11:52.227Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/6e/2394f8163360f8391f8f1b7e72d300a82724edb81a7b7084c799fbd4c91f/cuda_bindings-13.3.1-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9efb21c1ee64981e184b9e0ba5eb3179e5ba3d4b51665a6cb52b8ef3d01a7cbf", size = 5920504, upload-time = "2026-05-29T23:11:56.883Z" },
+    { url = "https://files.pythonhosted.org/packages/34/c2/ef9b6a63f7dc432712a462c816662e662e00d38caa9b861c8c2588195d03/cuda_bindings-13.3.1-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2732904099e0a4d4db774a5fc6d91ee95fae065b4d2ecabb4968c5fe2406c9d7", size = 6476660, upload-time = "2026-05-29T23:11:59.188Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/81/bff68ce829999c1e4209c761bbf903b1c06ec570416ddb25020864ad5907/cuda_bindings-13.3.1-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1ab2f74ed65bfef4163ba07a8db16f1085e0729291db12a2423aff84ee8278b8", size = 6013639, upload-time = "2026-05-29T23:12:03.509Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/e0/c8a1f0c8f9ffdea4f5fe6dbab89b326cef4d85caf489dad39e209da89416/cuda_bindings-13.3.1-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:efd4c814d311ec08c981f6dded1dbe7d4b371067ee4f6c14cccec4bde9590f80", size = 6534419, upload-time = "2026-05-29T23:12:05.633Z" },
+    { url = "https://files.pythonhosted.org/packages/52/b8/83b1f563925b290f2d11a01a77a84013ba56052fe3653a5bef3ccfbb43d6/cuda_bindings-13.3.1-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c3c772dfff49681541d59630c90f858e173ac926b9c593a2b7123f2a1043cc76", size = 5809771, upload-time = "2026-05-29T23:12:10.422Z" },
+    { url = "https://files.pythonhosted.org/packages/12/20/e79b4bfe98f075195afb6343d41c498f9dbd2d161d7021d4d28bceb83581/cuda_bindings-13.3.1-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:36febb7c1079d68a981dbbd8d5a67235b399802b82075c9388624719607e52b9", size = 6358584, upload-time = "2026-05-29T23:12:12.767Z" },
+]
+
+[[package]]
+name = "cuda-pathfinder"
+version = "1.5.5"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/11/c8/26f2e4aae92f11522a96043892ba39a90eac610d5242523aa863212bc1c7/cuda_pathfinder-1.5.5-py3-none-any.whl", hash = "sha256:0228c023f95d1480f143ef5c8922d27a2ab052087a942e81dc289c9eb8f91689", size = 51671, upload-time = "2026-05-27T01:21:25.413Z" },
+]
+
+[[package]]
+name = "cuda-toolkit"
+version = "13.0.2"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/57/b2/453099f5f3b698d7d0eab38916aac44c7f76229f451709e2eb9db6615dcd/cuda_toolkit-13.0.2-py2.py3-none-any.whl", hash = "sha256:b198824cf2f54003f50d64ada3a0f184b42ca0846c1c94192fa269ecd97a66eb", size = 2364, upload-time = "2025-12-19T23:24:07.328Z" },
+]
+
+[package.optional-dependencies]
+cudart = [
+    { name = "nvidia-cuda-runtime", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+cufft = [
+    { name = "nvidia-cufft", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+cufile = [
+    { name = "nvidia-cufile", marker = "sys_platform == 'linux'" },
+]
+cupti = [
+    { name = "nvidia-cuda-cupti", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+curand = [
+    { name = "nvidia-curand", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+cusolver = [
+    { name = "nvidia-cusolver", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+cusparse = [
+    { name = "nvidia-cusparse", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+nvjitlink = [
+    { name = "nvidia-nvjitlink", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+nvrtc = [
+    { name = "nvidia-cuda-nvrtc", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+nvtx = [
+    { name = "nvidia-nvtx", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+
+[[package]]
+name = "dbfread"
+version = "2.0.7"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ad/ae/a5891681f5012724d062a4ca63ec2ff539c73d5804ba594e7e0e72099d3f/dbfread-2.0.7.tar.gz", hash = "sha256:07c8a9af06ffad3f6f03e8fe91ad7d2733e31a26d2b72c4dd4cfbae07ee3b73d", size = 33212, upload-time = "2016-11-25T11:25:01.408Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/4c/94/51349e43503e30ed7b4ecfe68a8809cdb58f722c0feb79d18b1f1e36fe74/dbfread-2.0.7-py2.py3-none-any.whl", hash = "sha256:f604def58c59694fa0160d7be5d0b8d594467278d2bb6a47d46daf7162c84cec", size = 20018, upload-time = "2016-11-25T11:27:36.001Z" },
+]
+
+[[package]]
+name = "et-xmlfile"
+version = "2.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d3/38/af70d7ab1ae9d4da450eeec1fa3918940a5fafb9055e934af8d6eb0c2313/et_xmlfile-2.0.0.tar.gz", hash = "sha256:dab3f4764309081ce75662649be815c4c9081e88f0837825f90fd28317d4da54", size = 17234, upload-time = "2024-10-25T17:25:40.039Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c1/8b/5fe2cc11fee489817272089c4203e679c63b570a5aaeb18d852ae3cbba6a/et_xmlfile-2.0.0-py3-none-any.whl", hash = "sha256:7a91720bc756843502c3b7504c77b8fe44217c85c537d85037f0f536151b2caa", size = 18059, upload-time = "2024-10-25T17:25:39.051Z" },
+]
+
+[[package]]
+name = "exceptiongroup"
+version = "1.3.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/50/79/66800aadf48771f6b62f7eb014e352e5d06856655206165d775e675a02c9/exceptiongroup-1.3.1.tar.gz", hash = "sha256:8b412432c6055b0b7d14c310000ae93352ed6754f70fa8f7c34141f91c4e3219", size = 30371, upload-time = "2025-11-21T23:01:54.787Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8a/0e/97c33bf5009bdbac74fd2beace167cab3f978feb69cc36f1ef79360d6c4e/exceptiongroup-1.3.1-py3-none-any.whl", hash = "sha256:a7a39a3bd276781e98394987d3a5701d0c4edffb633bb7a5144577f82c773598", size = 16740, upload-time = "2025-11-21T23:01:53.443Z" },
+]
+
+[[package]]
+name = "fastparquet"
+version = "2024.11.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cramjam" },
+    { name = "fsspec" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "packaging" },
+    { name = "pandas", version = "2.3.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "pandas", version = "3.0.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b4/66/862da14f5fde4eff2cedc0f51a8dc34ba145088e5041b45b2d57ac54f922/fastparquet-2024.11.0.tar.gz", hash = "sha256:e3b1fc73fd3e1b70b0de254bae7feb890436cb67e99458b88cb9bd3cc44db419", size = 467192, upload-time = "2024-11-15T19:30:10.413Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3d/56/476f5b83476a256489879b78513bee737691a80905e246a2daa30ebcc362/fastparquet-2024.11.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:60ccf587410f0979105e17036df61bb60e1c2b81880dc91895cdb4ee65b71e7f", size = 910272, upload-time = "2024-11-12T20:37:19.594Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/ad/4ce73440df874479f7205fe5445090f71ed4e9bd77fdb3b740253ce82703/fastparquet-2024.11.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a5ad5fc14b0567e700bea3cd528a0bd45a6f9371370b49de8889fb3d10a6574a", size = 684095, upload-time = "2024-11-12T20:37:22.957Z" },
+    { url = "https://files.pythonhosted.org/packages/20/37/c3164261d6183d529a59afef2749821b262c8581d837faa91043837c6f76/fastparquet-2024.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0b74333914f454344458dab9d1432fda9b70d62e28dc7acb1512d937ef1424ee", size = 1700355, upload-time = "2024-11-12T20:37:25.792Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/95/cf4b175c22160ec21e4664830763bfaa80b2cf05133ef854c3f436d01c16/fastparquet-2024.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41d1610130b5cb1ce36467766191c5418cba8631e2bfe3affffaf13f9be4e7a8", size = 1714663, upload-time = "2024-11-12T20:37:28.369Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/31/b6c8cdb6d5df964a192e4e8c8ecd979718afb9ca7e2dc9243a4368b370e9/fastparquet-2024.11.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d281edd625c33628ba028d3221180283d6161bc5ceb55eae1f0ca1678f864f26", size = 1666729, upload-time = "2024-11-12T20:37:30.243Z" },
+    { url = "https://files.pythonhosted.org/packages/31/e5/8a0575c46a7973849f8f2a88af16618b9c7efe98f249f03e3e3de69c2b86/fastparquet-2024.11.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fa56b19a29008c34cfe8831e810f770080debcbffc69aabd1df4d47572181f9c", size = 1741669, upload-time = "2024-11-12T20:37:32.067Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/6a/669f8c9cf2fc6e30c9353832f870e5a2e170b458d12c5080837f742d963d/fastparquet-2024.11.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5914ecfa766b7763201b9f49d832a5e89c2dccad470ca4f9c9b228d9a8349756", size = 1782359, upload-time = "2024-11-12T20:37:33.806Z" },
+    { url = "https://files.pythonhosted.org/packages/70/c0/1374cb43924739f4542e39d972481c1f4c7dd96808a1947450808e4e7df7/fastparquet-2024.11.0-cp310-cp310-win_amd64.whl", hash = "sha256:561202e8f0e859ccc1aa77c4aaad1d7901b2d50fd6f624ca018bae4c3c7a62ce", size = 670700, upload-time = "2024-11-12T20:37:35.312Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/51/e0d6e702523ac923ede6c05e240f4a02533ccf2cea9fec7a43491078e920/fastparquet-2024.11.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:374cdfa745aa7d5188430528d5841cf823eb9ad16df72ad6dadd898ccccce3be", size = 909934, upload-time = "2024-11-12T20:37:37.049Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/c8/5c0fb644c19a8d80b2ae4d8aa7d90c2d85d0bd4a948c5c700bea5c2802ea/fastparquet-2024.11.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4c8401bfd86cccaf0ab7c0ade58c91ae19317ff6092e1d4ad96c2178197d8124", size = 683844, upload-time = "2024-11-12T20:37:38.456Z" },
+    { url = "https://files.pythonhosted.org/packages/33/4a/1e532fd1a0d4d8af7ffc7e3a8106c0bcd13ed914a93a61e299b3832dd3d2/fastparquet-2024.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f9cca4c6b5969df5561c13786f9d116300db1ec22c7941e237cfca4ce602f59b", size = 1791698, upload-time = "2024-11-12T20:37:41.101Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/e8/e1ede861bea68394a755d8be1aa2e2d60a3b9f6b551bfd56aeca74987e2e/fastparquet-2024.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9a9387e77ac608d8978774caaf1e19de67eaa1386806e514dcb19f741b19cfe5", size = 1804289, upload-time = "2024-11-12T20:37:43.08Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/1e/957090cccaede805583ca3f3e46e2762d0f9bf8860ecbce65197e47d84c1/fastparquet-2024.11.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6595d3771b3d587a31137e985f751b4d599d5c8e9af9c4858e373fdf5c3f8720", size = 1753638, upload-time = "2024-11-12T20:37:45.498Z" },
+    { url = "https://files.pythonhosted.org/packages/85/72/344787c685fd1531f07ae712a855a7c34d13deaa26c3fd4a9231bea7dbab/fastparquet-2024.11.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:053695c2f730b78a2d3925df7cd5c6444d6c1560076af907993361cc7accf3e2", size = 1814407, upload-time = "2024-11-12T20:37:47.25Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/ec/ab9d5685f776a1965797eb68c4364c72edf57cd35beed2df49b34425d1df/fastparquet-2024.11.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0a52eecc6270ae15f0d51347c3f762703dd667ca486f127dc0a21e7e59856ae5", size = 1874462, upload-time = "2024-11-12T20:37:49.755Z" },
+    { url = "https://files.pythonhosted.org/packages/90/4f/7a4ea9a7ddf0a3409873f0787f355806f9e0b73f42f2acecacdd9a8eff0a/fastparquet-2024.11.0-cp311-cp311-win_amd64.whl", hash = "sha256:e29ff7a367fafa57c6896fb6abc84126e2466811aefd3e4ad4070b9e18820e54", size = 671023, upload-time = "2024-11-12T20:37:51.461Z" },
+    { url = "https://files.pythonhosted.org/packages/08/76/068ac7ec9b4fc783be21a75a6a90b8c0654da4d46934d969e524ce287787/fastparquet-2024.11.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:dbad4b014782bd38b58b8e9f514fe958cfa7a6c4e187859232d29fd5c5ddd849", size = 915968, upload-time = "2024-11-12T20:37:52.861Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/9e/6d3b4188ad64ed51173263c07109a5f18f9c84a44fa39ab524fca7420cda/fastparquet-2024.11.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:403d31109d398b6be7ce84fa3483fc277c6a23f0b321348c0a505eb098a041cb", size = 685399, upload-time = "2024-11-12T20:37:54.899Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/6c/809220bc9fbe83d107df2d664c3fb62fb81867be8f5218ac66c2e6b6a358/fastparquet-2024.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cbbb9057a26acf0abad7adf58781ee357258b7708ee44a289e3bee97e2f55d42", size = 1758557, upload-time = "2024-11-12T20:37:56.553Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/2c/b3b3e6ca2e531484289024138cd4709c22512b3fe68066d7f9849da4a76c/fastparquet-2024.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:63e0e416e25c15daa174aad8ba991c2e9e5b0dc347e5aed5562124261400f87b", size = 1781052, upload-time = "2024-11-12T20:37:58.339Z" },
+    { url = "https://files.pythonhosted.org/packages/21/fe/97ed45092d0311c013996dae633122b7a51c5d9fe8dcbc2c840dc491201e/fastparquet-2024.11.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0e2d7f02f57231e6c86d26e9ea71953737202f20e948790e5d4db6d6a1a150dc", size = 1715797, upload-time = "2024-11-12T20:38:00.694Z" },
+    { url = "https://files.pythonhosted.org/packages/24/df/02fa6aee6c0d53d1563b5bc22097076c609c4c5baa47056b0b4bed456fcf/fastparquet-2024.11.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:fbe4468146b633d8f09d7b196fea0547f213cb5ce5f76e9d1beb29eaa9593a93", size = 1795682, upload-time = "2024-11-12T20:38:02.38Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/25/f4f87557589e1923ee0e3bebbc84f08b7c56962bf90f51b116ddc54f2c9f/fastparquet-2024.11.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:29d5c718817bcd765fc519b17f759cad4945974421ecc1931d3bdc3e05e57fa9", size = 1857842, upload-time = "2024-11-12T20:38:04.196Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/f9/98cd0c39115879be1044d59c9b76e8292776e99bb93565bf990078fd11c4/fastparquet-2024.11.0-cp312-cp312-win_amd64.whl", hash = "sha256:74a0b3c40ab373442c0fda96b75a36e88745d8b138fcc3a6143e04682cbbb8ca", size = 673269, upload-time = "2024-12-11T21:22:48.073Z" },
+    { url = "https://files.pythonhosted.org/packages/47/e3/e7db38704be5db787270d43dde895eaa1a825ab25dc245e71df70860ec12/fastparquet-2024.11.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:59e5c5b51083d5b82572cdb7aed0346e3181e3ac9d2e45759da2e804bdafa7ee", size = 912523, upload-time = "2024-11-12T20:38:06.003Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/66/e3387c99293dae441634e7724acaa425b27de19a00ee3d546775dace54a9/fastparquet-2024.11.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:bdadf7b6bad789125b823bfc5b0a719ba5c4a2ef965f973702d3ea89cff057f6", size = 683779, upload-time = "2024-11-12T20:38:07.442Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/21/d112d0573d086b578bf04302a502e9a7605ea8f1244a7b8577cd945eec78/fastparquet-2024.11.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46b2db02fc2a1507939d35441c8ab211d53afd75d82eec9767d1c3656402859b", size = 1751113, upload-time = "2024-11-12T20:38:09.36Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/a7/040507cee3a7798954e8fdbca21d2dbc532774b02b882d902b8a4a6849ef/fastparquet-2024.11.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a3afdef2895c9f459135a00a7ed3ceafebfbce918a9e7b5d550e4fae39c1b64d", size = 1780496, upload-time = "2024-11-12T20:38:11.022Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/75/d0d9f7533d780ec167eede16ad88073ee71696150511126c31940e7f73aa/fastparquet-2024.11.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:36b5c9bd2ffaaa26ff45d59a6cefe58503dd748e0c7fad80dd905749da0f2b9e", size = 1713608, upload-time = "2024-11-12T20:38:12.848Z" },
+    { url = "https://files.pythonhosted.org/packages/30/fa/1d95bc86e45e80669c4f374b2ca26a9e5895a1011bb05d6341b4a7414693/fastparquet-2024.11.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:6b7df5d3b61a19d76e209fe8d3133759af1c139e04ebc6d43f3cc2d8045ef338", size = 1792779, upload-time = "2024-11-12T20:38:14.5Z" },
+    { url = "https://files.pythonhosted.org/packages/13/3d/c076beeb926c79593374c04662a9422a76650eef17cd1c8e10951340764a/fastparquet-2024.11.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8b35823ac7a194134e5f82fa4a9659e42e8f9ad1f2d22a55fbb7b9e4053aabbb", size = 1851322, upload-time = "2024-11-12T20:38:16.231Z" },
+    { url = "https://files.pythonhosted.org/packages/09/5a/1d0d47e64816002824d4a876644e8c65540fa23f91b701f0daa726931545/fastparquet-2024.11.0-cp313-cp313-win_amd64.whl", hash = "sha256:d20632964e65530374ff7cddd42cc06aa0a1388934903693d6d22592a5ba827b", size = 673266, upload-time = "2024-11-12T20:38:17.661Z" },
+]
+
+[[package]]
+name = "filelock"
+version = "3.29.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/91/f5/3557bf28e0f1943e4849154c821533706e6dea010f96fb6aa0b6949037d1/filelock-3.29.3.tar.gz", hash = "sha256:7fc1b3f39cf172fd8203812043c57b8a65aef9969f38b6704f628b881f761a84", size = 61956, upload-time = "2026-06-10T17:37:11.832Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/81/8f/b61d427c4f49a8bdadc93f4e7e74df8a6df6f77ee6e26bf0df53d3925363/filelock-3.29.3-py3-none-any.whl", hash = "sha256:e58333029cc9b925f39aad59b1d8f0a1ad836af4e60d7217f4a4dba87461261d", size = 42324, upload-time = "2026-06-10T17:37:10.37Z" },
+]
+
+[[package]]
+name = "flatbuffers"
+version = "25.12.19"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e8/2d/d2a548598be01649e2d46231d151a6c56d10b964d94043a335ae56ea2d92/flatbuffers-25.12.19-py2.py3-none-any.whl", hash = "sha256:7634f50c427838bb021c2d66a3d1168e9d199b0607e6329399f04846d42e20b4", size = 26661, upload-time = "2025-12-19T23:16:13.622Z" },
+]
+
+[[package]]
+name = "fsspec"
+version = "2026.4.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d5/8d/1c51c094345df128ca4a990d633fe1a0ff28726c9e6b3c41ba65087bba1d/fsspec-2026.4.0.tar.gz", hash = "sha256:301d8ac70ae90ef3ad05dcf94d6c3754a097f9b5fe4667d2787aa359ec7df7e4", size = 312760, upload-time = "2026-04-29T20:42:38.635Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d5/0c/043d5e551459da400957a1395e0febbf771446ff34291afcbe3d8be2a279/fsspec-2026.4.0-py3-none-any.whl", hash = "sha256:11ef7bb35dab8a394fde6e608221d5cf3e8499401c249bebaeaad760a1a8dec2", size = 203402, upload-time = "2026-04-29T20:42:36.842Z" },
+]
+
+[[package]]
+name = "greenlet"
+version = "3.5.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6d/6e/802acd792aebb2256fbbee8cacf2727faaeb6f240ac11008f09eae4414bc/greenlet-3.5.1.tar.gz", hash = "sha256:5a56aeb7d5d9cc4b3a735efb5095bd4b4f6f0e4f93e5ca876d0e2315137b7829", size = 197356, upload-time = "2026-05-20T15:05:03.917Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1d/21/117c8710abb7f146d804a124c07eb5964a60b90d02b72452885aecc18efa/greenlet-3.5.1-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:7eacb17a9d41538a2bc4912eba5ef13823c83cb69e4d141d0813debe7163187f", size = 283510, upload-time = "2026-05-20T13:12:26.475Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/f7/6762a56fa5f6c2295c449c6524e10ce481e381c994cc44d9d03aef0700fb/greenlet-3.5.1-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e5cc9606aa5f4e0bde0d3bd502b44f743864c3ffa5cfa1011b1e30f5aa02366f", size = 599696, upload-time = "2026-05-20T14:00:02.906Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/05/85a511e68ee109aff0aa00b4b497806091dd2d82ce209e49c6e801bd5d92/greenlet-3.5.1-cp310-cp310-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c3d35f87c7253b715d13d679e0783d845910144f282cb939fe1ba4ac8616269c", size = 612618, upload-time = "2026-05-20T14:05:39.202Z" },
+    { url = "https://files.pythonhosted.org/packages/89/b8/8b83d18ae07c46c019617f35afd7b47aab7f9b4fbb12fc637d681e10bdd8/greenlet-3.5.1-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:540dae7b956209af4d70a3be35927b4055f617763771e5e84a5255bea934d2f5", size = 612947, upload-time = "2026-05-20T13:14:23.469Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/14/ad1f9fc9b82384c010212464a3702bd911f95dab2f1180bc6fbcfb1f958c/greenlet-3.5.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ed8cdb691169715a9a492844a83246f090182247d1a5031dc78a403f68ba1e97", size = 1571425, upload-time = "2026-05-20T14:02:22.671Z" },
+    { url = "https://files.pythonhosted.org/packages/46/1c/43b8203cf10f4292c9e3d270e9e5f5ade79115a0a0ca5ea6f1be5f8915a7/greenlet-3.5.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:9d59e840387076a51016777a9328b3f2c427c6f9208a6e958bad251be50a648d", size = 1638688, upload-time = "2026-05-20T13:14:30.026Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/6e/0344b1e99f58f71715456e46492101fd2daa408957b8186ade0a4b515da7/greenlet-3.5.1-cp310-cp310-win_amd64.whl", hash = "sha256:b9152fca4a6466e114aaec745ae61cba739903a109754a9d4e1262f01e9259b1", size = 237763, upload-time = "2026-05-20T13:11:35.659Z" },
+    { url = "https://files.pythonhosted.org/packages/42/3c/ff890b466eaba2b0f5e6bdfff025f8c75f41b8ffdc3dbc3d24ad261e764a/greenlet-3.5.1-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:73f78f9b9f0a5c06e5c946ba1e8e36f5114923b6be109ee618c54f079c3ea14f", size = 284764, upload-time = "2026-05-20T13:09:10.204Z" },
+    { url = "https://files.pythonhosted.org/packages/81/0e/5e5457be3d256918f6a4756f073548a3f0190836e2cc94aa6d0d617a940b/greenlet-3.5.1-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a0cbed8bb44e23c5b199f888f4e4ce096b45ad9f25ff74a7ad0213875e936bb2", size = 603479, upload-time = "2026-05-20T14:00:04.757Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/e1/f89a21d58d308298e6f275f13a1b472ed96c680b601a371b08be6a725989/greenlet-3.5.1-cp311-cp311-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a203a8bd0acb0701653d3bbb26e404854a68674139ed5cbb778830f42b09bb33", size = 615495, upload-time = "2026-05-20T14:05:40.87Z" },
+    { url = "https://files.pythonhosted.org/packages/75/de/af6cef182862d2ccd6975440d21c9058a77c3f9b469abf94e322dfd2e0e3/greenlet-3.5.1-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a271fcd66c74615cda6a964fda3f304267a12e50a084472218a39bb0376f563", size = 614754, upload-time = "2026-05-20T13:14:24.947Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/c6/50e520283a9f19388a7326b05f9e8637e566003475eacaadad04f558c68d/greenlet-3.5.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ded7b068c7c31c1a8657d4fd42d886b3e051ae29f88b80c5ff9d502257b0f071", size = 1574097, upload-time = "2026-05-20T14:02:24.003Z" },
+    { url = "https://files.pythonhosted.org/packages/21/1c/13abd1f4860d987fa5e1170a01930d6e6cd40d328de487a3c9fdaff0ffd0/greenlet-3.5.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:d0932b81d72f552ded9d810d00021b64d89f2195a91ce115b893f943b7a4ab3c", size = 1641058, upload-time = "2026-05-20T13:14:31.83Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/56/5f332b7705545eac2dc01b4e9254d24a793f2656d55d5cc6b94ee59d22ae/greenlet-3.5.1-cp311-cp311-win_amd64.whl", hash = "sha256:88e300d136eac057b2397aa1cfd7328b4c87c7eb66a09c7bc6a1292234db474e", size = 238089, upload-time = "2026-05-20T13:14:03.229Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/a9/a3c2fa886c5b94863fb0e61b3bc14610b7aa94cf4f17f8741b11708305fc/greenlet-3.5.1-cp311-cp311-win_arm64.whl", hash = "sha256:cc6ab7e555c8a112ad3a76e368e86e12a2754bcae1652a5602e133ec7b635523", size = 234989, upload-time = "2026-05-20T13:08:27.715Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/37/4549f149c9797c21b32c2683c33522af22522099de128b2406672526d005/greenlet-3.5.1-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:fa4f98af3a528f0c3fd592a26df7f376f93329c8f4d987f6bb979057af8bf5e2", size = 286220, upload-time = "2026-05-20T13:07:28.463Z" },
+    { url = "https://files.pythonhosted.org/packages/38/ff/a4f436709716965eaab9f36ea7b906c8a927fbe32fb1372a2071d964f6b1/greenlet-3.5.1-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ffea73584b216150eab159b6d12348fb253e68757974de1e2c40d8a318ac89ed", size = 601585, upload-time = "2026-05-20T14:00:06.141Z" },
+    { url = "https://files.pythonhosted.org/packages/65/ad/54bc3fcee3ad368a61b19b67d88117f7a8c29727bf71fffdeda81fbd946e/greenlet-3.5.1-cp312-cp312-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:1072b4f9edcc1e192d9283a66a3e68d6b84c561de33a83d7858beb9ba1effe10", size = 614215, upload-time = "2026-05-20T14:05:42.675Z" },
+    { url = "https://files.pythonhosted.org/packages/40/69/b91cda0647df839483201545913514c2827ebea5e5ccdf931842763bc127/greenlet-3.5.1-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:add5217d68b31130f0beca584d7fef4878327d2e31642b66618a14eef312b63b", size = 611358, upload-time = "2026-05-20T13:14:26.37Z" },
+    { url = "https://files.pythonhosted.org/packages/59/90/3cf77e080350cd02fa307bb2abf05df48f4482c240275bbd2c203ba8bb1c/greenlet-3.5.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a5ea42a752d47a145eae922b605cd1634665ac3d5ec1e72402d5048e8d60d207", size = 1570475, upload-time = "2026-05-20T14:02:25.29Z" },
+    { url = "https://files.pythonhosted.org/packages/65/2c/18cece62045e74598c3c393f70dce4a63f56222015ba29a5d4eeb04f764c/greenlet-3.5.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:c5551170cf4f5ff5623e9af81323751979fee2c731e2287b61f73cd27257b823", size = 1635625, upload-time = "2026-05-20T13:14:34.027Z" },
+    { url = "https://files.pythonhosted.org/packages/30/f5/310d104ddf41eb5a70f4c268d22508dfb0c3c8e86fec152be34d0d2ed819/greenlet-3.5.1-cp312-cp312-win_amd64.whl", hash = "sha256:3c8bb982ad117d29478ef8f5533e97df21f1e2befd17a299257b0c96d1371c0b", size = 238791, upload-time = "2026-05-20T13:10:39.018Z" },
+    { url = "https://files.pythonhosted.org/packages/62/90/ceca11f504cd23a8047a3dea31919adc48df9b626dd0c13f0d858734fdfd/greenlet-3.5.1-cp312-cp312-win_arm64.whl", hash = "sha256:80eb4b04dadc4e67df3fae179a32c4706a3f495bc7f22fc8a81115d5f5512188", size = 235580, upload-time = "2026-05-20T13:08:45.056Z" },
+    { url = "https://files.pythonhosted.org/packages/27/69/7f7e5372d998b81001899b1c0823c957aa413ba0f2662e65821611cc31e4/greenlet-3.5.1-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:51518ff74664078fc51bffcc6fc529b0df5ae58da192691cee765d45ce944a2b", size = 285060, upload-time = "2026-05-20T13:08:51.899Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/bf/387f9b6b865fd2ae0d0be09e0004827295a01b71be76ed350dd1e28a91a4/greenlet-3.5.1-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1ffdb3c0bb002c99cd8f298957e046c3dbf6006b5b7cdf11a4e19194624a0a0a", size = 604370, upload-time = "2026-05-20T14:00:07.492Z" },
+    { url = "https://files.pythonhosted.org/packages/32/f5/169ce3d4e4c67291bd18f8cbe0299c9f3e45102c7f1fb3c14780c93e4532/greenlet-3.5.1-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7715a5a2c3378ba602c3a440558261e13a820bb53a82693aacd7b7f6d964e283", size = 616987, upload-time = "2026-05-20T14:05:44.237Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/e5/7f2e41d5273be07e77560d61ea4e56485b4d6c316d2a84518c62d1364061/greenlet-3.5.1-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:dc71ff466927a201b08305acac451ebe1aedfcea002f62f1f2f2ac2ac1e6a135", size = 613911, upload-time = "2026-05-20T13:14:27.539Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/a4/fbdc67579b73615a1f91615e814303cc71e06128f7baaba87be79b8fb90c/greenlet-3.5.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:cd443683db272ebaaca03af98c0b063ab30db70ea8a31a1559f35e3f7b744ccd", size = 1570689, upload-time = "2026-05-20T14:02:27.225Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/b4/77abbe35078be39718a46cd49caf16bceb35662f97a34101dca28aa98e47/greenlet-3.5.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:089fff7a6ce8d9316d1f65ebc00273a56be258c1725b32b94de90a3a979557e1", size = 1635602, upload-time = "2026-05-20T13:14:36.344Z" },
+    { url = "https://files.pythonhosted.org/packages/37/f7/129f27ca700845b8ee8ca88ce7f43435a1239c2eddb7677fc938822762cf/greenlet-3.5.1-cp313-cp313-win_amd64.whl", hash = "sha256:110a1ca7b49b014b097f6078272c3f4ed31af45b254de5228b79adba879f6af9", size = 238683, upload-time = "2026-05-20T13:11:50.57Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/5c/a485a36e87df8d8fd0632ee01511244f5156a20ed3746cc6599340326395/greenlet-3.5.1-cp313-cp313-win_arm64.whl", hash = "sha256:f16ba1efc0715b680a18b8123d90dad887c6112ae3555b4b5c32c149540c6b4e", size = 235499, upload-time = "2026-05-20T13:12:42.028Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/cb/c62454606daf5640369c94d8a9dd540599b1bfc090e2d2180cb77f4038d2/greenlet-3.5.1-cp314-cp314-macosx_11_0_universal2.whl", hash = "sha256:d8ab31c9de8651a2facdd5c5bb0011f2380dd1a7af78ce2adf4b56095294fc07", size = 285579, upload-time = "2026-05-20T13:08:56.396Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/71/c4270398c2eba968a6071af1dfbdcaeee6ec1c24bc8b435b8cc452700da6/greenlet-3.5.1-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5e300185139abc337ade480c327183adf42a875ac7181bfe66d7d4efea31fbea", size = 651106, upload-time = "2026-05-20T14:00:09.448Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/ab/71e34b78a44ec271fb5f550c17bc46d301ddc5953890d935f270b0dcdb5a/greenlet-3.5.1-cp314-cp314-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7ffdb990dcaa0234cf9845aead5df2e3c3a8b6507d409274dd87e0d5ab05ffc2", size = 663478, upload-time = "2026-05-20T14:05:45.88Z" },
+    { url = "https://files.pythonhosted.org/packages/77/96/4efd6fa5c62c85426a0c19077a586258ebc3a2a146ff2493e4312a697a22/greenlet-3.5.1-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2f82b3597e9d83b63408affed0b48fd0f54935edac4302237b9a837be0dae33c", size = 660800, upload-time = "2026-05-20T13:14:29.129Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/e0/6c71401a25cac7000261304e866a2f2cc04dc74810d40e2f118aa4799495/greenlet-3.5.1-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:c0141e37414c10164e702b8fb1473304221ad98f71600850c6ef7ff4880feba0", size = 1617518, upload-time = "2026-05-20T14:02:28.662Z" },
+    { url = "https://files.pythonhosted.org/packages/41/26/c5c06643e8c0af9e7bf18e16cb51d0ab7625155f0392e1c9015d66d556cd/greenlet-3.5.1-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:50ae25a67bea74ea41fb14b960bc532df73eb713417b2d61892dced82fe8d3bc", size = 1681593, upload-time = "2026-05-20T13:14:39.417Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/bd/e11a108317485075e68af9d23039619b86b28130c3b50d227d42edece64b/greenlet-3.5.1-cp314-cp314-win_amd64.whl", hash = "sha256:8a17c42330e261299766b75ac1ea32caa437a9453c8f65d16a13140db378ecd3", size = 239800, upload-time = "2026-05-20T13:09:30.128Z" },
+    { url = "https://files.pythonhosted.org/packages/47/f8/8e8e8417b7bf28639a5a56356ef934d0375e1d0c70a57e04d7701e870ffe/greenlet-3.5.1-cp314-cp314-win_arm64.whl", hash = "sha256:7b5f5fae05b8ac6d176a61b60c394a8cbdc2b5b91b81793066e68745cf165e54", size = 236862, upload-time = "2026-05-20T13:09:10.498Z" },
+    { url = "https://files.pythonhosted.org/packages/90/12/41bf27fde4d3605d3773ae57751eda182b8be2f5398011c041173b1d9534/greenlet-3.5.1-cp314-cp314t-macosx_11_0_universal2.whl", hash = "sha256:ea8da1e900d758d078810d4255d8c6aa572181896a31ec79d779eb79c3adc9ad", size = 293637, upload-time = "2026-05-20T13:12:35.529Z" },
+    { url = "https://files.pythonhosted.org/packages/44/44/ba14b23e9757707050c2f397d305bbcae62e5d7cad122f8b6baec5ae4a1f/greenlet-3.5.1-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a19570c52a21420dcbc94e661994bc325c0b5b11304540fed514586da5dc8f2e", size = 650840, upload-time = "2026-05-20T14:00:11.079Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/37/5ddc2b686a6844f91abecef43411842426da2e1573f60b49ecf2547f4ae1/greenlet-3.5.1-cp314-cp314t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3d955c89b75eeca4723d7cc14135f393cd47c32e2a6cb4a8e4c6e760a26b0986", size = 656416, upload-time = "2026-05-20T14:05:47.118Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/f0/d17510297c35a2992712f0bf84de3779749999f7d3d63aa1f09db7c62dbe/greenlet-3.5.1-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:de2daaaebd1a5aa88c49045b6baf9310b3263796bd88db713edf37cf53e7bb4e", size = 654397, upload-time = "2026-05-20T13:14:30.696Z" },
+    { url = "https://files.pythonhosted.org/packages/37/eb/147387705bb89092645b012586e7273cb5ed3c90ef7eaf3a69173eaf0209/greenlet-3.5.1-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:3bfbd69cc349e43bf3a8ae1c85548ff0718efc887615c2db16c3833d7b0b072d", size = 1614469, upload-time = "2026-05-20T14:02:30.192Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/4e/37ee0da7732b7aa9896f17e15579a9df34b9fcb9dd494f0adfa749af6623/greenlet-3.5.1-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:4378720dd888136c27215a0214d32a4d37c3852765d45bc37aad0623423cfd78", size = 1675115, upload-time = "2026-05-20T13:14:40.972Z" },
+    { url = "https://files.pythonhosted.org/packages/57/f3/97dfcf4a6eb5077f8a672234216fb5923eb89f2cab7081cb10b2cf75b605/greenlet-3.5.1-cp314-cp314t-win_amd64.whl", hash = "sha256:45718441607f9325d948db98cbc691276059316d0358c188c246da4e1d4d23d2", size = 245246, upload-time = "2026-05-20T13:12:22.646Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/73/d7f72e34b582f694f4a9b248162db7b09cc458a259ba8f0c0bfa1a34ea7d/greenlet-3.5.1-cp315-cp315-macosx_11_0_universal2.whl", hash = "sha256:2baee5ca02031757ffe8cc3d69f0cc0aec7065ce362622da74f32d3bcab1c541", size = 285575, upload-time = "2026-05-20T13:12:07.043Z" },
+    { url = "https://files.pythonhosted.org/packages/df/59/fa9c6e87dc8ad27a95dabe2f29f372b733d05a8a67470f6c901ed9975655/greenlet-3.5.1-cp315-cp315-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9b1ec3274918a81d3ea778b9e75b56b72b33f300edb6cf7f3a7fe1dae56683de", size = 656428, upload-time = "2026-05-20T14:00:12.556Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/f9/e753408871eaa61dfe35e619cfc67512b036fde99893685d50eea9e07146/greenlet-3.5.1-cp315-cp315-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:111e2390ffffc47d5840b01711dd7fac07d4c09283d0283e7f3264b14e284c64", size = 667064, upload-time = "2026-05-20T14:05:48.662Z" },
+    { url = "https://files.pythonhosted.org/packages/96/27/5565b5b40389f1c7753003a07e21892fda8660926787036d5bc0308b8113/greenlet-3.5.1-cp315-cp315-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e630136e905fe5ff43e86945ae41220b6d1470956a39220e708110ac48d01ea5", size = 665697, upload-time = "2026-05-20T13:14:32.943Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/82/e7de4178c0c2d1c9a5a3be3cc0b33e46a85b3ee4a77c071bf7ad8600e079/greenlet-3.5.1-cp315-cp315-musllinux_1_2_aarch64.whl", hash = "sha256:975eac34b44a7077ca4d421348455b94f0f518246a7f14bc6d2fdcfe5b584368", size = 1621256, upload-time = "2026-05-20T14:02:31.91Z" },
+    { url = "https://files.pythonhosted.org/packages/00/10/f2dddcf7dacac17dfc68691809589adad06135eb28930429cf58a6467a2f/greenlet-3.5.1-cp315-cp315-musllinux_1_2_x86_64.whl", hash = "sha256:9ab3c3a0b2ae6198e67c898dad5215a49f9ae0d0081b3c3ec59f333e39eeca26", size = 1685956, upload-time = "2026-05-20T13:14:42.55Z" },
+    { url = "https://files.pythonhosted.org/packages/22/17/4a232b32133230ada52f70e9d7f5b65b0caef8772f01849bd8d149e7e4ca/greenlet-3.5.1-cp315-cp315-win_amd64.whl", hash = "sha256:cbfc69be86e10dcfef5b1e6269d1d6926552aa89ee39e1de3353360c1b6989ab", size = 239802, upload-time = "2026-05-20T13:13:15.481Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/ae/4e623a7e6d4d2a5f4cb8e4c82de4169fc637942caae68d6e676b8a128ac5/greenlet-3.5.1-cp315-cp315-win_arm64.whl", hash = "sha256:92fd6d44ac5e5a887c8a5dc4a8ba0ba908527c31c12f78c6bc7dcfe8aab279f6", size = 236853, upload-time = "2026-05-20T13:15:37.301Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/57/816d9cff29119da3505b3d6a5e14a8af89006ac36f47f891ff293ee05af1/greenlet-3.5.1-cp315-cp315t-macosx_11_0_universal2.whl", hash = "sha256:a6fdf2433a5441ef9a95464f7c3e674775da1c8c1177fff311cee1acad4626ed", size = 293877, upload-time = "2026-05-20T13:10:19.078Z" },
+    { url = "https://files.pythonhosted.org/packages/23/a1/59b0a7c7d140ff1a75626680b9a9899b79a9176cab298b394968fb023295/greenlet-3.5.1-cp315-cp315t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:7546556f0d649f99f6a361098a55f761181bb2ea12ff150bb16d26092ad88244", size = 655333, upload-time = "2026-05-20T14:00:14.758Z" },
+    { url = "https://files.pythonhosted.org/packages/72/1b/5efe127597625042218939d01855109f352779050768b670b52edcc16a6c/greenlet-3.5.1-cp315-cp315t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:d5ee3ea898009fa898f85f9982255d35278c477bebe185beca249cab42d4526c", size = 659443, upload-time = "2026-05-20T14:05:50.159Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/6d/c404246ea4d22d097a7426d0efb5b781bd7eb67715f09e79001bd552ab18/greenlet-3.5.1-cp315-cp315t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a5c81f74d204d3edd136ebfd50dce53acbb776995d721a0fe801626cfc93b8cd", size = 658356, upload-time = "2026-05-20T13:14:35.091Z" },
+    { url = "https://files.pythonhosted.org/packages/51/02/f8ee37fb6d2219329f350af241c27fcf12df57e723d11f6fc6d3bacdadaa/greenlet-3.5.1-cp315-cp315t-musllinux_1_2_aarch64.whl", hash = "sha256:2c18ef16bf6d4dd410e4dd52996888ea1497be26892fe5bbc73580aba4287b8e", size = 1619216, upload-time = "2026-05-20T14:02:33.403Z" },
+    { url = "https://files.pythonhosted.org/packages/93/c5/3dc9475ace2c7a3680da12372cddd7f1ac874eb410a1ac48d3e9dab83782/greenlet-3.5.1-cp315-cp315t-musllinux_1_2_x86_64.whl", hash = "sha256:17d86354f0ae6b61bf9be5148d0dd34e06c3cb7c602c671f79f29ac3b150e659", size = 1678427, upload-time = "2026-05-20T13:14:43.71Z" },
+    { url = "https://files.pythonhosted.org/packages/df/4e/750c15c317a41ffb36f0bf40b933e3d744a7dede61889f74443ea69690cf/greenlet-3.5.1-cp315-cp315t-win_amd64.whl", hash = "sha256:e7516cf6ae6b8a582c2770a0caed47b8a48373ed732c33d69a72913ae6ac923e", size = 245225, upload-time = "2026-05-20T13:13:59.366Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/fd/d3baea2eeb7b617efd47e87ca06e2ec2c6118d303aa9e918e0ce16eadc10/greenlet-3.5.1-cp315-cp315t-win_arm64.whl", hash = "sha256:5028648bf2253ec4745add746129d3904121fa7fe871a76bed23c5720573ce0a", size = 239590, upload-time = "2026-05-20T13:13:37.382Z" },
+]
+
+[[package]]
+name = "iniconfig"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/34/14ca021ce8e5dfedc35312d08ba8bf51fdd999c576889fc2c24cb97f4f10/iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730", size = 20503, upload-time = "2025-10-18T21:55:43.219Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cb/b1/3846dd7f199d53cb17f49cba7e651e9ce294d8497c8c150530ed11865bb8/iniconfig-2.3.0-py3-none-any.whl", hash = "sha256:f631c04d2c48c52b84d0d0549c99ff3859c98df65b3101406327ecc7d53fbf12", size = 7484, upload-time = "2025-10-18T21:55:41.639Z" },
+]
+
+[[package]]
+name = "isodate"
+version = "0.7.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/54/4d/e940025e2ce31a8ce1202635910747e5a87cc3a6a6bb2d00973375014749/isodate-0.7.2.tar.gz", hash = "sha256:4cd1aa0f43ca76f4a6c6c0292a85f40b35ec2e43e315b59f06e6d32171a953e6", size = 29705, upload-time = "2024-10-08T23:04:11.5Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/15/aa/0aca39a37d3c7eb941ba736ede56d689e7be91cab5d9ca846bde3999eba6/isodate-0.7.2-py3-none-any.whl", hash = "sha256:28009937d8031054830160fce6d409ed342816b543597cece116d966c6d99e15", size = 22320, upload-time = "2024-10-08T23:04:09.501Z" },
+]
+
+[[package]]
+name = "jinja2"
+version = "3.1.6"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "markupsafe" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/df/bf/f7da0350254c0ed7c72f3e33cef02e048281fec7ecec5f032d4aac52226b/jinja2-3.1.6.tar.gz", hash = "sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d", size = 245115, upload-time = "2025-03-05T20:05:02.478Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/62/a1/3d680cbfd5f4b8f15abc1d571870c5fc3e594bb582bc3b64ea099db13e56/jinja2-3.1.6-py3-none-any.whl", hash = "sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67", size = 134899, upload-time = "2025-03-05T20:05:00.369Z" },
+]
+
+[[package]]
+name = "leather"
+version = "0.4.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9e/09/849cf129d7eae1e42f873f2dbd60323267c738390b686a7384fb3fb289ad/leather-0.4.1.tar.gz", hash = "sha256:67119c2aee93be821f077193bd8534e296c05b38bd174d9c5a80c4aa31d1a4d3", size = 44072, upload-time = "2025-12-15T19:01:42.224Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1a/d4/c4dcb02ed11f8884e169b3350fc40aa4c08edf8bed77a8f0f267542e6452/leather-0.4.1-py3-none-any.whl", hash = "sha256:ec61cba1ca3ccb96ed90e38b116fc58757d97d352171006b3288c47ce3fbd183", size = 30340, upload-time = "2025-12-15T19:01:40.823Z" },
+]
+
+[[package]]
+name = "loguru"
+version = "0.7.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "win32-setctime", marker = "sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/3a/05/a1dae3dffd1116099471c643b8924f5aa6524411dc6c63fdae648c4f1aca/loguru-0.7.3.tar.gz", hash = "sha256:19480589e77d47b8d85b2c827ad95d49bf31b0dcde16593892eb51dd18706eb6", size = 63559, upload-time = "2024-12-06T11:20:56.608Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0c/29/0348de65b8cc732daa3e33e67806420b2ae89bdce2b04af740289c5c6c8c/loguru-0.7.3-py3-none-any.whl", hash = "sha256:31a33c10c8e1e10422bfd431aeb5d351c7cf7fa671e3c4df004162264b28220c", size = 61595, upload-time = "2024-12-06T11:20:54.538Z" },
+]
+
+[[package]]
+name = "mako"
+version = "1.3.12"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "markupsafe" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/00/62/791b31e69ae182791ec67f04850f2f062716bbd205483d63a215f3e062d3/mako-1.3.12.tar.gz", hash = "sha256:9f778e93289bd410bb35daadeb4fc66d95a746f0b75777b942088b7fd7af550a", size = 400219, upload-time = "2026-04-28T19:01:08.512Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bc/b1/a0ec7a5a9db730a08daef1fdfb8090435b82465abbf758a596f0ea88727e/mako-1.3.12-py3-none-any.whl", hash = "sha256:8f61569480282dbf557145ce441e4ba888be453c30989f879f0d652e39f53ea9", size = 78521, upload-time = "2026-04-28T19:01:10.393Z" },
+]
+
+[[package]]
+name = "markupsafe"
+version = "3.0.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7e/99/7690b6d4034fffd95959cbe0c02de8deb3098cc577c67bb6a24fe5d7caa7/markupsafe-3.0.3.tar.gz", hash = "sha256:722695808f4b6457b320fdc131280796bdceb04ab50fe1795cd540799ebe1698", size = 80313, upload-time = "2025-09-27T18:37:40.426Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e8/4b/3541d44f3937ba468b75da9eebcae497dcf67adb65caa16760b0a6807ebb/markupsafe-3.0.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2f981d352f04553a7171b8e44369f2af4055f888dfb147d55e42d29e29e74559", size = 11631, upload-time = "2025-09-27T18:36:05.558Z" },
+    { url = "https://files.pythonhosted.org/packages/98/1b/fbd8eed11021cabd9226c37342fa6ca4e8a98d8188a8d9b66740494960e4/markupsafe-3.0.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e1c1493fb6e50ab01d20a22826e57520f1284df32f2d8601fdd90b6304601419", size = 12057, upload-time = "2025-09-27T18:36:07.165Z" },
+    { url = "https://files.pythonhosted.org/packages/40/01/e560d658dc0bb8ab762670ece35281dec7b6c1b33f5fbc09ebb57a185519/markupsafe-3.0.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1ba88449deb3de88bd40044603fafffb7bc2b055d626a330323a9ed736661695", size = 22050, upload-time = "2025-09-27T18:36:08.005Z" },
+    { url = "https://files.pythonhosted.org/packages/af/cd/ce6e848bbf2c32314c9b237839119c5a564a59725b53157c856e90937b7a/markupsafe-3.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f42d0984e947b8adf7dd6dde396e720934d12c506ce84eea8476409563607591", size = 20681, upload-time = "2025-09-27T18:36:08.881Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/2a/b5c12c809f1c3045c4d580b035a743d12fcde53cf685dbc44660826308da/markupsafe-3.0.3-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:c0c0b3ade1c0b13b936d7970b1d37a57acde9199dc2aecc4c336773e1d86049c", size = 20705, upload-time = "2025-09-27T18:36:10.131Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/e3/9427a68c82728d0a88c50f890d0fc072a1484de2f3ac1ad0bfc1a7214fd5/markupsafe-3.0.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:0303439a41979d9e74d18ff5e2dd8c43ed6c6001fd40e5bf2e43f7bd9bbc523f", size = 21524, upload-time = "2025-09-27T18:36:11.324Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/36/23578f29e9e582a4d0278e009b38081dbe363c5e7165113fad546918a232/markupsafe-3.0.3-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:d2ee202e79d8ed691ceebae8e0486bd9a2cd4794cec4824e1c99b6f5009502f6", size = 20282, upload-time = "2025-09-27T18:36:12.573Z" },
+    { url = "https://files.pythonhosted.org/packages/56/21/dca11354e756ebd03e036bd8ad58d6d7168c80ce1fe5e75218e4945cbab7/markupsafe-3.0.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:177b5253b2834fe3678cb4a5f0059808258584c559193998be2601324fdeafb1", size = 20745, upload-time = "2025-09-27T18:36:13.504Z" },
+    { url = "https://files.pythonhosted.org/packages/87/99/faba9369a7ad6e4d10b6a5fbf71fa2a188fe4a593b15f0963b73859a1bbd/markupsafe-3.0.3-cp310-cp310-win32.whl", hash = "sha256:2a15a08b17dd94c53a1da0438822d70ebcd13f8c3a95abe3a9ef9f11a94830aa", size = 14571, upload-time = "2025-09-27T18:36:14.779Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/25/55dc3ab959917602c96985cb1253efaa4ff42f71194bddeb61eb7278b8be/markupsafe-3.0.3-cp310-cp310-win_amd64.whl", hash = "sha256:c4ffb7ebf07cfe8931028e3e4c85f0357459a3f9f9490886198848f4fa002ec8", size = 15056, upload-time = "2025-09-27T18:36:16.125Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/9e/0a02226640c255d1da0b8d12e24ac2aa6734da68bff14c05dd53b94a0fc3/markupsafe-3.0.3-cp310-cp310-win_arm64.whl", hash = "sha256:e2103a929dfa2fcaf9bb4e7c091983a49c9ac3b19c9061b6d5427dd7d14d81a1", size = 13932, upload-time = "2025-09-27T18:36:17.311Z" },
+    { url = "https://files.pythonhosted.org/packages/08/db/fefacb2136439fc8dd20e797950e749aa1f4997ed584c62cfb8ef7c2be0e/markupsafe-3.0.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:1cc7ea17a6824959616c525620e387f6dd30fec8cb44f649e31712db02123dad", size = 11631, upload-time = "2025-09-27T18:36:18.185Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/2e/5898933336b61975ce9dc04decbc0a7f2fee78c30353c5efba7f2d6ff27a/markupsafe-3.0.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4bd4cd07944443f5a265608cc6aab442e4f74dff8088b0dfc8238647b8f6ae9a", size = 12058, upload-time = "2025-09-27T18:36:19.444Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/09/adf2df3699d87d1d8184038df46a9c80d78c0148492323f4693df54e17bb/markupsafe-3.0.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6b5420a1d9450023228968e7e6a9ce57f65d148ab56d2313fcd589eee96a7a50", size = 24287, upload-time = "2025-09-27T18:36:20.768Z" },
+    { url = "https://files.pythonhosted.org/packages/30/ac/0273f6fcb5f42e314c6d8cd99effae6a5354604d461b8d392b5ec9530a54/markupsafe-3.0.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0bf2a864d67e76e5c9a34dc26ec616a66b9888e25e7b9460e1c76d3293bd9dbf", size = 22940, upload-time = "2025-09-27T18:36:22.249Z" },
+    { url = "https://files.pythonhosted.org/packages/19/ae/31c1be199ef767124c042c6c3e904da327a2f7f0cd63a0337e1eca2967a8/markupsafe-3.0.3-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:bc51efed119bc9cfdf792cdeaa4d67e8f6fcccab66ed4bfdd6bde3e59bfcbb2f", size = 21887, upload-time = "2025-09-27T18:36:23.535Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/76/7edcab99d5349a4532a459e1fe64f0b0467a3365056ae550d3bcf3f79e1e/markupsafe-3.0.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:068f375c472b3e7acbe2d5318dea141359e6900156b5b2ba06a30b169086b91a", size = 23692, upload-time = "2025-09-27T18:36:24.823Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/28/6e74cdd26d7514849143d69f0bf2399f929c37dc2b31e6829fd2045b2765/markupsafe-3.0.3-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:7be7b61bb172e1ed687f1754f8e7484f1c8019780f6f6b0786e76bb01c2ae115", size = 21471, upload-time = "2025-09-27T18:36:25.95Z" },
+    { url = "https://files.pythonhosted.org/packages/62/7e/a145f36a5c2945673e590850a6f8014318d5577ed7e5920a4b3448e0865d/markupsafe-3.0.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:f9e130248f4462aaa8e2552d547f36ddadbeaa573879158d721bbd33dfe4743a", size = 22923, upload-time = "2025-09-27T18:36:27.109Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/62/d9c46a7f5c9adbeeeda52f5b8d802e1094e9717705a645efc71b0913a0a8/markupsafe-3.0.3-cp311-cp311-win32.whl", hash = "sha256:0db14f5dafddbb6d9208827849fad01f1a2609380add406671a26386cdf15a19", size = 14572, upload-time = "2025-09-27T18:36:28.045Z" },
+    { url = "https://files.pythonhosted.org/packages/83/8a/4414c03d3f891739326e1783338e48fb49781cc915b2e0ee052aa490d586/markupsafe-3.0.3-cp311-cp311-win_amd64.whl", hash = "sha256:de8a88e63464af587c950061a5e6a67d3632e36df62b986892331d4620a35c01", size = 15077, upload-time = "2025-09-27T18:36:29.025Z" },
+    { url = "https://files.pythonhosted.org/packages/35/73/893072b42e6862f319b5207adc9ae06070f095b358655f077f69a35601f0/markupsafe-3.0.3-cp311-cp311-win_arm64.whl", hash = "sha256:3b562dd9e9ea93f13d53989d23a7e775fdfd1066c33494ff43f5418bc8c58a5c", size = 13876, upload-time = "2025-09-27T18:36:29.954Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/72/147da192e38635ada20e0a2e1a51cf8823d2119ce8883f7053879c2199b5/markupsafe-3.0.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d53197da72cc091b024dd97249dfc7794d6a56530370992a5e1a08983ad9230e", size = 11615, upload-time = "2025-09-27T18:36:30.854Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/81/7e4e08678a1f98521201c3079f77db69fb552acd56067661f8c2f534a718/markupsafe-3.0.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1872df69a4de6aead3491198eaf13810b565bdbeec3ae2dc8780f14458ec73ce", size = 12020, upload-time = "2025-09-27T18:36:31.971Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/2c/799f4742efc39633a1b54a92eec4082e4f815314869865d876824c257c1e/markupsafe-3.0.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3a7e8ae81ae39e62a41ec302f972ba6ae23a5c5396c8e60113e9066ef893da0d", size = 24332, upload-time = "2025-09-27T18:36:32.813Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/2e/8d0c2ab90a8c1d9a24f0399058ab8519a3279d1bd4289511d74e909f060e/markupsafe-3.0.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d6dd0be5b5b189d31db7cda48b91d7e0a9795f31430b7f271219ab30f1d3ac9d", size = 22947, upload-time = "2025-09-27T18:36:33.86Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/54/887f3092a85238093a0b2154bd629c89444f395618842e8b0c41783898ea/markupsafe-3.0.3-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:94c6f0bb423f739146aec64595853541634bde58b2135f27f61c1ffd1cd4d16a", size = 21962, upload-time = "2025-09-27T18:36:35.099Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/2f/336b8c7b6f4a4d95e91119dc8521402461b74a485558d8f238a68312f11c/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:be8813b57049a7dc738189df53d69395eba14fb99345e0a5994914a3864c8a4b", size = 23760, upload-time = "2025-09-27T18:36:36.001Z" },
+    { url = "https://files.pythonhosted.org/packages/32/43/67935f2b7e4982ffb50a4d169b724d74b62a3964bc1a9a527f5ac4f1ee2b/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:83891d0e9fb81a825d9a6d61e3f07550ca70a076484292a70fde82c4b807286f", size = 21529, upload-time = "2025-09-27T18:36:36.906Z" },
+    { url = "https://files.pythonhosted.org/packages/89/e0/4486f11e51bbba8b0c041098859e869e304d1c261e59244baa3d295d47b7/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:77f0643abe7495da77fb436f50f8dab76dbc6e5fd25d39589a0f1fe6548bfa2b", size = 23015, upload-time = "2025-09-27T18:36:37.868Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/e1/78ee7a023dac597a5825441ebd17170785a9dab23de95d2c7508ade94e0e/markupsafe-3.0.3-cp312-cp312-win32.whl", hash = "sha256:d88b440e37a16e651bda4c7c2b930eb586fd15ca7406cb39e211fcff3bf3017d", size = 14540, upload-time = "2025-09-27T18:36:38.761Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/5b/bec5aa9bbbb2c946ca2733ef9c4ca91c91b6a24580193e891b5f7dbe8e1e/markupsafe-3.0.3-cp312-cp312-win_amd64.whl", hash = "sha256:26a5784ded40c9e318cfc2bdb30fe164bdb8665ded9cd64d500a34fb42067b1c", size = 15105, upload-time = "2025-09-27T18:36:39.701Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/f1/216fc1bbfd74011693a4fd837e7026152e89c4bcf3e77b6692fba9923123/markupsafe-3.0.3-cp312-cp312-win_arm64.whl", hash = "sha256:35add3b638a5d900e807944a078b51922212fb3dedb01633a8defc4b01a3c85f", size = 13906, upload-time = "2025-09-27T18:36:40.689Z" },
+    { url = "https://files.pythonhosted.org/packages/38/2f/907b9c7bbba283e68f20259574b13d005c121a0fa4c175f9bed27c4597ff/markupsafe-3.0.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:e1cf1972137e83c5d4c136c43ced9ac51d0e124706ee1c8aa8532c1287fa8795", size = 11622, upload-time = "2025-09-27T18:36:41.777Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/d9/5f7756922cdd676869eca1c4e3c0cd0df60ed30199ffd775e319089cb3ed/markupsafe-3.0.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:116bb52f642a37c115f517494ea5feb03889e04df47eeff5b130b1808ce7c219", size = 12029, upload-time = "2025-09-27T18:36:43.257Z" },
+    { url = "https://files.pythonhosted.org/packages/00/07/575a68c754943058c78f30db02ee03a64b3c638586fba6a6dd56830b30a3/markupsafe-3.0.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:133a43e73a802c5562be9bbcd03d090aa5a1fe899db609c29e8c8d815c5f6de6", size = 24374, upload-time = "2025-09-27T18:36:44.508Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/21/9b05698b46f218fc0e118e1f8168395c65c8a2c750ae2bab54fc4bd4e0e8/markupsafe-3.0.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ccfcd093f13f0f0b7fdd0f198b90053bf7b2f02a3927a30e63f3ccc9df56b676", size = 22980, upload-time = "2025-09-27T18:36:45.385Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/71/544260864f893f18b6827315b988c146b559391e6e7e8f7252839b1b846a/markupsafe-3.0.3-cp313-cp313-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:509fa21c6deb7a7a273d629cf5ec029bc209d1a51178615ddf718f5918992ab9", size = 21990, upload-time = "2025-09-27T18:36:46.916Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/28/b50fc2f74d1ad761af2f5dcce7492648b983d00a65b8c0e0cb457c82ebbe/markupsafe-3.0.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:a4afe79fb3de0b7097d81da19090f4df4f8d3a2b3adaa8764138aac2e44f3af1", size = 23784, upload-time = "2025-09-27T18:36:47.884Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/76/104b2aa106a208da8b17a2fb72e033a5a9d7073c68f7e508b94916ed47a9/markupsafe-3.0.3-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:795e7751525cae078558e679d646ae45574b47ed6e7771863fcc079a6171a0fc", size = 21588, upload-time = "2025-09-27T18:36:48.82Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/99/16a5eb2d140087ebd97180d95249b00a03aa87e29cc224056274f2e45fd6/markupsafe-3.0.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8485f406a96febb5140bfeca44a73e3ce5116b2501ac54fe953e488fb1d03b12", size = 23041, upload-time = "2025-09-27T18:36:49.797Z" },
+    { url = "https://files.pythonhosted.org/packages/19/bc/e7140ed90c5d61d77cea142eed9f9c303f4c4806f60a1044c13e3f1471d0/markupsafe-3.0.3-cp313-cp313-win32.whl", hash = "sha256:bdd37121970bfd8be76c5fb069c7751683bdf373db1ed6c010162b2a130248ed", size = 14543, upload-time = "2025-09-27T18:36:51.584Z" },
+    { url = "https://files.pythonhosted.org/packages/05/73/c4abe620b841b6b791f2edc248f556900667a5a1cf023a6646967ae98335/markupsafe-3.0.3-cp313-cp313-win_amd64.whl", hash = "sha256:9a1abfdc021a164803f4d485104931fb8f8c1efd55bc6b748d2f5774e78b62c5", size = 15113, upload-time = "2025-09-27T18:36:52.537Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/3a/fa34a0f7cfef23cf9500d68cb7c32dd64ffd58a12b09225fb03dd37d5b80/markupsafe-3.0.3-cp313-cp313-win_arm64.whl", hash = "sha256:7e68f88e5b8799aa49c85cd116c932a1ac15caaa3f5db09087854d218359e485", size = 13911, upload-time = "2025-09-27T18:36:53.513Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/d7/e05cd7efe43a88a17a37b3ae96e79a19e846f3f456fe79c57ca61356ef01/markupsafe-3.0.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:218551f6df4868a8d527e3062d0fb968682fe92054e89978594c28e642c43a73", size = 11658, upload-time = "2025-09-27T18:36:54.819Z" },
+    { url = "https://files.pythonhosted.org/packages/99/9e/e412117548182ce2148bdeacdda3bb494260c0b0184360fe0d56389b523b/markupsafe-3.0.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:3524b778fe5cfb3452a09d31e7b5adefeea8c5be1d43c4f810ba09f2ceb29d37", size = 12066, upload-time = "2025-09-27T18:36:55.714Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/e6/fa0ffcda717ef64a5108eaa7b4f5ed28d56122c9a6d70ab8b72f9f715c80/markupsafe-3.0.3-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4e885a3d1efa2eadc93c894a21770e4bc67899e3543680313b09f139e149ab19", size = 25639, upload-time = "2025-09-27T18:36:56.908Z" },
+    { url = "https://files.pythonhosted.org/packages/96/ec/2102e881fe9d25fc16cb4b25d5f5cde50970967ffa5dddafdb771237062d/markupsafe-3.0.3-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8709b08f4a89aa7586de0aadc8da56180242ee0ada3999749b183aa23df95025", size = 23569, upload-time = "2025-09-27T18:36:57.913Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/30/6f2fce1f1f205fc9323255b216ca8a235b15860c34b6798f810f05828e32/markupsafe-3.0.3-cp313-cp313t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:b8512a91625c9b3da6f127803b166b629725e68af71f8184ae7e7d54686a56d6", size = 23284, upload-time = "2025-09-27T18:36:58.833Z" },
+    { url = "https://files.pythonhosted.org/packages/58/47/4a0ccea4ab9f5dcb6f79c0236d954acb382202721e704223a8aafa38b5c8/markupsafe-3.0.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:9b79b7a16f7fedff2495d684f2b59b0457c3b493778c9eed31111be64d58279f", size = 24801, upload-time = "2025-09-27T18:36:59.739Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/70/3780e9b72180b6fecb83a4814d84c3bf4b4ae4bf0b19c27196104149734c/markupsafe-3.0.3-cp313-cp313t-musllinux_1_2_riscv64.whl", hash = "sha256:12c63dfb4a98206f045aa9563db46507995f7ef6d83b2f68eda65c307c6829eb", size = 22769, upload-time = "2025-09-27T18:37:00.719Z" },
+    { url = "https://files.pythonhosted.org/packages/98/c5/c03c7f4125180fc215220c035beac6b9cb684bc7a067c84fc69414d315f5/markupsafe-3.0.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:8f71bc33915be5186016f675cd83a1e08523649b0e33efdb898db577ef5bb009", size = 23642, upload-time = "2025-09-27T18:37:01.673Z" },
+    { url = "https://files.pythonhosted.org/packages/80/d6/2d1b89f6ca4bff1036499b1e29a1d02d282259f3681540e16563f27ebc23/markupsafe-3.0.3-cp313-cp313t-win32.whl", hash = "sha256:69c0b73548bc525c8cb9a251cddf1931d1db4d2258e9599c28c07ef3580ef354", size = 14612, upload-time = "2025-09-27T18:37:02.639Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/98/e48a4bfba0a0ffcf9925fe2d69240bfaa19c6f7507b8cd09c70684a53c1e/markupsafe-3.0.3-cp313-cp313t-win_amd64.whl", hash = "sha256:1b4b79e8ebf6b55351f0d91fe80f893b4743f104bff22e90697db1590e47a218", size = 15200, upload-time = "2025-09-27T18:37:03.582Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/72/e3cc540f351f316e9ed0f092757459afbc595824ca724cbc5a5d4263713f/markupsafe-3.0.3-cp313-cp313t-win_arm64.whl", hash = "sha256:ad2cf8aa28b8c020ab2fc8287b0f823d0a7d8630784c31e9ee5edea20f406287", size = 13973, upload-time = "2025-09-27T18:37:04.929Z" },
+    { url = "https://files.pythonhosted.org/packages/33/8a/8e42d4838cd89b7dde187011e97fe6c3af66d8c044997d2183fbd6d31352/markupsafe-3.0.3-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:eaa9599de571d72e2daf60164784109f19978b327a3910d3e9de8c97b5b70cfe", size = 11619, upload-time = "2025-09-27T18:37:06.342Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/64/7660f8a4a8e53c924d0fa05dc3a55c9cee10bbd82b11c5afb27d44b096ce/markupsafe-3.0.3-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:c47a551199eb8eb2121d4f0f15ae0f923d31350ab9280078d1e5f12b249e0026", size = 12029, upload-time = "2025-09-27T18:37:07.213Z" },
+    { url = "https://files.pythonhosted.org/packages/da/ef/e648bfd021127bef5fa12e1720ffed0c6cbb8310c8d9bea7266337ff06de/markupsafe-3.0.3-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f34c41761022dd093b4b6896d4810782ffbabe30f2d443ff5f083e0cbbb8c737", size = 24408, upload-time = "2025-09-27T18:37:09.572Z" },
+    { url = "https://files.pythonhosted.org/packages/41/3c/a36c2450754618e62008bf7435ccb0f88053e07592e6028a34776213d877/markupsafe-3.0.3-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:457a69a9577064c05a97c41f4e65148652db078a3a509039e64d3467b9e7ef97", size = 23005, upload-time = "2025-09-27T18:37:10.58Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/20/b7fdf89a8456b099837cd1dc21974632a02a999ec9bf7ca3e490aacd98e7/markupsafe-3.0.3-cp314-cp314-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:e8afc3f2ccfa24215f8cb28dcf43f0113ac3c37c2f0f0806d8c70e4228c5cf4d", size = 22048, upload-time = "2025-09-27T18:37:11.547Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/a7/591f592afdc734f47db08a75793a55d7fbcc6902a723ae4cfbab61010cc5/markupsafe-3.0.3-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:ec15a59cf5af7be74194f7ab02d0f59a62bdcf1a537677ce67a2537c9b87fcda", size = 23821, upload-time = "2025-09-27T18:37:12.48Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/33/45b24e4f44195b26521bc6f1a82197118f74df348556594bd2262bda1038/markupsafe-3.0.3-cp314-cp314-musllinux_1_2_riscv64.whl", hash = "sha256:0eb9ff8191e8498cca014656ae6b8d61f39da5f95b488805da4bb029cccbfbaf", size = 21606, upload-time = "2025-09-27T18:37:13.485Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/0e/53dfaca23a69fbfbbf17a4b64072090e70717344c52eaaaa9c5ddff1e5f0/markupsafe-3.0.3-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:2713baf880df847f2bece4230d4d094280f4e67b1e813eec43b4c0e144a34ffe", size = 23043, upload-time = "2025-09-27T18:37:14.408Z" },
+    { url = "https://files.pythonhosted.org/packages/46/11/f333a06fc16236d5238bfe74daccbca41459dcd8d1fa952e8fbd5dccfb70/markupsafe-3.0.3-cp314-cp314-win32.whl", hash = "sha256:729586769a26dbceff69f7a7dbbf59ab6572b99d94576a5592625d5b411576b9", size = 14747, upload-time = "2025-09-27T18:37:15.36Z" },
+    { url = "https://files.pythonhosted.org/packages/28/52/182836104b33b444e400b14f797212f720cbc9ed6ba34c800639d154e821/markupsafe-3.0.3-cp314-cp314-win_amd64.whl", hash = "sha256:bdc919ead48f234740ad807933cdf545180bfbe9342c2bb451556db2ed958581", size = 15341, upload-time = "2025-09-27T18:37:16.496Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/18/acf23e91bd94fd7b3031558b1f013adfa21a8e407a3fdb32745538730382/markupsafe-3.0.3-cp314-cp314-win_arm64.whl", hash = "sha256:5a7d5dc5140555cf21a6fefbdbf8723f06fcd2f63ef108f2854de715e4422cb4", size = 14073, upload-time = "2025-09-27T18:37:17.476Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/f0/57689aa4076e1b43b15fdfa646b04653969d50cf30c32a102762be2485da/markupsafe-3.0.3-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:1353ef0c1b138e1907ae78e2f6c63ff67501122006b0f9abad68fda5f4ffc6ab", size = 11661, upload-time = "2025-09-27T18:37:18.453Z" },
+    { url = "https://files.pythonhosted.org/packages/89/c3/2e67a7ca217c6912985ec766c6393b636fb0c2344443ff9d91404dc4c79f/markupsafe-3.0.3-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:1085e7fbddd3be5f89cc898938f42c0b3c711fdcb37d75221de2666af647c175", size = 12069, upload-time = "2025-09-27T18:37:19.332Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/00/be561dce4e6ca66b15276e184ce4b8aec61fe83662cce2f7d72bd3249d28/markupsafe-3.0.3-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1b52b4fb9df4eb9ae465f8d0c228a00624de2334f216f178a995ccdcf82c4634", size = 25670, upload-time = "2025-09-27T18:37:20.245Z" },
+    { url = "https://files.pythonhosted.org/packages/50/09/c419f6f5a92e5fadde27efd190eca90f05e1261b10dbd8cbcb39cd8ea1dc/markupsafe-3.0.3-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fed51ac40f757d41b7c48425901843666a6677e3e8eb0abcff09e4ba6e664f50", size = 23598, upload-time = "2025-09-27T18:37:21.177Z" },
+    { url = "https://files.pythonhosted.org/packages/22/44/a0681611106e0b2921b3033fc19bc53323e0b50bc70cffdd19f7d679bb66/markupsafe-3.0.3-cp314-cp314t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:f190daf01f13c72eac4efd5c430a8de82489d9cff23c364c3ea822545032993e", size = 23261, upload-time = "2025-09-27T18:37:22.167Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/57/1b0b3f100259dc9fffe780cfb60d4be71375510e435efec3d116b6436d43/markupsafe-3.0.3-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:e56b7d45a839a697b5eb268c82a71bd8c7f6c94d6fd50c3d577fa39a9f1409f5", size = 24835, upload-time = "2025-09-27T18:37:23.296Z" },
+    { url = "https://files.pythonhosted.org/packages/26/6a/4bf6d0c97c4920f1597cc14dd720705eca0bf7c787aebc6bb4d1bead5388/markupsafe-3.0.3-cp314-cp314t-musllinux_1_2_riscv64.whl", hash = "sha256:f3e98bb3798ead92273dc0e5fd0f31ade220f59a266ffd8a4f6065e0a3ce0523", size = 22733, upload-time = "2025-09-27T18:37:24.237Z" },
+    { url = "https://files.pythonhosted.org/packages/14/c7/ca723101509b518797fedc2fdf79ba57f886b4aca8a7d31857ba3ee8281f/markupsafe-3.0.3-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:5678211cb9333a6468fb8d8be0305520aa073f50d17f089b5b4b477ea6e67fdc", size = 23672, upload-time = "2025-09-27T18:37:25.271Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/df/5bd7a48c256faecd1d36edc13133e51397e41b73bb77e1a69deab746ebac/markupsafe-3.0.3-cp314-cp314t-win32.whl", hash = "sha256:915c04ba3851909ce68ccc2b8e2cd691618c4dc4c4232fb7982bca3f41fd8c3d", size = 14819, upload-time = "2025-09-27T18:37:26.285Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/8a/0402ba61a2f16038b48b39bccca271134be00c5c9f0f623208399333c448/markupsafe-3.0.3-cp314-cp314t-win_amd64.whl", hash = "sha256:4faffd047e07c38848ce017e8725090413cd80cbc23d86e55c587bf979e579c9", size = 15426, upload-time = "2025-09-27T18:37:27.316Z" },
+    { url = "https://files.pythonhosted.org/packages/70/bc/6f1c2f612465f5fa89b95bead1f44dcb607670fd42891d8fdcd5d039f4f4/markupsafe-3.0.3-cp314-cp314t-win_arm64.whl", hash = "sha256:32001d6a8fc98c8cb5c947787c5d08b0a50663d139f1305bac5885d98d9b40fa", size = 14146, upload-time = "2025-09-27T18:37:28.327Z" },
+]
+
+[[package]]
+name = "ml-dtypes"
+version = "0.5.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/0e/4a/c27b42ed9b1c7d13d9ba8b6905dece787d6259152f2309338aed29b2447b/ml_dtypes-0.5.4.tar.gz", hash = "sha256:8ab06a50fb9bf9666dd0fe5dfb4676fa2b0ac0f31ecff72a6c3af8e22c063453", size = 692314, upload-time = "2025-11-17T22:32:31.031Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fe/3a/c5b855752a70267ff729c349e650263adb3c206c29d28cc8ea7ace30a1d5/ml_dtypes-0.5.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b95e97e470fe60ed493fd9ae3911d8da4ebac16bd21f87ffa2b7c588bf22ea2c", size = 679735, upload-time = "2025-11-17T22:31:31.367Z" },
+    { url = "https://files.pythonhosted.org/packages/41/79/7433f30ee04bd4faa303844048f55e1eb939131c8e5195a00a96a0939b64/ml_dtypes-0.5.4-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b4b801ebe0b477be666696bda493a9be8356f1f0057a57f1e35cd26928823e5a", size = 5051883, upload-time = "2025-11-17T22:31:33.658Z" },
+    { url = "https://files.pythonhosted.org/packages/10/b1/8938e8830b0ee2e167fc75a094dea766a1152bde46752cd9bfc57ee78a82/ml_dtypes-0.5.4-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:388d399a2152dd79a3f0456a952284a99ee5c93d3e2f8dfe25977511e0515270", size = 5030369, upload-time = "2025-11-17T22:31:35.595Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/a3/51886727bd16e2f47587997b802dd56398692ce8c6c03c2e5bb32ecafe26/ml_dtypes-0.5.4-cp310-cp310-win_amd64.whl", hash = "sha256:4ff7f3e7ca2972e7de850e7b8fcbb355304271e2933dd90814c1cb847414d6e2", size = 210738, upload-time = "2025-11-17T22:31:37.43Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/5e/712092cfe7e5eb667b8ad9ca7c54442f21ed7ca8979745f1000e24cf8737/ml_dtypes-0.5.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6c7ecb74c4bd71db68a6bea1edf8da8c34f3d9fe218f038814fd1d310ac76c90", size = 679734, upload-time = "2025-11-17T22:31:39.223Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/cf/912146dfd4b5c0eea956836c01dcd2fce6c9c844b2691f5152aca196ce4f/ml_dtypes-0.5.4-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bc11d7e8c44a65115d05e2ab9989d1e045125d7be8e05a071a48bc76eb6d6040", size = 5056165, upload-time = "2025-11-17T22:31:41.071Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/80/19189ea605017473660e43762dc853d2797984b3c7bf30ce656099add30c/ml_dtypes-0.5.4-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:19b9a53598f21e453ea2fbda8aa783c20faff8e1eeb0d7ab899309a0053f1483", size = 5034975, upload-time = "2025-11-17T22:31:42.758Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/24/70bd59276883fdd91600ca20040b41efd4902a923283c4d6edcb1de128d2/ml_dtypes-0.5.4-cp311-cp311-win_amd64.whl", hash = "sha256:7c23c54a00ae43edf48d44066a7ec31e05fdc2eee0be2b8b50dd1903a1db94bb", size = 210742, upload-time = "2025-11-17T22:31:44.068Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/c9/64230ef14e40aa3f1cb254ef623bf812735e6bec7772848d19131111ac0d/ml_dtypes-0.5.4-cp311-cp311-win_arm64.whl", hash = "sha256:557a31a390b7e9439056644cb80ed0735a6e3e3bb09d67fd5687e4b04238d1de", size = 160709, upload-time = "2025-11-17T22:31:46.557Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/b8/3c70881695e056f8a32f8b941126cf78775d9a4d7feba8abcb52cb7b04f2/ml_dtypes-0.5.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:a174837a64f5b16cab6f368171a1a03a27936b31699d167684073ff1c4237dac", size = 676927, upload-time = "2025-11-17T22:31:48.182Z" },
+    { url = "https://files.pythonhosted.org/packages/54/0f/428ef6881782e5ebb7eca459689448c0394fa0a80bea3aa9262cba5445ea/ml_dtypes-0.5.4-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a7f7c643e8b1320fd958bf098aa7ecf70623a42ec5154e3be3be673f4c34d900", size = 5028464, upload-time = "2025-11-17T22:31:50.135Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/cb/28ce52eb94390dda42599c98ea0204d74799e4d8047a0eb559b6fd648056/ml_dtypes-0.5.4-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9ad459e99793fa6e13bd5b7e6792c8f9190b4e5a1b45c63aba14a4d0a7f1d5ff", size = 5009002, upload-time = "2025-11-17T22:31:52.001Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/f0/0cfadd537c5470378b1b32bd859cf2824972174b51b873c9d95cfd7475a5/ml_dtypes-0.5.4-cp312-cp312-win_amd64.whl", hash = "sha256:c1a953995cccb9e25a4ae19e34316671e4e2edaebe4cf538229b1fc7109087b7", size = 212222, upload-time = "2025-11-17T22:31:53.742Z" },
+    { url = "https://files.pythonhosted.org/packages/16/2e/9acc86985bfad8f2c2d30291b27cd2bb4c74cea08695bd540906ed744249/ml_dtypes-0.5.4-cp312-cp312-win_arm64.whl", hash = "sha256:9bad06436568442575beb2d03389aa7456c690a5b05892c471215bfd8cf39460", size = 160793, upload-time = "2025-11-17T22:31:55.358Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/a1/4008f14bbc616cfb1ac5b39ea485f9c63031c4634ab3f4cf72e7541f816a/ml_dtypes-0.5.4-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8c760d85a2f82e2bed75867079188c9d18dae2ee77c25a54d60e9cc79be1bc48", size = 676888, upload-time = "2025-11-17T22:31:56.907Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/b7/dff378afc2b0d5a7d6cd9d3209b60474d9819d1189d347521e1688a60a53/ml_dtypes-0.5.4-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ce756d3a10d0c4067172804c9cc276ba9cc0ff47af9078ad439b075d1abdc29b", size = 5036993, upload-time = "2025-11-17T22:31:58.497Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/33/40cd74219417e78b97c47802037cf2d87b91973e18bb968a7da48a96ea44/ml_dtypes-0.5.4-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:533ce891ba774eabf607172254f2e7260ba5f57bdd64030c9a4fcfbd99815d0d", size = 5010956, upload-time = "2025-11-17T22:31:59.931Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/8b/200088c6859d8221454825959df35b5244fa9bdf263fd0249ac5fb75e281/ml_dtypes-0.5.4-cp313-cp313-win_amd64.whl", hash = "sha256:f21c9219ef48ca5ee78402d5cc831bd58ea27ce89beda894428bc67a52da5328", size = 212224, upload-time = "2025-11-17T22:32:01.349Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/75/dfc3775cb36367816e678f69a7843f6f03bd4e2bcd79941e01ea960a068e/ml_dtypes-0.5.4-cp313-cp313-win_arm64.whl", hash = "sha256:35f29491a3e478407f7047b8a4834e4640a77d2737e0b294d049746507af5175", size = 160798, upload-time = "2025-11-17T22:32:02.864Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/74/e9ddb35fd1dd43b1106c20ced3f53c2e8e7fc7598c15638e9f80677f81d4/ml_dtypes-0.5.4-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:304ad47faa395415b9ccbcc06a0350800bc50eda70f0e45326796e27c62f18b6", size = 702083, upload-time = "2025-11-17T22:32:04.08Z" },
+    { url = "https://files.pythonhosted.org/packages/74/f5/667060b0aed1aa63166b22897fdf16dca9eb704e6b4bbf86848d5a181aa7/ml_dtypes-0.5.4-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6a0df4223b514d799b8a1629c65ddc351b3efa833ccf7f8ea0cf654a61d1e35d", size = 5354111, upload-time = "2025-11-17T22:32:05.546Z" },
+    { url = "https://files.pythonhosted.org/packages/40/49/0f8c498a28c0efa5f5c95a9e374c83ec1385ca41d0e85e7cf40e5d519a21/ml_dtypes-0.5.4-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:531eff30e4d368cb6255bc2328d070e35836aa4f282a0fb5f3a0cd7260257298", size = 5366453, upload-time = "2025-11-17T22:32:07.115Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/27/12607423d0a9c6bbbcc780ad19f1f6baa2b68b18ce4bddcdc122c4c68dc9/ml_dtypes-0.5.4-cp313-cp313t-win_amd64.whl", hash = "sha256:cb73dccfc991691c444acc8c0012bee8f2470da826a92e3a20bb333b1a7894e6", size = 225612, upload-time = "2025-11-17T22:32:08.615Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/80/5a5929e92c72936d5b19872c5fb8fc09327c1da67b3b68c6a13139e77e20/ml_dtypes-0.5.4-cp313-cp313t-win_arm64.whl", hash = "sha256:3bbbe120b915090d9dd1375e4684dd17a20a2491ef25d640a908281da85e73f1", size = 164145, upload-time = "2025-11-17T22:32:09.782Z" },
+    { url = "https://files.pythonhosted.org/packages/72/4e/1339dc6e2557a344f5ba5590872e80346f76f6cb2ac3dd16e4666e88818c/ml_dtypes-0.5.4-cp314-cp314-macosx_10_13_universal2.whl", hash = "sha256:2b857d3af6ac0d39db1de7c706e69c7f9791627209c3d6dedbfca8c7e5faec22", size = 673781, upload-time = "2025-11-17T22:32:11.364Z" },
+    { url = "https://files.pythonhosted.org/packages/04/f9/067b84365c7e83bda15bba2b06c6ca250ce27b20630b1128c435fb7a09aa/ml_dtypes-0.5.4-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:805cef3a38f4eafae3a5bf9ebdcdb741d0bcfd9e1bd90eb54abd24f928cd2465", size = 5036145, upload-time = "2025-11-17T22:32:12.783Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/bb/82c7dcf38070b46172a517e2334e665c5bf374a262f99a283ea454bece7c/ml_dtypes-0.5.4-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:14a4fd3228af936461db66faccef6e4f41c1d82fcc30e9f8d58a08916b1d811f", size = 5010230, upload-time = "2025-11-17T22:32:14.38Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/93/2bfed22d2498c468f6bcd0d9f56b033eaa19f33320389314c19ef6766413/ml_dtypes-0.5.4-cp314-cp314-win_amd64.whl", hash = "sha256:8c6a2dcebd6f3903e05d51960a8058d6e131fe69f952a5397e5dbabc841b6d56", size = 221032, upload-time = "2025-11-17T22:32:15.763Z" },
+    { url = "https://files.pythonhosted.org/packages/76/a3/9c912fe6ea747bb10fe2f8f54d027eb265db05dfb0c6335e3e063e74e6e8/ml_dtypes-0.5.4-cp314-cp314-win_arm64.whl", hash = "sha256:5a0f68ca8fd8d16583dfa7793973feb86f2fbb56ce3966daf9c9f748f52a2049", size = 163353, upload-time = "2025-11-17T22:32:16.932Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/02/48aa7d84cc30ab4ee37624a2fd98c56c02326785750cd212bc0826c2f15b/ml_dtypes-0.5.4-cp314-cp314t-macosx_10_13_universal2.whl", hash = "sha256:bfc534409c5d4b0bf945af29e5d0ab075eae9eecbb549ff8a29280db822f34f9", size = 702085, upload-time = "2025-11-17T22:32:18.175Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/e7/85cb99fe80a7a5513253ec7faa88a65306be071163485e9a626fce1b6e84/ml_dtypes-0.5.4-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2314892cdc3fcf05e373d76d72aaa15fda9fb98625effa73c1d646f331fcecb7", size = 5355358, upload-time = "2025-11-17T22:32:19.7Z" },
+    { url = "https://files.pythonhosted.org/packages/79/2b/a826ba18d2179a56e144aef69e57fb2ab7c464ef0b2111940ee8a3a223a2/ml_dtypes-0.5.4-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0d2ffd05a2575b1519dc928c0b93c06339eb67173ff53acb00724502cda231cf", size = 5366332, upload-time = "2025-11-17T22:32:21.193Z" },
+    { url = "https://files.pythonhosted.org/packages/84/44/f4d18446eacb20ea11e82f133ea8f86e2bf2891785b67d9da8d0ab0ef525/ml_dtypes-0.5.4-cp314-cp314t-win_amd64.whl", hash = "sha256:4381fe2f2452a2d7589689693d3162e876b3ddb0a832cde7a414f8e1adf7eab1", size = 236612, upload-time = "2025-11-17T22:32:22.579Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/3f/3d42e9a78fe5edf792a83c074b13b9b770092a4fbf3462872f4303135f09/ml_dtypes-0.5.4-cp314-cp314t-win_arm64.whl", hash = "sha256:11942cbf2cf92157db91e5022633c0d9474d4dfd813a909383bd23ce828a4b7d", size = 168825, upload-time = "2025-11-17T22:32:23.766Z" },
+]
+
+[[package]]
+name = "mpmath"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106, upload-time = "2023-03-07T16:47:11.061Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198, upload-time = "2023-03-07T16:47:09.197Z" },
+]
+
+[[package]]
+name = "narwhals"
+version = "2.22.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/62/3c/c4ef2164a71c1a63d7f1ae411c4082c5fa872405106db60a4b7114989ad7/narwhals-2.22.1.tar.gz", hash = "sha256:d62920805a0a43b7ff8b54b0c0d3142d796f8a9301836ada37e573d6a33cbcd9", size = 647493, upload-time = "2026-06-05T12:34:34.051Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/48/ca/36339329c4604adbcc99c899b7eb1ce1a555c499b6a6860757dc9bfed36d/narwhals-2.22.1-py3-none-any.whl", hash = "sha256:60567d774edf77db53906f89d9fbd164e66e56d66d388e1e6990f17ac33cfb53", size = 454815, upload-time = "2026-06-05T12:34:32.289Z" },
+]
+
+[[package]]
+name = "networkx"
+version = "3.4.2"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.11' and platform_machine != 's390x'",
+    "python_full_version < '3.11' and platform_machine == 's390x'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fd/1d/06475e1cd5264c0b870ea2cc6fdb3e37177c1e565c43f56ff17a10e3937f/networkx-3.4.2.tar.gz", hash = "sha256:307c3669428c5362aab27c8a1260aa8f47c4e91d3891f48be0141738d8d053e1", size = 2151368, upload-time = "2024-10-21T12:39:38.695Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/54/dd730b32ea14ea797530a4479b2ed46a6fb250f682a9cfb997e968bf0261/networkx-3.4.2-py3-none-any.whl", hash = "sha256:df5d4365b724cf81b8c6a7312509d0c22386097011ad1abe274afd5e9d3bbc5f", size = 1723263, upload-time = "2024-10-21T12:39:36.247Z" },
+]
+
+[[package]]
+name = "networkx"
+version = "3.6.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6a/51/63fe664f3908c97be9d2e4f1158eb633317598cfa6e1fc14af5383f17512/networkx-3.6.1.tar.gz", hash = "sha256:26b7c357accc0c8cde558ad486283728b65b6a95d85ee1cd66bafab4c8168509", size = 2517025, upload-time = "2025-12-08T17:02:39.908Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9e/c9/b2622292ea83fbb4ec318f5b9ab867d0a28ab43c5717bb85b0a5f6b3b0a4/networkx-3.6.1-py3-none-any.whl", hash = "sha256:d47fbf302e7d9cbbb9e2555a0d267983d2aa476bac30e90dfbe5669bd57f3762", size = 2068504, upload-time = "2025-12-08T17:02:38.159Z" },
+]
+
+[[package]]
+name = "numpy"
+version = "2.2.6"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.11' and platform_machine != 's390x'",
+    "python_full_version < '3.11' and platform_machine == 's390x'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/76/21/7d2a95e4bba9dc13d043ee156a356c0a8f0c6309dff6b21b4d71a073b8a8/numpy-2.2.6.tar.gz", hash = "sha256:e29554e2bef54a90aa5cc07da6ce955accb83f21ab5de01a62c8478897b264fd", size = 20276440, upload-time = "2025-05-17T22:38:04.611Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9a/3e/ed6db5be21ce87955c0cbd3009f2803f59fa08df21b5df06862e2d8e2bdd/numpy-2.2.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b412caa66f72040e6d268491a59f2c43bf03eb6c96dd8f0307829feb7fa2b6fb", size = 21165245, upload-time = "2025-05-17T21:27:58.555Z" },
+    { url = "https://files.pythonhosted.org/packages/22/c2/4b9221495b2a132cc9d2eb862e21d42a009f5a60e45fc44b00118c174bff/numpy-2.2.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:8e41fd67c52b86603a91c1a505ebaef50b3314de0213461c7a6e99c9a3beff90", size = 14360048, upload-time = "2025-05-17T21:28:21.406Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/77/dc2fcfc66943c6410e2bf598062f5959372735ffda175b39906d54f02349/numpy-2.2.6-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:37e990a01ae6ec7fe7fa1c26c55ecb672dd98b19c3d0e1d1f326fa13cb38d163", size = 5340542, upload-time = "2025-05-17T21:28:30.931Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/4f/1cb5fdc353a5f5cc7feb692db9b8ec2c3d6405453f982435efc52561df58/numpy-2.2.6-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:5a6429d4be8ca66d889b7cf70f536a397dc45ba6faeb5f8c5427935d9592e9cf", size = 6878301, upload-time = "2025-05-17T21:28:41.613Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/17/96a3acd228cec142fcb8723bd3cc39c2a474f7dcf0a5d16731980bcafa95/numpy-2.2.6-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:efd28d4e9cd7d7a8d39074a4d44c63eda73401580c5c76acda2ce969e0a38e83", size = 14297320, upload-time = "2025-05-17T21:29:02.78Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/63/3de6a34ad7ad6646ac7d2f55ebc6ad439dbbf9c4370017c50cf403fb19b5/numpy-2.2.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fc7b73d02efb0e18c000e9ad8b83480dfcd5dfd11065997ed4c6747470ae8915", size = 16801050, upload-time = "2025-05-17T21:29:27.675Z" },
+    { url = "https://files.pythonhosted.org/packages/07/b6/89d837eddef52b3d0cec5c6ba0456c1bf1b9ef6a6672fc2b7873c3ec4e2e/numpy-2.2.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:74d4531beb257d2c3f4b261bfb0fc09e0f9ebb8842d82a7b4209415896adc680", size = 15807034, upload-time = "2025-05-17T21:29:51.102Z" },
+    { url = "https://files.pythonhosted.org/packages/01/c8/dc6ae86e3c61cfec1f178e5c9f7858584049b6093f843bca541f94120920/numpy-2.2.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:8fc377d995680230e83241d8a96def29f204b5782f371c532579b4f20607a289", size = 18614185, upload-time = "2025-05-17T21:30:18.703Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/c5/0064b1b7e7c89137b471ccec1fd2282fceaae0ab3a9550f2568782d80357/numpy-2.2.6-cp310-cp310-win32.whl", hash = "sha256:b093dd74e50a8cba3e873868d9e93a85b78e0daf2e98c6797566ad8044e8363d", size = 6527149, upload-time = "2025-05-17T21:30:29.788Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/dd/4b822569d6b96c39d1215dbae0582fd99954dcbcf0c1a13c61783feaca3f/numpy-2.2.6-cp310-cp310-win_amd64.whl", hash = "sha256:f0fd6321b839904e15c46e0d257fdd101dd7f530fe03fd6359c1ea63738703f3", size = 12904620, upload-time = "2025-05-17T21:30:48.994Z" },
+    { url = "https://files.pythonhosted.org/packages/da/a8/4f83e2aa666a9fbf56d6118faaaf5f1974d456b1823fda0a176eff722839/numpy-2.2.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:f9f1adb22318e121c5c69a09142811a201ef17ab257a1e66ca3025065b7f53ae", size = 21176963, upload-time = "2025-05-17T21:31:19.36Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/2b/64e1affc7972decb74c9e29e5649fac940514910960ba25cd9af4488b66c/numpy-2.2.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:c820a93b0255bc360f53eca31a0e676fd1101f673dda8da93454a12e23fc5f7a", size = 14406743, upload-time = "2025-05-17T21:31:41.087Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/9f/0121e375000b5e50ffdd8b25bf78d8e1a5aa4cca3f185d41265198c7b834/numpy-2.2.6-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:3d70692235e759f260c3d837193090014aebdf026dfd167834bcba43e30c2a42", size = 5352616, upload-time = "2025-05-17T21:31:50.072Z" },
+    { url = "https://files.pythonhosted.org/packages/31/0d/b48c405c91693635fbe2dcd7bc84a33a602add5f63286e024d3b6741411c/numpy-2.2.6-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:481b49095335f8eed42e39e8041327c05b0f6f4780488f61286ed3c01368d491", size = 6889579, upload-time = "2025-05-17T21:32:01.712Z" },
+    { url = "https://files.pythonhosted.org/packages/52/b8/7f0554d49b565d0171eab6e99001846882000883998e7b7d9f0d98b1f934/numpy-2.2.6-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b64d8d4d17135e00c8e346e0a738deb17e754230d7e0810ac5012750bbd85a5a", size = 14312005, upload-time = "2025-05-17T21:32:23.332Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/dd/2238b898e51bd6d389b7389ffb20d7f4c10066d80351187ec8e303a5a475/numpy-2.2.6-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ba10f8411898fc418a521833e014a77d3ca01c15b0c6cdcce6a0d2897e6dbbdf", size = 16821570, upload-time = "2025-05-17T21:32:47.991Z" },
+    { url = "https://files.pythonhosted.org/packages/83/6c/44d0325722cf644f191042bf47eedad61c1e6df2432ed65cbe28509d404e/numpy-2.2.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:bd48227a919f1bafbdda0583705e547892342c26fb127219d60a5c36882609d1", size = 15818548, upload-time = "2025-05-17T21:33:11.728Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/9d/81e8216030ce66be25279098789b665d49ff19eef08bfa8cb96d4957f422/numpy-2.2.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:9551a499bf125c1d4f9e250377c1ee2eddd02e01eac6644c080162c0c51778ab", size = 18620521, upload-time = "2025-05-17T21:33:39.139Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/fd/e19617b9530b031db51b0926eed5345ce8ddc669bb3bc0044b23e275ebe8/numpy-2.2.6-cp311-cp311-win32.whl", hash = "sha256:0678000bb9ac1475cd454c6b8c799206af8107e310843532b04d49649c717a47", size = 6525866, upload-time = "2025-05-17T21:33:50.273Z" },
+    { url = "https://files.pythonhosted.org/packages/31/0a/f354fb7176b81747d870f7991dc763e157a934c717b67b58456bc63da3df/numpy-2.2.6-cp311-cp311-win_amd64.whl", hash = "sha256:e8213002e427c69c45a52bbd94163084025f533a55a59d6f9c5b820774ef3303", size = 12907455, upload-time = "2025-05-17T21:34:09.135Z" },
+    { url = "https://files.pythonhosted.org/packages/82/5d/c00588b6cf18e1da539b45d3598d3557084990dcc4331960c15ee776ee41/numpy-2.2.6-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:41c5a21f4a04fa86436124d388f6ed60a9343a6f767fced1a8a71c3fbca038ff", size = 20875348, upload-time = "2025-05-17T21:34:39.648Z" },
+    { url = "https://files.pythonhosted.org/packages/66/ee/560deadcdde6c2f90200450d5938f63a34b37e27ebff162810f716f6a230/numpy-2.2.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:de749064336d37e340f640b05f24e9e3dd678c57318c7289d222a8a2f543e90c", size = 14119362, upload-time = "2025-05-17T21:35:01.241Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/65/4baa99f1c53b30adf0acd9a5519078871ddde8d2339dc5a7fde80d9d87da/numpy-2.2.6-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:894b3a42502226a1cac872f840030665f33326fc3dac8e57c607905773cdcde3", size = 5084103, upload-time = "2025-05-17T21:35:10.622Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/89/e5a34c071a0570cc40c9a54eb472d113eea6d002e9ae12bb3a8407fb912e/numpy-2.2.6-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:71594f7c51a18e728451bb50cc60a3ce4e6538822731b2933209a1f3614e9282", size = 6625382, upload-time = "2025-05-17T21:35:21.414Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/35/8c80729f1ff76b3921d5c9487c7ac3de9b2a103b1cd05e905b3090513510/numpy-2.2.6-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f2618db89be1b4e05f7a1a847a9c1c0abd63e63a1607d892dd54668dd92faf87", size = 14018462, upload-time = "2025-05-17T21:35:42.174Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/3d/1e1db36cfd41f895d266b103df00ca5b3cbe965184df824dec5c08c6b803/numpy-2.2.6-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd83c01228a688733f1ded5201c678f0c53ecc1006ffbc404db9f7a899ac6249", size = 16527618, upload-time = "2025-05-17T21:36:06.711Z" },
+    { url = "https://files.pythonhosted.org/packages/61/c6/03ed30992602c85aa3cd95b9070a514f8b3c33e31124694438d88809ae36/numpy-2.2.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:37c0ca431f82cd5fa716eca9506aefcabc247fb27ba69c5062a6d3ade8cf8f49", size = 15505511, upload-time = "2025-05-17T21:36:29.965Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/25/5761d832a81df431e260719ec45de696414266613c9ee268394dd5ad8236/numpy-2.2.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:fe27749d33bb772c80dcd84ae7e8df2adc920ae8297400dabec45f0dedb3f6de", size = 18313783, upload-time = "2025-05-17T21:36:56.883Z" },
+    { url = "https://files.pythonhosted.org/packages/57/0a/72d5a3527c5ebffcd47bde9162c39fae1f90138c961e5296491ce778e682/numpy-2.2.6-cp312-cp312-win32.whl", hash = "sha256:4eeaae00d789f66c7a25ac5f34b71a7035bb474e679f410e5e1a94deb24cf2d4", size = 6246506, upload-time = "2025-05-17T21:37:07.368Z" },
+    { url = "https://files.pythonhosted.org/packages/36/fa/8c9210162ca1b88529ab76b41ba02d433fd54fecaf6feb70ef9f124683f1/numpy-2.2.6-cp312-cp312-win_amd64.whl", hash = "sha256:c1f9540be57940698ed329904db803cf7a402f3fc200bfe599334c9bd84a40b2", size = 12614190, upload-time = "2025-05-17T21:37:26.213Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/5c/6657823f4f594f72b5471f1db1ab12e26e890bb2e41897522d134d2a3e81/numpy-2.2.6-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:0811bb762109d9708cca4d0b13c4f67146e3c3b7cf8d34018c722adb2d957c84", size = 20867828, upload-time = "2025-05-17T21:37:56.699Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/9e/14520dc3dadf3c803473bd07e9b2bd1b69bc583cb2497b47000fed2fa92f/numpy-2.2.6-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:287cc3162b6f01463ccd86be154f284d0893d2b3ed7292439ea97eafa8170e0b", size = 14143006, upload-time = "2025-05-17T21:38:18.291Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/06/7e96c57d90bebdce9918412087fc22ca9851cceaf5567a45c1f404480e9e/numpy-2.2.6-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:f1372f041402e37e5e633e586f62aa53de2eac8d98cbfb822806ce4bbefcb74d", size = 5076765, upload-time = "2025-05-17T21:38:27.319Z" },
+    { url = "https://files.pythonhosted.org/packages/73/ed/63d920c23b4289fdac96ddbdd6132e9427790977d5457cd132f18e76eae0/numpy-2.2.6-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:55a4d33fa519660d69614a9fad433be87e5252f4b03850642f88993f7b2ca566", size = 6617736, upload-time = "2025-05-17T21:38:38.141Z" },
+    { url = "https://files.pythonhosted.org/packages/85/c5/e19c8f99d83fd377ec8c7e0cf627a8049746da54afc24ef0a0cb73d5dfb5/numpy-2.2.6-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f92729c95468a2f4f15e9bb94c432a9229d0d50de67304399627a943201baa2f", size = 14010719, upload-time = "2025-05-17T21:38:58.433Z" },
+    { url = "https://files.pythonhosted.org/packages/19/49/4df9123aafa7b539317bf6d342cb6d227e49f7a35b99c287a6109b13dd93/numpy-2.2.6-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1bc23a79bfabc5d056d106f9befb8d50c31ced2fbc70eedb8155aec74a45798f", size = 16526072, upload-time = "2025-05-17T21:39:22.638Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/6c/04b5f47f4f32f7c2b0e7260442a8cbcf8168b0e1a41ff1495da42f42a14f/numpy-2.2.6-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:e3143e4451880bed956e706a3220b4e5cf6172ef05fcc397f6f36a550b1dd868", size = 15503213, upload-time = "2025-05-17T21:39:45.865Z" },
+    { url = "https://files.pythonhosted.org/packages/17/0a/5cd92e352c1307640d5b6fec1b2ffb06cd0dabe7d7b8227f97933d378422/numpy-2.2.6-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b4f13750ce79751586ae2eb824ba7e1e8dba64784086c98cdbbcc6a42112ce0d", size = 18316632, upload-time = "2025-05-17T21:40:13.331Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/3b/5cba2b1d88760ef86596ad0f3d484b1cbff7c115ae2429678465057c5155/numpy-2.2.6-cp313-cp313-win32.whl", hash = "sha256:5beb72339d9d4fa36522fc63802f469b13cdbe4fdab4a288f0c441b74272ebfd", size = 6244532, upload-time = "2025-05-17T21:43:46.099Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/3b/d58c12eafcb298d4e6d0d40216866ab15f59e55d148a5658bb3132311fcf/numpy-2.2.6-cp313-cp313-win_amd64.whl", hash = "sha256:b0544343a702fa80c95ad5d3d608ea3599dd54d4632df855e4c8d24eb6ecfa1c", size = 12610885, upload-time = "2025-05-17T21:44:05.145Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/9e/4bf918b818e516322db999ac25d00c75788ddfd2d2ade4fa66f1f38097e1/numpy-2.2.6-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:0bca768cd85ae743b2affdc762d617eddf3bcf8724435498a1e80132d04879e6", size = 20963467, upload-time = "2025-05-17T21:40:44Z" },
+    { url = "https://files.pythonhosted.org/packages/61/66/d2de6b291507517ff2e438e13ff7b1e2cdbdb7cb40b3ed475377aece69f9/numpy-2.2.6-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:fc0c5673685c508a142ca65209b4e79ed6740a4ed6b2267dbba90f34b0b3cfda", size = 14225144, upload-time = "2025-05-17T21:41:05.695Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/25/480387655407ead912e28ba3a820bc69af9adf13bcbe40b299d454ec011f/numpy-2.2.6-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:5bd4fc3ac8926b3819797a7c0e2631eb889b4118a9898c84f585a54d475b7e40", size = 5200217, upload-time = "2025-05-17T21:41:15.903Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/4a/6e313b5108f53dcbf3aca0c0f3e9c92f4c10ce57a0a721851f9785872895/numpy-2.2.6-cp313-cp313t-macosx_14_0_x86_64.whl", hash = "sha256:fee4236c876c4e8369388054d02d0e9bb84821feb1a64dd59e137e6511a551f8", size = 6712014, upload-time = "2025-05-17T21:41:27.321Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/30/172c2d5c4be71fdf476e9de553443cf8e25feddbe185e0bd88b096915bcc/numpy-2.2.6-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e1dda9c7e08dc141e0247a5b8f49cf05984955246a327d4c48bda16821947b2f", size = 14077935, upload-time = "2025-05-17T21:41:49.738Z" },
+    { url = "https://files.pythonhosted.org/packages/12/fb/9e743f8d4e4d3c710902cf87af3512082ae3d43b945d5d16563f26ec251d/numpy-2.2.6-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f447e6acb680fd307f40d3da4852208af94afdfab89cf850986c3ca00562f4fa", size = 16600122, upload-time = "2025-05-17T21:42:14.046Z" },
+    { url = "https://files.pythonhosted.org/packages/12/75/ee20da0e58d3a66f204f38916757e01e33a9737d0b22373b3eb5a27358f9/numpy-2.2.6-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:389d771b1623ec92636b0786bc4ae56abafad4a4c513d36a55dce14bd9ce8571", size = 15586143, upload-time = "2025-05-17T21:42:37.464Z" },
+    { url = "https://files.pythonhosted.org/packages/76/95/bef5b37f29fc5e739947e9ce5179ad402875633308504a52d188302319c8/numpy-2.2.6-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:8e9ace4a37db23421249ed236fdcdd457d671e25146786dfc96835cd951aa7c1", size = 18385260, upload-time = "2025-05-17T21:43:05.189Z" },
+    { url = "https://files.pythonhosted.org/packages/09/04/f2f83279d287407cf36a7a8053a5abe7be3622a4363337338f2585e4afda/numpy-2.2.6-cp313-cp313t-win32.whl", hash = "sha256:038613e9fb8c72b0a41f025a7e4c3f0b7a1b5d768ece4796b674c8f3fe13efff", size = 6377225, upload-time = "2025-05-17T21:43:16.254Z" },
+    { url = "https://files.pythonhosted.org/packages/67/0e/35082d13c09c02c011cf21570543d202ad929d961c02a147493cb0c2bdf5/numpy-2.2.6-cp313-cp313t-win_amd64.whl", hash = "sha256:6031dd6dfecc0cf9f668681a37648373bddd6421fff6c66ec1624eed0180ee06", size = 12771374, upload-time = "2025-05-17T21:43:35.479Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/3b/d94a75f4dbf1ef5d321523ecac21ef23a3cd2ac8b78ae2aac40873590229/numpy-2.2.6-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0b605b275d7bd0c640cad4e5d30fa701a8d59302e127e5f79138ad62762c3e3d", size = 21040391, upload-time = "2025-05-17T21:44:35.948Z" },
+    { url = "https://files.pythonhosted.org/packages/17/f4/09b2fa1b58f0fb4f7c7963a1649c64c4d315752240377ed74d9cd878f7b5/numpy-2.2.6-pp310-pypy310_pp73-macosx_14_0_x86_64.whl", hash = "sha256:7befc596a7dc9da8a337f79802ee8adb30a552a94f792b9c9d18c840055907db", size = 6786754, upload-time = "2025-05-17T21:44:47.446Z" },
+    { url = "https://files.pythonhosted.org/packages/af/30/feba75f143bdc868a1cc3f44ccfa6c4b9ec522b36458e738cd00f67b573f/numpy-2.2.6-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ce47521a4754c8f4593837384bd3424880629f718d87c5d44f8ed763edd63543", size = 16643476, upload-time = "2025-05-17T21:45:11.871Z" },
+    { url = "https://files.pythonhosted.org/packages/37/48/ac2a9584402fb6c0cd5b5d1a91dcf176b15760130dd386bbafdbfe3640bf/numpy-2.2.6-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:d042d24c90c41b54fd506da306759e06e568864df8ec17ccc17e9e884634fd00", size = 12812666, upload-time = "2025-05-17T21:45:31.426Z" },
+]
+
+[[package]]
+name = "numpy"
+version = "2.4.6"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/d0/ad/fed0499ce6a338d2a03ebae59cd15093910c8875328855781952abf6c2fe/numpy-2.4.6.tar.gz", hash = "sha256:f3a3570c4a2a16746ac2c31a7c7c7b0c186b95ce902e33db6f28094ed7387dda", size = 20735807, upload-time = "2026-05-18T23:37:14.07Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b3/49/ec46835a70be8fa6446c495126ac84fdb28cb2558e1620ffb87a10c8b64c/numpy-2.4.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:0280e0356c0829a18d9de1cb7eee50ec22ca639878d7240307ca0943d73cd2c4", size = 16969194, upload-time = "2026-05-18T23:33:13.503Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/0d/f5957185c0ee2f3e12f78715aa9e3b353fd83633316c8532b38faa37e3f6/numpy-2.4.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:110f8b71aacb688ec69062bb7f6938a0f8acb01b7c1c4beb453c65b6d234584d", size = 14964111, upload-time = "2026-05-18T23:33:17.795Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/40/40a40ee0ddf7ceb782c49af278894b686e586d65d8c1889c8b5da01a3d7d/numpy-2.4.6-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:4cfe66903cc32a9921a6733d96b19bb6abf310397581bbad89c228f5abaf0ee8", size = 5469159, upload-time = "2026-05-18T23:33:20.654Z" },
+    { url = "https://files.pythonhosted.org/packages/63/13/f9a8046535cb21deae82f8d03de9617e08882d274fad2539630761888228/numpy-2.4.6-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:8155154c7c691289fe18f510b5d4657c68c67989f293f0535a91360392ff6538", size = 6798936, upload-time = "2026-05-18T23:33:22.987Z" },
+    { url = "https://files.pythonhosted.org/packages/33/a8/6fa8c1a345a8c85dbb21932c447bee07c30a2c2a3f31e369c0a84b300147/numpy-2.4.6-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0ab0a9c4ffb1a6d95ef519fe4247dba8eb6b18ad93999f76b7f657039acabd47", size = 15966692, upload-time = "2026-05-18T23:33:26.62Z" },
+    { url = "https://files.pythonhosted.org/packages/02/03/74fe2a4cb3817d94d86402f2506554130a2f01414e299b5a843e5a8a957f/numpy-2.4.6-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:89cd468399cfd2504718f0ba50e410dca55a170b61a02ad92bb18c8a65186e93", size = 16918164, upload-time = "2026-05-18T23:33:29.955Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/80/3615be3313f7e7696609bc194b9f0101da809df79e859bdb84e0cd043f46/numpy-2.4.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c2d37ab77531417474168eb79d6d80b14f821a966818505d03013d0833edb7a8", size = 17322877, upload-time = "2026-05-18T23:33:34.724Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/ac/a691e0fe2675e370d0e08ff905adc49a1c8830e8cae03efe4477e92cd55d/numpy-2.4.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:f407cb6b8e9d6d8c626bc73c945db1706035af8fd632295547bf1c9e46d092d6", size = 18651487, upload-time = "2026-05-18T23:33:38.217Z" },
+    { url = "https://files.pythonhosted.org/packages/15/a7/9bc1cd626d7bf6869bfedf27b91b6ab5dd607758bf8e959d6fa80c6a59cb/numpy-2.4.6-cp311-cp311-win32.whl", hash = "sha256:ddea102b48f9e339f3948bf22040944184627a30fdf7f858667673b9c5f033c8", size = 6233945, upload-time = "2026-05-18T23:33:41.331Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/31/7fc6239c12bce7e931463251cca4426c465e1876ba3cc785402ef4dd8f4e/numpy-2.4.6-cp311-cp311-win_amd64.whl", hash = "sha256:1e254a00cdf42b1e4d5b3d68d33af63268d41340d8885df2ab6470f2e1500147", size = 12608406, upload-time = "2026-05-18T23:33:44.131Z" },
+    { url = "https://files.pythonhosted.org/packages/27/83/140f85a466595a16382996a1bf06b2b54bcd597488921b0c9daaeeda72af/numpy-2.4.6-cp311-cp311-win_arm64.whl", hash = "sha256:ed9749eef4cbd126da3dc1d6bcb3a57f5eb7ac6a6484146bdbf743f552dfc577", size = 10479528, upload-time = "2026-05-18T23:33:50.725Z" },
+    { url = "https://files.pythonhosted.org/packages/95/2a/3d7b5ac8aac24feaf9ad7ed58f45b0bbc06d37e4338ae84c9f2298b570f9/numpy-2.4.6-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:001fbb8e08d942dd57599e781f2472269ee7f2755fae407b4f67b2f0b17da3f1", size = 16689119, upload-time = "2026-05-18T23:33:54.065Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/12/92c4c131527599e8288d6918e888d88726f84d805d784b771f32408aeaef/numpy-2.4.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ebfb099f8dcf083deef3ac1ca4c1503f387cf76296fcb3816b66f5ecb5f54fdb", size = 14699246, upload-time = "2026-05-18T23:33:57.621Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/fe/c0a6b7b2ca128a8fb228575147073b660656734b8ebe4d76c8fd748dcc79/numpy-2.4.6-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:3213d622a0283a39a93d188f3cf72b26862df52fbb4ca3697f51705016523d41", size = 5204410, upload-time = "2026-05-18T23:34:00.302Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/d4/9770d14ba719432bb90a421bfd443872ed0f70f7264b64bec12ea363d5fd/numpy-2.4.6-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:357cc07a6d7b0b182ff02249616a03742827ebb1277546b5c7cd7f7620a45698", size = 6551240, upload-time = "2026-05-18T23:34:02.852Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/c6/50a46a6205feba2343f1d6d17438107c5dc491ed1c736e6ea68689fd906b/numpy-2.4.6-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5f9fb9157b4ce2971008323afe46053787b526ef624fea915b261468a8421a0f", size = 15671012, upload-time = "2026-05-18T23:34:05.485Z" },
+    { url = "https://files.pythonhosted.org/packages/99/60/14115e6364fa676c5397c2ad3004e527e9aa487abf5d0706ec81bbd08529/numpy-2.4.6-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:90f9849678c75fe7afa2d348ac842c168b0a4d3d61919687216dfc547976d853", size = 16645538, upload-time = "2026-05-18T23:34:09.265Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/c5/693cbe59e57db94d2231fa519ca3978dc9e19da5a8f088588f5c6e947ff2/numpy-2.4.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:c1a2af6c6ef86344a6b0db6b97834208bf598db514f2b155042439b62605601a", size = 17020706, upload-time = "2026-05-18T23:34:13.053Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/fc/85b7c4eff9b4966ade25c2273cf7e7012e92366c032058653934b37de044/numpy-2.4.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e5805d5a22fd19c8ccff10a9561f9df94436b0545619ea579db2d3c35294bce2", size = 18368541, upload-time = "2026-05-18T23:34:17.024Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/81/e1b27545deedce7f4a0b348618c6b62d74e36a4dc9ccd42f3eb2f85eee32/numpy-2.4.6-cp312-cp312-win32.whl", hash = "sha256:e3eeb0aabd6bd5ce64faae67e9935203a6991b4bc2a485a767fbafb2c5125f45", size = 5962825, upload-time = "2026-05-18T23:34:20.3Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/ca/feab00bd44aa5fe1ad2c18f08b4d3bb92e26484b0b1d1443897809ed528c/numpy-2.4.6-cp312-cp312-win_amd64.whl", hash = "sha256:d8e8286dd7cea7895157318d1b91cdacac64c479f3cbc8dce548331728484751", size = 12321687, upload-time = "2026-05-18T23:34:23.095Z" },
+    { url = "https://files.pythonhosted.org/packages/63/cf/5a6d34850a39d1093558564f77ee8e8e0bee5061151b8f05a55711001ec7/numpy-2.4.6-cp312-cp312-win_arm64.whl", hash = "sha256:4081eb135ac24158bd51cdfbef16f1c64df7063b1143f24731387137c092bec8", size = 10221482, upload-time = "2026-05-18T23:34:25.876Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/82/bdab26d7438c6791ca31b7c024ca37c1eab8b726ba236129005cd4a06e45/numpy-2.4.6-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:511dbaf848decaaaf4b4ca48032619fb3138710c4bf7da7617765edad1ef96b0", size = 16684648, upload-time = "2026-05-18T23:34:29.41Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/30/a80189bcc7f5e4258b3fbc3968d909d1756f54d023299ecc39ad6fdb9ef8/numpy-2.4.6-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:bf162abab1c1a736333192707cef898e735a5ca00f38f27eeedf44b39d9e85eb", size = 14693902, upload-time = "2026-05-18T23:34:33.013Z" },
+    { url = "https://files.pythonhosted.org/packages/97/12/70b5d0d7c15e1ebb8a6a84a8caa1d19e181d84fb58bb6d70aca29099dec1/numpy-2.4.6-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:043191bfa8eab18c776647b62723ac9dddece59743b13f49b2016094129c2b3f", size = 5198992, upload-time = "2026-05-18T23:34:36.132Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/8c/ebd2a8f8a83541f8d38cc5667e8c2b69cecfd30da6e45693e8158857d44b/numpy-2.4.6-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:6180d8b35af935aed8ece3a85e0a43f87393ae0ac87c8d2c8bd2c993f7270ef3", size = 6546944, upload-time = "2026-05-18T23:34:38.484Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/c5/7b863a97a91671a0338f4253bd3b5a3d3852f0692dae91711c9f4a10e787/numpy-2.4.6-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:72fbe16c6fac95aedf5937fa873445cec2110be35d8a4e9433d7501fd98dae6b", size = 15669392, upload-time = "2026-05-18T23:34:41.257Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/9d/3584b9984ca4c047aea75214ce1a4c4c73d849bd71b604264b7f5653f8a8/numpy-2.4.6-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a7830bab239b79cda9c08c2da014761cafb48da6150e1da17ac06283f43b6089", size = 16633220, upload-time = "2026-05-18T23:34:45.075Z" },
+    { url = "https://files.pythonhosted.org/packages/05/ae/7c67fba23bd98caec7c99261f3a16072ade14813486b0282cb29846de832/numpy-2.4.6-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ef4aea96ce4d3b074422cb4f2f64e216bf9e213004bb58ecfdf50ea02ea8eb9a", size = 17020800, upload-time = "2026-05-18T23:34:49.065Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/5d/3b6725cb31d983c5e66916f5d36f6d7e5521129e4c4404d64f918292a5b6/numpy-2.4.6-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:dfa20cc6ca228e6b155b11da03825975ce66aea520985dbbddf0f2a5a495c605", size = 18357600, upload-time = "2026-05-18T23:34:52.709Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/da/2ccc6c2fe8898dee01d90c75c5f5f914a23daf99e3e0f59516a08760c8b5/numpy-2.4.6-cp313-cp313-win32.whl", hash = "sha256:56b39e5e0622a09a25bf5baf62f4bcf0cb8a41ae6e2819cf49bbc5a74c083f91", size = 5961134, upload-time = "2026-05-18T23:34:55.618Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/cd/9cc4dc876fb065d5c220aae4d5e14826b2715331bb7618ce1fb07a679d99/numpy-2.4.6-cp313-cp313-win_amd64.whl", hash = "sha256:c4fc99836233ea196540b17ab0983aff60ed07941751930f5f4d05bc3b3b7359", size = 12318598, upload-time = "2026-05-18T23:34:58.928Z" },
+    { url = "https://files.pythonhosted.org/packages/39/1e/c0bcba1f8694116485fe28fd1be698c278fcda4141c5b0e53a2aed8b12a8/numpy-2.4.6-cp313-cp313-win_arm64.whl", hash = "sha256:a7c711e21628b52034bb5ab8d1bce291f752fcc5e92accc615778acee1ff4778", size = 10222272, upload-time = "2026-05-18T23:35:02.167Z" },
+    { url = "https://files.pythonhosted.org/packages/63/6d/cc5619247c8f4204e507f5883528372e4ac4bb189e579fb859a12e480b1f/numpy-2.4.6-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:112b06a867b235ef466ed3508ddf0238050df9c727cafb5301ac385b899189a1", size = 14821197, upload-time = "2026-05-18T23:35:05.468Z" },
+    { url = "https://files.pythonhosted.org/packages/00/58/f1c39161c87d9e9bed660f1ed4bafc0e403d5ec9650b6dd77aead07d489b/numpy-2.4.6-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:eaf7fa2de5c0be8ae6ff8e9bea2ccd725e980541244521d8d4b5f3354a27babe", size = 5326287, upload-time = "2026-05-18T23:35:08.693Z" },
+    { url = "https://files.pythonhosted.org/packages/af/57/3917ab0fd97f271a8694513581b8a36c655f111c446852c302f04ccdb6fc/numpy-2.4.6-cp313-cp313t-macosx_14_0_x86_64.whl", hash = "sha256:7265a2f3d436e54ef9f2b52b5c937e6be778781bd97a590319d7348f1c1ca997", size = 6646763, upload-time = "2026-05-18T23:35:11.459Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/0f/037e64c494b67581ae18193d770adef354c41f3f2c8ebf865602d949bf8f/numpy-2.4.6-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f74a575920ab21fe304421a3fc28793d82e299cae9eccb37084e9fc7f3617c20", size = 15728070, upload-time = "2026-05-18T23:35:14.79Z" },
+    { url = "https://files.pythonhosted.org/packages/21/a6/5d2bae9c9542eb4df16dc9c46dc79c186e9bad53805dfa5399a6023c6db0/numpy-2.4.6-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ede83e07a75dd06bc501566c1eca2afc0d61677c1472ac9ad93fdee6e638a48d", size = 16681752, upload-time = "2026-05-18T23:35:18.836Z" },
+    { url = "https://files.pythonhosted.org/packages/92/14/23d1dfb410ae362cd59ce53e936b1513d545eb40db3949ced632e19a459e/numpy-2.4.6-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:68bb27509ac1b9a3443094260f6326150663b06abe40b73a2f81160623da5b67", size = 17086024, upload-time = "2026-05-18T23:35:22.52Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/6e/23595a2c642cdf3bc567877064bdd7f91c8b0038a4453cf2daf7248eafe9/numpy-2.4.6-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:a0df0043bdb289bde1f62da130d20df23d58b45429f752bc7a8fc5325a225ecd", size = 18403398, upload-time = "2026-05-18T23:35:26.398Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/90/0ac3bc947217e66dec77e7cbc6a1979d1af70b6461b82f620d3bccd5e4c8/numpy-2.4.6-cp313-cp313t-win32.whl", hash = "sha256:29a287e0cf63ff528da061de6b9f64a4618da591ca1046aafc54062e40ca7eab", size = 6084971, upload-time = "2026-05-18T23:35:29.387Z" },
+    { url = "https://files.pythonhosted.org/packages/77/71/5673e351671a1d2bd6063b91b44f70c0affea7d1516fa7a6572941ba4aa1/numpy-2.4.6-cp313-cp313t-win_amd64.whl", hash = "sha256:25c692919ac5a01f170a3bfcd62d745b24fd095c353d50812637d6fcab442e75", size = 12458532, upload-time = "2026-05-18T23:35:32.175Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/88/19d3503c5046e688f049274b27a3ef3d771152fa80d3ba3d01a3dff61abe/numpy-2.4.6-cp313-cp313t-win_arm64.whl", hash = "sha256:1e978ec1e8bd0e0e4de6bb75de9d30cbb74db6b6a2bb727618613703ca0167dd", size = 10291881, upload-time = "2026-05-18T23:35:35.465Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/91/3ab2044d05fd16d343c5ac2e69b127f1b2854040dd20b193257c78028bd3/numpy-2.4.6-cp314-cp314-macosx_10_15_x86_64.whl", hash = "sha256:06ca2f61ec4385a07a6977c55ba998a4466c123642b4a32694d3128fce18c079", size = 16683458, upload-time = "2026-05-18T23:35:38.353Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/62/764ce66fa4147ae6d73071a3abf804ffe606f174618697c571acdf26a7c9/numpy-2.4.6-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:38efbc8de75c7a0fc1ac190162d892787f3f47b57cc291231aafee36b80982b7", size = 14704559, upload-time = "2026-05-18T23:35:42.14Z" },
+    { url = "https://files.pythonhosted.org/packages/60/61/23f27c172f022e04025b7dc2367f4d63c1a398120607ec896228649a6f48/numpy-2.4.6-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:d581b735e177fdcdce6fed8e7e8880a3fb6ee4e3653a3ac6af01c6f4c03effc5", size = 5209716, upload-time = "2026-05-18T23:35:45.377Z" },
+    { url = "https://files.pythonhosted.org/packages/03/71/21cf70dc6ea3e3acb95fc53a265b2fc248b981f0194ceb5b475271b8809d/numpy-2.4.6-cp314-cp314-macosx_14_0_x86_64.whl", hash = "sha256:0a041d3d761dc3c35cc56ce0351506a02bcbc25f7b169f652435141a17db9096", size = 6543947, upload-time = "2026-05-18T23:35:47.926Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/91/64288395ee1799bd2e0b04a305dce9666da90c961e1f3fe982a05ee1c036/numpy-2.4.6-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:40fdc1ae7125e518ea98e53e69a4ebc27e1fd50510c47b7ea130cf21e5e1d42b", size = 15685197, upload-time = "2026-05-18T23:35:50.863Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/eb/ebffaa97dc55502df69584a8f0dcf07f69a3e0b3e2323670a2722db9aa39/numpy-2.4.6-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a2c306dea656c12c68f51f4cea133cbe78ca7435eb28c735eac1d3ebe73be6e8", size = 16638245, upload-time = "2026-05-18T23:35:54.752Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/0b/54f9da33128d7e350fab89c7455902eeae70349ee52bddb448dc4a576f45/numpy-2.4.6-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:33111801a01c12a8a1e3721f0a9232f8cfc8ae2c6b7098167e6f623c6073f402", size = 17036587, upload-time = "2026-05-18T23:35:58.355Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/f0/fdebc1052db1cc37c64beb22072d67cd6d1c71adca1299f53dec2b5e20d3/numpy-2.4.6-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:ae506e6902902557576a26ff33eda8695e7ecb3cb36c3b573a0765dee114ebdb", size = 18363226, upload-time = "2026-05-18T23:36:02.845Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/b4/298628d98c72b57e57f7165ae6a481a1deaf6f3c28262a6e4c739c275930/numpy-2.4.6-cp314-cp314-win32.whl", hash = "sha256:aaf159caa35993cb1f56fb9b8e4610d35758e7ca005412eb1daa856a78c9c4b1", size = 6010196, upload-time = "2026-05-18T23:36:05.92Z" },
+    { url = "https://files.pythonhosted.org/packages/df/ac/46de6dda46478f7942f839e094970be2d4a861e005c4b3bf07c92e291a09/numpy-2.4.6-cp314-cp314-win_amd64.whl", hash = "sha256:b507f5c4c1d508876d1819b6bf9a49d365b96320b5d4993426b33a23ca4b8261", size = 12450334, upload-time = "2026-05-18T23:36:09.107Z" },
+    { url = "https://files.pythonhosted.org/packages/78/92/b8b798ac784102c0da830d2257d59358e3d3d90d1e2b3f2575dad976c5cf/numpy-2.4.6-cp314-cp314-win_arm64.whl", hash = "sha256:6f41ae150c4e32db4f3310cdaf64b1593a03dbabe29eec77fc9b50fe64061df6", size = 10495678, upload-time = "2026-05-18T23:36:12.766Z" },
+    { url = "https://files.pythonhosted.org/packages/30/34/ec28d1aa8115971537c01469ab2011ee96827930f0a124de1000cc2a7ed7/numpy-2.4.6-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:ece3d2cfe132e7d51f44a832b303895e6f2d499c5e74dfbdb06ee246147a304a", size = 14823672, upload-time = "2026-05-18T23:36:16.473Z" },
+    { url = "https://files.pythonhosted.org/packages/16/bd/f6d1fede4e54e8042a7ff97bb495510f3c220f94bcd9e8b228e87c92cc0d/numpy-2.4.6-cp314-cp314t-macosx_14_0_arm64.whl", hash = "sha256:e3e5193ef5a3dc73bceee50f7fdc2c90dbb76c42df8d8fae3d1067a583df579e", size = 5328731, upload-time = "2026-05-18T23:36:19.767Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/f0/e105b9e2fd728a9910103884decd6951d9dd73896b914a98d9a231de02ee/numpy-2.4.6-cp314-cp314t-macosx_14_0_x86_64.whl", hash = "sha256:17f9ade344e7d9b464a084d69bcf18fc691cb1db67c62ed80820bf4926d78f0e", size = 6649805, upload-time = "2026-05-18T23:36:22.266Z" },
+    { url = "https://files.pythonhosted.org/packages/82/dd/1206a7ca6ab15e3f02069707ca96222e202af681bb73756da7527f3cb837/numpy-2.4.6-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9cd5ffd25db4e7ba6a375693b3fc0fc1791ec636c17db3720da19bde7180ec43", size = 15730496, upload-time = "2026-05-18T23:36:25.713Z" },
+    { url = "https://files.pythonhosted.org/packages/51/e7/38d3ea825dcab85a591734decb2f6c67caa7c8367d374df1a1c3842f9b07/numpy-2.4.6-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7d92c3819208a60205a12a245c91ad70cb0a85336659b19b834205573ac8456e", size = 16679616, upload-time = "2026-05-18T23:36:29.652Z" },
+    { url = "https://files.pythonhosted.org/packages/93/b7/caabfdf53edf663e0b4eb74d7d405d83baef09eb5e83bcd32d601d72b93e/numpy-2.4.6-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:e85b752a1e912b70eaad4fafbd4d1238007ab221de2009b9a2f5ae7461239895", size = 17085145, upload-time = "2026-05-18T23:36:33.449Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/45/68d7c33a6bcf3e5aa3bdbd57a367e6f615286dfd6482f97e8ffeb734306e/numpy-2.4.6-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:29cb7f67d10b479ff07c17d33e39f78c07f71c40ef30d63c153d340e96cd3fb4", size = 18403813, upload-time = "2026-05-18T23:36:37.369Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/50/0753655aa844c99cd9e018aacf76f130f1bd81d881bb74bc0aef5d73a8ba/numpy-2.4.6-cp314-cp314t-win32.whl", hash = "sha256:260a5d70215b61ab4fadf5c7baacd64821842975eea312125ed3c39a6391b063", size = 6156982, upload-time = "2026-05-18T23:36:40.817Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/d4/7c67becf668f973cb490cec3e98dfd799d866f9c989a54d355672cfa0db6/numpy-2.4.6-cp314-cp314t-win_amd64.whl", hash = "sha256:81a1cca95ed5bb92aa8b10dd2cdc9a0d3853a50fad926c28b5d7e8ea54389627", size = 12638908, upload-time = "2026-05-18T23:36:43.996Z" },
+    { url = "https://files.pythonhosted.org/packages/43/bb/e1c71a4295b1b1d1393d50dbb4f2a36283c6859d9d3892e84f00ec5a91d5/numpy-2.4.6-cp314-cp314t-win_arm64.whl", hash = "sha256:0c9136e14ed34a9e343a31c533d78a9813a69a3148332bce5e9821cb2f996e66", size = 10565867, upload-time = "2026-05-18T23:36:47.114Z" },
+    { url = "https://files.pythonhosted.org/packages/de/12/b422cc84439adc0d00de605bf4a308890ae5c26f2c71fbd73e5d08fbb0dd/numpy-2.4.6-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:55cced7c52e981362f708ad635198e97a752dfba412cc03c23bbf3bd8d5cd662", size = 16847511, upload-time = "2026-05-18T23:36:50.673Z" },
+    { url = "https://files.pythonhosted.org/packages/44/53/f481bef68011740f8849418d82db07230e825013f31f4eef5ba5b805316a/numpy-2.4.6-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:d6da64deb6b8ed903e7560180a92f2d804ee1ba5eeb849ac2748b8c1aba1f6d7", size = 14889064, upload-time = "2026-05-18T23:36:53.879Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/57/42ed575c10ced8af951d426bc4e1f8aff16fd851db33f067036215a7f860/numpy-2.4.6-pp311-pypy311_pp73-macosx_14_0_arm64.whl", hash = "sha256:68a5124b13fa6cc2086764a20005d30bc0548146f7f5322f02fce212ca14317f", size = 5394157, upload-time = "2026-05-18T23:36:57.194Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/ef/f66cc724fcc36c1e364c67f51ae9146090b8b584f27d58b97fdae3edd737/numpy-2.4.6-pp311-pypy311_pp73-macosx_14_0_x86_64.whl", hash = "sha256:948424b06129ce883307e8cff868c31396d8dc7630a59c61d70d98dbe70f222c", size = 6708728, upload-time = "2026-05-18T23:36:59.575Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/9c/c531f2293b91265d8b48e9b329f54fdd7ffae73cb4134ea10cca4237e9cc/numpy-2.4.6-pp311-pypy311_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5dbbdb29840ca3d91ee0fece42fc29278886d908280bfec0a5846c6f901a3eb0", size = 15798374, upload-time = "2026-05-18T23:37:02.674Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/b0/413077f6b1153ed3cba361401c6783bbad6114804a000cc22eb71c13e190/numpy-2.4.6-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8ad03c0965fb3c692200e74d458ca28c1dbb4ce96f9a479a8aa041ad5fabca02", size = 16747286, upload-time = "2026-05-18T23:37:06.327Z" },
+    { url = "https://files.pythonhosted.org/packages/15/ce/e5ec180bc41812edcd8daeb8639d205622c0e8c02259d8ab25a0201b3c2a/numpy-2.4.6-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:2803abfebfc990042cd494d8ce2d5f82e9d847af6d35ec486923aa19dbad5e73", size = 12504263, upload-time = "2026-05-18T23:37:09.715Z" },
+]
+
+[[package]]
+name = "nvidia-cublas"
+version = "13.1.1.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-cuda-nvrtc" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a7/a1/0bd24ee8c8d03adac032fd2909426a00c88f8c57961b1277ded97f91119f/nvidia_cublas-13.1.1.3-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:b7a210458267ac818974c53038fbec2e969d5c99f305ab15c72522fa9f001dd5", size = 542848918, upload-time = "2026-04-08T18:46:22.985Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/cd/154ca20c38269e05eff77c1464e6c1da89f50a6390b565e9d82e06bc11e1/nvidia_cublas-13.1.1.3-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:37936a16db8fe4ac1f065c2139360608a543a09275cb1a1af612e08cfa065436", size = 423138758, upload-time = "2026-04-08T18:46:58.655Z" },
+]
+
+[[package]]
+name = "nvidia-cuda-cupti"
+version = "13.0.85"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2a/2a/80353b103fc20ce05ef51e928daed4b6015db4aaa9162ed0997090fe2250/nvidia_cuda_cupti-13.0.85-py3-none-manylinux_2_25_aarch64.whl", hash = "sha256:796bd679890ee55fb14a94629b698b6db54bcfd833d391d5e94017dd9d7d3151", size = 10310827, upload-time = "2025-09-04T08:26:42.012Z" },
+    { url = "https://files.pythonhosted.org/packages/33/6d/737d164b4837a9bbd202f5ae3078975f0525a55730fe871d8ed4e3b952b0/nvidia_cuda_cupti-13.0.85-py3-none-manylinux_2_25_x86_64.whl", hash = "sha256:4eb01c08e859bf924d222250d2e8f8b8ff6d3db4721288cf35d14252a4d933c8", size = 10715597, upload-time = "2025-09-04T08:26:51.312Z" },
+]
+
+[[package]]
+name = "nvidia-cuda-nvrtc"
+version = "13.0.88"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c3/68/483a78f5e8f31b08fb1bb671559968c0ca3a065ac7acabfc7cee55214fd6/nvidia_cuda_nvrtc-13.0.88-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:ad9b6d2ead2435f11cbb6868809d2adeeee302e9bb94bcf0539c7a40d80e8575", size = 90215200, upload-time = "2025-09-04T08:28:44.204Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/dc/6bb80850e0b7edd6588d560758f17e0550893a1feaf436807d64d2da040f/nvidia_cuda_nvrtc-13.0.88-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d27f20a0ca67a4bb34268a5e951033496c5b74870b868bacd046b1b8e0c3267b", size = 43015449, upload-time = "2025-09-04T08:28:20.239Z" },
+]
+
+[[package]]
+name = "nvidia-cuda-runtime"
+version = "13.0.96"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/87/4f/17d7b9b8e285199c58ce28e31b5c5bbaa4d8271af06a89b6405258245de2/nvidia_cuda_runtime-13.0.96-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ef9bcbe90493a2b9d810e43d249adb3d02e98dd30200d86607d8d02687c43f55", size = 2261060, upload-time = "2025-10-09T08:55:15.78Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/24/d1558f3b68b1d26e706813b1d10aa1d785e4698c425af8db8edc3dced472/nvidia_cuda_runtime-13.0.96-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7f82250d7782aa23b6cfe765ecc7db554bd3c2870c43f3d1821f1d18aebf0548", size = 2243632, upload-time = "2025-10-09T08:55:36.117Z" },
+]
+
+[[package]]
+name = "nvidia-cudnn-cu13"
+version = "9.20.0.48"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-cublas" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/56/c5/83384d846b2fd17c44bd499b36c75a45ed4f095fbbb2252294e89cea5c5c/nvidia_cudnn_cu13-9.20.0.48-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:e31454ae00094b0c55319d9d15b6fa2fc50a9e1c0f5c8c80fb75258234e731e1", size = 444574296, upload-time = "2026-03-09T19:28:27.751Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/5e/edb9c0ae051602c3ccaffe424256463636d639e27d7f302dde9975ef9e7a/nvidia_cudnn_cu13-9.20.0.48-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:0c45dd8eeb50b603f07995b1b300c62ffe6a1980482b82b3bcf94a4ca9d49304", size = 366173588, upload-time = "2026-03-09T19:29:34.474Z" },
+]
+
+[[package]]
+name = "nvidia-cufft"
+version = "12.0.0.61"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-nvjitlink" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8b/ae/f417a75c0259e85c1d2f83ca4e960289a5f814ed0cea74d18c353d3e989d/nvidia_cufft-12.0.0.61-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:2708c852ef8cd89d1d2068bdbece0aa188813a0c934db3779b9b1faa8442e5f5", size = 214053554, upload-time = "2025-09-04T08:31:38.196Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/2f/7b57e29836ea8714f81e9898409196f47d772d5ddedddf1592eadb8ab743/nvidia_cufft-12.0.0.61-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:6c44f692dce8fd5ffd3e3df134b6cdb9c2f72d99cf40b62c32dde45eea9ddad3", size = 214085489, upload-time = "2025-09-04T08:31:56.044Z" },
+]
+
+[[package]]
+name = "nvidia-cufile"
+version = "1.15.1.6"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3f/70/4f193de89a48b71714e74602ee14d04e4019ad36a5a9f20c425776e72cd6/nvidia_cufile-1.15.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:08a3ecefae5a01c7f5117351c64f17c7c62efa5fffdbe24fc7d298da19cd0b44", size = 1223672, upload-time = "2025-09-04T08:32:22.779Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/73/cc4a14c9813a8a0d509417cf5f4bdaba76e924d58beb9864f5a7baceefbf/nvidia_cufile-1.15.1.6-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:bdc0deedc61f548bddf7733bdc216456c2fdb101d020e1ab4b88d232d5e2f6d1", size = 1136992, upload-time = "2025-09-04T08:32:14.119Z" },
+]
+
+[[package]]
+name = "nvidia-curand"
+version = "10.4.0.35"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1e/72/7c2ae24fb6b63a32e6ae5d241cc65263ea18d08802aaae087d9f013335a2/nvidia_curand-10.4.0.35-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:133df5a7509c3e292aaa2b477afd0194f06ce4ea24d714d616ff36439cee349a", size = 61962106, upload-time = "2025-08-04T10:21:41.128Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/9f/be0a41ca4a4917abf5cb9ae0daff1a6060cc5de950aec0396de9f3b52bc5/nvidia_curand-10.4.0.35-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:1aee33a5da6e1db083fe2b90082def8915f30f3248d5896bcec36a579d941bfc", size = 59544258, upload-time = "2025-08-04T10:22:03.992Z" },
+]
+
+[[package]]
+name = "nvidia-cusolver"
+version = "12.0.4.66"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-cublas" },
+    { name = "nvidia-cusparse" },
+    { name = "nvidia-nvjitlink" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c8/c3/b30c9e935fc01e3da443ec0116ed1b2a009bb867f5324d3f2d7e533e776b/nvidia_cusolver-12.0.4.66-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:02c2457eaa9e39de20f880f4bd8820e6a1cfb9f9a34f820eb12a155aa5bc92d2", size = 223467760, upload-time = "2025-09-04T08:33:04.222Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/67/cba3777620cdacb99102da4042883709c41c709f4b6323c10781a9c3aa34/nvidia_cusolver-12.0.4.66-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:0a759da5dea5c0ea10fd307de75cdeb59e7ea4fcb8add0924859b944babf1112", size = 200941980, upload-time = "2025-09-04T08:33:22.767Z" },
+]
+
+[[package]]
+name = "nvidia-cusparse"
+version = "12.6.3.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-nvjitlink" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f8/94/5c26f33738ae35276672f12615a64bd008ed5be6d1ebcb23579285d960a9/nvidia_cusparse-12.6.3.3-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:80bcc4662f23f1054ee334a15c72b8940402975e0eab63178fc7e670aa59472c", size = 162155568, upload-time = "2025-09-04T08:33:42.864Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/18/623c77619c31d62efd55302939756966f3ecc8d724a14dab2b75f1508850/nvidia_cusparse-12.6.3.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:2b3c89c88d01ee0e477cb7f82ef60a11a4bcd57b6b87c33f789350b59759360b", size = 145942937, upload-time = "2025-09-04T08:33:58.029Z" },
+]
+
+[[package]]
+name = "nvidia-cusparselt-cu13"
+version = "0.8.1"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/46/e1/cdc1797eadf82d3a9a575a19b33fdc871a97edbec42c00b5b5e914f4aff4/nvidia_cusparselt_cu13-0.8.1-py3-none-manylinux2014_aarch64.whl", hash = "sha256:4dca476c50bf4780d46cd0bfbd82e2bc10a08e4fef7950917ce8d7578d22a23f", size = 221051344, upload-time = "2025-09-05T18:49:51.289Z" },
+    { url = "https://files.pythonhosted.org/packages/34/7d/2661f2fb3ac4302f3a246f5fc030213ac60c1fe0bce84f9783dbd831dbb7/nvidia_cusparselt_cu13-0.8.1-py3-none-manylinux2014_x86_64.whl", hash = "sha256:786ce87568c303fadb5afcc7102d454cd3040d75f6f8626f5db460d1871f4dd0", size = 170148586, upload-time = "2025-09-05T18:50:50.248Z" },
+]
+
+[[package]]
+name = "nvidia-nccl-cu13"
+version = "2.29.7"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/72/0d/daf50d44177ee0cbc7ff0a0c91eb5ff676c82be42f9a970bc7597f440c3a/nvidia_nccl_cu13-2.29.7-py3-none-manylinux_2_18_aarch64.whl", hash = "sha256:674a12383e3c38a1bcccae7d4f3633b37852230b6047883cb2f4c2d1b36d9bf5", size = 206014712, upload-time = "2026-03-03T05:34:20.843Z" },
+    { url = "https://files.pythonhosted.org/packages/67/f4/58e4e91b6919367c7aafb8e36fce9aad1a3047e536bf7e2fd560927d3a4c/nvidia_nccl_cu13-2.29.7-py3-none-manylinux_2_18_x86_64.whl", hash = "sha256:edd81538446786ec3b73972543e53bb43bcaf0bfc8ef76cb679fcc390ffe136d", size = 205976000, upload-time = "2026-03-03T05:36:24.472Z" },
+]
+
+[[package]]
+name = "nvidia-nvjitlink"
+version = "13.0.88"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/56/7a/123e033aaff487c77107195fa5a2b8686795ca537935a24efae476c41f05/nvidia_nvjitlink-13.0.88-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:13a74f429e23b921c1109976abefacc69835f2f433ebd323d3946e11d804e47b", size = 40713933, upload-time = "2025-09-04T08:35:43.553Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/2c/93c5250e64df4f894f1cbb397c6fd71f79813f9fd79d7cd61de3f97b3c2d/nvidia_nvjitlink-13.0.88-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:e931536ccc7d467a98ba1d8b89ff7fa7f1fa3b13f2b0069118cd7f47bff07d0c", size = 38768748, upload-time = "2025-09-04T08:35:20.008Z" },
+]
+
+[[package]]
+name = "nvidia-nvshmem-cu13"
+version = "3.4.5"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/dc/0f/05cc9c720236dcd2db9c1ab97fff629e96821be2e63103569da0c9b72f19/nvidia_nvshmem_cu13-3.4.5-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:6dc2a197f38e5d0376ad52cd1a2a3617d3cdc150fd5966f4aee9bcebb1d68fe9", size = 60215947, upload-time = "2025-09-06T00:32:20.022Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/35/a9bf80a609e74e3b000fef598933235c908fcefcef9026042b8e6dfde2a9/nvidia_nvshmem_cu13-3.4.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:290f0a2ee94c9f3687a02502f3b9299a9f9fe826e6d0287ee18482e78d495b80", size = 60412546, upload-time = "2025-09-06T00:32:41.564Z" },
+]
+
+[[package]]
+name = "nvidia-nvtx"
+version = "13.0.85"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c2/f3/d86c845465a2723ad7e1e5c36dcd75ddb82898b3f53be47ebd429fb2fa5d/nvidia_nvtx-13.0.85-py3-none-manylinux1_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:4936d1d6780fbe68db454f5e72a42ff64d1fd6397df9f363ae786930fd5c1cd4", size = 148047, upload-time = "2025-09-04T08:29:01.761Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/64/3708a90d1ebe202ffdeb7185f878a3c84d15c2b2c31858da2ce0583e2def/nvidia_nvtx-13.0.85-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cb7780edb6b14107373c835bf8b72e7a178bac7367e23da7acb108f973f157a6", size = 148878, upload-time = "2025-09-04T08:28:53.627Z" },
+]
+
+[[package]]
+name = "olefile"
+version = "0.47"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/69/1b/077b508e3e500e1629d366249c3ccb32f95e50258b231705c09e3c7a4366/olefile-0.47.zip", hash = "sha256:599383381a0bf3dfbd932ca0ca6515acd174ed48870cbf7fee123d698c192c1c", size = 112240, upload-time = "2023-12-01T16:22:53.025Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/17/d3/b64c356a907242d719fc668b71befd73324e47ab46c8ebbbede252c154b2/olefile-0.47-py2.py3-none-any.whl", hash = "sha256:543c7da2a7adadf21214938bb79c83ea12b473a4b6ee4ad4bf854e7715e13d1f", size = 114565, upload-time = "2023-12-01T16:22:51.518Z" },
+]
+
+[[package]]
+name = "onnx"
+version = "1.21.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "ml-dtypes" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "protobuf" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/c5/93/942d2a0f6a70538eea042ce0445c8aefd46559ad153469986f29a743c01c/onnx-1.21.0.tar.gz", hash = "sha256:4d8b67d0aaec5864c87633188b91cc520877477ec0254eda122bef8be43cd764", size = 12074608, upload-time = "2026-03-27T21:33:36.118Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a8/28/a14b1845bf9302c3a787221e8f37cde4e7f930e10d95a8e22dd910aeb41d/onnx-1.21.0-cp310-cp310-macosx_12_0_universal2.whl", hash = "sha256:e0c21cc5c7a41d1a509828e2b14fe9c30e807c6df611ec0fd64a47b8d4b16abd", size = 17966899, upload-time = "2026-03-27T21:32:15.53Z" },
+    { url = "https://files.pythonhosted.org/packages/41/7b/788881bf022a4cfb7b0843782f88415ea51c805cee4a909dcf2e49bb8129/onnx-1.21.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e1931bfcc222a4c9da6475f2ffffb84b97ab3876041ec639171c11ce802bee6a", size = 17534297, upload-time = "2026-03-27T21:32:18.343Z" },
+    { url = "https://files.pythonhosted.org/packages/16/51/eb64d4f2ec6caa98909aab5fbcfa24be9c059081e804bbb0012cc549ef89/onnx-1.21.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c9b56ad04039fac6b028c07e54afa1ec7f75dd340f65311f2c292e41ed7aa4d9", size = 17616697, upload-time = "2026-03-27T21:32:21Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/4e/6b1f7800dae3407dc850e7e59d591ed8c83e9b3401e4cd57a1f612e400c6/onnx-1.21.0-cp310-cp310-win32.whl", hash = "sha256:3abd09872523c7e0362d767e4e63bd7c6bac52a5e2c3edbf061061fe540e2027", size = 16288893, upload-time = "2026-03-27T21:32:23.864Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/a8/89273e581d3943e20314af19b1596ab4d763f9c2eb07d4eaf4fb0593219b/onnx-1.21.0-cp310-cp310-win_amd64.whl", hash = "sha256:f2c7c234c568402e10db74e33d787e4144e394ae2bcbbf11000fbfe2e017ad68", size = 16443416, upload-time = "2026-03-27T21:32:26.655Z" },
+    { url = "https://files.pythonhosted.org/packages/45/48/32e383aa6bc40b72a9fd419937aaa647078190c9bfccdc97b316d2dee687/onnx-1.21.0-cp311-cp311-macosx_12_0_universal2.whl", hash = "sha256:2aca19949260875c14866fc77ea0bc37e4e809b24976108762843d328c92d3ce", size = 17968053, upload-time = "2026-03-27T21:32:29.558Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/26/5726e8df7d36e96bb3c679912d1a86af42f393d77aa17d6b98a97d4289ce/onnx-1.21.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:82aa6ab51144df07c58c4850cb78d4f1ae969d8c0bf657b28041796d49ba6974", size = 17534821, upload-time = "2026-03-27T21:32:32.351Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/2b/021dcd2dd50c3c71b7959d7368526da384a295c162fb4863f36057973f78/onnx-1.21.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:10c3185a232089335581fabb98fba4e86d3e8246b8140f2e406082438100ebda", size = 17616664, upload-time = "2026-03-27T21:32:34.921Z" },
+    { url = "https://files.pythonhosted.org/packages/12/00/afa32a46fa122a7ed42df1cfe8796922156a3725ba8fc581c4779c96e2fc/onnx-1.21.0-cp311-cp311-win32.whl", hash = "sha256:f53b3c15a3b539c16b99655c43c365622046d68c49b680c48eba4da2a4fb6f27", size = 16289035, upload-time = "2026-03-27T21:32:37.783Z" },
+    { url = "https://files.pythonhosted.org/packages/73/8d/483cc980a24d4c0131d0af06d0ff6a37fb08ae90a7848ece8cef645194f1/onnx-1.21.0-cp311-cp311-win_amd64.whl", hash = "sha256:5f78c411743db317a76e5d009f84f7e3d5380411a1567a868e82461a1e5c775d", size = 16443748, upload-time = "2026-03-27T21:32:40.337Z" },
+    { url = "https://files.pythonhosted.org/packages/38/78/9d06fd5aaaed1ec9cb8a3b70fbbf00c1bdc18db610771e96379f0ed58112/onnx-1.21.0-cp311-cp311-win_arm64.whl", hash = "sha256:ab6a488dabbb172eebc9f3b3e7ac68763f32b0c571626d4a5004608f866cc83d", size = 16406123, upload-time = "2026-03-27T21:32:45.159Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/ae/cb644ec84c25e63575d9d8790fdcc5d1a11d67d3f62f872edb35fa38d158/onnx-1.21.0-cp312-abi3-macosx_12_0_universal2.whl", hash = "sha256:fc2635400fe39ff37ebc4e75342cc54450eadadf39c540ff132c319bf4960095", size = 17965930, upload-time = "2026-03-27T21:32:48.089Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/b6/eeb5903586645ef8a49b4b7892580438741acc3df91d7a5bd0f3a59ea9cb/onnx-1.21.0-cp312-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9003d5206c01fa2ff4b46311566865d8e493e1a6998d4009ec6de39843f1b59b", size = 17531344, upload-time = "2026-03-27T21:32:50.837Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/00/4823f06357892d1e60d6f34e7299d2ba4ed2108c487cc394f7ce85a3ff14/onnx-1.21.0-cp312-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a9261bd580fb8548c9c37b3c6750387eb8f21ea43c63880d37b2c622e1684285", size = 17613697, upload-time = "2026-03-27T21:32:54.222Z" },
+    { url = "https://files.pythonhosted.org/packages/23/1d/391f3c567ae068c8ac4f1d1316bae97c9eb45e702f05975fe0e17ad441f0/onnx-1.21.0-cp312-abi3-win32.whl", hash = "sha256:9ea4e824964082811938a9250451d89c4ec474fe42dd36c038bfa5df31993d1e", size = 16287200, upload-time = "2026-03-27T21:32:57.277Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/a6/5eefbe5b40ea96de95a766bd2e0e751f35bdea2d4b951991ec9afaa69531/onnx-1.21.0-cp312-abi3-win_amd64.whl", hash = "sha256:458d91948ad9a7729a347550553b49ab6939f9af2cddf334e2116e45467dc61f", size = 16441045, upload-time = "2026-03-27T21:33:00.081Z" },
+    { url = "https://files.pythonhosted.org/packages/63/c4/0ed8dc037a39113d2a4d66e0005e07751c299c46b993f1ad5c2c35664c20/onnx-1.21.0-cp312-abi3-win_arm64.whl", hash = "sha256:ca14bc4842fccc3187eb538f07eabeb25a779b39388b006db4356c07403a7bbb", size = 16403134, upload-time = "2026-03-27T21:33:03.987Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/89/0e1a9beb536401e2f45ac88735e123f2735e12fc7b56ff6c11727e097526/onnx-1.21.0-cp313-cp313t-macosx_12_0_universal2.whl", hash = "sha256:257d1d1deb6a652913698f1e3f33ef1ca0aa69174892fe38946d4572d89dd94f", size = 17975430, upload-time = "2026-03-27T21:33:07.005Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/46/e6dc71a7b3b317265591b20a5f71d0ff5c0d26c24e52283139dc90c66038/onnx-1.21.0-cp313-cp313t-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:7cd7cb8f6459311bdb557cbf6c0ccc6d8ace11c304d1bba0a30b4a4688e245f8", size = 17537435, upload-time = "2026-03-27T21:33:09.765Z" },
+    { url = "https://files.pythonhosted.org/packages/49/2e/27affcac63eaf2ef183a44fd1a1354b11da64a6c72fe6f3fdcf5571bcee5/onnx-1.21.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7b58a4cfec8d9311b73dc083e4c1fa362069267881144c05139b3eba5dc3a840", size = 17617687, upload-time = "2026-03-27T21:33:12.619Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/5c/ac8ed15e941593a3672ce424280b764979026317811f2e8508432bfc3429/onnx-1.21.0-cp313-cp313t-win_amd64.whl", hash = "sha256:1a9baf882562c4cebf79589bebb7cd71a20e30b51158cac3e3bbaf27da6163bd", size = 16449402, upload-time = "2026-03-27T21:33:15.555Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/aa/d2231e0dcaad838217afc64c306c8152a080134d2034e247cc973d577674/onnx-1.21.0-cp313-cp313t-win_arm64.whl", hash = "sha256:bba12181566acf49b35875838eba49536a327b2944664b17125577d230c637ad", size = 16408273, upload-time = "2026-03-27T21:33:18.599Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/0a/8905b14694def6ad23edf1011fdd581500384062f8c4c567e114be7aa272/onnx-1.21.0-cp314-cp314t-macosx_12_0_universal2.whl", hash = "sha256:7ee9d8fd6a4874a5fa8b44bbcabea104ce752b20469b88bc50c7dcf9030779ad", size = 17975331, upload-time = "2026-03-27T21:33:21.69Z" },
+    { url = "https://files.pythonhosted.org/packages/61/28/f4e401e5199d1b9c8b76c7e7ae1169e050515258e877b58fa8bb49d3bdcc/onnx-1.21.0-cp314-cp314t-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5489f25fe461e7f32128218251a466cabbeeaf1eaa791c79daebf1a80d5a2cc9", size = 17537430, upload-time = "2026-03-27T21:33:24.547Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/cf/5d13320eb3660d5af360ea3b43aa9c63a70c92a9b4d1ea0d34501a32fcb8/onnx-1.21.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:db17fc0fec46180b6acbd1d5d8650a04e5527c02b09381da0b5b888d02a204c8", size = 17617662, upload-time = "2026-03-27T21:33:27.418Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/50/3eaa1878338247be021e6423696813d61e77e534dccbd15a703a144e703d/onnx-1.21.0-cp314-cp314t-win_amd64.whl", hash = "sha256:19d9971a3e52a12968ae6c70fd0f86c349536de0b0c33922ecdbe52d1972fe60", size = 16463688, upload-time = "2026-03-27T21:33:30.229Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/48/38d46b43bbb525e0b6a4c2c4204cc6795d67e45687a2f7403e06d8e7053d/onnx-1.21.0-cp314-cp314t-win_arm64.whl", hash = "sha256:efba467efb316baf2a9452d892c2f982b9b758c778d23e38c7f44fa211b30bb9", size = 16423387, upload-time = "2026-03-27T21:33:33.446Z" },
+]
+
+[[package]]
+name = "onnx-ir"
+version = "0.2.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "ml-dtypes" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "onnx" },
+    { name = "sympy" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/35/e6/672fefb2f108d077f58181a7babf4c0f8d1182a30353ffc9c79c63afc5ee/onnx_ir-0.2.1.tar.gz", hash = "sha256:8b8b10a93f43e65962104de6070c43c5dacb0e3cdfefc7c8059dd83c9db64f35", size = 144279, upload-time = "2026-04-20T20:21:47.735Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8c/aa/f7a53321c60b9ad9ee184b6018292ed6b5389947592a2c8c09c736bb7f9e/onnx_ir-0.2.1-py3-none-any.whl", hash = "sha256:c7285da889312f91882de2092e298a9eeeefbfc1d1951c49d983992967eb09a7", size = 166792, upload-time = "2026-04-20T20:21:46.357Z" },
+]
+
+[[package]]
+name = "onnxruntime"
+version = "1.24.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.11' and platform_machine != 's390x'",
+    "python_full_version < '3.11' and platform_machine == 's390x'",
+]
+dependencies = [
+    { name = "flatbuffers", marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "packaging", marker = "python_full_version < '3.11'" },
+    { name = "protobuf", marker = "python_full_version < '3.11'" },
+    { name = "sympy", marker = "python_full_version < '3.11'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/15/41/3253db975a90c3ce1d475e2a230773a21cd7998537f0657947df6fb79861/onnxruntime-1.24.3-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:3e6456801c66b095c5cd68e690ca25db970ea5202bd0c5b84a2c3ef7731c5a3c", size = 17332766, upload-time = "2026-03-05T17:18:59.714Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/c5/3af6b325f1492d691b23844d88ed26844c1164620860c5efe95c0e22782d/onnxruntime-1.24.3-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b2ebc54c6d8281dccff78d4b06e47d4cf07535937584ab759448390a70f4978", size = 15130330, upload-time = "2026-03-05T16:34:53.831Z" },
+    { url = "https://files.pythonhosted.org/packages/03/4b/f96b46c1866a293ed23ca2cf5e5a63d413ad3a951da60dd877e3c56cbbca/onnxruntime-1.24.3-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fb56575d7794bf0781156955610c9e651c9504c64d42ec880784b6106244882d", size = 17213247, upload-time = "2026-03-05T17:17:59.812Z" },
+    { url = "https://files.pythonhosted.org/packages/36/13/27cf4d8df2578747584e8758aeb0b673b60274048510257f1f084b15e80e/onnxruntime-1.24.3-cp311-cp311-win_amd64.whl", hash = "sha256:c958222ef9eff54018332beecd32d5d94a3ab079d8821937b333811bf4da0d39", size = 12595530, upload-time = "2026-03-05T17:18:49.356Z" },
+    { url = "https://files.pythonhosted.org/packages/19/8c/6d9f31e6bae72a8079be12ed8ba36c4126a571fad38ded0a1b96f60f6896/onnxruntime-1.24.3-cp311-cp311-win_arm64.whl", hash = "sha256:a8f761857ebaf58a85b9e42422d03207f1d39e6bb8fecfdbf613bac5b9710723", size = 12261715, upload-time = "2026-03-05T17:18:39.699Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/7f/dfdc4e52600fde4c02d59bfe98c4b057931c1114b701e175aee311a9bc11/onnxruntime-1.24.3-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:0d244227dc5e00a9ae15a7ac1eba4c4460d7876dfecafe73fb00db9f1d914d91", size = 17342578, upload-time = "2026-03-05T17:19:02.403Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/dc/1f5489f7b21817d4ad352bf7a92a252bd5b438bcbaa7ad20ea50814edc79/onnxruntime-1.24.3-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a9847b870b6cb462652b547bc98c49e0efb67553410a082fde1918a38707452", size = 15150105, upload-time = "2026-03-05T16:34:56.897Z" },
+    { url = "https://files.pythonhosted.org/packages/28/7c/fd253da53594ab8efbefdc85b3638620ab1a6aab6eb7028a513c853559ce/onnxruntime-1.24.3-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b354afce3333f2859c7e8706d84b6c552beac39233bcd3141ce7ab77b4cabb5d", size = 17237101, upload-time = "2026-03-05T17:18:02.561Z" },
+    { url = "https://files.pythonhosted.org/packages/71/5f/eaabc5699eeed6a9188c5c055ac1948ae50138697a0428d562ac970d7db5/onnxruntime-1.24.3-cp312-cp312-win_amd64.whl", hash = "sha256:44ea708c34965439170d811267c51281d3897ecfc4aa0087fa25d4a4c3eb2e4a", size = 12597638, upload-time = "2026-03-05T17:18:52.141Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/5c/d8066c320b90610dbeb489a483b132c3b3879b2f93f949fb5d30cfa9b119/onnxruntime-1.24.3-cp312-cp312-win_arm64.whl", hash = "sha256:48d1092b44ca2ba6f9543892e7c422c15a568481403c10440945685faf27a8d8", size = 12270943, upload-time = "2026-03-05T17:18:42.006Z" },
+    { url = "https://files.pythonhosted.org/packages/51/8d/487ece554119e2991242d4de55de7019ac6e47ee8dfafa69fcf41d37f8ed/onnxruntime-1.24.3-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:34a0ea5ff191d8420d9c1332355644148b1bf1a0d10c411af890a63a9f662aa7", size = 17342706, upload-time = "2026-03-05T16:35:10.813Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/25/8b444f463c1ac6106b889f6235c84f01eec001eaf689c3eff8c69cf48fae/onnxruntime-1.24.3-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1fd2ec7bb0fabe42f55e8337cfc9b1969d0d14622711aac73d69b4bd5abb5ed7", size = 15149956, upload-time = "2026-03-05T16:34:59.264Z" },
+    { url = "https://files.pythonhosted.org/packages/34/fc/c9182a3e1ab46940dd4f30e61071f59eee8804c1f641f37ce6e173633fb6/onnxruntime-1.24.3-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:df8e70e732fe26346faaeec9147fa38bef35d232d2495d27e93dd221a2d473a9", size = 17237370, upload-time = "2026-03-05T17:18:05.258Z" },
+    { url = "https://files.pythonhosted.org/packages/05/7e/3b549e1f4538514118bff98a1bcd6481dd9a17067f8c9af77151621c9a5c/onnxruntime-1.24.3-cp313-cp313-win_amd64.whl", hash = "sha256:2d3706719be6ad41d38a2250998b1d87758a20f6ea4546962e21dc79f1f1fd2b", size = 12597939, upload-time = "2026-03-05T17:18:54.772Z" },
+    { url = "https://files.pythonhosted.org/packages/80/41/9696a5c4631a0caa75cc8bc4efd30938fd483694aa614898d087c3ee6d29/onnxruntime-1.24.3-cp313-cp313-win_arm64.whl", hash = "sha256:b082f3ba9519f0a1a1e754556bc7e635c7526ef81b98b3f78da4455d25f0437b", size = 12270705, upload-time = "2026-03-05T17:18:44.774Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/65/a26c5e59e3b210852ee04248cf8843c81fe7d40d94cf95343b66efe7eec9/onnxruntime-1.24.3-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:72f956634bc2e4bd2e8b006bef111849bd42c42dea37bd0a4c728404fdaf4d34", size = 15161796, upload-time = "2026-03-05T16:35:02.871Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/25/2035b4aa2ccb5be6acf139397731ec507c5f09e199ab39d3262b22ffa1ac/onnxruntime-1.24.3-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:78d1f25eed4ab9959db70a626ed50ee24cf497e60774f59f1207ac8556399c4d", size = 17240936, upload-time = "2026-03-05T17:18:09.534Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/a4/b3240ea84b92a3efb83d49cc16c04a17ade1ab47a6a95c4866d15bf0ac35/onnxruntime-1.24.3-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:a6b4bce87d96f78f0a9bf5cefab3303ae95d558c5bfea53d0bf7f9ea207880a8", size = 17344149, upload-time = "2026-03-05T16:35:13.382Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/4a/4b56757e51a56265e8c56764d9c36d7b435045e05e3b8a38bedfc5aedba3/onnxruntime-1.24.3-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d48f36c87b25ab3b2b4c88826c96cf1399a5631e3c2c03cc27d6a1e5d6b18eb4", size = 15151571, upload-time = "2026-03-05T16:35:05.679Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/14/c6fb84980cec8f682a523fcac7c2bdd6b311e7f342c61ce48d3a9cb87fc6/onnxruntime-1.24.3-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e104d33a409bf6e3f30f0e8198ec2aaf8d445b8395490a80f6e6ad56da98e400", size = 17238951, upload-time = "2026-03-05T17:18:12.394Z" },
+    { url = "https://files.pythonhosted.org/packages/57/14/447e1400165aca8caf35dabd46540eb943c92f3065927bb4d9bcbc91e221/onnxruntime-1.24.3-cp314-cp314-win_amd64.whl", hash = "sha256:e785d73fbd17421c2513b0bb09eb25d88fa22c8c10c3f5d6060589efa5537c5b", size = 12903820, upload-time = "2026-03-05T17:18:57.123Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/ec/6b2fa5702e4bbba7339ca5787a9d056fc564a16079f8833cc6ba4798da1c/onnxruntime-1.24.3-cp314-cp314-win_arm64.whl", hash = "sha256:951e897a275f897a05ffbcaa615d98777882decaeb80c9216c68cdc62f849f53", size = 12594089, upload-time = "2026-03-05T17:18:47.169Z" },
+    { url = "https://files.pythonhosted.org/packages/12/dc/cd06cba3ddad92ceb17b914a8e8d49836c79e38936e26bde6e368b62c1fe/onnxruntime-1.24.3-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4d4e70ce578aa214c74c7a7a9226bc8e229814db4a5b2d097333b81279ecde36", size = 15162789, upload-time = "2026-03-05T16:35:08.282Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/d6/413e98ab666c6fb9e8be7d1c6eb3bd403b0bea1b8d42db066dab98c7df07/onnxruntime-1.24.3-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:02aaf6ddfa784523b6873b4176a79d508e599efe12ab0ea1a3a6e7314408b7aa", size = 17240738, upload-time = "2026-03-05T17:18:15.203Z" },
+]
+
+[[package]]
+name = "onnxruntime"
+version = "1.26.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+]
+dependencies = [
+    { name = "flatbuffers", marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "packaging", marker = "python_full_version >= '3.11'" },
+    { name = "protobuf", marker = "python_full_version >= '3.11'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d4/81/29a9eb470994a75eb7b3ccf32be314d7c66675a00ac7b50294816cc2db27/onnxruntime-1.26.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:ee1109ef4ef27cad90e823399e61e03b3c6c7bfe0fb820b4baf3678c15be8b3c", size = 18005108, upload-time = "2026-05-08T19:08:11.728Z" },
+    { url = "https://files.pythonhosted.org/packages/66/c7/73efa6c8a4000c38fcc14947d84f234a17e5d66f203b37b7f1ad4a7b46eb/onnxruntime-1.26.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:35c7c7b0ac2e02001d28fab6c9fc24e9abc5e6faa35e6e19c63cecf1406ba89f", size = 16043752, upload-time = "2026-05-08T19:07:10.707Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/3f/8de630f595daf6ce884d4dd95afd2a60e70ec6572e52bfee3aa2229befab/onnxruntime-1.26.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:11a8df4dcfe9ad5ff0bd71a7571dbed019fabc7594676c89fe8b86ea029c246f", size = 18176043, upload-time = "2026-05-08T19:07:33.735Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/21/9f041de20787cd85498bd48e0ec4d098bf2a6c486e25b24b8dae1bf492b2/onnxruntime-1.26.0-cp311-cp311-win_amd64.whl", hash = "sha256:e6456718125fd777c673f3b78d4a9ab58d6adea641e9afae85ee6444f0e0e9a9", size = 13023165, upload-time = "2026-05-08T19:08:00.633Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/82/3b9fe0ead2557cc3adf74c74c141bd1c7c4c6a9548c610af37df199f4512/onnxruntime-1.26.0-cp311-cp311-win_arm64.whl", hash = "sha256:cd920e45b730e4a87833e2910d8ca375aaca9da6ccc09e24bce463b3356d637f", size = 12789514, upload-time = "2026-05-08T19:07:49.433Z" },
+    { url = "https://files.pythonhosted.org/packages/81/b1/d111b1df656761f980d9e298a60039a9cb66036b1d039e777537743d0ac3/onnxruntime-1.26.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:05b028781b322ad74b57ce5b50aa5280bb1fe96ceec334628ade681e0b24c1ac", size = 18016624, upload-time = "2026-05-12T00:41:01.735Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/a0/3f9d896a0385a36bd04345d6d0b802821a5782adde562e7e135f6bb71c73/onnxruntime-1.26.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:91f2bb870a4b9224eba0a6728c1fa7a9e552b8e59e1083c51fbbc3d013f2b5c0", size = 16052692, upload-time = "2026-05-08T19:07:13.829Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/43/2a4e04f8dbeffad19bbcced4bcd4289bf478921518437404d6b92bdf213b/onnxruntime-1.26.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9b6dd70599005bd1bf29779f04a91978b92b5e719c11a20068a8f8e535f725b6", size = 18185439, upload-time = "2026-05-08T19:07:36.299Z" },
+    { url = "https://files.pythonhosted.org/packages/44/fc/026d0a7162b9c2153dac292baea9e027c42304dc1d9dc6f8ff5b4cfbaedd/onnxruntime-1.26.0-cp312-cp312-win_amd64.whl", hash = "sha256:a26374dc7fbcaae593601086b242120e13f2310558df0991da6dd8b8fac00414", size = 13026427, upload-time = "2026-05-08T19:08:03.503Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/27/1dcf88e45e4c69db5f7b106f2dacc3801ba98994e082ca03e1dfdf7bfe57/onnxruntime-1.26.0-cp312-cp312-win_arm64.whl", hash = "sha256:54a8053410fd31fd66469bd754fcfe8a4df9f7eb44756b4b5479bf50c842d948", size = 12796647, upload-time = "2026-05-08T19:07:52.108Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/a2/c801242685e0ce48a4ca51dfafbb588765e0446397e123be53ba5598f3f5/onnxruntime-1.26.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:ccce19c5f771b8268902f77d9fed9e88f9499465d6780808faa6611a789d33f0", size = 18016563, upload-time = "2026-05-08T19:07:28.081Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/64/0492c0b1db04e29b2630c87cfa36f9d6872b1ca8614b90c5cad58fac7d76/onnxruntime-1.26.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bdbed8cf3b672b66acb032f33a253bc27f42bce6ece48ae3fab4fa483a5e96e0", size = 16052634, upload-time = "2026-05-08T19:07:16.885Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/26/4d09ddc755a84fc8d5e192991626b0e0680e8f6c5d58f4f1d05c42bc48cf/onnxruntime-1.26.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c07af6fc6d5557835f2b6ee7a96d8b3235d0c57a8e230efdedaee106a8a3cbc6", size = 18185632, upload-time = "2026-05-08T19:07:38.756Z" },
+    { url = "https://files.pythonhosted.org/packages/77/89/3e52249aa08fa301e217ecba07b5246a8338fa2b401e109326e3fc5be0f9/onnxruntime-1.26.0-cp313-cp313-win_amd64.whl", hash = "sha256:61bec80655efa460591c2bc655392d57d2650ce85533a6b9b3b7a790d7ea7916", size = 13026751, upload-time = "2026-05-08T19:08:06.2Z" },
+    { url = "https://files.pythonhosted.org/packages/06/b3/c1c8782b14af6797c303de132d6eef26a9fb80dfacd3750ce57911d11c6b/onnxruntime-1.26.0-cp313-cp313-win_arm64.whl", hash = "sha256:a6677545ff451e3539a02746d2f207d8c5baa4a0a818886bb9d6a6eb9511ee89", size = 12796807, upload-time = "2026-05-08T19:07:54.879Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/f5/47b0676408abec652c14b84d7173e389837832d850c24f87184277313e8d/onnxruntime-1.26.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5e016edc15d3c19f36807e1c6b10be5b27807688c32720f91b5ae480a95215d0", size = 16057265, upload-time = "2026-05-08T19:07:19.603Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/45/33ab6deeef010ca844c877dd618cebc079590bbe52d2a3678e7223b1b908/onnxruntime-1.26.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f5fc48a91a046a6a5c9b147f83fb41d65d24d24923373b222cdd248f0f4f4aac", size = 18197590, upload-time = "2026-05-08T19:07:41.422Z" },
+    { url = "https://files.pythonhosted.org/packages/40/89/17546c1c20f6bfc3ae41c22152378a26edfea918af3129e2139dcd7c99f3/onnxruntime-1.26.0-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:33a791f31432a3af1a96db5e54818b37aba5e5eefc2e6af5794c10a9118a9993", size = 18019724, upload-time = "2026-05-08T19:07:30.723Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/24/89457a35f6af29538a76647f2c18c3a28277e6c19234c847e7b4b7c19860/onnxruntime-1.26.0-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e90c00732c4553618103149d93f688e8c3063017938f8983e21a71d9f3b6d22e", size = 16054821, upload-time = "2026-05-08T19:07:22.348Z" },
+    { url = "https://files.pythonhosted.org/packages/12/f9/15b2e1815cf570d238e0135529f80d2dce64e8e8818a1489cae83823c5c6/onnxruntime-1.26.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:01498e80ba8988428d08c2d51b1338f89e3de2a93e6ffe555f79c68f26a5c06b", size = 18185815, upload-time = "2026-05-08T19:07:44.179Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/65/2e11055faf015e4b07f45b513fa49b391baf2e19d92d77d73ebee13c1004/onnxruntime-1.26.0-cp314-cp314-win_amd64.whl", hash = "sha256:7ead61450d8405167c87dd3a31d8da1d576b490a57dab1aa8b82a7da6825f5aa", size = 13349887, upload-time = "2026-05-08T19:08:08.671Z" },
+    { url = "https://files.pythonhosted.org/packages/19/e4/0f9d1a5718b1781c610c1e354765a3820597081754277a6a9a2b50705702/onnxruntime-1.26.0-cp314-cp314-win_arm64.whl", hash = "sha256:31d71a53490e46910877d0902b5ad99c69a5955e5c7ea6c82863519410e1ba7c", size = 13140121, upload-time = "2026-05-08T19:07:57.804Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/42/3b8e635f067d06d9f45bede470b8d539d101a4166c272213158dfd08b6ce/onnxruntime-1.26.0-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d7b6d258fb78fdfcf049795bcfaa74dcb90ae7baa277afd21e6fd28b83f2c496", size = 16057240, upload-time = "2026-05-08T19:07:25.163Z" },
+    { url = "https://files.pythonhosted.org/packages/93/99/f2be40a31b908d96b861ae0ce98582fa376c18a7f816b9d5eb4cd6aa0a4c/onnxruntime-1.26.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4eefd386a45202aefb7a5132b94f32df9d506c9edcc7faf2fc60d65183f4b183", size = 18197382, upload-time = "2026-05-08T19:07:46.965Z" },
+]
+
+[[package]]
+name = "onnxscript"
+version = "0.7.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "ml-dtypes" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "onnx" },
+    { name = "onnx-ir" },
+    { name = "packaging" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9b/99/fd948eba63ba65b52265a4cd09a14f96bb9f5b730fcef58876c4358bf406/onnxscript-0.7.0.tar.gz", hash = "sha256:c95ed7b339b02cface56ee27689565c46612e1fc542c562298dddfdad5268dc5", size = 612032, upload-time = "2026-04-20T17:09:19.775Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/ce/2ed92575cc3be4ea1db5f38f16f20765f9b20b69b14d6c1d9972658a8ee9/onnxscript-0.7.0-py3-none-any.whl", hash = "sha256:5b356907d4501e9919f8599c91d8da967406a37b1fac2b40caa55a49acf242ea", size = 714842, upload-time = "2026-04-20T17:09:22.089Z" },
+]
+
+[[package]]
+name = "openpyxl"
+version = "3.1.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "et-xmlfile" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/3d/f9/88d94a75de065ea32619465d2f77b29a0469500e99012523b91cc4141cd1/openpyxl-3.1.5.tar.gz", hash = "sha256:cf0e3cf56142039133628b5acffe8ef0c12bc902d2aadd3e0fe5878dc08d1050", size = 186464, upload-time = "2024-06-28T14:03:44.161Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c0/da/977ded879c29cbd04de313843e76868e6e13408a94ed6b987245dc7c8506/openpyxl-3.1.5-py2.py3-none-any.whl", hash = "sha256:5282c12b107bffeef825f4617dc029afaf41d0ea60823bbb665ef3079dc79de2", size = 250910, upload-time = "2024-06-28T14:03:41.161Z" },
+]
+
+[[package]]
+name = "optuna"
+version = "4.9.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "alembic" },
+    { name = "colorlog" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "packaging" },
+    { name = "pyyaml" },
+    { name = "sqlalchemy" },
+    { name = "tqdm" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f4/aa/05f5e3f662cc96a4c478fc3446b8ed6359825a2b504ecb614a9ac84e4a4d/optuna-4.9.0.tar.gz", hash = "sha256:b322e5cbdf1655fb84c37646c4a7a1f391de1b47806bbe222e015825d0a82b87", size = 485834, upload-time = "2026-06-01T06:23:30.424Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ab/f3/e5fcd5d9b15771ed6dc10e3a7eeddc672e418f4f4c4653d216cc1d857e2d/optuna-4.9.0-py3-none-any.whl", hash = "sha256:f52f3be6148654850c92a5860d398fd88ec6b2c84ab68d9c3d07dcff02e7afee", size = 425553, upload-time = "2026-06-01T06:23:28.804Z" },
+]
+
+[[package]]
+name = "packaging"
+version = "26.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d7/f1/e7a6dd94a8d4a5626c03e4e99c87f241ba9e350cd9e6d75123f992427270/packaging-26.2.tar.gz", hash = "sha256:ff452ff5a3e828ce110190feff1178bb1f2ea2281fa2075aadb987c2fb221661", size = 228134, upload-time = "2026-04-24T20:15:23.917Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/df/b2/87e62e8c3e2f4b32e5fe99e0b86d576da1312593b39f47d8ceef365e95ed/packaging-26.2-py3-none-any.whl", hash = "sha256:5fc45236b9446107ff2415ce77c807cee2862cb6fac22b8a73826d0693b0980e", size = 100195, upload-time = "2026-04-24T20:15:22.081Z" },
+]
+
+[[package]]
+name = "pandas"
+version = "2.3.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.11' and platform_machine != 's390x'",
+    "python_full_version < '3.11' and platform_machine == 's390x'",
+]
+dependencies = [
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "python-dateutil", marker = "python_full_version < '3.11'" },
+    { name = "pytz", marker = "python_full_version < '3.11'" },
+    { name = "tzdata", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/33/01/d40b85317f86cf08d853a4f495195c73815fdf205eef3993821720274518/pandas-2.3.3.tar.gz", hash = "sha256:e05e1af93b977f7eafa636d043f9f94c7ee3ac81af99c13508215942e64c993b", size = 4495223, upload-time = "2025-09-29T23:34:51.853Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3d/f7/f425a00df4fcc22b292c6895c6831c0c8ae1d9fac1e024d16f98a9ce8749/pandas-2.3.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:376c6446ae31770764215a6c937f72d917f214b43560603cd60da6408f183b6c", size = 11555763, upload-time = "2025-09-29T23:16:53.287Z" },
+    { url = "https://files.pythonhosted.org/packages/13/4f/66d99628ff8ce7857aca52fed8f0066ce209f96be2fede6cef9f84e8d04f/pandas-2.3.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e19d192383eab2f4ceb30b412b22ea30690c9e618f78870357ae1d682912015a", size = 10801217, upload-time = "2025-09-29T23:17:04.522Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/03/3fc4a529a7710f890a239cc496fc6d50ad4a0995657dccc1d64695adb9f4/pandas-2.3.3-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5caf26f64126b6c7aec964f74266f435afef1c1b13da3b0636c7518a1fa3e2b1", size = 12148791, upload-time = "2025-09-29T23:17:18.444Z" },
+    { url = "https://files.pythonhosted.org/packages/40/a8/4dac1f8f8235e5d25b9955d02ff6f29396191d4e665d71122c3722ca83c5/pandas-2.3.3-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:dd7478f1463441ae4ca7308a70e90b33470fa593429f9d4c578dd00d1fa78838", size = 12769373, upload-time = "2025-09-29T23:17:35.846Z" },
+    { url = "https://files.pythonhosted.org/packages/df/91/82cc5169b6b25440a7fc0ef3a694582418d875c8e3ebf796a6d6470aa578/pandas-2.3.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4793891684806ae50d1288c9bae9330293ab4e083ccd1c5e383c34549c6e4250", size = 13200444, upload-time = "2025-09-29T23:17:49.341Z" },
+    { url = "https://files.pythonhosted.org/packages/10/ae/89b3283800ab58f7af2952704078555fa60c807fff764395bb57ea0b0dbd/pandas-2.3.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:28083c648d9a99a5dd035ec125d42439c6c1c525098c58af0fc38dd1a7a1b3d4", size = 13858459, upload-time = "2025-09-29T23:18:03.722Z" },
+    { url = "https://files.pythonhosted.org/packages/85/72/530900610650f54a35a19476eca5104f38555afccda1aa11a92ee14cb21d/pandas-2.3.3-cp310-cp310-win_amd64.whl", hash = "sha256:503cf027cf9940d2ceaa1a93cfb5f8c8c7e6e90720a2850378f0b3f3b1e06826", size = 11346086, upload-time = "2025-09-29T23:18:18.505Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/fa/7ac648108144a095b4fb6aa3de1954689f7af60a14cf25583f4960ecb878/pandas-2.3.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:602b8615ebcc4a0c1751e71840428ddebeb142ec02c786e8ad6b1ce3c8dec523", size = 11578790, upload-time = "2025-09-29T23:18:30.065Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/35/74442388c6cf008882d4d4bdfc4109be87e9b8b7ccd097ad1e7f006e2e95/pandas-2.3.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:8fe25fc7b623b0ef6b5009149627e34d2a4657e880948ec3c840e9402e5c1b45", size = 10833831, upload-time = "2025-09-29T23:38:56.071Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/e4/de154cbfeee13383ad58d23017da99390b91d73f8c11856f2095e813201b/pandas-2.3.3-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b468d3dad6ff947df92dcb32ede5b7bd41a9b3cceef0a30ed925f6d01fb8fa66", size = 12199267, upload-time = "2025-09-29T23:18:41.627Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/c9/63f8d545568d9ab91476b1818b4741f521646cbdd151c6efebf40d6de6f7/pandas-2.3.3-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b98560e98cb334799c0b07ca7967ac361a47326e9b4e5a7dfb5ab2b1c9d35a1b", size = 12789281, upload-time = "2025-09-29T23:18:56.834Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/00/a5ac8c7a0e67fd1a6059e40aa08fa1c52cc00709077d2300e210c3ce0322/pandas-2.3.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:1d37b5848ba49824e5c30bedb9c830ab9b7751fd049bc7914533e01c65f79791", size = 13240453, upload-time = "2025-09-29T23:19:09.247Z" },
+    { url = "https://files.pythonhosted.org/packages/27/4d/5c23a5bc7bd209231618dd9e606ce076272c9bc4f12023a70e03a86b4067/pandas-2.3.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:db4301b2d1f926ae677a751eb2bd0e8c5f5319c9cb3f88b0becbbb0b07b34151", size = 13890361, upload-time = "2025-09-29T23:19:25.342Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/59/712db1d7040520de7a4965df15b774348980e6df45c129b8c64d0dbe74ef/pandas-2.3.3-cp311-cp311-win_amd64.whl", hash = "sha256:f086f6fe114e19d92014a1966f43a3e62285109afe874f067f5abbdcbb10e59c", size = 11348702, upload-time = "2025-09-29T23:19:38.296Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/fb/231d89e8637c808b997d172b18e9d4a4bc7bf31296196c260526055d1ea0/pandas-2.3.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:6d21f6d74eb1725c2efaa71a2bfc661a0689579b58e9c0ca58a739ff0b002b53", size = 11597846, upload-time = "2025-09-29T23:19:48.856Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/bd/bf8064d9cfa214294356c2d6702b716d3cf3bb24be59287a6a21e24cae6b/pandas-2.3.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3fd2f887589c7aa868e02632612ba39acb0b8948faf5cc58f0850e165bd46f35", size = 10729618, upload-time = "2025-09-29T23:39:08.659Z" },
+    { url = "https://files.pythonhosted.org/packages/57/56/cf2dbe1a3f5271370669475ead12ce77c61726ffd19a35546e31aa8edf4e/pandas-2.3.3-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ecaf1e12bdc03c86ad4a7ea848d66c685cb6851d807a26aa245ca3d2017a1908", size = 11737212, upload-time = "2025-09-29T23:19:59.765Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/63/cd7d615331b328e287d8233ba9fdf191a9c2d11b6af0c7a59cfcec23de68/pandas-2.3.3-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b3d11d2fda7eb164ef27ffc14b4fcab16a80e1ce67e9f57e19ec0afaf715ba89", size = 12362693, upload-time = "2025-09-29T23:20:14.098Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/de/8b1895b107277d52f2b42d3a6806e69cfef0d5cf1d0ba343470b9d8e0a04/pandas-2.3.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a68e15f780eddf2b07d242e17a04aa187a7ee12b40b930bfdd78070556550e98", size = 12771002, upload-time = "2025-09-29T23:20:26.76Z" },
+    { url = "https://files.pythonhosted.org/packages/87/21/84072af3187a677c5893b170ba2c8fbe450a6ff911234916da889b698220/pandas-2.3.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:371a4ab48e950033bcf52b6527eccb564f52dc826c02afd9a1bc0ab731bba084", size = 13450971, upload-time = "2025-09-29T23:20:41.344Z" },
+    { url = "https://files.pythonhosted.org/packages/86/41/585a168330ff063014880a80d744219dbf1dd7a1c706e75ab3425a987384/pandas-2.3.3-cp312-cp312-win_amd64.whl", hash = "sha256:a16dcec078a01eeef8ee61bf64074b4e524a2a3f4b3be9326420cabe59c4778b", size = 10992722, upload-time = "2025-09-29T23:20:54.139Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/4b/18b035ee18f97c1040d94debd8f2e737000ad70ccc8f5513f4eefad75f4b/pandas-2.3.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:56851a737e3470de7fa88e6131f41281ed440d29a9268dcbf0002da5ac366713", size = 11544671, upload-time = "2025-09-29T23:21:05.024Z" },
+    { url = "https://files.pythonhosted.org/packages/31/94/72fac03573102779920099bcac1c3b05975c2cb5f01eac609faf34bed1ca/pandas-2.3.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:bdcd9d1167f4885211e401b3036c0c8d9e274eee67ea8d0758a256d60704cfe8", size = 10680807, upload-time = "2025-09-29T23:21:15.979Z" },
+    { url = "https://files.pythonhosted.org/packages/16/87/9472cf4a487d848476865321de18cc8c920b8cab98453ab79dbbc98db63a/pandas-2.3.3-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e32e7cc9af0f1cc15548288a51a3b681cc2a219faa838e995f7dc53dbab1062d", size = 11709872, upload-time = "2025-09-29T23:21:27.165Z" },
+    { url = "https://files.pythonhosted.org/packages/15/07/284f757f63f8a8d69ed4472bfd85122bd086e637bf4ed09de572d575a693/pandas-2.3.3-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:318d77e0e42a628c04dc56bcef4b40de67918f7041c2b061af1da41dcff670ac", size = 12306371, upload-time = "2025-09-29T23:21:40.532Z" },
+    { url = "https://files.pythonhosted.org/packages/33/81/a3afc88fca4aa925804a27d2676d22dcd2031c2ebe08aabd0ae55b9ff282/pandas-2.3.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4e0a175408804d566144e170d0476b15d78458795bb18f1304fb94160cabf40c", size = 12765333, upload-time = "2025-09-29T23:21:55.77Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/0f/b4d4ae743a83742f1153464cf1a8ecfafc3ac59722a0b5c8602310cb7158/pandas-2.3.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:93c2d9ab0fc11822b5eece72ec9587e172f63cff87c00b062f6e37448ced4493", size = 13418120, upload-time = "2025-09-29T23:22:10.109Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/c7/e54682c96a895d0c808453269e0b5928a07a127a15704fedb643e9b0a4c8/pandas-2.3.3-cp313-cp313-win_amd64.whl", hash = "sha256:f8bfc0e12dc78f777f323f55c58649591b2cd0c43534e8355c51d3fede5f4dee", size = 10993991, upload-time = "2025-09-29T23:25:04.889Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/ca/3f8d4f49740799189e1395812f3bf23b5e8fc7c190827d55a610da72ce55/pandas-2.3.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:75ea25f9529fdec2d2e93a42c523962261e567d250b0013b16210e1d40d7c2e5", size = 12048227, upload-time = "2025-09-29T23:22:24.343Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/5a/f43efec3e8c0cc92c4663ccad372dbdff72b60bdb56b2749f04aa1d07d7e/pandas-2.3.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:74ecdf1d301e812db96a465a525952f4dde225fdb6d8e5a521d47e1f42041e21", size = 11411056, upload-time = "2025-09-29T23:22:37.762Z" },
+    { url = "https://files.pythonhosted.org/packages/46/b1/85331edfc591208c9d1a63a06baa67b21d332e63b7a591a5ba42a10bb507/pandas-2.3.3-cp313-cp313t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6435cb949cb34ec11cc9860246ccb2fdc9ecd742c12d3304989017d53f039a78", size = 11645189, upload-time = "2025-09-29T23:22:51.688Z" },
+    { url = "https://files.pythonhosted.org/packages/44/23/78d645adc35d94d1ac4f2a3c4112ab6f5b8999f4898b8cdf01252f8df4a9/pandas-2.3.3-cp313-cp313t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:900f47d8f20860de523a1ac881c4c36d65efcb2eb850e6948140fa781736e110", size = 12121912, upload-time = "2025-09-29T23:23:05.042Z" },
+    { url = "https://files.pythonhosted.org/packages/53/da/d10013df5e6aaef6b425aa0c32e1fc1f3e431e4bcabd420517dceadce354/pandas-2.3.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:a45c765238e2ed7d7c608fc5bc4a6f88b642f2f01e70c0c23d2224dd21829d86", size = 12712160, upload-time = "2025-09-29T23:23:28.57Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/17/e756653095a083d8a37cbd816cb87148debcfcd920129b25f99dd8d04271/pandas-2.3.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:c4fc4c21971a1a9f4bdb4c73978c7f7256caa3e62b323f70d6cb80db583350bc", size = 13199233, upload-time = "2025-09-29T23:24:24.876Z" },
+    { url = "https://files.pythonhosted.org/packages/04/fd/74903979833db8390b73b3a8a7d30d146d710bd32703724dd9083950386f/pandas-2.3.3-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:ee15f284898e7b246df8087fc82b87b01686f98ee67d85a17b7ab44143a3a9a0", size = 11540635, upload-time = "2025-09-29T23:25:52.486Z" },
+    { url = "https://files.pythonhosted.org/packages/21/00/266d6b357ad5e6d3ad55093a7e8efc7dd245f5a842b584db9f30b0f0a287/pandas-2.3.3-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:1611aedd912e1ff81ff41c745822980c49ce4a7907537be8692c8dbc31924593", size = 10759079, upload-time = "2025-09-29T23:26:33.204Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/05/d01ef80a7a3a12b2f8bbf16daba1e17c98a2f039cbc8e2f77a2c5a63d382/pandas-2.3.3-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6d2cefc361461662ac48810cb14365a365ce864afe85ef1f447ff5a1e99ea81c", size = 11814049, upload-time = "2025-09-29T23:27:15.384Z" },
+    { url = "https://files.pythonhosted.org/packages/15/b2/0e62f78c0c5ba7e3d2c5945a82456f4fac76c480940f805e0b97fcbc2f65/pandas-2.3.3-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ee67acbbf05014ea6c763beb097e03cd629961c8a632075eeb34247120abcb4b", size = 12332638, upload-time = "2025-09-29T23:27:51.625Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/33/dd70400631b62b9b29c3c93d2feee1d0964dc2bae2e5ad7a6c73a7f25325/pandas-2.3.3-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:c46467899aaa4da076d5abc11084634e2d197e9460643dd455ac3db5856b24d6", size = 12886834, upload-time = "2025-09-29T23:28:21.289Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/18/b5d48f55821228d0d2692b34fd5034bb185e854bdb592e9c640f6290e012/pandas-2.3.3-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:6253c72c6a1d990a410bc7de641d34053364ef8bcd3126f7e7450125887dffe3", size = 13409925, upload-time = "2025-09-29T23:28:58.261Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/3d/124ac75fcd0ecc09b8fdccb0246ef65e35b012030defb0e0eba2cbbbe948/pandas-2.3.3-cp314-cp314-win_amd64.whl", hash = "sha256:1b07204a219b3b7350abaae088f451860223a52cfb8a6c53358e7948735158e5", size = 11109071, upload-time = "2025-09-29T23:32:27.484Z" },
+    { url = "https://files.pythonhosted.org/packages/89/9c/0e21c895c38a157e0faa1fb64587a9226d6dd46452cac4532d80c3c4a244/pandas-2.3.3-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:2462b1a365b6109d275250baaae7b760fd25c726aaca0054649286bcfbb3e8ec", size = 12048504, upload-time = "2025-09-29T23:29:31.47Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/82/b69a1c95df796858777b68fbe6a81d37443a33319761d7c652ce77797475/pandas-2.3.3-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:0242fe9a49aa8b4d78a4fa03acb397a58833ef6199e9aa40a95f027bb3a1b6e7", size = 11410702, upload-time = "2025-09-29T23:29:54.591Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/88/702bde3ba0a94b8c73a0181e05144b10f13f29ebfc2150c3a79062a8195d/pandas-2.3.3-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a21d830e78df0a515db2b3d2f5570610f5e6bd2e27749770e8bb7b524b89b450", size = 11634535, upload-time = "2025-09-29T23:30:21.003Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/1e/1bac1a839d12e6a82ec6cb40cda2edde64a2013a66963293696bbf31fbbb/pandas-2.3.3-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2e3ebdb170b5ef78f19bfb71b0dc5dc58775032361fa188e814959b74d726dd5", size = 12121582, upload-time = "2025-09-29T23:30:43.391Z" },
+    { url = "https://files.pythonhosted.org/packages/44/91/483de934193e12a3b1d6ae7c8645d083ff88dec75f46e827562f1e4b4da6/pandas-2.3.3-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:d051c0e065b94b7a3cea50eb1ec32e912cd96dba41647eb24104b6c6c14c5788", size = 12699963, upload-time = "2025-09-29T23:31:10.009Z" },
+    { url = "https://files.pythonhosted.org/packages/70/44/5191d2e4026f86a2a109053e194d3ba7a31a2d10a9c2348368c63ed4e85a/pandas-2.3.3-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:3869faf4bd07b3b66a9f462417d0ca3a9df29a9f6abd5d0d0dbab15dac7abe87", size = 13202175, upload-time = "2025-09-29T23:31:59.173Z" },
+]
+
+[[package]]
+name = "pandas"
+version = "3.0.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine != 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version >= '3.14' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'win32'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform == 'emscripten'",
+    "python_full_version == '3.13.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.12.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+    "python_full_version == '3.11.*' and platform_machine == 's390x' and sys_platform != 'emscripten' and sys_platform != 'win32'",
+]
+dependencies = [
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "python-dateutil", marker = "python_full_version >= '3.11'" },
+    { name = "tzdata", marker = "(python_full_version >= '3.11' and sys_platform == 'emscripten') or (python_full_version >= '3.11' and sys_platform == 'win32')" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f8/87/4341c6252d1c47b08768c3d25ac487362bf403f0313ddae4a2a26c9b1b4c/pandas-3.0.3.tar.gz", hash = "sha256:696a4a00a2a2a35d4e5deb3fc946641b96c944f02230e4f76137fe35d806c4fc", size = 4651414, upload-time = "2026-05-11T18:54:29.21Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/42/16/b5c76b838fd9bf6ce84d3a53346b8874ec05c5f0040d75ef2c320100cd2a/pandas-3.0.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:455f6f8139d4282188f526868dbc3c828470e88a3d9d59a891bd46a455f21b98", size = 10338495, upload-time = "2026-05-11T18:52:11.558Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/b0/a4ffc4ae74d2d822200dcc46898987d8eb6032d1e2b219cae39da6f5cbcc/pandas-3.0.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4e15135e2ee5df1063313e2425ceef8ac0f4ae775893815b0923651b806a5639", size = 9938250, upload-time = "2026-05-11T18:52:17.005Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/b2/3323601a52caee42c019e370090ca4544b241437240ca04f786cce82b0cf/pandas-3.0.3-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:05f1f1752b8533ea03f7f39a9c15b1a058d067bb48f4748948e7a8691e0510f2", size = 10770558, upload-time = "2026-05-11T18:52:19.865Z" },
+    { url = "https://files.pythonhosted.org/packages/32/f1/bbecd2f867b97abebe0f9b53d750f862251b40337e061b36676ded3d920f/pandas-3.0.3-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a1e45c80cceb3b4a21bc5939d52e8cbd8d9b7305309219d59e9754d9ce09e27", size = 11274611, upload-time = "2026-05-11T18:52:22.622Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/4f/eafabf2d5fae5adf143b4d18d3706c5efdc368a7c4eb1ee8a3eddabbd0f6/pandas-3.0.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:14da8316da4d0c5a77618425996bfb1248ca87fc2c1486e6fde4652bd18b5824", size = 11784670, upload-time = "2026-05-11T18:52:25.4Z" },
+    { url = "https://files.pythonhosted.org/packages/49/44/1eb20389301b57b19cc099a1c2f662501f72f08a65f912d05822613c1532/pandas-3.0.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a55066a0505dae0ba2b50a46637db34b46f9094c65c5d4800794ef6335010938", size = 12353708, upload-time = "2026-05-11T18:52:28.139Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/62/c321f13b5ba1819fc8dca456c7fce578da2dcfecff1abbf0eaddf8406c0f/pandas-3.0.3-cp311-cp311-win_amd64.whl", hash = "sha256:6674ab18ad8c57802867264b00e15e7bb904700cdd9046e3b2fa1fce237439ea", size = 9907609, upload-time = "2026-05-11T18:52:30.982Z" },
+    { url = "https://files.pythonhosted.org/packages/53/85/1b7f563ebc6357c27233a02a96b589bcce1fa9c6eb89fb4f0e56421d277e/pandas-3.0.3-cp311-cp311-win_arm64.whl", hash = "sha256:5cc09a68b3120e0f54870dede8287a7bb1fa463907e4fcec1ea77cab6179bf7a", size = 9165596, upload-time = "2026-05-11T18:52:33.334Z" },
+    { url = "https://files.pythonhosted.org/packages/24/f1/392f8c5bfc16f66a0d2d41561c01627c228fe7ed2a0d056ef11315042570/pandas-3.0.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fed2ff7fd9779120e388e285fc029bd5cf9490cdd2e4166a9ee22c0e49a9ab09", size = 10357846, upload-time = "2026-05-11T18:52:36.143Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/3d/b16412745651e855f357e5e66930248688378853a6e2698a214e331fba1f/pandas-3.0.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b168fc218fd80a6cbdbdbc1a97ddc7889ed057d7eb45f50d866ceab5f39904c4", size = 9899550, upload-time = "2026-05-11T18:52:38.976Z" },
+    { url = "https://files.pythonhosted.org/packages/31/a8/fa2535168fffcedf67f4f6de28d2dd903a747ca7c8ea6989451aaeb3a92f/pandas-3.0.3-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0383c72c75cdcca61a9e116e611143902dbfd08bff356829c2f6d1cf40a9ca8c", size = 10412965, upload-time = "2026-05-11T18:52:41.915Z" },
+    { url = "https://files.pythonhosted.org/packages/65/b6/09b01cdbc15224e2850365192d17b7bdebb8bdbd8780ed221fcdf0d9a515/pandas-3.0.3-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6dc0b3fd2169c9157deed50b4d519553a3655c8c6a96027136d654592be973a9", size = 10894600, upload-time = "2026-05-11T18:52:45.02Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/a4/2eb28f2fccb4ced4a2c79ab2a5dee9ade1ebf44922ebad6fea158c9f95d4/pandas-3.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:7e65d5407dc0b394f509699650e4a2ec01c0514f21850f453fa60f3be79a5dbf", size = 11422824, upload-time = "2026-05-11T18:52:48.058Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/45/830bb57f533a4604b355e07edcb8ea18cf88b5f94e5fca92f27052d7c597/pandas-3.0.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:f8894dc474d648fe7b6ff0ca9b0bd73950d19952bc1a6534540762c5d79d305c", size = 11950889, upload-time = "2026-05-11T18:52:50.905Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/c5/fc1b368f303087d20e8c9bf3d6ceb186263cfac0ade735cd938538bea839/pandas-3.0.3-cp312-cp312-win_amd64.whl", hash = "sha256:c7be265b62cef88e253a941e4698604973736dcfe242fdb5198f0f7bc473cdcc", size = 9755463, upload-time = "2026-05-11T18:52:53.386Z" },
+    { url = "https://files.pythonhosted.org/packages/86/bd/fda8f9705b1b09c6ebe14bfc0fa0e4ec8584d54ea673628f157ff55131af/pandas-3.0.3-cp312-cp312-win_arm64.whl", hash = "sha256:557409bc4178e70ee8d9ddb494798e51ebf6ea59330f6be22c51bab2a7db6c49", size = 9066158, upload-time = "2026-05-11T18:52:56.038Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/90/62d8302883c44308c477e222c3daf7c813a34c8e96985882fbd53d964352/pandas-3.0.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:67b3b64c11910cfa29f4e94a14d3bff9ee693b6fc76055e7cad549cee0aec5fa", size = 10331071, upload-time = "2026-05-11T18:52:58.838Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/ae/6a6493c783a101f165e4356953ba3c74d6f77f0042fa7d753da9dfbb640c/pandas-3.0.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:39436b377d56d2a2e52d0395bdbee171f01068e99af5250509aceeb929f765c7", size = 9875690, upload-time = "2026-05-11T18:53:01.431Z" },
+    { url = "https://files.pythonhosted.org/packages/62/7c/5df8e9f56c69a2769fbe9382a5ef8f2658c007e376434e1e2cbb57ad895f/pandas-3.0.3-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d4be06d68f9ddcfc645b87534911da79a8fbffc7573c80e0edcf42a5020624d8", size = 10381634, upload-time = "2026-05-11T18:53:04.393Z" },
+    { url = "https://files.pythonhosted.org/packages/99/68/1237369725aa617bb358263d535803e3053fdbc593513ec5ed9c9896b5b6/pandas-3.0.3-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a4eeb6830daf35a71cc09649bd823e2b542dac246cdee9614c6e4bd65028cd6a", size = 10891243, upload-time = "2026-05-11T18:53:07.643Z" },
+    { url = "https://files.pythonhosted.org/packages/25/93/77d108e8af7222b4a503ebde0e30215b1c2e4f8e53a526431890f22d5586/pandas-3.0.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:1928e07221f82db493cd4af1e23c1bfca524a19a4699887975bff68f49a72bfb", size = 11388659, upload-time = "2026-05-11T18:53:10.634Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/bd/eff5b4399f332ac386c853f6cd2bd3fa2ca0061b9f36ecd9c4d7c4265649/pandas-3.0.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:51b1fe551acb77dac643c6fda86084d8d446c10fe64b06a9cc29c4cc8540e7f2", size = 11942880, upload-time = "2026-05-11T18:53:13.536Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/20/559ace4200982c3887d0b86bfd0d856a2143ef8ddab63cc07934951a964c/pandas-3.0.3-cp313-cp313-win_amd64.whl", hash = "sha256:a82d532a3351d435432cd913edbccaf8b8e01d4dd0e5ced5a8d2e8ecd94c7e44", size = 9757091, upload-time = "2026-05-11T18:53:16.306Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/66/69055a09fe200f29f922a3eeec4804611900b95f52d932ece3393c3c0c19/pandas-3.0.3-cp313-cp313-win_arm64.whl", hash = "sha256:275c14e0fce14a2ec20eee474aecd305478ea3c1e6f6a9d8fe219a165542717e", size = 9057282, upload-time = "2026-05-11T18:53:18.768Z" },
+    { url = "https://files.pythonhosted.org/packages/57/0e/efe801b0e6811e8e650cd21b7f2608e30f08a7067e2bf6e8752b0d56ee3c/pandas-3.0.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:46997386d528eb40376ecd6b033cf4a8a1e5282580f68f43de875b78cba2199d", size = 10767016, upload-time = "2026-05-11T18:53:21.227Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/dc/eb55135a1d5f0f0519f28da1f609a206d2cad1f9c35c32d51e38dd7261ae/pandas-3.0.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:261e308dfb22448384b7580cf719d2f998fe2966c92893c3e77d14008af1f066", size = 10420210, upload-time = "2026-05-11T18:53:23.982Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/3e/b1d5d955ce33ffecb407465a60bc32769d74fcf68224b7ae67ae11d4dea4/pandas-3.0.3-cp313-cp313t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:dd1a5d1def6a46002e964510bdc67c368aa0951df5d1d9f8365336f5a1f490cd", size = 10336126, upload-time = "2026-05-11T18:53:26.731Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/76/a01261711ab60a22d71b862f0de20e4c504bf80457270ad8cb42110f6abc/pandas-3.0.3-cp313-cp313t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d72828c20c6d6e83e1e22a6a3b47b326b71664112fa9705dcbccfd7a39b62085", size = 10728051, upload-time = "2026-05-11T18:53:29.125Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/21/ea191195e587b18cf682e97f433f81b2d0fbe341380e80a3e0d6e4403c8e/pandas-3.0.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:d26cbe1fcfc12e8fd900e2454163e466b2d3af84f7c75481df7683ffc073d870", size = 11350796, upload-time = "2026-05-11T18:53:32.056Z" },
+    { url = "https://files.pythonhosted.org/packages/64/69/f0eaaf54939f0e8c6768fd06be9af2cef9b36048b96dfb9e1b2c685a807e/pandas-3.0.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:3e91cec1879ada0624fc3dc9953c5cbd60208e59c0db28f540c5d6d47502422f", size = 11799741, upload-time = "2026-05-11T18:53:34.985Z" },
+    { url = "https://files.pythonhosted.org/packages/45/a4/865e0e510cae5fc2194de4db28be638952de942571ba9125934fd9c01d47/pandas-3.0.3-cp313-cp313t-win_amd64.whl", hash = "sha256:08d789b41f87e0905880e293cedf6197ce71fe67cc081358b1e148a491b9bd13", size = 10499958, upload-time = "2026-05-11T18:53:37.857Z" },
+    { url = "https://files.pythonhosted.org/packages/86/54/effdcc3c0ff7a08037889200e148ebe94c16c4f653be078c7b3675955df1/pandas-3.0.3-cp314-cp314-macosx_10_15_x86_64.whl", hash = "sha256:3650109c0f22879df8bd6179ab9ee3d7f1d1d4e7e0094a3f0032d9f51e2e64ac", size = 10336065, upload-time = "2026-05-11T18:53:41.099Z" },
+    { url = "https://files.pythonhosted.org/packages/68/10/bf2d6738d72748b961a3751ab89522d58c54efc36a8e1a12161216cd45cf/pandas-3.0.3-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:bab900348131a7db1f69a7309ef141fd5680f1487094193bcbbb61791573bf8f", size = 9926101, upload-time = "2026-05-11T18:53:43.515Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/e9/e35cf11c8a136e757b956f5f0efdcaa50aecde85ea055f1898dfc68262f3/pandas-3.0.3-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ba7e08b9ac1d54569cd1e256e3668975ed624d6826f7b68df0342b012007bddb", size = 10457553, upload-time = "2026-05-11T18:53:46.394Z" },
+    { url = "https://files.pythonhosted.org/packages/58/3b/1cdec6772bdbaf7b25dab360c59f03cadf05492dd724c6540af905389b07/pandas-3.0.3-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9d71c63ae4ebdbf70209742096f1fc46a83a0613c99d4b23766cced9ff8cd62a", size = 10914065, upload-time = "2026-05-11T18:53:49.134Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/c2/1ef644445fcd72e3627bceec77e3560636f87ddce4ed841afe76b83b5bf9/pandas-3.0.3-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:e3a2ec42c98ffa2565a67e08e218d06d72576d758d90facb7c00805194d8f360", size = 11459188, upload-time = "2026-05-11T18:53:52.527Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/49/4d8d4f42cbc9c4adc7a1870f269c02cbd6cd40d059622c06fb298addcbad/pandas-3.0.3-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:335f62418ed562cfc3c49e9e196375c28b729dcef8543abf4f9438e381bf3c76", size = 11982966, upload-time = "2026-05-11T18:53:55.043Z" },
+    { url = "https://files.pythonhosted.org/packages/38/55/792619469bab9882d8bbd5865d45a72f6478762d04a9af4bf0d08c503e95/pandas-3.0.3-cp314-cp314-win_amd64.whl", hash = "sha256:3c20a521bbb85902f79f7270c80a59e1b5452d96d170c034f207181870f97ac5", size = 9876755, upload-time = "2026-05-11T18:53:58.067Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/af/33c469653b0ba03b50c3a98192d4c07f0c75c66b263ceb097fce0ee97d31/pandas-3.0.3-cp314-cp314-win_arm64.whl", hash = "sha256:a2d2dff8a04f3917b55ab3910c32990f8ddf7eceba114947838cefa976a68977", size = 9198658, upload-time = "2026-05-11T18:54:00.733Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/fa/b8c257bd76b8bd060c3a9151c1fca05e9b9c5e3af5d0f549c0356f6d143d/pandas-3.0.3-cp314-cp314t-macosx_10_15_x86_64.whl", hash = "sha256:0d589105b3c14645af1738ff279b2995102d8f7a03b0a66dc8d95550eb513e04", size = 10787242, upload-time = "2026-05-11T18:54:03.564Z" },
+    { url = "https://files.pythonhosted.org/packages/54/eb/f19206ffb0bf1919002969aa448b4702c6594845156a6f8050674855aac3/pandas-3.0.3-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:13fc1e853d9e04743d11ba75a985ccbc2a317fe07d8af61e445a6fd24dacd6a6", size = 10436369, upload-time = "2026-05-11T18:54:06.311Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/24/c7c39fb4fe22b71a0c2d78bf0c585c600092d85f94f086d2b3b2f6ca27e2/pandas-3.0.3-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:819959dab7bbd0049c15623fbac4e29a191b9528160a61fb1032242d8ced2d9c", size = 10358306, upload-time = "2026-05-11T18:54:09.085Z" },
+    { url = "https://files.pythonhosted.org/packages/16/ec/dd2a9eb7fa1204df88c0864164e35b228ac581062ac612ba0a67fd812e4c/pandas-3.0.3-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:60ae316d3fd75d1858d450d0db0103ea2be3e7d4a95ec2f064f7e2ae63f7b028", size = 10758394, upload-time = "2026-05-11T18:54:11.956Z" },
+    { url = "https://files.pythonhosted.org/packages/95/6e/00c61ea8e85b4f6d8d35e11852a1a4998fc7fafc91c6a602d1cc9c972d64/pandas-3.0.3-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:bd3a518890b400d32f9023722dc9a9a5c969f00b415419a3c06c043f09bb5d7d", size = 11375717, upload-time = "2026-05-11T18:54:14.539Z" },
+    { url = "https://files.pythonhosted.org/packages/31/89/8fc1c268969fac43688d65fd92e67df24bd128d53cb4d2eee534cd307399/pandas-3.0.3-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:9c39be2d709d01fa972a0cabc522389fceca4f3969332ba25a7d6c5802cf976a", size = 11828897, upload-time = "2026-05-11T18:54:17.146Z" },
+    { url = "https://files.pythonhosted.org/packages/56/3b/e7d20dea247a3e6dc0bd8a6953854afbedc03951def4e7371e05e7263e25/pandas-3.0.3-cp314-cp314t-win_amd64.whl", hash = "sha256:4db8c527972a821cf5286b40ccc57642a39bc62e62022b42f99f8a67fca8c3a1", size = 10900855, upload-time = "2026-05-11T18:54:19.72Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/54/68a0978d1ef8502b8492099beaa6e7a0c1b32e3b5d4f677f5810cb08711c/pandas-3.0.3-cp314-cp314t-win_arm64.whl", hash = "sha256:b2c95f8bfc1ee412bf482605d7bfd30c12d1d26bd59fdd91efeef1d4718decb1", size = 9466464, upload-time = "2026-05-11T18:54:22.754Z" },
+]
+
+[[package]]
+name = "parsedatetime"
+version = "2.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a8/20/cb587f6672dbe585d101f590c3871d16e7aec5a576a1694997a3777312ac/parsedatetime-2.6.tar.gz", hash = "sha256:4cb368fbb18a0b7231f4d76119165451c8d2e35951455dfee97c62a87b04d455", size = 60114, upload-time = "2020-05-31T23:50:57.443Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9d/a4/3dd804926a42537bf69fb3ebb9fd72a50ba84f807d95df5ae016606c976c/parsedatetime-2.6-py3-none-any.whl", hash = "sha256:cb96edd7016872f58479e35879294258c71437195760746faffedb692aef000b", size = 42548, upload-time = "2020-05-31T23:50:56.315Z" },
+]
+
+[[package]]
+name = "plotly"
+version = "6.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "narwhals" },
+    { name = "packaging" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/94/fd/d72c292d78aadb93d1a9bcd76bf3c678271040c7cf10abe5788b33040a39/plotly-6.8.0.tar.gz", hash = "sha256:e088e7ddc68d4f70e3d66659224727a45296d71d2b8284181862d3d8f1f0d88f", size = 6915161, upload-time = "2026-06-03T18:33:40.226Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f9/14/abe5ce876ab5b66ee3c691bf537fcd43d037aea55d447aacf74630a8f31e/plotly-6.8.0-py3-none-any.whl", hash = "sha256:13c5c4a0f70b74cab1913eda0de49b826df5931708eb6f9c3010040614700ec8", size = 9902055, upload-time = "2026-06-03T18:33:34.26Z" },
+]
+
+[[package]]
+name = "pluggy"
+version = "1.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/e2/3e91f31a7d2b083fe6ef3fa267035b518369d9511ffab804f839851d2779/pluggy-1.6.0.tar.gz", hash = "sha256:7dcc130b76258d33b90f61b658791dede3486c3e6bfb003ee5c9bfb396dd22f3", size = 69412, upload-time = "2025-05-15T12:30:07.975Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/54/20/4d324d65cc6d9205fabedc306948156824eb9f0ee1633355a8f7ec5c66bf/pluggy-1.6.0-py3-none-any.whl", hash = "sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746", size = 20538, upload-time = "2025-05-15T12:30:06.134Z" },
+]
+
+[[package]]
+name = "polars"
+version = "1.41.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "polars-runtime-32" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ff/f9/aeda46259b0669247a160315d2d51269de9504b9dd2f70acadbcb22f46b7/polars-1.41.2.tar.gz", hash = "sha256:256d6731162371b77f3f29a55eacb8c0fc740ddb1a293a01d2ef5b5393c5c708", size = 737996, upload-time = "2026-05-29T17:39:15.604Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1f/22/28f62d24f7db56ac4343588f9362d49b7b4177e55ac47a466fe696b0099b/polars-1.41.2-py3-none-any.whl", hash = "sha256:23ce9a2910b6e3e8d4258770bf44aa17170958df7af6e85feedf4458a04d8d29", size = 833445, upload-time = "2026-05-29T17:37:05.576Z" },
+]
+
+[[package]]
+name = "polars-runtime-32"
+version = "1.41.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/56/54e3ea0e9b64f327179049e4742241cc6b1d3e8fa414b05a057dd26df367/polars_runtime_32-1.41.2.tar.gz", hash = "sha256:7af09ec1ab053da2c9669e8d15f809a4083a29be05db57111688b8051062af56", size = 2989474, upload-time = "2026-05-29T17:39:17.257Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d6/9b/fe72a3811c0357cdb06c67bdc7695fa1623ad47948fc523195f5ac31037f/polars_runtime_32-1.41.2-cp310-abi3-macosx_10_12_x86_64.whl", hash = "sha256:95a08346dac337357cdb825c8076df7d36da54c4caa59a5cb41d0a30691c5edd", size = 52265283, upload-time = "2026-05-29T17:37:09.407Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/93/fab9da803fd80d9e83ef88c20932f637a10bc611b20415fc322eec84bc44/polars_runtime_32-1.41.2-cp310-abi3-macosx_11_0_arm64.whl", hash = "sha256:dedfaeec2c7f995298da7319dd9431d662e5dd1d0ec51b1459df4a0234ceff52", size = 46571222, upload-time = "2026-05-29T17:37:13.698Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/2a/8843f34a8ac57acd058a39b87b03b580dd352a490e9dae0415e02033bdd4/polars_runtime_32-1.41.2-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:18eea22c5cc34e27f8a60950458ad81e6a9ea75e89363ca1367e14e7e7f781fc", size = 50409372, upload-time = "2026-05-29T17:37:17.875Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/c6/92b352fe88cf51bd0a19fb99e1c0cbe46aa26c14dcf7995b89869cd932ae/polars_runtime_32-1.41.2-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2630540dfdfb0f36f9b04a07c7c2e3f50bf2ad384113263c1c812007ee9141e0", size = 56405484, upload-time = "2026-05-29T17:37:22.684Z" },
+    { url = "https://files.pythonhosted.org/packages/74/c4/bae3174c3b02f6b441d2e58594387abcd509f67a098f682a83b195f08966/polars_runtime_32-1.41.2-cp310-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:20e969e08f9b137e233c04cc04de73d9795f89eb77d34854e40a025965a43763", size = 50603512, upload-time = "2026-05-29T17:37:27.422Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/ed/f2d26ae02d92c2689056838ed59e2a626326ad23c2831d58637d25f6c82a/polars_runtime_32-1.41.2-cp310-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:e7016a3deb641b64a31447abbbee0f34bd020a6a9ae34ee6b743837def15e2a4", size = 54328561, upload-time = "2026-05-29T17:37:32.587Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/c4/9c3831cc885dc7769e59abf8f583821a5fb4403fd0e4eba0ccc6d47a3d4b/polars_runtime_32-1.41.2-cp310-abi3-win_amd64.whl", hash = "sha256:1e5e5377c315e0dcafdfb2a31adc546abbaeb3f9cb1864e6536523d2af473265", size = 51978643, upload-time = "2026-05-29T17:37:37.443Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/c6/79e9f3f270270d7ed5575d92b7bfef49f01abd9275447161275b23b553a8/polars_runtime_32-1.41.2-cp310-abi3-win_arm64.whl", hash = "sha256:843d96f69d18eca53429c1198e58891db7f18111f83b9c419bb45ad9d73eaed5", size = 46006901, upload-time = "2026-05-29T17:37:42.522Z" },
+]
+
+[[package]]
+name = "protobuf"
+version = "7.35.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/da/01/9ef0afd7999eb9badb3a768b4aedd78c86d4c65cfaf1958ab276199e76b4/protobuf-7.35.1.tar.gz", hash = "sha256:ce115a26fe0c39a2c29973d914d327e516a6455464489fe3cd1e51a1b354f81a", size = 458717, upload-time = "2026-06-11T21:55:40.257Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/10/03/8aeeb7458d22546bf64b5250ca1daeb5ff757d900e8e4a7476c6f0db843e/protobuf-7.35.1-cp310-abi3-macosx_10_9_universal2.whl", hash = "sha256:24f857477359a85c0c235261b8ba905fd51b2562f4a64ca1df5473f29850cbf6", size = 433226, upload-time = "2026-06-11T21:55:31.719Z" },
+    { url = "https://files.pythonhosted.org/packages/37/4b/dfb89eb0e652a1ff073c39a59fb5e3a83cfe9b57a2c83fa6d78270101767/protobuf-7.35.1-cp310-abi3-manylinux2014_aarch64.whl", hash = "sha256:11d6b0ec246892d85215b0a13ca6e0233cf5284b68f0ac02646427f4ff88a799", size = 328847, upload-time = "2026-06-11T21:55:34.035Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/58/dc12f2cd484951524af6e3382c785869b9b3fb5e52ee95ae23add53ee8f9/protobuf-7.35.1-cp310-abi3-manylinux2014_s390x.whl", hash = "sha256:b73f9489a4b8b1c9cb1f8ed951c736392592edb24b9d6819f36d2e10b171d5b4", size = 344030, upload-time = "2026-06-11T21:55:34.941Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/be/5b3cfe508bfab6761414ff944e3366eb13be4fd71efcd69450f89ba39f43/protobuf-7.35.1-cp310-abi3-manylinux2014_x86_64.whl", hash = "sha256:74758715c53d7158fb76caf4f0cfdacc5329a4b1bb994f865d6cf302d413a1c4", size = 327130, upload-time = "2026-06-11T21:55:35.921Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/bc/6d6c7ba8709c85f8f2c390b2b118d6fb08a783676a572271851bf45a7d22/protobuf-7.35.1-cp310-abi3-win32.whl", hash = "sha256:353652e4efd0bca5b5fc2656abf8307ef351f0cf938c9eba09f0e09c20a25c30", size = 428945, upload-time = "2026-06-11T21:55:37.034Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/19/8d0cb6f20a1ef7b18f1c8986ad5783f22f84cce39c6ce9a6e645ea55192e/protobuf-7.35.1-cp310-abi3-win_amd64.whl", hash = "sha256:230a75ddfc2de4806e56696ce9640c1cdfdb6543b7cfce98d42a4c0a0e7bdb87", size = 439996, upload-time = "2026-06-11T21:55:38.123Z" },
+    { url = "https://files.pythonhosted.org/packages/19/c7/5f7c636ec43e0c545e28d1f1db71990108306f7bdcb89f069ba97e428e7f/protobuf-7.35.1-py3-none-any.whl", hash = "sha256:4bc97768d8fe4ad6743c8a19403e314511ed9f6d13205b687e52421c023ac1b9", size = 171659, upload-time = "2026-06-11T21:55:39.155Z" },
+]
+
+[[package]]
+name = "psutil"
+version = "7.2.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/aa/c6/d1ddf4abb55e93cebc4f2ed8b5d6dbad109ecb8d63748dd2b20ab5e57ebe/psutil-7.2.2.tar.gz", hash = "sha256:0746f5f8d406af344fd547f1c8daa5f5c33dbc293bb8d6a16d80b4bb88f59372", size = 493740, upload-time = "2026-01-28T18:14:54.428Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/51/08/510cbdb69c25a96f4ae523f733cdc963ae654904e8db864c07585ef99875/psutil-7.2.2-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:2edccc433cbfa046b980b0df0171cd25bcaeb3a68fe9022db0979e7aa74a826b", size = 130595, upload-time = "2026-01-28T18:14:57.293Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/f5/97baea3fe7a5a9af7436301f85490905379b1c6f2dd51fe3ecf24b4c5fbf/psutil-7.2.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:e78c8603dcd9a04c7364f1a3e670cea95d51ee865e4efb3556a3a63adef958ea", size = 131082, upload-time = "2026-01-28T18:14:59.732Z" },
+    { url = "https://files.pythonhosted.org/packages/37/d6/246513fbf9fa174af531f28412297dd05241d97a75911ac8febefa1a53c6/psutil-7.2.2-cp313-cp313t-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1a571f2330c966c62aeda00dd24620425d4b0cc86881c89861fbc04549e5dc63", size = 181476, upload-time = "2026-01-28T18:15:01.884Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/b5/9182c9af3836cca61696dabe4fd1304e17bc56cb62f17439e1154f225dd3/psutil-7.2.2-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:917e891983ca3c1887b4ef36447b1e0873e70c933afc831c6b6da078ba474312", size = 184062, upload-time = "2026-01-28T18:15:04.436Z" },
+    { url = "https://files.pythonhosted.org/packages/16/ba/0756dca669f5a9300d0cbcbfae9a4c30e446dfc7440ffe43ded5724bfd93/psutil-7.2.2-cp313-cp313t-win_amd64.whl", hash = "sha256:ab486563df44c17f5173621c7b198955bd6b613fb87c71c161f827d3fb149a9b", size = 139893, upload-time = "2026-01-28T18:15:06.378Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/61/8fa0e26f33623b49949346de05ec1ddaad02ed8ba64af45f40a147dbfa97/psutil-7.2.2-cp313-cp313t-win_arm64.whl", hash = "sha256:ae0aefdd8796a7737eccea863f80f81e468a1e4cf14d926bd9b6f5f2d5f90ca9", size = 135589, upload-time = "2026-01-28T18:15:08.03Z" },
+    { url = "https://files.pythonhosted.org/packages/81/69/ef179ab5ca24f32acc1dac0c247fd6a13b501fd5534dbae0e05a1c48b66d/psutil-7.2.2-cp314-cp314t-macosx_10_15_x86_64.whl", hash = "sha256:eed63d3b4d62449571547b60578c5b2c4bcccc5387148db46e0c2313dad0ee00", size = 130664, upload-time = "2026-01-28T18:15:09.469Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/64/665248b557a236d3fa9efc378d60d95ef56dd0a490c2cd37dafc7660d4a9/psutil-7.2.2-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:7b6d09433a10592ce39b13d7be5a54fbac1d1228ed29abc880fb23df7cb694c9", size = 131087, upload-time = "2026-01-28T18:15:11.724Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/2e/e6782744700d6759ebce3043dcfa661fb61e2fb752b91cdeae9af12c2178/psutil-7.2.2-cp314-cp314t-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1fa4ecf83bcdf6e6c8f4449aff98eefb5d0604bf88cb883d7da3d8d2d909546a", size = 182383, upload-time = "2026-01-28T18:15:13.445Z" },
+    { url = "https://files.pythonhosted.org/packages/57/49/0a41cefd10cb7505cdc04dab3eacf24c0c2cb158a998b8c7b1d27ee2c1f5/psutil-7.2.2-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e452c464a02e7dc7822a05d25db4cde564444a67e58539a00f929c51eddda0cf", size = 185210, upload-time = "2026-01-28T18:15:16.002Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/2c/ff9bfb544f283ba5f83ba725a3c5fec6d6b10b8f27ac1dc641c473dc390d/psutil-7.2.2-cp314-cp314t-win_amd64.whl", hash = "sha256:c7663d4e37f13e884d13994247449e9f8f574bc4655d509c3b95e9ec9e2b9dc1", size = 141228, upload-time = "2026-01-28T18:15:18.385Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/fc/f8d9c31db14fcec13748d373e668bc3bed94d9077dbc17fb0eebc073233c/psutil-7.2.2-cp314-cp314t-win_arm64.whl", hash = "sha256:11fe5a4f613759764e79c65cf11ebdf26e33d6dd34336f8a337aa2996d71c841", size = 136284, upload-time = "2026-01-28T18:15:19.912Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/36/5ee6e05c9bd427237b11b3937ad82bb8ad2752d72c6969314590dd0c2f6e/psutil-7.2.2-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:ed0cace939114f62738d808fdcecd4c869222507e266e574799e9c0faa17d486", size = 129090, upload-time = "2026-01-28T18:15:22.168Z" },
+    { url = "https://files.pythonhosted.org/packages/80/c4/f5af4c1ca8c1eeb2e92ccca14ce8effdeec651d5ab6053c589b074eda6e1/psutil-7.2.2-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:1a7b04c10f32cc88ab39cbf606e117fd74721c831c98a27dc04578deb0c16979", size = 129859, upload-time = "2026-01-28T18:15:23.795Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/70/5d8df3b09e25bce090399cf48e452d25c935ab72dad19406c77f4e828045/psutil-7.2.2-cp36-abi3-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:076a2d2f923fd4821644f5ba89f059523da90dc9014e85f8e45a5774ca5bc6f9", size = 155560, upload-time = "2026-01-28T18:15:25.976Z" },
+    { url = "https://files.pythonhosted.org/packages/63/65/37648c0c158dc222aba51c089eb3bdfa238e621674dc42d48706e639204f/psutil-7.2.2-cp36-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b0726cecd84f9474419d67252add4ac0cd9811b04d61123054b9fb6f57df6e9e", size = 156997, upload-time = "2026-01-28T18:15:27.794Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/13/125093eadae863ce03c6ffdbae9929430d116a246ef69866dad94da3bfbc/psutil-7.2.2-cp36-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:fd04ef36b4a6d599bbdb225dd1d3f51e00105f6d48a28f006da7f9822f2606d8", size = 148972, upload-time = "2026-01-28T18:15:29.342Z" },
+    { url = "https://files.pythonhosted.org/packages/04/78/0acd37ca84ce3ddffaa92ef0f571e073faa6d8ff1f0559ab1272188ea2be/psutil-7.2.2-cp36-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:b58fabe35e80b264a4e3bb23e6b96f9e45a3df7fb7eed419ac0e5947c61e47cc", size = 148266, upload-time = "2026-01-28T18:15:31.597Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/90/e2159492b5426be0c1fef7acba807a03511f97c5f86b3caeda6ad92351a7/psutil-7.2.2-cp37-abi3-win_amd64.whl", hash = "sha256:eb7e81434c8d223ec4a219b5fc1c47d0417b12be7ea866e24fb5ad6e84b3d988", size = 137737, upload-time = "2026-01-28T18:15:33.849Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/c7/7bb2e321574b10df20cbde462a94e2b71d05f9bbda251ef27d104668306a/psutil-7.2.2-cp37-abi3-win_arm64.whl", hash = "sha256:8c233660f575a5a89e6d4cb65d9f938126312bca76d8fe087b947b3a1aaac9ee", size = 134617, upload-time = "2026-01-28T18:15:36.514Z" },
+]
+
+[[package]]
+name = "pyarrow"
+version = "24.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/91/13/13e1069b351bdc3881266e11147ffccf687505dbb0ea74036237f5d454a5/pyarrow-24.0.0.tar.gz", hash = "sha256:85fe721a14dd823aca09127acbb06c3ca723efbd436c004f16bca601b04dcc83", size = 1180261, upload-time = "2026-04-21T10:51:25.837Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a5/bf/a34fee1d624152124fa8355c42f34195ad5fe5233ce5bb87946432047d52/pyarrow-24.0.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:7c2b98645d576a0b9616892ead22b64a83a5f043c5e2ca15ebcefcb5b70c80cb", size = 35076681, upload-time = "2026-04-21T08:51:46.845Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/41/64180033d7027afce12dc96d0fe1f504c6fa112190582b458acea2399530/pyarrow-24.0.0-cp310-cp310-macosx_12_0_x86_64.whl", hash = "sha256:644a246325b8c69c595ad1dd4b463eba4b0cdb731370e4a86137d433208d6147", size = 36684260, upload-time = "2026-04-21T08:51:53.642Z" },
+    { url = "https://files.pythonhosted.org/packages/57/02/9b9320e673dd8a99411fac78690f3df92f6dd6f59754c750110bca66d64e/pyarrow-24.0.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:3a577bd840ca83f646f0a625dbc571dba7044c43c2d1503afc378b570954345c", size = 45698566, upload-time = "2026-04-21T10:46:02.133Z" },
+    { url = "https://files.pythonhosted.org/packages/67/33/f75e91b9a64c3f33c787e263c93b871ad91b8a4a68c1d5cebddd9840e835/pyarrow-24.0.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:e3268e43984d0b1a185c89b4cfff282a7ead12fc93f56cfd7088bdbcbe727041", size = 48835562, upload-time = "2026-04-21T10:46:10.278Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/63/097510448e47e4091faa41c43ba92f97cecaab8f4535b56a3d149578f634/pyarrow-24.0.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:2392d954fcb920f42d230284b677605e4e2fbb11f2821e823e642abd67fbb491", size = 49394997, upload-time = "2026-04-21T10:46:18.08Z" },
+    { url = "https://files.pythonhosted.org/packages/60/6b/c047d6222ab279024a062742d1807e2fbaf27bba88a98637299ff47b9236/pyarrow-24.0.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:bec9373df11544592b0ba7ec2af0e35059e5f0e7647c6183a854dedd193298f1", size = 51911424, upload-time = "2026-04-21T10:46:25.347Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/ba/464cc70761c2a525d97ebd84e21c31ebd47f3ef4bdcee117009f51c46f24/pyarrow-24.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:c42ab9439498270139cc63e18847a02afe5c8b3ed9c931266533cfe378bd3591", size = 27251730, upload-time = "2026-04-21T10:46:30.913Z" },
+    { url = "https://files.pythonhosted.org/packages/62/c9/a47ab7ece0d86cbe6678418a0fbd1ac4bb493b9184a3891dfa0e7f287ae0/pyarrow-24.0.0-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:b0e131f880cda8d04e076cee175a46fc0e8bc8b65c99c6c09dff6669335fde74", size = 35068898, upload-time = "2026-04-21T10:46:36.599Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/bc/8db86617a9a58008acf8913d6fed68ea2a46acb6de928db28d724c891a68/pyarrow-24.0.0-cp311-cp311-macosx_12_0_x86_64.whl", hash = "sha256:1b2fe7f9a5566401a0ef2571f197eb92358925c1f0c8dba305d6e43ea0871bb3", size = 36679915, upload-time = "2026-04-21T10:46:42.602Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/8e/fb178720400ef69db251eb4a9c3ccf4af269bc1feb5055529b8fc87170d1/pyarrow-24.0.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:0b3537c00fb8d384f15ac1e79b6eb6db04a16514c8c1d22e59a9b95c8ba42868", size = 45697931, upload-time = "2026-04-21T10:46:48.403Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/27/99c42abe8e21b44f4917f62631f3aa31404882a2c41d8a4cd5c110e13d52/pyarrow-24.0.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:14e31a3c9e35f1ab6356c6378f6f72830e6d2d5f1791df3774a7b097d18a6a1e", size = 48837449, upload-time = "2026-04-21T10:46:55.329Z" },
+    { url = "https://files.pythonhosted.org/packages/36/b6/333749e2666e9032891125bf9c691146e92901bece62030ac1430e2e7c88/pyarrow-24.0.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b7d9a514e73bc42711e6a35aaccf3587c520024fe0a25d830a1a8a27c15f4f57", size = 49395949, upload-time = "2026-04-21T10:47:01.869Z" },
+    { url = "https://files.pythonhosted.org/packages/17/25/c5201706a2dd374e8ba6ee3fd7a8c89fb7ffc16eed5217a91fd2bd7f7626/pyarrow-24.0.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:b196eb3f931862af3fa84c2a253514d859c08e0d8fe020e07be12e75a5a9780c", size = 51912986, upload-time = "2026-04-21T10:47:09.872Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/d2/4d1bbba65320b21a49678d6fbdc6ff7c649251359fdcfc03568c4136231d/pyarrow-24.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:35405aecb474e683fb36af650618fd5340ee5471fc65a21b36076a18bbc6c981", size = 27255371, upload-time = "2026-04-21T10:47:15.943Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/a9/9686d9f07837f91f775e8932659192e02c74f9d8920524b480b85212cc68/pyarrow-24.0.0-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:6233c9ed9ab9d1db47de57d9753256d9dcffbf42db341576099f0fd9f6bf4810", size = 34981559, upload-time = "2026-04-21T10:47:22.17Z" },
+    { url = "https://files.pythonhosted.org/packages/80/b6/0ddf0e9b6ead3474ab087ae598c76b031fc45532bf6a63f3a553440fb258/pyarrow-24.0.0-cp312-cp312-macosx_12_0_x86_64.whl", hash = "sha256:f7616236ec1bc2b15bfdec22a71ab38851c86f8f05ff64f379e1278cf20c634a", size = 36663654, upload-time = "2026-04-21T10:47:28.315Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/3b/926382efe8ce27ba729071d3566ade6dfb86bdf112f366000196b2f5780a/pyarrow-24.0.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:1617043b99bd33e5318ae18eb2919af09c71322ef1ca46566cdafc6e6712fb66", size = 45679394, upload-time = "2026-04-21T10:47:34.821Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/7a/829f7d9dfd37c207206081d6dad474d81dde29952401f07f2ba507814818/pyarrow-24.0.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:6165461f55ef6314f026de6638d661188e3455d3ec49834556a0ebbdbace18bb", size = 48863122, upload-time = "2026-04-21T10:47:42.056Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/e8/f88ce625fe8babaae64e8db2d417c7653adb3019b08aae85c5ed787dc816/pyarrow-24.0.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3b13dedfe76a0ad2d1d859b0811b53827a4e9d93a0bcb05cf59333ab4980cc7e", size = 49376032, upload-time = "2026-04-21T10:47:48.967Z" },
+    { url = "https://files.pythonhosted.org/packages/36/7a/82c363caa145fff88fb475da50d3bf52bb024f61917be5424c3392eaf878/pyarrow-24.0.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:25ea65d868eb04015cd18e6df2fbe98f07e5bda2abefabcb88fce39a947716f6", size = 51929490, upload-time = "2026-04-21T10:47:55.981Z" },
+    { url = "https://files.pythonhosted.org/packages/66/1c/e3e72c8014ad2743ca64a701652c733cc5cbcee15c0463a32a8c55518d9e/pyarrow-24.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:295f0a7f2e242dabd513737cf076007dc5b2d59237e3eca37b05c0c6446f3826", size = 27355660, upload-time = "2026-04-21T10:48:01.718Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/d3/a1abf004482026ddc17f4503db227787fa3cfe41ec5091ff20e4fea55e57/pyarrow-24.0.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:02b001b3ed4723caa44f6cd1af2d5c86aa2cf9971dacc2ffa55b21237713dfba", size = 34976759, upload-time = "2026-04-21T10:48:07.258Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/4a/34f0a36d28a2dd32225301b79daad44e243dc1a2bb77d43b60749be255c4/pyarrow-24.0.0-cp313-cp313-macosx_12_0_x86_64.whl", hash = "sha256:04920d6a71aabd08a0417709efce97d45ea8e6fb733d9ca9ecffb13c67839f68", size = 36658471, upload-time = "2026-04-21T10:48:13.347Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/78/543b94712ae8bb1a6023bcc1acf1a740fbff8286747c289cd9468fced2a5/pyarrow-24.0.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:a964266397740257f16f7bb2e4f08a0c81454004beab8ff59dd531b73610e9f2", size = 45675981, upload-time = "2026-04-21T10:48:20.201Z" },
+    { url = "https://files.pythonhosted.org/packages/84/9f/8fb7c222b100d314137fa40ec050de56cd8c6d957d1cfff685ce72f15b17/pyarrow-24.0.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:6f066b179d68c413374294bc1735f68475457c933258df594443bb9d88ddc2a0", size = 48859172, upload-time = "2026-04-21T10:48:27.541Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/d3/1ea72538e6c8b3b475ed78d1049a2c518e655761ea50fe1171fc855fcab7/pyarrow-24.0.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:1183baeb14c5f587b1ec52831e665718ce632caab84b7cd6b85fd44f96114495", size = 49385733, upload-time = "2026-04-21T10:48:34.7Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/be/c3d8b06a1ba35f2260f8e1f771abbee7d5e345c0937aab90675706b1690a/pyarrow-24.0.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:806f24b4085453c197a5078218d1ee08783ebbba271badd153d1ae22a3ee804f", size = 51934335, upload-time = "2026-04-21T10:48:42.099Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/62/89e07a1e7329d2cde3e3c6994ba0839a24977a2beda8be6005ea3d860b99/pyarrow-24.0.0-cp313-cp313-win_amd64.whl", hash = "sha256:e4505fc6583f7b05ab854934896bcac8253b04ac1171a77dfb73efef92076d91", size = 27271748, upload-time = "2026-04-21T10:49:42.532Z" },
+    { url = "https://files.pythonhosted.org/packages/17/1a/cff3a59f80b5b1658549d46611b67163f65e0664431c076ad728bf9d5af4/pyarrow-24.0.0-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:1a4e45017efbf115032e4475ee876d525e0e36c742214fbe405332480ecd6275", size = 35238554, upload-time = "2026-04-21T10:48:48.526Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/99/cce0f42a327bfef2c420fb6078a3eb834826e5d6697bf3009fe11d2ad051/pyarrow-24.0.0-cp313-cp313t-macosx_12_0_x86_64.whl", hash = "sha256:7986f1fa71cee060ad00758bcc79d3a93bab8559bf978fab9e53472a2e25a17b", size = 36782301, upload-time = "2026-04-21T10:48:55.181Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/66/8e560d5ff6793ca29aca213c53eec0dd482dd46cb93b2819e5aab52e4252/pyarrow-24.0.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:d3e0b61e8efb24ed38898e5cdc5fffa9124be480008d401a1f8071500494ae42", size = 45721929, upload-time = "2026-04-21T10:49:03.676Z" },
+    { url = "https://files.pythonhosted.org/packages/27/0c/a26e25505d030716e078d9f16eb74973cbf0b33b672884e9f9da1c83b871/pyarrow-24.0.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:55a3bc1e3df3b5567b7d27ef551b2283f0c68a5e86f1cd56abc569da4f31335b", size = 48825365, upload-time = "2026-04-21T10:49:11.714Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/eb/771f9ecb0c65e73fe9dccdd1717901b9594f08c4515d000c7c62df573811/pyarrow-24.0.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:641f795b361874ac9da5294f8f443dfdbee355cf2bd9e3b8d97aaac2306b9b37", size = 49451819, upload-time = "2026-04-21T10:49:21.474Z" },
+    { url = "https://files.pythonhosted.org/packages/48/da/61ae89a88732f5a785646f3ec6125dbb640fa98a540eb2b9889caa561403/pyarrow-24.0.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:8adc8e6ce5fccf5dc707046ae4914fd537def529709cc0d285d37a7f9cd442ca", size = 51909252, upload-time = "2026-04-21T10:49:31.164Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/1a/8dd5cafab7b66573fa91c03d06d213356ad4edd71813aa75e08ce2b3a844/pyarrow-24.0.0-cp313-cp313t-win_amd64.whl", hash = "sha256:9b18371ad2f44044b81a8d23bc2d8a9b6a6226dca775e8e16cfee640473d6c5d", size = 27388127, upload-time = "2026-04-21T10:49:37.334Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/80/d022a34ff05d2cbedd8ccf841fc1f532ecfa9eb5ed1711b56d0e0ea71fc9/pyarrow-24.0.0-cp314-cp314-macosx_12_0_arm64.whl", hash = "sha256:1cc9057f0319e26333b357e17f3c2c022f1a83739b48a88b25bfd5fa2dc18838", size = 35007997, upload-time = "2026-04-21T10:49:48.796Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/ff/f01485fda6f4e5d441afb8dd5e7681e4db18826c1e271852f5d3957d6a80/pyarrow-24.0.0-cp314-cp314-macosx_12_0_x86_64.whl", hash = "sha256:e6f1278ee4785b6db21229374a1c9e54ec7c549de5d1efc9630b6207de7e170b", size = 36678720, upload-time = "2026-04-21T10:49:55.858Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/c2/2d2d5fea814237923f71b36495211f20b43a1576f9a4d6da7e751a64ec6f/pyarrow-24.0.0-cp314-cp314-manylinux_2_28_aarch64.whl", hash = "sha256:adbbedc55506cbdabb830890444fb856bfb0060c46c6f8026c6c2f2cf86ae795", size = 45741852, upload-time = "2026-04-21T10:50:04.624Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/3a/28ba9c1c1ebdbb5f1b94dfebb46f207e52e6a554b7fe4132540fde29a3a0/pyarrow-24.0.0-cp314-cp314-manylinux_2_28_x86_64.whl", hash = "sha256:ae8a1145af31d903fa9bb166824d7abe9b4681a000b0159c9fb99c11bc11ad26", size = 48889852, upload-time = "2026-04-21T10:50:12.293Z" },
+    { url = "https://files.pythonhosted.org/packages/df/51/4a389acfd31dca009f8fb82d7f510bb4130f2b3a8e18cf00194d0687d8ac/pyarrow-24.0.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:d7027eba1df3b2069e2e8d80f644fa0918b68c46432af3d088ddd390d063ecde", size = 49445207, upload-time = "2026-04-21T10:50:20.677Z" },
+    { url = "https://files.pythonhosted.org/packages/19/4b/0bab2b23d2ae901b1b9a03c0efd4b2d070256f8ce3fc43f6e58c167b2081/pyarrow-24.0.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:e56a1ffe9bf7b727432b89104cc0849c21582949dd7bdcb34f17b2001a351a76", size = 51954117, upload-time = "2026-04-21T10:50:29.14Z" },
+    { url = "https://files.pythonhosted.org/packages/29/88/f4e9145da0417b3d2c12035a8492b35ff4a3dbc653e614fcfb51d9dedb38/pyarrow-24.0.0-cp314-cp314-win_amd64.whl", hash = "sha256:38be1808cdd068605b787e6ca9119b27eb275a0234e50212c3492331680c3b1e", size = 28001155, upload-time = "2026-04-21T10:51:22.337Z" },
+    { url = "https://files.pythonhosted.org/packages/79/4f/46a49a63f43526da895b1a45bbb51d5baf8e4d77159f8528fc3e5490007f/pyarrow-24.0.0-cp314-cp314t-macosx_12_0_arm64.whl", hash = "sha256:418e48ce50a45a6a6c73c454677203a9c75c966cb1e92ca3370959185f197a05", size = 35250387, upload-time = "2026-04-21T10:50:35.552Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/da/d5e0cd5ef00796922404806d5f00325cdadc3441ce2c13fe7115f2df9a64/pyarrow-24.0.0-cp314-cp314t-macosx_12_0_x86_64.whl", hash = "sha256:2f16197705a230a78270cdd4ea8a1d57e86b2fdcbc34a1f6aebc72e65c986f9a", size = 36797102, upload-time = "2026-04-21T10:50:42.417Z" },
+    { url = "https://files.pythonhosted.org/packages/34/c7/5904145b0a593a05236c882933d439b5720f0a145381179063722fbfc123/pyarrow-24.0.0-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:fb24ac194bfc5e86839d7dcd52092ee31e5fe6733fe11f5e3b06ef0812b20072", size = 45745118, upload-time = "2026-04-21T10:50:49.324Z" },
+    { url = "https://files.pythonhosted.org/packages/13/d3/cca42fe166d1c6e4d5b80e530b7949104d10e17508a90ae202dac205ce2a/pyarrow-24.0.0-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:9700ebd9a51f5895ce75ff4ac4b3c47a7d4b42bc618be8e713e5d56bacf5f931", size = 48844765, upload-time = "2026-04-21T10:50:55.579Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/49/942c3b79878ba928324d1e17c274ed84581db8c0a749b24bcf4cbdf15bd3/pyarrow-24.0.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:d8ddd2768da81d3ee08cfea9b597f4abb4e8e1dc8ae7e204b608d23a0d3ab699", size = 49471890, upload-time = "2026-04-21T10:51:02.439Z" },
+    { url = "https://files.pythonhosted.org/packages/76/97/ff71431000a75d84135a1ace5ca4ba11726a231a8007bbb320a4c54075d5/pyarrow-24.0.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:61a3d7eaa97a14768b542f3d284dc6400dd2470d9f080708b13cd46b6ae18136", size = 51932250, upload-time = "2026-04-21T10:51:10.576Z" },
+    { url = "https://files.pythonhosted.org/packages/51/be/6f79d55816d5c22557cf27533543d5d70dfe692adfbee4b99f2760674f38/pyarrow-24.0.0-cp314-cp314t-win_amd64.whl", hash = "sha256:c91d00057f23b8d353039520dc3a6c09d8608164c692e9f59a175a42b2ae0c19", size = 28131282, upload-time = "2026-04-21T10:51:16.815Z" },
+]
+
+[[package]]
+name = "pydantic"
+version = "2.13.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "annotated-types" },
+    { name = "pydantic-core" },
+    { name = "typing-extensions" },
+    { name = "typing-inspection" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/18/a5/b60d21ac674192f8ab0ba4e9fd860690f9b4a6e51ca5df118733b487d8d6/pydantic-2.13.4.tar.gz", hash = "sha256:c40756b57adaa8b1efeeced5c196f3f3b7c435f90e84ea7f443901bec8099ef6", size = 844775, upload-time = "2026-05-06T13:43:05.343Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fd/7b/122376b1fd3c62c1ed9dc80c931ace4844b3c55407b6fb2d199377c9736f/pydantic-2.13.4-py3-none-any.whl", hash = "sha256:45a282cde31d808236fd7ea9d919b128653c8b38b393d1c4ab335c62924d9aba", size = 472262, upload-time = "2026-05-06T13:43:02.641Z" },
+]
+
+[[package]]
+name = "pydantic-core"
+version = "2.46.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9d/56/921726b776ace8d8f5db44c4ef961006580d91dc52b803c489fafd1aa249/pydantic_core-2.46.4.tar.gz", hash = "sha256:62f875393d7f270851f20523dd2e29f082bcc82292d66db2b64ea71f64b6e1c1", size = 471464, upload-time = "2026-05-06T13:37:06.98Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e7/08/f1ba952f1c8ae5581c70fa9c6da89f247b83e3dd8c09c035d5d7931fc23d/pydantic_core-2.46.4-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:a396dcc17e5a0b164dbe026896245a4fa9ff402edca1dff0be3d53a517f74de4", size = 2113146, upload-time = "2026-05-06T13:37:36.537Z" },
+    { url = "https://files.pythonhosted.org/packages/56/c6/65f646c7ff09bd257f660434adb45c4dfcbbcebcc030562fecf6f5bf887d/pydantic_core-2.46.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:da4b951fe36dc7c3a1ccb4e3cd1747c3542b8c9ceede8fc86cae054e764485f5", size = 1949769, upload-time = "2026-05-06T13:37:46.365Z" },
+    { url = "https://files.pythonhosted.org/packages/64/ba/bfb1d928fd5b49e1258935ff104ae356e9fd89384a55bf9f847e9193ad40/pydantic_core-2.46.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bb63e0198ca18aad131c089b9204c23079c3afa95487e561f4c522d519e55aba", size = 1974958, upload-time = "2026-05-06T13:37:28.611Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/74/76223bfb117b64af743c9b6670d1364516f5c0604f96b48f3272f6af6cc6/pydantic_core-2.46.4-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:f47286a97f0bc9b8859519809077b91b2cefe4ae47fcbf5e466a009c1c5d742b", size = 2042118, upload-time = "2026-05-06T13:36:55.216Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/7b/848732968bc8f48f3187542f08358b9d842db564147b256669426ebb1652/pydantic_core-2.46.4-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:905a0ed8ea6f2d61c1738835f99b699348d7857379083e5fc497fa0c967a407c", size = 2222876, upload-time = "2026-05-06T13:38:25.455Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/2f/e90b63ee2e14bd8d3db8f705a6d75d64e6ee1b7c2c8833747ce706e1e0ce/pydantic_core-2.46.4-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ea793e075b70290d89d8142074262885d3f7da19634845135751bd6344f73b50", size = 2286703, upload-time = "2026-05-06T13:37:53.304Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/1e/acc4d70f88a0a277e4a1fa77ebb985ceabaf900430f875bf9338e11c9420/pydantic_core-2.46.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:395aebd9183f9d112f569aeb5b2214d1a10a33bec8456447f7fbdfa51d38d4cd", size = 2092042, upload-time = "2026-05-06T13:38:46.981Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/da/0a422b57bf8504102bf3c4ccea9c41bab5a5cee6a54650acf8faf67f5a24/pydantic_core-2.46.4-cp310-cp310-manylinux_2_31_riscv64.whl", hash = "sha256:b078afbc25f3a1436c7a1d2cd3e322497ee99615ba97c563566fdf46aff1ee01", size = 2117231, upload-time = "2026-05-06T13:39:23.146Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/2a/2ac13c3af305843e23c5078c53d135656b3f05a2fd78cb7bbbb12e97b473/pydantic_core-2.46.4-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f747929cf940cddb5b3668a390056ddd5ba2e5010615ea2dcf4f9c4f3ab8791d", size = 2168388, upload-time = "2026-05-06T13:40:08.06Z" },
+    { url = "https://files.pythonhosted.org/packages/72/04/2beacf7e1607e93eefe4aed1b4709f079b905fb77530179d4f7c71745f22/pydantic_core-2.46.4-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:daa27d92c36f24388fe3ad306b174781c747627f134452e4f128ea00ce1fe8c4", size = 2184769, upload-time = "2026-05-06T13:38:13.901Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/29/d2b9fd9f539133548eaf622c06a4ce176cb46ac59f32d0359c4abc0de047/pydantic_core-2.46.4-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:19e51f073cd3df251856a8a4189fbdf1de4012c3ebacfb1884f94f1eb406079f", size = 2319312, upload-time = "2026-05-06T13:39:08.24Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/af/0f7a5b85fec6075bea96e3ef9187de38fccced0de92c1e7feda8d5cc7bb9/pydantic_core-2.46.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:c1747f85cee84c26985853c6f3d9bd3e75da5212912443fa111c113b9c246f39", size = 2361817, upload-time = "2026-05-06T13:38:43.2Z" },
+    { url = "https://files.pythonhosted.org/packages/25/a4/73363fec545fd3ec025490bdda2743c56d0dd5b6266b1a53bbe9e4265375/pydantic_core-2.46.4-cp310-cp310-win32.whl", hash = "sha256:2f84c03c8607173d16b5a854ec68a2f9079ae03237a54fb506d13af47e1d018d", size = 1987085, upload-time = "2026-05-06T13:39:25.497Z" },
+    { url = "https://files.pythonhosted.org/packages/01/aa/62f082da2c91fac1c234bc9ee0066257ce83f0604abd72e4c9d5991f2d84/pydantic_core-2.46.4-cp310-cp310-win_amd64.whl", hash = "sha256:8358a950c8909158e3df31538a7e4edc2d7265a7c54b47f0864d9e5bae9dcebf", size = 2074311, upload-time = "2026-05-06T13:39:59.922Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/fa/6d7708d2cfc1a832acb6aeb0cd16e801902df8a0f583bb3b4b527fde022e/pydantic_core-2.46.4-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:0e96592440881c74a213e5ad528e2b24d3d4f940de2766bed9010ab1d9e51594", size = 2111872, upload-time = "2026-05-06T13:40:27.596Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/6f/aa064a3e74b5745afbdf250594f38e7ead05e2d651bcb35994b9417a0d4d/pydantic_core-2.46.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:e0d65b8c354be7fb5f720c3caa8bc940bc2d20ce749c8e06135f07f8ed95dd7c", size = 1948255, upload-time = "2026-05-06T13:39:12.574Z" },
+    { url = "https://files.pythonhosted.org/packages/43/3a/41114a9f7569b84b4d84e7a018c57c56347dac30c0d4a872946ec4e36c46/pydantic_core-2.46.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7bfb192b3f4b9e8a89b6277b6ce787564f62cfd272055f6e685726b111dc7826", size = 1972827, upload-time = "2026-05-06T13:38:19.841Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/25/1ab42e8048fe551934d9884e8d64daa7e990ad386f310a15981aeb6a5b08/pydantic_core-2.46.4-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:9037063db01f09b09e237c282b6792bd4da634b5402c4e7f0c61effed7701a04", size = 2041051, upload-time = "2026-05-06T13:38:10.447Z" },
+    { url = "https://files.pythonhosted.org/packages/94/c2/1a934597ddf08da410385b3b7aae91956a5a76c635effef456074fad7e88/pydantic_core-2.46.4-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:fc010ab034c8c7452522748bf937df58020d256ccae0874463d1f4d01758af8e", size = 2221314, upload-time = "2026-05-06T13:40:13.089Z" },
+    { url = "https://files.pythonhosted.org/packages/02/6d/9e8ad178c9c4df27ad3c8f25d1fe2a7ab0d2ba0559fad4aee5d3d1f16771/pydantic_core-2.46.4-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8c5dac79fa1614d1e06ca695109c6105923bd9c7d1d6c918d4e637b7e6b32fd3", size = 2285146, upload-time = "2026-05-06T13:38:59.224Z" },
+    { url = "https://files.pythonhosted.org/packages/80/50/540cd3aeefc041beb111125c4bff779831a2111fc6b15a9138cda277d32c/pydantic_core-2.46.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f9fa868638bf362d3d138ea55829cefb3d5f4b0d7f142234382a15e2485dbec4", size = 2089685, upload-time = "2026-05-06T13:38:17.762Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/a4/b440ad35f05f6a38f89fa0f149accb3f0e02be94ca5e15f3c449a61b4bc9/pydantic_core-2.46.4-cp311-cp311-manylinux_2_31_riscv64.whl", hash = "sha256:17299feefe090f2caa5b8e37222bb5f663e4935a8bfa6931d4102e5df1a9f398", size = 2115420, upload-time = "2026-05-06T13:37:58.195Z" },
+    { url = "https://files.pythonhosted.org/packages/99/61/de4f55db8dfd57bfdfa9a12ec90fe1b57c4f41062f7ca86f08586b3e0ac0/pydantic_core-2.46.4-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:4c63ebc82684aa89d9a3bcbd13d515b3be44250dc68dd3bd81526c1cb31286c3", size = 2165122, upload-time = "2026-05-06T13:37:01.167Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/52/7c529d7bdb2d1068bd52f51fe32572c8301f9a4febf1948f10639f1436f5/pydantic_core-2.46.4-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:aaa2a54443eff1950ba5ddc6b6ccda0d9c84a364276a62f969bdf2a390650848", size = 2182573, upload-time = "2026-05-06T13:38:45.04Z" },
+    { url = "https://files.pythonhosted.org/packages/37/b3/7c40325848ba78247f2812dcf9c7274e38cd801820ca6dd9fe63bcfb0eb4/pydantic_core-2.46.4-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:18e5ceec2ab67e6d5f1a9085e5a24c9c4e2ac4545730bfe668680bca05e555f3", size = 2317139, upload-time = "2026-05-06T13:37:15.539Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/37/f913f81a657c865b75da6c0dbed79876073c2a43b5bd9edbe8da785e4d49/pydantic_core-2.46.4-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:a0f62d0a58f4e7da165457e995725421e0064f2255d8eccebc49f41bbc23b109", size = 2360433, upload-time = "2026-05-06T13:37:30.099Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/67/6acaa1be2567f9256b056d8477158cac7240813956ce86e49deae8e173b4/pydantic_core-2.46.4-cp311-cp311-win32.whl", hash = "sha256:041bde0a48fd37cf71cab1c9d56d3e8625a3793fef1f7dd232b3ff37e978ecda", size = 1985513, upload-time = "2026-05-06T13:38:15.669Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/e6/c505f83dfeda9a2e5c995cfd872949e4d05e12f7feb3dca72f633daefa94/pydantic_core-2.46.4-cp311-cp311-win_amd64.whl", hash = "sha256:6f2eeda33a839975441c86a4119e1383c50b47faf0cbb5176985565c6bb02c33", size = 2071114, upload-time = "2026-05-06T13:40:35.416Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/da/7a263a96d965d9d0df5e8de8a475f33495451117035b09acb110288c381f/pydantic_core-2.46.4-cp311-cp311-win_arm64.whl", hash = "sha256:14f4c5d6db102bd796a627bbb3a17b4cf4574b9ae861d8b7c9a9661c6dd3362d", size = 2044298, upload-time = "2026-05-06T13:38:29.754Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/8c/af022f0af448d7747c5154288d46b5f2bc5f17366eaa0e23e9aa04d59f3b/pydantic_core-2.46.4-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:3245406455a5d98187ec35530fd772b1d799b26667980872c8d4614991e2c4a2", size = 2106158, upload-time = "2026-05-06T13:38:57.215Z" },
+    { url = "https://files.pythonhosted.org/packages/19/95/6195171e385007300f0f5574592e467c568becce2d937a0b6804f218bc49/pydantic_core-2.46.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:962ccbab7b642487b1d8b7df90ef677e03134cf1fd8880bf698649b22a69371f", size = 1951724, upload-time = "2026-05-06T13:37:02.697Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/bc/f47d1ff9cbb1620e1b5b697eef06010035735f07820180e74178226b27b3/pydantic_core-2.46.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8233f2947cf85404441fd7e0085f53b10c93e0ee78611099b5c7237e36aacbf7", size = 1975742, upload-time = "2026-05-06T13:37:09.448Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/11/9b9a5b0306345664a2da6410877af6e8082481b5884b3ddd78d47c6013ce/pydantic_core-2.46.4-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3a233125ac121aa3ffba9a2b59edfc4a985a76092dc8279586ab4b71390875e7", size = 2052418, upload-time = "2026-05-06T13:37:38.234Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/b7/a65fec226f5d78fc39f4a13c4cc0c768c22b113438f60c14adc9d2865038/pydantic_core-2.46.4-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5b712b53160b79a5850310b912a5ef8e57e56947c8ad690c227f5c9d7e561712", size = 2232274, upload-time = "2026-05-06T13:38:27.753Z" },
+    { url = "https://files.pythonhosted.org/packages/68/f0/92039db98b907ef49269a8271f67db9cb78ae2fc68062ef7e4e77adb5f61/pydantic_core-2.46.4-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9401557acd873c3a7f3eb9383edef8ac4968f9510e340f4808d427e75667e7b4", size = 2309940, upload-time = "2026-05-06T13:38:05.353Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/97/2aab507d3d00ca626e8e57c1eac6a79e4e5fbcc63eb99733ff55d1717f65/pydantic_core-2.46.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:926c9541b14b12b1681dca8a0b75feb510b06c6341b70a8e500c2fdcff837cce", size = 2094516, upload-time = "2026-05-06T13:39:10.577Z" },
+    { url = "https://files.pythonhosted.org/packages/22/37/a8aca44d40d737dde2bc05b3c6c07dff0de07ce6f82e9f3167aeaf4d5dea/pydantic_core-2.46.4-cp312-cp312-manylinux_2_31_riscv64.whl", hash = "sha256:56cb4851bcaf3d117eddcef4fe66afd750a50274b0da8e22be256d10e5611987", size = 2136854, upload-time = "2026-05-06T13:40:22.59Z" },
+    { url = "https://files.pythonhosted.org/packages/24/99/fcef1b79238c06a8cbec70819ac722ba76e02bc8ada9b0fd66eba40da01b/pydantic_core-2.46.4-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c68fcd102d71ea85c5b2dfac3f4f8476eff42a9e078fd5faefff6d145063536b", size = 2180306, upload-time = "2026-05-06T13:40:10.666Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/6c/fc44000918855b42779d007ae63b0532794739027b2f417321cddbc44f6a/pydantic_core-2.46.4-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:b2f69dec1725e79a012d920df1707de5caf7ed5e08f3be4435e25803efc47458", size = 2190044, upload-time = "2026-05-06T13:40:43.231Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/65/d9cadc9f1920d7a127ad2edba16c1db7916e59719285cd6c94600b0080ba/pydantic_core-2.46.4-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:8d0820e8192167f80d88d64038e609c31452eeca865b4e1d9950a27a4609b00b", size = 2329133, upload-time = "2026-05-06T13:39:57.365Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/cf/c873d91679f3a30bcf5e7ac280ce5573483e72295307685120d0d5ad3416/pydantic_core-2.46.4-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:fbdb89b3e1c94a30cc5edfce477c6e6a5dc4d8f84665b455c27582f211a1c72c", size = 2374464, upload-time = "2026-05-06T13:38:06.976Z" },
+    { url = "https://files.pythonhosted.org/packages/47/bd/6f2fc8188f31bf10590f1e98e7b306336161fac930a8c514cd7bd828c7dc/pydantic_core-2.46.4-cp312-cp312-win32.whl", hash = "sha256:9aa768456404a8bf48a4406685ac2bec8e72b62c69313734fa3b73cf33b3a894", size = 1974823, upload-time = "2026-05-06T13:40:47.985Z" },
+    { url = "https://files.pythonhosted.org/packages/40/8c/985c1d41ea1107c2534abd9870e4ed5c8e7669b5c308297835c001e7a1c4/pydantic_core-2.46.4-cp312-cp312-win_amd64.whl", hash = "sha256:e9c26f834c65f5752f3f06cb08cb86a913ceb7274d0db6e267808a708b46bc89", size = 2072919, upload-time = "2026-05-06T13:39:21.153Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/ba/f463d006e0c47373ca7ec5e1a261c59dc01ef4d62b2657af925fb0deee3a/pydantic_core-2.46.4-cp312-cp312-win_arm64.whl", hash = "sha256:4fc73cb559bdb54b1134a706a2802a4cddd27a0633f5abb7e53056268751ac6a", size = 2027604, upload-time = "2026-05-06T13:39:03.753Z" },
+    { url = "https://files.pythonhosted.org/packages/51/a2/5d30b469c5267a17b39dec53208222f76a8d351dfac4af661888c5aee77d/pydantic_core-2.46.4-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:5d5902252db0d3cedf8d4a1bc68f70eeb430f7e4c7104c8c476753519b423008", size = 2106306, upload-time = "2026-05-06T13:37:48.029Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/81/4fa520eaffa8bd7d1525e644cd6d39e7d60b1592bc5b516693c7340b50f1/pydantic_core-2.46.4-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:c94f0688e7b8d0a67abf40e57a7eaaecd17cc9586706a31b76c031f63df052b4", size = 1951906, upload-time = "2026-05-06T13:37:17.012Z" },
+    { url = "https://files.pythonhosted.org/packages/03/d5/fd02da45b659668b05923b17ba3a0100a0a3d5541e3bd8fcc4ecb711309e/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f027324c56cd5406ca49c124b0db10e56c69064fec039acc571c29020cc87c76", size = 1976802, upload-time = "2026-05-06T13:37:35.113Z" },
+    { url = "https://files.pythonhosted.org/packages/21/f2/95727e1368be3d3ed485eaab7adbd7dda408f33f7a36e8b48e0144002b91/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e739fee756ba1010f8bcccb534252e85a35fe45ae92c295a06059ce58b74ccd3", size = 2052446, upload-time = "2026-05-06T13:37:12.313Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/86/5d99feea3f77c7234b8718075b23db11532773c1a0dbd9b9490215dc2eeb/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9d56801be94b86a9da183e5f3766e6310752b99ff647e38b09a9500d88e46e76", size = 2232757, upload-time = "2026-05-06T13:39:01.149Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/3a/508ac615935ef7588cf6d9e9b91309fdc2da751af865e02a9098de88258c/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2412e734dcb48da14d4e4006b82b46b74f2518b8a26ee7e58c6844a6cd6d03c4", size = 2309275, upload-time = "2026-05-06T13:37:41.406Z" },
+    { url = "https://files.pythonhosted.org/packages/07/f8/41db9de19d7987d6b04715a02b3b40aea467000275d9d758ffaa31af7d50/pydantic_core-2.46.4-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9551187363ffc0de2a00b2e47c25aeaeb1020b69b668762966df15fc5659dd5a", size = 2094467, upload-time = "2026-05-06T13:39:18.847Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/e2/f35033184cb11d0052daf4416e8e10a502ea2ac006fc4f459aee872727d1/pydantic_core-2.46.4-cp313-cp313-manylinux_2_31_riscv64.whl", hash = "sha256:0186750b482eefa11d7f435892b09c5c606193ef3375bcf94aa00ae6bfb66262", size = 2134417, upload-time = "2026-05-06T13:40:17.944Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/7b/6ceeb1cc90e193862f444ebe373d8fdf613f0a82572dde03fb10734c6c71/pydantic_core-2.46.4-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:5855698a4856556d86e8e6cd8434bc3ac0314ee8e12089ae0e143f64c6256e4e", size = 2179782, upload-time = "2026-05-06T13:40:32.618Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/f2/c8d7773ede6af08036423a00ae0ceffce266c3c52a096c435d68c896083f/pydantic_core-2.46.4-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:cbaf13819775b7f769bf4a1f066cb6df7a28d4480081a589828ef190226881cd", size = 2188782, upload-time = "2026-05-06T13:36:51.018Z" },
+    { url = "https://files.pythonhosted.org/packages/59/31/0c864784e31f09f05cdd87606f08923b9c9e7f6e51dd27f20f62f975ce9f/pydantic_core-2.46.4-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:633147d34cf4550417f12e2b1a0383973bdf5cdfde212cb09e9a581cf10820be", size = 2328334, upload-time = "2026-05-06T13:40:37.764Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/eb/4f6c8a41efa30baa755590f4141abf3a8c370fab610915733e74134a7270/pydantic_core-2.46.4-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:82cf5301172168103724d49a1444d3378cb20cdee30b116a1bd6031236298a5d", size = 2372986, upload-time = "2026-05-06T13:39:34.152Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/24/b375a480d53113860c299764bfe9f349a3dc9108b3adc0d7f0d786492ebf/pydantic_core-2.46.4-cp313-cp313-win32.whl", hash = "sha256:9fa8ae11da9e2b3126c6426f147e0fba88d96d65921799bb30c6abd1cb2c97fb", size = 1973693, upload-time = "2026-05-06T13:37:55.072Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/e8/cff247591966f2d22ec8c003cd7587e27b7ba7b81ab2fb888e3ab75dc285/pydantic_core-2.46.4-cp313-cp313-win_amd64.whl", hash = "sha256:6b3ace8194b0e5204818c92802dcdca7fc6d88aabbb799d7c795540d9cd6d292", size = 2071819, upload-time = "2026-05-06T13:38:49.139Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/1a/f4aee670d5670e9e148e0c82c7db98d780be566c6e6a97ee8035528ca0b3/pydantic_core-2.46.4-cp313-cp313-win_arm64.whl", hash = "sha256:184c081504d17f1c1066e430e117142b2c77d9448a97f7b65c6ac9fd9aee238d", size = 2027411, upload-time = "2026-05-06T13:40:45.796Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/74/228a26ddad29c6672b805d9fd78e8d251cd04004fa7eed0e622096cd0250/pydantic_core-2.46.4-cp314-cp314-macosx_10_12_x86_64.whl", hash = "sha256:428e04521a40150c85216fc8b85e8d39fece235a9cf5e383761238c7fa9b96fb", size = 2102079, upload-time = "2026-05-06T13:38:41.019Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/1f/8970b150a4b4365623ae00fc88603491f763c627311ae8031e3111356d6e/pydantic_core-2.46.4-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:23ace664830ee0bfe014a0c7bc248b1f7f25ed7ad103852c317624a1083af462", size = 1952179, upload-time = "2026-05-06T13:36:59.812Z" },
+    { url = "https://files.pythonhosted.org/packages/95/30/5211a831ae054928054b2f79731661087a2bc5c01e825c672b3a4a8f1b3e/pydantic_core-2.46.4-cp314-cp314-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ce5c1d2a8b27468f433ca974829c44060b8097eedc39933e3c206a90ee49c4a9", size = 1978926, upload-time = "2026-05-06T13:37:39.933Z" },
+    { url = "https://files.pythonhosted.org/packages/57/e9/689668733b1eb67adeef047db3c2e8788fcf65a7fd9c9e2b46b7744fe245/pydantic_core-2.46.4-cp314-cp314-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7283d57845ecf5a163403eb0702dfc220cc4fbdd18919cb5ccea4f95ee1cdab4", size = 2046785, upload-time = "2026-05-06T13:38:01.995Z" },
+    { url = "https://files.pythonhosted.org/packages/60/d9/6715260422ff50a2109878fd24d948a6c3446bb2664f34ee78cd972b3acd/pydantic_core-2.46.4-cp314-cp314-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8daafc69c93ee8a0204506a3b6b30f586ef54028f52aeeeb5c4cfc5184fd5914", size = 2228733, upload-time = "2026-05-06T13:40:50.371Z" },
+    { url = "https://files.pythonhosted.org/packages/18/ae/fdb2f64316afca925640f8e70bb1a564b0ec2721c1389e25b8eb4bf9a299/pydantic_core-2.46.4-cp314-cp314-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cd2213145bcc2ba85884d0ac63d222fece9209678f77b9b4d76f054c561adb28", size = 2307534, upload-time = "2026-05-06T13:37:21.531Z" },
+    { url = "https://files.pythonhosted.org/packages/89/1d/8eff589b45bb8190a9d12c49cfad0f176a5cbd1534908a6b5125e2886239/pydantic_core-2.46.4-cp314-cp314-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a5f930472650a82629163023e630d160863fce524c616f4e5186e5de9d9a49b", size = 2099732, upload-time = "2026-05-06T13:39:31.942Z" },
+    { url = "https://files.pythonhosted.org/packages/06/d5/ee5a3366637fee41dee51a1fc91562dcf12ddbc68fda34e6b253da2324bb/pydantic_core-2.46.4-cp314-cp314-manylinux_2_31_riscv64.whl", hash = "sha256:c1b3f518abeca3aa13c712fd202306e145abf59a18b094a6bafb2d2bbf59192c", size = 2129627, upload-time = "2026-05-06T13:37:25.033Z" },
+    { url = "https://files.pythonhosted.org/packages/94/33/2414be571d2c6a6c4d08be21f9292b6d3fdb08949a97b6dfe985017821db/pydantic_core-2.46.4-cp314-cp314-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1a7dd0b3ee80d90150e3495a3a13ac34dbcbfd4f012996a6a1d8900e91b5c0fb", size = 2179141, upload-time = "2026-05-06T13:37:14.046Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/79/7daa95be995be0eecc4cf75064cb33f9bbbfe3fe0158caf2f0d4a996a5c7/pydantic_core-2.46.4-cp314-cp314-musllinux_1_1_aarch64.whl", hash = "sha256:3fb702cd90b0446a3a1c5e470bfa0dd23c0233b676a9099ddcc964fa6ca13898", size = 2184325, upload-time = "2026-05-06T13:36:53.615Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/cb/d0a382f5c0de8a222dc61c65348e0ce831b1f68e0a018450d31c2cace3a5/pydantic_core-2.46.4-cp314-cp314-musllinux_1_1_armv7l.whl", hash = "sha256:b8458003118a712e66286df6a707db01c52c0f52f7db8e4a38f0da1d3b94fc4e", size = 2323990, upload-time = "2026-05-06T13:40:29.971Z" },
+    { url = "https://files.pythonhosted.org/packages/05/db/d9ba624cc4a5aced1598e88c04fdbd8310c8a69b9d38b9a3d39ce3a61ed7/pydantic_core-2.46.4-cp314-cp314-musllinux_1_1_x86_64.whl", hash = "sha256:372429a130e469c9cd698925ce5fc50940b7a1336b0d82038e63d5bbc4edc519", size = 2369978, upload-time = "2026-05-06T13:37:23.027Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/20/d15df15ba918c423461905802bfd2981c3af0bfa0e40d05e13edbfa48bc3/pydantic_core-2.46.4-cp314-cp314-win32.whl", hash = "sha256:85bb3611ff1802f3ee7fdd7dbff26b56f343fb432d57a4728fdd49b6ef35e2f4", size = 1966354, upload-time = "2026-05-06T13:38:03.499Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/b6/6b8de4c0a7d7ab3004c439c80c5c1e0a3e8d78bbae19379b01960383d9e5/pydantic_core-2.46.4-cp314-cp314-win_amd64.whl", hash = "sha256:811ff8e9c313ab425368bcbb36e5c4ebd7108c2bbf4e4089cfbb0b01eff63fac", size = 2072238, upload-time = "2026-05-06T13:39:40.807Z" },
+    { url = "https://files.pythonhosted.org/packages/32/36/51eb763beec1f4cf59b1db243a7dcc39cbb41230f050a09b9d69faaf0a48/pydantic_core-2.46.4-cp314-cp314-win_arm64.whl", hash = "sha256:bfec22eab3c8cc2ceec0248aec886624116dc079afa027ecc8ad4a7e62010f8a", size = 2018251, upload-time = "2026-05-06T13:37:26.72Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/91/855af51d625b23aa987116a19e231d2aaef9c4a415273ddc189b79a45fee/pydantic_core-2.46.4-cp314-cp314t-macosx_10_12_x86_64.whl", hash = "sha256:af8244b2bef6aaad6d92cda81372de7f8c8d36c9f0c3ea36e827c60e7d9467a0", size = 2099593, upload-time = "2026-05-06T13:39:47.682Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/1b/8784a54c65edb5f49f0a14d6977cf1b209bba85a4c77445b255c2de58ab3/pydantic_core-2.46.4-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:5a4330cdbc57162e4b3aa303f588ba752257694c9c9be3e7ebb11b4aca659b5d", size = 1935226, upload-time = "2026-05-06T13:40:40.428Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/e7/1955d28d1afc56dd4b3ad7cc0cf39df1b9852964cf16e5d13912756d6d6b/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:29c61fc04a3d840155ff08e475a04809278972fe6aef51e2720554e96367e34b", size = 1974605, upload-time = "2026-05-06T13:37:32.029Z" },
+    { url = "https://files.pythonhosted.org/packages/93/e2/3fedbf0ba7a22850e6e9fd78117f1c0f10f950182344d8a6c535d468fdd8/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:c50f2528cf200c5eed56faf3f4e22fcd5f38c157a8b78576e6ba3168ec35f000", size = 2030777, upload-time = "2026-05-06T13:38:55.239Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/61/46be275fcaaba0b4f5b9669dd852267ce1ff616592dccf7a7845588df091/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0cbe8b01f948de4286c74cdd6c667aceb38f5c1e26f0693b3983d9d74887c65e", size = 2236641, upload-time = "2026-05-06T13:37:08.096Z" },
+    { url = "https://files.pythonhosted.org/packages/60/db/12e93e46a8bac9988be3c016860f83293daea8c716c029c9ace279036f2f/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:617d7e2ca7dcb8c5cf6bcb8c59b8832c94b36196bbf1cbd1bfb56ed341905edd", size = 2286404, upload-time = "2026-05-06T13:40:20.221Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/4a/4d8b19008f38d31c53b8219cfedc2e3d5de5fe99d90076b7e767de29274f/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7027560ee92211647d0d34e3f7cd6f50da56399d26a9c8ad0da286d3869a53f3", size = 2109219, upload-time = "2026-05-06T13:38:12.153Z" },
+    { url = "https://files.pythonhosted.org/packages/88/70/3cbc40978fefb7bb09c6708d40d4ad1a5d70fd7213c3d17f971de868ec1f/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_31_riscv64.whl", hash = "sha256:f99626688942fb746e545232e7726926f3be91b5975f8b55327665fafda991c7", size = 2110594, upload-time = "2026-05-06T13:40:02.971Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/20/b8d36736216e29491125531685b2f9e61aa5b4b2599893f8268551da3338/pydantic_core-2.46.4-cp314-cp314t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:fc3e9034a63de20e15e8ade85358bc6efc614008cab72898b4b4952bea0509ff", size = 2159542, upload-time = "2026-05-06T13:39:27.506Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/a2/367df868eb584dacf6bf82a389272406d7178e301c4ac82545ab98bc2dd9/pydantic_core-2.46.4-cp314-cp314t-musllinux_1_1_aarch64.whl", hash = "sha256:97e7cf2be5c77b7d1a9713a05605d49460d02c6078d38d8bef3cbe323c548424", size = 2168146, upload-time = "2026-05-06T13:38:31.93Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/b8/4460f77f7e201893f649a29ab355dddd3beee8a97bcb1a320db414f9a06e/pydantic_core-2.46.4-cp314-cp314t-musllinux_1_1_armv7l.whl", hash = "sha256:3bf92c5d0e00fefaab325a4d27828fe6b6e2a21848686b5b60d2d9eeb09d76c6", size = 2306309, upload-time = "2026-05-06T13:37:44.717Z" },
+    { url = "https://files.pythonhosted.org/packages/64/c4/be2639293acd87dc8ddbcec41a73cee9b2ebf996fe6d892a1a74e88ad3f7/pydantic_core-2.46.4-cp314-cp314t-musllinux_1_1_x86_64.whl", hash = "sha256:3ecbc122d18468d06ca279dc26a8c2e2d5acb10943bb35e36ae92096dc3b5565", size = 2369736, upload-time = "2026-05-06T13:37:05.645Z" },
+    { url = "https://files.pythonhosted.org/packages/30/a6/9f9f380dbb301f67023bf8f707aaa75daadf84f7152d95c410fd7e81d994/pydantic_core-2.46.4-cp314-cp314t-win32.whl", hash = "sha256:e846ae7835bf0703ae43f534ab79a867146dadd59dc9ca5c8b53d5c8f7c9ef02", size = 1955575, upload-time = "2026-05-06T13:38:51.116Z" },
+    { url = "https://files.pythonhosted.org/packages/40/1f/f1eb9eb350e795d1af8586289746f5c5677d16043040d63710e22abc43c9/pydantic_core-2.46.4-cp314-cp314t-win_amd64.whl", hash = "sha256:2108ba5c1c1eca18030634489dc544844144ee36357f2f9f780b93e7ddbb44b5", size = 2051624, upload-time = "2026-05-06T13:38:21.672Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/d2/42dd53d0a85c27606f316d3aa5d2869c4e8470a5ed6dec30e4a1abe19192/pydantic_core-2.46.4-cp314-cp314t-win_arm64.whl", hash = "sha256:4fcbe087dbc2068af7eda3aa87634eba216dbda64d1ae73c8684b621d33f6596", size = 2017325, upload-time = "2026-05-06T13:40:52.723Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/a4/73995fd4ebbb46ba0ee51e6fa049b8f02c40daebb762208feda8a6b7894d/pydantic_core-2.46.4-graalpy311-graalpy242_311_native-macosx_10_12_x86_64.whl", hash = "sha256:14d4edf427bdcf950a8a02d7cb44a08614388dd6e1bdcbf4f67504fa7887da9c", size = 2111589, upload-time = "2026-05-06T13:37:10.817Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/7f/f37d3a5e8bfcc2e403f5c57a730f2d815693fb42119e8ea48b3789335af1/pydantic_core-2.46.4-graalpy311-graalpy242_311_native-macosx_11_0_arm64.whl", hash = "sha256:0ce40cd7b21210e99342afafbd4d0f76d784eb5b1d60f3bdc566be4983c6c73b", size = 1944552, upload-time = "2026-05-06T13:36:56.717Z" },
+    { url = "https://files.pythonhosted.org/packages/15/3c/d7eb777b3ff43e8433a4efb39a17aa8fd98a4ee8561a24a67ef5db07b2d6/pydantic_core-2.46.4-graalpy311-graalpy242_311_native-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:90884113d8b48f760e9587002789ddd741e76ab9f89518cd1e43b1f1a52ec44b", size = 1982984, upload-time = "2026-05-06T13:39:06.207Z" },
+    { url = "https://files.pythonhosted.org/packages/63/87/70b9f40170a81afd55ca26c9b2acb25c20d64bcfbf888fafecb3ba077d4c/pydantic_core-2.46.4-graalpy311-graalpy242_311_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:66ce7632c22d837c95301830e111ad0128a32b8207533b60896a96c4915192ea", size = 2138417, upload-time = "2026-05-06T13:39:45.476Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/1d/8987ad40f65ae1432753072f214fb5c74fe47ffbd0698bb9cbbb585664f8/pydantic_core-2.46.4-graalpy312-graalpy250_312_native-macosx_10_12_x86_64.whl", hash = "sha256:1d8ba486450b14f3b1d63bc521d410ec7565e52f887b9fb671791886436a42f7", size = 2095527, upload-time = "2026-05-06T13:39:52.283Z" },
+    { url = "https://files.pythonhosted.org/packages/64/d3/84c282a7eee1d3ac4c0377546ef5a1ea436ce26840d9ac3b7ed54a377507/pydantic_core-2.46.4-graalpy312-graalpy250_312_native-macosx_11_0_arm64.whl", hash = "sha256:3009f12e4e90b7f88b4f9adb1b0c4a3d58fe7820f3238c190047209d148026df", size = 1936024, upload-time = "2026-05-06T13:40:15.671Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/ca/eac61596cdeb4d7e174d3dc0bd8a6238f14f75f97a24e7b7db4c7e7340a0/pydantic_core-2.46.4-graalpy312-graalpy250_312_native-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ad785e92e6dc634c21555edc8bd6b64957ab844541bcb96a1366c202951ae526", size = 1990696, upload-time = "2026-05-06T13:38:34.717Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/c3/7c8b240552251faf6b3a957db200fcfbbcec36763c050428b601e0c9b83b/pydantic_core-2.46.4-graalpy312-graalpy250_312_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:00c603d540afdd6b80eb39f078f33ebd46211f02f33e34a32d9f053bba711de0", size = 2147590, upload-time = "2026-05-06T13:39:29.883Z" },
+    { url = "https://files.pythonhosted.org/packages/11/cb/428de0385b6c8d44b716feba566abfacfbd23ee3c4439faa789a1456242f/pydantic_core-2.46.4-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:0c563b08bca408dc7f65f700633d8442fffb2421fc47b8101377e9fd65051ff0", size = 2112782, upload-time = "2026-05-06T13:37:04.016Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/b5/6a17bdadd0fc1f170adfd05a20d37c832f52b117b4d9131da1f41bb097ce/pydantic_core-2.46.4-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:db06ffe51636ffe9ca531fe9023dd64bdd794be8754cb5df57c5498ae5b518a7", size = 1952146, upload-time = "2026-05-06T13:39:43.092Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/dc/03734d80e362cd43ef65428e9de77c730ce7f2f11c60d2b1e1b39f0fbf99/pydantic_core-2.46.4-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:133878133d271ade3d41d1bfb2a45ec38dbdbda40bc065921c6b04e4630127e2", size = 2134492, upload-time = "2026-05-06T13:36:58.124Z" },
+    { url = "https://files.pythonhosted.org/packages/de/df/5e5ffc085ed07cc22d298134d3d911c63e91f6a0eb91fe646750a3209910/pydantic_core-2.46.4-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:9bc519fbf2b7578398853d815009ae5e4d4603d12f4e3f91da8c06852d3da3e9", size = 2156604, upload-time = "2026-05-06T13:37:49.88Z" },
+    { url = "https://files.pythonhosted.org/packages/81/44/6e112a4253e56f5705467cbab7ab5e91ee7398ba3d56d358635958893d3e/pydantic_core-2.46.4-pp311-pypy311_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:c7a7bd4e39e8e4c12c39cd480356842b6a8a06e41b23a55a5e3e191718838ddf", size = 2183828, upload-time = "2026-05-06T13:37:43.053Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/ad/5565071e937d8e752842ac241463944c9eb14c87e2d269f2658a5bd05e98/pydantic_core-2.46.4-pp311-pypy311_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:d396ec2b979760aaf3218e76c24e65bd0aca24983298653b3a9d7a45f9e47b30", size = 2310000, upload-time = "2026-05-06T13:37:56.694Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/c3/66883a5cec183e7fba4d024b4cbbe61851a63750ef606b0afecc46d1f2bf/pydantic_core-2.46.4-pp311-pypy311_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:86e1a4418c6cd97d60c95c71164158eaf7324fae7b0923264016baa993eba6fc", size = 2361286, upload-time = "2026-05-06T13:40:05.667Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/2d/69abac8f838090bbecd5df894befb2c2619e7996a98ddb949db9f3b93225/pydantic_core-2.46.4-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:d51026d73fcfd93610abc7b27789c26b313920fcfb20e27462d74a7f8b06e983", size = 2193071, upload-time = "2026-05-06T13:38:08.682Z" },
+]
+
+[[package]]
+name = "pytest"
+version = "7.4.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
+    { name = "iniconfig" },
+    { name = "packaging" },
+    { name = "pluggy" },
+    { name = "tomli", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/80/1f/9d8e98e4133ffb16c90f3b405c43e38d3abb715bb5d7a63a5a684f7e46a3/pytest-7.4.4.tar.gz", hash = "sha256:2cf0005922c6ace4a3e2ec8b4080eb0d9753fdc93107415332f50ce9e7994280", size = 1357116, upload-time = "2023-12-31T12:00:18.035Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/51/ff/f6e8b8f39e08547faece4bd80f89d5a8de68a38b2d179cc1c4490ffa3286/pytest-7.4.4-py3-none-any.whl", hash = "sha256:b090cdf5ed60bf4c45261be03239c2c1c22df034fbffe691abe93cd80cea01d8", size = 325287, upload-time = "2023-12-31T12:00:13.963Z" },
+]
+
+[[package]]
+name = "python-dateutil"
+version = "2.9.0.post0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "six" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/66/c0/0c8b6ad9f17a802ee498c46e004a0eb49bc148f2fd230864601a86dcf6db/python-dateutil-2.9.0.post0.tar.gz", hash = "sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3", size = 342432, upload-time = "2024-03-01T18:36:20.211Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892, upload-time = "2024-03-01T18:36:18.57Z" },
+]
+
+[[package]]
+name = "python-slugify"
+version = "8.0.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "text-unidecode" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/87/c7/5e1547c44e31da50a460df93af11a535ace568ef89d7a811069ead340c4a/python-slugify-8.0.4.tar.gz", hash = "sha256:59202371d1d05b54a9e7720c5e038f928f45daaffe41dd10822f3907b937c856", size = 10921, upload-time = "2024-02-08T18:32:45.488Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a4/62/02da182e544a51a5c3ccf4b03ab79df279f9c60c5e82d5e8bec7ca26ac11/python_slugify-8.0.4-py2.py3-none-any.whl", hash = "sha256:276540b79961052b66b7d116620b36518847f52d5fd9e3a70164fc8c50faa6b8", size = 10051, upload-time = "2024-02-08T18:32:43.911Z" },
+]
+
+[[package]]
+name = "pytimeparse"
+version = "1.1.8"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/37/5d/231f5f33c81e09682708fb323f9e4041408d8223e2f0fb9742843328778f/pytimeparse-1.1.8.tar.gz", hash = "sha256:e86136477be924d7e670646a98561957e8ca7308d44841e21f5ddea757556a0a", size = 9403, upload-time = "2018-05-18T17:40:42.76Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1b/b4/afd75551a3b910abd1d922dbd45e49e5deeb4d47dc50209ce489ba9844dd/pytimeparse-1.1.8-py2.py3-none-any.whl", hash = "sha256:04b7be6cc8bd9f5647a6325444926c3ac34ee6bc7e69da4367ba282f076036bd", size = 9969, upload-time = "2018-05-18T17:40:41.28Z" },
+]
+
+[[package]]
+name = "pytorch-ranger"
+version = "0.1.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "torch" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/bb/32/9269ee5981995e760c3bf51d6cf7f84a2ce051eca2315753910585bce50d/pytorch_ranger-0.1.1.tar.gz", hash = "sha256:aa7115431cef11b57d7dd7bc86e7302a911dae467f62ec5d0b10e1ff744875db", size = 7865, upload-time = "2020-03-30T07:37:22.194Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0d/70/12256257d861bbc3e176130d25be1de085ce7a9e60594064888a950f2154/pytorch_ranger-0.1.1-py3-none-any.whl", hash = "sha256:1e69156c9cc8439185cb8ba4725b18c91947fbe72743e25aca937da8aeb0c8ec", size = 14436, upload-time = "2020-03-30T07:37:21.198Z" },
+]
+
+[[package]]
+name = "pytz"
+version = "2026.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ff/46/dd499ec9038423421951e4fad73051febaa13d2df82b4064f87af8b8c0c3/pytz-2026.2.tar.gz", hash = "sha256:0e60b47b29f21574376f218fe21abc009894a2321ea16c6754f3cad6eb7cdd6a", size = 320861, upload-time = "2026-05-04T01:35:29.667Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ec/dd/96da98f892250475bdf2328112d7468abdd4acc7b902b6af23f4ed958ea0/pytz-2026.2-py2.py3-none-any.whl", hash = "sha256:04156e608bee23d3792fd45c94ae47fae1036688e75032eea2e3bf0323d1f126", size = 510141, upload-time = "2026-05-04T01:35:27.408Z" },
+]
+
+[[package]]
+name = "pyyaml"
+version = "6.0.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/05/8e/961c0007c59b8dd7729d542c61a4d537767a59645b82a0b521206e1e25c2/pyyaml-6.0.3.tar.gz", hash = "sha256:d76623373421df22fb4cf8817020cbb7ef15c725b9d5e45f17e189bfc384190f", size = 130960, upload-time = "2025-09-25T21:33:16.546Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f4/a0/39350dd17dd6d6c6507025c0e53aef67a9293a6d37d3511f23ea510d5800/pyyaml-6.0.3-cp310-cp310-macosx_10_13_x86_64.whl", hash = "sha256:214ed4befebe12df36bcc8bc2b64b396ca31be9304b8f59e25c11cf94a4c033b", size = 184227, upload-time = "2025-09-25T21:31:46.04Z" },
+    { url = "https://files.pythonhosted.org/packages/05/14/52d505b5c59ce73244f59c7a50ecf47093ce4765f116cdb98286a71eeca2/pyyaml-6.0.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:02ea2dfa234451bbb8772601d7b8e426c2bfa197136796224e50e35a78777956", size = 174019, upload-time = "2025-09-25T21:31:47.706Z" },
+    { url = "https://files.pythonhosted.org/packages/43/f7/0e6a5ae5599c838c696adb4e6330a59f463265bfa1e116cfd1fbb0abaaae/pyyaml-6.0.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b30236e45cf30d2b8e7b3e85881719e98507abed1011bf463a8fa23e9c3e98a8", size = 740646, upload-time = "2025-09-25T21:31:49.21Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/3a/61b9db1d28f00f8fd0ae760459a5c4bf1b941baf714e207b6eb0657d2578/pyyaml-6.0.3-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:66291b10affd76d76f54fad28e22e51719ef9ba22b29e1d7d03d6777a9174198", size = 840793, upload-time = "2025-09-25T21:31:50.735Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/1e/7acc4f0e74c4b3d9531e24739e0ab832a5edf40e64fbae1a9c01941cabd7/pyyaml-6.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9c7708761fccb9397fe64bbc0395abcae8c4bf7b0eac081e12b809bf47700d0b", size = 770293, upload-time = "2025-09-25T21:31:51.828Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/ef/abd085f06853af0cd59fa5f913d61a8eab65d7639ff2a658d18a25d6a89d/pyyaml-6.0.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:418cf3f2111bc80e0933b2cd8cd04f286338bb88bdc7bc8e6dd775ebde60b5e0", size = 732872, upload-time = "2025-09-25T21:31:53.282Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/15/2bc9c8faf6450a8b3c9fc5448ed869c599c0a74ba2669772b1f3a0040180/pyyaml-6.0.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5e0b74767e5f8c593e8c9b5912019159ed0533c70051e9cce3e8b6aa699fcd69", size = 758828, upload-time = "2025-09-25T21:31:54.807Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/00/531e92e88c00f4333ce359e50c19b8d1de9fe8d581b1534e35ccfbc5f393/pyyaml-6.0.3-cp310-cp310-win32.whl", hash = "sha256:28c8d926f98f432f88adc23edf2e6d4921ac26fb084b028c733d01868d19007e", size = 142415, upload-time = "2025-09-25T21:31:55.885Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/fa/926c003379b19fca39dd4634818b00dec6c62d87faf628d1394e137354d4/pyyaml-6.0.3-cp310-cp310-win_amd64.whl", hash = "sha256:bdb2c67c6c1390b63c6ff89f210c8fd09d9a1217a465701eac7316313c915e4c", size = 158561, upload-time = "2025-09-25T21:31:57.406Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/16/a95b6757765b7b031c9374925bb718d55e0a9ba8a1b6a12d25962ea44347/pyyaml-6.0.3-cp311-cp311-macosx_10_13_x86_64.whl", hash = "sha256:44edc647873928551a01e7a563d7452ccdebee747728c1080d881d68af7b997e", size = 185826, upload-time = "2025-09-25T21:31:58.655Z" },
+    { url = "https://files.pythonhosted.org/packages/16/19/13de8e4377ed53079ee996e1ab0a9c33ec2faf808a4647b7b4c0d46dd239/pyyaml-6.0.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:652cb6edd41e718550aad172851962662ff2681490a8a711af6a4d288dd96824", size = 175577, upload-time = "2025-09-25T21:32:00.088Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/62/d2eb46264d4b157dae1275b573017abec435397aa59cbcdab6fc978a8af4/pyyaml-6.0.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:10892704fc220243f5305762e276552a0395f7beb4dbf9b14ec8fd43b57f126c", size = 775556, upload-time = "2025-09-25T21:32:01.31Z" },
+    { url = "https://files.pythonhosted.org/packages/10/cb/16c3f2cf3266edd25aaa00d6c4350381c8b012ed6f5276675b9eba8d9ff4/pyyaml-6.0.3-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:850774a7879607d3a6f50d36d04f00ee69e7fc816450e5f7e58d7f17f1ae5c00", size = 882114, upload-time = "2025-09-25T21:32:03.376Z" },
+    { url = "https://files.pythonhosted.org/packages/71/60/917329f640924b18ff085ab889a11c763e0b573da888e8404ff486657602/pyyaml-6.0.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b8bb0864c5a28024fac8a632c443c87c5aa6f215c0b126c449ae1a150412f31d", size = 806638, upload-time = "2025-09-25T21:32:04.553Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/6f/529b0f316a9fd167281a6c3826b5583e6192dba792dd55e3203d3f8e655a/pyyaml-6.0.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:1d37d57ad971609cf3c53ba6a7e365e40660e3be0e5175fa9f2365a379d6095a", size = 767463, upload-time = "2025-09-25T21:32:06.152Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/6a/b627b4e0c1dd03718543519ffb2f1deea4a1e6d42fbab8021936a4d22589/pyyaml-6.0.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:37503bfbfc9d2c40b344d06b2199cf0e96e97957ab1c1b546fd4f87e53e5d3e4", size = 794986, upload-time = "2025-09-25T21:32:07.367Z" },
+    { url = "https://files.pythonhosted.org/packages/45/91/47a6e1c42d9ee337c4839208f30d9f09caa9f720ec7582917b264defc875/pyyaml-6.0.3-cp311-cp311-win32.whl", hash = "sha256:8098f252adfa6c80ab48096053f512f2321f0b998f98150cea9bd23d83e1467b", size = 142543, upload-time = "2025-09-25T21:32:08.95Z" },
+    { url = "https://files.pythonhosted.org/packages/da/e3/ea007450a105ae919a72393cb06f122f288ef60bba2dc64b26e2646fa315/pyyaml-6.0.3-cp311-cp311-win_amd64.whl", hash = "sha256:9f3bfb4965eb874431221a3ff3fdcddc7e74e3b07799e0e84ca4a0f867d449bf", size = 158763, upload-time = "2025-09-25T21:32:09.96Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/33/422b98d2195232ca1826284a76852ad5a86fe23e31b009c9886b2d0fb8b2/pyyaml-6.0.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7f047e29dcae44602496db43be01ad42fc6f1cc0d8cd6c83d342306c32270196", size = 182063, upload-time = "2025-09-25T21:32:11.445Z" },
+    { url = "https://files.pythonhosted.org/packages/89/a0/6cf41a19a1f2f3feab0e9c0b74134aa2ce6849093d5517a0c550fe37a648/pyyaml-6.0.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:fc09d0aa354569bc501d4e787133afc08552722d3ab34836a80547331bb5d4a0", size = 173973, upload-time = "2025-09-25T21:32:12.492Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/23/7a778b6bd0b9a8039df8b1b1d80e2e2ad78aa04171592c8a5c43a56a6af4/pyyaml-6.0.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9149cad251584d5fb4981be1ecde53a1ca46c891a79788c0df828d2f166bda28", size = 775116, upload-time = "2025-09-25T21:32:13.652Z" },
+    { url = "https://files.pythonhosted.org/packages/65/30/d7353c338e12baef4ecc1b09e877c1970bd3382789c159b4f89d6a70dc09/pyyaml-6.0.3-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:5fdec68f91a0c6739b380c83b951e2c72ac0197ace422360e6d5a959d8d97b2c", size = 844011, upload-time = "2025-09-25T21:32:15.21Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/9d/b3589d3877982d4f2329302ef98a8026e7f4443c765c46cfecc8858c6b4b/pyyaml-6.0.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ba1cc08a7ccde2d2ec775841541641e4548226580ab850948cbfda66a1befcdc", size = 807870, upload-time = "2025-09-25T21:32:16.431Z" },
+    { url = "https://files.pythonhosted.org/packages/05/c0/b3be26a015601b822b97d9149ff8cb5ead58c66f981e04fedf4e762f4bd4/pyyaml-6.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:8dc52c23056b9ddd46818a57b78404882310fb473d63f17b07d5c40421e47f8e", size = 761089, upload-time = "2025-09-25T21:32:17.56Z" },
+    { url = "https://files.pythonhosted.org/packages/be/8e/98435a21d1d4b46590d5459a22d88128103f8da4c2d4cb8f14f2a96504e1/pyyaml-6.0.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:41715c910c881bc081f1e8872880d3c650acf13dfa8214bad49ed4cede7c34ea", size = 790181, upload-time = "2025-09-25T21:32:18.834Z" },
+    { url = "https://files.pythonhosted.org/packages/74/93/7baea19427dcfbe1e5a372d81473250b379f04b1bd3c4c5ff825e2327202/pyyaml-6.0.3-cp312-cp312-win32.whl", hash = "sha256:96b533f0e99f6579b3d4d4995707cf36df9100d67e0c8303a0c55b27b5f99bc5", size = 137658, upload-time = "2025-09-25T21:32:20.209Z" },
+    { url = "https://files.pythonhosted.org/packages/86/bf/899e81e4cce32febab4fb42bb97dcdf66bc135272882d1987881a4b519e9/pyyaml-6.0.3-cp312-cp312-win_amd64.whl", hash = "sha256:5fcd34e47f6e0b794d17de1b4ff496c00986e1c83f7ab2fb8fcfe9616ff7477b", size = 154003, upload-time = "2025-09-25T21:32:21.167Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/08/67bd04656199bbb51dbed1439b7f27601dfb576fb864099c7ef0c3e55531/pyyaml-6.0.3-cp312-cp312-win_arm64.whl", hash = "sha256:64386e5e707d03a7e172c0701abfb7e10f0fb753ee1d773128192742712a98fd", size = 140344, upload-time = "2025-09-25T21:32:22.617Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/11/0fd08f8192109f7169db964b5707a2f1e8b745d4e239b784a5a1dd80d1db/pyyaml-6.0.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:8da9669d359f02c0b91ccc01cac4a67f16afec0dac22c2ad09f46bee0697eba8", size = 181669, upload-time = "2025-09-25T21:32:23.673Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/16/95309993f1d3748cd644e02e38b75d50cbc0d9561d21f390a76242ce073f/pyyaml-6.0.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:2283a07e2c21a2aa78d9c4442724ec1eb15f5e42a723b99cb3d822d48f5f7ad1", size = 173252, upload-time = "2025-09-25T21:32:25.149Z" },
+    { url = "https://files.pythonhosted.org/packages/50/31/b20f376d3f810b9b2371e72ef5adb33879b25edb7a6d072cb7ca0c486398/pyyaml-6.0.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ee2922902c45ae8ccada2c5b501ab86c36525b883eff4255313a253a3160861c", size = 767081, upload-time = "2025-09-25T21:32:26.575Z" },
+    { url = "https://files.pythonhosted.org/packages/49/1e/a55ca81e949270d5d4432fbbd19dfea5321eda7c41a849d443dc92fd1ff7/pyyaml-6.0.3-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a33284e20b78bd4a18c8c2282d549d10bc8408a2a7ff57653c0cf0b9be0afce5", size = 841159, upload-time = "2025-09-25T21:32:27.727Z" },
+    { url = "https://files.pythonhosted.org/packages/74/27/e5b8f34d02d9995b80abcef563ea1f8b56d20134d8f4e5e81733b1feceb2/pyyaml-6.0.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0f29edc409a6392443abf94b9cf89ce99889a1dd5376d94316ae5145dfedd5d6", size = 801626, upload-time = "2025-09-25T21:32:28.878Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/11/ba845c23988798f40e52ba45f34849aa8a1f2d4af4b798588010792ebad6/pyyaml-6.0.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f7057c9a337546edc7973c0d3ba84ddcdf0daa14533c2065749c9075001090e6", size = 753613, upload-time = "2025-09-25T21:32:30.178Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/e0/7966e1a7bfc0a45bf0a7fb6b98ea03fc9b8d84fa7f2229e9659680b69ee3/pyyaml-6.0.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:eda16858a3cab07b80edaf74336ece1f986ba330fdb8ee0d6c0d68fe82bc96be", size = 794115, upload-time = "2025-09-25T21:32:31.353Z" },
+    { url = "https://files.pythonhosted.org/packages/de/94/980b50a6531b3019e45ddeada0626d45fa85cbe22300844a7983285bed3b/pyyaml-6.0.3-cp313-cp313-win32.whl", hash = "sha256:d0eae10f8159e8fdad514efdc92d74fd8d682c933a6dd088030f3834bc8e6b26", size = 137427, upload-time = "2025-09-25T21:32:32.58Z" },
+    { url = "https://files.pythonhosted.org/packages/97/c9/39d5b874e8b28845e4ec2202b5da735d0199dbe5b8fb85f91398814a9a46/pyyaml-6.0.3-cp313-cp313-win_amd64.whl", hash = "sha256:79005a0d97d5ddabfeeea4cf676af11e647e41d81c9a7722a193022accdb6b7c", size = 154090, upload-time = "2025-09-25T21:32:33.659Z" },
+    { url = "https://files.pythonhosted.org/packages/73/e8/2bdf3ca2090f68bb3d75b44da7bbc71843b19c9f2b9cb9b0f4ab7a5a4329/pyyaml-6.0.3-cp313-cp313-win_arm64.whl", hash = "sha256:5498cd1645aa724a7c71c8f378eb29ebe23da2fc0d7a08071d89469bf1d2defb", size = 140246, upload-time = "2025-09-25T21:32:34.663Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/8c/f4bd7f6465179953d3ac9bc44ac1a8a3e6122cf8ada906b4f96c60172d43/pyyaml-6.0.3-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:8d1fab6bb153a416f9aeb4b8763bc0f22a5586065f86f7664fc23339fc1c1fac", size = 181814, upload-time = "2025-09-25T21:32:35.712Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/9c/4d95bb87eb2063d20db7b60faa3840c1b18025517ae857371c4dd55a6b3a/pyyaml-6.0.3-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:34d5fcd24b8445fadc33f9cf348c1047101756fd760b4dacb5c3e99755703310", size = 173809, upload-time = "2025-09-25T21:32:36.789Z" },
+    { url = "https://files.pythonhosted.org/packages/92/b5/47e807c2623074914e29dabd16cbbdd4bf5e9b2db9f8090fa64411fc5382/pyyaml-6.0.3-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:501a031947e3a9025ed4405a168e6ef5ae3126c59f90ce0cd6f2bfc477be31b7", size = 766454, upload-time = "2025-09-25T21:32:37.966Z" },
+    { url = "https://files.pythonhosted.org/packages/02/9e/e5e9b168be58564121efb3de6859c452fccde0ab093d8438905899a3a483/pyyaml-6.0.3-cp314-cp314-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:b3bc83488de33889877a0f2543ade9f70c67d66d9ebb4ac959502e12de895788", size = 836355, upload-time = "2025-09-25T21:32:39.178Z" },
+    { url = "https://files.pythonhosted.org/packages/88/f9/16491d7ed2a919954993e48aa941b200f38040928474c9e85ea9e64222c3/pyyaml-6.0.3-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c458b6d084f9b935061bc36216e8a69a7e293a2f1e68bf956dcd9e6cbcd143f5", size = 794175, upload-time = "2025-09-25T21:32:40.865Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/3f/5989debef34dc6397317802b527dbbafb2b4760878a53d4166579111411e/pyyaml-6.0.3-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:7c6610def4f163542a622a73fb39f534f8c101d690126992300bf3207eab9764", size = 755228, upload-time = "2025-09-25T21:32:42.084Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/ce/af88a49043cd2e265be63d083fc75b27b6ed062f5f9fd6cdc223ad62f03e/pyyaml-6.0.3-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:5190d403f121660ce8d1d2c1bb2ef1bd05b5f68533fc5c2ea899bd15f4399b35", size = 789194, upload-time = "2025-09-25T21:32:43.362Z" },
+    { url = "https://files.pythonhosted.org/packages/23/20/bb6982b26a40bb43951265ba29d4c246ef0ff59c9fdcdf0ed04e0687de4d/pyyaml-6.0.3-cp314-cp314-win_amd64.whl", hash = "sha256:4a2e8cebe2ff6ab7d1050ecd59c25d4c8bd7e6f400f5f82b96557ac0abafd0ac", size = 156429, upload-time = "2025-09-25T21:32:57.844Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/f4/a4541072bb9422c8a883ab55255f918fa378ecf083f5b85e87fc2b4eda1b/pyyaml-6.0.3-cp314-cp314-win_arm64.whl", hash = "sha256:93dda82c9c22deb0a405ea4dc5f2d0cda384168e466364dec6255b293923b2f3", size = 143912, upload-time = "2025-09-25T21:32:59.247Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/f9/07dd09ae774e4616edf6cda684ee78f97777bdd15847253637a6f052a62f/pyyaml-6.0.3-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:02893d100e99e03eda1c8fd5c441d8c60103fd175728e23e431db1b589cf5ab3", size = 189108, upload-time = "2025-09-25T21:32:44.377Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/78/8d08c9fb7ce09ad8c38ad533c1191cf27f7ae1effe5bb9400a46d9437fcf/pyyaml-6.0.3-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:c1ff362665ae507275af2853520967820d9124984e0f7466736aea23d8611fba", size = 183641, upload-time = "2025-09-25T21:32:45.407Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/5b/3babb19104a46945cf816d047db2788bcaf8c94527a805610b0289a01c6b/pyyaml-6.0.3-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6adc77889b628398debc7b65c073bcb99c4a0237b248cacaf3fe8a557563ef6c", size = 831901, upload-time = "2025-09-25T21:32:48.83Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/cc/dff0684d8dc44da4d22a13f35f073d558c268780ce3c6ba1b87055bb0b87/pyyaml-6.0.3-cp314-cp314t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a80cb027f6b349846a3bf6d73b5e95e782175e52f22108cfa17876aaeff93702", size = 861132, upload-time = "2025-09-25T21:32:50.149Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/5e/f77dc6b9036943e285ba76b49e118d9ea929885becb0a29ba8a7c75e29fe/pyyaml-6.0.3-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:00c4bdeba853cc34e7dd471f16b4114f4162dc03e6b7afcc2128711f0eca823c", size = 839261, upload-time = "2025-09-25T21:32:51.808Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/88/a9db1376aa2a228197c58b37302f284b5617f56a5d959fd1763fb1675ce6/pyyaml-6.0.3-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:66e1674c3ef6f541c35191caae2d429b967b99e02040f5ba928632d9a7f0f065", size = 805272, upload-time = "2025-09-25T21:32:52.941Z" },
+    { url = "https://files.pythonhosted.org/packages/da/92/1446574745d74df0c92e6aa4a7b0b3130706a4142b2d1a5869f2eaa423c6/pyyaml-6.0.3-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:16249ee61e95f858e83976573de0f5b2893b3677ba71c9dd36b9cf8be9ac6d65", size = 829923, upload-time = "2025-09-25T21:32:54.537Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/7a/1c7270340330e575b92f397352af856a8c06f230aa3e76f86b39d01b416a/pyyaml-6.0.3-cp314-cp314t-win_amd64.whl", hash = "sha256:4ad1906908f2f5ae4e5a8ddfce73c320c2a1429ec52eafd27138b7f1cbe341c9", size = 174062, upload-time = "2025-09-25T21:32:55.767Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/12/de94a39c2ef588c7e6455cfbe7343d3b2dc9d6b6b2f40c4c6565744c873d/pyyaml-6.0.3-cp314-cp314t-win_arm64.whl", hash = "sha256:ebc55a14a21cb14062aa4162f906cd962b28e2e9ea38f9b4391244cd8de4ae0b", size = 149341, upload-time = "2025-09-25T21:32:56.828Z" },
+]
+
+[[package]]
+name = "sequifier"
+version = "2.0.0.0"
+source = { editable = "." }
+dependencies = [
+    { name = "beartype" },
+    { name = "csvkit" },
+    { name = "fastparquet" },
+    { name = "loguru" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "onnx" },
+    { name = "onnxruntime", version = "1.24.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "onnxruntime", version = "1.26.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "onnxscript" },
+    { name = "optuna" },
+    { name = "plotly" },
+    { name = "polars" },
+    { name = "psutil" },
+    { name = "pyarrow" },
+    { name = "pydantic" },
+    { name = "pyyaml" },
+    { name = "torch" },
+    { name = "torch-optimizer" },
+]
+
+[package.optional-dependencies]
+test = [
+    { name = "plotly" },
+    { name = "pytest" },
+]
+
+[package.metadata]
+requires-dist = [
+    { name = "beartype", specifier = ">=0.18.5,<0.19.0" },
+    { name = "csvkit", specifier = ">=1.0,<2.0" },
+    { name = "fastparquet", specifier = ">=2024.2.0,<2025.0.0" },
+    { name = "loguru", specifier = ">=0.7.3,<1.0.0" },
+    { name = "numpy", specifier = ">=1.23" },
+    { name = "onnx", specifier = ">=1.15.0,<2.0.0" },
+    { name = "onnxruntime", specifier = ">=1.17" },
+    { name = "onnxscript", specifier = ">=0.5.4" },
+    { name = "optuna", specifier = ">=2.10.0" },
+    { name = "plotly", specifier = ">=4.0.0" },
+    { name = "plotly", marker = "extra == 'test'", specifier = ">=6.0,<7.0" },
+    { name = "polars", specifier = ">=1.0.0,<2.0.0" },
+    { name = "polars", specifier = ">=1.31.0,<2.0" },
+    { name = "psutil", specifier = ">=7.0.0,<8.0.0" },
+    { name = "pyarrow", specifier = ">=16.1" },
+    { name = "pydantic", specifier = ">=2.0,<3.0" },
+    { name = "pytest", marker = "extra == 'test'", specifier = ">=7.2,<8.0" },
+    { name = "pyyaml", specifier = ">=6.0,<7.0" },
+    { name = "torch", specifier = ">=2.4" },
+    { name = "torch-optimizer", specifier = ">=0.3.0,<0.4.0" },
+]
+provides-extras = ["test"]
+
+[[package]]
+name = "setuptools"
+version = "81.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0d/1c/73e719955c59b8e424d015ab450f51c0af856ae46ea2da83eba51cc88de1/setuptools-81.0.0.tar.gz", hash = "sha256:487b53915f52501f0a79ccfd0c02c165ffe06631443a886740b91af4b7a5845a", size = 1198299, upload-time = "2026-02-06T21:10:39.601Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e1/e3/c164c88b2e5ce7b24d667b9bd83589cf4f3520d97cad01534cd3c4f55fdb/setuptools-81.0.0-py3-none-any.whl", hash = "sha256:fdd925d5c5d9f62e4b74b30d6dd7828ce236fd6ed998a08d81de62ce5a6310d6", size = 1062021, upload-time = "2026-02-06T21:10:37.175Z" },
+]
+
+[[package]]
+name = "six"
+version = "1.17.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/94/e7/b2c673351809dca68a0e064b6af791aa332cf192da575fd474ed7d6f16a2/six-1.17.0.tar.gz", hash = "sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81", size = 34031, upload-time = "2024-12-04T17:35:28.174Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b7/ce/149a00dd41f10bc29e5921b496af8b574d8413afcd5e30dfa0ed46c2cc5e/six-1.17.0-py2.py3-none-any.whl", hash = "sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274", size = 11050, upload-time = "2024-12-04T17:35:26.475Z" },
+]
+
+[[package]]
+name = "sqlalchemy"
+version = "2.0.50"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "greenlet", marker = "platform_machine == 'AMD64' or platform_machine == 'WIN32' or platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'ppc64le' or platform_machine == 'win32' or platform_machine == 'x86_64'" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/57/da/6fbf010c8ebb347679d0d100b22fe9ba5e13fd04046c5df7280d2f0bf706/sqlalchemy-2.0.50.tar.gz", hash = "sha256:af5607d11ef90fd6a5c0549fe0045dce1663d427426bcfb506dcb5346a85a3b9", size = 9907424, upload-time = "2026-05-24T19:20:04.018Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/70/a9/812a775bd8c1af0966d660238d005baf25e9bced1f038c8e71f00aa637a7/sqlalchemy-2.0.50-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:7af6eeb84985bf840ba779018ff9424d61ff69b52e66b8789d3c8da7bf5341b2", size = 2161617, upload-time = "2026-05-24T20:00:00.761Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/74/5a6bc5496e9be8f740fbf80f9e6bd4ab965c8a80870eb07ab015e360957a/sqlalchemy-2.0.50-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0fe7822866f3a9fc5f3db21a290ce8961a53050115f05edf9402b6a5feb92a9f", size = 3244104, upload-time = "2026-05-24T20:07:38.158Z" },
+    { url = "https://files.pythonhosted.org/packages/81/55/b260d8df2adc9bb0bf294f67b5f802ff0d84d99442b536b9efd0ea72d447/sqlalchemy-2.0.50-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e8e1b0f6a4dcd9b4839e2320afb5df37a6981cbc20ff9c423ae11c5537bdbd21", size = 3243039, upload-time = "2026-05-24T20:14:23.765Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/6d/58714005cbf370f16c3f30d30324a43be10069efcfe764f7236a2e851947/sqlalchemy-2.0.50-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:e195687f1af431c9515416288373b323b6eb599f774409814e89e9d603a56e39", size = 3195017, upload-time = "2026-05-24T20:07:40.086Z" },
+    { url = "https://files.pythonhosted.org/packages/30/e8/67527fee039bd3e1a6ce3f03d2b62fd87ab9099c17052810d79496727b66/sqlalchemy-2.0.50-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:ea1a8a2db4b2217d456c8d7a873bfc605f06fe3584d315264ea18c2a17585d0b", size = 3215308, upload-time = "2026-05-24T20:14:26.034Z" },
+    { url = "https://files.pythonhosted.org/packages/94/b2/dd3155a6a6706cb89adecf5ee6e0512f7b0ee5cf3e6f4cde67d3c20ebfda/sqlalchemy-2.0.50-cp310-cp310-win32.whl", hash = "sha256:68b154b08088b4ec32bb4d2958bfbb50e57549f91a4cd3e7f928e3553ed69031", size = 2121637, upload-time = "2026-05-24T20:08:06.401Z" },
+    { url = "https://files.pythonhosted.org/packages/93/a1/a09c463ee3e7764b5ce5bd19a7f0b6eefbde62e637439ab58498cdbd6b47/sqlalchemy-2.0.50-cp310-cp310-win_amd64.whl", hash = "sha256:66e374271ecb7101273f57af1a62446a953d327eec4f8089147de57c591bbacc", size = 2144673, upload-time = "2026-05-24T20:08:07.936Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/5d/3172686af1770e4de2805f919a51441085f589ddadf3dd76ec582f84f497/sqlalchemy-2.0.50-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1aa6e403663a9c43c8fef7ce4bdb4cf48bcd8d352e91deda2a99f963270bd508", size = 2161366, upload-time = "2026-05-24T20:00:02.061Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/90/e98dedea3c3e663a17afcd003a34ba45efdac2cea3b6f2e4585e2b1e2537/sqlalchemy-2.0.50-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:51b637a84f9fa35ae1f9017e786cb142974a25305085e1b378b3647a67f65ad3", size = 3318926, upload-time = "2026-05-24T20:07:42.369Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/4f/501308c2babb62c11753ecb4ee88ba9eef019419a4d6cbf7cb13e2bad353/sqlalchemy-2.0.50-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2dab927761d9108550f0cf8e66ff21af56f907a0ce0a689793db615e2b55f62c", size = 3319199, upload-time = "2026-05-24T20:14:28.551Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/39/d88996c5e03ed6248c3a788d20f0b8d8b376b9f8a495e4bab9df7c72d2f8/sqlalchemy-2.0.50-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:545eae198d37bcf837a10ede3684e2af32458d6f35c597c35c2de7502dc38fc4", size = 3270301, upload-time = "2026-05-24T20:07:44.917Z" },
+    { url = "https://files.pythonhosted.org/packages/42/1b/1ae0e65161b51cc43e5ca75430ef79d80e23b5042d645586c2c342c3b92e/sqlalchemy-2.0.50-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0fec460e18cdbb4c7773531122ce9a27e96c6ca17af3933941d94da475ad2c86", size = 3293465, upload-time = "2026-05-24T20:14:30.501Z" },
+    { url = "https://files.pythonhosted.org/packages/83/29/17c0003f2c0dfa6d1b97672475707e3ec5980db09defd7fa20beb6833bbd/sqlalchemy-2.0.50-cp311-cp311-win32.whl", hash = "sha256:e6e814658818fd165e749e3d8490ef16cc7f379a118c37ada8b0589ffbaaac22", size = 2120694, upload-time = "2026-05-24T20:08:09.237Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/18/280d00654cc19d1fccf236fa5070f6dd04b84dde6f1b2e637bde0ff340a7/sqlalchemy-2.0.50-cp311-cp311-win_amd64.whl", hash = "sha256:1c5f858fe79c9f5d8fda065c06186356acb7f8df3cd52dbd5ee3f200e4b144f5", size = 2145315, upload-time = "2026-05-24T20:08:10.952Z" },
+    { url = "https://files.pythonhosted.org/packages/be/b0/a9d19b43f38f878b1278bca5b00b909f7540d41494396dd2561f9ad0956d/sqlalchemy-2.0.50-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:23ae23d8b9d344d30d0a92f06d45825024a5790f1c1dd4cf452636a50d3e58cb", size = 2159807, upload-time = "2026-05-24T19:27:53.086Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/2c/191dd58a248fd2cfd4780fa82c375c505e4ad98c8b522fa69ec492130d77/sqlalchemy-2.0.50-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:47b71b933e7b4ebad407c8fdfd70d2c4f08b78b3238bb30eebdd6eb32ca51b89", size = 3343358, upload-time = "2026-05-24T20:09:29.279Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/2b/514fce8a7df81cf5bad7ff7865de7ac0c5776a38cc043475c4703eb7fe8b/sqlalchemy-2.0.50-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:110fdac56ace278949f00de805edacbd6141e382d992f9ba28238b3a0827a600", size = 3357994, upload-time = "2026-05-24T20:17:13.495Z" },
+    { url = "https://files.pythonhosted.org/packages/35/a6/a0e283f5494f92b0d77e319ff77e437b1ffe4a051ba67c81d53234825475/sqlalchemy-2.0.50-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:0f5e4ac70e9e757f6b3e87c0491ff034442ecd8dfd36d041a50564c322dafc0e", size = 3289399, upload-time = "2026-05-24T20:09:32.239Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/96/1b07325ba71752d6a028b77d07bed1483ad545f794e8b1dc89b3ba3b3c68/sqlalchemy-2.0.50-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:724f3dcbe53dd0151e3cb5e7ec4ba4c620bede579caacd16275dc35ce06e8615", size = 3321216, upload-time = "2026-05-24T20:17:15.581Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/8e/bad6ed253e8a99edfc99af02f7173ec48a1d3ed1b9b35a1b8bc1700900cc/sqlalchemy-2.0.50-cp312-cp312-win32.whl", hash = "sha256:1208050441471d003b7c8cb4054fb084f185cf35ac3f0ea270803865bca9939a", size = 2119194, upload-time = "2026-05-24T19:50:04.943Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/2d/314a6690dda4b9cfc571eab1a63cf6fe6e1470aa3759ccda6aa016ee0f5a/sqlalchemy-2.0.50-cp312-cp312-win_amd64.whl", hash = "sha256:9d1af51558029a156a70986b7df88f042b3d158d7c8d8fb5072912d4b32d89c7", size = 2146186, upload-time = "2026-05-24T19:50:06.74Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/c4/c42356b527296e9862f67990efce31ef78b4cf69cd3f80873a528a060320/sqlalchemy-2.0.50-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:06a9210bdc5f4298cff0781087e2ff45683922252dacc452846373a58761f093", size = 2156697, upload-time = "2026-05-24T19:27:54.764Z" },
+    { url = "https://files.pythonhosted.org/packages/60/a1/b1a70e3c4365ac7fe9e347f3710f19b562c866fb96d45e3c891588789a7b/sqlalchemy-2.0.50-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b53784972ade4f8174b9aa661f31a06f8a936d2cfdd602913ff3c6dd40ae873", size = 3284260, upload-time = "2026-05-24T20:09:34.195Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/4a/f3ac3caa19f263d57b0a47f8c91bbf56583dc2d3fc63acfbf644abb24fe0/sqlalchemy-2.0.50-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:31648fa14460537e768a7303b078e4344d208e0d23e06867c1f376a227ed82db", size = 3302280, upload-time = "2026-05-24T20:17:17.825Z" },
+    { url = "https://files.pythonhosted.org/packages/66/55/ccada3e3d62254587819749a0bc69f41173eb48a6e385d10e66d32a9c88e/sqlalchemy-2.0.50-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:03f4323c980ad0e918cc9e5369b015f759f4e534db5bbaf4dc36832c10d05064", size = 3231580, upload-time = "2026-05-24T20:09:36.406Z" },
+    { url = "https://files.pythonhosted.org/packages/05/f6/6809349130a2de0e109e7f00fd7d431da9565b9b2868b32ee684754f672b/sqlalchemy-2.0.50-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:2b9dcc43afef8ac157cd92fce96985d6b8b0cfbd3df4d666f66b4d55a75d202f", size = 3269375, upload-time = "2026-05-24T20:17:20.34Z" },
+    { url = "https://files.pythonhosted.org/packages/48/84/278a811ef4e07be9c89dc5cdd7be833268509a66a68c4897cf585e67428f/sqlalchemy-2.0.50-cp313-cp313-win32.whl", hash = "sha256:60922d6599065ddca2c6f376b9aa2f41a6b85a271725e0909490bbc50b1998a5", size = 2117229, upload-time = "2026-05-24T19:50:08.215Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/1c/067cc6187ed32d2ec222fe6d2643acc1659a6d0659f8a7cbc5ad3ae83280/sqlalchemy-2.0.50-cp313-cp313-win_amd64.whl", hash = "sha256:287086e67275a212c4582d166a6fb03a65ccc5551d80866270ce0dd9f34eccd3", size = 2143126, upload-time = "2026-05-24T19:50:09.691Z" },
+    { url = "https://files.pythonhosted.org/packages/df/32/10ac51b4be7cdecd7e93d069251c86dfbf70b7adbd7c67b48ccea6c49e1c/sqlalchemy-2.0.50-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:c966932507a4d7d0a37314927dbfcd89720e3f37d2a1e3352e7ae7939fa8e8a0", size = 2158519, upload-time = "2026-05-24T19:27:56.472Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/76/e703d2f7681d7d66c4c891af3f07c7ccf4c76ad7f18351de035b5eda007a/sqlalchemy-2.0.50-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:faffef4bcc20a1892e65e155293d99d60855bbbc79250ab712819cfd56a8e6bb", size = 3282063, upload-time = "2026-05-24T20:09:38.57Z" },
+    { url = "https://files.pythonhosted.org/packages/31/26/ef168b184a25701f9995e8fb7e503fafd7a99c1c77cda1bc1a26ea2ed486/sqlalchemy-2.0.50-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6c206aec519a2e7bd08abbfb33436e325fd22c632d9c21a9047e376ce241646e", size = 3287069, upload-time = "2026-05-24T20:17:21.942Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/15/765acc2bc693bccc43ca4a95d5b69750da8aaf6db1b5c616536e087f8920/sqlalchemy-2.0.50-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:bef4ac756363227ef6402a75fee025a4bc690f92328e825868939b3b3a446a6d", size = 3230453, upload-time = "2026-05-24T20:09:40.398Z" },
+    { url = "https://files.pythonhosted.org/packages/63/61/08e03c3adbf5db0087a0b6816746fec8f3032fb2f7fc899a9bb9b2a48ce4/sqlalchemy-2.0.50-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:96fbee6b19c19cd1556c8bf9419447cf2ec149ffcab7ab64348c23e54ef8547f", size = 3252413, upload-time = "2026-05-24T20:17:24.067Z" },
+    { url = "https://files.pythonhosted.org/packages/03/0c/370a1f2db38436c615e10134c8a37de3688e74084792380695f3f5083860/sqlalchemy-2.0.50-cp314-cp314-win32.whl", hash = "sha256:8f00e3eb43ba30eb1b238ee03a8a62309486d1321eda3328bb611e0340033ad8", size = 2120063, upload-time = "2026-05-24T19:50:11.08Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/a0/fe92bb9817863bc13ba093bda931979a26cc2ca69f8e8f26d07add3d7c6f/sqlalchemy-2.0.50-cp314-cp314-win_amd64.whl", hash = "sha256:15708c613cd5005b7dffe1f66ee6a63ee8f5e46799f71c70ebad74178c676a39", size = 2145830, upload-time = "2026-05-24T19:50:12.452Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/ff/e5640a98a0b2f491eb8fde10fb6c773621a2e44340de231fafcc9370f4a9/sqlalchemy-2.0.50-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:3699dac4be410e97049a1658e9480da9cde956594aa0f3aebc60b88f21c5ba70", size = 2178435, upload-time = "2026-05-24T19:42:58.889Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/85/337116e186f1236375b5fb70c21cfac98e8e8ab0d3a47be838dc47a59e08/sqlalchemy-2.0.50-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f96233858e3df43932ac11589e22520da6e8aeb624b03fedfeebb0e8ea213086", size = 3566059, upload-time = "2026-05-24T20:01:20.848Z" },
+    { url = "https://files.pythonhosted.org/packages/96/34/bb0e190e161c3c2c24314a65add57218be14a4a9486886b7f5047c1ff7c8/sqlalchemy-2.0.50-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c4e70c46fad30c3bcc6a4708bc0130a3173e11a5b25f0ea4a9d8911b450f1f52", size = 3535366, upload-time = "2026-05-24T20:03:56.768Z" },
+    { url = "https://files.pythonhosted.org/packages/df/5a/a7f759f97e4fd499c5d4e4488c760d5a7fbecf3028b465a04274fcd52384/sqlalchemy-2.0.50-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:1918a3cf564d16d95bca7301005f41ab2ad50b07cd3b9da50d3ed986db148d6a", size = 3474879, upload-time = "2026-05-24T20:01:23.058Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/d9/2907ea38eb60687d297bf9c39e5ee58053c87b57fe8a9cae97090cecbf10/sqlalchemy-2.0.50-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:b00098cdbdbd38c7be3d568b0c9c3122b8c0ec62b911b57cd5e6e0254d60a76d", size = 3486117, upload-time = "2026-05-24T20:03:59.052Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/e3/5aa06f167559f8c0bdae487e297d23ba548150ab016a3418265d617a4985/sqlalchemy-2.0.50-cp314-cp314t-win32.whl", hash = "sha256:1fbd55a969d7ac44a98e3dec75016074f809fa08f871585ace58dde110d1bf3e", size = 2150823, upload-time = "2026-05-24T20:08:58.644Z" },
+    { url = "https://files.pythonhosted.org/packages/65/9b/112fb8f977582d7489d036e409e3723948bcf5320b3ac465f3c481bbe8f9/sqlalchemy-2.0.50-cp314-cp314t-win_amd64.whl", hash = "sha256:c5c3cdb753a9004183e1ccb634b41611654c989e61bc68617ce878e46d6f1e51", size = 2185794, upload-time = "2026-05-24T20:09:00.319Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/10/f7220e9b784d295d241c86ed99aeb537f92afcd469a64861f2717e9bb077/sqlalchemy-2.0.50-py3-none-any.whl", hash = "sha256:92064363517a3ff8212b5a93b8c62876579d8dfd1ca5b561335f30152d884fa9", size = 1943861, upload-time = "2026-05-24T19:59:01.119Z" },
+]
+
+[[package]]
+name = "sympy"
+version = "1.14.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "mpmath" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/83/d3/803453b36afefb7c2bb238361cd4ae6125a569b4db67cd9e79846ba2d68c/sympy-1.14.0.tar.gz", hash = "sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517", size = 7793921, upload-time = "2025-04-27T18:05:01.611Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl", hash = "sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5", size = 6299353, upload-time = "2025-04-27T18:04:59.103Z" },
+]
+
+[[package]]
+name = "text-unidecode"
+version = "1.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ab/e2/e9a00f0ccb71718418230718b3d900e71a5d16e701a3dae079a21e9cd8f8/text-unidecode-1.3.tar.gz", hash = "sha256:bad6603bb14d279193107714b288be206cac565dfa49aa5b105294dd5c4aab93", size = 76885, upload-time = "2019-08-30T21:36:45.405Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a6/a5/c0b6468d3824fe3fde30dbb5e1f687b291608f9473681bbf7dabbf5a87d7/text_unidecode-1.3-py2.py3-none-any.whl", hash = "sha256:1311f10e8b895935241623731c2ba64f4c455287888b18189350b67134a822e8", size = 78154, upload-time = "2019-08-30T21:37:03.543Z" },
+]
+
+[[package]]
+name = "tomli"
+version = "2.4.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/22/de/48c59722572767841493b26183a0d1cc411d54fd759c5607c4590b6563a6/tomli-2.4.1.tar.gz", hash = "sha256:7c7e1a961a0b2f2472c1ac5b69affa0ae1132c39adcb67aba98568702b9cc23f", size = 17543, upload-time = "2026-03-25T20:22:03.828Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f4/11/db3d5885d8528263d8adc260bb2d28ebf1270b96e98f0e0268d32b8d9900/tomli-2.4.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:f8f0fc26ec2cc2b965b7a3b87cd19c5c6b8c5e5f436b984e85f486d652285c30", size = 154704, upload-time = "2026-03-25T20:21:10.473Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/f7/675db52c7e46064a9aa928885a9b20f4124ecb9bc2e1ce74c9106648d202/tomli-2.4.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4ab97e64ccda8756376892c53a72bd1f964e519c77236368527f758fbc36a53a", size = 149454, upload-time = "2026-03-25T20:21:12.036Z" },
+    { url = "https://files.pythonhosted.org/packages/61/71/81c50943cf953efa35bce7646caab3cf457a7d8c030b27cfb40d7235f9ee/tomli-2.4.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:96481a5786729fd470164b47cdb3e0e58062a496f455ee41b4403be77cb5a076", size = 237561, upload-time = "2026-03-25T20:21:13.098Z" },
+    { url = "https://files.pythonhosted.org/packages/48/c1/f41d9cb618acccca7df82aaf682f9b49013c9397212cb9f53219e3abac37/tomli-2.4.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5a881ab208c0baf688221f8cecc5401bd291d67e38a1ac884d6736cbcd8247e9", size = 243824, upload-time = "2026-03-25T20:21:14.569Z" },
+    { url = "https://files.pythonhosted.org/packages/22/e4/5a816ecdd1f8ca51fb756ef684b90f2780afc52fc67f987e3c61d800a46d/tomli-2.4.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:47149d5bd38761ac8be13a84864bf0b7b70bc051806bc3669ab1cbc56216b23c", size = 242227, upload-time = "2026-03-25T20:21:15.712Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/49/2b2a0ef529aa6eec245d25f0c703e020a73955ad7edf73e7f54ddc608aa5/tomli-2.4.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ec9bfaf3ad2df51ace80688143a6a4ebc09a248f6ff781a9945e51937008fcbc", size = 247859, upload-time = "2026-03-25T20:21:17.001Z" },
+    { url = "https://files.pythonhosted.org/packages/83/bd/6c1a630eaca337e1e78c5903104f831bda934c426f9231429396ce3c3467/tomli-2.4.1-cp311-cp311-win32.whl", hash = "sha256:ff2983983d34813c1aeb0fa89091e76c3a22889ee83ab27c5eeb45100560c049", size = 97204, upload-time = "2026-03-25T20:21:18.079Z" },
+    { url = "https://files.pythonhosted.org/packages/42/59/71461df1a885647e10b6bb7802d0b8e66480c61f3f43079e0dcd315b3954/tomli-2.4.1-cp311-cp311-win_amd64.whl", hash = "sha256:5ee18d9ebdb417e384b58fe414e8d6af9f4e7a0ae761519fb50f721de398dd4e", size = 108084, upload-time = "2026-03-25T20:21:18.978Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/83/dceca96142499c069475b790e7913b1044c1a4337e700751f48ed723f883/tomli-2.4.1-cp311-cp311-win_arm64.whl", hash = "sha256:c2541745709bad0264b7d4705ad453b76ccd191e64aa6f0fc66b69a293a45ece", size = 95285, upload-time = "2026-03-25T20:21:20.309Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/ba/42f134a3fe2b370f555f44b1d72feebb94debcab01676bf918d0cb70e9aa/tomli-2.4.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c742f741d58a28940ce01d58f0ab2ea3ced8b12402f162f4d534dfe18ba1cd6a", size = 155924, upload-time = "2026-03-25T20:21:21.626Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/c7/62d7a17c26487ade21c5422b646110f2162f1fcc95980ef7f63e73c68f14/tomli-2.4.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:7f86fd587c4ed9dd76f318225e7d9b29cfc5a9d43de44e5754db8d1128487085", size = 150018, upload-time = "2026-03-25T20:21:23.002Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/05/79d13d7c15f13bdef410bdd49a6485b1c37d28968314eabee452c22a7fda/tomli-2.4.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ff18e6a727ee0ab0388507b89d1bc6a22b138d1e2fa56d1ad494586d61d2eae9", size = 244948, upload-time = "2026-03-25T20:21:24.04Z" },
+    { url = "https://files.pythonhosted.org/packages/10/90/d62ce007a1c80d0b2c93e02cab211224756240884751b94ca72df8a875ca/tomli-2.4.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:136443dbd7e1dee43c68ac2694fde36b2849865fa258d39bf822c10e8068eac5", size = 253341, upload-time = "2026-03-25T20:21:25.177Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/7e/caf6496d60152ad4ed09282c1885cca4eea150bfd007da84aea07bcc0a3e/tomli-2.4.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5e262d41726bc187e69af7825504c933b6794dc3fbd5945e41a79bb14c31f585", size = 248159, upload-time = "2026-03-25T20:21:26.364Z" },
+    { url = "https://files.pythonhosted.org/packages/99/e7/c6f69c3120de34bbd882c6fba7975f3d7a746e9218e56ab46a1bc4b42552/tomli-2.4.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:5cb41aa38891e073ee49d55fbc7839cfdb2bc0e600add13874d048c94aadddd1", size = 253290, upload-time = "2026-03-25T20:21:27.46Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/2f/4a3c322f22c5c66c4b836ec58211641a4067364f5dcdd7b974b4c5da300c/tomli-2.4.1-cp312-cp312-win32.whl", hash = "sha256:da25dc3563bff5965356133435b757a795a17b17d01dbc0f42fb32447ddfd917", size = 98141, upload-time = "2026-03-25T20:21:28.492Z" },
+    { url = "https://files.pythonhosted.org/packages/24/22/4daacd05391b92c55759d55eaee21e1dfaea86ce5c571f10083360adf534/tomli-2.4.1-cp312-cp312-win_amd64.whl", hash = "sha256:52c8ef851d9a240f11a88c003eacb03c31fc1c9c4ec64a99a0f922b93874fda9", size = 108847, upload-time = "2026-03-25T20:21:29.386Z" },
+    { url = "https://files.pythonhosted.org/packages/68/fd/70e768887666ddd9e9f5d85129e84910f2db2796f9096aa02b721a53098d/tomli-2.4.1-cp312-cp312-win_arm64.whl", hash = "sha256:f758f1b9299d059cc3f6546ae2af89670cb1c4d48ea29c3cacc4fe7de3058257", size = 95088, upload-time = "2026-03-25T20:21:30.677Z" },
+    { url = "https://files.pythonhosted.org/packages/07/06/b823a7e818c756d9a7123ba2cda7d07bc2dd32835648d1a7b7b7a05d848d/tomli-2.4.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:36d2bd2ad5fb9eaddba5226aa02c8ec3fa4f192631e347b3ed28186d43be6b54", size = 155866, upload-time = "2026-03-25T20:21:31.65Z" },
+    { url = "https://files.pythonhosted.org/packages/14/6f/12645cf7f08e1a20c7eb8c297c6f11d31c1b50f316a7e7e1e1de6e2e7b7e/tomli-2.4.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:eb0dc4e38e6a1fd579e5d50369aa2e10acfc9cace504579b2faabb478e76941a", size = 149887, upload-time = "2026-03-25T20:21:33.028Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/e0/90637574e5e7212c09099c67ad349b04ec4d6020324539297b634a0192b0/tomli-2.4.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c7f2c7f2b9ca6bdeef8f0fa897f8e05085923eb091721675170254cbc5b02897", size = 243704, upload-time = "2026-03-25T20:21:34.51Z" },
+    { url = "https://files.pythonhosted.org/packages/10/8f/d3ddb16c5a4befdf31a23307f72828686ab2096f068eaf56631e136c1fdd/tomli-2.4.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f3c6818a1a86dd6dca7ddcaaf76947d5ba31aecc28cb1b67009a5877c9a64f3f", size = 251628, upload-time = "2026-03-25T20:21:36.012Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/f1/dbeeb9116715abee2485bf0a12d07a8f31af94d71608c171c45f64c0469d/tomli-2.4.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:d312ef37c91508b0ab2cee7da26ec0b3ed2f03ce12bd87a588d771ae15dcf82d", size = 247180, upload-time = "2026-03-25T20:21:37.136Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/74/16336ffd19ed4da28a70959f92f506233bd7cfc2332b20bdb01591e8b1d1/tomli-2.4.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:51529d40e3ca50046d7606fa99ce3956a617f9b36380da3b7f0dd3dd28e68cb5", size = 251674, upload-time = "2026-03-25T20:21:38.298Z" },
+    { url = "https://files.pythonhosted.org/packages/16/f9/229fa3434c590ddf6c0aa9af64d3af4b752540686cace29e6281e3458469/tomli-2.4.1-cp313-cp313-win32.whl", hash = "sha256:2190f2e9dd7508d2a90ded5ed369255980a1bcdd58e52f7fe24b8162bf9fedbd", size = 97976, upload-time = "2026-03-25T20:21:39.316Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/1e/71dfd96bcc1c775420cb8befe7a9d35f2e5b1309798f009dca17b7708c1e/tomli-2.4.1-cp313-cp313-win_amd64.whl", hash = "sha256:8d65a2fbf9d2f8352685bc1364177ee3923d6baf5e7f43ea4959d7d8bc326a36", size = 108755, upload-time = "2026-03-25T20:21:40.248Z" },
+    { url = "https://files.pythonhosted.org/packages/83/7a/d34f422a021d62420b78f5c538e5b102f62bea616d1d75a13f0a88acb04a/tomli-2.4.1-cp313-cp313-win_arm64.whl", hash = "sha256:4b605484e43cdc43f0954ddae319fb75f04cc10dd80d830540060ee7cd0243cd", size = 95265, upload-time = "2026-03-25T20:21:41.219Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/fb/9a5c8d27dbab540869f7c1f8eb0abb3244189ce780ba9cd73f3770662072/tomli-2.4.1-cp314-cp314-macosx_10_15_x86_64.whl", hash = "sha256:fd0409a3653af6c147209d267a0e4243f0ae46b011aa978b1080359fddc9b6cf", size = 155726, upload-time = "2026-03-25T20:21:42.23Z" },
+    { url = "https://files.pythonhosted.org/packages/62/05/d2f816630cc771ad836af54f5001f47a6f611d2d39535364f148b6a92d6b/tomli-2.4.1-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:a120733b01c45e9a0c34aeef92bf0cf1d56cfe81ed9d47d562f9ed591a9828ac", size = 149859, upload-time = "2026-03-25T20:21:43.386Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/48/66341bdb858ad9bd0ceab5a86f90eddab127cf8b046418009f2125630ecb/tomli-2.4.1-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:559db847dc486944896521f68d8190be1c9e719fced785720d2216fe7022b662", size = 244713, upload-time = "2026-03-25T20:21:44.474Z" },
+    { url = "https://files.pythonhosted.org/packages/df/6d/c5fad00d82b3c7a3ab6189bd4b10e60466f22cfe8a08a9394185c8a8111c/tomli-2.4.1-cp314-cp314-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:01f520d4f53ef97964a240a035ec2a869fe1a37dde002b57ebc4417a27ccd853", size = 252084, upload-time = "2026-03-25T20:21:45.62Z" },
+    { url = "https://files.pythonhosted.org/packages/00/71/3a69e86f3eafe8c7a59d008d245888051005bd657760e96d5fbfb0b740c2/tomli-2.4.1-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:7f94b27a62cfad8496c8d2513e1a222dd446f095fca8987fceef261225538a15", size = 247973, upload-time = "2026-03-25T20:21:46.937Z" },
+    { url = "https://files.pythonhosted.org/packages/67/50/361e986652847fec4bd5e4a0208752fbe64689c603c7ae5ea7cb16b1c0ca/tomli-2.4.1-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:ede3e6487c5ef5d28634ba3f31f989030ad6af71edfb0055cbbd14189ff240ba", size = 256223, upload-time = "2026-03-25T20:21:48.467Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/9a/b4173689a9203472e5467217e0154b00e260621caa227b6fa01feab16998/tomli-2.4.1-cp314-cp314-win32.whl", hash = "sha256:3d48a93ee1c9b79c04bb38772ee1b64dcf18ff43085896ea460ca8dec96f35f6", size = 98973, upload-time = "2026-03-25T20:21:49.526Z" },
+    { url = "https://files.pythonhosted.org/packages/14/58/640ac93bf230cd27d002462c9af0d837779f8773bc03dee06b5835208214/tomli-2.4.1-cp314-cp314-win_amd64.whl", hash = "sha256:88dceee75c2c63af144e456745e10101eb67361050196b0b6af5d717254dddf7", size = 109082, upload-time = "2026-03-25T20:21:50.506Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/2f/702d5e05b227401c1068f0d386d79a589bb12bf64c3d2c72ce0631e3bc49/tomli-2.4.1-cp314-cp314-win_arm64.whl", hash = "sha256:b8c198f8c1805dc42708689ed6864951fd2494f924149d3e4bce7710f8eb5232", size = 96490, upload-time = "2026-03-25T20:21:51.474Z" },
+    { url = "https://files.pythonhosted.org/packages/45/4b/b877b05c8ba62927d9865dd980e34a755de541eb65fffba52b4cc495d4d2/tomli-2.4.1-cp314-cp314t-macosx_10_15_x86_64.whl", hash = "sha256:d4d8fe59808a54658fcc0160ecfb1b30f9089906c50b23bcb4c69eddc19ec2b4", size = 164263, upload-time = "2026-03-25T20:21:52.543Z" },
+    { url = "https://files.pythonhosted.org/packages/24/79/6ab420d37a270b89f7195dec5448f79400d9e9c1826df982f3f8e97b24fd/tomli-2.4.1-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:7008df2e7655c495dd12d2a4ad038ff878d4ca4b81fccaf82b714e07eae4402c", size = 160736, upload-time = "2026-03-25T20:21:53.674Z" },
+    { url = "https://files.pythonhosted.org/packages/02/e0/3630057d8eb170310785723ed5adcdfb7d50cb7e6455f85ba8a3deed642b/tomli-2.4.1-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1d8591993e228b0c930c4bb0db464bdad97b3289fb981255d6c9a41aedc84b2d", size = 270717, upload-time = "2026-03-25T20:21:55.129Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/b4/1613716072e544d1a7891f548d8f9ec6ce2faf42ca65acae01d76ea06bb0/tomli-2.4.1-cp314-cp314t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:734e20b57ba95624ecf1841e72b53f6e186355e216e5412de414e3c51e5e3c41", size = 278461, upload-time = "2026-03-25T20:21:56.228Z" },
+    { url = "https://files.pythonhosted.org/packages/05/38/30f541baf6a3f6df77b3df16b01ba319221389e2da59427e221ef417ac0c/tomli-2.4.1-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:8a650c2dbafa08d42e51ba0b62740dae4ecb9338eefa093aa5c78ceb546fcd5c", size = 274855, upload-time = "2026-03-25T20:21:57.653Z" },
+    { url = "https://files.pythonhosted.org/packages/77/a3/ec9dd4fd2c38e98de34223b995a3b34813e6bdadf86c75314c928350ed14/tomli-2.4.1-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:504aa796fe0569bb43171066009ead363de03675276d2d121ac1a4572397870f", size = 283144, upload-time = "2026-03-25T20:21:59.089Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/be/605a6261cac79fba2ec0c9827e986e00323a1945700969b8ee0b30d85453/tomli-2.4.1-cp314-cp314t-win32.whl", hash = "sha256:b1d22e6e9387bf4739fbe23bfa80e93f6b0373a7f1b96c6227c32bef95a4d7a8", size = 108683, upload-time = "2026-03-25T20:22:00.214Z" },
+    { url = "https://files.pythonhosted.org/packages/12/64/da524626d3b9cc40c168a13da8335fe1c51be12c0a63685cc6db7308daae/tomli-2.4.1-cp314-cp314t-win_amd64.whl", hash = "sha256:2c1c351919aca02858f740c6d33adea0c5deea37f9ecca1cc1ef9e884a619d26", size = 121196, upload-time = "2026-03-25T20:22:01.169Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/cd/e80b62269fc78fc36c9af5a6b89c835baa8af28ff5ad28c7028d60860320/tomli-2.4.1-cp314-cp314t-win_arm64.whl", hash = "sha256:eab21f45c7f66c13f2a9e0e1535309cee140182a9cdae1e041d02e47291e8396", size = 100393, upload-time = "2026-03-25T20:22:02.137Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/61/cceae43728b7de99d9b847560c262873a1f6c98202171fd5ed62640b494b/tomli-2.4.1-py3-none-any.whl", hash = "sha256:0d85819802132122da43cb86656f8d1f8c6587d54ae7dcaf30e90533028b49fe", size = 14583, upload-time = "2026-03-25T20:22:03.012Z" },
+]
+
+[[package]]
+name = "torch"
+version = "2.12.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cuda-bindings", marker = "sys_platform == 'linux'" },
+    { name = "cuda-toolkit", extra = ["cudart", "cufft", "cufile", "cupti", "curand", "cusolver", "cusparse", "nvjitlink", "nvrtc", "nvtx"], marker = "sys_platform == 'linux'" },
+    { name = "filelock" },
+    { name = "fsspec" },
+    { name = "jinja2" },
+    { name = "networkx", version = "3.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "networkx", version = "3.6.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "nvidia-cublas", marker = "sys_platform == 'linux'" },
+    { name = "nvidia-cudnn-cu13", marker = "sys_platform == 'linux'" },
+    { name = "nvidia-cusparselt-cu13", marker = "sys_platform == 'linux'" },
+    { name = "nvidia-nccl-cu13", marker = "sys_platform == 'linux'" },
+    { name = "nvidia-nvshmem-cu13", marker = "sys_platform == 'linux'" },
+    { name = "setuptools" },
+    { name = "sympy" },
+    { name = "triton", marker = "sys_platform == 'linux'" },
+    { name = "typing-extensions" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c2/b7/53fe0436586716ab7aecff41e26b9302d57c85ded481fd83a2cd741e6b4e/torch-2.12.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:1834bd984f8a2f4f16bdfbeecca9146184b220aa46276bf5756735b5dae12812", size = 87981887, upload-time = "2026-05-13T14:55:53.234Z" },
+    { url = "https://files.pythonhosted.org/packages/34/60/d930eac44c30de06ed16f6d1ba4e785e1632532b50d8f0bf9bf699a4d0c7/torch-2.12.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:d4d029801cb7b6df858804a2a21b00cc2aa0bf0ee5d2ab18d343c9e9e5681f35", size = 426355000, upload-time = "2026-05-13T14:54:31.944Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/0c/c76b6a087820bab55705b94dfc074e520de9ae91f5ef90da2ecbf2a3ef12/torch-2.12.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:d47e7dee68ac4cd7a068b26bcd6b989935427709fae1c8f7bd0019978f829e15", size = 532144998, upload-time = "2026-05-13T14:56:05.523Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/64/8a0d036e166a6aa85ee09bef072f3655d1ba5d5486a68d1b03b6813c01b3/torch-2.12.0-cp310-cp310-win_amd64.whl", hash = "sha256:cf9839790285dd472e7a16aafcb4a4e6bf58ec1b494045044b0eefb0eb4bd1f2", size = 122949877, upload-time = "2026-05-13T14:55:46.841Z" },
+    { url = "https://files.pythonhosted.org/packages/18/62/131124fb95df03811b8260d1d43dcc5ee85ea1a344b964613d7efe77fb08/torch-2.12.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:10802fd383bbfed646212e765a72c37d2185205d4f26eb197a254e8ac7ddcb25", size = 87990344, upload-time = "2026-05-13T14:55:42.154Z" },
+    { url = "https://files.pythonhosted.org/packages/12/9c/dda0dbd547dc549839824135f223792fd0e725f28ed0715dda366b7acaa2/torch-2.12.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:c12592630aef72feaf18bd3f197ef587bbfa21131b31c38b23ab2e55fce92e36", size = 426362932, upload-time = "2026-05-13T14:54:15.295Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/d2/a7dd5a3f9bdaa7842124e8e2359202b317c48d47d2fc5816fafdf2049adb/torch-2.12.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:415c1b8d0412f67551c8e89a2daca0fb3e56694af0281ba155eaa9da481f58b4", size = 532170085, upload-time = "2026-05-13T14:55:20.788Z" },
+    { url = "https://files.pythonhosted.org/packages/12/1b/a61ce2004f9ab0ea8964a6e6168133a127795667639e2ff4f8f2bdb16a65/torch-2.12.0-cp311-cp311-win_amd64.whl", hash = "sha256:dd37188ea325042cb1f6cafa56822b11ada2520c04791a52629b0af25bdfbfd9", size = 122953128, upload-time = "2026-05-13T14:54:52.744Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/bb/285d643f254731294c9b595a007eac39db4600a98682d7bca688f42ca164/torch-2.12.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:b41339df93d491435e790ff8bcbae1c0ce777175889bfd1281d119862793e6a2", size = 88010197, upload-time = "2026-05-13T14:55:35.414Z" },
+    { url = "https://files.pythonhosted.org/packages/79/81/76debf1db1343bd929bbb5d74c89fb437c2ed88eb144712557e7bd3eea45/torch-2.12.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:8fbef9f108a863e7722a73740998967e3b074742a834fc5be3a535a2befa7057", size = 426376751, upload-time = "2026-05-13T14:55:03.353Z" },
+    { url = "https://files.pythonhosted.org/packages/de/f0/80026028b603c4650ff270fc3785bdef4bd6738765a9cc5a0f5a637d65a2/torch-2.12.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:4b4f64c2c2b11f7510d93dd6412b87025ff6eddd6bb61c3b5a3d892ea20c4756", size = 532261691, upload-time = "2026-05-13T14:52:54.453Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/c2/64b06cbb7830fb3cd9be13e1158b31a3f36b68e6a209105ee3c9d9480be0/torch-2.12.0-cp312-cp312-win_amd64.whl", hash = "sha256:8b958caff4a14d3a3b0b2dfc6a378f64dda9728a9dad28c08a0db9ce4dafb549", size = 122988114, upload-time = "2026-05-13T14:54:42.153Z" },
+    { url = "https://files.pythonhosted.org/packages/86/ca/01896c80ba921676aa45886b2c5b8d774912de2a1f719de48169c6f755cd/torch-2.12.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:90dd587a5f61bfe1307148b581e2084fc5bc4a06e2b90a20e9a36b81087ff16b", size = 88009511, upload-time = "2026-05-13T14:54:47.411Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/04/52bdaf4787eab6ac7d7f5851dff934e4def0bc8ead9c8fd2b69b3e529699/torch-2.12.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:864392c73b7654f4d2b3ae712f607937d0dbb1101c4555fbb41848106b297f39", size = 426383231, upload-time = "2026-05-13T14:53:32.129Z" },
+    { url = "https://files.pythonhosted.org/packages/49/8a/94bdecd13f5aaa90d45920b89789d9fe7c6f4af8c3cdd7ce01fcb59908fc/torch-2.12.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:5d6b560dfa7d56291c07d615c3bb73e8d9943d9b6d87f76cd0d9d570c4797fa6", size = 532269288, upload-time = "2026-05-13T14:53:49.423Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/2f/bdbaaa267de519ef1b73054bf590d8c93c37a266c9a4e24a01bd38b6918f/torch-2.12.0-cp313-cp313-win_amd64.whl", hash = "sha256:3fee918902090ade827643e758e98363278815de583c75d111fdd665ebffde9f", size = 122987706, upload-time = "2026-05-13T14:54:00.335Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/ad/e95e822f3538171e22640a7fbe839a1fdb666600bf6487025de2ff03b11a/torch-2.12.0-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:10ee1448a9f304d3b987eb4656f664ba6e4d7b410ca7a5a7c642199777a2cf88", size = 88319556, upload-time = "2026-05-13T14:54:05.574Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/07/055d06d985b445d67422d25b033c11cf55bbb81785d4c4e68e28bca5820e/torch-2.12.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:af68dbf403439cae9ceaeaaf92f8352b460787dcd27b92aa05c40dd4a19c0f1e", size = 426397656, upload-time = "2026-05-13T14:52:38.84Z" },
+    { url = "https://files.pythonhosted.org/packages/43/94/b0b4fdc3014122e0a7302fb90086d352aa48f2576f0b252561ebb38c01a8/torch-2.12.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:a6a2eebb237d3b1d9ad3b378e86d9b9e0782afdea8b1e0eba6a13646b9b49c07", size = 532183124, upload-time = "2026-05-13T14:53:16.178Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/c8/052405e6ad05d3237bfe5a4df78f917773956f8e17813a2d44c059068b74/torch-2.12.0-cp313-cp313t-win_amd64.whl", hash = "sha256:2140e373e9a51a3e22ef62e8d14366d0b470d18f0adf19fdc757368077133a34", size = 123232462, upload-time = "2026-05-13T14:52:27.26Z" },
+    { url = "https://files.pythonhosted.org/packages/67/dc/ac069f8d6e8be701535921141055293b0d4819d3d7f224a4612cf157c7f9/torch-2.12.0-cp314-cp314-macosx_14_0_arm64.whl", hash = "sha256:f7dfae4a519197dfa050e98d8e36378a0fb5899625a875c2b54445005a2e404e", size = 88027282, upload-time = "2026-05-13T14:53:05.258Z" },
+    { url = "https://files.pythonhosted.org/packages/33/c3/1c1eb00e34555b536dddf792676026a988d710ed36981aa00499b36b0620/torch-2.12.0-cp314-cp314-manylinux_2_28_aarch64.whl", hash = "sha256:891c769072637c74e9a5a77a3bc782894696d8ffec83b938df8536dee7f0ba78", size = 426386961, upload-time = "2026-05-13T14:51:28.406Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/d4/7e730dba0c7032a4154dc9056b76cf9625515e030e269cfbf8098fcfee7d/torch-2.12.0-cp314-cp314-manylinux_2_28_x86_64.whl", hash = "sha256:e2ad3eb85d39c3cab62dfa93ed5a73516e6a53c6713cb97d004004fe089f0f1f", size = 532272265, upload-time = "2026-05-13T14:51:59.308Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/b4/92c80d1bbfee1c0036c06d1d2155a3065bd2423134c83bf8a47e65cd6b9b/torch-2.12.0-cp314-cp314-win_amd64.whl", hash = "sha256:c66696857e987efb8bc1777a37357ec4f60ab5e8af6250b83d6034437fa2d8f3", size = 122987138, upload-time = "2026-05-13T14:51:45.942Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/78/2e12b37ce50a19a037d7bc62d652a5a8f27385a7b05859d6bc9204f20cfe/torch-2.12.0-cp314-cp314t-macosx_14_0_arm64.whl", hash = "sha256:b4556715c8572758625d62b6e0ae3b1f76c440221913a6fb5e100f321fb4fb02", size = 88320100, upload-time = "2026-05-13T14:51:39.955Z" },
+    { url = "https://files.pythonhosted.org/packages/56/5e/83c450ec7b0bb40a7b74611c1b5440f9260e33c54c90d556fd4a1f0fd955/torch-2.12.0-cp314-cp314t-manylinux_2_28_aarch64.whl", hash = "sha256:a43ac605a5e13116c72b64c359644cce0229f213dde48d2ae0ae5eb5becf7feb", size = 426391871, upload-time = "2026-05-13T14:52:14.989Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/e9/1a0b575d98d0afedd8f157d23fa3d2759421483660448e60d0a4b10b6daa/torch-2.12.0-cp314-cp314t-manylinux_2_28_x86_64.whl", hash = "sha256:6a7512adfdd7f6732e40de1c620831e3c75b39b98cef60b11d0c5f0a76473ec5", size = 532192241, upload-time = "2026-05-13T14:51:07.795Z" },
+    { url = "https://files.pythonhosted.org/packages/88/21/afadd25ecd81b3cea1e11c73cf1ab41a983a50271548c3ec7ec3b9efc3e9/torch-2.12.0-cp314-cp314t-win_amd64.whl", hash = "sha256:5f96b63f8287f66a005dd1b5a6abba2920f11156c5e5c4d815f3e2050fd1aa16", size = 123231092, upload-time = "2026-05-13T14:51:18.854Z" },
+]
+
+[[package]]
+name = "torch-optimizer"
+version = "0.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pytorch-ranger" },
+    { name = "torch" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/18/13/c4c0a206131e978d8ceaa095ad1e3153d7daf48efad207b6057efe3491a2/torch-optimizer-0.3.0.tar.gz", hash = "sha256:b2180629df9d6cd7a2aeabe71fa4a872bba938e8e275965092568cd9931b924c", size = 54409, upload-time = "2021-10-31T03:00:22.084Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f6/54/bbb1b4c15afc2dac525c8359c340ade685542113394fd4c6564ee3c71da3/torch_optimizer-0.3.0-py3-none-any.whl", hash = "sha256:7de8e57315e43561cdd0370a1b67303cc8ef1b053f9b5573de629a62390f2af9", size = 61897, upload-time = "2021-10-31T03:00:19.812Z" },
+]
+
+[[package]]
+name = "tqdm"
+version = "4.68.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/85/05/0d5260f1f1ca784f4a4a0def9cbe6affe587f5b4025328d446c3d67765f4/tqdm-4.68.2.tar.gz", hash = "sha256:89c230e8dbc67c7615c142487111222f878c77427ea09549960f62389e258add", size = 171923, upload-time = "2026-06-09T13:26:42.539Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/eb/75/1a0392bcc21c44dcdf87b3cf2d137e7829be2c083a1e38d44efca3d57a16/tqdm-4.68.2-py3-none-any.whl", hash = "sha256:d4240441fb5353290b87d6a85968c9decc131a99b8c7faa28269d829de669ede", size = 78578, upload-time = "2026-06-09T13:26:40.731Z" },
+]
+
+[[package]]
+name = "triton"
+version = "3.7.0"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3e/97/dcd1f2a0f8336691bff74abc59b2ed9c69a0c0f8f65cd77109c49e05f068/triton-3.7.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:223ac302091491436c248a34ee1e6c47a1026486579103c906ffd805be50cb89", size = 188367104, upload-time = "2026-05-07T19:04:56.68Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/c0/c2ac4fd2d8809b7579d4a820a0f9e5de62a9bc8a757ed4b3abf4f7ee964a/triton-3.7.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c631b65668d4951213b948a413c0564184305b77bb45cc9d686d3e1ecc4701a3", size = 201313191, upload-time = "2026-05-07T18:45:58.444Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/c1/5d842314bb6c78442cc60437928781701c6050b8d479bc2a1aed691d37ca/triton-3.7.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a9e71fc392675fac364e0ecf4ef3f76f85b7f5433a16f4c3c5fe5f05a52c85fe", size = 188480277, upload-time = "2026-05-07T19:05:03.231Z" },
+    { url = "https://files.pythonhosted.org/packages/13/31/8315ea5f8dd18e60970b3022e3a8b93fd37e0b784fbbef86e10c8e6e5ca1/triton-3.7.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:22bacffce443f54593dd20f05294d5a40622e0ea9ab632816f87154504356221", size = 201415942, upload-time = "2026-05-07T18:46:06.479Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/13/ec05adfcd87311d532ba61e3af143e8be59fcd26675884c4682841406a20/triton-3.7.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a4bf49b00a7a377a68a6da603a876e797614e6455a80e9021669c476a953ad9a", size = 188505104, upload-time = "2026-05-07T19:05:09.843Z" },
+    { url = "https://files.pythonhosted.org/packages/62/7b/468a576e35beef1426e0828e28e9ba9e65f5474d496f16ee126c15646324/triton-3.7.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8f111161d49bf903c0eaedde3962353a3d841c08a836839b7cc1025b8426efcf", size = 201457567, upload-time = "2026-05-07T18:46:13.505Z" },
+    { url = "https://files.pythonhosted.org/packages/01/e1/a59a583de59b8f62c495d67c80ee3ea97d09e91ac80c4c6e76456ed8d8ac/triton-3.7.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:abdf6beaa89b1bcfb9a43cd990536ce66091a997841a4814b260b7bee4c88c3c", size = 188503209, upload-time = "2026-05-07T19:05:17.935Z" },
+    { url = "https://files.pythonhosted.org/packages/30/b1/b7507bb9815d403927c8dd51d4158ed2e11751a92dbc118a044f247b6848/triton-3.7.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a35d7afe3f3f058e7ec49fcce09794049e0ffc5c59019ac25ec3413741b8c4e7", size = 201453566, upload-time = "2026-05-07T18:46:20.427Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/8f/0bea7a6a0c989315c9135a1d7fb37e41905cfb3a17cbc1f10044ebd4cc3a/triton-3.7.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cc1d61c172d257db80ddf42595131fb196ad2e9bdd751e90fe2ef13531734e8b", size = 188612899, upload-time = "2026-05-07T19:05:24.955Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/02/d96f57828d0912aec733b9bc7e0e7dbfd2c6f079a8fa433ac25cb93d1a30/triton-3.7.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:70fb9bbdc9f400afc54bbf6eb2670af28829a6ae3996863317964783141daf56", size = 201553816, upload-time = "2026-05-07T18:46:27.49Z" },
+    { url = "https://files.pythonhosted.org/packages/40/fb/82a802dac4689f2a2fb2e69302e6a138eecc3e175bbe976ba3cfc717683a/triton-3.7.0-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c4a44a8476d0d3571eac4e4d1048e1ff75aad81a09ff4602ccfc56c6dea1672e", size = 188507879, upload-time = "2026-05-07T19:05:32.209Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/af/9904ec6d3c93d9b24e5ec360445bbdf758b7f00bfbeedb89cb0eb64eb8bb/triton-3.7.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b9b85e72968a9d8bba5ddb24e9b64aaabaf48affb042f2755cb7cfa92b7531ce", size = 201460637, upload-time = "2026-05-07T18:46:34.749Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/f9/4835a8ea746b88727d8899f4e3ccce4f9cacb38abfc3bb0a638266c53111/triton-3.7.0-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:18a160de426fd99f92b0baf509045360afbd3bfaa0b4a5171dde800ec9f09684", size = 188608706, upload-time = "2026-05-07T19:05:39.218Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/68/fa86e5a39608000f645535b2c124920126327ab731f8c4fafd5b07ff8d4b/triton-3.7.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ce061073102714b725f3660ec6939d94a1da7984b3aa99c921417cae273672f5", size = 201546766, upload-time = "2026-05-07T18:46:42.088Z" },
+]
+
+[[package]]
+name = "typing-extensions"
+version = "4.15.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/94/1a15dd82efb362ac84269196e94cf00f187f7ed21c242792a923cdb1c61f/typing_extensions-4.15.0.tar.gz", hash = "sha256:0cea48d173cc12fa28ecabc3b837ea3cf6f38c6d1136f85cbaaf598984861466", size = 109391, upload-time = "2025-08-25T13:49:26.313Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/18/67/36e9267722cc04a6b9f15c7f3441c2363321a3ea07da7ae0c0707beb2a9c/typing_extensions-4.15.0-py3-none-any.whl", hash = "sha256:f0fa19c6845758ab08074a0cfa8b7aecb71c999ca73d62883bc25cc018c4e548", size = 44614, upload-time = "2025-08-25T13:49:24.86Z" },
+]
+
+[[package]]
+name = "typing-inspection"
+version = "0.4.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/55/e3/70399cb7dd41c10ac53367ae42139cf4b1ca5f36bb3dc6c9d33acdb43655/typing_inspection-0.4.2.tar.gz", hash = "sha256:ba561c48a67c5958007083d386c3295464928b01faa735ab8547c5692e87f464", size = 75949, upload-time = "2025-10-01T02:14:41.687Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/dc/9b/47798a6c91d8bdb567fe2698fe81e0c6b7cb7ef4d13da4114b41d239f65d/typing_inspection-0.4.2-py3-none-any.whl", hash = "sha256:4ed1cacbdc298c220f1bd249ed5287caa16f34d44ef4e9c3d0cbad5b521545e7", size = 14611, upload-time = "2025-10-01T02:14:40.154Z" },
+]
+
+[[package]]
+name = "tzdata"
+version = "2026.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ba/19/1b9b0e29f30c6d35cb345486df41110984ea67ae69dddbc0e8a100999493/tzdata-2026.2.tar.gz", hash = "sha256:9173fde7d80d9018e02a662e168e5a2d04f87c41ea174b139fbef642eda62d10", size = 198254, upload-time = "2026-04-24T15:22:08.651Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/e4/dccd7f47c4b64213ac01ef921a1337ee6e30e8c6466046018326977efd95/tzdata-2026.2-py2.py3-none-any.whl", hash = "sha256:bbe9af844f658da81a5f95019480da3a89415801f6cc966806612cc7169bffe7", size = 349321, upload-time = "2026-04-24T15:22:05.876Z" },
+]
+
+[[package]]
+name = "win32-setctime"
+version = "1.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/b3/8f/705086c9d734d3b663af0e9bb3d4de6578d08f46b1b101c2442fd9aecaa2/win32_setctime-1.2.0.tar.gz", hash = "sha256:ae1fdf948f5640aae05c511ade119313fb6a30d7eabe25fef9764dca5873c4c0", size = 4867, upload-time = "2024-12-07T15:28:28.314Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e1/07/c6fe3ad3e685340704d314d765b7912993bcb8dc198f0e7a89382d37974b/win32_setctime-1.2.0-py3-none-any.whl", hash = "sha256:95d644c4e708aba81dc3704a116d8cbc974d70b3bdb8be1d150e36be6e9d1390", size = 4083, upload-time = "2024-12-07T15:28:26.465Z" },
+]
+
+[[package]]
+name = "xlrd"
+version = "2.0.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/07/5a/377161c2d3538d1990d7af382c79f3b2372e880b65de21b01b1a2b78691e/xlrd-2.0.2.tar.gz", hash = "sha256:08b5e25de58f21ce71dc7db3b3b8106c1fa776f3024c54e45b45b374e89234c9", size = 100167, upload-time = "2025-06-14T08:46:39.039Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1a/62/c8d562e7766786ba6587d09c5a8ba9f718ed3fa8af7f4553e8f91c36f302/xlrd-2.0.2-py2.py3-none-any.whl", hash = "sha256:ea762c3d29f4cca48d82df517b6d89fbce4db3107f9d78713e48cd321d5c9aa9", size = 96555, upload-time = "2025-06-14T08:46:37.766Z" },
+]

From bd6890844c50558866325d40c5e2cfdd475d19d6 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 15:20:00 +0200
Subject: [PATCH 51/81] Separate storage and modelling window size

---
 README.md                                     |   2 +-
 documentation/configs/infer.md                |   2 +-
 documentation/configs/preprocess.md           |   8 +-
 documentation/configs/train.md                |   2 +-
 documentation/consolidated-docs.md            |  14 +-
 .../config/hyperparameter_search_config.py    |  45 ++--
 src/sequifier/config/infer_config.py          |  55 +++--
 src/sequifier/config/preprocess_config.py     |  12 +-
 src/sequifier/config/train_config.py          |  55 +++--
 src/sequifier/helpers.py                      | 203 +++++++++---------
 src/sequifier/hyperparameter_search.py        |  10 +-
 src/sequifier/infer.py                        |  92 +++-----
 .../io/sequifier_dataset_from_file.py         |  15 +-
 .../sequifier_dataset_from_folder_parquet.py  |  43 ++--
 ...uifier_dataset_from_folder_parquet_lazy.py |  36 +---
 .../io/sequifier_dataset_from_folder_pt.py    |  32 +--
 .../sequifier_dataset_from_folder_pt_lazy.py  |  32 +--
 src/sequifier/io/yaml.py                      |   5 +-
 src/sequifier/make.py                         |   7 +-
 src/sequifier/preprocess.py                   |  97 ++++-----
 src/sequifier/train.py                        |  11 +-
 tests/configs/hyperparameter-search-bert.yaml |   1 -
 ...infer-test-categorical-autoregression.yaml |   1 -
 ...infer-test-categorical-bert-embedding.yaml |   1 -
 .../configs/infer-test-categorical-bert.yaml  |   1 -
 .../infer-test-categorical-embedding.yaml     |   1 -
 .../infer-test-categorical-inf-size-1.yaml    |   1 -
 ...test-categorical-inf-size-3-embedding.yaml |   1 -
 .../infer-test-categorical-inf-size-3.yaml    |   1 -
 .../infer-test-categorical-multitarget.yaml   |   1 -
 tests/configs/infer-test-categorical.yaml     |   1 -
 .../infer-test-distributed-parquet.yaml       |   1 -
 tests/configs/infer-test-distributed.yaml     |   1 -
 tests/configs/infer-test-lazy.yaml            |   1 -
 .../infer-test-real-autoregression.yaml       |   1 -
 tests/configs/infer-test-real.yaml            |   1 -
 .../preprocess-test-categorical-exact-pt.yaml |   3 +-
 .../preprocess-test-categorical-exact.yaml    |   3 +-
 ...eprocess-test-categorical-interrupted.yaml |   3 +-
 ...eprocess-test-categorical-lookahead-0.yaml |   4 +-
 ...eprocess-test-categorical-multitarget.yaml |   3 +-
 ...ategorical-precomputed-stats-negative.yaml |   3 +-
 ...ss-test-categorical-precomputed-stats.yaml |   3 +-
 .../configs/preprocess-test-categorical.yaml  |   3 +-
 tests/configs/preprocess-test-multi-file.yaml |   3 +-
 tests/configs/preprocess-test-real.yaml       |   3 +-
 .../configs/train-test-categorical-bert.yaml  |   2 -
 .../train-test-categorical-inf-size-1.yaml    |   2 -
 .../train-test-categorical-inf-size-3.yaml    |   2 -
 ...in-test-categorical-multitarget-eager.yaml |   2 -
 .../train-test-categorical-multitarget.yaml   |   2 -
 tests/configs/train-test-categorical.yaml     |   2 -
 .../train-test-distributed-lazy-parquet.yaml  |   2 -
 tests/configs/train-test-distributed.yaml     |   2 -
 tests/configs/train-test-lazy.yaml            |   2 -
 tests/configs/train-test-real-bert.yaml       |   2 -
 tests/configs/train-test-real.yaml            |   2 -
 tests/configs/train-test-resume-epoch.yaml    |   2 -
 .../configs/train-test-resume-mid-epoch.yaml  |   2 -
 .../integration/test_hyperparameter_search.py |   9 +-
 tests/integration/test_make.py                |   4 +-
 tests/integration/test_onnx_export.py         |  24 ++-
 tests/integration/test_preprocessing.py       |  13 +-
 .../preprocess-manifest.json                  |   7 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  22 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  22 +-
 tests/unit/test_helpers.py                    |  57 +++--
 tests/unit/test_infer.py                      |  38 ++--
 tests/unit/test_preprocess.py                 |  41 ++--
 tests/unit/test_train.py                      |  57 +++--
 70 files changed, 564 insertions(+), 580 deletions(-)

diff --git a/README.md b/README.md
index 8aa1c7f9..183d05bd 100644
--- a/README.md
+++ b/README.md
@@ -97,7 +97,7 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `context_length + max_lookahead` (`max_lookahead` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Preprocessing defines the physical `stored_width` and `future_capacity`; training and inference choose the model-facing `context_length` from that stored capacity:
 
 |sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
 |----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
diff --git a/documentation/configs/infer.md b/documentation/configs/infer.md
index 6a80934e..b4c63f53 100644
--- a/documentation/configs/infer.md
+++ b/documentation/configs/infer.md
@@ -40,7 +40,7 @@ These fields tell the inference engine which columns to extract from the new dat
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `model_type` | `str` | **Yes** | - | `generative` (predict next value) or `embedding` (extract vector representation). |
-| `context_length` | `int` | **Yes** | - | The context window size. Must match training. |
+| `context_length` | `int` | **Yes** | - | The model context window size. It must match the trained model view and fit inside the stored metadata capacity. |
 | `prediction_length` | `int` | No | `1` | Number of steps to predict *simultaneously*. **Must be 1** if `autoregression: true`. |
 | `inference_batch_size`| `int` | **Yes** | - | Number of sequences to process at once. |
 | `autoregression` | `bool` | No | `false` | If `true`, feeds predictions back into the model to predict further into the future. |
diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index 8185baec..a9ee943e 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -45,10 +45,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
-| `context_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
-| `max_lookahead` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `context_length + max_lookahead`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
+| `stored_width` | `int` | **Yes** | - | The physical serialized window width written to preprocessed data. |
+| `future_capacity` | `int` | No | `1` | Number of future items retained after the model input window. Use `0` for BERT-style same-width inputs and targets; use `1` for causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
-| `stride_by_split` | `list[int]` | No | `[context_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
+| `stride_by_split` | `list[int]` | No | `[stored_width]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
 
 ### 4\. Performance & System
@@ -76,7 +76,7 @@ This controls data augmentation and redundancy.
   * **Stride = `context_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
   * **Stride = 1 (Maximum Overlap):** Maximizes data volume. The model sees every possible sequence. This yields the highest accuracy but significantly increases the size of the preprocessed data and training time.
   * **Hybrid Approach:** It is common practice to set a large stride for the training and validation splits (index 0) to reduce the size on disk of the dataset, and a stride=1 for the test split to evaluate the model on each point in the test set. This supposes that the test split value is low.
-      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `context_length: 48`).
+      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `stored_width: 49`).
 
 ### 3\. `subsequence_start_mode`: `distribute` vs `exact`
 
diff --git a/documentation/configs/train.md b/documentation/configs/train.md
index 48df255a..4a66e7ca 100644
--- a/documentation/configs/train.md
+++ b/documentation/configs/train.md
@@ -30,7 +30,7 @@ The configuration is defined in a YAML file (e.g., `train.yaml`). The file is st
 | `target_columns` | `list[str]`| **Yes** | - | The specific column(s) the model should learn to predict. |
 | `target_column_types`| `dict` | **Yes** | - | Map of target columns to their type: `'categorical'` or `'real'`. The key order in target_column_types must exactly match the list order in target_columns |
 | `input_columns` | `list[str]`| No | All | Subset of columns to use as input features. Defaults to all available in metadata. |
-| `context_length` | `int` | **Yes** | - | Must match the `context_length` used in preprocessing. |
+| `context_length` | `int` | **Yes** | - | Model input context length. It must fit inside the metadata `stored_width` with the stored `future_capacity`. |
 
 ### 3\. Model Architecture (`model_spec`)
 
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 0baf670c..d8a0b1a6 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -97,7 +97,7 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Each stored window has width `context_length + max_lookahead` (`max_lookahead` defaults to `1`; set it to `0` for BERT-style same-width inputs and targets):
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Preprocessing defines the physical `stored_width` and `future_capacity`; training and inference choose the model-facing `context_length` from that stored capacity:
 
 |sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
 |----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
@@ -253,10 +253,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
-| `context_length` | `int` | **Yes** | - | The length of the context window (history) fed into the model. |
-| `max_lookahead` | `int` | No | `1` | Number of future items retained after the input context. The stored window width is `context_length + max_lookahead`. Use `0` for BERT-style preprocessing where inputs and targets have the same width; keep `1` for legacy causal next-item training. |
+| `stored_width` | `int` | **Yes** | - | The physical serialized window width written to preprocessed data. |
+| `future_capacity` | `int` | No | `1` | Number of future items retained after the model input window. Use `0` for BERT-style same-width inputs and targets; use `1` for causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
-| `stride_by_split` | `list[int]` | No | `[context_length]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
+| `stride_by_split` | `list[int]` | No | `[stored_width]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
 
 ### 4\. Performance & System
@@ -284,7 +284,7 @@ This controls data augmentation and redundancy.
   * **Stride = `context_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
   * **Stride = 1 (Maximum Overlap):** Maximizes data volume. The model sees every possible sequence. This yields the highest accuracy but significantly increases the size of the preprocessed data and training time.
   * **Hybrid Approach:** It is common practice to set a large stride for the training and validation splits (index 0) to reduce the size on disk of the dataset, and a stride=1 for the test split to evaluate the model on each point in the test set. This supposes that the test split value is low.
-      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `context_length: 48`).
+      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `stored_width: 49`).
 
 ### 3\. `subsequence_start_mode`: `distribute` vs `exact`
 
@@ -379,7 +379,7 @@ The configuration is defined in a YAML file (e.g., `train.yaml`). The file is st
 | `target_columns` | `list[str]`| **Yes** | - | The specific column(s) the model should learn to predict. |
 | `target_column_types`| `dict` | **Yes** | - | Map of target columns to their type: `'categorical'` or `'real'`. The key order in target_column_types must exactly match the list order in target_columns |
 | `input_columns` | `list[str]`| No | All | Subset of columns to use as input features. Defaults to all available in metadata. |
-| `context_length` | `int` | **Yes** | - | Must match the `context_length` used in preprocessing. |
+| `context_length` | `int` | **Yes** | - | Model input context length. It must fit inside the metadata `stored_width` with the stored `future_capacity`. |
 
 ### 3\. Model Architecture (`model_spec`)
 
@@ -604,7 +604,7 @@ These fields tell the inference engine which columns to extract from the new dat
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `model_type` | `str` | **Yes** | - | `generative` (predict next value) or `embedding` (extract vector representation). |
-| `context_length` | `int` | **Yes** | - | The context window size. Must match training. |
+| `context_length` | `int` | **Yes** | - | The model context window size. It must match the trained model view and fit inside the stored metadata capacity. |
 | `prediction_length` | `int` | No | `1` | Number of steps to predict *simultaneously*. **Must be 1** if `autoregression: true`. |
 | `inference_batch_size`| `int` | **Yes** | - | Number of sequences to process at once. |
 | `autoregression` | `bool` | No | `false` | If `true`, feeds predictions back into the model to predict further into the future. |
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index c2b6c5ff..f097e507 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -18,9 +18,11 @@
     TrainModel,
 )
 from sequifier.helpers import (
-    SequenceLayout,
+    ModelWindowView,
+    StoredWindowLayout,
     normalize_path,
-    sequence_layout_from_metadata,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
     try_catch_excess_keys,
 )
 from sequifier.special_tokens import validate_special_token_ids
@@ -207,18 +209,14 @@ def load_hyperparameter_search_config(
             "n_classes", metadata_config["n_classes"]
         )
 
-        sequence_layout = sequence_layout_from_metadata(metadata_config)
-        if sequence_layout.sequence_layout_version != 2:
+        storage_layout = stored_window_layout_from_metadata(metadata_config)
+        if storage_layout.version != 2:
             raise ValueError(
-                "Hyperparameter search requires metadata sequence_layout_version=2, "
-                f"got {sequence_layout.sequence_layout_version}."
+                "Hyperparameter search requires metadata stored_window_layout_version=2, "
+                f"got {storage_layout.version}."
             )
 
-        config_values["max_lookahead"] = sequence_layout.max_lookahead
-        config_values["sample_length"] = sequence_layout.sample_length
-        config_values["sequence_layout_version"] = (
-            sequence_layout.sequence_layout_version
-        )
+        config_values["storage_layout"] = storage_layout
 
         config_values["training_data_path"] = normalize_path(
             config_values.get("training_data_path", metadata_config["split_paths"][0]),
@@ -718,9 +716,7 @@ class HyperparameterSearchConfig(BaseModel):
     id_maps: dict[str, dict[str | int, int]]
 
     context_length: list[int]
-    max_lookahead: int = Field(default=1, ge=0)
-    sample_length: int
-    sequence_layout_version: int
+    storage_layout: StoredWindowLayout
     n_classes: dict[str, int]
     inference_batch_size: int
 
@@ -743,10 +739,13 @@ class HyperparameterSearchConfig(BaseModel):
     @model_validator(mode="after")
     def validate_sequence_layout(self):
         for cl in self.context_length:
-            if cl + self.max_lookahead > self.sample_length:
+            if (
+                cl + self.storage_layout.future_capacity
+                > self.storage_layout.stored_width
+            ):
                 raise ValueError(
-                    f"Sample length mismatch: context_length ({cl}) + max_lookahead "
-                    f"({self.max_lookahead}) > stored sample_length ({self.sample_length}). "
+                    f"Window capacity mismatch: context_length ({cl}) + future_capacity "
+                    f"({self.storage_layout.future_capacity}) > stored_width ({self.storage_layout.stored_width}). "
                     "Model inputs cannot exceed the preprocessed sequence length."
                 )
         return self
@@ -856,12 +855,13 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
         context_length = trial.suggest_categorical(
             "context_length", self.context_length
         )
-        layout = SequenceLayout(
+        training_spec = self.training_hyperparameter_sampling.sample_trial(trial)
+        window_view = ModelWindowView(
             context_length=context_length,
-            max_lookahead=self.max_lookahead,
-            sequence_layout_version=self.sequence_layout_version,
+            objective=training_spec.training_objective,
+            target_shift=0 if training_spec.training_objective == "bert" else 1,
         )
-        training_spec = self.training_hyperparameter_sampling.sample_trial(trial)
+        resolve_window_view(self.storage_layout, window_view)
 
         logger.info(f"{input_columns_index = } - {context_length = }")
 
@@ -879,7 +879,8 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             target_columns=self.target_columns,
             target_column_types=self.target_column_types,
             id_maps=self.id_maps,
-            layout=layout,
+            storage_layout=self.storage_layout,
+            window_view=window_view,
             n_classes=self.n_classes,
             inference_batch_size=self.inference_batch_size,
             seed=101,
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 92a63b9e..77deaee1 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -15,9 +15,11 @@
 )
 
 from sequifier.helpers import (
-    SequenceLayout,
+    ModelWindowView,
+    StoredWindowLayout,
     normalize_path,
-    sequence_layout_from_metadata,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
     try_catch_excess_keys,
 )
 from sequifier.special_tokens import validate_special_token_ids
@@ -56,18 +58,31 @@ def load_inferer_config(
             metadata_config["special_token_ids"],
             source=f"metadata config '{metadata_config_path}'",
         )
-        sequence_layout = sequence_layout_from_metadata(metadata_config)
-        if sequence_layout.sequence_layout_version != 2:
+        storage_layout = stored_window_layout_from_metadata(metadata_config)
+        if storage_layout.version != 2:
             raise ValueError(
-                "Inference requires metadata sequence_layout_version=2, "
-                f"got {sequence_layout.sequence_layout_version}."
+                "Inference requires metadata stored_window_layout_version=2, "
+                f"got {storage_layout.version}."
             )
-        config_values["layout"] = sequence_layout
+        training_objective = config_values["training_objective"]
+        target_shift = (
+            0
+            if training_objective == "bert"
+            else int(config_values.pop("target_shift", 1))
+        )
+        window_view = ModelWindowView(
+            context_length=int(config_values.pop("context_length")),
+            objective=training_objective,
+            target_shift=target_shift,
+        )
+        resolve_window_view(storage_layout, window_view)
+        config_values["storage_layout"] = storage_layout
+        config_values["window_view"] = window_view
         for key in (
-            "context_length",
-            "max_lookahead",
-            "sample_length",
-            "sequence_layout_version",
+            "target_shift",
+            "stored_width",
+            "future_capacity",
+            "stored_window_layout_version",
         ):
             config_values.pop(key, None)
 
@@ -161,7 +176,8 @@ class InfererModel(BaseModel):
     map_to_id: bool = Field(default=True)
     seed: int
     device: str
-    layout: SequenceLayout
+    storage_layout: StoredWindowLayout
+    window_view: ModelWindowView
     prediction_length: Optional[int] = None
     inference_batch_size: int
 
@@ -172,18 +188,25 @@ class InfererModel(BaseModel):
 
     @model_validator(mode="after")
     def normalize_prediction_length(self):
+        if self.window_view.objective != self.training_objective:
+            raise ValueError(
+                "window_view objective must match training_objective "
+                f"({self.window_view.objective} != {self.training_objective})."
+            )
         if self.prediction_length is None:
             self.prediction_length = (
-                self.layout.context_length if self.training_objective == "bert" else 1
+                self.window_view.context_length
+                if self.training_objective == "bert"
+                else 1
             )
         if self.training_objective == "bert":
-            if self.prediction_length != self.layout.context_length:
+            if self.prediction_length != self.window_view.context_length:
                 raise ValueError(
                     "For BERT inference, prediction_length must be equal to context_length "
-                    f"(got prediction_length={self.prediction_length}, context_length={self.layout.context_length})."
+                    f"(got prediction_length={self.prediction_length}, context_length={self.window_view.context_length})."
                 )
         else:
-            self.layout.get_target_offset(self.training_objective)
+            resolve_window_view(self.storage_layout, self.window_view)
         return self
 
     @field_validator("training_objective")
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index ed1aa518..da53eb8b 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -54,8 +54,8 @@ class PreprocessorModel(BaseModel):
         selected_columns: A list of columns to be included in the preprocessing. If None, all columns are used.
         split_ratios: A list of floats that define the relative sizes of data splits (e.g., for train, validation, test).
                            The sum of proportions must be 1.0.
-        context_length: The sequence length for the model inputs.
-        max_lookahead: The maximum retained target offset after the model input window.
+        stored_width: The physical serialized window width.
+        future_capacity: The number of future items retained after the input window.
         stride_by_split: A list of step sizes for creating subsequences within each data split.
         max_rows: The maximum number of input rows to process. If None, all rows are processed.
         seed: A random seed for reproducibility.
@@ -78,8 +78,8 @@ class PreprocessorModel(BaseModel):
     selected_columns: Optional[list[str]] = None
 
     split_ratios: list[float]
-    context_length: int
-    max_lookahead: int = Field(default=1, ge=0)
+    stored_width: int = Field(gt=0)
+    future_capacity: int = Field(default=1, ge=0)
     stride_by_split: Optional[list[int]] = None
     max_rows: Optional[int] = None
     seed: int
@@ -194,10 +194,12 @@ def validate_mask_column_requires_metadata(self) -> "PreprocessorModel":
             raise ValueError("metadata_config_path must be set when mask_column is set")
         if self.mask_column in ("sequenceId", "itemPosition"):
             raise ValueError("mask_column cannot be sequenceId or itemPosition")
+        if self.future_capacity >= self.stored_width:
+            raise ValueError("future_capacity must be smaller than stored_width")
         return self
 
     def __init__(self, **kwargs):
-        default_stride_for_split = [kwargs["context_length"]] * len(
+        default_stride_for_split = [kwargs["stored_width"]] * len(
             kwargs["split_ratios"]
         )
         kwargs["stride_by_split"] = kwargs.get(
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index eabfaa2f..34409a69 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -23,9 +23,11 @@
 import sequifier
 from sequifier.config.probabilities import ProbabilityDistribution
 from sequifier.helpers import (
-    SequenceLayout,
+    ModelWindowView,
+    StoredWindowLayout,
     normalize_path,
-    sequence_layout_from_metadata,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
     try_catch_excess_keys,
 )
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS, validate_special_token_ids
@@ -63,21 +65,36 @@ def load_train_config(
         ) as f:
             metadata_config = json.loads(f.read())
 
-        sequence_layout = sequence_layout_from_metadata(metadata_config)
-        if sequence_layout.sequence_layout_version != 2:
+        storage_layout = stored_window_layout_from_metadata(metadata_config)
+        if storage_layout.version != 2:
             raise ValueError(
-                "Training requires metadata sequence_layout_version=2, "
-                f"got {sequence_layout.sequence_layout_version}."
+                "Training requires metadata stored_window_layout_version=2, "
+                f"got {storage_layout.version}."
             )
-        config_values["layout"] = sequence_layout
+        training_objective = config_values["training_spec"]["training_objective"]
+        target_shift = (
+            0
+            if training_objective == "bert"
+            else int(config_values.pop("target_shift", 1))
+        )
+        window_view = ModelWindowView(
+            context_length=int(config_values.pop("context_length")),
+            objective=training_objective,
+            target_shift=target_shift,
+        )
+        resolve_window_view(storage_layout, window_view)
+        config_values["storage_layout"] = storage_layout
+        config_values["window_view"] = window_view
+        for key in ("stored_width", "future_capacity", "stored_window_layout_version"):
+            config_values.pop(key, None)
         for key in (
-            "context_length",
-            "max_lookahead",
-            "sample_length",
-            "sequence_layout_version",
+            "target_shift",
+            "stored_width",
+            "future_capacity",
+            "stored_window_layout_version",
         ):
             config_values.pop(key, None)
-        for key in ("max_lookahead", "sample_length"):
+        for key in ("target_shift", "stored_width", "future_capacity"):
             config_values.get("training_spec", {}).pop(key, None)
 
         split_paths = metadata_config["split_paths"]
@@ -576,7 +593,8 @@ class TrainModel(BaseModel):
         default_factory=lambda: SPECIAL_TOKEN_IDS.ids_by_label
     )
 
-    layout: SequenceLayout
+    storage_layout: StoredWindowLayout
+    window_view: ModelWindowView
     n_classes: dict[str, int]
     inference_batch_size: int
     seed: int
@@ -597,14 +615,19 @@ def validate_special_token_ids_match_runtime(cls, v):
 
     @model_validator(mode="after")
     def validate_bert_prediction_length_matches_context_length(self):
-        self.layout.get_target_offset(self.training_spec.training_objective)
+        if self.window_view.objective != self.training_spec.training_objective:
+            raise ValueError(
+                "window_view objective must match training_spec.training_objective "
+                f"({self.window_view.objective} != {self.training_spec.training_objective})."
+            )
+        resolve_window_view(self.storage_layout, self.window_view)
         if (
             self.training_spec.training_objective == "bert"
-            and self.model_spec.prediction_length != self.layout.context_length
+            and self.model_spec.prediction_length != self.window_view.context_length
         ):
             raise ValueError(
                 "For BERT training, model_spec.prediction_length must be equal to context_length "
-                f"(got prediction_length={self.model_spec.prediction_length}, context_length={self.layout.context_length})."
+                f"(got prediction_length={self.model_spec.prediction_length}, context_length={self.window_view.context_length})."
             )
         return self
 
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index ae7f4022..a94967f1 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -44,95 +44,116 @@
 
 
 @dataclass(frozen=True)
-class SequenceLayout:
+class StoredWindowLayout:
+    stored_width: int
+    future_capacity: int
+    version: int
+
+    def __post_init__(self) -> None:
+        if self.stored_width < 1:
+            raise ValueError("stored_width must be a positive integer")
+        if self.future_capacity < 0:
+            raise ValueError("future_capacity must be non-negative")
+        if self.future_capacity >= self.stored_width:
+            raise ValueError("future_capacity must be smaller than stored_width")
+
+
+@dataclass(frozen=True)
+class ModelWindowView:
     context_length: int
-    max_lookahead: int
-    sequence_layout_version: int
+    objective: str
+    target_shift: int = 1
 
     def __post_init__(self) -> None:
         if self.context_length < 1:
             raise ValueError("context_length must be a positive integer")
-        if self.max_lookahead < 0:
-            raise ValueError("max_lookahead must be non-negative")
-
-    @property
-    def sample_length(self) -> int:
-        return self.context_length + self.max_lookahead
-
-    @property
-    def input_offset(self) -> int:
-        return self.max_lookahead
-
-    def get_target_offset(self, training_objective: str) -> int:
-        if training_objective == "bert":
-            return self.max_lookahead
-        if training_objective == "causal":
-            if self.max_lookahead < 1:
-                raise ValueError(
-                    "Causal training requires data preprocessed with max_lookahead >= 1"
-                )
-            return self.max_lookahead - 1
-        raise ValueError(
-            f"Only 'causal' and 'bert' are allowed, found {training_objective}"
-        )
+        if self.objective not in {"causal", "bert"}:
+            raise ValueError(
+                f"Only 'causal' and 'bert' are allowed, found {self.objective}"
+            )
+        if self.target_shift < 0:
+            raise ValueError("target_shift must be non-negative")
+        if self.objective == "bert" and self.target_shift != 0:
+            raise ValueError("BERT views require target_shift=0")
+        if self.objective == "causal" and self.target_shift < 1:
+            raise ValueError("Causal views require target_shift >= 1")
 
 
-@beartype
-def sequence_column_names(context_length: int, offset: int) -> list[str]:
-    if offset < 0:
-        raise ValueError("offset must be non-negative")
-    return [str(i) for i in range(context_length - 1 + offset, offset - 1, -1)]
+@dataclass(frozen=True)
+class ResolvedWindowView:
+    storage: StoredWindowLayout
+    view: ModelWindowView
+    required_width: int
+    input_slice: slice
+    target_slice: slice
+
+    def build_masks(self, left_pad_lengths: Tensor) -> dict[str, Tensor]:
+        """Build explicit input-attention and target-validity masks for this view."""
+        return {
+            "attention_valid_mask": build_valid_mask(
+                left_pad_lengths, self.storage.stored_width, self.input_slice
+            ),
+            "target_valid_mask": build_valid_mask(
+                left_pad_lengths, self.storage.stored_width, self.target_slice
+            ),
+        }
 
 
 @beartype
-def slice_window(tensor: Tensor, context_length: int, offset: int) -> Tensor:
-    if offset < 0:
-        raise ValueError("offset must be non-negative")
+def _right_aligned_slice(width: int, length: int, offset: int) -> slice:
+    start = width - (length + offset)
+    stop = width - offset if offset else width
+    return slice(start, stop)
 
-    end = -offset if offset else None
-    result = tensor[:, -(context_length + offset) : end]
 
-    if result.shape[1] != context_length:
+@beartype
+def resolve_window_view(
+    storage: StoredWindowLayout, view: ModelWindowView
+) -> ResolvedWindowView:
+    if view.target_shift > storage.future_capacity:
         raise ValueError(
-            f"Stored window width {tensor.shape[1]} cannot provide "
-            f"context_length={context_length} at offset={offset}"
+            f"Model target_shift={view.target_shift} exceeds stored "
+            f"future_capacity={storage.future_capacity}."
         )
-    return result
 
-
-@beartype
-def validate_stored_window_width(tensor: Tensor, sample_length: int) -> None:
-    if tensor.shape[1] != sample_length:
+    input_offset = storage.future_capacity
+    target_offset = storage.future_capacity - view.target_shift
+    required_width = view.context_length + max(input_offset, target_offset)
+    if required_width > storage.stored_width:
         raise ValueError(
-            f"Stored window width {tensor.shape[1]} does not match "
-            f"metadata sample_length={sample_length}."
+            f"Model view requires width {required_width}, but storage only has "
+            f"stored_width={storage.stored_width}."
         )
 
+    return ResolvedWindowView(
+        storage=storage,
+        view=view,
+        required_width=required_width,
+        input_slice=_right_aligned_slice(
+            storage.stored_width, view.context_length, input_offset
+        ),
+        target_slice=_right_aligned_slice(
+            storage.stored_width, view.context_length, target_offset
+        ),
+    )
 
-@beartype
-def sequence_layout_from_metadata(metadata: dict) -> SequenceLayout:
-    metadata_context_length = int(metadata["context_length"])
 
-    max_lookahead = int(metadata["max_lookahead"])
-    sample_length = int(metadata["sample_length"])
-    if sample_length != metadata_context_length + max_lookahead:
+@beartype
+def validate_stored_window_width(tensor: Tensor, stored_width: int) -> None:
+    if tensor.shape[1] != stored_width:
         raise ValueError(
-            "Invalid sequence layout metadata: sample_length must equal "
-            "context_length + max_lookahead "
-            f"({sample_length} != {metadata_context_length} + {max_lookahead})."
+            f"Stored window width {tensor.shape[1]} does not match "
+            f"metadata stored_width={stored_width}."
         )
 
-    layout = SequenceLayout(
-        context_length=metadata_context_length,
-        max_lookahead=max_lookahead,
-        sequence_layout_version=int(metadata["sequence_layout_version"]),
+
+@beartype
+def stored_window_layout_from_metadata(metadata: dict) -> StoredWindowLayout:
+    return StoredWindowLayout(
+        stored_width=int(metadata["stored_width"]),
+        future_capacity=int(metadata["future_capacity"]),
+        version=int(metadata["stored_window_layout_version"]),
     )
-    if layout.sample_length != sample_length:
-        raise ValueError(
-            f"Resolved layout sample_length={layout.sample_length} does not match "
-            f"metadata sample_length={sample_length}."
-        )
-    return layout
 
 
 # Check an environment variable to see if we are in a testing context
@@ -331,10 +352,7 @@ def numpy_to_pytorch(
     data: pl.DataFrame,
     column_types: dict[str, torch.dtype],
     all_columns: list[str],
-    context_length: int,
-    sample_length: int,
-    data_offset: int,
-    target_offset: int,
+    resolved_view: ResolvedWindowView,
 ) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
     """Converts a long-format Polars DataFrame to a dict of sequence tensors.
 
@@ -370,8 +388,12 @@ def numpy_to_pytorch(
           (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
         - a metadata dictionary containing any explicit masks.
     """
-    input_seq_cols = sequence_column_names(context_length, data_offset)
-    target_seq_cols = sequence_column_names(context_length, target_offset)
+    input_seq_cols = columns_from_slice(
+        resolved_view.input_slice, resolved_view.storage.stored_width
+    )
+    target_seq_cols = columns_from_slice(
+        resolved_view.target_slice, resolved_view.storage.stored_width
+    )
 
     # We will create a unified dictionary
     unified_tensors = {}
@@ -397,13 +419,7 @@ def numpy_to_pytorch(
         unified_tensors[f"{col_name}_target"] = target_tensor
 
     left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(data)
-    metadata = generate_padding_masks(
-        left_pad_lengths,
-        context_length,
-        sample_length,
-        data_offset,
-        target_offset,
-    )
+    metadata = resolved_view.build_masks(left_pad_lengths)
 
     return unified_tensors, metadata
 
@@ -412,8 +428,7 @@ def numpy_to_pytorch(
 def build_valid_mask(
     left_pad_lengths: Tensor,
     full_length: int,
-    offset: int,
-    context_length: int,
+    view_slice: slice,
 ) -> Tensor:
     """Builds a boolean validity mask from explicit left-padding metadata."""
 
@@ -422,7 +437,14 @@ def build_valid_mask(
     )
     full_valid = full_positions[None, :] >= left_pad_lengths[:, None]
 
-    return full_valid[:, -(context_length + offset) : (-offset if offset > 0 else None)]
+    return full_valid[:, view_slice]
+
+
+@beartype
+def columns_from_slice(view_slice: slice, stored_width: int) -> list[str]:
+    if view_slice.start is None or view_slice.stop is None:
+        raise ValueError("Resolved window slices must have concrete bounds")
+    return [str(stored_width - 1 - i) for i in range(view_slice.start, view_slice.stop)]
 
 
 @beartype
@@ -445,25 +467,6 @@ def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Tensor:
     return torch.tensor(lengths.to_numpy(), dtype=torch.int64)
 
 
-@beartype
-def generate_padding_masks(
-    left_pad_lengths: Tensor,
-    context_length: int,
-    sample_length: int,
-    data_offset: int,
-    target_offset: int,
-) -> dict[str, Tensor]:
-    """Generates explicit attention and target masks as a metadata dictionary."""
-    return {
-        "attention_valid_mask": build_valid_mask(
-            left_pad_lengths, sample_length, data_offset, context_length
-        ),
-        "target_valid_mask": build_valid_mask(
-            left_pad_lengths, sample_length, target_offset, context_length
-        ),
-    }
-
-
 @beartype
 def normalize_path(path: str, project_root: str) -> str:
     """Normalizes a path to be relative to a project path, then joins them.
diff --git a/src/sequifier/hyperparameter_search.py b/src/sequifier/hyperparameter_search.py
index 7948b682..d39028cb 100644
--- a/src/sequifier/hyperparameter_search.py
+++ b/src/sequifier/hyperparameter_search.py
@@ -58,8 +58,16 @@ def objective(trial: optuna.Trial, config) -> Union[float, tuple[float, ...]]:
         config.project_root, config.model_config_write_path, f"{run_name}.yaml"
     )
     os.makedirs(os.path.dirname(config_path), exist_ok=True)
+
+    run_config_dict = run_config.model_dump()
+    run_config_dict["context_length"] = run_config_dict["window_view"]["context_length"]
+    run_config_dict["target_shift"] = run_config_dict["window_view"]["target_shift"]
+
+    del run_config_dict["window_view"]
+    del run_config_dict["storage_layout"]
+
     with open(config_path, "w") as f:
-        yaml.dump(run_config, f, Dumper=TrainModelDumper, sort_keys=False)
+        yaml.dump(run_config_dict, f, Dumper=TrainModelDumper, sort_keys=False)
 
     os.environ["SEQUIFIER_HYPERPARAMETER_SEARCH_RUN"] = "1"
 
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 4514ab76..804eca47 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -17,10 +17,9 @@
     PANDAS_TO_TORCH_TYPES,
     configure_determinism,
     construct_index_maps,
-    generate_padding_masks,
     normalize_path,
     numpy_to_pytorch,
-    slice_window,
+    resolve_window_view,
     subset_to_input_columns,
     validate_stored_window_width,
     write_data,
@@ -221,7 +220,10 @@ def infer_worker(
                 f"Unsupported input type or read format: {config.read_format}"
             )
 
-        default_prediction_length = {"causal": 1, "bert": config.layout.context_length}
+        default_prediction_length = {
+            "causal": 1,
+            "bert": config.window_view.context_length,
+        }
         prediction_length = (
             config.prediction_length
             if config.prediction_length is not None
@@ -361,13 +363,8 @@ def _bert_target_valid_mask_from_preprocessed_data(
         left_pad_lengths = torch.tensor(
             reference_rows.get_column("leftPadLength").to_numpy(), dtype=torch.int64
         )
-        metadata = generate_padding_masks(
-            left_pad_lengths,
-            config.layout.context_length,
-            config.layout.sample_length,
-            data_offset=config.layout.input_offset,
-            target_offset=config.layout.get_target_offset(config.training_objective),
-        )
+        resolved_view = resolve_window_view(config.storage_layout, config.window_view)
+        metadata = resolved_view.build_masks(left_pad_lengths)
         return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
 
     return None
@@ -495,17 +492,12 @@ def infer_embedding(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.layout.sample_length)
+                validate_stored_window_width(tensor, config.storage_layout.stored_width)
 
-            metadata = generate_padding_masks(
-                left_pad_lengths_tensor,
-                config.layout.context_length,
-                config.layout.sample_length,
-                data_offset=config.layout.input_offset,
-                target_offset=config.layout.get_target_offset(
-                    config.training_objective
-                ),
+            resolved_view = resolve_window_view(
+                config.storage_layout, config.window_view
             )
+            metadata = resolved_view.build_masks(left_pad_lengths_tensor)
             embeddings = get_embeddings_pt(
                 config, inferer, sequences_dict, metadata=metadata
             )
@@ -523,8 +515,8 @@ def infer_embedding(
         # Step 2: Calculate absolute positions and repeat IDs
         # (e.g., for seq_len=50, inf_size=5, offsets are [45, 46, 47, 48, 49])
         base_offsets = np.arange(
-            config.layout.context_length - prediction_length,
-            config.layout.context_length,
+            config.window_view.context_length - prediction_length,
+            config.window_view.context_length,
         )
 
         # Tile these offsets for each sample in the batch
@@ -671,7 +663,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.layout.context_length,
+                    config.window_view.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -688,7 +680,7 @@ def infer_generative(
                         inferer,
                         data,
                         column_types,
-                        config.layout.context_length,
+                        config.window_view.context_length,
                     )
                 )
         elif config.read_format == "parquet" and is_folder_input:
@@ -726,7 +718,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.layout.context_length,
+                    config.window_view.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -742,7 +734,7 @@ def infer_generative(
                         inferer,
                         data,
                         column_types,
-                        config.layout.context_length,
+                        config.window_view.context_length,
                     )
                 )
         elif config.read_format == "pt":
@@ -754,30 +746,23 @@ def infer_generative(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.layout.sample_length)
+                validate_stored_window_width(tensor, config.storage_layout.stored_width)
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
                 else config.autoregression_total_steps
             )
 
+            resolved_view = resolve_window_view(
+                config.storage_layout, config.window_view
+            )
             sequences_dict = {
-                key: slice_window(
-                    tensor, config.layout.context_length, config.layout.input_offset
-                )
+                key: tensor[:, resolved_view.input_slice]
                 for key, tensor in sequences_dict.items()
                 if key in config.input_columns
             }
 
-            metadata = generate_padding_masks(
-                left_pad_lengths_tensor,
-                config.layout.context_length,
-                config.layout.sample_length,
-                data_offset=config.layout.input_offset,
-                target_offset=config.layout.get_target_offset(
-                    config.training_objective
-                ),
-            )
+            metadata = resolved_view.build_masks(left_pad_lengths_tensor)
 
             probs, preds = get_probs_preds_from_dict(
                 config, inferer, sequences_dict, metadata, total_steps
@@ -800,7 +785,7 @@ def infer_generative(
                 # Invoke the unified positioning engine
                 item_positions_for_preds = calculate_item_positions(
                     item_positions_base_raw,
-                    config.layout.context_length,
+                    config.window_view.context_length,
                     prediction_length,
                     config.training_objective,
                 )
@@ -810,10 +795,10 @@ def infer_generative(
                     sequence_ids_tensor.numpy(), total_steps
                 )
                 item_position_boundaries = zip(
-                    list(start_positions_tensor + config.layout.context_length),
+                    list(start_positions_tensor + config.window_view.context_length),
                     list(
                         start_positions_tensor
-                        + config.layout.context_length
+                        + config.window_view.context_length
                         + total_steps
                     ),
                 )
@@ -959,12 +944,11 @@ def get_embeddings_pt(
     Returns:
         A NumPy array containing the computed embeddings for the batch.
     """
+    resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     for tensor in data.values():
-        validate_stored_window_width(tensor, config.layout.sample_length)
+        validate_stored_window_width(tensor, config.storage_layout.stored_width)
     X = {
-        key: slice_window(
-            val, config.layout.context_length, config.layout.input_offset
-        ).numpy()
+        key: val[:, resolved_view.input_slice].numpy()
         for key, val in data.items()
         if key in config.input_columns
     }
@@ -1114,14 +1098,12 @@ def get_embeddings(
         A NumPy array containing the computed embeddings for the batch.
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
+    resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     X, metadata = numpy_to_pytorch(
         data,
         column_types,
         all_columns,
-        config.layout.context_length,
-        config.layout.sample_length,
-        config.layout.input_offset,
-        config.layout.get_target_offset(config.training_objective),
+        resolved_view,
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     metadata_np = {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
@@ -1163,14 +1145,12 @@ def get_probs_preds_from_df(
     """
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
 
+    resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     X, metadata = numpy_to_pytorch(
         data,
         column_types,
         all_columns,
-        config.layout.context_length,
-        config.layout.sample_length,
-        config.layout.input_offset,
-        config.layout.get_target_offset(config.training_objective),
+        resolved_view,
     )
     X = {col: X_col.numpy() for col, X_col in X.items()}
     metadata_np = {col: metadata_col.numpy() for col, metadata_col in metadata.items()}
@@ -1289,14 +1269,12 @@ def get_probs_preds_autoregression(
         + context_length
     )
 
+    resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     head_data, metadata = numpy_to_pytorch(
         head_data_df,
         column_types,
         config.input_columns,
-        context_length,
-        config.layout.sample_length,
-        data_offset=config.layout.input_offset,
-        target_offset=0,
+        resolved_view,
     )
 
     # Run the autoregressive PyTorch inference
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index 71ce10ac..6dc03b16 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -7,7 +7,12 @@
 from torch.utils.data import IterableDataset
 
 from sequifier.config.train_config import TrainModel
-from sequifier.helpers import PANDAS_TO_TORCH_TYPES, numpy_to_pytorch, read_data
+from sequifier.helpers import (
+    PANDAS_TO_TORCH_TYPES,
+    numpy_to_pytorch,
+    read_data,
+    resolve_window_view,
+)
 from sequifier.io.batch import SequifierBatch
 
 
@@ -42,16 +47,12 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         }
 
         # self.all_tensors now holds both inputs and targets
+        resolved_view = resolve_window_view(config.storage_layout, config.window_view)
         all_tensors, metadata_tensors = numpy_to_pytorch(
             data=data_df,
             column_types=column_types,
             all_columns=all_columns,
-            context_length=config.layout.context_length,
-            sample_length=config.layout.sample_length,
-            data_offset=config.layout.input_offset,
-            target_offset=config.layout.get_target_offset(
-                config.training_spec.training_objective
-            ),
+            resolved_view=resolved_view,
         )
         self.n_samples = all_tensors[all_columns[0]].shape[0]
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 6eebda70..24fe6e12 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -12,11 +12,11 @@
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
-    generate_padding_masks,
+    columns_from_slice,
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
-    sequence_column_names,
-    sequence_layout_from_metadata,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
 )
 from sequifier.io.batch import SequifierBatch
 
@@ -49,12 +49,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.layout.sample_length:
-            raise ValueError(
-                f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.layout.sample_length}."
-            )
+        self.folder_layout = stored_window_layout_from_metadata(metadata)
+        self.resolved_view = resolve_window_view(self.folder_layout, config.window_view)
 
         self.n_samples = metadata["total_samples"]
 
@@ -68,15 +64,11 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         }
 
         # Sequence formatting structures matching long-format schema boundaries
-        train_seq_len = self.config.layout.context_length
-        input_seq_cols = sequence_column_names(
-            train_seq_len, self.config.layout.input_offset
+        input_seq_cols = columns_from_slice(
+            self.resolved_view.input_slice, self.folder_layout.stored_width
         )
-        target_seq_cols = sequence_column_names(
-            train_seq_len,
-            self.config.layout.get_target_offset(
-                self.config.training_spec.training_objective
-            ),
+        target_seq_cols = columns_from_slice(
+            self.resolved_view.target_slice, self.folder_layout.stored_width
         )
         all_sequences: Dict[str, list[torch.Tensor]] = {
             col: [] for col in config.input_columns
@@ -211,29 +203,20 @@ def __iter__(
         indices_for_worker = indices_for_rank[worker_id::num_workers]
 
         # 5. Extract and pass unified data frames
-        train_seq_len = self.config.layout.context_length
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
             data_batch = {
-                key: tensor[batch_indices, -train_seq_len:]
-                for key, tensor in self.sequences.items()
+                key: tensor[batch_indices] for key, tensor in self.sequences.items()
             }
             targets_batch = {
-                key: tensor[batch_indices, -train_seq_len:]
-                for key, tensor in self.targets.items()
+                key: tensor[batch_indices] for key, tensor in self.targets.items()
             }
 
             metadata_batch = {}
             if self.left_pad_lengths is not None:
-                metadata_batch = generate_padding_masks(
-                    self.left_pad_lengths[batch_indices],
-                    train_seq_len,
-                    self.config.layout.sample_length,
-                    self.config.layout.input_offset,
-                    self.config.layout.get_target_offset(
-                        self.config.training_spec.training_objective
-                    ),
+                metadata_batch = self.resolved_view.build_masks(
+                    self.left_pad_lengths[batch_indices]
                 )
 
             yield SequifierBatch(
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 496fbf6c..7f2e5bcc 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -13,11 +13,11 @@
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
-    generate_padding_masks,
+    columns_from_slice,
     get_left_pad_lengths_from_preprocessed_data,
     normalize_path,
-    sequence_column_names,
-    sequence_layout_from_metadata,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
 )
 from sequifier.io.batch import SequifierBatch
 
@@ -65,12 +65,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.layout.sample_length:
-            raise ValueError(
-                f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.layout.sample_length}."
-            )
+        self.folder_layout = stored_window_layout_from_metadata(metadata)
+        self.resolved_view = resolve_window_view(self.folder_layout, config.window_view)
 
         self.batch_files_info = metadata["batch_files"]
         self.total_samples = metadata["total_samples"]
@@ -219,17 +215,13 @@ def __iter__(
 
         # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
-        train_seq_len = self.config.layout.context_length
         global_file_start_sample = 0
 
-        input_seq_cols = sequence_column_names(
-            train_seq_len, self.config.layout.input_offset
+        input_seq_cols = columns_from_slice(
+            self.resolved_view.input_slice, self.folder_layout.stored_width
         )
-        target_seq_cols = sequence_column_names(
-            train_seq_len,
-            self.config.layout.get_target_offset(
-                self.config.training_spec.training_objective
-            ),
+        target_seq_cols = columns_from_slice(
+            self.resolved_view.target_slice, self.folder_layout.stored_width
         )
 
         # Initialize cross-file buffers
@@ -313,15 +305,7 @@ def __iter__(
                         f"Missing required column {col_name} in Parquet partition"
                     )
 
-            new_meta = generate_padding_masks(
-                left_pad_lengths[worker_indices],
-                train_seq_len,
-                self.config.layout.sample_length,
-                self.config.layout.input_offset,
-                self.config.layout.get_target_offset(
-                    self.config.training_spec.training_objective
-                ),
-            )
+            new_meta = self.resolved_view.build_masks(left_pad_lengths[worker_indices])
 
             del df
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index a1647fea..e60ea981 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -10,10 +10,9 @@
 
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
-    generate_padding_masks,
     normalize_path,
-    sequence_layout_from_metadata,
-    slice_window,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
     validate_stored_window_width,
 )
 from sequifier.io.batch import SequifierBatch
@@ -47,12 +46,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.layout.sample_length:
-            raise ValueError(
-                f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.layout.sample_length}."
-            )
+        self.folder_layout = stored_window_layout_from_metadata(metadata)
+        self.resolved_view = resolve_window_view(self.folder_layout, config.window_view)
 
         self.n_samples = metadata["total_samples"]
 
@@ -78,7 +73,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in all_sequences.keys():
                 if col in sequences_batch:
                     validate_stored_window_width(
-                        sequences_batch[col], config.layout.sample_length
+                        sequences_batch[col], self.folder_layout.stored_width
                     )
                     all_sequences[col].append(sequences_batch[col])
             all_left_pad_lengths.append(left_pad_lengths_batch)
@@ -169,33 +164,24 @@ def __iter__(
         indices_for_worker = indices_for_rank[worker_id::num_workers]
 
         # 5. Yield full batches
-        train_seq_len = self.config.layout.context_length
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
 
-            data_offset = self.config.layout.input_offset
-            target_offset = self.config.layout.get_target_offset(
-                self.config.training_spec.training_objective
-            )
             data_batch = {
-                key: slice_window(tensor[batch_indices], train_seq_len, data_offset)
+                key: tensor[batch_indices, self.resolved_view.input_slice]
                 for key, tensor in self.sequences.items()
                 if key in self.config.input_columns
             }
             targets_batch = {
-                key: slice_window(tensor[batch_indices], train_seq_len, target_offset)
+                key: tensor[batch_indices, self.resolved_view.target_slice]
                 for key, tensor in self.sequences.items()
                 if key in self.config.target_columns
             }
 
             metadata_batch = {}
             if self.left_pad_lengths is not None:
-                metadata_batch = generate_padding_masks(
-                    self.left_pad_lengths[batch_indices],
-                    train_seq_len,
-                    self.config.layout.sample_length,
-                    data_offset,
-                    target_offset,
+                metadata_batch = self.resolved_view.build_masks(
+                    self.left_pad_lengths[batch_indices]
                 )
 
             yield SequifierBatch(
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 4c50e903..df67ee66 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -11,10 +11,9 @@
 
 from sequifier.config.train_config import TrainModel
 from sequifier.helpers import (
-    generate_padding_masks,
     normalize_path,
-    sequence_layout_from_metadata,
-    slice_window,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
     validate_stored_window_width,
 )
 from sequifier.io.batch import SequifierBatch
@@ -63,12 +62,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         with open(metadata_path, "r") as f:
             metadata = json.load(f)
 
-        folder_layout = sequence_layout_from_metadata(metadata)
-        if folder_layout.sample_length != config.layout.sample_length:
-            raise ValueError(
-                f"Preprocessed folder sample_length={folder_layout.sample_length} "
-                f"does not match config sample_length={config.layout.sample_length}."
-            )
+        self.folder_layout = stored_window_layout_from_metadata(metadata)
+        self.resolved_view = resolve_window_view(self.folder_layout, config.window_view)
 
         self.batch_files_info = metadata["batch_files"]
         self.total_samples = metadata["total_samples"]
@@ -212,7 +207,6 @@ def __iter__(
 
         # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
-        train_seq_len = self.config.layout.context_length
         global_file_start_sample = 0
 
         # Initialize cross-file buffers
@@ -244,7 +238,7 @@ def __iter__(
                 left_pad_lengths_batch,
             ) = torch.load(file_path, map_location="cpu", weights_only=False)
             for tensor in sequences_batch.values():
-                validate_stored_window_width(tensor, self.config.layout.sample_length)
+                validate_stored_window_width(tensor, self.folder_layout.stored_width)
 
             # Generate indices for the whole file
             indices = torch.arange(file_samples)
@@ -265,27 +259,19 @@ def __iter__(
                 continue
 
             # Extract the data subset for this worker (Advanced indexing copies the data)
-            data_offset = self.config.layout.input_offset
-            target_offset = self.config.layout.get_target_offset(
-                self.config.training_spec.training_objective
-            )
             new_seq = {
-                k: slice_window(v[worker_indices], train_seq_len, data_offset)
+                k: v[worker_indices, self.resolved_view.input_slice]
                 for k, v in sequences_batch.items()
                 if k in self.config.input_columns
             }
             new_tgt = {
-                k: slice_window(v[worker_indices], train_seq_len, target_offset)
+                k: v[worker_indices, self.resolved_view.target_slice]
                 for k, v in sequences_batch.items()
                 if k in self.config.target_columns
             }
 
-            new_meta = generate_padding_masks(
-                left_pad_lengths_batch[worker_indices],
-                train_seq_len,
-                self.config.layout.sample_length,
-                data_offset,
-                target_offset,
+            new_meta = self.resolved_view.build_masks(
+                left_pad_lengths_batch[worker_indices]
             )
 
             # Free the large file immediately to keep RAM down
diff --git a/src/sequifier/io/yaml.py b/src/sequifier/io/yaml.py
index b35497b4..934f40bf 100644
--- a/src/sequifier/io/yaml.py
+++ b/src/sequifier/io/yaml.py
@@ -8,7 +8,7 @@
     TrainingSpecModel,
     TrainModel,
 )
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import ModelWindowView, StoredWindowLayout
 
 
 def represent_sequifier_object(dumper, data):
@@ -79,7 +79,8 @@ def increase_indent(self, flow=False, indentless=False):
 TrainModelDumper.add_representer(TrainModel, represent_sequifier_object)
 TrainModelDumper.add_representer(ModelSpecModel, represent_sequifier_object)
 TrainModelDumper.add_representer(TrainingSpecModel, represent_sequifier_object)
-TrainModelDumper.add_representer(SequenceLayout, represent_sequifier_object)
+TrainModelDumper.add_representer(StoredWindowLayout, represent_sequifier_object)
+TrainModelDumper.add_representer(ModelWindowView, represent_sequifier_object)
 TrainModelDumper.add_multi_representer(BaseModel, represent_sequifier_object)
 TrainModelDumper.add_representer(DotDict, represent_dot_dict)
 TrainModelDumper.add_representer(numpy.float64, represent_numpy_float)
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 8f27cf17..79d5cc42 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -11,8 +11,8 @@
 - 0.8
 - 0.1
 - 0.1
-context_length: 48
-max_lookahead: 1
+stored_width: 49
+future_capacity: 1
 stride_by_split:
 - 1
 - 1
@@ -31,7 +31,6 @@
   EXAMPLE_TARGET_COLUMN_NAME: real
 
 context_length: 48
-sample_length: 49
 inference_batch_size: 10
 
 export_generative_model: PLEASE FILL # true or false
@@ -49,7 +48,6 @@
   num_layers: 3
   prediction_length: 1
 training_spec:
-  sample_length: 49
   training_objective: causal
   device: cuda
   epochs: 10
@@ -91,7 +89,6 @@
 map_to_id: true
 device: cpu
 context_length: 48
-sample_length: 49
 inference_batch_size: 10
 
 autoregression: true
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 8232688a..4172ce18 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -18,7 +18,7 @@
 from sequifier.config.preprocess_config import load_preprocessor_config
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
-    SequenceLayout,
+    StoredWindowLayout,
     read_data,
     write_data,
 )
@@ -31,7 +31,7 @@
 
 INPUT_METADATA_COLUMNS = ("sequenceId", "itemPosition")
 REAL_MASK_VALUE = 0.0
-CURRENT_SEQUENCE_LAYOUT_VERSION = 2
+CURRENT_STORED_WINDOW_LAYOUT_VERSION = 2
 
 
 @beartype
@@ -86,7 +86,7 @@ def __init__(
         merge_output: bool,
         selected_columns: Optional[list[str]],
         split_ratios: list[float],
-        context_length: int,
+        stored_width: int,
         stride_by_split: list[int],
         max_rows: Optional[int],
         seed: int,
@@ -96,7 +96,7 @@ def __init__(
         subsequence_start_mode: str,
         use_precomputed_maps: Optional[list[str]],
         metadata_config_path: Optional[str],
-        max_lookahead: int = 1,
+        future_capacity: int = 1,
         mask_column: Optional[str] = None,
     ):
         """Initializes the Preprocessor with the given parameters.
@@ -109,8 +109,8 @@ def __init__(
             merge_output: Whether to combine the output into a single file.
             selected_columns: A list of columns to be included in the preprocessing.
             split_ratios: A list of floats that define the relative sizes of data splits.
-            context_length: The sequence length for the model inputs.
-            max_lookahead: The maximum target horizon retained after each input window.
+            stored_width: The physical serialized window width.
+            future_capacity: The number of future items retained after each input window.
             stride_by_split: A list of step sizes for creating subsequences.
             max_rows: The maximum number of input rows to process.
             seed: A random seed for reproducibility.
@@ -145,10 +145,10 @@ def __init__(
         np.random.seed(seed)
         self.n_cores = n_cores or multiprocessing.cpu_count()
         self.continue_preprocessing = continue_preprocessing
-        self.layout = SequenceLayout(
-            context_length,
-            max_lookahead,
-            sequence_layout_version=CURRENT_SEQUENCE_LAYOUT_VERSION,
+        self.storage_layout = StoredWindowLayout(
+            stored_width=stored_width,
+            future_capacity=future_capacity,
+            version=CURRENT_STORED_WINDOW_LAYOUT_VERSION,
         )
         self._setup_directories()
 
@@ -251,7 +251,7 @@ def __init__(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
 
-            schema = self._create_schema(col_types, self.layout.sample_length)
+            schema = self._create_schema(col_types, self.storage_layout.stored_width)
 
             data = data.sort(["sequenceId", "itemPosition"])
             n_batches = _process_batches_single_file(
@@ -260,7 +260,7 @@ def __init__(
                 data,
                 schema,
                 self.n_cores,
-                self.layout,
+                self.storage_layout,
                 stride_by_split,
                 data_columns,
                 col_types,
@@ -347,7 +347,7 @@ def __init__(
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
-            schema = self._create_schema(col_types, self.layout.sample_length)
+            schema = self._create_schema(col_types, self.storage_layout.stored_width)
 
             self._process_batches_multiple_files(
                 files_to_process,
@@ -356,7 +356,7 @@ def __init__(
                 max_rows,
                 schema,
                 self.n_cores,
-                self.layout,
+                self.storage_layout,
                 stride_by_split,
                 data_columns,
                 n_classes,
@@ -373,7 +373,7 @@ def __init__(
 
     @beartype
     def _create_schema(
-        self, col_types: dict[str, str], sample_length: int
+        self, col_types: dict[str, str], stored_width: int
     ) -> dict[str, Any]:
         """Creates the Polars schema for the intermediate sequence DataFrame.
 
@@ -385,7 +385,7 @@ def _create_schema(
         Args:
             col_types: A dictionary mapping data column names to their Polars
                 string representations (e.g., "Int64", "Float64").
-            sample_length: The number of items stored for each extracted window.
+            stored_width: The number of items stored for each extracted window.
 
         Returns:
             A dictionary defining the Polars schema. Keys are column names
@@ -415,7 +415,7 @@ def _create_schema(
             sequence_position_type = pl.Float64
 
         schema.update(
-            {str(i): sequence_position_type for i in range(sample_length - 1, -1, -1)}
+            {str(i): sequence_position_type for i in range(stored_width - 1, -1, -1)}
         )
 
         return schema
@@ -618,7 +618,7 @@ def _process_batches_multiple_files(
         max_rows: Optional[int],
         schema: Any,
         n_cores: int,
-        layout: SequenceLayout,
+        layout: StoredWindowLayout,
         stride_by_split: list[int],
         data_columns: list[str],
         n_classes: dict[str, int],
@@ -817,10 +817,9 @@ def _cleanup(self, write_format: str) -> None:
     @beartype
     def _layout_metadata(self) -> dict[str, int]:
         return {
-            "context_length": self.layout.context_length,
-            "max_lookahead": self.layout.max_lookahead,
-            "sample_length": self.layout.sample_length,
-            "sequence_layout_version": self.layout.sequence_layout_version,
+            "stored_width": self.storage_layout.stored_width,
+            "future_capacity": self.storage_layout.future_capacity,
+            "stored_window_layout_version": self.storage_layout.version,
         }
 
     @beartype
@@ -1465,7 +1464,7 @@ def _process_batches_multiple_files_inner(
     max_rows: Optional[int],
     schema: Any,
     n_cores: int,
-    layout: SequenceLayout,
+    layout: StoredWindowLayout,
     stride_by_split: list[int],
     data_columns: list[str],
     n_classes: dict[str, int],
@@ -1623,7 +1622,7 @@ def _process_batches_single_file(
     data: pl.DataFrame,
     schema: Any,
     n_cores: Optional[int],
-    layout: SequenceLayout,
+    layout: StoredWindowLayout,
     stride_by_split: list[int],
     data_columns: list[str],
     col_types: dict[str, str],
@@ -1902,7 +1901,7 @@ def get_group_bounds(data_subset: pl.DataFrame, split_ratios: list[float]):
 @beartype
 def process_and_write_data_pt(
     data: pl.DataFrame,
-    sample_length: int,
+    stored_width: int,
     path: str,
     column_types: dict[str, str],
 ):
@@ -1915,13 +1914,13 @@ def process_and_write_data_pt(
 
     It then converts these lists into NumPy arrays and stores one full
     sequence tensor per feature. Each tensor has shape
-    `(batch_size, sample_length)`. The final five-element tuple
+    `(batch_size, stored_width)`. The final five-element tuple
     `(sequences_dict, sequence_ids_tensor, subsequence_ids_tensor, start_item_positions_tensor, left_pad_lengths_tensor)`
     is saved to a .pt file using `torch.save`.
 
     Args:
         data: The long-format Polars DataFrame of extracted sequences.
-        sample_length: The stored serialized window width.
+        stored_width: The stored serialized window width.
         path: The output file path (e.g., "data/batch_0.pt").
         column_types: A dictionary mapping column names to their
             string data types, used to determine the correct torch dtype.
@@ -1929,7 +1928,7 @@ def process_and_write_data_pt(
     if data.is_empty():
         return
 
-    sequence_cols = [str(c) for c in range(sample_length - 1, -1, -1)]
+    sequence_cols = [str(c) for c in range(stored_width - 1, -1, -1)]
 
     all_feature_cols = data.get_column("inputCol").unique().to_list()
 
@@ -1998,7 +1997,7 @@ def _write_accumulated_sequences(
     process_id: int,
     file_index_str: str,
     target_dir: str,
-    layout: SequenceLayout,
+    layout: StoredWindowLayout,
     col_types: dict[str, str],
 ):
     """Helper to write a batch of accumulated sequences to a single .pt file.
@@ -2029,9 +2028,7 @@ def _write_accumulated_sequences(
     out_path = insert_top_folder(split_path_batch_seq, target_dir)
 
     if write_format == "pt":
-        process_and_write_data_pt(
-            combined_df, layout.sample_length, out_path, col_types
-        )
+        process_and_write_data_pt(combined_df, layout.stored_width, out_path, col_types)
     elif write_format == "parquet":
         combined_df.write_parquet(out_path)
 
@@ -2044,7 +2041,7 @@ def preprocess_batch(
     batch: pl.DataFrame,
     schema: Any,
     split_paths: list[str],
-    layout: SequenceLayout,
+    layout: StoredWindowLayout,
     stride_by_split: list[int],
     data_columns: list[str],
     col_types: dict[str, str],
@@ -2183,7 +2180,7 @@ def preprocess_batch(
 def extract_sequences(
     data: pl.DataFrame,
     schema: Any,
-    layout: SequenceLayout,
+    layout: StoredWindowLayout,
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
@@ -2225,7 +2222,7 @@ def extract_sequences(
 
         subsequences, left_pad_lengths, subsequence_starts = extract_subsequences(
             in_seq_lists_only,
-            layout.sample_length,
+            layout.stored_width,
             stride_for_split,
             columns,
             subsequence_start_mode,
@@ -2241,7 +2238,7 @@ def extract_sequences(
                     left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
-                expected_row_length = 5 + layout.sample_length
+                expected_row_length = 5 + layout.stored_width
                 if len(row) != expected_row_length:
                     raise RuntimeError(
                         f"Row length mismatch. Expected {expected_row_length}, got {len(row)}. Row: {row}"
@@ -2259,21 +2256,21 @@ def extract_sequences(
 @beartype
 def get_subsequence_starts(
     in_context_length: int,
-    sample_length: int,
+    stored_width: int,
     stride_for_split: int,
     subsequence_start_mode: str,
 ) -> np.ndarray:
     """Calculates the start indices for extracting subsequences.
 
     This function determines the starting indices for sliding a window of
-    `sample_length` over an input sequence of `in_context_length`. It aims to
+    `stored_width` over an input sequence of `in_context_length`. It aims to
     use `stride_for_split`, but adjusts the step size slightly to ensure
     that the windows are distributed as evenly as possible and cover the
     full sequence from the beginning to the end.
 
     Args:
         in_context_length: The length of the original input sequence.
-        sample_length: The stored window length to extract.
+        stored_width: The stored window length to extract.
         stride_for_split: The *desired* step size between subsequences.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
 
@@ -2286,7 +2283,7 @@ def get_subsequence_starts(
         )
 
     if subsequence_start_mode == "distribute":
-        last_available_start = in_context_length - sample_length
+        last_available_start = in_context_length - stored_width
         raw_starts = np.arange(
             0, last_available_start + stride_for_split, stride_for_split
         )
@@ -2297,11 +2294,11 @@ def get_subsequence_starts(
         return np.unique(starts)
 
     if subsequence_start_mode == "exact":
-        if (in_context_length - sample_length) % stride_for_split != 0:
+        if (in_context_length - stored_width) % stride_for_split != 0:
             raise ValueError(
-                f"'exact' mode requires sequence length alignment, i.e. if: (in_context_length - sample_length) % stride_for_split == 0, {in_context_length = }, {sample_length = }, {stride_for_split = }"
+                f"'exact' mode requires sequence length alignment, i.e. if: (in_context_length - stored_width) % stride_for_split == 0, {in_context_length = }, {stored_width = }, {stride_for_split = }"
             )
-        last_possible_start = in_context_length - sample_length
+        last_possible_start = in_context_length - stored_width
         return np.arange(0, last_possible_start + 1, stride_for_split)
     return np.array([])
 
@@ -2309,7 +2306,7 @@ def get_subsequence_starts(
 @beartype
 def extract_subsequences(
     in_seq: dict[str, list],
-    sample_length: int,
+    stored_width: int,
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
@@ -2319,14 +2316,14 @@ def extract_subsequences(
     This function takes a dictionary `in_seq` where keys are column
     names and values are lists of items for a single full sequence.
     It first pads the sequences with 0s at the beginning if they are
-    shorter than `sample_length`. Then, it calculates the subsequence
+    shorter than `stored_width`. Then, it calculates the subsequence
     start indices using `get_subsequence_starts` and extracts all
     subsequences.
 
     Args:
         in_seq: A dictionary mapping column names to lists of items
             (e.g., `{'col_A': [1, 2, 3, 4, 5], 'col_B': [6, 7, 8, 9, 10]}`).
-        sample_length: The stored window length to extract.
+        stored_width: The stored window length to extract.
         stride_for_split: The desired step size between subsequences.
         columns: A list of the column names (keys in `in_seq`) to process.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
@@ -2337,13 +2334,13 @@ def extract_subsequences(
     """
     in_seq_len = len(in_seq[columns[0]])
     pad_len = 0
-    if in_seq_len < sample_length:
-        pad_len = sample_length - in_seq_len
+    if in_seq_len < stored_width:
+        pad_len = stored_width - in_seq_len
         in_seq = {col: ([0] * pad_len) + in_seq[col] for col in columns}
     in_context_length = len(in_seq[columns[0]])
 
     subsequence_starts = get_subsequence_starts(
-        in_context_length, sample_length, stride_for_split, subsequence_start_mode
+        in_context_length, stored_width, stride_for_split, subsequence_start_mode
     )
     subsequence_starts_diff = subsequence_starts[1:] - subsequence_starts[:-1]
     if not np.all(subsequence_starts_diff <= stride_for_split):
@@ -2352,7 +2349,7 @@ def extract_subsequences(
         )
 
     result = {
-        col: [list(in_seq[col][i : i + sample_length]) for i in subsequence_starts]
+        col: [list(in_seq[col][i : i + stored_width]) for i in subsequence_starts]
         for col in columns
     }
     left_pad_lengths = [pad_len] * len(subsequence_starts)
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index f6b58150..bb78c930 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -97,14 +97,14 @@ def create_dummy_data_and_metadata(
     for col in config.input_columns:
         dtype = torch.int64 if col in config.categorical_columns else torch.float32
         dummy_data[col] = torch.ones(
-            (config.training_spec.batch_size, config.layout.context_length),
+            (config.training_spec.batch_size, config.window_view.context_length),
             dtype=dtype,
             device=local_rank,
         )
 
     dummy_metadata = {
         "attention_valid_mask": torch.ones(
-            (config.training_spec.batch_size, config.layout.context_length),
+            (config.training_spec.batch_size, config.window_view.context_length),
             dtype=torch.bool,
             device=local_rank,
         )
@@ -613,8 +613,9 @@ def __init__(
         self.target_columns = hparams.target_columns
         self.target_column_types = hparams.target_column_types
         self.loss_weights = hparams.training_spec.loss_weights
-        self.layout = hparams.layout
-        self.context_length = hparams.layout.context_length
+        self.storage_layout = hparams.storage_layout
+        self.window_view = hparams.window_view
+        self.context_length = hparams.window_view.context_length
         self.n_classes = hparams.n_classes
         self.inference_batch_size = hparams.inference_batch_size
         self.log_interval = hparams.training_spec.log_interval
@@ -691,7 +692,7 @@ def __init__(
                     hparams.model_spec.n_head,
                     hparams.model_spec.dim_feedforward,
                     hparams.training_spec.dropout,
-                    hparams.layout.context_length,
+                    hparams.window_view.context_length,
                 )
                 for _ in range(hparams.model_spec.num_layers)
             ]
diff --git a/tests/configs/hyperparameter-search-bert.yaml b/tests/configs/hyperparameter-search-bert.yaml
index b4490817..99a4024e 100644
--- a/tests/configs/hyperparameter-search-bert.yaml
+++ b/tests/configs/hyperparameter-search-bert.yaml
@@ -8,7 +8,6 @@ target_columns: [itemId]
 target_column_types:
   itemId: categorical
 context_length: [8]
-max_lookahead: 0
 inference_batch_size: 10
 
 search_strategy: sample
diff --git a/tests/configs/infer-test-categorical-autoregression.yaml b/tests/configs/infer-test-categorical-autoregression.yaml
index 842cd2ea..7112cd9e 100644
--- a/tests/configs/infer-test-categorical-autoregression.yaml
+++ b/tests/configs/infer-test-categorical-autoregression.yaml
@@ -17,7 +17,6 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical-bert-embedding.yaml b/tests/configs/infer-test-categorical-bert-embedding.yaml
index b74491a0..61724d81 100644
--- a/tests/configs/infer-test-categorical-bert-embedding.yaml
+++ b/tests/configs/infer-test-categorical-bert-embedding.yaml
@@ -17,7 +17,6 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 prediction_length: null
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-bert.yaml b/tests/configs/infer-test-categorical-bert.yaml
index ca23aba1..5a459841 100644
--- a/tests/configs/infer-test-categorical-bert.yaml
+++ b/tests/configs/infer-test-categorical-bert.yaml
@@ -17,7 +17,6 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 prediction_length: null
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-embedding.yaml b/tests/configs/infer-test-categorical-embedding.yaml
index 2c2bac57..e8ce1543 100644
--- a/tests/configs/infer-test-categorical-embedding.yaml
+++ b/tests/configs/infer-test-categorical-embedding.yaml
@@ -17,7 +17,6 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical-inf-size-1.yaml b/tests/configs/infer-test-categorical-inf-size-1.yaml
index 89af4da8..8c8c14b9 100644
--- a/tests/configs/infer-test-categorical-inf-size-1.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-1.yaml
@@ -16,7 +16,6 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
index 25f9f86e..61732fa0 100644
--- a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
@@ -19,7 +19,6 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-inf-size-3.yaml b/tests/configs/infer-test-categorical-inf-size-3.yaml
index 014e53a4..6bdad176 100644
--- a/tests/configs/infer-test-categorical-inf-size-3.yaml
+++ b/tests/configs/infer-test-categorical-inf-size-3.yaml
@@ -19,7 +19,6 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 prediction_length: 3
 enforce_determinism: true
diff --git a/tests/configs/infer-test-categorical-multitarget.yaml b/tests/configs/infer-test-categorical-multitarget.yaml
index ec7064f4..9202b7ed 100644
--- a/tests/configs/infer-test-categorical-multitarget.yaml
+++ b/tests/configs/infer-test-categorical-multitarget.yaml
@@ -19,7 +19,6 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-categorical.yaml b/tests/configs/infer-test-categorical.yaml
index 72214fba..dafe686e 100644
--- a/tests/configs/infer-test-categorical.yaml
+++ b/tests/configs/infer-test-categorical.yaml
@@ -17,7 +17,6 @@ output_probabilities: true
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-distributed-parquet.yaml b/tests/configs/infer-test-distributed-parquet.yaml
index a7f2d423..52ce0a4e 100644
--- a/tests/configs/infer-test-distributed-parquet.yaml
+++ b/tests/configs/infer-test-distributed-parquet.yaml
@@ -18,7 +18,6 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/infer-test-distributed.yaml b/tests/configs/infer-test-distributed.yaml
index a89c17c8..61f43431 100644
--- a/tests/configs/infer-test-distributed.yaml
+++ b/tests/configs/infer-test-distributed.yaml
@@ -20,5 +20,4 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 2
diff --git a/tests/configs/infer-test-lazy.yaml b/tests/configs/infer-test-lazy.yaml
index 20bf3ec5..3429ecf5 100644
--- a/tests/configs/infer-test-lazy.yaml
+++ b/tests/configs/infer-test-lazy.yaml
@@ -18,5 +18,4 @@ output_probabilities: false
 map_to_id: true
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 2
diff --git a/tests/configs/infer-test-real-autoregression.yaml b/tests/configs/infer-test-real-autoregression.yaml
index 79896d4c..07d3b21f 100644
--- a/tests/configs/infer-test-real-autoregression.yaml
+++ b/tests/configs/infer-test-real-autoregression.yaml
@@ -17,7 +17,6 @@ output_probabilities: false
 map_to_id: false
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: false
 
diff --git a/tests/configs/infer-test-real.yaml b/tests/configs/infer-test-real.yaml
index e8610c0c..74b40e23 100644
--- a/tests/configs/infer-test-real.yaml
+++ b/tests/configs/infer-test-real.yaml
@@ -15,7 +15,6 @@ output_probabilities: false
 map_to_id: false
 device: cpu
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 enforce_determinism: true
 
diff --git a/tests/configs/preprocess-test-categorical-exact-pt.yaml b/tests/configs/preprocess-test-categorical-exact-pt.yaml
index 905eaf7e..0d6f82c1 100644
--- a/tests/configs/preprocess-test-categorical-exact-pt.yaml
+++ b/tests/configs/preprocess-test-categorical-exact-pt.yaml
@@ -7,7 +7,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-context_length: 5
+stored_width: 6
+future_capacity: 1
 stride_by_split:
 - 5
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-exact.yaml b/tests/configs/preprocess-test-categorical-exact.yaml
index d0cb1349..b448ef32 100644
--- a/tests/configs/preprocess-test-categorical-exact.yaml
+++ b/tests/configs/preprocess-test-categorical-exact.yaml
@@ -7,7 +7,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-context_length: 5
+stored_width: 6
+future_capacity: 1
 stride_by_split:
 - 5
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-interrupted.yaml b/tests/configs/preprocess-test-categorical-interrupted.yaml
index 16487d15..6d244e94 100644
--- a/tests/configs/preprocess-test-categorical-interrupted.yaml
+++ b/tests/configs/preprocess-test-categorical-interrupted.yaml
@@ -9,7 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-lookahead-0.yaml b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
index 520f4b8d..be13f25a 100644
--- a/tests/configs/preprocess-test-categorical-lookahead-0.yaml
+++ b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
@@ -9,8 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-context_length: 8
-max_lookahead: 0
+stored_width: 8
+future_capacity: 0
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-multitarget.yaml b/tests/configs/preprocess-test-categorical-multitarget.yaml
index b90646c8..1fc95074 100644
--- a/tests/configs/preprocess-test-categorical-multitarget.yaml
+++ b/tests/configs/preprocess-test-categorical-multitarget.yaml
@@ -9,7 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
index 3a088619..24c12914 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
@@ -8,7 +8,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
index 4c730a1e..8e7fd362 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
@@ -7,7 +7,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical.yaml b/tests/configs/preprocess-test-categorical.yaml
index c9ff762b..7ca2774b 100644
--- a/tests/configs/preprocess-test-categorical.yaml
+++ b/tests/configs/preprocess-test-categorical.yaml
@@ -9,7 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-multi-file.yaml b/tests/configs/preprocess-test-multi-file.yaml
index 4ac995c6..a7792689 100644
--- a/tests/configs/preprocess-test-multi-file.yaml
+++ b/tests/configs/preprocess-test-multi-file.yaml
@@ -12,7 +12,8 @@ split_ratios:
 - 0.8
 - 0.1
 - 0.1
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-real.yaml b/tests/configs/preprocess-test-real.yaml
index 3265dccc..f0015e26 100644
--- a/tests/configs/preprocess-test-real.yaml
+++ b/tests/configs/preprocess-test-real.yaml
@@ -8,7 +8,8 @@ selected_columns: null
 split_ratios:
 - 0.5
 - 0.5
-context_length: 8
+stored_width: 9
+future_capacity: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/train-test-categorical-bert.yaml b/tests/configs/train-test-categorical-bert.yaml
index af929462..c6b4187c 100644
--- a/tests/configs/train-test-categorical-bert.yaml
+++ b/tests/configs/train-test-categorical-bert.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +32,6 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
-  sample_length: 9
   training_objective: bert
   bert_spec:
     masking_probability: 0.5
diff --git a/tests/configs/train-test-categorical-inf-size-1.yaml b/tests/configs/train-test-categorical-inf-size-1.yaml
index bc3a78f5..5c915af2 100644
--- a/tests/configs/train-test-categorical-inf-size-1.yaml
+++ b/tests/configs/train-test-categorical-inf-size-1.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -32,7 +31,6 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-inf-size-3.yaml b/tests/configs/train-test-categorical-inf-size-3.yaml
index 9e0e1ebc..93165300 100644
--- a/tests/configs/train-test-categorical-inf-size-3.yaml
+++ b/tests/configs/train-test-categorical-inf-size-3.yaml
@@ -11,7 +11,6 @@ target_column_types:
   supCat2: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -39,7 +38,6 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 10000.0
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-multitarget-eager.yaml b/tests/configs/train-test-categorical-multitarget-eager.yaml
index 4439f4e4..2dfbd2c5 100644
--- a/tests/configs/train-test-categorical-multitarget-eager.yaml
+++ b/tests/configs/train-test-categorical-multitarget-eager.yaml
@@ -10,7 +10,6 @@ target_column_types:
   supCat1: categorical
   supReal3: real
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -39,7 +38,6 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical-multitarget.yaml b/tests/configs/train-test-categorical-multitarget.yaml
index f7594939..4f1a83c1 100644
--- a/tests/configs/train-test-categorical-multitarget.yaml
+++ b/tests/configs/train-test-categorical-multitarget.yaml
@@ -11,7 +11,6 @@ target_column_types:
   supCat1: categorical
   supReal3: real
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -40,7 +39,6 @@ model_spec:
   rope_theta: 10000.0
 
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-categorical.yaml b/tests/configs/train-test-categorical.yaml
index 018226c2..e10b673f 100644
--- a/tests/configs/train-test-categorical.yaml
+++ b/tests/configs/train-test-categorical.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +32,6 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "outer"
diff --git a/tests/configs/train-test-distributed-lazy-parquet.yaml b/tests/configs/train-test-distributed-lazy-parquet.yaml
index ba63aaec..1e50b40f 100644
--- a/tests/configs/train-test-distributed-lazy-parquet.yaml
+++ b/tests/configs/train-test-distributed-lazy-parquet.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -26,7 +25,6 @@ model_spec:
   prediction_length: 1
 
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-distributed.yaml b/tests/configs/train-test-distributed.yaml
index 3c0bb02d..ea0efe24 100644
--- a/tests/configs/train-test-distributed.yaml
+++ b/tests/configs/train-test-distributed.yaml
@@ -11,7 +11,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -28,7 +27,6 @@ model_spec:
   prediction_length: 1
 
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-lazy.yaml b/tests/configs/train-test-lazy.yaml
index 6367027d..348c1ada 100644
--- a/tests/configs/train-test-lazy.yaml
+++ b/tests/configs/train-test-lazy.yaml
@@ -11,7 +11,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 2
 
 export_generative_model: true
@@ -28,7 +27,6 @@ model_spec:
   prediction_length: 1
 
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-real-bert.yaml b/tests/configs/train-test-real-bert.yaml
index 4a8ef116..89ec8ad2 100644
--- a/tests/configs/train-test-real-bert.yaml
+++ b/tests/configs/train-test-real-bert.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemValue: real
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +32,6 @@ model_spec:
   n_kv_heads: 2
   rope_theta: 10000.0
 training_spec:
-  sample_length: 9
   training_objective: bert
   bert_spec:
     masking_probability: 0.5
diff --git a/tests/configs/train-test-real.yaml b/tests/configs/train-test-real.yaml
index 33a0c67d..e44db6a1 100644
--- a/tests/configs/train-test-real.yaml
+++ b/tests/configs/train-test-real.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemValue: real
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +32,6 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "inner"
diff --git a/tests/configs/train-test-resume-epoch.yaml b/tests/configs/train-test-resume-epoch.yaml
index f229fb1d..1ab8bc7f 100644
--- a/tests/configs/train-test-resume-epoch.yaml
+++ b/tests/configs/train-test-resume-epoch.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemValue: real
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: false
@@ -33,7 +32,6 @@ model_spec:
   rope_theta: 10000.0
   prediction_length: 1
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   epochs: 3
diff --git a/tests/configs/train-test-resume-mid-epoch.yaml b/tests/configs/train-test-resume-mid-epoch.yaml
index b43c8c2b..7f4e6c2c 100644
--- a/tests/configs/train-test-resume-mid-epoch.yaml
+++ b/tests/configs/train-test-resume-mid-epoch.yaml
@@ -9,7 +9,6 @@ target_column_types:
   itemId: categorical
 
 context_length: 8
-sample_length: 9
 inference_batch_size: 10
 
 export_generative_model: true
@@ -33,7 +32,6 @@ model_spec:
   n_kv_heads: 1
   rope_theta: 1000.0
 training_spec:
-  sample_length: 9
   training_objective: causal
   device: cpu
   torch_compile: "outer"
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index 991205fd..1beeeb51 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -40,8 +40,13 @@ def test_hp_search_bert_outputs(run_hp_search, project_root):
     assert generated_config["metadata_config_path"].endswith(
         "test-data-categorical-1-lookahead-0.json"
     )
-    assert generated_config["layout"]["max_lookahead"] == 0
-    assert generated_config["layout"]["sequence_layout_version"] == 2
+    assert generated_config["metadata_config_path"].endswith(
+        "test-data-categorical-1-lookahead-0.json"
+    )
+    assert "storage_layout" not in generated_config
+    assert "window_view" not in generated_config
+    assert generated_config["target_shift"] == 0
+    assert generated_config["training_spec"]["training_objective"] == "bert"
 
 
 def test_hp_search_bayesian_outputs(run_hp_search, project_root):
diff --git a/tests/integration/test_make.py b/tests/integration/test_make.py
index 471de7a0..47eaad7e 100644
--- a/tests/integration/test_make.py
+++ b/tests/integration/test_make.py
@@ -54,7 +54,7 @@ def adapt_configs(config_strings):
                 "selected_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "selected_columns: "
             )
             .replace("input_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "input_columns: ")
-            .replace("context_length: 48", "context_length: 10")
+            .replace("stored_width: 49", "stored_width: 11")
             .replace("max_rows: null", "max_rows: null\nn_cores: 1")
         )
 
@@ -88,7 +88,6 @@ def adapt_configs(config_strings):
             .replace("epochs: 10", "epochs: 3")
             .replace("device: cuda", "device: cpu")
             .replace("context_length: 48", "context_length: 10")
-            .replace("sample_length: 49", "sample_length: 11")
             .replace("total_steps: PLEASE FILL", "total_steps: 10000")
         )
 
@@ -119,7 +118,6 @@ def adapt_configs(config_strings):
             .replace("[EXAMPLE_INPUT_COLUMN_NAME]", "[itemId]")
             .replace("[EXAMPLE_TARGET_COLUMN_NAME]", "[itemId]")
             .replace("context_length: 48", "context_length: 10")
-            .replace("sample_length: 49", "sample_length: 11")
             .replace("autoregression: true", "autoregression: false")
             .replace("EXAMPLE_TARGET_COLUMN_NAME: real", "itemId: categorical")
         )
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index d3c0839b..0036de9c 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -4,7 +4,7 @@
 import onnxruntime
 
 from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import ModelWindowView, StoredWindowLayout
 from sequifier.train import TransformerModel
 
 
@@ -28,10 +28,15 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         real_columns=["real_col"],
         id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
         n_classes={"cat_col": 6},
-        layout=SequenceLayout(
+        storage_layout=StoredWindowLayout(
+            stored_width=context_length + 1,
+            future_capacity=1,
+            version=2,
+        ),
+        window_view=ModelWindowView(
             context_length=context_length,
-            max_lookahead=1,
-            sequence_layout_version=2,
+            objective="bert",
+            target_shift=0,
         ),
         inference_batch_size=inference_batch_size,
         seed=42,
@@ -130,10 +135,15 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
             "cat10": {"x": 3, "y": 4, "z": 5},
         },
         n_classes={"cat2": 7, "cat10": 6},
-        layout=SequenceLayout(
+        storage_layout=StoredWindowLayout(
+            stored_width=context_length + 1,
+            future_capacity=1,
+            version=2,
+        ),
+        window_view=ModelWindowView(
             context_length=context_length,
-            max_lookahead=1,
-            sequence_layout_version=2,
+            objective="causal",
+            target_shift=1,
         ),
         inference_batch_size=inference_batch_size,
         seed=42,
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 9eeb3069..67508669 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -31,16 +31,15 @@ def test_metadata_config(metadata_configs):
         "split_paths",
         "column_types",
         "selected_columns_statistics",
-        "context_length",
-        "max_lookahead",
-        "sample_length",
-        "sequence_layout_version",
+        "stored_width",
+        "future_capacity",
+        "stored_window_layout_version",
     ]
 
     for file_name, metadata_config in metadata_configs.items():
         print(f"Verifying metadata_config for: {file_name}")
         assert list(metadata_config.keys()) == expected_metadata_keys
-        assert metadata_config["sequence_layout_version"] == 2
+        assert metadata_config["stored_window_layout_version"] == 2
         assert metadata_config["special_token_ids"] == {
             "[unknown]": 0,
             "[other]": 1,
@@ -273,8 +272,8 @@ def test_preprocessed_data_categorical_lookahead_0(run_preprocessing, project_ro
     with open(metadata_path, "r") as f:
         metadata_config = json.load(f)
 
-    assert metadata_config["max_lookahead"] == 0
-    assert metadata_config["sample_length"] == metadata_config["context_length"]
+    assert metadata_config["future_capacity"] == 0
+    assert metadata_config["stored_width"] == 8
 
     for split in range(3):
         data = read_preprocessing_outputs(
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
index 03a72927..1ec4aee0 100644
--- a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
+++ b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
@@ -1,8 +1,7 @@
 {
-    "context_length": 8,
-    "max_lookahead": 1,
-    "sample_length": 9,
-    "sequence_layout_version": 2,
+    "stored_width": 9,
+    "future_capacity": 1,
+    "stored_window_layout_version": 2,
     "selected_columns": null,
     "data_columns": [
         "itemId"
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 3cbe5f33..24ae734b 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -5,24 +5,23 @@
 import pytest
 import torch
 
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import ModelWindowView, StoredWindowLayout
 from sequifier.io.sequifier_dataset_from_folder_parquet_lazy import (
     SequifierDatasetFromFolderParquetLazy,
 )
 
 CONTEXT_LENGTH = 2
-MAX_LOOKAHEAD = 1
-SAMPLE_LENGTH = CONTEXT_LENGTH + MAX_LOOKAHEAD
+FUTURE_CAPACITY = 1
+STORED_WIDTH = CONTEXT_LENGTH + FUTURE_CAPACITY
 
 
 def _folder_metadata(total_samples, batch_files):
     return {
         "total_samples": total_samples,
         "batch_files": batch_files,
-        "context_length": CONTEXT_LENGTH,
-        "max_lookahead": MAX_LOOKAHEAD,
-        "sample_length": SAMPLE_LENGTH,
-        "sequence_layout_version": 2,
+        "stored_width": STORED_WIDTH,
+        "future_capacity": FUTURE_CAPACITY,
+        "stored_window_layout_version": 2,
     }
 
 
@@ -31,10 +30,11 @@ def mock_config():
     config = MagicMock()
     config.project_root = "."
     config.seed = 42
-    config.layout = SequenceLayout(
-        context_length=CONTEXT_LENGTH,
-        max_lookahead=MAX_LOOKAHEAD,
-        sequence_layout_version=2,
+    config.storage_layout = StoredWindowLayout(
+        stored_width=STORED_WIDTH, future_capacity=FUTURE_CAPACITY, version=2
+    )
+    config.window_view = ModelWindowView(
+        context_length=CONTEXT_LENGTH, objective="causal", target_shift=1
     )
     config.column_types = {"item": "Float64"}
     config.training_spec.batch_size = 5
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 3eed7bcd..2b62d861 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -4,24 +4,23 @@
 import pytest
 import torch
 
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import ModelWindowView, StoredWindowLayout
 from sequifier.io.sequifier_dataset_from_folder_pt_lazy import (
     SequifierDatasetFromFolderPtLazy,
 )
 
 CONTEXT_LENGTH = 5
-MAX_LOOKAHEAD = 1
-SAMPLE_LENGTH = CONTEXT_LENGTH + MAX_LOOKAHEAD
+FUTURE_CAPACITY = 1
+STORED_WIDTH = CONTEXT_LENGTH + FUTURE_CAPACITY
 
 
 def _folder_metadata(total_samples, batch_files):
     return {
         "total_samples": total_samples,
         "batch_files": batch_files,
-        "context_length": CONTEXT_LENGTH,
-        "max_lookahead": MAX_LOOKAHEAD,
-        "sample_length": SAMPLE_LENGTH,
-        "sequence_layout_version": 2,
+        "stored_width": STORED_WIDTH,
+        "future_capacity": FUTURE_CAPACITY,
+        "stored_window_layout_version": 2,
     }
 
 
@@ -35,10 +34,11 @@ def mock_config(tmp_path):
     config.training_spec.sampling_strategy = "exact"
     config.training_spec.num_workers = 0
     config.seed = 42
-    config.layout = SequenceLayout(
-        context_length=CONTEXT_LENGTH,
-        max_lookahead=MAX_LOOKAHEAD,
-        sequence_layout_version=2,
+    config.storage_layout = StoredWindowLayout(
+        stored_width=STORED_WIDTH, future_capacity=FUTURE_CAPACITY, version=2
+    )
+    config.window_view = ModelWindowView(
+        context_length=CONTEXT_LENGTH, objective="causal", target_shift=1
     )
     config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index 2b9f6b0e..635669d1 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -4,11 +4,13 @@
 import torch
 
 from sequifier.helpers import (
+    ModelWindowView,
+    StoredWindowLayout,
     apply_bert_masking,
     build_valid_mask,
     construct_index_maps,
     numpy_to_pytorch,
-    slice_window,
+    resolve_window_view,
 )
 
 # ==========================================
@@ -88,16 +90,16 @@ def test_numpy_to_pytorch_shapes_and_shifting():
 
     column_types = {"A": torch.float32}
     all_columns = ["A"]
-    context_length = 3
+    resolved_view = resolve_window_view(
+        StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        ModelWindowView(context_length=3, objective="causal", target_shift=1),
+    )
 
     tensors, metadata = numpy_to_pytorch(
         data,
         column_types,
         all_columns,
-        context_length,
-        context_length + 1,
-        data_offset=1,
-        target_offset=0,
+        resolved_view,
     )
 
     # 1. Check Keys
@@ -149,10 +151,10 @@ def test_numpy_to_pytorch_dtypes():
         data_int,
         {"int_col": torch.int64},
         ["int_col"],
-        context_length=1,
-        sample_length=2,
-        data_offset=1,
-        target_offset=0,
+        resolve_window_view(
+            StoredWindowLayout(stored_width=2, future_capacity=1, version=2),
+            ModelWindowView(context_length=1, objective="causal", target_shift=1),
+        ),
     )
     assert tensors_int["int_col"].dtype == torch.int64
 
@@ -172,10 +174,10 @@ def test_numpy_to_pytorch_dtypes():
         data_float,
         {"float_col": torch.float32},
         ["float_col"],
-        context_length=1,
-        sample_length=2,
-        data_offset=1,
-        target_offset=0,
+        resolve_window_view(
+            StoredWindowLayout(stored_width=2, future_capacity=1, version=2),
+            ModelWindowView(context_length=1, objective="causal", target_shift=1),
+        ),
     )
     assert tensors_float["float_col"].dtype == torch.float32
 
@@ -183,11 +185,15 @@ def test_numpy_to_pytorch_dtypes():
 def test_build_valid_mask_from_left_pad_lengths():
     left_pad_lengths = torch.tensor([0, 2, 5], dtype=torch.int64)
 
+    resolved_view = resolve_window_view(
+        StoredWindowLayout(stored_width=6, future_capacity=1, version=2),
+        ModelWindowView(context_length=5, objective="causal", target_shift=1),
+    )
     input_mask = build_valid_mask(
-        left_pad_lengths, full_length=6, offset=1, context_length=5
+        left_pad_lengths, full_length=6, view_slice=resolved_view.input_slice
     )
     target_mask = build_valid_mask(
-        left_pad_lengths, full_length=6, offset=0, context_length=5
+        left_pad_lengths, full_length=6, view_slice=resolved_view.target_slice
     )
 
     assert torch.equal(
@@ -212,15 +218,20 @@ def test_build_valid_mask_from_left_pad_lengths():
     )
 
 
-def test_slice_window_validates_stored_width():
+def test_resolve_window_view_slices_tensors():
     tensor = torch.arange(8).reshape(2, 4)
+    resolved_view = resolve_window_view(
+        StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        ModelWindowView(context_length=3, objective="causal", target_shift=1),
+    )
 
     assert torch.equal(
-        slice_window(tensor, context_length=3, offset=1),
+        tensor[:, resolved_view.input_slice],
         torch.tensor([[0, 1, 2], [4, 5, 6]]),
     )
     assert torch.equal(
-        slice_window(tensor[:, :3], context_length=3, offset=0), tensor[:, :3]
+        tensor[:, resolved_view.target_slice],
+        torch.tensor([[1, 2, 3], [5, 6, 7]]),
     )
 
 
@@ -243,10 +254,10 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
         data,
         {"A": torch.float32},
         ["A"],
-        context_length=3,
-        sample_length=4,
-        data_offset=1,
-        target_offset=0,
+        resolve_window_view(
+            StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+            ModelWindowView(context_length=3, objective="causal", target_shift=1),
+        ),
     )
 
     assert torch.equal(
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 97ba2fb8..92f86dc3 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -8,7 +8,7 @@
 import yaml
 
 from sequifier.config.infer_config import InfererModel, load_inferer_config
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import ModelWindowView, StoredWindowLayout
 from sequifier.infer import (
     Inferer,
     calculate_item_positions,
@@ -174,16 +174,15 @@ def test_infer_config_defaults_bert_prediction_length_to_context_length():
         seed=42,
         device="cpu",
         prediction_length=None,
-        layout=SequenceLayout(
-            context_length=3, max_lookahead=1, sequence_layout_version=2
-        ),
+        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        window_view=ModelWindowView(context_length=3, objective="bert", target_shift=0),
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
         autoregression=False,
     )
 
-    assert config.prediction_length == config.layout.context_length
+    assert config.prediction_length == config.window_view.context_length
 
 
 def test_infer_config_defaults_causal_prediction_length_to_one():
@@ -203,8 +202,9 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
         seed=42,
         device="cpu",
         prediction_length=None,
-        layout=SequenceLayout(
-            context_length=3, max_lookahead=1, sequence_layout_version=2
+        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        window_view=ModelWindowView(
+            context_length=3, objective="causal", target_shift=1
         ),
         inference_batch_size=2,
         output_probabilities=False,
@@ -233,8 +233,11 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             seed=42,
             device="cpu",
             prediction_length=1,
-            layout=SequenceLayout(
-                context_length=3, max_lookahead=1, sequence_layout_version=2
+            storage_layout=StoredWindowLayout(
+                stored_width=4, future_capacity=1, version=2
+            ),
+            window_view=ModelWindowView(
+                context_length=3, objective="bert", target_shift=0
             ),
             inference_batch_size=2,
             output_probabilities=False,
@@ -253,10 +256,9 @@ def test_load_inferer_config_rejects_mismatched_metadata_special_token_ids(tmp_p
         json.dumps(
             {
                 "split_paths": [str(data_path)],
-                "context_length": 3,
-                "max_lookahead": 1,
-                "sample_length": 4,
-                "sequence_layout_version": 2,
+                "stored_width": 4,
+                "future_capacity": 1,
+                "stored_window_layout_version": 2,
                 "column_types": {"target_col": "int64"},
                 "id_maps": {"target_col": {"A": 3}},
                 "selected_columns_statistics": {},
@@ -399,9 +401,8 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         seed=42,
         device="cpu",
         prediction_length=None,
-        layout=SequenceLayout(
-            context_length=3, max_lookahead=1, sequence_layout_version=2
-        ),
+        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        window_view=ModelWindowView(context_length=3, objective="bert", target_shift=0),
         inference_batch_size=4,
         output_probabilities=False,
         map_to_id=True,
@@ -616,8 +617,9 @@ def ar_config():
         seed=42,
         device="cpu",
         prediction_length=1,
-        layout=SequenceLayout(
-            context_length=3, max_lookahead=1, sequence_layout_version=2
+        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        window_view=ModelWindowView(
+            context_length=3, objective="causal", target_shift=1
         ),
         inference_batch_size=2,
         output_probabilities=False,
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index a90ae1c8..eb2a6281 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -7,7 +7,7 @@
 from pydantic import ValidationError
 
 from sequifier.config.preprocess_config import PreprocessorModel
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import StoredWindowLayout
 from sequifier.preprocess import (
     Preprocessor,
     _apply_column_statistics,
@@ -61,7 +61,7 @@ def test_extract_subsequences_bert_width():
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        sample_length=context_length,
+        stored_width=context_length,
         stride_for_split=stride,
         columns=columns,
         subsequence_start_mode="distribute",
@@ -75,14 +75,14 @@ def test_extract_subsequences_bert_width():
 def test_extract_subsequences_padding():
     """Tests that sequences shorter than context_length are padded with 0s."""
     input_data = {"col1": [1, 2]}  # Length 2
-    sample_length = 5
+    stored_width = 5
     stride = 1
     columns = ["col1"]
 
     # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data, sample_length, stride, columns, subsequence_start_mode="distribute"
+        input_data, stored_width, stride, columns, subsequence_start_mode="distribute"
     )
 
     assert len(result["col1"]) == 1
@@ -93,7 +93,7 @@ def test_extract_subsequences_returns_left_pad_lengths_when_requested():
     input_data = {"col1": [0.0, 1.5]}
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        sample_length=5,
+        stored_width=5,
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -127,7 +127,7 @@ def test_extract_sequences_persists_left_pad_length_metadata():
     sequences = extract_sequences(
         data,
         schema,
-        layout=SequenceLayout(4, 1, 2),
+        layout=StoredWindowLayout(stored_width=5, future_capacity=1, version=2),
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -154,7 +154,7 @@ def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
 
     process_and_write_data_pt(
         data,
-        sample_length=4,
+        stored_width=4,
         path=str(out_path),
         column_types={"col1": "Float64"},
     )
@@ -218,10 +218,9 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
                 "selected_columns_statistics": {
                     "itemValue": {"mean": 60.0, "std": 10.0}
                 },
-                "context_length": 2,
-                "max_lookahead": 1,
-                "sample_length": 3,
-                "sequence_layout_version": 2,
+                "stored_width": 3,
+                "future_capacity": 1,
+                "stored_window_layout_version": 2,
                 "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
             }
         )
@@ -246,7 +245,8 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
         merge_output=True,
         selected_columns=["itemId", "itemValue"],
         split_ratios=[1.0],
-        context_length=2,
+        stored_width=3,
+        future_capacity=1,
         stride_by_split=[1],
         max_rows=None,
         seed=1010,
@@ -383,7 +383,8 @@ def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
             merge_output=True,
             selected_columns=["itemId"],
             split_ratios=[1.0],
-            context_length=1,
+            stored_width=2,
+            future_capacity=1,
             stride_by_split=[1],
             max_rows=None,
             seed=1010,
@@ -412,7 +413,8 @@ def test_preprocessor_config_defaults_mask_column_to_none(tmp_path):
         project_root=str(tmp_path),
         data_path=str(data_path),
         split_ratios=[1.0],
-        context_length=1,
+        stored_width=2,
+        future_capacity=1,
         seed=1010,
     )
 
@@ -440,7 +442,8 @@ def test_preprocessor_config_requires_metadata_config_for_mask_column(
             project_root=str(tmp_path),
             data_path=str(data_path),
             split_ratios=[1.0],
-            context_length=1,
+            stored_width=2,
+            future_capacity=1,
             seed=1010,
             mask_column=RESERVED_MASK_COLUMN,
         )
@@ -454,14 +457,14 @@ def test_extract_subsequences_modes(mode):
     # exact: strictly adheres to stride, throws error if misalignment.
     input_data = {"col1": list(range(10))}
     context_length = 2
-    sample_length = context_length + 1
+    stored_width = context_length + 1
     columns = ["col1"]
 
     if mode == "distribute":
         stride = 4
         # distribute might adjust indices to maximize coverage
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data, sample_length, stride, columns, mode
+            input_data, stored_width, stride, columns, mode
         )
         assert len(result["col1"]) > 0
 
@@ -471,13 +474,13 @@ def test_extract_subsequences_modes(mode):
         )
         # Testing a failing exact case
         with pytest.raises(ValueError):
-            extract_subsequences(input_data, sample_length, 4, columns, mode)
+            extract_subsequences(input_data, stored_width, 4, columns, mode)
 
         # Testing a passing exact case
         # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
         input_data_exact = {"col1": list(range(11))}
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data_exact, sample_length, 4, columns, mode
+            input_data_exact, stored_width, 4, columns, mode
         )
         assert len(result["col1"]) > 0
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index f6bedbc4..42be51f8 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -15,7 +15,7 @@
     TrainModel,
     load_train_config,
 )
-from sequifier.helpers import SequenceLayout
+from sequifier.helpers import ModelWindowView, StoredWindowLayout
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS
 from sequifier.train import TransformerModel
 
@@ -77,7 +77,7 @@ def test_training_spec_model_dump_excludes_runtime_offsets():
 
     assert "data_offset" not in dumped
     assert "target_offset" not in dumped
-    assert "sample_length" not in dumped
+    assert "stored_width" not in dumped
     assert "max_lookahead" not in dumped
 
 
@@ -136,8 +136,11 @@ def model_config(tmp_path):
         # id_maps is needed for constructing index_maps in model init
         id_maps={"cat_col": {"a": 1, "b": 2, "c": 3, "d": 4}},
         n_classes={"cat_col": 5},  # 0 + 4 classes
-        layout=SequenceLayout(
-            context_length=10, max_lookahead=1, sequence_layout_version=2
+        storage_layout=StoredWindowLayout(
+            stored_width=11, future_capacity=1, version=2
+        ),
+        window_view=ModelWindowView(
+            context_length=10, objective="causal", target_shift=1
         ),
         inference_batch_size=4,
         seed=42,
@@ -166,12 +169,17 @@ def causal_model(model_config):
 def bert_model(model_config):
     config_values = model_config.model_dump()
     config_values["model_spec"]["prediction_length"] = (
-        model_config.layout.context_length
+        model_config.window_view.context_length
     )
     config_values["training_spec"] = _training_spec_kwargs(
         training_objective="bert",
         bert_spec=_bert_spec(),
     )
+    config_values["window_view"] = {
+        **config_values["window_view"],
+        "objective": "bert",
+        "target_shift": 0,
+    }
     config = TrainModel(**config_values)
     return TransformerModel(config)
 
@@ -221,12 +229,17 @@ def test_train_model_requires_bert_prediction_length_to_equal_context_length(
 ):
     config_values = model_config.model_dump()
     config_values["model_spec"]["prediction_length"] = (
-        model_config.layout.context_length - 1
+        model_config.window_view.context_length - 1
     )
     config_values["training_spec"] = _training_spec_kwargs(
         training_objective="bert",
         bert_spec=_bert_spec(),
     )
+    config_values["window_view"] = {
+        **config_values["window_view"],
+        "objective": "bert",
+        "target_shift": 0,
+    }
 
     with pytest.raises(ValidationError, match="prediction_length must be equal"):
         TrainModel(**config_values)
@@ -252,16 +265,17 @@ def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
     config_values = model_config.model_dump()
     config_values["project_root"] = str(tmp_path)
     config_values["metadata_config_path"] = metadata_path.name
-    layout = config_values["layout"]
+    storage_layout = config_values.pop("storage_layout")
+    config_values.pop("window_view")
+    config_values["context_length"] = model_config.window_view.context_length
     config_path.write_text(yaml.safe_dump(config_values))
     metadata_path.write_text(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
-                "context_length": layout["context_length"],
-                "max_lookahead": layout["max_lookahead"],
-                "sample_length": layout["context_length"] + layout["max_lookahead"],
-                "sequence_layout_version": layout["sequence_layout_version"],
+                "stored_width": storage_layout["stored_width"],
+                "future_capacity": storage_layout["future_capacity"],
+                "stored_window_layout_version": storage_layout["version"],
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
@@ -286,16 +300,17 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
     config_values = model_config.model_dump()
     config_values["project_root"] = str(tmp_path)
     config_values["metadata_config_path"] = metadata_path.name
-    layout = config_values["layout"]
+    storage_layout = config_values.pop("storage_layout")
+    config_values.pop("window_view")
+    config_values["context_length"] = model_config.window_view.context_length
     config_path.write_text(yaml.safe_dump(config_values))
     metadata_path.write_text(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
-                "context_length": layout["context_length"],
-                "max_lookahead": layout["max_lookahead"],
-                "sample_length": layout["context_length"] + layout["max_lookahead"],
-                "sequence_layout_version": layout["sequence_layout_version"],
+                "stored_width": storage_layout["stored_width"],
+                "future_capacity": storage_layout["future_capacity"],
+                "stored_window_layout_version": storage_layout["version"],
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
@@ -312,7 +327,7 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
 def test_forward_train_shapes(model, model_config):
     """Tests the output shapes of the forward_train method."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.layout.context_length
+    seq_len = model_config.window_view.context_length
 
     # Create dummy inputs
     # Categorical: (batch, seq_len) integers
@@ -343,7 +358,7 @@ def test_forward_train_shapes(model, model_config):
 def test_forward_inference_shapes(model, model_config):
     """Tests the output shapes of the forward (inference) method."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.layout.context_length
+    seq_len = model_config.window_view.context_length
     prediction_length = model_config.model_spec.prediction_length  # 1
 
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
@@ -376,7 +391,7 @@ def test_forward_inference_shapes(model, model_config):
 def test_calculate_loss(model, model_config):
     """Tests that loss calculation returns a scalar tensor."""
     batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.layout.context_length
+    seq_len = model_config.window_view.context_length
 
     # Inputs
     x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
@@ -465,7 +480,7 @@ def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
 
 
 def test_padding_keys_are_masked(bert_model):
-    seq_len = bert_model.layout.context_length
+    seq_len = bert_model.window_view.context_length
 
     valid_mask = torch.ones(
         2,
@@ -494,7 +509,7 @@ def test_padding_keys_are_masked(bert_model):
 
 
 def test_causal_and_padding_masks_are_combined(causal_model):
-    seq_len = causal_model.layout.context_length
+    seq_len = causal_model.window_view.context_length
 
     valid_mask = torch.ones(
         1,

From 6da8f270d456f503b8712f0a436815d8a0db5bbc Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 15:42:02 +0200
Subject: [PATCH 52/81] Rename params

---
 README.md                                     |  2 +-
 documentation/configs/preprocess.md           |  8 +-
 documentation/configs/train.md                |  2 +-
 documentation/consolidated-docs.md            | 12 +--
 .../config/hyperparameter_search_config.py    | 10 +--
 src/sequifier/config/infer_config.py          | 12 +--
 src/sequifier/config/preprocess_config.py     | 16 ++--
 src/sequifier/config/train_config.py          | 20 +++--
 src/sequifier/helpers.py                      | 75 +++++++++--------
 src/sequifier/hyperparameter_search.py        |  2 +-
 src/sequifier/infer.py                        | 10 ++-
 .../sequifier_dataset_from_folder_parquet.py  |  4 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  4 +-
 .../io/sequifier_dataset_from_folder_pt.py    |  2 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  4 +-
 src/sequifier/make.py                         |  4 +-
 src/sequifier/preprocess.py                   | 82 +++++++++++--------
 .../preprocess-test-categorical-exact-pt.yaml |  4 +-
 .../preprocess-test-categorical-exact.yaml    |  4 +-
 ...eprocess-test-categorical-interrupted.yaml |  4 +-
 ...eprocess-test-categorical-lookahead-0.yaml |  4 +-
 ...eprocess-test-categorical-multitarget.yaml |  4 +-
 ...ategorical-precomputed-stats-negative.yaml |  4 +-
 ...ss-test-categorical-precomputed-stats.yaml |  4 +-
 .../configs/preprocess-test-categorical.yaml  |  4 +-
 tests/configs/preprocess-test-multi-file.yaml |  4 +-
 tests/configs/preprocess-test-real.yaml       |  4 +-
 .../integration/test_hyperparameter_search.py |  2 +-
 tests/integration/test_make.py                |  2 +-
 tests/integration/test_onnx_export.py         | 12 +--
 tests/integration/test_preprocessing.py       |  8 +-
 .../preprocess-manifest.json                  |  4 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  8 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  8 +-
 tests/unit/test_helpers.py                    | 24 +++---
 tests/unit/test_infer.py                      | 36 +++++---
 tests/unit/test_preprocess.py                 | 46 ++++++-----
 tests/unit/test_train.py                      | 20 ++---
 38 files changed, 264 insertions(+), 215 deletions(-)

diff --git a/README.md b/README.md
index 183d05bd..10cb0085 100644
--- a/README.md
+++ b/README.md
@@ -97,7 +97,7 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Preprocessing defines the physical `stored_width` and `future_capacity`; training and inference choose the model-facing `context_length` from that stored capacity:
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Preprocessing defines the physical `stored_context_width` and `max_target_offset`; training and inference choose the model-facing `context_length` from that stored capacity:
 
 |sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
 |----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index a9ee943e..8837d3f0 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -45,10 +45,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
-| `stored_width` | `int` | **Yes** | - | The physical serialized window width written to preprocessed data. |
-| `future_capacity` | `int` | No | `1` | Number of future items retained after the model input window. Use `0` for BERT-style same-width inputs and targets; use `1` for causal next-item training. |
+| `stored_context_width` | `int` | **Yes** | - | The physical serialized window width written to preprocessed data. |
+| `max_target_offset` | `int` | No | `1` | Number of future items retained after the model input window. Use `0` for BERT-style same-width inputs and targets; use `1` for causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
-| `stride_by_split` | `list[int]` | No | `[stored_width]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
+| `stride_by_split` | `list[int]` | No | `[stored_context_width]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
 
 ### 4\. Performance & System
@@ -76,7 +76,7 @@ This controls data augmentation and redundancy.
   * **Stride = `context_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
   * **Stride = 1 (Maximum Overlap):** Maximizes data volume. The model sees every possible sequence. This yields the highest accuracy but significantly increases the size of the preprocessed data and training time.
   * **Hybrid Approach:** It is common practice to set a large stride for the training and validation splits (index 0) to reduce the size on disk of the dataset, and a stride=1 for the test split to evaluate the model on each point in the test set. This supposes that the test split value is low.
-      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `stored_width: 49`).
+      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `stored_context_width: 49`).
 
 ### 3\. `subsequence_start_mode`: `distribute` vs `exact`
 
diff --git a/documentation/configs/train.md b/documentation/configs/train.md
index 4a66e7ca..63d986d1 100644
--- a/documentation/configs/train.md
+++ b/documentation/configs/train.md
@@ -30,7 +30,7 @@ The configuration is defined in a YAML file (e.g., `train.yaml`). The file is st
 | `target_columns` | `list[str]`| **Yes** | - | The specific column(s) the model should learn to predict. |
 | `target_column_types`| `dict` | **Yes** | - | Map of target columns to their type: `'categorical'` or `'real'`. The key order in target_column_types must exactly match the list order in target_columns |
 | `input_columns` | `list[str]`| No | All | Subset of columns to use as input features. Defaults to all available in metadata. |
-| `context_length` | `int` | **Yes** | - | Model input context length. It must fit inside the metadata `stored_width` with the stored `future_capacity`. |
+| `context_length` | `int` | **Yes** | - | Model input context length. It must fit inside the metadata `stored_context_width` with the stored `max_target_offset`. |
 
 ### 3\. Model Architecture (`model_spec`)
 
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index d8a0b1a6..c02aa5bf 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -97,7 +97,7 @@ Let's start with the data format expected by sequifier. The basic data format th
 
 The two columns "sequenceId" and "itemPosition" have to be present, and then there must be at least one feature column. There can also be many feature columns, and these can be categorical or real valued.
 
-Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Preprocessing defines the physical `stored_width` and `future_capacity`; training and inference choose the model-facing `context_length` from that stored capacity:
+Data of this input format can be transformed into the format that is used for model training and inference using `sequifier preprocess`. Preprocessing defines the physical `stored_context_width` and `max_target_offset`; training and inference choose the model-facing `context_length` from that stored capacity:
 
 |sequenceId|subsequenceId|startItemPosition|leftPadLength|inputCol|[Window Length - 1]|[Window Length - 2]|...|0|
 |----------|-------------|-----------------|-------------|--------|-------------------|-------------------| - |-|
@@ -253,10 +253,10 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
-| `stored_width` | `int` | **Yes** | - | The physical serialized window width written to preprocessed data. |
-| `future_capacity` | `int` | No | `1` | Number of future items retained after the model input window. Use `0` for BERT-style same-width inputs and targets; use `1` for causal next-item training. |
+| `stored_context_width` | `int` | **Yes** | - | The physical serialized window width written to preprocessed data. |
+| `max_target_offset` | `int` | No | `1` | Number of future items retained after the model input window. Use `0` for BERT-style same-width inputs and targets; use `1` for causal next-item training. |
 | `split_ratios` | `list[float]`| **Yes** | - | Proportions for data splits (e.g., `[0.8, 0.1, 0.1]` for train/val/test). Must sum to 1.0. |
-| `stride_by_split` | `list[int]` | No | `[stored_width]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
+| `stride_by_split` | `list[int]` | No | `[stored_context_width]*N` | The step size used to slide the window for each split. Corresponds to `split_ratios`. |
 | `subsequence_start_mode`| `str` | No | `distribute` | Strategy for selecting start indices (`distribute` or `exact`). |
 
 ### 4\. Performance & System
@@ -284,7 +284,7 @@ This controls data augmentation and redundancy.
   * **Stride = `context_length` (Non-overlapping):** The model sees every data point exactly once as a target. Training is faster, but the model might miss patterns that cross the window boundary.
   * **Stride = 1 (Maximum Overlap):** Maximizes data volume. The model sees every possible sequence. This yields the highest accuracy but significantly increases the size of the preprocessed data and training time.
   * **Hybrid Approach:** It is common practice to set a large stride for the training and validation splits (index 0) to reduce the size on disk of the dataset, and a stride=1 for the test split to evaluate the model on each point in the test set. This supposes that the test split value is low.
-      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `stored_width: 49`).
+      * *Example:* `stride_by_split: [24, 24, 1]` (assuming `stored_context_width: 49`).
 
 ### 3\. `subsequence_start_mode`: `distribute` vs `exact`
 
@@ -379,7 +379,7 @@ The configuration is defined in a YAML file (e.g., `train.yaml`). The file is st
 | `target_columns` | `list[str]`| **Yes** | - | The specific column(s) the model should learn to predict. |
 | `target_column_types`| `dict` | **Yes** | - | Map of target columns to their type: `'categorical'` or `'real'`. The key order in target_column_types must exactly match the list order in target_columns |
 | `input_columns` | `list[str]`| No | All | Subset of columns to use as input features. Defaults to all available in metadata. |
-| `context_length` | `int` | **Yes** | - | Model input context length. It must fit inside the metadata `stored_width` with the stored `future_capacity`. |
+| `context_length` | `int` | **Yes** | - | Model input context length. It must fit inside the metadata `stored_context_width` with the stored `max_target_offset`. |
 
 ### 3\. Model Architecture (`model_spec`)
 
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index f097e507..b93c8ea5 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -740,12 +740,12 @@ class HyperparameterSearchConfig(BaseModel):
     def validate_sequence_layout(self):
         for cl in self.context_length:
             if (
-                cl + self.storage_layout.future_capacity
-                > self.storage_layout.stored_width
+                cl + self.storage_layout.max_target_offset
+                > self.storage_layout.stored_context_width
             ):
                 raise ValueError(
-                    f"Window capacity mismatch: context_length ({cl}) + future_capacity "
-                    f"({self.storage_layout.future_capacity}) > stored_width ({self.storage_layout.stored_width}). "
+                    f"Window capacity mismatch: context_length ({cl}) + max_target_offset "
+                    f"({self.storage_layout.max_target_offset}) > stored_context_width ({self.storage_layout.stored_context_width}). "
                     "Model inputs cannot exceed the preprocessed sequence length."
                 )
         return self
@@ -859,7 +859,7 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
         window_view = ModelWindowView(
             context_length=context_length,
             objective=training_spec.training_objective,
-            target_shift=0 if training_spec.training_objective == "bert" else 1,
+            target_offset=0 if training_spec.training_objective == "bert" else 1,
         )
         resolve_window_view(self.storage_layout, window_view)
 
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 77deaee1..84fa6e8e 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -65,23 +65,23 @@ def load_inferer_config(
                 f"got {storage_layout.version}."
             )
         training_objective = config_values["training_objective"]
-        target_shift = (
+        target_offset = (
             0
             if training_objective == "bert"
-            else int(config_values.pop("target_shift", 1))
+            else int(config_values.pop("target_offset", 1))
         )
         window_view = ModelWindowView(
             context_length=int(config_values.pop("context_length")),
             objective=training_objective,
-            target_shift=target_shift,
+            target_offset=target_offset,
         )
         resolve_window_view(storage_layout, window_view)
         config_values["storage_layout"] = storage_layout
         config_values["window_view"] = window_view
         for key in (
-            "target_shift",
-            "stored_width",
-            "future_capacity",
+            "target_offset",
+            "stored_context_width",
+            "max_target_offset",
             "stored_window_layout_version",
         ):
             config_values.pop(key, None)
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index da53eb8b..5745ead0 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -54,8 +54,8 @@ class PreprocessorModel(BaseModel):
         selected_columns: A list of columns to be included in the preprocessing. If None, all columns are used.
         split_ratios: A list of floats that define the relative sizes of data splits (e.g., for train, validation, test).
                            The sum of proportions must be 1.0.
-        stored_width: The physical serialized window width.
-        future_capacity: The number of future items retained after the input window.
+        stored_context_width: The physical serialized window width.
+        max_target_offset: The number of future items retained after the input window.
         stride_by_split: A list of step sizes for creating subsequences within each data split.
         max_rows: The maximum number of input rows to process. If None, all rows are processed.
         seed: A random seed for reproducibility.
@@ -78,8 +78,8 @@ class PreprocessorModel(BaseModel):
     selected_columns: Optional[list[str]] = None
 
     split_ratios: list[float]
-    stored_width: int = Field(gt=0)
-    future_capacity: int = Field(default=1, ge=0)
+    stored_context_width: int = Field(gt=0)
+    max_target_offset: int = Field(default=1, ge=0)
     stride_by_split: Optional[list[int]] = None
     max_rows: Optional[int] = None
     seed: int
@@ -194,12 +194,14 @@ def validate_mask_column_requires_metadata(self) -> "PreprocessorModel":
             raise ValueError("metadata_config_path must be set when mask_column is set")
         if self.mask_column in ("sequenceId", "itemPosition"):
             raise ValueError("mask_column cannot be sequenceId or itemPosition")
-        if self.future_capacity >= self.stored_width:
-            raise ValueError("future_capacity must be smaller than stored_width")
+        if self.max_target_offset >= self.stored_context_width:
+            raise ValueError(
+                "max_target_offset must be smaller than stored_context_width"
+            )
         return self
 
     def __init__(self, **kwargs):
-        default_stride_for_split = [kwargs["stored_width"]] * len(
+        default_stride_for_split = [kwargs["stored_context_width"]] * len(
             kwargs["split_ratios"]
         )
         kwargs["stride_by_split"] = kwargs.get(
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 34409a69..b8886d71 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -72,29 +72,33 @@ def load_train_config(
                 f"got {storage_layout.version}."
             )
         training_objective = config_values["training_spec"]["training_objective"]
-        target_shift = (
+        target_offset = (
             0
             if training_objective == "bert"
-            else int(config_values.pop("target_shift", 1))
+            else int(config_values.pop("target_offset", 1))
         )
         window_view = ModelWindowView(
             context_length=int(config_values.pop("context_length")),
             objective=training_objective,
-            target_shift=target_shift,
+            target_offset=target_offset,
         )
         resolve_window_view(storage_layout, window_view)
         config_values["storage_layout"] = storage_layout
         config_values["window_view"] = window_view
-        for key in ("stored_width", "future_capacity", "stored_window_layout_version"):
+        for key in (
+            "stored_context_width",
+            "max_target_offset",
+            "stored_window_layout_version",
+        ):
             config_values.pop(key, None)
         for key in (
-            "target_shift",
-            "stored_width",
-            "future_capacity",
+            "target_offset",
+            "stored_context_width",
+            "max_target_offset",
             "stored_window_layout_version",
         ):
             config_values.pop(key, None)
-        for key in ("target_shift", "stored_width", "future_capacity"):
+        for key in ("target_offset", "stored_context_width", "max_target_offset"):
             config_values.get("training_spec", {}).pop(key, None)
 
         split_paths = metadata_config["split_paths"]
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index a94967f1..5f3a52fb 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -45,24 +45,26 @@
 
 @dataclass(frozen=True)
 class StoredWindowLayout:
-    stored_width: int
-    future_capacity: int
+    stored_context_width: int
+    max_target_offset: int
     version: int
 
     def __post_init__(self) -> None:
-        if self.stored_width < 1:
-            raise ValueError("stored_width must be a positive integer")
-        if self.future_capacity < 0:
-            raise ValueError("future_capacity must be non-negative")
-        if self.future_capacity >= self.stored_width:
-            raise ValueError("future_capacity must be smaller than stored_width")
+        if self.stored_context_width < 1:
+            raise ValueError("stored_context_width must be a positive integer")
+        if self.max_target_offset < 0:
+            raise ValueError("max_target_offset must be non-negative")
+        if self.max_target_offset >= self.stored_context_width:
+            raise ValueError(
+                "max_target_offset must be smaller than stored_context_width"
+            )
 
 
 @dataclass(frozen=True)
 class ModelWindowView:
     context_length: int
     objective: str
-    target_shift: int = 1
+    target_offset: int
 
     def __post_init__(self) -> None:
         if self.context_length < 1:
@@ -71,12 +73,12 @@ def __post_init__(self) -> None:
             raise ValueError(
                 f"Only 'causal' and 'bert' are allowed, found {self.objective}"
             )
-        if self.target_shift < 0:
-            raise ValueError("target_shift must be non-negative")
-        if self.objective == "bert" and self.target_shift != 0:
-            raise ValueError("BERT views require target_shift=0")
-        if self.objective == "causal" and self.target_shift < 1:
-            raise ValueError("Causal views require target_shift >= 1")
+        if self.target_offset < 0:
+            raise ValueError("target_offset must be non-negative")
+        if self.objective == "bert" and self.target_offset != 0:
+            raise ValueError("BERT views require target_offset=0")
+        if self.objective == "causal" and self.target_offset < 1:
+            raise ValueError("Causal views require target_offset >= 1")
 
 
 @dataclass(frozen=True)
@@ -91,10 +93,10 @@ def build_masks(self, left_pad_lengths: Tensor) -> dict[str, Tensor]:
         """Build explicit input-attention and target-validity masks for this view."""
         return {
             "attention_valid_mask": build_valid_mask(
-                left_pad_lengths, self.storage.stored_width, self.input_slice
+                left_pad_lengths, self.storage.stored_context_width, self.input_slice
             ),
             "target_valid_mask": build_valid_mask(
-                left_pad_lengths, self.storage.stored_width, self.target_slice
+                left_pad_lengths, self.storage.stored_context_width, self.target_slice
             ),
         }
 
@@ -110,19 +112,19 @@ def _right_aligned_slice(width: int, length: int, offset: int) -> slice:
 def resolve_window_view(
     storage: StoredWindowLayout, view: ModelWindowView
 ) -> ResolvedWindowView:
-    if view.target_shift > storage.future_capacity:
+    if view.target_offset > storage.max_target_offset:
         raise ValueError(
-            f"Model target_shift={view.target_shift} exceeds stored "
-            f"future_capacity={storage.future_capacity}."
+            f"Model target_offset={view.target_offset} exceeds stored "
+            f"max_target_offset={storage.max_target_offset}."
         )
 
-    input_offset = storage.future_capacity
-    target_offset = storage.future_capacity - view.target_shift
+    input_offset = storage.max_target_offset
+    target_offset = storage.max_target_offset - view.target_offset
     required_width = view.context_length + max(input_offset, target_offset)
-    if required_width > storage.stored_width:
+    if required_width > storage.stored_context_width:
         raise ValueError(
             f"Model view requires width {required_width}, but storage only has "
-            f"stored_width={storage.stored_width}."
+            f"stored_context_width={storage.stored_context_width}."
         )
 
     return ResolvedWindowView(
@@ -130,28 +132,28 @@ def resolve_window_view(
         view=view,
         required_width=required_width,
         input_slice=_right_aligned_slice(
-            storage.stored_width, view.context_length, input_offset
+            storage.stored_context_width, view.context_length, input_offset
         ),
         target_slice=_right_aligned_slice(
-            storage.stored_width, view.context_length, target_offset
+            storage.stored_context_width, view.context_length, target_offset
         ),
     )
 
 
 @beartype
-def validate_stored_window_width(tensor: Tensor, stored_width: int) -> None:
-    if tensor.shape[1] != stored_width:
+def validate_stored_window_width(tensor: Tensor, stored_context_width: int) -> None:
+    if tensor.shape[1] != stored_context_width:
         raise ValueError(
             f"Stored window width {tensor.shape[1]} does not match "
-            f"metadata stored_width={stored_width}."
+            f"metadata stored_context_width={stored_context_width}."
         )
 
 
 @beartype
 def stored_window_layout_from_metadata(metadata: dict) -> StoredWindowLayout:
     return StoredWindowLayout(
-        stored_width=int(metadata["stored_width"]),
-        future_capacity=int(metadata["future_capacity"]),
+        stored_context_width=int(metadata["stored_context_width"]),
+        max_target_offset=int(metadata["max_target_offset"]),
         version=int(metadata["stored_window_layout_version"]),
     )
 
@@ -389,10 +391,10 @@ def numpy_to_pytorch(
         - a metadata dictionary containing any explicit masks.
     """
     input_seq_cols = columns_from_slice(
-        resolved_view.input_slice, resolved_view.storage.stored_width
+        resolved_view.input_slice, resolved_view.storage.stored_context_width
     )
     target_seq_cols = columns_from_slice(
-        resolved_view.target_slice, resolved_view.storage.stored_width
+        resolved_view.target_slice, resolved_view.storage.stored_context_width
     )
 
     # We will create a unified dictionary
@@ -441,10 +443,13 @@ def build_valid_mask(
 
 
 @beartype
-def columns_from_slice(view_slice: slice, stored_width: int) -> list[str]:
+def columns_from_slice(view_slice: slice, stored_context_width: int) -> list[str]:
     if view_slice.start is None or view_slice.stop is None:
         raise ValueError("Resolved window slices must have concrete bounds")
-    return [str(stored_width - 1 - i) for i in range(view_slice.start, view_slice.stop)]
+    return [
+        str(stored_context_width - 1 - i)
+        for i in range(view_slice.start, view_slice.stop)
+    ]
 
 
 @beartype
diff --git a/src/sequifier/hyperparameter_search.py b/src/sequifier/hyperparameter_search.py
index d39028cb..0fe83532 100644
--- a/src/sequifier/hyperparameter_search.py
+++ b/src/sequifier/hyperparameter_search.py
@@ -61,7 +61,7 @@ def objective(trial: optuna.Trial, config) -> Union[float, tuple[float, ...]]:
 
     run_config_dict = run_config.model_dump()
     run_config_dict["context_length"] = run_config_dict["window_view"]["context_length"]
-    run_config_dict["target_shift"] = run_config_dict["window_view"]["target_shift"]
+    run_config_dict["target_offset"] = run_config_dict["window_view"]["target_offset"]
 
     del run_config_dict["window_view"]
     del run_config_dict["storage_layout"]
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 804eca47..18a8c3ad 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -492,7 +492,9 @@ def infer_embedding(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.storage_layout.stored_width)
+                validate_stored_window_width(
+                    tensor, config.storage_layout.stored_context_width
+                )
 
             resolved_view = resolve_window_view(
                 config.storage_layout, config.window_view
@@ -746,7 +748,9 @@ def infer_generative(
                 left_pad_lengths_tensor,
             ) = data
             for tensor in sequences_dict.values():
-                validate_stored_window_width(tensor, config.storage_layout.stored_width)
+                validate_stored_window_width(
+                    tensor, config.storage_layout.stored_context_width
+                )
             total_steps = (
                 1
                 if config.autoregression_total_steps is None
@@ -946,7 +950,7 @@ def get_embeddings_pt(
     """
     resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     for tensor in data.values():
-        validate_stored_window_width(tensor, config.storage_layout.stored_width)
+        validate_stored_window_width(tensor, config.storage_layout.stored_context_width)
     X = {
         key: val[:, resolved_view.input_slice].numpy()
         for key, val in data.items()
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 24fe6e12..e1d060a7 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -65,10 +65,10 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
 
         # Sequence formatting structures matching long-format schema boundaries
         input_seq_cols = columns_from_slice(
-            self.resolved_view.input_slice, self.folder_layout.stored_width
+            self.resolved_view.input_slice, self.folder_layout.stored_context_width
         )
         target_seq_cols = columns_from_slice(
-            self.resolved_view.target_slice, self.folder_layout.stored_width
+            self.resolved_view.target_slice, self.folder_layout.stored_context_width
         )
         all_sequences: Dict[str, list[torch.Tensor]] = {
             col: [] for col in config.input_columns
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 7f2e5bcc..06ee64bb 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -218,10 +218,10 @@ def __iter__(
         global_file_start_sample = 0
 
         input_seq_cols = columns_from_slice(
-            self.resolved_view.input_slice, self.folder_layout.stored_width
+            self.resolved_view.input_slice, self.folder_layout.stored_context_width
         )
         target_seq_cols = columns_from_slice(
-            self.resolved_view.target_slice, self.folder_layout.stored_width
+            self.resolved_view.target_slice, self.folder_layout.stored_context_width
         )
 
         # Initialize cross-file buffers
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index e60ea981..a3f93d73 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -73,7 +73,7 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in all_sequences.keys():
                 if col in sequences_batch:
                     validate_stored_window_width(
-                        sequences_batch[col], self.folder_layout.stored_width
+                        sequences_batch[col], self.folder_layout.stored_context_width
                     )
                     all_sequences[col].append(sequences_batch[col])
             all_left_pad_lengths.append(left_pad_lengths_batch)
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index df67ee66..f11fe985 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -238,7 +238,9 @@ def __iter__(
                 left_pad_lengths_batch,
             ) = torch.load(file_path, map_location="cpu", weights_only=False)
             for tensor in sequences_batch.values():
-                validate_stored_window_width(tensor, self.folder_layout.stored_width)
+                validate_stored_window_width(
+                    tensor, self.folder_layout.stored_context_width
+                )
 
             # Generate indices for the whole file
             indices = torch.arange(file_samples)
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 79d5cc42..5b7da268 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -11,8 +11,8 @@
 - 0.8
 - 0.1
 - 0.1
-stored_width: 49
-future_capacity: 1
+stored_context_width: 49
+max_target_offset: 1
 stride_by_split:
 - 1
 - 1
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 4172ce18..9f8964f7 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -86,7 +86,7 @@ def __init__(
         merge_output: bool,
         selected_columns: Optional[list[str]],
         split_ratios: list[float],
-        stored_width: int,
+        stored_context_width: int,
         stride_by_split: list[int],
         max_rows: Optional[int],
         seed: int,
@@ -96,7 +96,7 @@ def __init__(
         subsequence_start_mode: str,
         use_precomputed_maps: Optional[list[str]],
         metadata_config_path: Optional[str],
-        future_capacity: int = 1,
+        max_target_offset: int = 1,
         mask_column: Optional[str] = None,
     ):
         """Initializes the Preprocessor with the given parameters.
@@ -109,8 +109,8 @@ def __init__(
             merge_output: Whether to combine the output into a single file.
             selected_columns: A list of columns to be included in the preprocessing.
             split_ratios: A list of floats that define the relative sizes of data splits.
-            stored_width: The physical serialized window width.
-            future_capacity: The number of future items retained after each input window.
+            stored_context_width: The physical serialized window width.
+            max_target_offset: The number of future items retained after each input window.
             stride_by_split: A list of step sizes for creating subsequences.
             max_rows: The maximum number of input rows to process.
             seed: A random seed for reproducibility.
@@ -146,8 +146,8 @@ def __init__(
         self.n_cores = n_cores or multiprocessing.cpu_count()
         self.continue_preprocessing = continue_preprocessing
         self.storage_layout = StoredWindowLayout(
-            stored_width=stored_width,
-            future_capacity=future_capacity,
+            stored_context_width=stored_context_width,
+            max_target_offset=max_target_offset,
             version=CURRENT_STORED_WINDOW_LAYOUT_VERSION,
         )
         self._setup_directories()
@@ -251,7 +251,9 @@ def __init__(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
 
-            schema = self._create_schema(col_types, self.storage_layout.stored_width)
+            schema = self._create_schema(
+                col_types, self.storage_layout.stored_context_width
+            )
 
             data = data.sort(["sequenceId", "itemPosition"])
             n_batches = _process_batches_single_file(
@@ -347,7 +349,9 @@ def __init__(
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
             )
-            schema = self._create_schema(col_types, self.storage_layout.stored_width)
+            schema = self._create_schema(
+                col_types, self.storage_layout.stored_context_width
+            )
 
             self._process_batches_multiple_files(
                 files_to_process,
@@ -373,7 +377,7 @@ def __init__(
 
     @beartype
     def _create_schema(
-        self, col_types: dict[str, str], stored_width: int
+        self, col_types: dict[str, str], stored_context_width: int
     ) -> dict[str, Any]:
         """Creates the Polars schema for the intermediate sequence DataFrame.
 
@@ -385,7 +389,7 @@ def _create_schema(
         Args:
             col_types: A dictionary mapping data column names to their Polars
                 string representations (e.g., "Int64", "Float64").
-            stored_width: The number of items stored for each extracted window.
+            stored_context_width: The number of items stored for each extracted window.
 
         Returns:
             A dictionary defining the Polars schema. Keys are column names
@@ -415,7 +419,10 @@ def _create_schema(
             sequence_position_type = pl.Float64
 
         schema.update(
-            {str(i): sequence_position_type for i in range(stored_width - 1, -1, -1)}
+            {
+                str(i): sequence_position_type
+                for i in range(stored_context_width - 1, -1, -1)
+            }
         )
 
         return schema
@@ -817,8 +824,8 @@ def _cleanup(self, write_format: str) -> None:
     @beartype
     def _layout_metadata(self) -> dict[str, int]:
         return {
-            "stored_width": self.storage_layout.stored_width,
-            "future_capacity": self.storage_layout.future_capacity,
+            "stored_context_width": self.storage_layout.stored_context_width,
+            "max_target_offset": self.storage_layout.max_target_offset,
             "stored_window_layout_version": self.storage_layout.version,
         }
 
@@ -1901,7 +1908,7 @@ def get_group_bounds(data_subset: pl.DataFrame, split_ratios: list[float]):
 @beartype
 def process_and_write_data_pt(
     data: pl.DataFrame,
-    stored_width: int,
+    stored_context_width: int,
     path: str,
     column_types: dict[str, str],
 ):
@@ -1914,13 +1921,13 @@ def process_and_write_data_pt(
 
     It then converts these lists into NumPy arrays and stores one full
     sequence tensor per feature. Each tensor has shape
-    `(batch_size, stored_width)`. The final five-element tuple
+    `(batch_size, stored_context_width)`. The final five-element tuple
     `(sequences_dict, sequence_ids_tensor, subsequence_ids_tensor, start_item_positions_tensor, left_pad_lengths_tensor)`
     is saved to a .pt file using `torch.save`.
 
     Args:
         data: The long-format Polars DataFrame of extracted sequences.
-        stored_width: The stored serialized window width.
+        stored_context_width: The stored serialized window width.
         path: The output file path (e.g., "data/batch_0.pt").
         column_types: A dictionary mapping column names to their
             string data types, used to determine the correct torch dtype.
@@ -1928,7 +1935,7 @@ def process_and_write_data_pt(
     if data.is_empty():
         return
 
-    sequence_cols = [str(c) for c in range(stored_width - 1, -1, -1)]
+    sequence_cols = [str(c) for c in range(stored_context_width - 1, -1, -1)]
 
     all_feature_cols = data.get_column("inputCol").unique().to_list()
 
@@ -2028,7 +2035,9 @@ def _write_accumulated_sequences(
     out_path = insert_top_folder(split_path_batch_seq, target_dir)
 
     if write_format == "pt":
-        process_and_write_data_pt(combined_df, layout.stored_width, out_path, col_types)
+        process_and_write_data_pt(
+            combined_df, layout.stored_context_width, out_path, col_types
+        )
     elif write_format == "parquet":
         combined_df.write_parquet(out_path)
 
@@ -2222,7 +2231,7 @@ def extract_sequences(
 
         subsequences, left_pad_lengths, subsequence_starts = extract_subsequences(
             in_seq_lists_only,
-            layout.stored_width,
+            layout.stored_context_width,
             stride_for_split,
             columns,
             subsequence_start_mode,
@@ -2238,7 +2247,7 @@ def extract_sequences(
                     left_pad_lengths[subsequence_id],
                     col,
                 ] + subseqs[subsequence_id]
-                expected_row_length = 5 + layout.stored_width
+                expected_row_length = 5 + layout.stored_context_width
                 if len(row) != expected_row_length:
                     raise RuntimeError(
                         f"Row length mismatch. Expected {expected_row_length}, got {len(row)}. Row: {row}"
@@ -2256,21 +2265,21 @@ def extract_sequences(
 @beartype
 def get_subsequence_starts(
     in_context_length: int,
-    stored_width: int,
+    stored_context_width: int,
     stride_for_split: int,
     subsequence_start_mode: str,
 ) -> np.ndarray:
     """Calculates the start indices for extracting subsequences.
 
     This function determines the starting indices for sliding a window of
-    `stored_width` over an input sequence of `in_context_length`. It aims to
+    `stored_context_width` over an input sequence of `in_context_length`. It aims to
     use `stride_for_split`, but adjusts the step size slightly to ensure
     that the windows are distributed as evenly as possible and cover the
     full sequence from the beginning to the end.
 
     Args:
         in_context_length: The length of the original input sequence.
-        stored_width: The stored window length to extract.
+        stored_context_width: The stored window length to extract.
         stride_for_split: The *desired* step size between subsequences.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
 
@@ -2283,7 +2292,7 @@ def get_subsequence_starts(
         )
 
     if subsequence_start_mode == "distribute":
-        last_available_start = in_context_length - stored_width
+        last_available_start = in_context_length - stored_context_width
         raw_starts = np.arange(
             0, last_available_start + stride_for_split, stride_for_split
         )
@@ -2294,11 +2303,11 @@ def get_subsequence_starts(
         return np.unique(starts)
 
     if subsequence_start_mode == "exact":
-        if (in_context_length - stored_width) % stride_for_split != 0:
+        if (in_context_length - stored_context_width) % stride_for_split != 0:
             raise ValueError(
-                f"'exact' mode requires sequence length alignment, i.e. if: (in_context_length - stored_width) % stride_for_split == 0, {in_context_length = }, {stored_width = }, {stride_for_split = }"
+                f"'exact' mode requires sequence length alignment, i.e. if: (in_context_length - stored_context_width) % stride_for_split == 0, {in_context_length = }, {stored_context_width = }, {stride_for_split = }"
             )
-        last_possible_start = in_context_length - stored_width
+        last_possible_start = in_context_length - stored_context_width
         return np.arange(0, last_possible_start + 1, stride_for_split)
     return np.array([])
 
@@ -2306,7 +2315,7 @@ def get_subsequence_starts(
 @beartype
 def extract_subsequences(
     in_seq: dict[str, list],
-    stored_width: int,
+    stored_context_width: int,
     stride_for_split: int,
     columns: list[str],
     subsequence_start_mode: str,
@@ -2316,14 +2325,14 @@ def extract_subsequences(
     This function takes a dictionary `in_seq` where keys are column
     names and values are lists of items for a single full sequence.
     It first pads the sequences with 0s at the beginning if they are
-    shorter than `stored_width`. Then, it calculates the subsequence
+    shorter than `stored_context_width`. Then, it calculates the subsequence
     start indices using `get_subsequence_starts` and extracts all
     subsequences.
 
     Args:
         in_seq: A dictionary mapping column names to lists of items
             (e.g., `{'col_A': [1, 2, 3, 4, 5], 'col_B': [6, 7, 8, 9, 10]}`).
-        stored_width: The stored window length to extract.
+        stored_context_width: The stored window length to extract.
         stride_for_split: The desired step size between subsequences.
         columns: A list of the column names (keys in `in_seq`) to process.
         subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
@@ -2334,13 +2343,16 @@ def extract_subsequences(
     """
     in_seq_len = len(in_seq[columns[0]])
     pad_len = 0
-    if in_seq_len < stored_width:
-        pad_len = stored_width - in_seq_len
+    if in_seq_len < stored_context_width:
+        pad_len = stored_context_width - in_seq_len
         in_seq = {col: ([0] * pad_len) + in_seq[col] for col in columns}
     in_context_length = len(in_seq[columns[0]])
 
     subsequence_starts = get_subsequence_starts(
-        in_context_length, stored_width, stride_for_split, subsequence_start_mode
+        in_context_length,
+        stored_context_width,
+        stride_for_split,
+        subsequence_start_mode,
     )
     subsequence_starts_diff = subsequence_starts[1:] - subsequence_starts[:-1]
     if not np.all(subsequence_starts_diff <= stride_for_split):
@@ -2349,7 +2361,9 @@ def extract_subsequences(
         )
 
     result = {
-        col: [list(in_seq[col][i : i + stored_width]) for i in subsequence_starts]
+        col: [
+            list(in_seq[col][i : i + stored_context_width]) for i in subsequence_starts
+        ]
         for col in columns
     }
     left_pad_lengths = [pad_len] * len(subsequence_starts)
diff --git a/tests/configs/preprocess-test-categorical-exact-pt.yaml b/tests/configs/preprocess-test-categorical-exact-pt.yaml
index 0d6f82c1..2e233b9b 100644
--- a/tests/configs/preprocess-test-categorical-exact-pt.yaml
+++ b/tests/configs/preprocess-test-categorical-exact-pt.yaml
@@ -7,8 +7,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-stored_width: 6
-future_capacity: 1
+stored_context_width: 6
+max_target_offset: 1
 stride_by_split:
 - 5
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-exact.yaml b/tests/configs/preprocess-test-categorical-exact.yaml
index b448ef32..3e58bef4 100644
--- a/tests/configs/preprocess-test-categorical-exact.yaml
+++ b/tests/configs/preprocess-test-categorical-exact.yaml
@@ -7,8 +7,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-stored_width: 6
-future_capacity: 1
+stored_context_width: 6
+max_target_offset: 1
 stride_by_split:
 - 5
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-interrupted.yaml b/tests/configs/preprocess-test-categorical-interrupted.yaml
index 6d244e94..7dae9dcf 100644
--- a/tests/configs/preprocess-test-categorical-interrupted.yaml
+++ b/tests/configs/preprocess-test-categorical-interrupted.yaml
@@ -9,8 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-lookahead-0.yaml b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
index be13f25a..c9653bea 100644
--- a/tests/configs/preprocess-test-categorical-lookahead-0.yaml
+++ b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
@@ -9,8 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-stored_width: 8
-future_capacity: 0
+stored_context_width: 8
+max_target_offset: 0
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-multitarget.yaml b/tests/configs/preprocess-test-categorical-multitarget.yaml
index 1fc95074..0dca4827 100644
--- a/tests/configs/preprocess-test-categorical-multitarget.yaml
+++ b/tests/configs/preprocess-test-categorical-multitarget.yaml
@@ -9,8 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
index 24c12914..7372bc0e 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
@@ -8,8 +8,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
index 8e7fd362..1f565c32 100644
--- a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
+++ b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
@@ -7,8 +7,8 @@ selected_columns: null
 
 split_ratios:
 - 1.0
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 max_rows: null
diff --git a/tests/configs/preprocess-test-categorical.yaml b/tests/configs/preprocess-test-categorical.yaml
index 7ca2774b..4767f7e6 100644
--- a/tests/configs/preprocess-test-categorical.yaml
+++ b/tests/configs/preprocess-test-categorical.yaml
@@ -9,8 +9,8 @@ split_ratios:
 - 0.6
 - 0.2
 - 0.2
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-multi-file.yaml b/tests/configs/preprocess-test-multi-file.yaml
index a7792689..cc76ee4c 100644
--- a/tests/configs/preprocess-test-multi-file.yaml
+++ b/tests/configs/preprocess-test-multi-file.yaml
@@ -12,8 +12,8 @@ split_ratios:
 - 0.8
 - 0.1
 - 0.1
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/configs/preprocess-test-real.yaml b/tests/configs/preprocess-test-real.yaml
index f0015e26..538ba285 100644
--- a/tests/configs/preprocess-test-real.yaml
+++ b/tests/configs/preprocess-test-real.yaml
@@ -8,8 +8,8 @@ selected_columns: null
 split_ratios:
 - 0.5
 - 0.5
-stored_width: 9
-future_capacity: 1
+stored_context_width: 9
+max_target_offset: 1
 stride_by_split:
 - 1
 - 1
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index 1beeeb51..ebe52dd2 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -45,7 +45,7 @@ def test_hp_search_bert_outputs(run_hp_search, project_root):
     )
     assert "storage_layout" not in generated_config
     assert "window_view" not in generated_config
-    assert generated_config["target_shift"] == 0
+    assert generated_config["target_offset"] == 0
     assert generated_config["training_spec"]["training_objective"] == "bert"
 
 
diff --git a/tests/integration/test_make.py b/tests/integration/test_make.py
index 47eaad7e..5b09d088 100644
--- a/tests/integration/test_make.py
+++ b/tests/integration/test_make.py
@@ -54,7 +54,7 @@ def adapt_configs(config_strings):
                 "selected_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "selected_columns: "
             )
             .replace("input_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "input_columns: ")
-            .replace("stored_width: 49", "stored_width: 11")
+            .replace("stored_context_width: 49", "stored_context_width: 11")
             .replace("max_rows: null", "max_rows: null\nn_cores: 1")
         )
 
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index 0036de9c..4a4be821 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -29,14 +29,14 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
         n_classes={"cat_col": 6},
         storage_layout=StoredWindowLayout(
-            stored_width=context_length + 1,
-            future_capacity=1,
+            stored_context_width=context_length + 1,
+            max_target_offset=1,
             version=2,
         ),
         window_view=ModelWindowView(
             context_length=context_length,
             objective="bert",
-            target_shift=0,
+            target_offset=0,
         ),
         inference_batch_size=inference_batch_size,
         seed=42,
@@ -136,14 +136,14 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
         },
         n_classes={"cat2": 7, "cat10": 6},
         storage_layout=StoredWindowLayout(
-            stored_width=context_length + 1,
-            future_capacity=1,
+            stored_context_width=context_length + 1,
+            max_target_offset=1,
             version=2,
         ),
         window_view=ModelWindowView(
             context_length=context_length,
             objective="causal",
-            target_shift=1,
+            target_offset=1,
         ),
         inference_batch_size=inference_batch_size,
         seed=42,
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 67508669..708dc24d 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -31,8 +31,8 @@ def test_metadata_config(metadata_configs):
         "split_paths",
         "column_types",
         "selected_columns_statistics",
-        "stored_width",
-        "future_capacity",
+        "stored_context_width",
+        "max_target_offset",
         "stored_window_layout_version",
     ]
 
@@ -272,8 +272,8 @@ def test_preprocessed_data_categorical_lookahead_0(run_preprocessing, project_ro
     with open(metadata_path, "r") as f:
         metadata_config = json.load(f)
 
-    assert metadata_config["future_capacity"] == 0
-    assert metadata_config["stored_width"] == 8
+    assert metadata_config["max_target_offset"] == 0
+    assert metadata_config["stored_context_width"] == 8
 
     for split in range(3):
         data = read_preprocessing_outputs(
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
index 1ec4aee0..f578cde1 100644
--- a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
+++ b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
@@ -1,6 +1,6 @@
 {
-    "stored_width": 9,
-    "future_capacity": 1,
+    "stored_context_width": 9,
+    "max_target_offset": 1,
     "stored_window_layout_version": 2,
     "selected_columns": null,
     "data_columns": [
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 24ae734b..6fb7b0b5 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -19,8 +19,8 @@ def _folder_metadata(total_samples, batch_files):
     return {
         "total_samples": total_samples,
         "batch_files": batch_files,
-        "stored_width": STORED_WIDTH,
-        "future_capacity": FUTURE_CAPACITY,
+        "stored_context_width": STORED_WIDTH,
+        "max_target_offset": FUTURE_CAPACITY,
         "stored_window_layout_version": 2,
     }
 
@@ -31,10 +31,10 @@ def mock_config():
     config.project_root = "."
     config.seed = 42
     config.storage_layout = StoredWindowLayout(
-        stored_width=STORED_WIDTH, future_capacity=FUTURE_CAPACITY, version=2
+        stored_context_width=STORED_WIDTH, max_target_offset=FUTURE_CAPACITY, version=2
     )
     config.window_view = ModelWindowView(
-        context_length=CONTEXT_LENGTH, objective="causal", target_shift=1
+        context_length=CONTEXT_LENGTH, objective="causal", target_offset=1
     )
     config.column_types = {"item": "Float64"}
     config.training_spec.batch_size = 5
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 2b62d861..215ddc99 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -18,8 +18,8 @@ def _folder_metadata(total_samples, batch_files):
     return {
         "total_samples": total_samples,
         "batch_files": batch_files,
-        "stored_width": STORED_WIDTH,
-        "future_capacity": FUTURE_CAPACITY,
+        "stored_context_width": STORED_WIDTH,
+        "max_target_offset": FUTURE_CAPACITY,
         "stored_window_layout_version": 2,
     }
 
@@ -35,10 +35,10 @@ def mock_config(tmp_path):
     config.training_spec.num_workers = 0
     config.seed = 42
     config.storage_layout = StoredWindowLayout(
-        stored_width=STORED_WIDTH, future_capacity=FUTURE_CAPACITY, version=2
+        stored_context_width=STORED_WIDTH, max_target_offset=FUTURE_CAPACITY, version=2
     )
     config.window_view = ModelWindowView(
-        context_length=CONTEXT_LENGTH, objective="causal", target_shift=1
+        context_length=CONTEXT_LENGTH, objective="causal", target_offset=1
     )
     config.input_columns = ["col1"]
     config.target_columns = ["col1", "tgt1"]
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index 635669d1..46bea614 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -91,8 +91,8 @@ def test_numpy_to_pytorch_shapes_and_shifting():
     column_types = {"A": torch.float32}
     all_columns = ["A"]
     resolved_view = resolve_window_view(
-        StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
-        ModelWindowView(context_length=3, objective="causal", target_shift=1),
+        StoredWindowLayout(stored_context_width=4, max_target_offset=1, version=2),
+        ModelWindowView(context_length=3, objective="causal", target_offset=1),
     )
 
     tensors, metadata = numpy_to_pytorch(
@@ -152,8 +152,8 @@ def test_numpy_to_pytorch_dtypes():
         {"int_col": torch.int64},
         ["int_col"],
         resolve_window_view(
-            StoredWindowLayout(stored_width=2, future_capacity=1, version=2),
-            ModelWindowView(context_length=1, objective="causal", target_shift=1),
+            StoredWindowLayout(stored_context_width=2, max_target_offset=1, version=2),
+            ModelWindowView(context_length=1, objective="causal", target_offset=1),
         ),
     )
     assert tensors_int["int_col"].dtype == torch.int64
@@ -175,8 +175,8 @@ def test_numpy_to_pytorch_dtypes():
         {"float_col": torch.float32},
         ["float_col"],
         resolve_window_view(
-            StoredWindowLayout(stored_width=2, future_capacity=1, version=2),
-            ModelWindowView(context_length=1, objective="causal", target_shift=1),
+            StoredWindowLayout(stored_context_width=2, max_target_offset=1, version=2),
+            ModelWindowView(context_length=1, objective="causal", target_offset=1),
         ),
     )
     assert tensors_float["float_col"].dtype == torch.float32
@@ -186,8 +186,8 @@ def test_build_valid_mask_from_left_pad_lengths():
     left_pad_lengths = torch.tensor([0, 2, 5], dtype=torch.int64)
 
     resolved_view = resolve_window_view(
-        StoredWindowLayout(stored_width=6, future_capacity=1, version=2),
-        ModelWindowView(context_length=5, objective="causal", target_shift=1),
+        StoredWindowLayout(stored_context_width=6, max_target_offset=1, version=2),
+        ModelWindowView(context_length=5, objective="causal", target_offset=1),
     )
     input_mask = build_valid_mask(
         left_pad_lengths, full_length=6, view_slice=resolved_view.input_slice
@@ -221,8 +221,8 @@ def test_build_valid_mask_from_left_pad_lengths():
 def test_resolve_window_view_slices_tensors():
     tensor = torch.arange(8).reshape(2, 4)
     resolved_view = resolve_window_view(
-        StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
-        ModelWindowView(context_length=3, objective="causal", target_shift=1),
+        StoredWindowLayout(stored_context_width=4, max_target_offset=1, version=2),
+        ModelWindowView(context_length=3, objective="causal", target_offset=1),
     )
 
     assert torch.equal(
@@ -255,8 +255,8 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
         {"A": torch.float32},
         ["A"],
         resolve_window_view(
-            StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
-            ModelWindowView(context_length=3, objective="causal", target_shift=1),
+            StoredWindowLayout(stored_context_width=4, max_target_offset=1, version=2),
+            ModelWindowView(context_length=3, objective="causal", target_offset=1),
         ),
     )
 
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 92f86dc3..91d3c910 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -174,8 +174,12 @@ def test_infer_config_defaults_bert_prediction_length_to_context_length():
         seed=42,
         device="cpu",
         prediction_length=None,
-        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
-        window_view=ModelWindowView(context_length=3, objective="bert", target_shift=0),
+        storage_layout=StoredWindowLayout(
+            stored_context_width=4, max_target_offset=1, version=2
+        ),
+        window_view=ModelWindowView(
+            context_length=3, objective="bert", target_offset=0
+        ),
         inference_batch_size=2,
         output_probabilities=False,
         map_to_id=False,
@@ -202,9 +206,11 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
         seed=42,
         device="cpu",
         prediction_length=None,
-        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        storage_layout=StoredWindowLayout(
+            stored_context_width=4, max_target_offset=1, version=2
+        ),
         window_view=ModelWindowView(
-            context_length=3, objective="causal", target_shift=1
+            context_length=3, objective="causal", target_offset=1
         ),
         inference_batch_size=2,
         output_probabilities=False,
@@ -234,10 +240,10 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             device="cpu",
             prediction_length=1,
             storage_layout=StoredWindowLayout(
-                stored_width=4, future_capacity=1, version=2
+                stored_context_width=4, max_target_offset=1, version=2
             ),
             window_view=ModelWindowView(
-                context_length=3, objective="bert", target_shift=0
+                context_length=3, objective="bert", target_offset=0
             ),
             inference_batch_size=2,
             output_probabilities=False,
@@ -256,8 +262,8 @@ def test_load_inferer_config_rejects_mismatched_metadata_special_token_ids(tmp_p
         json.dumps(
             {
                 "split_paths": [str(data_path)],
-                "stored_width": 4,
-                "future_capacity": 1,
+                "stored_context_width": 4,
+                "max_target_offset": 1,
                 "stored_window_layout_version": 2,
                 "column_types": {"target_col": "int64"},
                 "id_maps": {"target_col": {"A": 3}},
@@ -401,8 +407,12 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         seed=42,
         device="cpu",
         prediction_length=None,
-        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
-        window_view=ModelWindowView(context_length=3, objective="bert", target_shift=0),
+        storage_layout=StoredWindowLayout(
+            stored_context_width=4, max_target_offset=1, version=2
+        ),
+        window_view=ModelWindowView(
+            context_length=3, objective="bert", target_offset=0
+        ),
         inference_batch_size=4,
         output_probabilities=False,
         map_to_id=True,
@@ -617,9 +627,11 @@ def ar_config():
         seed=42,
         device="cpu",
         prediction_length=1,
-        storage_layout=StoredWindowLayout(stored_width=4, future_capacity=1, version=2),
+        storage_layout=StoredWindowLayout(
+            stored_context_width=4, max_target_offset=1, version=2
+        ),
         window_view=ModelWindowView(
-            context_length=3, objective="causal", target_shift=1
+            context_length=3, objective="causal", target_offset=1
         ),
         inference_batch_size=2,
         output_probabilities=False,
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index eb2a6281..436d3f04 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -61,7 +61,7 @@ def test_extract_subsequences_bert_width():
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        stored_width=context_length,
+        stored_context_width=context_length,
         stride_for_split=stride,
         columns=columns,
         subsequence_start_mode="distribute",
@@ -75,14 +75,18 @@ def test_extract_subsequences_bert_width():
 def test_extract_subsequences_padding():
     """Tests that sequences shorter than context_length are padded with 0s."""
     input_data = {"col1": [1, 2]}  # Length 2
-    stored_width = 5
+    stored_context_width = 5
     stride = 1
     columns = ["col1"]
 
     # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
 
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data, stored_width, stride, columns, subsequence_start_mode="distribute"
+        input_data,
+        stored_context_width,
+        stride,
+        columns,
+        subsequence_start_mode="distribute",
     )
 
     assert len(result["col1"]) == 1
@@ -93,7 +97,7 @@ def test_extract_subsequences_returns_left_pad_lengths_when_requested():
     input_data = {"col1": [0.0, 1.5]}
     result, left_pad_lengths, subsequence_starts = extract_subsequences(
         input_data,
-        stored_width=5,
+        stored_context_width=5,
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -127,7 +131,9 @@ def test_extract_sequences_persists_left_pad_length_metadata():
     sequences = extract_sequences(
         data,
         schema,
-        layout=StoredWindowLayout(stored_width=5, future_capacity=1, version=2),
+        layout=StoredWindowLayout(
+            stored_context_width=5, max_target_offset=1, version=2
+        ),
         stride_for_split=1,
         columns=["col1"],
         subsequence_start_mode="distribute",
@@ -154,7 +160,7 @@ def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
 
     process_and_write_data_pt(
         data,
-        stored_width=4,
+        stored_context_width=4,
         path=str(out_path),
         column_types={"col1": "Float64"},
     )
@@ -218,8 +224,8 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
                 "selected_columns_statistics": {
                     "itemValue": {"mean": 60.0, "std": 10.0}
                 },
-                "stored_width": 3,
-                "future_capacity": 1,
+                "stored_context_width": 3,
+                "max_target_offset": 1,
                 "stored_window_layout_version": 2,
                 "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
             }
@@ -245,8 +251,8 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
         merge_output=True,
         selected_columns=["itemId", "itemValue"],
         split_ratios=[1.0],
-        stored_width=3,
-        future_capacity=1,
+        stored_context_width=3,
+        max_target_offset=1,
         stride_by_split=[1],
         max_rows=None,
         seed=1010,
@@ -383,8 +389,8 @@ def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
             merge_output=True,
             selected_columns=["itemId"],
             split_ratios=[1.0],
-            stored_width=2,
-            future_capacity=1,
+            stored_context_width=2,
+            max_target_offset=1,
             stride_by_split=[1],
             max_rows=None,
             seed=1010,
@@ -413,8 +419,8 @@ def test_preprocessor_config_defaults_mask_column_to_none(tmp_path):
         project_root=str(tmp_path),
         data_path=str(data_path),
         split_ratios=[1.0],
-        stored_width=2,
-        future_capacity=1,
+        stored_context_width=2,
+        max_target_offset=1,
         seed=1010,
     )
 
@@ -442,8 +448,8 @@ def test_preprocessor_config_requires_metadata_config_for_mask_column(
             project_root=str(tmp_path),
             data_path=str(data_path),
             split_ratios=[1.0],
-            stored_width=2,
-            future_capacity=1,
+            stored_context_width=2,
+            max_target_offset=1,
             seed=1010,
             mask_column=RESERVED_MASK_COLUMN,
         )
@@ -457,14 +463,14 @@ def test_extract_subsequences_modes(mode):
     # exact: strictly adheres to stride, throws error if misalignment.
     input_data = {"col1": list(range(10))}
     context_length = 2
-    stored_width = context_length + 1
+    stored_context_width = context_length + 1
     columns = ["col1"]
 
     if mode == "distribute":
         stride = 4
         # distribute might adjust indices to maximize coverage
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data, stored_width, stride, columns, mode
+            input_data, stored_context_width, stride, columns, mode
         )
         assert len(result["col1"]) > 0
 
@@ -474,13 +480,13 @@ def test_extract_subsequences_modes(mode):
         )
         # Testing a failing exact case
         with pytest.raises(ValueError):
-            extract_subsequences(input_data, stored_width, 4, columns, mode)
+            extract_subsequences(input_data, stored_context_width, 4, columns, mode)
 
         # Testing a passing exact case
         # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
         input_data_exact = {"col1": list(range(11))}
         result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data_exact, stored_width, 4, columns, mode
+            input_data_exact, stored_context_width, 4, columns, mode
         )
         assert len(result["col1"]) > 0
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 42be51f8..872b2e76 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -77,8 +77,8 @@ def test_training_spec_model_dump_excludes_runtime_offsets():
 
     assert "data_offset" not in dumped
     assert "target_offset" not in dumped
-    assert "stored_width" not in dumped
-    assert "max_lookahead" not in dumped
+    assert "stored_context_width" not in dumped
+    assert "max_target_offset" not in dumped
 
 
 def test_poisson_span_masking_samples_at_least_one_token():
@@ -137,10 +137,10 @@ def model_config(tmp_path):
         id_maps={"cat_col": {"a": 1, "b": 2, "c": 3, "d": 4}},
         n_classes={"cat_col": 5},  # 0 + 4 classes
         storage_layout=StoredWindowLayout(
-            stored_width=11, future_capacity=1, version=2
+            stored_context_width=11, max_target_offset=1, version=2
         ),
         window_view=ModelWindowView(
-            context_length=10, objective="causal", target_shift=1
+            context_length=10, objective="causal", target_offset=1
         ),
         inference_batch_size=4,
         seed=42,
@@ -178,7 +178,7 @@ def bert_model(model_config):
     config_values["window_view"] = {
         **config_values["window_view"],
         "objective": "bert",
-        "target_shift": 0,
+        "target_offset": 0,
     }
     config = TrainModel(**config_values)
     return TransformerModel(config)
@@ -238,7 +238,7 @@ def test_train_model_requires_bert_prediction_length_to_equal_context_length(
     config_values["window_view"] = {
         **config_values["window_view"],
         "objective": "bert",
-        "target_shift": 0,
+        "target_offset": 0,
     }
 
     with pytest.raises(ValidationError, match="prediction_length must be equal"):
@@ -273,8 +273,8 @@ def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
-                "stored_width": storage_layout["stored_width"],
-                "future_capacity": storage_layout["future_capacity"],
+                "stored_context_width": storage_layout["stored_context_width"],
+                "max_target_offset": storage_layout["max_target_offset"],
                 "stored_window_layout_version": storage_layout["version"],
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
@@ -308,8 +308,8 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
         json.dumps(
             {
                 "split_paths": ["data/train.pt", "data/val.pt"],
-                "stored_width": storage_layout["stored_width"],
-                "future_capacity": storage_layout["future_capacity"],
+                "stored_context_width": storage_layout["stored_context_width"],
+                "max_target_offset": storage_layout["max_target_offset"],
                 "stored_window_layout_version": storage_layout["version"],
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],

From dcd92f70d1a9e3e4c0f2e3239f8818c7f5182d38 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 16:51:40 +0200
Subject: [PATCH 53/81] make valid_mask compulsory

---
 src/sequifier/helpers.py |   2 +-
 src/sequifier/infer.py   | 135 ++++++++++++++++++++-------------------
 tests/unit/test_infer.py |  29 ---------
 3 files changed, 72 insertions(+), 94 deletions(-)

diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 5f3a52fb..a7c4f1e4 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -104,7 +104,7 @@ def build_masks(self, left_pad_lengths: Tensor) -> dict[str, Tensor]:
 @beartype
 def _right_aligned_slice(width: int, length: int, offset: int) -> slice:
     start = width - (length + offset)
-    stop = width - offset if offset else width
+    stop = width - offset
     return slice(start, stop)
 
 
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 18a8c3ad..b3ed690f 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -305,30 +305,21 @@ def calculate_item_positions(
 
 
 @beartype
-def _flatten_bert_target_valid_mask(
+def _flatten_valid_mask(
     config: InfererModel,
     metadata: dict[str, Any],
     prediction_length: int,
-) -> Optional[np.ndarray]:
-    if config.training_objective != "bert" or not metadata:
-        return None
-    if "target_valid_mask" not in metadata:
-        return None
-
-    valid_mask = metadata["target_valid_mask"]
+    mask_key: str = "target_valid_mask",
+) -> np.ndarray:
+    valid_mask = metadata.get(mask_key)
     if isinstance(valid_mask, torch.Tensor):
-        valid_mask = valid_mask.detach().cpu().numpy()
+        valid_mask = valid_mask.detach().cpu().numpy()  # type: ignore
     valid_mask = np.asarray(valid_mask, dtype=bool)
 
     if valid_mask.ndim != 2:
-        raise ValueError(f"target_valid_mask must be 2D, got shape {valid_mask.shape}.")
-    if valid_mask.shape[1] != prediction_length:
-        raise ValueError(
-            "target_valid_mask width must match prediction_length "
-            f"(got {valid_mask.shape[1]} and {prediction_length})."
-        )
+        raise ValueError(f"{mask_key} must be 2D, got shape {valid_mask.shape}.")
 
-    return valid_mask.reshape(-1)
+    return valid_mask[:, -prediction_length:].reshape(-1)
 
 
 @beartype
@@ -347,41 +338,34 @@ def _bert_reference_column(config: InfererModel, data_columns: set[str]) -> str:
 
 
 @beartype
-def _bert_target_valid_mask_from_preprocessed_data(
+def _valid_mask_from_preprocessed_data(
     config: InfererModel,
     data: pl.DataFrame,
     prediction_length: int,
-) -> Optional[np.ndarray]:
-    if config.training_objective != "bert":
-        return None
-
+    mask_key: str = "target_valid_mask",
+) -> np.ndarray:
     data_columns = set(data.get_column("inputCol").unique())
     column_name = _bert_reference_column(config, data_columns)
     reference_rows = data.filter(pl.col("inputCol") == column_name)
 
-    if "leftPadLength" in reference_rows.columns:
-        left_pad_lengths = torch.tensor(
-            reference_rows.get_column("leftPadLength").to_numpy(), dtype=torch.int64
-        )
-        resolved_view = resolve_window_view(config.storage_layout, config.window_view)
-        metadata = resolved_view.build_masks(left_pad_lengths)
-        return _flatten_bert_target_valid_mask(config, metadata, prediction_length)
-
-    return None
+    left_pad_lengths = torch.tensor(
+        reference_rows.get_column("leftPadLength").to_numpy(), dtype=torch.int64
+    )
+    resolved_view = resolve_window_view(config.storage_layout, config.window_view)
+    metadata = resolved_view.build_masks(left_pad_lengths)
+    return _flatten_valid_mask(config, metadata, prediction_length, mask_key)
 
 
 @beartype
 def _apply_valid_prediction_mask(
     values: np.ndarray,
-    valid_prediction_mask: Optional[np.ndarray],
+    valid_prediction_mask: np.ndarray,
     label: str,
 ) -> np.ndarray:
     values = np.asarray(values)
-    if valid_prediction_mask is None:
-        return values
     if values.shape[0] != valid_prediction_mask.shape[0]:
         raise ValueError(
-            f"{label} has {values.shape[0]} rows, but target_valid_mask has "
+            f"{label} has {values.shape[0]} rows, but valid_prediction_mask has "
             f"{valid_prediction_mask.shape[0]} rows."
         )
     return values[valid_prediction_mask]
@@ -390,7 +374,7 @@ def _apply_valid_prediction_mask(
 @beartype
 def _apply_valid_prediction_mask_to_dict(
     values: Optional[dict[str, np.ndarray]],
-    valid_prediction_mask: Optional[np.ndarray],
+    valid_prediction_mask: np.ndarray,
     label: str,
 ) -> Optional[dict[str, np.ndarray]]:
     if values is None:
@@ -451,8 +435,8 @@ def infer_embedding(
             mask = pl.arange(0, data.height, eager=True) % n_input_cols == 0
 
             embeddings = get_embeddings(config, inferer, data, column_types)
-            valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
-                config, data, prediction_length
+            valid_prediction_mask = _valid_mask_from_preprocessed_data(
+                config, data, prediction_length, mask_key="attention_valid_mask"
             )
 
             sequence_ids_for_preds = data.get_column("sequenceId").filter(mask)
@@ -469,8 +453,8 @@ def infer_embedding(
             mask = pl.arange(0, data.height, eager=True) % n_input_cols == 0
 
             embeddings = get_embeddings(config, inferer, data, column_types)
-            valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
-                config, data, prediction_length
+            valid_prediction_mask = _valid_mask_from_preprocessed_data(
+                config, data, prediction_length, mask_key="attention_valid_mask"
             )
 
             sequence_ids_for_preds = (
@@ -503,8 +487,8 @@ def infer_embedding(
             embeddings = get_embeddings_pt(
                 config, inferer, sequences_dict, metadata=metadata
             )
-            valid_prediction_mask = _flatten_bert_target_valid_mask(
-                config, metadata, prediction_length
+            valid_prediction_mask = _flatten_valid_mask(
+                config, metadata, prediction_length, mask_key="attention_valid_mask"
             )
 
             sequence_ids_for_preds = sequence_ids_tensor.numpy()
@@ -538,6 +522,7 @@ def infer_embedding(
         subsequence_ids_repeated = np.repeat(
             subsequence_ids_for_preds, prediction_length
         )
+        assert valid_prediction_mask is not None
         embeddings = _apply_valid_prediction_mask(
             embeddings, valid_prediction_mask, "embeddings"
         )
@@ -628,8 +613,6 @@ def infer_generative(
             `torch.dtype`.
     """
     for data_id, data in enumerate(dataset):
-        valid_prediction_mask = None
-
         # Step 1: Adapt Data Subsetting (now works on Polars DF)
         is_folder_input = os.path.isdir(
             normalize_path(config.data_path, config.project_root)
@@ -653,7 +636,7 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
-                valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
+                valid_prediction_mask = _valid_mask_from_preprocessed_data(
                     config, data, prediction_length
                 )
 
@@ -676,14 +659,18 @@ def infer_generative(
                         f"prediction_length must be 1 for autoregression, got {inferer.prediction_length}"
                     )
                 # Unpack the new third return value
-                probs, preds, sequence_ids_for_preds, item_positions_for_preds = (
-                    get_probs_preds_autoregression(
-                        config,
-                        inferer,
-                        data,
-                        column_types,
-                        config.window_view.context_length,
-                    )
+                (
+                    probs,
+                    preds,
+                    sequence_ids_for_preds,
+                    item_positions_for_preds,
+                    valid_prediction_mask,
+                ) = get_probs_preds_autoregression(
+                    config,
+                    inferer,
+                    data,
+                    column_types,
+                    config.window_view.context_length,
                 )
         elif config.read_format == "parquet" and is_folder_input:
             # Folder-based Parquet chunk logic
@@ -709,7 +696,7 @@ def infer_generative(
                     data.get_column("startItemPosition").filter(mask).to_numpy()
                 )
                 prediction_length = inferer.prediction_length
-                valid_prediction_mask = _bert_target_valid_mask_from_preprocessed_data(
+                valid_prediction_mask = _valid_mask_from_preprocessed_data(
                     config, data, prediction_length
                 )
 
@@ -730,14 +717,18 @@ def infer_generative(
                         f"prediction_length must be 1 for autoregression, got {inferer.prediction_length}"
                     )
 
-                probs, preds, sequence_ids_for_preds, item_positions_for_preds = (
-                    get_probs_preds_autoregression(
-                        config,
-                        inferer,
-                        data,
-                        column_types,
-                        config.window_view.context_length,
-                    )
+                (
+                    probs,
+                    preds,
+                    sequence_ids_for_preds,
+                    item_positions_for_preds,
+                    valid_prediction_mask,
+                ) = get_probs_preds_autoregression(
+                    config,
+                    inferer,
+                    data,
+                    column_types,
+                    config.window_view.context_length,
                 )
         elif config.read_format == "pt":
             (
@@ -773,7 +764,7 @@ def infer_generative(
             )
 
             prediction_length = inferer.prediction_length  # Get prediction_length
-            valid_prediction_mask = _flatten_bert_target_valid_mask(
+            valid_prediction_mask = _flatten_valid_mask(
                 config, metadata, prediction_length
             )
 
@@ -798,6 +789,7 @@ def infer_generative(
                 sequence_ids_for_preds = np.repeat(
                     sequence_ids_tensor.numpy(), total_steps
                 )
+                valid_prediction_mask = np.repeat(valid_prediction_mask, total_steps)
                 item_position_boundaries = zip(
                     list(start_positions_tensor + config.window_view.context_length),
                     list(
@@ -826,6 +818,8 @@ def infer_generative(
                     predictions, target_column
                 )
 
+        assert valid_prediction_mask is not None
+
         sequence_ids_for_preds = _apply_valid_prediction_mask(
             sequence_ids_for_preds, valid_prediction_mask, "sequenceId"
         )
@@ -1235,7 +1229,11 @@ def get_probs_preds_autoregression(
     column_types: dict[str, torch.dtype],
     context_length: int,
 ) -> tuple[
-    Optional[dict[str, np.ndarray]], dict[str, np.ndarray], np.ndarray, np.ndarray
+    Optional[dict[str, np.ndarray]],
+    dict[str, np.ndarray],
+    np.ndarray,
+    np.ndarray,
+    np.ndarray,
 ]:
     """Generates autoregressive predictions and aligns them with sequence IDs and positions.
 
@@ -1303,7 +1301,16 @@ def get_probs_preds_autoregression(
         aligned_sequence_ids, config.autoregression_total_steps
     )
 
-    return probs, preds, sequence_ids_for_preds, item_positions_for_preds
+    base_mask = _flatten_valid_mask(config, metadata, 1)
+    valid_prediction_mask = np.repeat(base_mask, config.autoregression_total_steps)
+
+    return (
+        probs,
+        preds,
+        sequence_ids_for_preds,
+        item_positions_for_preds,
+        valid_prediction_mask,
+    )
 
 
 class Inferer:
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 91d3c910..9afb97ea 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -544,35 +544,6 @@ def capture_write(dataframe, path, write_format):
     assert predictions.get_column("target_col").to_list() == ["B", "A", "B"]
 
 
-def test_bert_generative_inference_keeps_data_without_left_pad(tmp_path):
-    config = _bert_inference_config(tmp_path)
-    data = _bert_preprocessed_frame().drop("leftPadLength")
-    written = []
-
-    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
-        "sequifier.infer.load_inference_model"
-    ):
-        inferer = _mock_bert_inferer(config)
-
-    def capture_write(dataframe, path, write_format):
-        written.append((path, dataframe, write_format))
-
-    preds = {"target_col": np.array([3, 4, 3])}
-    with patch(
-        "sequifier.infer.get_probs_preds_from_df", return_value=(None, preds)
-    ), patch("sequifier.infer.write_data", side_effect=capture_write):
-        infer_generative(
-            config, inferer, "bert-model", [data], {"target_col": torch.int64}
-        )
-
-    predictions = next(frame for path, frame, _ in written if "predictions" in path)
-
-    assert predictions.height == 3
-    assert predictions.get_column("sequenceId").to_list() == [7, 7, 7]
-    assert predictions.get_column("itemPosition").to_list() == [100, 101, 102]
-    assert predictions.get_column("target_col").to_list() == ["A", "B", "A"]
-
-
 def test_bert_embedding_inference_filters_left_padded_positions(tmp_path):
     config = _bert_inference_config(tmp_path, model_type="embedding")
     data = _bert_preprocessed_frame()

From 6b78f3ea1657f0b282933d752c45e7b0d6ee799a Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 17:02:04 +0200
Subject: [PATCH 54/81] set prediction_length from context_length in
 hyperparameter tuning of bert models

---
 src/sequifier/config/hyperparameter_search_config.py | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index b93c8ea5..94c20f00 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -748,6 +748,13 @@ def validate_sequence_layout(self):
                     f"({self.storage_layout.max_target_offset}) > stored_context_width ({self.storage_layout.stored_context_width}). "
                     "Model inputs cannot exceed the preprocessed sequence length."
                 )
+        if self.training_hyperparameter_sampling.training_objective == "causal":
+            if self.storage_layout.max_target_offset < 1:
+                raise ValueError(
+                    "The hyperparameter search space includes the 'causal' objective, "
+                    "but the preprocessed dataset has max_target_offset=0. "
+                    "Causal modeling requires max_target_offset >= 1."
+                )
         return self
 
     @model_validator(mode="after")
@@ -856,6 +863,11 @@ def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
             "context_length", self.context_length
         )
         training_spec = self.training_hyperparameter_sampling.sample_trial(trial)
+        if training_spec.training_objective == "bert":
+            model_spec = model_spec.model_copy(
+                update={"prediction_length": context_length}
+            )
+
         window_view = ModelWindowView(
             context_length=context_length,
             objective=training_spec.training_objective,

From 4c899e9b0336877af34d8278070ad0d8f5d37a01 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 18:02:42 +0200
Subject: [PATCH 55/81] Expand preprocessing digest

---
 src/sequifier/preprocess.py                   |  86 ++++++++++++--
 .../preprocess-manifest.json                  |  41 +++++--
 tests/unit/test_preprocess.py                 | 111 ++++++++++++++++++
 3 files changed, 223 insertions(+), 15 deletions(-)

diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 9f8964f7..cde9834d 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -1,3 +1,4 @@
+import hashlib
 import json
 import math
 import multiprocessing
@@ -34,6 +35,30 @@
 CURRENT_STORED_WINDOW_LAYOUT_VERSION = 2
 
 
+def _stable_json_value(value: Any) -> Any:
+    if isinstance(value, dict):
+        return {
+            str(key): _stable_json_value(val)
+            for key, val in sorted(value.items(), key=lambda item: str(item[0]))
+        }
+    if isinstance(value, (list, tuple)):
+        return [_stable_json_value(item) for item in value]
+    if isinstance(value, np.generic):
+        return value.item()
+    if isinstance(value, float) and math.isnan(value):
+        return "NaN"
+    return value
+
+
+def _stable_json_digest(value: Any) -> str:
+    encoded = json.dumps(
+        _stable_json_value(value),
+        sort_keys=True,
+        separators=(",", ":"),
+    ).encode("utf-8")
+    return hashlib.sha256(encoded).hexdigest()
+
+
 @beartype
 def preprocess(args: Any, args_config: dict[str, Any]) -> None:
     """Runs the main data preprocessing pipeline.
@@ -123,6 +148,9 @@ def __init__(
         """
         self.project_root = project_root
         self.batches_per_file = batches_per_file
+        self.data_path = data_path
+        self.read_format = read_format
+        self.write_format = write_format
 
         self.data_name_root = os.path.splitext(os.path.basename(data_path))[0]
         self.merge_output = merge_output
@@ -138,6 +166,11 @@ def __init__(
         self.use_precomputed_maps = use_precomputed_maps
         self.metadata_config_path = metadata_config_path
         self.mask_column = mask_column
+        self.split_ratios = split_ratios
+        self.stride_by_split = stride_by_split
+        self.max_rows = max_rows
+        self.process_by_file = process_by_file
+        self.subsequence_start_mode = subsequence_start_mode
         if self.mask_column is not None and self.metadata_config_path is None:
             raise ValueError("metadata_config_path must be set when mask_column is set")
 
@@ -246,6 +279,10 @@ def __init__(
                 selected_columns,
                 write_format,
                 data_columns,
+                id_maps,
+                n_classes,
+                col_types,
+                selected_columns_statistics,
             )
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
@@ -345,6 +382,10 @@ def __init__(
                 selected_columns,
                 write_format,
                 data_columns,
+                id_maps,
+                n_classes,
+                col_types,
+                selected_columns_statistics,
             )
             self._export_metadata(
                 id_maps, n_classes, col_types, selected_columns_statistics
@@ -835,12 +876,40 @@ def _write_or_validate_resume_manifest(
         selected_columns: Optional[list[str]],
         write_format: str,
         data_columns: list[str],
+        id_maps: dict[str, dict[Union[str, int], int]],
+        n_classes: dict[str, int],
+        col_types: dict[str, str],
+        selected_columns_statistics: dict[str, dict[str, float]],
     ) -> None:
+        effective_metadata = {
+            "n_classes": n_classes,
+            "id_maps": id_maps,
+            "column_types": col_types,
+            "selected_columns_statistics": selected_columns_statistics,
+            "special_token_ids": SPECIAL_TOKEN_IDS.ids_by_label,
+        }
         manifest = {
-            **self._layout_metadata(),
-            "selected_columns": selected_columns,
-            "data_columns": data_columns,
-            "write_format": write_format,
+            "manifest_version": 1,
+            "preprocessing_config": {
+                **self._layout_metadata(),
+                "read_format": self.read_format,
+                "write_format": write_format,
+                "merge_output": self.merge_output,
+                "selected_columns": selected_columns,
+                "data_columns": data_columns,
+                "split_ratios": self.split_ratios,
+                "stride_by_split": self.stride_by_split,
+                "max_rows": self.max_rows,
+                "process_by_file": self.process_by_file,
+                "subsequence_start_mode": self.subsequence_start_mode,
+                "mask_column": self.mask_column,
+                "metadata_config_path": self.metadata_config_path,
+                "use_precomputed_maps": self.use_precomputed_maps,
+            },
+            "effective_metadata_digest": {
+                "algorithm": "sha256",
+                "value": _stable_json_digest(effective_metadata),
+            },
         }
         manifest_path = os.path.join(
             self.project_root, "data", self.target_dir, "preprocess-manifest.json"
@@ -853,10 +922,13 @@ def _write_or_validate_resume_manifest(
                 )
             with open(manifest_path, "r") as f:
                 previous_manifest = json.load(f)
-            if previous_manifest != manifest:
+            if _stable_json_value(previous_manifest) != _stable_json_value(manifest):
                 raise ValueError(
-                    "Cannot continue preprocessing with different sequence layout, "
-                    "selected columns, or write format."
+                    "Cannot continue preprocessing with a different preprocessing "
+                    "manifest. Check sequence layout, input path, selected/data "
+                    "columns, output format, mask/metadata settings, split/stride "
+                    "settings, max_rows, process_by_file, subsequence_start_mode, "
+                    "or metadata/maps/statistics."
                 )
             return
 
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
index f578cde1..78cac8f9 100644
--- a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
+++ b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
@@ -1,10 +1,35 @@
 {
-    "stored_context_width": 9,
-    "max_target_offset": 1,
-    "stored_window_layout_version": 2,
-    "selected_columns": null,
-    "data_columns": [
-        "itemId"
-    ],
-    "write_format": "pt"
+    "manifest_version": 1,
+    "preprocessing_config": {
+        "stored_context_width": 9,
+        "max_target_offset": 1,
+        "stored_window_layout_version": 2,
+        "read_format": "csv",
+        "write_format": "pt",
+        "merge_output": false,
+        "selected_columns": null,
+        "data_columns": [
+            "itemId"
+        ],
+        "split_ratios": [
+            0.6,
+            0.2,
+            0.2
+        ],
+        "stride_by_split": [
+            1,
+            1,
+            1
+        ],
+        "max_rows": null,
+        "process_by_file": true,
+        "subsequence_start_mode": "distribute",
+        "mask_column": null,
+        "metadata_config_path": null,
+        "use_precomputed_maps": null
+    },
+    "effective_metadata_digest": {
+        "algorithm": "sha256",
+        "value": "b245092f219b7b0bed5e4e24ec7aa7be6dead1885e61adc00f6395d6164c460c"
+    }
 }
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 436d3f04..9ac84b5a 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -292,6 +292,117 @@ def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
     }
 
 
+def test_preprocessor_writes_expanded_resume_manifest(tmp_path):
+    project_root = tmp_path / "project"
+    project_root.mkdir()
+    data_path = tmp_path / "manifest-input.csv"
+    pl.DataFrame(
+        {
+            "sequenceId": [0, 0, 0],
+            "itemPosition": [0, 1, 2],
+            "itemId": ["a", "b", "c"],
+        }
+    ).write_csv(data_path)
+
+    Preprocessor(
+        project_root=str(project_root),
+        continue_preprocessing=False,
+        data_path=str(data_path),
+        read_format="csv",
+        write_format="parquet",
+        merge_output=False,
+        selected_columns=["itemId"],
+        split_ratios=[1.0],
+        stored_context_width=3,
+        max_target_offset=1,
+        stride_by_split=[2],
+        max_rows=2,
+        seed=1010,
+        n_cores=1,
+        batches_per_file=1024,
+        process_by_file=True,
+        subsequence_start_mode="exact",
+        use_precomputed_maps=None,
+        metadata_config_path=None,
+        mask_column=None,
+    )
+
+    manifest_path = (
+        project_root / "data" / "manifest-input-temp" / "preprocess-manifest.json"
+    )
+    manifest = json.loads(manifest_path.read_text())
+    config = manifest["preprocessing_config"]
+
+    assert manifest["manifest_version"] == 1
+    assert config["read_format"] == "csv"
+    assert config["write_format"] == "parquet"
+    assert config["merge_output"] is False
+    assert config["selected_columns"] == ["sequenceId", "itemPosition", "itemId"]
+    assert config["data_columns"] == ["itemId"]
+    assert config["split_ratios"] == [1.0]
+    assert config["stride_by_split"] == [2]
+    assert config["max_rows"] == 2
+    assert config["process_by_file"] is True
+    assert config["subsequence_start_mode"] == "exact"
+    assert config["mask_column"] is None
+    assert config["metadata_config_path"] is None
+    assert config["use_precomputed_maps"] is None
+    assert manifest["effective_metadata_digest"]["algorithm"] == "sha256"
+    assert len(manifest["effective_metadata_digest"]["value"]) == 64
+
+
+def test_resume_manifest_rejects_changed_preprocessing_config(tmp_path):
+    project_root = tmp_path / "project"
+    manifest_dir = project_root / "data" / "input-temp"
+    manifest_dir.mkdir(parents=True)
+
+    preprocessor = object.__new__(Preprocessor)
+    preprocessor.project_root = str(project_root)
+    preprocessor.target_dir = "input-temp"
+    preprocessor.continue_preprocessing = False
+    preprocessor.storage_layout = StoredWindowLayout(
+        stored_context_width=3,
+        max_target_offset=1,
+        version=2,
+    )
+    preprocessor.data_path = "data/input.csv"
+    preprocessor.data_name_root = "input"
+    preprocessor.read_format = "csv"
+    preprocessor.merge_output = False
+    preprocessor.split_ratios = [1.0]
+    preprocessor.stride_by_split = [1]
+    preprocessor.max_rows = None
+    preprocessor.process_by_file = True
+    preprocessor.subsequence_start_mode = "distribute"
+    preprocessor.mask_column = None
+    preprocessor.metadata_config_path = None
+    preprocessor.use_precomputed_maps = None
+
+    preprocessor._write_or_validate_resume_manifest(
+        selected_columns=None,
+        write_format="parquet",
+        data_columns=["itemId"],
+        id_maps={"itemId": {"a": 3}},
+        n_classes={"itemId": 4},
+        col_types={"itemId": "Int64"},
+        selected_columns_statistics={},
+    )
+
+    preprocessor.continue_preprocessing = True
+    preprocessor.stride_by_split = [2]
+
+    with pytest.raises(ValueError, match="different preprocessing manifest"):
+        preprocessor._write_or_validate_resume_manifest(
+            selected_columns=None,
+            write_format="parquet",
+            data_columns=["itemId"],
+            id_maps={"itemId": {"a": 3}},
+            n_classes={"itemId": 4},
+            col_types={"itemId": "Int64"},
+            selected_columns_statistics={},
+        )
+
+
 def test_load_and_preprocess_data_requests_mask_column_for_csv_projection(tmp_path):
     data_path = tmp_path / "masked-input.csv"
     captured_columns = None

From da7c9346a13833b1897334bbd06b60de0748dafa Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 18:53:23 +0200
Subject: [PATCH 56/81] Correctly weighted validation loss

---
 .../sequifier_dataset_from_folder_parquet.py  |  23 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  13 +-
 .../io/sequifier_dataset_from_folder_pt.py    |  24 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  13 +-
 src/sequifier/make.py                         |   6 +-
 src/sequifier/train.py                        | 538 +++++++++++-------
 tests/unit/test_train.py                      |  81 ++-
 7 files changed, 462 insertions(+), 236 deletions(-)

diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index e1d060a7..17414dcf 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -189,22 +189,36 @@ def __iter__(
 
         # 2. Slice metrics based on GPU distribution metrics
         indices_for_rank = indices[rank::world_size].tolist()
+        sample_is_real = [True] * len(indices_for_rank)
 
         # 3. Synchronize cross-device oversampling/undersampling rules
         if self.sampling_strategy == "oversampling":
+            real_count = len(indices_for_rank)
+            if real_count == 0:
+                fallback_indices = indices.tolist()
+                while len(indices_for_rank) < self.target_samples and fallback_indices:
+                    n = min(
+                        len(fallback_indices),
+                        self.target_samples - len(indices_for_rank),
+                    )
+                    indices_for_rank.extend(fallback_indices[:n])
+                    sample_is_real.extend([False] * n)
             while len(indices_for_rank) < self.target_samples:
-                indices_for_rank.extend(
-                    indices_for_rank[: self.target_samples - len(indices_for_rank)]
-                )
+                n = min(real_count, self.target_samples - len(indices_for_rank))
+                indices_for_rank.extend(indices_for_rank[:n])
+                sample_is_real.extend([False] * n)
         elif self.sampling_strategy == "undersampling":
             indices_for_rank = indices_for_rank[: self.target_samples]
+            sample_is_real = sample_is_real[: self.target_samples]
 
         # 4. Map worker task splits
         indices_for_worker = indices_for_rank[worker_id::num_workers]
+        sample_is_real_for_worker = sample_is_real[worker_id::num_workers]
 
         # 5. Extract and pass unified data frames
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
+            batch_sample_is_real = sample_is_real_for_worker[i : i + self.batch_size]
 
             data_batch = {
                 key: tensor[batch_indices] for key, tensor in self.sequences.items()
@@ -218,6 +232,9 @@ def __iter__(
                 metadata_batch = self.resolved_view.build_masks(
                     self.left_pad_lengths[batch_indices]
                 )
+            metadata_batch["sample_valid_mask"] = torch.tensor(
+                batch_sample_is_real, dtype=torch.bool
+            )
 
             yield SequifierBatch(
                 inputs=data_batch,
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 06ee64bb..2d03ffc4 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -171,7 +171,11 @@ def __iter__(
 
         # 1. Distribute files among ranks
         num_files = len(self.batch_files_info)
-        files_for_this_rank = list(range(rank, num_files, world_size))
+        original_files_for_this_rank = list(range(rank, num_files, world_size))
+        rank_real_samples = sum(
+            self.batch_files_info[i]["samples"] for i in original_files_for_this_rank
+        )
+        files_for_this_rank = original_files_for_this_rank.copy()
 
         if not files_for_this_rank:
             if self.sampling_strategy == "oversampling":
@@ -260,6 +264,12 @@ def __iter__(
             worker_file_end_idx = min(file_samples, worker_end_sample - file_start)
 
             worker_indices = indices[worker_file_start_idx:worker_file_end_idx]
+            logical_positions = torch.arange(
+                file_start + worker_file_start_idx,
+                file_start + worker_file_end_idx,
+                dtype=torch.int64,
+            )
+            sample_is_real = logical_positions < rank_real_samples
 
             num_new_samples = len(worker_indices)
 
@@ -306,6 +316,7 @@ def __iter__(
                     )
 
             new_meta = self.resolved_view.build_masks(left_pad_lengths[worker_indices])
+            new_meta["sample_valid_mask"] = sample_is_real
 
             del df
 
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index a3f93d73..aff09a89 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -149,23 +149,36 @@ def __iter__(
 
         # 2. Distribute across GPU ranks
         indices_for_rank = indices[rank::world_size].tolist()
+        sample_is_real = [True] * len(indices_for_rank)
 
         # 3. Handle FSDP oversampling/undersampling sync requirements
         if self.sampling_strategy == "oversampling":
-            # Loop the indices to pad out the shorter ranks
+            real_count = len(indices_for_rank)
+            if real_count == 0:
+                fallback_indices = indices.tolist()
+                while len(indices_for_rank) < self.target_samples and fallback_indices:
+                    n = min(
+                        len(fallback_indices),
+                        self.target_samples - len(indices_for_rank),
+                    )
+                    indices_for_rank.extend(fallback_indices[:n])
+                    sample_is_real.extend([False] * n)
             while len(indices_for_rank) < self.target_samples:
-                indices_for_rank.extend(
-                    indices_for_rank[: self.target_samples - len(indices_for_rank)]
-                )
+                n = min(real_count, self.target_samples - len(indices_for_rank))
+                indices_for_rank.extend(indices_for_rank[:n])
+                sample_is_real.extend([False] * n)
         elif self.sampling_strategy == "undersampling":
             indices_for_rank = indices_for_rank[: self.target_samples]
+            sample_is_real = sample_is_real[: self.target_samples]
 
         # 4. Distribute among CPU workers for this GPU
         indices_for_worker = indices_for_rank[worker_id::num_workers]
+        sample_is_real_for_worker = sample_is_real[worker_id::num_workers]
 
         # 5. Yield full batches
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
+            batch_sample_is_real = sample_is_real_for_worker[i : i + self.batch_size]
 
             data_batch = {
                 key: tensor[batch_indices, self.resolved_view.input_slice]
@@ -183,6 +196,9 @@ def __iter__(
                 metadata_batch = self.resolved_view.build_masks(
                     self.left_pad_lengths[batch_indices]
                 )
+            metadata_batch["sample_valid_mask"] = torch.tensor(
+                batch_sample_is_real, dtype=torch.bool
+            )
 
             yield SequifierBatch(
                 inputs=data_batch,
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index f11fe985..8193b793 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -163,7 +163,11 @@ def __iter__(
 
         # 1. Distribute files among ranks
         num_files = len(self.batch_files_info)
-        files_for_this_rank = list(range(rank, num_files, world_size))
+        original_files_for_this_rank = list(range(rank, num_files, world_size))
+        rank_real_samples = sum(
+            self.batch_files_info[i]["samples"] for i in original_files_for_this_rank
+        )
+        files_for_this_rank = original_files_for_this_rank.copy()
 
         if not files_for_this_rank:
             if self.sampling_strategy == "oversampling":
@@ -254,6 +258,12 @@ def __iter__(
             worker_file_end_idx = min(file_samples, worker_end_sample - file_start)
 
             worker_indices = indices[worker_file_start_idx:worker_file_end_idx]
+            logical_positions = torch.arange(
+                file_start + worker_file_start_idx,
+                file_start + worker_file_end_idx,
+                dtype=torch.int64,
+            )
+            sample_is_real = logical_positions < rank_real_samples
             num_new_samples = len(worker_indices)
 
             if num_new_samples == 0:
@@ -275,6 +285,7 @@ def __iter__(
             new_meta = self.resolved_view.build_masks(
                 left_pad_lengths_batch[worker_indices]
             )
+            new_meta["sample_valid_mask"] = sample_is_real
 
             # Free the large file immediately to keep RAM down
             del sequences_batch, left_pad_lengths_batch
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 5b7da268..c5bab86f 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -13,10 +13,6 @@
 - 0.1
 stored_context_width: 49
 max_target_offset: 1
-stride_by_split:
-- 1
-- 1
-- 1
 max_rows: null
 """
 
@@ -52,7 +48,7 @@
   device: cuda
   epochs: 10
   save_interval_epochs: 10
-  batch_size: 100
+  batch_size: 10
   log_interval: 10
   learning_rate: 0.0001
   accumulation_steps: 1
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index bb78c930..a9c41f7a 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -499,6 +499,27 @@ def format_number(number: Union[int, float, np.float32]) -> str:
     return f"{number_adjusted:5.2f}e{order_of_magnitude}"
 
 
+def _get_evaluation_loss_mask(metadata: dict[str, Tensor]) -> Tensor:
+    valid_mask = metadata["target_valid_mask"].bool()
+
+    if "bert_mask" in metadata:
+        valid_mask = valid_mask & metadata["bert_mask"].bool()
+
+    if "sample_valid_mask" in metadata:
+        sample_valid_mask = metadata["sample_valid_mask"].bool()
+
+        if sample_valid_mask.ndim != 1:
+            raise ValueError("sample_valid_mask must have shape [batch_size].")
+        if sample_valid_mask.shape[0] != valid_mask.shape[0]:
+            raise ValueError(
+                "sample_valid_mask batch dimension does not match target_valid_mask."
+            )
+
+        valid_mask = valid_mask & sample_valid_mask.unsqueeze(1)
+
+    return valid_mask
+
+
 class TransformerEmbeddingModel(nn.Module):
     """A wrapper around the TransformerModel to expose the embedding functionality."""
 
@@ -1730,40 +1751,41 @@ def _calculate_loss(
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
         valid_mask = metadata["target_valid_mask"].bool()  # type: ignore
-
         if "bert_mask" in metadata:
             valid_mask = valid_mask & metadata["bert_mask"].bool()
 
         mask = valid_mask.T.contiguous().reshape(-1)
+        token_count = mask.sum(dtype=torch.int64)
 
         losses = {}
         for target_column in target_names:
             target_column_type = self.target_column_types[target_column]
             if target_column_type == "categorical":
-                output[target_column] = (
+                output_tensor = (
                     output[target_column]
                     .float()
                     .reshape(-1, self.n_classes[target_column])
                 )
             elif target_column_type == "real":
-                output[target_column] = (
+                output_tensor = (
                     output[target_column].to(dtype=torch.float32).reshape(-1)
                 )
+            else:
+                raise ValueError(
+                    f"Target column type {target_column_type} not in ['categorical', 'real']"
+                )
 
             target_tensor = targets[target_column].T.contiguous().reshape(-1)
 
             if self.target_column_types[target_column] == "real":
-                target_tensor = target_tensor.to(dtype=output[target_column].dtype)
+                target_tensor = target_tensor.to(dtype=output_tensor.dtype)
 
-            raw_loss = self.criterion[target_column](
-                output[target_column], target_tensor
-            )
+            raw_loss = self.criterion[target_column](output_tensor, target_tensor)
 
             current_mask = mask.to(dtype=raw_loss.dtype)
+            denominator = token_count.clamp_min(1).to(dtype=raw_loss.dtype)
 
-            losses[target_column] = (raw_loss * current_mask).sum() / (
-                current_mask.sum() + 1e-9
-            )
+            losses[target_column] = (raw_loss * current_mask).sum() / denominator
 
         loss = None
         for target_column in target_names:
@@ -1784,6 +1806,57 @@ def _calculate_loss(
 
         return loss, losses
 
+    @beartype
+    def _calculate_loss_components(
+        self,
+        output: dict[str, Tensor],
+        targets: dict[str, Tensor],
+        valid_mask: Tensor,
+    ) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
+        target_names = [k for k in targets.keys() if k in self.target_column_types]
+        if not target_names:
+            raise RuntimeError("Loss calculation failed; no target columns were found.")
+
+        mask = valid_mask.bool().T.contiguous().reshape(-1)
+        token_count = mask.sum(dtype=torch.int64)
+
+        loss_sums = {}
+        token_counts = {}
+        for target_column in target_names:
+            target_column_type = self.target_column_types[target_column]
+            if target_column_type == "categorical":
+                output_tensor = (
+                    output[target_column]
+                    .float()
+                    .reshape(-1, self.n_classes[target_column])
+                )
+            elif target_column_type == "real":
+                output_tensor = (
+                    output[target_column].to(dtype=torch.float32).reshape(-1)
+                )
+            else:
+                raise ValueError(
+                    f"Target column type {target_column_type} not in ['categorical', 'real']"
+                )
+
+            target_tensor = targets[target_column].T.contiguous().reshape(-1)
+
+            if self.target_column_types[target_column] == "real":
+                target_tensor = target_tensor.to(dtype=output_tensor.dtype)
+
+            raw_loss = self.criterion[target_column](output_tensor, target_tensor)
+            current_mask = mask.to(dtype=raw_loss.dtype)
+
+            loss_sum = (raw_loss * current_mask).sum(dtype=torch.float64)
+            loss_sums[target_column] = loss_sum * (
+                self.loss_weights[target_column]
+                if self.loss_weights is not None
+                else 1.0
+            )
+            token_counts[target_column] = token_count
+
+        return loss_sums, token_counts
+
     @beartype
     def _copy_model(self):
         """Copies the model.
@@ -1853,232 +1926,246 @@ def _evaluate(
             - The output tensor dictionary from the last batch (used for class share logging).
         """
 
-        total_loss_collect = []
-        # Initialize a dict to hold lists of losses for each target
-        total_losses_collect = {col: [] for col in self.target_columns}
+        model_to_call = ddp_model if ddp_model is not None else self
+        target_names = list(self.target_columns)
         output = {}  # for type checking
 
-        model_to_call = ddp_model if ddp_model is not None else self
+        def new_accumulators() -> tuple[dict[str, Tensor], dict[str, Tensor]]:
+            return (
+                {
+                    col: torch.zeros((), device=self.device, dtype=torch.float64)
+                    for col in target_names
+                },
+                {
+                    col: torch.zeros((), device=self.device, dtype=torch.float64)
+                    for col in target_names
+                },
+            )
 
-        model_to_call.eval()
+        def accumulate_components(
+            sums: dict[str, Tensor],
+            counts: dict[str, Tensor],
+            batch_sums: dict[str, Tensor],
+            batch_counts: dict[str, Tensor],
+        ) -> None:
+            for col in batch_sums:
+                sums[col] = sums[col] + batch_sums[col].detach().double()
+                counts[col] = counts[col] + batch_counts[col].detach().double()
+
+        def finalize_components(
+            sums: dict[str, Tensor],
+            counts: dict[str, Tensor],
+            label: str,
+        ) -> tuple[Tensor, dict[str, Tensor]]:
+            packed = torch.stack(
+                [sums[col] for col in target_names]
+                + [counts[col] for col in target_names]
+            )
 
-        with torch.no_grad():
-            for batch_idx, batch in enumerate(valid_loader):
-                if not isinstance(batch, SequifierBatch):
-                    raise TypeError(
-                        "Validation DataLoader must yield SequifierBatch objects, "
-                        f"got {type(batch).__name__}."
-                    )
-                data = batch.inputs
-                targets = batch.targets
-                metadata = batch.metadata
-                # Move data to the current process's assigned GPU
-                data = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in data.items()
-                    if k in self.input_columns
-                }
-                targets = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in targets.items()
-                    if k in self.target_column_types
-                }
-                metadata = {
-                    k: v.to(self.device, non_blocking=True) for k, v in metadata.items()
-                }
-                if self.hparams.training_spec.training_objective == "bert":
-                    data, targets, metadata = apply_bert_masking(
-                        data,
-                        targets,
-                        metadata,
-                        self.hparams,
-                        eval_seed=self.hparams.seed + batch_idx,
+            if self.hparams.training_spec.distributed:
+                dist.all_reduce(packed, op=dist.ReduceOp.SUM)
+
+            n_targets = len(target_names)
+            reduced_sums = dict(zip(target_names, packed[:n_targets]))
+            reduced_counts = dict(zip(target_names, packed[n_targets:]))
+
+            losses = {}
+            total = torch.zeros((), device=self.device, dtype=torch.float64)
+            for col in target_names:
+                if reduced_counts[col].detach().cpu().item() == 0:
+                    raise RuntimeError(
+                        f"No valid {label} tokens found for target column {col!r}."
                     )
+                losses[col] = reduced_sums[col] / reduced_counts[col]
+                total = total + losses[col]
+            return total, losses
 
-                if (
-                    self.hparams.training_spec.layer_autocast
-                    and self.hparams.training_spec.data_parallelism != "FSDP"
-                ):
-                    amp_dtype = get_torch_dtype(
-                        self.hparams.training_spec.layer_type_dtypes.get(
-                            "linear", "bfloat16"
+        was_training = model_to_call.training
+        model_to_call.eval()
+
+        try:
+            total_loss_sums, total_loss_counts = new_accumulators()
+
+            with torch.no_grad():
+                for batch_idx, batch in enumerate(valid_loader):
+                    if not isinstance(batch, SequifierBatch):
+                        raise TypeError(
+                            "Validation DataLoader must yield SequifierBatch objects, "
+                            f"got {type(batch).__name__}."
                         )
-                        if self.hparams.training_spec.layer_type_dtypes
-                        else "bfloat16"
-                    )
-                    with torch.autocast(
-                        device_type=self.device.split(":")[0], dtype=amp_dtype
+                    data = batch.inputs
+                    targets = batch.targets
+                    metadata = batch.metadata
+                    # Move data to the current process's assigned GPU
+                    data = {
+                        k: v.to(self.device, non_blocking=True)
+                        for k, v in data.items()
+                        if k in self.input_columns
+                    }
+                    targets = {
+                        k: v.to(self.device, non_blocking=True)
+                        for k, v in targets.items()
+                        if k in self.target_column_types
+                    }
+                    metadata = {
+                        k: v.to(self.device, non_blocking=True)
+                        for k, v in metadata.items()
+                    }
+                    if self.hparams.training_spec.training_objective == "bert":
+                        data, targets, metadata = apply_bert_masking(
+                            data,
+                            targets,
+                            metadata,
+                            self.hparams,
+                            eval_seed=self.hparams.seed + batch_idx,
+                        )
+
+                    valid_mask = _get_evaluation_loss_mask(metadata)
+
+                    if (
+                        self.hparams.training_spec.layer_autocast
+                        and self.hparams.training_spec.data_parallelism != "FSDP"
                     ):
+                        amp_dtype = get_torch_dtype(
+                            self.hparams.training_spec.layer_type_dtypes.get(
+                                "linear", "bfloat16"
+                            )
+                            if self.hparams.training_spec.layer_type_dtypes
+                            else "bfloat16"
+                        )
+                        with torch.autocast(
+                            device_type=self.device.split(":")[0], dtype=amp_dtype
+                        ):
+                            output = model_to_call(
+                                data, metadata=metadata, return_logits=True
+                            )
+                            loss_sums, token_counts = self._calculate_loss_components(
+                                output, targets, valid_mask
+                            )
+                    else:
                         output = model_to_call(
                             data, metadata=metadata, return_logits=True
                         )
-                        loss, losses = self._calculate_loss(output, targets, metadata)
-                else:
-                    output = model_to_call(data, metadata=metadata, return_logits=True)
-                    loss, losses = self._calculate_loss(output, targets, metadata)
-
-                total_loss_collect.append(loss.item())
-                for col, loss in losses.items():
-                    total_losses_collect[col].append(loss.item())
-
-                # Free up GPU memory
-                del data, targets, metadata, loss, losses
-                if self.device == "cuda":
-                    torch.cuda.empty_cache()
-
-        if len(total_loss_collect) > 0:
-            total_loss_local = np.mean(total_loss_collect)
-            total_losses_local = {
-                col: np.mean(loss_list)
-                for col, loss_list in total_losses_collect.items()
-            }
-        else:
-            # Handle empty validation set case
-            total_loss_local = 0.0
-            total_losses_local = {col: 0.0 for col in self.target_columns}
+                        loss_sums, token_counts = self._calculate_loss_components(
+                            output, targets, valid_mask
+                        )
 
-        # 2. Aggregate losses across all GPUs if in distributed mode
-        if self.hparams.training_spec.distributed:
-            # Put local losses into tensors for reduction
-            total_loss_tensor = torch.tensor(
-                total_loss_local, device=self.device, dtype=torch.float32
-            )
+                    accumulate_components(
+                        total_loss_sums,
+                        total_loss_counts,
+                        loss_sums,
+                        token_counts,
+                    )
 
-            # Ensure consistent order for the losses tensor
-            loss_keys = sorted(total_losses_local.keys())
-            losses_values = [total_losses_local[k] for k in loss_keys]
-            losses_tensor = torch.tensor(
-                losses_values, device=self.device, dtype=torch.float32
+            total_loss_global, total_losses_global = finalize_components(
+                total_loss_sums, total_loss_counts, "validation"
             )
 
-            # Sum losses from all processes. The result is broadcast back to all processes.
-            dist.all_reduce(total_loss_tensor, op=dist.ReduceOp.SUM)
-            dist.all_reduce(losses_tensor, op=dist.ReduceOp.SUM)
+            # Handle one-time baseline loss calculation with the same aggregation semantics.
+            if not hasattr(self, "baseline_loss"):
+                baseline_loss_sums, baseline_loss_counts = new_accumulators()
 
-            world_size = dist.get_world_size()
-            total_loss_tensor /= world_size
-            losses_tensor /= world_size
-
-            # Update local variables with the aggregated global results
-            total_loss_global = total_loss_tensor.cpu().numpy()
-            losses_global_values = losses_tensor.cpu().numpy()
-            total_losses_global = dict(zip(loss_keys, losses_global_values))
-        else:
-            # If not distributed, local losses are the global losses
-            total_loss_global = total_loss_local
-            total_losses_global = total_losses_local
-
-        # 3. Handle one-time baseline loss calculation (must also be synchronized)
-        if not hasattr(self, "baseline_loss"):
-            baseline_loss_local_collect = []
-            baseline_losses_local_collect = {col: [] for col in self.target_columns}
-
-            # Iterate over the sharded validation loader
-            for batch_idx, batch in enumerate(valid_loader):
-                if not isinstance(batch, SequifierBatch):
-                    raise TypeError(
-                        "Validation DataLoader must yield SequifierBatch objects, "
-                        f"got {type(batch).__name__}."
-                    )
-                data = batch.inputs
-                targets = batch.targets
-                metadata = batch.metadata
-                data = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in data.items()
-                    if k in self.input_columns
-                }
-                targets = {
-                    k: v.to(self.device, non_blocking=True)
-                    for k, v in targets.items()
-                    if k in self.target_column_types
-                }
-                metadata = {
-                    k: v.to(self.device, non_blocking=True) for k, v in metadata.items()
-                }
+                with torch.no_grad():
+                    for batch_idx, batch in enumerate(valid_loader):
+                        if not isinstance(batch, SequifierBatch):
+                            raise TypeError(
+                                "Validation DataLoader must yield SequifierBatch objects, "
+                                f"got {type(batch).__name__}."
+                            )
+                        data = batch.inputs
+                        targets = batch.targets
+                        metadata = batch.metadata
+                        data = {
+                            k: v.to(self.device, non_blocking=True)
+                            for k, v in data.items()
+                            if k in self.input_columns
+                        }
+                        targets = {
+                            k: v.to(self.device, non_blocking=True)
+                            for k, v in targets.items()
+                            if k in self.target_column_types
+                        }
+                        metadata = {
+                            k: v.to(self.device, non_blocking=True)
+                            for k, v in metadata.items()
+                        }
 
-                if self.hparams.training_spec.training_objective == "bert":
-                    _, _, metadata = apply_bert_masking(
-                        data,
-                        targets,
-                        metadata,
-                        self.hparams,
-                        eval_seed=self.hparams.seed + batch_idx,
-                    )
+                        if self.hparams.training_spec.training_objective == "bert":
+                            _, _, metadata = apply_bert_masking(
+                                data,
+                                targets,
+                                metadata,
+                                self.hparams,
+                                eval_seed=self.hparams.seed + batch_idx,
+                            )
 
-                pseudo_output = {}
-                targets_for_baseline = {}
-                for col in self.target_columns:
-                    if col in targets:
-                        if self.hparams.training_spec.training_objective == "causal":
-                            pseudo_output[col] = self._transform_val(
-                                col, data[col].transpose(0, 1)
+                        valid_mask = _get_evaluation_loss_mask(metadata)
+
+                        pseudo_output = {}
+                        targets_for_baseline = {}
+                        for col in self.target_columns:
+                            if col in targets:
+                                if (
+                                    self.hparams.training_spec.training_objective
+                                    == "causal"
+                                ):
+                                    pseudo_output[col] = self._transform_val(
+                                        col, data[col].transpose(0, 1)
+                                    )
+                                elif (
+                                    self.hparams.training_spec.training_objective
+                                    == "bert"
+                                ):
+                                    shifted_targets = torch.roll(
+                                        targets[col], shifts=1, dims=1
+                                    )
+
+                                    if self.target_column_types[col] == "categorical":
+                                        shifted_targets[:, 0] = (
+                                            SPECIAL_TOKEN_IDS.unknown
+                                        )
+                                    else:
+                                        shifted_targets[:, 0] = 0.0
+                                    pseudo_output[col] = self._transform_val(
+                                        col, shifted_targets.transpose(0, 1)
+                                    )
+                                else:
+                                    raise ValueError("Impossible")
+                                targets_for_baseline[col] = targets[col]
+
+                        if len(pseudo_output) > 0:
+                            loss_sums, token_counts = self._calculate_loss_components(
+                                pseudo_output,
+                                targets_for_baseline,
+                                valid_mask,
                             )
-                        elif self.hparams.training_spec.training_objective == "bert":
-                            shifted_targets = torch.roll(targets[col], shifts=1, dims=1)
-
-                            if self.target_column_types[col] == "categorical":
-                                shifted_targets[:, 0] = SPECIAL_TOKEN_IDS.unknown
-                            else:
-                                shifted_targets[:, 0] = 0.0
-                            pseudo_output[col] = self._transform_val(
-                                col, shifted_targets.transpose(0, 1)
+                            accumulate_components(
+                                baseline_loss_sums,
+                                baseline_loss_counts,
+                                loss_sums,
+                                token_counts,
                             )
-                        else:
-                            raise ValueError("Impossible")
-                        targets_for_baseline[col] = targets[col]
-
-                if len(pseudo_output) > 0:
-                    loss, losses = self._calculate_loss(
-                        pseudo_output, targets_for_baseline, metadata
-                    )
-                    baseline_loss_local_collect.append(loss.item())
-                    for col, loss_ in losses.items():
-                        baseline_losses_local_collect[col].append(loss_.item())
-
-            # Sum the losses for the local shard
-            if len(baseline_loss_local_collect):
-                baseline_loss_local = np.mean(baseline_loss_local_collect)
-                baseline_losses_local = {
-                    col: np.mean(loss_list)
-                    for col, loss_list in baseline_losses_local_collect.items()
-                }
-            else:
-                baseline_loss_local = -1.0
-                baseline_losses_local = {col: -1.0 for col in self.target_columns}
 
-            # Broadcast the baseline values from the main process to all others
-            if self.hparams.training_spec.distributed:
-                total_loss_tensor = torch.tensor(
-                    baseline_loss_local, device=self.device, dtype=torch.float32
+                baseline_loss, baseline_losses = finalize_components(
+                    baseline_loss_sums, baseline_loss_counts, "baseline validation"
                 )
-                dist.all_reduce(total_loss_tensor, op=dist.ReduceOp.SUM)
-                loss_keys = sorted(baseline_losses_local.keys())
-                losses_values = [baseline_losses_local[k] for k in loss_keys]
-                losses_tensor = torch.tensor(
-                    losses_values, device=self.device, dtype=torch.float32
-                )
-                dist.all_reduce(losses_tensor, op=dist.ReduceOp.SUM)
-
-                world_size = dist.get_world_size()
-                total_loss_tensor /= world_size
-                losses_tensor /= world_size
-
-                self.baseline_loss = total_loss_tensor.item()
-                self.baseline_losses = dict(zip(loss_keys, losses_tensor.cpu().numpy()))
-            else:
-                # If not distributed, local is global
-                self.baseline_loss = baseline_loss_local
-                self.baseline_losses = baseline_losses_local
-
-        model_to_call.train()
-        torch.clear_autocast_cache()
+                self.baseline_loss = baseline_loss.detach().cpu().item()
+                self.baseline_losses = {
+                    col: loss.detach().cpu().item()
+                    for col, loss in baseline_losses.items()
+                }
 
-        return (
-            np.float32(total_loss_global),
-            {k: np.float32(v) for k, v in total_losses_global.items()},
-            output,
-        )
+            return (
+                np.float32(total_loss_global.detach().cpu().item()),
+                {
+                    k: np.float32(v.detach().cpu().item())
+                    for k, v in total_losses_global.items()
+                },
+                output,
+            )
+        finally:
+            model_to_call.train(was_training)
+            torch.clear_autocast_cache()
 
     @beartype
     def _export(
@@ -2457,20 +2544,29 @@ class share statistics (if configured) to the log file.
 
             for categorical_column in self.class_share_log_columns:
                 output_values = (
-                    output[categorical_column].argmax(1).cpu().detach().numpy()
+                    output[categorical_column]
+                    .argmax(dim=-1)  # class dimension
+                    .reshape(-1)  # scalar category ID per token
+                    .detach()
+                    .cpu()
+                    .numpy()
                 )
+
                 output_counts_df = (
-                    pl.Series("values", output_values).value_counts().sort("values")
+                    pl.Series("values", output_values)
+                    .value_counts()
+                    .with_columns(
+                        (pl.col("count") / pl.col("count").sum()).alias("share")
+                    )
+                    .sort("values")
                 )
-                output_counts = output_counts_df.get_column("count")
 
-                output_counts = output_counts / output_counts.sum()
                 value_shares = " | ".join(
-                    [
-                        f"{self.index_maps[categorical_column][row['values']]}: {row['count']:5.5f}"
-                        for row in output_counts_df.iter_rows(named=True)
-                    ]
+                    f"{self.index_maps[categorical_column][int(row['values'])]}: "
+                    f"{row['share']:5.5f}"
+                    for row in output_counts_df.iter_rows(named=True)
                 )
+
                 self.logger.info(f"[INFO] {categorical_column}: {value_shares}")
 
             self.logger.info("-" * 89)
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 872b2e76..e2527ab6 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -17,7 +17,7 @@
 )
 from sequifier.helpers import ModelWindowView, StoredWindowLayout
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS
-from sequifier.train import TransformerModel
+from sequifier.train import TransformerModel, _get_evaluation_loss_mask
 
 
 def _training_spec_kwargs(**overrides):
@@ -479,6 +479,85 @@ def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
     assert torch.isclose(component_losses["real_target"], torch.tensor(1.0))
 
 
+def test_evaluation_loss_mask_intersects_target_bert_and_sample_masks():
+    metadata = {
+        "target_valid_mask": torch.tensor(
+            [
+                [True, True, False],
+                [True, True, True],
+            ]
+        ),
+        "bert_mask": torch.tensor(
+            [
+                [True, False, True],
+                [True, True, True],
+            ]
+        ),
+        "sample_valid_mask": torch.tensor([True, False]),
+    }
+
+    loss_mask = _get_evaluation_loss_mask(metadata)
+
+    assert torch.equal(
+        loss_mask,
+        torch.tensor(
+            [
+                [True, False, False],
+                [False, False, False],
+            ]
+        ),
+    )
+
+
+def test_calculate_loss_components_aggregate_by_token_count():
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_column_types = {"real_col": "real"}
+    model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
+    model.loss_weights = {"real_col": 1.0}
+
+    first_sums, first_counts = TransformerModel._calculate_loss_components(
+        model,
+        {"real_col": torch.zeros(1, 1, 1)},
+        {"real_col": torch.tensor([[10.0]])},
+        torch.ones(1, 1, dtype=torch.bool),
+    )
+    second_sums, second_counts = TransformerModel._calculate_loss_components(
+        model,
+        {"real_col": torch.zeros(100, 1, 1)},
+        {"real_col": torch.ones(1, 100)},
+        torch.ones(1, 100, dtype=torch.bool),
+    )
+
+    aggregate = (first_sums["real_col"] + second_sums["real_col"]) / (
+        first_counts["real_col"] + second_counts["real_col"]
+    )
+
+    assert torch.isclose(aggregate, torch.tensor(200.0 / 101.0, dtype=torch.float64))
+
+
+def test_calculate_loss_zero_token_training_batch_is_differentiable():
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_column_types = {"real_col": "real"}
+    model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
+    model.loss_weights = None
+
+    output = {"real_col": torch.ones(3, 1, 1, requires_grad=True)}
+    targets = {"real_col": torch.zeros(1, 3)}
+    metadata = {
+        "attention_valid_mask": torch.zeros(1, 3, dtype=torch.bool),
+        "target_valid_mask": torch.zeros(1, 3, dtype=torch.bool),
+    }
+
+    total_loss, component_losses = TransformerModel._calculate_loss(
+        model, output, targets, metadata
+    )
+    total_loss.backward()
+
+    assert torch.equal(total_loss.detach(), torch.tensor(0.0))
+    assert torch.equal(component_losses["real_col"].detach(), torch.tensor(0.0))
+    assert torch.equal(output["real_col"].grad, torch.zeros_like(output["real_col"]))
+
+
 def test_padding_keys_are_masked(bert_model):
     seq_len = bert_model.window_view.context_length
 

From da4d0fa14f027f2dcfd6a1f4ce8f1487f0a17703 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 13 Jun 2026 19:07:49 +0200
Subject: [PATCH 57/81] update custom eval int test outputs

---
 .../hyperparameter-search-custom-eval.yaml    |   2 +-
 ...rical-distributed-best-3-6-predictions.csv |   2 +-
 ...al-1-best-3-autoregression-predictions.csv | 398 +++++++++---------
 ...custom-eval-run-0-best-1-0-predictions.csv |  31 --
 ...custom-eval-run-0-best-1-1-predictions.csv |  31 --
 ...custom-eval-run-0-best-1-2-predictions.csv |  31 --
 ...custom-eval-run-0-best-1-3-predictions.csv |  61 ---
 ...custom-eval-run-0-best-1-4-predictions.csv |  31 --
 ...custom-eval-run-0-best-1-5-predictions.csv |  61 ---
 ...custom-eval-run-0-best-1-6-predictions.csv |  31 --
 ...custom-eval-run-0-best-1-7-predictions.csv |  31 --
 ...custom-eval-run-1-best-1-0-predictions.csv |  31 --
 ...custom-eval-run-1-best-1-1-predictions.csv |  31 --
 ...custom-eval-run-1-best-1-2-predictions.csv |  31 --
 ...custom-eval-run-1-best-1-3-predictions.csv |  61 ---
 ...custom-eval-run-1-best-1-4-predictions.csv |  31 --
 ...custom-eval-run-1-best-1-5-predictions.csv |  61 ---
 ...custom-eval-run-1-best-1-6-predictions.csv |  31 --
 ...custom-eval-run-1-best-1-7-predictions.csv |  31 --
 ...custom-eval-run-2-best-1-2-predictions.csv |  31 --
 ...custom-eval-run-2-best-1-3-predictions.csv |  61 ---
 ...custom-eval-run-2-best-1-7-predictions.csv |  31 --
 ...custom-eval-run-3-best-1-0-predictions.csv |  31 --
 ...custom-eval-run-3-best-1-1-predictions.csv |  31 --
 ...custom-eval-run-3-best-1-2-predictions.csv |  31 --
 ...custom-eval-run-3-best-1-3-predictions.csv |  61 ---
 ...custom-eval-run-3-best-1-4-predictions.csv |  31 --
 ...custom-eval-run-3-best-1-5-predictions.csv |  61 ---
 ...custom-eval-run-3-best-1-6-predictions.csv |  31 --
 ...custom-eval-run-3-best-1-7-predictions.csv |  31 --
 ...custom-eval-run-4-best-3-0-predictions.csv |  31 ++
 ...custom-eval-run-4-best-3-1-predictions.csv |  31 ++
 ...custom-eval-run-4-best-3-2-predictions.csv |  31 ++
 ...custom-eval-run-4-best-3-3-predictions.csv |  61 +++
 ...custom-eval-run-4-best-3-4-predictions.csv |  31 ++
 ...custom-eval-run-4-best-3-5-predictions.csv |  61 +++
 ...custom-eval-run-4-best-3-6-predictions.csv |  31 ++
 ...custom-eval-run-4-best-3-7-predictions.csv |  31 ++
 ...custom-eval-run-5-best-3-0-predictions.csv |  31 ++
 ...custom-eval-run-5-best-3-1-predictions.csv |  31 ++
 ...custom-eval-run-5-best-3-2-predictions.csv |  31 ++
 ...custom-eval-run-5-best-3-3-predictions.csv |  61 +++
 ...custom-eval-run-5-best-3-4-predictions.csv |  31 ++
 ...custom-eval-run-5-best-3-5-predictions.csv |  61 +++
 ...custom-eval-run-5-best-3-6-predictions.csv |  31 ++
 ...custom-eval-run-5-best-3-7-predictions.csv |  31 ++
 ...ustom-eval-run-6-best-3-0-predictions.csv} |  16 +-
 ...ustom-eval-run-6-best-3-1-predictions.csv} |   8 +-
 ...custom-eval-run-6-best-3-2-predictions.csv |  31 ++
 ...custom-eval-run-6-best-3-3-predictions.csv |  61 +++
 ...ustom-eval-run-6-best-3-4-predictions.csv} |  12 +-
 ...ustom-eval-run-6-best-3-5-predictions.csv} |  52 +--
 ...ustom-eval-run-6-best-3-6-predictions.csv} |   4 +-
 ...custom-eval-run-6-best-3-7-predictions.csv |  31 ++
 ...custom-eval-run-7-best-3-0-predictions.csv |  31 ++
 ...custom-eval-run-7-best-3-1-predictions.csv |  31 ++
 ...custom-eval-run-7-best-3-2-predictions.csv |  31 ++
 ...custom-eval-run-7-best-3-3-predictions.csv |  61 +++
 ...custom-eval-run-7-best-3-4-predictions.csv |  31 ++
 ...custom-eval-run-7-best-3-5-predictions.csv |  61 +++
 ...custom-eval-run-7-best-3-6-predictions.csv |  31 ++
 ...custom-eval-run-7-best-3-7-predictions.csv |  31 ++
 62 files changed, 1294 insertions(+), 1294 deletions(-)
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv => sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv} (74%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv => sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv} (86%)
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv => sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv} (79%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv => sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv} (55%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv => sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv} (92%)
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv

diff --git a/tests/configs/hyperparameter-search-custom-eval.yaml b/tests/configs/hyperparameter-search-custom-eval.yaml
index 2dfc8d7a..abb6b043 100644
--- a/tests/configs/hyperparameter-search-custom-eval.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval.yaml
@@ -54,7 +54,7 @@ model_hyperparameter_sampling:
 training_hyperparameter_sampling:
   training_objective: ["causal"]
   device: cpu
-  epochs: [1, 1]
+  epochs: [3, 3]
   save_interval_epochs: 10
   batch_size: [5, 10]
   learning_rate: [0.001, 0.0001]
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
index 8eb4c4a0..59349bc1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,4,111
+8,4,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index 4305a322..fe0245bd 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,-0.16518183
-0,22,0.28807208
-0,23,0.26411152
-0,24,0.070929505
-0,25,0.106371164
-0,26,0.08790157
-0,27,0.28008524
-0,28,0.18923476
-0,29,-0.0023990057
-0,30,-0.007035904
-0,31,-0.20611446
-0,32,-0.07283385
-0,33,0.10886706
-0,34,0.016643867
-0,35,-0.0878092
-0,36,0.06843361
-0,37,0.032991957
-0,38,-0.0878092
-0,39,-0.12524757
-0,40,-0.18714568
-1,19,0.18324462
-1,20,-0.041136023
-1,21,0.41785845
-1,22,0.41386503
-1,23,0.3938979
-1,24,0.07841718
-1,25,0.0021052985
-1,26,0.06344183
-1,27,-0.04138561
-1,28,-0.045878217
-1,29,-0.03289958
-1,30,-0.06684371
-1,31,0.10537281
-1,32,-0.058856852
-1,33,-0.14820977
-1,34,-0.13023935
-1,35,0.035737436
-1,36,-0.11326729
-1,37,-0.039139308
-1,38,0.11785226
-2,17,0.34198335
-2,18,0.20121504
-2,19,0.07142868
-2,20,0.03773415
-2,21,0.3060425
-2,22,-0.11426565
-2,23,0.07192786
-2,24,0.013773591
-2,25,-0.036643416
-2,26,-0.027034234
-2,27,-0.021293685
-2,28,-0.034397114
-2,29,-0.09030509
-2,30,-0.15120484
-2,31,-0.18914239
-2,32,-0.18914239
-2,33,0.07991471
-2,34,-0.120754965
-2,35,0.011652084
-2,36,-0.14421634
-3,17,0.11385884
-3,18,0.3240129
-3,19,0.29406223
-3,20,0.094390884
-3,21,0.2251756
-3,22,0.36993733
-3,23,0.057202104
-3,24,0.22417724
-3,25,-0.07283385
-3,26,0.16427584
-3,27,0.05545498
-3,28,-0.22108981
-3,29,-0.06834124
-3,30,-0.083316594
-3,31,0.05021361
-3,32,-0.052117944
-3,33,-0.1222525
-3,34,0.1148572
-3,35,-0.011247721
-3,36,-0.0209193
-4,14,0.25612468
-4,15,0.09139582
-4,16,0.2092019
-4,17,0.17126435
-4,18,0.12234487
-4,19,0.22817066
-4,20,0.04047963
-4,21,-0.14022292
-4,22,0.041727576
-4,23,-0.07033796
-4,24,-0.058856852
-4,25,-0.061352745
-4,26,-0.0067239176
-4,27,-0.012994845
-4,28,-0.12874182
-4,29,0.16128078
-4,30,-0.11925743
-4,31,-0.071336314
-4,32,-0.11726072
-4,33,0.06843361
-5,14,-0.2260816
-5,15,0.026253048
-5,16,0.13232844
-5,17,0.17825285
-5,18,0.35196692
-5,19,0.31203264
-5,20,0.06943197
-5,21,-0.10428208
-5,22,-0.10178619
-5,23,0.019638937
-5,24,0.017018251
-5,25,-0.10577962
-5,26,-0.073333025
-5,27,-0.17716211
-5,28,-0.16917527
-5,29,-0.17217033
-5,30,0.23316245
-5,31,0.04047963
-5,32,-0.06784207
-5,33,0.13432515
-6,18,0.2531296
-6,19,0.19222984
-6,20,0.455796
-6,21,0.3958946
-6,22,-0.015490737
-6,23,-0.17316869
-6,24,-0.109273866
-6,25,-0.16717854
-6,26,0.1333268
-6,27,-0.35486957
-6,28,-0.07283385
-6,29,-0.15320155
-6,30,0.12833501
-6,31,-0.04512945
-6,32,-0.12774347
-6,33,0.142312
-6,34,0.018765375
-6,35,0.14630543
-6,36,-0.10428208
-6,37,-0.062351096
-7,15,-0.022167247
-7,16,0.2601181
-7,17,0.08041389
-7,18,-0.019172177
-7,19,-0.1841506
-7,20,0.068932794
-7,21,0.053707857
-7,22,0.028499352
-7,23,-0.20811117
-7,24,-0.039139308
-7,25,-0.033897936
-7,26,-0.07582892
-7,27,0.06593772
-7,28,-0.067342885
-7,29,-0.21609803
-7,30,-0.19712925
-7,31,0.19222984
-7,32,-0.11825907
-7,33,-0.022167247
-7,34,0.065438546
-8,20,0.18324462
-8,21,-0.14122127
-8,22,-0.018922588
-8,23,0.32001948
-8,24,0.060197175
-8,25,0.070929505
-8,26,-0.03639383
-8,27,0.16727091
-8,28,-0.018049026
-8,29,-0.06634453
-8,30,-0.056111373
-8,31,-0.07383221
-8,32,-0.14920813
-8,33,-0.05361548
-8,34,0.10736952
-8,35,0.06793443
-8,36,0.043973878
-8,37,0.3040458
-8,38,-0.077326454
-8,39,-0.1522032
-9,16,0.3060425
-9,17,0.28807208
-9,18,0.3180228
-9,19,0.0068474924
-9,20,0.25612468
-9,21,0.14331035
-9,22,0.22118217
-9,23,-0.16118841
-9,24,-0.06384864
-9,25,-0.04537904
-9,26,0.18923476
-9,27,0.0664369
-9,28,-0.20611446
-9,29,0.07991471
-9,30,-0.19613089
+0,21,0.03249278
+0,22,0.1493005
+0,23,0.12733665
+0,24,0.21319532
+0,25,0.16727091
+0,26,0.31602606
+0,27,0.15728734
+0,28,-0.1362295
+0,29,-0.1362295
+0,30,-0.17217033
+0,31,-0.13922456
+0,32,-0.15020649
+0,33,0.022259623
+0,34,-0.0992903
+0,35,-0.11825907
+0,36,-0.08631166
+0,37,0.029248118
+0,38,-0.1661802
+0,39,0.27209836
+0,40,-0.037392184
+1,19,-0.04438068
+1,20,0.23515917
+1,21,0.16327749
+1,22,-0.0032530688
+1,23,0.2411493
+1,24,0.0893991
+1,25,0.12084734
+1,26,-0.1751654
+1,27,0.12134651
+1,28,-0.15719499
+1,29,0.04671936
+1,30,-0.05860726
+1,31,-0.051868357
+1,32,0.06993115
+1,33,0.051711142
+1,34,-0.08631166
+1,35,-0.015865121
+1,36,-0.17117198
+1,37,-0.057359315
+1,38,0.015770305
+2,17,0.04996402
+2,18,-0.11176976
+2,19,0.17425941
+2,20,0.19422655
+2,21,-0.021293685
+2,22,-0.021044096
+2,23,0.1078687
+2,24,0.077918
+2,25,0.02899853
+2,26,-0.13922456
+2,27,-0.1751654
+2,28,-0.021792863
+2,29,0.09838431
+2,30,0.19722162
+2,31,-0.20611446
+2,32,-0.009344604
+2,33,-0.08481413
+2,34,-0.038140953
+2,35,0.078916356
+2,36,-0.124748394
+3,17,0.057451695
+3,18,0.2780885
+3,19,0.21020025
+3,20,0.30005237
+3,21,0.23016739
+3,22,0.028000174
+3,23,0.12084734
+3,24,0.17825285
+3,25,-0.114764825
+3,26,-0.0601048
+3,27,-0.07183549
+3,28,-0.037392184
+3,29,-0.13822621
+3,30,-0.13922456
+3,31,-0.101287015
+3,32,-0.0076598767
+3,33,-0.05935603
+3,34,-0.054364245
+3,35,0.04047963
+3,36,-0.048623696
+4,14,0.20820354
+4,15,0.16227913
+4,16,-0.17716211
+4,17,-0.08032152
+4,18,0.02600346
+4,19,0.14830214
+4,20,0.07242704
+4,21,-0.089306735
+4,22,-0.15020649
+4,23,0.036236614
+4,24,-0.27100763
+4,25,0.100381024
+4,26,-0.024039164
+4,27,0.19322819
+4,28,-0.09879112
+4,29,0.008969001
+4,30,-0.05860726
+4,31,-0.14421634
+4,32,0.036236614
+4,33,0.1403153
+5,14,0.26610824
+5,15,0.41586173
+5,16,0.30803922
+5,17,0.21519203
+5,18,0.2601181
+5,19,0.16128078
+5,20,-0.023040809
+5,21,0.027001817
+5,22,-0.16418348
+5,23,-0.0058503556
+5,24,-0.109273866
+5,25,-0.12574674
+5,26,0.13232844
+5,27,0.028499352
+5,28,0.02600346
+5,29,0.008157835
+5,30,0.042975523
+5,31,-0.07882399
+5,32,-0.25703064
+5,33,-0.2350668
+6,18,0.40987158
+6,19,0.11136295
+6,20,0.17126435
+6,21,0.30204907
+6,22,0.20720518
+6,23,0.07941554
+6,24,-0.028157387
+6,25,0.024006747
+6,26,0.015770305
+6,27,-0.13922456
+6,28,0.02575387
+6,29,-0.0043060225
+6,30,-0.08181906
+6,31,-0.052367534
+6,32,0.051211964
+6,33,-0.12025579
+6,34,0.056702927
+6,35,0.19222984
+6,36,0.017392633
+6,37,-0.1522032
+7,15,0.156289
+7,16,0.29805565
+7,17,-0.03289958
+7,18,0.16128078
+7,19,0.06943197
+7,20,-0.057608906
+7,21,-0.07283385
+7,22,0.1952249
+7,23,-0.032275606
+7,24,-0.06484699
+7,25,0.066936076
+7,26,0.023757158
+7,27,-0.13722785
+7,28,-0.057858497
+7,29,0.036236614
+7,30,-0.096794404
+7,31,-0.03864013
+7,32,-0.14721142
+7,33,0.06444018
+7,34,0.033740725
+8,20,-0.012370872
+8,21,0.0412284
+8,22,0.21419367
+8,23,-0.143218
+8,24,0.060696352
+8,25,-0.04762534
+8,26,0.114358015
+8,27,0.07741882
+8,28,0.09738596
+8,29,0.058699638
+8,30,-0.040137667
+8,31,-0.0878092
+8,32,0.059947584
+8,33,0.04147799
+8,34,0.05245991
+8,35,0.10137938
+8,36,-0.05411466
+8,37,-0.054613836
+8,38,-0.034397114
+8,39,0.03848292
+9,16,0.066936076
+9,17,0.25113288
+9,18,0.28807208
+9,19,-0.19213746
+9,20,0.060696352
+9,21,0.2092019
+9,22,0.29406223
+9,23,-0.12424921
+9,24,-0.062351096
+9,25,0.00085735274
+9,26,-0.11626236
+9,27,-0.065845355
+9,28,0.27010167
+9,29,0.0051627657
+9,30,-0.01287005
 9,31,-0.109273866
-9,32,0.07642046
-9,33,-0.11975661
-9,34,-0.018548204
-9,35,-0.2350668
+9,32,-0.097293586
+9,33,-0.055861782
+9,34,0.033740725
+9,35,-0.10228537
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
deleted file mode 100644
index 4e65dc53..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,104,4,9,7,0
-0,5,112,4,9,7,0
-0,6,104,9,9,7,0
-0,7,112,4,9,5,0
-0,8,112,9,9,5,0
-0,9,112,9,9,5,0
-0,10,112,9,9,5,0
-0,11,112,9,9,5,0
-0,12,112,9,9,5,0
-0,13,112,9,9,5,0
-0,14,112,9,9,5,0
-0,15,112,9,9,2,0
-0,16,112,9,9,7,0
-0,17,112,9,9,7,0
-0,18,112,9,9,7,0
-0,19,112,9,9,7,0
-0,20,112,9,9,7,0
-0,21,112,9,9,7,0
-0,22,112,9,9,7,0
-0,23,112,9,9,7,0
-0,24,112,9,9,7,0
-0,25,112,9,9,7,0
-0,26,112,9,9,7,0
-0,27,112,9,9,7,0
-0,28,112,9,9,7,0
-0,29,112,9,9,7,0
-0,30,112,9,9,7,0
-0,31,112,9,9,7,0
-0,32,112,9,9,7,0
-0,33,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
deleted file mode 100644
index bf364d0d..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,111,4,9,7,1
-1,5,111,4,9,7,1
-1,6,111,4,9,7,1
-1,7,111,4,9,7,1
-1,8,111,4,9,7,1
-1,9,111,4,9,7,1
-1,10,111,4,9,7,1
-1,11,111,4,9,7,1
-1,12,102,4,9,7,1
-1,13,102,4,9,7,1
-1,14,102,4,9,7,1
-1,15,102,4,9,7,1
-1,16,122,4,9,5,1
-1,17,122,9,9,2,0
-1,18,121,9,9,5,0
-1,19,121,9,9,5,0
-1,20,112,9,9,2,0
-1,21,112,9,9,2,0
-1,22,112,9,9,2,0
-1,23,112,9,9,2,0
-1,24,112,9,9,7,0
-1,25,112,9,9,7,0
-1,26,112,9,9,7,0
-1,27,112,9,9,7,0
-1,28,112,9,9,7,0
-1,29,112,9,9,7,0
-1,30,112,9,9,7,0
-1,31,112,9,9,7,0
-1,32,112,9,9,7,0
-1,33,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
deleted file mode 100644
index ab10690c..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,112,9,9,7,0
-2,4,112,9,9,2,0
-2,5,112,9,9,2,0
-2,6,112,9,9,2,0
-2,7,112,9,9,2,0
-2,8,112,9,9,2,0
-2,9,112,9,9,7,0
-2,10,112,9,9,7,0
-2,11,112,9,9,7,0
-2,12,112,9,9,7,0
-2,13,112,9,9,7,0
-2,14,112,9,9,7,0
-2,15,112,9,9,7,0
-2,16,112,9,9,7,0
-2,17,112,9,9,7,0
-2,18,112,9,9,7,0
-2,19,112,9,9,7,0
-2,20,112,9,9,7,0
-2,21,112,9,9,7,0
-2,22,112,9,9,7,0
-2,23,112,9,9,7,0
-2,24,112,9,9,7,0
-2,25,112,9,9,7,0
-2,26,112,9,9,7,0
-2,27,112,9,9,7,0
-2,28,112,9,9,7,0
-2,29,112,9,9,7,0
-2,30,112,9,9,7,0
-2,31,112,9,9,7,0
-2,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
deleted file mode 100644
index c3e044e5..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,112,9,9,5,0
-3,4,112,9,9,2,0
-3,5,112,9,9,2,0
-3,6,112,9,9,2,0
-3,7,112,9,9,2,0
-3,8,112,9,9,2,0
-3,9,112,9,9,2,0
-3,10,112,9,9,7,0
-3,11,112,9,9,7,0
-3,12,112,9,9,7,0
-3,13,112,9,9,7,0
-3,14,112,9,9,7,0
-3,15,112,9,9,7,0
-3,16,112,9,9,7,0
-3,17,112,9,9,7,0
-3,18,112,9,9,7,0
-3,19,112,9,9,7,0
-3,20,112,9,9,7,0
-3,21,112,9,9,7,0
-3,22,112,9,9,7,0
-3,23,112,9,9,7,0
-3,24,112,9,9,7,0
-3,25,112,9,9,7,0
-3,26,112,9,9,7,0
-3,27,112,9,9,7,0
-3,28,112,9,9,7,0
-3,29,112,9,9,7,0
-3,30,112,9,9,7,0
-3,31,112,9,9,7,0
-3,32,112,9,9,7,0
-4,3,112,9,9,7,0
-4,4,112,9,9,7,0
-4,5,112,9,9,7,0
-4,6,112,9,9,7,0
-4,7,112,9,9,7,0
-4,8,112,9,9,7,0
-4,9,112,9,9,5,0
-4,10,112,9,9,5,0
-4,11,112,9,9,5,0
-4,12,112,9,9,5,0
-4,13,112,9,9,5,0
-4,14,112,9,9,5,0
-4,15,112,9,9,7,0
-4,16,112,9,9,7,0
-4,17,112,9,9,7,0
-4,18,112,9,9,7,0
-4,19,112,9,9,7,0
-4,20,112,9,9,7,0
-4,21,112,9,9,7,0
-4,22,112,9,9,7,0
-4,23,112,9,9,7,0
-4,24,112,9,9,7,0
-4,25,112,9,9,7,0
-4,26,112,9,9,7,0
-4,27,112,9,9,7,0
-4,28,112,9,9,7,0
-4,29,112,9,9,7,0
-4,30,112,9,9,7,0
-4,31,112,9,9,7,0
-4,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
deleted file mode 100644
index c38b58c0..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,111,3,2,2,0
-5,4,111,4,9,8,1
-5,5,111,4,9,7,1
-5,6,111,4,9,7,1
-5,7,111,4,9,7,1
-5,8,111,4,9,7,1
-5,9,111,4,9,7,1
-5,10,102,4,9,7,1
-5,11,102,4,9,7,1
-5,12,102,4,9,7,1
-5,13,102,4,9,7,1
-5,14,122,4,9,5,1
-5,15,122,9,9,2,0
-5,16,121,9,9,5,0
-5,17,121,9,9,5,0
-5,18,112,9,9,2,0
-5,19,112,9,9,2,0
-5,20,112,9,9,2,0
-5,21,112,9,9,2,0
-5,22,112,9,9,7,0
-5,23,112,9,9,7,0
-5,24,112,9,9,7,0
-5,25,112,9,9,7,0
-5,26,112,9,9,7,0
-5,27,112,9,9,7,0
-5,28,112,9,9,7,0
-5,29,112,9,9,7,0
-5,30,112,9,9,7,0
-5,31,112,9,9,7,0
-5,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
deleted file mode 100644
index d8447e35..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,111,4,9,7,1
-6,5,111,4,9,7,1
-6,6,111,4,9,7,1
-6,7,111,4,9,7,1
-6,8,111,4,9,7,1
-6,9,111,4,9,7,1
-6,10,102,4,9,7,1
-6,11,102,4,9,7,1
-6,12,102,4,9,7,1
-6,13,102,4,9,7,1
-6,14,122,4,9,5,1
-6,15,122,9,9,2,0
-6,16,121,9,9,5,0
-6,17,121,9,9,5,0
-6,18,112,9,9,2,0
-6,19,112,9,9,2,0
-6,20,112,9,9,2,0
-6,21,112,9,9,2,0
-6,22,112,9,9,7,0
-6,23,112,9,9,7,0
-6,24,112,9,9,7,0
-6,25,112,9,9,7,0
-6,26,112,9,9,7,0
-6,27,112,9,9,7,0
-6,28,112,9,9,7,0
-6,29,112,9,9,7,0
-6,30,112,9,9,7,0
-6,31,112,9,9,7,0
-6,32,112,9,9,7,0
-6,33,112,9,9,7,0
-7,3,104,4,9,7,1
-7,4,102,4,9,7,1
-7,5,104,4,9,7,0
-7,6,121,4,9,5,1
-7,7,111,6,9,5,0
-7,8,121,4,9,5,1
-7,9,111,6,9,5,0
-7,10,122,4,9,5,1
-7,11,111,9,9,7,1
-7,12,111,9,9,7,1
-7,13,111,9,9,7,1
-7,14,111,3,9,7,1
-7,15,111,4,9,7,1
-7,16,111,4,9,7,1
-7,17,102,4,9,7,1
-7,18,102,4,9,7,1
-7,19,102,4,9,7,1
-7,20,102,4,9,7,1
-7,21,102,4,9,5,0
-7,22,121,4,9,5,0
-7,23,121,9,9,5,0
-7,24,121,9,9,5,0
-7,25,112,9,9,2,0
-7,26,112,9,9,2,0
-7,27,112,9,9,2,0
-7,28,112,9,9,2,0
-7,29,112,9,9,2,0
-7,30,112,9,9,7,0
-7,31,112,9,9,7,0
-7,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
deleted file mode 100644
index 2ed29b99..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,111,9,9,7,1
-8,5,111,4,9,7,1
-8,6,111,4,9,7,1
-8,7,111,4,9,7,1
-8,8,111,4,9,7,1
-8,9,111,4,9,7,1
-8,10,111,4,9,7,1
-8,11,111,4,9,7,1
-8,12,102,4,9,7,1
-8,13,102,4,9,7,1
-8,14,102,4,9,7,1
-8,15,102,4,9,7,1
-8,16,122,4,9,5,1
-8,17,122,9,9,2,0
-8,18,121,9,9,5,0
-8,19,121,9,9,5,0
-8,20,112,9,9,2,0
-8,21,112,9,9,2,0
-8,22,112,9,9,2,0
-8,23,112,9,9,2,0
-8,24,112,9,9,7,0
-8,25,112,9,9,7,0
-8,26,112,9,9,7,0
-8,27,112,9,9,7,0
-8,28,112,9,9,7,0
-8,29,112,9,9,7,0
-8,30,112,9,9,7,0
-8,31,112,9,9,7,0
-8,32,112,9,9,7,0
-8,33,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
deleted file mode 100644
index bdde68a7..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-predictions/sequifier-test-hp-search-custom-eval-run-0-best-1-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,113,4,2,7,1
-9,4,113,4,9,7,1
-9,5,113,4,9,7,1
-9,6,113,4,9,7,1
-9,7,111,4,9,7,1
-9,8,102,4,9,7,1
-9,9,102,4,9,7,0
-9,10,113,4,9,7,1
-9,11,113,4,9,7,1
-9,12,113,4,9,7,1
-9,13,113,9,9,7,1
-9,14,111,9,9,7,1
-9,15,111,9,9,7,1
-9,16,111,9,9,7,1
-9,17,111,4,9,7,1
-9,18,102,4,9,7,1
-9,19,102,4,9,7,1
-9,20,102,4,9,7,1
-9,21,102,4,9,7,1
-9,22,102,4,9,7,1
-9,23,104,6,9,5,0
-9,24,121,9,9,5,1
-9,25,122,9,9,2,0
-9,26,112,9,9,2,0
-9,27,112,9,9,2,0
-9,28,112,9,9,2,0
-9,29,112,9,9,2,0
-9,30,112,9,9,7,0
-9,31,112,9,9,7,0
-9,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
deleted file mode 100644
index 2132af3a..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,127,1,1,2,1
-0,5,113,4,9,8,1
-0,6,113,1,1,8,1
-0,7,113,1,9,8,1
-0,8,113,1,9,8,1
-0,9,113,1,9,8,1
-0,10,113,1,8,8,1
-0,11,113,1,0,8,1
-0,12,113,1,9,7,1
-0,13,121,1,8,7,1
-0,14,105,1,8,7,1
-0,15,105,1,8,7,1
-0,16,105,1,8,7,1
-0,17,113,1,7,7,1
-0,18,113,6,7,7,1
-0,19,113,6,7,2,1
-0,20,113,1,7,8,1
-0,21,113,1,7,8,1
-0,22,113,1,7,8,1
-0,23,113,1,0,8,1
-0,24,113,1,0,8,1
-0,25,113,1,0,8,1
-0,26,113,1,0,8,1
-0,27,113,1,0,8,1
-0,28,113,1,0,7,1
-0,29,108,1,8,7,1
-0,30,108,1,0,7,1
-0,31,108,6,0,7,1
-0,32,108,1,0,7,1
-0,33,108,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
deleted file mode 100644
index c473bc6b..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,121,6,0,8,0
-1,5,121,1,0,8,0
-1,6,121,1,0,8,0
-1,7,121,1,0,8,0
-1,8,121,1,0,8,0
-1,9,126,1,0,8,0
-1,10,127,1,0,8,0
-1,11,127,0,0,8,0
-1,12,127,0,9,8,0
-1,13,127,0,7,7,0
-1,14,105,8,9,7,1
-1,15,127,0,7,2,1
-1,16,113,4,7,8,1
-1,17,113,1,9,8,1
-1,18,113,1,7,8,1
-1,19,113,1,9,8,1
-1,20,113,1,0,8,1
-1,21,113,1,0,8,1
-1,22,113,1,0,8,1
-1,23,113,1,0,8,1
-1,24,113,1,0,7,1
-1,25,108,1,8,7,1
-1,26,108,1,8,7,1
-1,27,105,6,8,7,1
-1,28,113,1,8,7,1
-1,29,113,6,7,7,1
-1,30,113,6,7,7,1
-1,31,113,6,7,2,1
-1,32,113,1,7,8,1
-1,33,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
deleted file mode 100644
index 4f612da8..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,121,1,8,0,0
-2,4,105,0,8,7,0
-2,5,121,1,8,2,0
-2,6,113,4,7,8,0
-2,7,121,1,9,8,0
-2,8,127,1,0,8,0
-2,9,127,4,0,8,1
-2,10,127,1,0,8,0
-2,11,127,4,7,8,1
-2,12,127,1,0,8,1
-2,13,103,1,0,8,1
-2,14,113,1,0,8,1
-2,15,103,1,9,7,1
-2,16,105,1,7,7,1
-2,17,113,1,9,7,1
-2,18,105,1,9,7,1
-2,19,113,1,7,7,1
-2,20,113,6,9,7,1
-2,21,113,1,7,7,1
-2,22,113,6,5,7,1
-2,23,113,1,8,8,1
-2,24,113,1,0,8,1
-2,25,113,1,0,8,1
-2,26,113,1,0,8,1
-2,27,113,1,0,8,1
-2,28,113,1,0,8,1
-2,29,113,1,0,7,1
-2,30,108,1,8,7,1
-2,31,105,1,8,7,1
-2,32,105,1,9,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
deleted file mode 100644
index 52804bbb..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,127,4,0,2,0
-3,4,127,4,0,8,1
-3,5,127,4,0,8,1
-3,6,127,1,0,2,1
-3,7,113,4,9,8,1
-3,8,121,1,1,2,1
-3,9,113,0,9,8,1
-3,10,113,1,1,8,1
-3,11,113,1,9,7,1
-3,12,105,1,9,7,1
-3,13,105,1,8,7,1
-3,14,105,1,9,7,1
-3,15,113,1,8,7,1
-3,16,105,8,9,7,1
-3,17,113,1,8,8,1
-3,18,113,1,9,8,1
-3,19,113,1,7,8,1
-3,20,113,1,9,7,1
-3,21,113,1,8,7,1
-3,22,113,1,8,7,1
-3,23,113,1,8,7,1
-3,24,113,1,8,7,1
-3,25,113,1,5,7,1
-3,26,113,1,5,8,1
-3,27,113,1,0,8,1
-3,28,113,1,0,8,1
-3,29,113,1,0,8,1
-3,30,113,1,0,8,1
-3,31,113,1,0,7,1
-3,32,108,1,8,7,1
-4,3,127,4,0,8,0
-4,4,127,4,0,8,1
-4,5,127,1,0,2,1
-4,6,127,4,7,8,1
-4,7,127,1,1,8,1
-4,8,127,1,0,8,1
-4,9,113,1,0,8,1
-4,10,113,1,9,8,1
-4,11,113,1,9,7,1
-4,12,105,1,9,7,1
-4,13,105,1,9,7,1
-4,14,105,1,9,7,1
-4,15,105,1,9,7,1
-4,16,113,1,9,7,1
-4,17,113,1,9,7,1
-4,18,113,1,9,7,1
-4,19,113,1,9,8,1
-4,20,113,1,9,8,1
-4,21,113,1,9,8,1
-4,22,113,1,9,8,1
-4,23,113,1,9,7,1
-4,24,113,1,8,7,1
-4,25,105,1,8,7,1
-4,26,113,1,8,7,1
-4,27,113,1,8,7,1
-4,28,113,1,8,7,1
-4,29,113,1,9,7,1
-4,30,113,1,7,8,1
-4,31,113,1,7,8,1
-4,32,113,1,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
deleted file mode 100644
index 75c58ebf..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,121,1,5,4,1
-5,4,127,1,0,8,1
-5,5,127,1,0,2,1
-5,6,127,1,7,2,1
-5,7,127,1,1,0,1
-5,8,105,1,9,2,1
-5,9,113,1,7,0,1
-5,10,113,1,9,7,1
-5,11,105,8,9,7,1
-5,12,105,1,8,2,1
-5,13,113,0,7,8,1
-5,14,113,1,9,8,1
-5,15,113,1,9,8,1
-5,16,113,1,9,8,1
-5,17,113,1,0,8,1
-5,18,113,1,0,8,1
-5,19,113,1,0,8,1
-5,20,113,1,0,7,1
-5,21,108,1,8,7,1
-5,22,105,1,8,7,1
-5,23,105,1,8,7,1
-5,24,105,1,9,7,1
-5,25,113,1,7,7,1
-5,26,113,6,7,7,1
-5,27,113,6,7,2,1
-5,28,113,1,7,8,1
-5,29,113,1,7,8,1
-5,30,113,1,9,8,1
-5,31,113,1,0,8,1
-5,32,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
deleted file mode 100644
index d4edc9b2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,121,4,0,8,0
-6,5,113,1,0,8,0
-6,6,113,1,0,8,1
-6,7,113,1,0,8,1
-6,8,113,1,0,8,1
-6,9,113,1,0,8,1
-6,10,113,1,0,8,1
-6,11,113,1,9,7,1
-6,12,105,8,9,7,1
-6,13,105,1,8,7,1
-6,14,105,1,9,7,1
-6,15,105,1,9,7,1
-6,16,113,1,7,7,1
-6,17,113,6,9,7,1
-6,18,113,1,3,2,1
-6,19,113,6,7,8,1
-6,20,113,1,9,8,1
-6,21,113,1,0,8,1
-6,22,113,1,0,8,1
-6,23,113,1,0,8,1
-6,24,113,1,0,8,1
-6,25,113,1,0,8,1
-6,26,113,1,0,7,1
-6,27,108,1,8,7,1
-6,28,105,1,8,7,1
-6,29,105,1,9,7,1
-6,30,105,1,7,7,1
-6,31,113,6,7,7,1
-6,32,113,6,7,7,1
-6,33,113,6,7,2,1
-7,3,105,4,9,8,0
-7,4,121,0,8,8,0
-7,5,113,0,8,8,0
-7,6,121,4,8,8,0
-7,7,127,0,0,8,0
-7,8,121,4,8,8,0
-7,9,127,4,0,8,0
-7,10,127,1,0,8,0
-7,11,127,4,0,8,1
-7,12,127,1,0,8,0
-7,13,127,1,0,8,1
-7,14,127,1,7,8,1
-7,15,113,1,7,8,1
-7,16,113,1,9,7,1
-7,17,103,1,7,7,1
-7,18,113,8,7,7,1
-7,19,113,1,9,7,1
-7,20,113,1,7,7,1
-7,21,113,6,5,7,1
-7,22,113,1,5,8,1
-7,23,113,1,0,8,1
-7,24,113,1,0,8,1
-7,25,113,1,0,8,1
-7,26,113,1,0,8,1
-7,27,113,1,0,7,1
-7,28,108,1,8,7,1
-7,29,108,1,8,7,1
-7,30,108,1,8,7,1
-7,31,122,6,7,7,1
-7,32,113,6,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
deleted file mode 100644
index 42ff3af8..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,113,1,0,8,0
-8,5,113,1,0,8,1
-8,6,113,1,0,2,1
-8,7,113,1,8,2,1
-8,8,113,1,0,8,1
-8,9,113,1,0,8,1
-8,10,113,1,0,2,1
-8,11,113,1,0,2,1
-8,12,113,1,8,7,1
-8,13,105,8,9,7,1
-8,14,105,1,8,2,1
-8,15,113,1,9,7,1
-8,16,105,1,9,7,1
-8,17,113,1,7,7,1
-8,18,113,1,9,7,1
-8,19,113,1,7,8,1
-8,20,113,1,9,8,1
-8,21,113,1,0,8,1
-8,22,113,1,9,8,1
-8,23,113,1,0,7,1
-8,24,113,1,8,7,1
-8,25,105,1,8,7,1
-8,26,113,1,8,7,1
-8,27,105,1,9,7,1
-8,28,113,1,8,7,1
-8,29,113,6,9,7,1
-8,30,113,1,8,2,1
-8,31,113,1,7,8,1
-8,32,113,1,7,8,1
-8,33,113,1,7,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
deleted file mode 100644
index 0c5df9b9..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-predictions/sequifier-test-hp-search-custom-eval-run-1-best-1-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,121,1,8,8,1
-9,4,113,1,0,8,1
-9,5,113,1,8,8,1
-9,6,113,1,0,7,1
-9,7,105,1,8,7,1
-9,8,105,1,8,7,1
-9,9,105,1,7,7,1
-9,10,113,1,7,7,1
-9,11,113,1,7,7,1
-9,12,113,6,7,7,1
-9,13,113,6,5,2,1
-9,14,113,1,0,8,1
-9,15,113,1,0,8,1
-9,16,113,1,0,8,1
-9,17,113,1,0,8,1
-9,18,113,1,0,8,1
-9,19,113,1,0,8,1
-9,20,113,1,0,8,1
-9,21,113,1,0,7,1
-9,22,105,1,8,7,1
-9,23,105,1,8,7,1
-9,24,105,1,9,7,1
-9,25,105,1,7,7,1
-9,26,113,6,7,7,1
-9,27,113,6,7,2,1
-9,28,113,1,7,8,1
-9,29,113,1,7,8,1
-9,30,113,1,9,8,1
-9,31,113,1,0,8,1
-9,32,113,1,0,8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
deleted file mode 100644
index cffacbd7..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,115,9,2,2,[unknown]
-2,4,118,[mask],2,[mask],[mask]
-2,5,[unknown],9,2,7,[mask]
-2,6,115,9,9,7,[mask]
-2,7,115,9,9,7,[mask]
-2,8,115,9,9,7,[mask]
-2,9,119,9,9,7,[other]
-2,10,104,[mask],9,7,[other]
-2,11,112,1,9,7,[other]
-2,12,104,[mask],9,7,[other]
-2,13,112,1,9,7,[other]
-2,14,104,9,9,7,[other]
-2,15,112,6,9,7,[other]
-2,16,104,9,9,7,[other]
-2,17,104,1,9,7,[other]
-2,18,104,1,9,7,[other]
-2,19,104,6,9,7,[other]
-2,20,104,6,9,7,[other]
-2,21,104,6,9,7,[other]
-2,22,104,6,9,7,[other]
-2,23,104,6,9,7,[other]
-2,24,104,6,9,7,[other]
-2,25,104,6,9,7,[other]
-2,26,104,6,9,7,[other]
-2,27,104,6,9,7,[other]
-2,28,104,6,9,7,[other]
-2,29,104,6,9,7,[other]
-2,30,104,6,9,7,[other]
-2,31,104,6,9,7,[other]
-2,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
deleted file mode 100644
index 9bf3ad90..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,112,9,2,[mask],[mask]
-3,4,115,9,2,9,[other]
-3,5,112,9,2,2,[mask]
-3,6,112,9,2,2,[mask]
-3,7,112,9,2,2,[mask]
-3,8,112,9,9,2,[other]
-3,9,112,9,2,2,[mask]
-3,10,112,9,9,7,[other]
-3,11,112,9,9,7,[other]
-3,12,112,9,9,7,[other]
-3,13,112,9,9,7,[other]
-3,14,112,9,9,7,[other]
-3,15,112,9,9,7,[other]
-3,16,112,9,9,7,[other]
-3,17,112,9,9,7,[other]
-3,18,112,9,9,8,[other]
-3,19,112,9,9,7,[other]
-3,20,112,[mask],9,7,[other]
-3,21,112,[mask],9,8,[other]
-3,22,112,[mask],9,7,[other]
-3,23,112,[mask],9,7,[other]
-3,24,112,[mask],9,7,[other]
-3,25,129,[mask],9,7,[other]
-3,26,129,1,9,7,[other]
-3,27,129,1,9,7,[other]
-3,28,129,1,9,7,[other]
-3,29,129,1,9,[unknown],[other]
-3,30,112,6,9,7,[other]
-3,31,104,9,9,7,[other]
-3,32,112,9,9,7,[other]
-4,3,107,9,2,7,[other]
-4,4,107,9,9,7,[other]
-4,5,107,9,9,7,[other]
-4,6,107,9,9,7,[other]
-4,7,107,9,9,7,[other]
-4,8,107,9,9,7,[other]
-4,9,104,6,9,7,[other]
-4,10,104,6,9,7,[other]
-4,11,104,6,9,7,[other]
-4,12,104,6,9,7,[other]
-4,13,104,6,9,7,[other]
-4,14,104,6,9,7,[other]
-4,15,104,6,9,7,[other]
-4,16,104,6,9,7,[other]
-4,17,104,6,9,7,[other]
-4,18,104,6,9,7,[other]
-4,19,104,6,9,7,[other]
-4,20,104,6,9,7,[other]
-4,21,104,6,9,7,[other]
-4,22,104,6,9,7,[other]
-4,23,104,6,9,7,[other]
-4,24,104,6,9,7,[other]
-4,25,104,6,9,7,[other]
-4,26,104,6,9,7,[other]
-4,27,104,6,9,7,[other]
-4,28,104,6,9,7,[other]
-4,29,104,6,9,7,[other]
-4,30,104,6,9,7,[other]
-4,31,104,6,9,7,[other]
-4,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
deleted file mode 100644
index 12d83f71..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,113,3,2,[other],[other]
-9,4,119,9,2,7,[mask]
-9,5,129,9,9,[other],[other]
-9,6,119,9,9,[other],[mask]
-9,7,119,9,9,9,[mask]
-9,8,129,9,9,8,[other]
-9,9,111,9,9,7,[other]
-9,10,112,9,9,7,[other]
-9,11,112,[mask],9,7,[unknown]
-9,12,112,[mask],9,7,[other]
-9,13,112,[mask],9,8,[unknown]
-9,14,104,[mask],9,7,[other]
-9,15,129,1,9,7,[other]
-9,16,129,1,9,7,[other]
-9,17,129,1,9,7,[other]
-9,18,104,1,9,7,[other]
-9,19,104,1,9,7,[other]
-9,20,104,1,9,[unknown],[other]
-9,21,104,6,9,[unknown],[other]
-9,22,104,6,9,[unknown],[other]
-9,23,104,6,9,[unknown],[other]
-9,24,104,6,9,7,[other]
-9,25,104,6,9,7,[other]
-9,26,104,6,9,7,[other]
-9,27,104,6,9,7,[other]
-9,28,104,6,9,7,[other]
-9,29,104,6,9,7,[other]
-9,30,104,6,9,7,[other]
-9,31,104,6,9,7,[other]
-9,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
deleted file mode 100644
index 54189e82..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,121,1,1,9,1
-0,5,105,0,[other],8,2
-0,6,113,1,[other],[mask],0
-0,7,105,8,[other],8,1
-0,8,113,1,1,8,0
-0,9,105,3,[other],1,1
-0,10,113,1,8,8,1
-0,11,113,1,[other],1,1
-0,12,113,1,9,1,1
-0,13,113,1,[other],1,1
-0,14,113,0,9,1,1
-0,15,113,1,8,1,1
-0,16,103,0,9,1,1
-0,17,113,0,8,1,1
-0,18,113,0,[other],8,1
-0,19,113,0,[other],0,[mask]
-0,20,108,0,[other],1,[other]
-0,21,108,0,[other],7,1
-0,22,108,5,[other],[mask],1
-0,23,124,[mask],[other],7,1
-0,24,124,1,[other],2,1
-0,25,113,1,[other],8,1
-0,26,113,1,[other],[mask],1
-0,27,122,8,[other],8,1
-0,28,113,1,[other],8,1
-0,29,113,1,[other],0,1
-0,30,113,8,[other],2,1
-0,31,113,1,[other],2,1
-0,32,113,8,9,1,1
-0,33,113,1,[other],[mask],1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
deleted file mode 100644
index 6342f656..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,107,5,[other],[mask],[other]
-1,5,121,3,[other],7,[other]
-1,6,121,5,[other],8,[other]
-1,7,121,5,[other],7,[other]
-1,8,121,5,[other],7,[other]
-1,9,121,1,[other],8,1
-1,10,122,1,7,8,0
-1,11,105,1,7,8,[other]
-1,12,122,1,7,[unknown],0
-1,13,105,0,7,1,1
-1,14,113,[other],7,5,0
-1,15,113,[mask],1,5,1
-1,16,113,1,1,2,1
-1,17,120,[mask],4,1,1
-1,18,113,[mask],1,7,1
-1,19,102,1,[other],2,1
-1,20,113,1,7,8,[unknown]
-1,21,113,1,1,0,1
-1,22,128,1,0,1,1
-1,23,113,1,[other],7,1
-1,24,105,8,[other],2,1
-1,25,113,1,1,2,1
-1,26,113,8,[other],1,1
-1,27,105,1,1,2,1
-1,28,113,0,[other],1,1
-1,29,113,1,[other],2,1
-1,30,113,[mask],9,1,1
-1,31,113,1,[other],1,1
-1,32,113,1,9,1,1
-1,33,113,1,9,1,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
deleted file mode 100644
index dba66fc2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,128,0,7,[unknown],2
-2,4,113,8,[other],6,2
-2,5,109,1,[other],6,1
-2,6,128,1,[other],[mask],2
-2,7,104,[mask],[other],6,1
-2,8,104,1,1,2,1
-2,9,113,1,8,0,2
-2,10,113,1,[other],[mask],1
-2,11,121,1,[other],0,1
-2,12,101,1,[other],7,1
-2,13,105,8,[other],0,1
-2,14,122,1,1,2,1
-2,15,113,[mask],[other],0,1
-2,16,113,1,[other],2,1
-2,17,113,[mask],[other],2,1
-2,18,113,1,[other],2,1
-2,19,113,[mask],[other],1,1
-2,20,113,1,[other],2,1
-2,21,113,1,[other],1,1
-2,22,113,1,9,1,1
-2,23,113,1,[other],1,1
-2,24,103,1,9,1,1
-2,25,103,1,8,1,1
-2,26,103,0,9,7,1
-2,27,105,5,8,0,1
-2,28,105,0,8,1,1
-2,29,113,0,8,9,2
-2,30,105,3,[other],1,[other]
-2,31,113,0,[other],8,[other]
-2,32,113,0,[other],8,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
deleted file mode 100644
index 20206ce0..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,102,5,1,1,2
-3,4,128,0,9,5,[unknown]
-3,5,105,0,1,5,2
-3,6,105,[other],9,[mask],[unknown]
-3,7,128,0,1,[mask],1
-3,8,105,0,[other],8,[unknown]
-3,9,121,1,8,[mask],[other]
-3,10,104,0,[other],8,1
-3,11,113,1,[other],[mask],0
-3,12,104,3,[other],1,1
-3,13,113,1,[other],8,2
-3,14,113,1,[other],1,1
-3,15,113,1,[other],1,1
-3,16,113,1,[other],1,1
-3,17,105,0,[other],1,1
-3,18,120,0,8,0,1
-3,19,105,0,8,2,1
-3,20,113,0,8,0,1
-3,21,113,1,[other],2,1
-3,22,113,0,[other],1,1
-3,23,113,0,[other],8,1
-3,24,113,1,[other],8,1
-3,25,113,1,[other],8,1
-3,26,113,1,[other],8,1
-3,27,113,1,[other],7,1
-3,28,105,8,[other],1,1
-3,29,113,1,1,0,1
-3,30,105,1,[other],1,1
-3,31,113,1,9,0,1
-3,32,105,1,[other],1,1
-4,3,102,0,[other],[unknown],[unknown]
-4,4,102,8,[other],8,[unknown]
-4,5,102,0,[other],6,[unknown]
-4,6,105,8,[other],6,[unknown]
-4,7,115,8,[other],2,1
-4,8,127,[mask],7,8,[unknown]
-4,9,113,2,1,0,1
-4,10,122,1,[other],2,1
-4,11,113,[mask],7,2,1
-4,12,113,1,1,2,1
-4,13,113,[mask],[other],1,1
-4,14,113,1,1,2,1
-4,15,113,1,[other],1,1
-4,16,113,1,1,1,1
-4,17,105,1,[other],1,1
-4,18,105,1,9,0,1
-4,19,105,1,8,0,1
-4,20,105,1,9,0,1
-4,21,105,1,8,0,1
-4,22,105,0,9,1,[other]
-4,23,113,0,8,0,[other]
-4,24,113,0,[other],1,[other]
-4,25,113,0,[other],8,2
-4,26,113,1,[other],8,1
-4,27,113,1,[other],8,1
-4,28,113,1,[other],8,1
-4,29,113,1,[other],8,1
-4,30,113,1,[other],[mask],1
-4,31,121,1,[other],1,1
-4,32,126,1,0,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
deleted file mode 100644
index 051cc930..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,128,3,[other],4,1
-5,4,128,1,[other],6,1
-5,5,105,1,[other],7,1
-5,6,105,1,[other],0,1
-5,7,105,1,[other],0,1
-5,8,105,1,[other],0,1
-5,9,113,1,7,2,1
-5,10,113,1,[other],1,1
-5,11,113,1,[other],1,1
-5,12,113,1,[other],1,1
-5,13,113,1,[other],1,1
-5,14,113,0,[other],1,1
-5,15,113,0,[other],2,1
-5,16,113,0,[other],2,1
-5,17,113,0,1,1,1
-5,18,108,0,9,1,1
-5,19,113,0,8,0,1
-5,20,113,0,[other],[mask],1
-5,21,108,3,[other],2,1
-5,22,113,1,[other],8,1
-5,23,113,1,[other],8,1
-5,24,113,1,[other],8,1
-5,25,113,1,[other],0,1
-5,26,113,1,[other],7,1
-5,27,105,8,[other],0,1
-5,28,113,1,[other],2,1
-5,29,113,8,[other],1,1
-5,30,113,1,[other],2,1
-5,31,113,[mask],[other],1,1
-5,32,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
deleted file mode 100644
index d92da9a7..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,122,2,[other],8,[unknown]
-6,5,113,[mask],[other],8,[other]
-6,6,113,1,[other],[unknown],1
-6,7,113,[mask],[other],2,1
-6,8,113,1,[other],2,1
-6,9,113,3,[other],1,1
-6,10,121,1,[other],[mask],1
-6,11,108,0,7,1,1
-6,12,113,1,1,[mask],1
-6,13,108,8,9,1,1
-6,14,113,1,1,0,1
-6,15,105,1,9,1,[other]
-6,16,113,0,7,0,[other]
-6,17,113,0,[other],0,[other]
-6,18,113,8,[other],2,[other]
-6,19,113,[mask],[other],8,1
-6,20,113,1,[other],[mask],1
-6,21,113,1,[other],1,1
-6,22,113,1,[other],[mask],1
-6,23,113,1,[other],[mask],1
-6,24,121,1,[other],2,1
-6,25,113,1,[other],8,1
-6,26,105,1,1,[mask],1
-6,27,105,[other],9,1,[other]
-6,28,113,0,7,[unknown],1
-6,29,105,0,[other],1,1
-6,30,113,1,[other],0,1
-6,31,105,1,9,1,1
-6,32,113,0,7,1,1
-6,33,113,1,[other],1,1
-7,3,105,0,[other],8,[mask]
-7,4,105,0,[other],1,[other]
-7,5,105,0,9,6,[unknown]
-7,6,105,0,[other],1,1
-7,7,113,0,[other],1,2
-7,8,113,0,[other],1,[other]
-7,9,113,0,[other],[mask],1
-7,10,113,1,[other],[mask],1
-7,11,113,[mask],[other],8,1
-7,12,113,1,[other],[mask],1
-7,13,113,3,[other],1,1
-7,14,113,1,[other],[mask],1
-7,15,108,1,[other],0,1
-7,16,113,1,[other],7,1
-7,17,105,8,[other],2,1
-7,18,113,1,[other],2,1
-7,19,113,8,[other],1,1
-7,20,113,1,[other],2,1
-7,21,113,3,[other],1,1
-7,22,113,1,[other],2,1
-7,23,113,8,[other],1,1
-7,24,113,1,[other],2,1
-7,25,113,1,9,1,1
-7,26,103,1,9,1,1
-7,27,103,1,8,1,1
-7,28,103,0,9,7,1
-7,29,105,5,8,0,1
-7,30,105,1,8,1,1
-7,31,113,0,9,1,2
-7,32,113,0,[other],1,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
deleted file mode 100644
index acc3eb04..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,101,1,[other],8,2
-8,5,104,1,[other],[unknown],2
-8,6,104,3,[other],[other],1
-8,7,120,5,1,0,1
-8,8,128,1,[other],0,1
-8,9,105,1,[other],7,[other]
-8,10,120,8,7,0,[other]
-8,11,120,5,[other],0,1
-8,12,120,1,7,2,1
-8,13,113,[mask],7,1,1
-8,14,113,1,[other],2,1
-8,15,113,1,[other],2,1
-8,16,113,1,[other],8,1
-8,17,113,1,[other],2,1
-8,18,113,[mask],1,8,1
-8,19,102,1,1,2,1
-8,20,128,1,4,7,1
-8,21,105,8,[other],7,1
-8,22,105,5,5,2,1
-8,23,113,1,8,0,1
-8,24,105,1,[other],7,1
-8,25,105,8,9,2,1
-8,26,113,1,[other],0,1
-8,27,113,8,[other],2,1
-8,28,113,1,[other],8,1
-8,29,113,1,[other],7,1
-8,30,105,8,[other],1,1
-8,31,113,1,[other],2,1
-8,32,113,[mask],[other],1,1
-8,33,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
deleted file mode 100644
index ff465a4a..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-predictions/sequifier-test-hp-search-custom-eval-run-3-best-1-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,108,5,[other],8,2
-9,4,122,5,[other],7,1
-9,5,105,5,[other],7,[other]
-9,6,122,5,7,7,[other]
-9,7,122,5,[other],7,1
-9,8,122,5,[other],2,1
-9,9,113,[mask],7,2,1
-9,10,113,1,[other],2,1
-9,11,113,[mask],[other],2,1
-9,12,113,1,[other],2,1
-9,13,113,[mask],[other],2,1
-9,14,113,1,1,2,1
-9,15,113,[mask],[other],1,1
-9,16,113,1,1,2,1
-9,17,129,1,0,7,1
-9,18,108,1,8,7,1
-9,19,108,1,0,7,1
-9,20,108,1,[other],7,1
-9,21,124,8,[other],7,1
-9,22,105,8,[other],2,1
-9,23,113,1,[other],2,1
-9,24,113,8,[other],8,2
-9,25,113,[mask],[other],8,1
-9,26,113,1,[other],2,1
-9,27,113,8,[other],2,1
-9,28,113,1,[other],2,1
-9,29,113,1,[other],7,1
-9,30,108,8,9,2,1
-9,31,113,1,1,7,1
-9,32,108,8,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv
new file mode 100644
index 00000000..453ac47f
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+0,4,102,3,3,5,0
+0,5,102,3,9,0,1
+0,6,102,1,9,0,0
+0,7,102,1,0,0,2
+0,8,111,1,2,0,2
+0,9,102,1,9,0,2
+0,10,102,1,3,0,2
+0,11,102,1,3,0,2
+0,12,111,1,3,0,2
+0,13,111,1,9,0,2
+0,14,105,1,0,0,2
+0,15,105,1,0,0,2
+0,16,105,1,0,0,2
+0,17,105,1,0,0,2
+0,18,105,1,0,0,2
+0,19,105,1,0,0,2
+0,20,105,1,0,8,2
+0,21,105,1,0,5,2
+0,22,105,5,0,2,2
+0,23,105,5,0,8,2
+0,24,105,5,0,8,2
+0,25,105,5,0,8,2
+0,26,105,5,0,8,2
+0,27,105,3,0,8,2
+0,28,105,3,0,5,1
+0,29,105,3,0,2,1
+0,30,105,3,0,2,1
+0,31,105,3,0,2,1
+0,32,110,3,0,2,1
+0,33,110,3,0,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv
new file mode 100644
index 00000000..6eac25ae
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+1,4,102,7,0,5,[other]
+1,5,111,3,0,1,0
+1,6,111,1,0,1,1
+1,7,111,7,0,1,0
+1,8,111,5,0,1,1
+1,9,[unknown],7,0,1,0
+1,10,102,1,0,5,0
+1,11,110,5,0,1,2
+1,12,102,5,0,5,0
+1,13,102,5,0,2,1
+1,14,[unknown],5,0,1,1
+1,15,110,2,0,8,0
+1,16,102,3,0,5,0
+1,17,111,3,0,2,2
+1,18,111,3,3,5,2
+1,19,111,3,3,1,2
+1,20,111,1,3,0,2
+1,21,102,1,2,0,2
+1,22,111,[mask],2,0,2
+1,23,112,[mask],2,5,2
+1,24,109,[mask],0,5,2
+1,25,109,[mask],0,2,2
+1,26,109,[mask],0,[other],2
+1,27,109,3,8,[other],2
+1,28,109,1,8,5,2
+1,29,109,1,8,[other],2
+1,30,104,1,8,[other],2
+1,31,104,1,9,[other],2
+1,32,111,3,9,0,2
+1,33,127,[mask],9,[other],2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv
new file mode 100644
index 00000000..9b51b012
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+2,3,112,2,0,5,1
+2,4,112,7,2,0,[other]
+2,5,112,7,3,5,2
+2,6,104,7,3,0,2
+2,7,104,7,9,0,2
+2,8,104,7,3,0,0
+2,9,104,4,3,[other],0
+2,10,112,7,3,0,0
+2,11,104,7,3,0,0
+2,12,116,4,3,6,0
+2,13,112,4,3,6,0
+2,14,112,7,3,6,0
+2,15,116,7,3,6,0
+2,16,116,4,3,6,0
+2,17,112,4,3,6,0
+2,18,112,8,3,6,0
+2,19,112,8,3,6,0
+2,20,118,4,7,6,0
+2,21,112,4,9,6,1
+2,22,118,8,7,6,0
+2,23,112,4,9,6,0
+2,24,112,8,7,0,1
+2,25,112,4,4,6,2
+2,26,112,4,3,[other],1
+2,27,104,4,4,8,[other]
+2,28,112,4,4,[other],1
+2,29,104,7,4,8,1
+2,30,104,4,4,[other],1
+2,31,104,8,4,[other],1
+2,32,104,8,4,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv
new file mode 100644
index 00000000..4c107703
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+3,3,111,3,3,5,0
+3,4,111,2,3,5,0
+3,5,112,3,3,2,0
+3,6,102,3,3,5,1
+3,7,118,2,3,0,0
+3,8,102,6,3,2,0
+3,9,102,6,3,0,0
+3,10,102,6,3,0,0
+3,11,102,6,3,2,0
+3,12,102,6,3,2,0
+3,13,102,3,3,2,2
+3,14,102,6,3,0,2
+3,15,102,6,3,0,0
+3,16,102,1,3,2,2
+3,17,111,3,3,0,2
+3,18,102,1,3,0,2
+3,19,111,1,3,0,2
+3,20,111,1,3,0,2
+3,21,111,1,2,0,2
+3,22,111,1,2,0,2
+3,23,102,1,2,0,2
+3,24,111,1,2,0,2
+3,25,111,1,2,5,2
+3,26,105,5,9,5,2
+3,27,105,[mask],0,8,2
+3,28,105,[mask],0,8,2
+3,29,105,[mask],0,8,2
+3,30,105,[mask],0,[other],2
+3,31,105,[mask],9,[other],2
+3,32,105,[mask],9,[other],2
+4,3,113,7,9,5,2
+4,4,104,4,0,0,1
+4,5,112,6,2,0,1
+4,6,104,7,9,0,1
+4,7,118,7,9,0,1
+4,8,104,4,9,0,0
+4,9,102,7,0,0,0
+4,10,102,7,3,0,0
+4,11,111,6,3,0,0
+4,12,102,4,3,0,0
+4,13,102,6,3,0,0
+4,14,102,6,3,0,0
+4,15,102,6,3,0,0
+4,16,102,6,3,6,0
+4,17,102,1,3,2,0
+4,18,102,3,3,2,2
+4,19,102,6,3,0,0
+4,20,102,1,3,2,0
+4,21,102,3,3,2,2
+4,22,111,3,3,0,2
+4,23,102,1,3,0,2
+4,24,111,1,3,0,2
+4,25,102,1,3,0,2
+4,26,111,1,3,0,2
+4,27,102,1,2,0,2
+4,28,111,1,2,0,2
+4,29,111,1,2,0,2
+4,30,111,1,2,5,2
+4,31,105,1,9,5,2
+4,32,105,1,0,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv
new file mode 100644
index 00000000..d03238fa
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+5,3,104,7,9,8,2
+5,4,104,1,9,0,2
+5,5,104,1,9,0,2
+5,6,104,1,9,0,2
+5,7,104,1,9,0,2
+5,8,104,1,9,0,2
+5,9,104,1,0,0,2
+5,10,104,7,9,0,0
+5,11,122,7,0,7,0
+5,12,122,7,3,7,0
+5,13,119,4,3,7,0
+5,14,119,4,3,0,0
+5,15,112,3,9,6,0
+5,16,122,7,3,0,0
+5,17,102,4,9,6,0
+5,18,102,8,3,0,0
+5,19,102,8,9,6,0
+5,20,102,8,3,0,0
+5,21,102,3,3,6,0
+5,22,102,4,3,0,0
+5,23,102,3,3,6,0
+5,24,102,4,3,6,0
+5,25,102,3,3,6,0
+5,26,102,4,3,6,0
+5,27,102,3,3,6,0
+5,28,102,4,3,2,0
+5,29,102,3,3,6,2
+5,30,111,6,3,0,2
+5,31,102,3,3,6,2
+5,32,102,6,3,6,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv
new file mode 100644
index 00000000..244d29b4
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+6,4,110,1,0,5,1
+6,5,110,2,0,0,1
+6,6,112,1,0,[other],1
+6,7,111,7,[unknown],[other],2
+6,8,111,3,2,5,1
+6,9,112,2,9,5,1
+6,10,108,7,0,1,[other]
+6,11,104,7,0,[other],1
+6,12,119,7,9,0,1
+6,13,104,7,9,0,0
+6,14,122,7,0,0,1
+6,15,111,7,4,[other],0
+6,16,119,7,3,0,0
+6,17,122,7,9,0,0
+6,18,102,7,0,5,0
+6,19,111,7,0,0,0
+6,20,111,8,0,0,0
+6,21,112,8,0,2,0
+6,22,102,7,0,5,0
+6,23,111,3,0,2,0
+6,24,111,3,0,1,0
+6,25,111,1,0,1,0
+6,26,111,1,0,1,0
+6,27,110,1,0,1,0
+6,28,102,1,0,4,0
+6,29,111,1,0,[unknown],2
+6,30,110,1,0,1,1
+6,31,110,1,0,1,1
+6,32,110,1,0,1,1
+6,33,110,1,0,1,1
+7,3,102,3,0,5,0
+7,4,102,3,0,[unknown],0
+7,5,111,3,0,[unknown],0
+7,6,102,8,0,1,[other]
+7,7,102,3,0,2,0
+7,8,111,3,0,2,1
+7,9,112,4,0,1,[other]
+7,10,102,1,0,2,0
+7,11,111,1,8,2,2
+7,12,110,1,[mask],1,2
+7,13,[mask],1,0,1,2
+7,14,111,1,0,1,2
+7,15,111,1,2,1,2
+7,16,110,1,2,1,2
+7,17,[mask],1,0,5,1
+7,18,109,1,0,8,2
+7,19,102,1,0,5,1
+7,20,110,7,0,8,2
+7,21,127,3,0,5,1
+7,22,109,3,0,5,1
+7,23,109,7,0,5,1
+7,24,109,7,0,5,1
+7,25,104,7,0,5,1
+7,26,109,7,0,5,1
+7,27,119,7,0,0,[other]
+7,28,111,7,0,[other],0
+7,29,111,7,0,0,0
+7,30,111,7,9,0,0
+7,31,111,7,0,0,0
+7,32,111,7,0,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv
new file mode 100644
index 00000000..0735ca91
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+8,4,111,3,0,6,[other]
+8,5,102,7,9,7,[other]
+8,6,104,7,0,7,0
+8,7,102,8,0,5,1
+8,8,112,3,0,2,0
+8,9,111,7,0,5,2
+8,10,111,1,3,0,2
+8,11,102,6,9,0,2
+8,12,111,1,3,0,2
+8,13,111,1,3,0,2
+8,14,102,1,9,0,2
+8,15,111,1,2,0,2
+8,16,111,1,2,5,2
+8,17,111,5,0,5,2
+8,18,105,5,0,[other],2
+8,19,105,[mask],0,2,2
+8,20,105,[mask],0,8,2
+8,21,105,[mask],0,8,2
+8,22,105,[mask],0,[other],2
+8,23,105,3,0,[other],2
+8,24,109,5,9,0,2
+8,25,105,[mask],0,6,2
+8,26,105,[mask],3,[other],2
+8,27,109,[mask],0,[other],2
+8,28,109,3,0,[other],2
+8,29,109,3,0,2,2
+8,30,109,3,0,6,1
+8,31,109,3,0,2,1
+8,32,109,3,0,[other],[other]
+8,33,109,3,8,2,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv
new file mode 100644
index 00000000..20d90721
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,108,4,9,1,1
+9,4,113,7,9,8,1
+9,5,104,4,9,0,1
+9,6,119,8,9,0,1
+9,7,119,7,9,0,2
+9,8,104,7,9,0,0
+9,9,122,7,9,0,0
+9,10,102,7,3,0,0
+9,11,122,7,3,0,0
+9,12,102,4,3,6,0
+9,13,102,7,3,0,0
+9,14,111,6,3,0,0
+9,15,112,3,3,6,0
+9,16,102,7,3,5,0
+9,17,102,8,3,6,0
+9,18,102,3,3,2,0
+9,19,102,4,3,6,0
+9,20,102,3,3,6,0
+9,21,102,6,3,2,0
+9,22,102,3,3,2,1
+9,23,118,3,3,6,2
+9,24,102,6,3,0,0
+9,25,102,3,3,2,2
+9,26,111,3,3,0,2
+9,27,102,3,3,6,2
+9,28,102,6,3,0,2
+9,29,111,1,3,2,2
+9,30,102,1,3,0,2
+9,31,102,1,2,0,2
+9,32,111,1,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv
new file mode 100644
index 00000000..0b011a28
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+0,4,104,7,[mask],[unknown],1
+0,5,104,7,[mask],[unknown],0
+0,6,104,7,1,3,0
+0,7,104,0,1,3,0
+0,8,112,7,1,[unknown],0
+0,9,108,5,1,[unknown],0
+0,10,112,1,1,5,2
+0,11,129,2,1,5,2
+0,12,129,2,1,5,2
+0,13,112,2,1,5,2
+0,14,107,7,[mask],[unknown],1
+0,15,107,7,[mask],[unknown],1
+0,16,107,7,[mask],[unknown],1
+0,17,107,7,[mask],[unknown],1
+0,18,107,7,[mask],[unknown],1
+0,19,104,7,[mask],[unknown],1
+0,20,104,7,[mask],[unknown],1
+0,21,104,9,[mask],[unknown],1
+0,22,104,9,[mask],3,2
+0,23,104,9,[mask],3,2
+0,24,104,9,[mask],3,2
+0,25,104,9,[mask],3,2
+0,26,104,9,[mask],1,2
+0,27,104,9,[other],1,[mask]
+0,28,104,[unknown],[other],1,[mask]
+0,29,112,[unknown],[other],1,[mask]
+0,30,112,2,[mask],[unknown],[mask]
+0,31,110,2,[mask],[unknown],[mask]
+0,32,110,2,[other],7,[mask]
+0,33,104,2,[other],7,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv
new file mode 100644
index 00000000..531427bf
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+1,4,104,7,[mask],1,1
+1,5,104,7,9,1,1
+1,6,104,7,9,[unknown],1
+1,7,104,7,9,[unknown],1
+1,8,104,7,9,[unknown],1
+1,9,118,7,9,[unknown],[other]
+1,10,112,7,9,[unknown],0
+1,11,118,7,1,[unknown],[other]
+1,12,118,7,1,[unknown],[other]
+1,13,110,7,1,[unknown],[other]
+1,14,118,7,1,[unknown],[other]
+1,15,112,7,1,[unknown],2
+1,16,108,7,[mask],[unknown],2
+1,17,129,1,[mask],8,2
+1,18,129,2,[mask],5,2
+1,19,129,2,1,1,2
+1,20,129,7,1,5,[mask]
+1,21,104,9,1,5,[mask]
+1,22,129,9,1,5,[other]
+1,23,112,9,1,5,[mask]
+1,24,112,9,1,5,[mask]
+1,25,112,0,1,5,[mask]
+1,26,112,6,[mask],[unknown],[mask]
+1,27,123,7,[mask],[unknown],[mask]
+1,28,109,7,[mask],5,1
+1,29,104,7,[mask],1,[mask]
+1,30,104,7,[mask],8,1
+1,31,104,7,[mask],8,1
+1,32,104,0,[mask],8,1
+1,33,104,7,[mask],8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv
new file mode 100644
index 00000000..4350e88b
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+2,3,112,2,[mask],5,[mask]
+2,4,104,7,[mask],[unknown],[mask]
+2,5,104,7,1,8,1
+2,6,104,7,[mask],[unknown],0
+2,7,129,7,1,8,1
+2,8,104,7,[mask],1,0
+2,9,112,7,1,1,1
+2,10,107,7,[mask],[unknown],2
+2,11,129,9,1,1,2
+2,12,112,9,[other],1,2
+2,13,104,7,1,1,2
+2,14,118,7,1,1,[other]
+2,15,112,7,1,5,2
+2,16,129,7,1,1,[other]
+2,17,112,7,1,5,[other]
+2,18,112,0,1,5,[other]
+2,19,112,7,1,5,[other]
+2,20,112,7,1,[unknown],[other]
+2,21,112,7,1,[unknown],0
+2,22,108,7,1,[unknown],2
+2,23,119,7,[mask],[unknown],0
+2,24,129,0,1,5,2
+2,25,112,7,[mask],8,2
+2,26,104,7,[mask],5,2
+2,27,129,7,1,8,2
+2,28,104,9,1,5,2
+2,29,112,9,1,8,[mask]
+2,30,112,9,1,5,[mask]
+2,31,104,9,1,5,[mask]
+2,32,112,0,[mask],8,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv
new file mode 100644
index 00000000..b312c008
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+3,3,104,0,[mask],8,[mask]
+3,4,104,0,[mask],8,[other]
+3,5,104,[mask],[mask],6,0
+3,6,104,0,[mask],8,[unknown]
+3,7,104,0,[mask],8,0
+3,8,104,0,[mask],6,[mask]
+3,9,[other],1,[mask],8,[unknown]
+3,10,129,2,1,6,[unknown]
+3,11,129,2,1,6,[unknown]
+3,12,129,2,1,6,[unknown]
+3,13,129,2,1,5,2
+3,14,129,2,1,5,2
+3,15,129,2,1,5,2
+3,16,129,2,1,5,2
+3,17,112,2,1,5,2
+3,18,107,7,[mask],5,1
+3,19,107,7,[mask],[unknown],1
+3,20,107,7,[mask],[unknown],1
+3,21,107,7,[mask],[unknown],1
+3,22,107,7,[mask],[unknown],1
+3,23,104,7,[mask],[unknown],1
+3,24,104,7,[mask],[unknown],1
+3,25,104,[mask],[mask],[unknown],1
+3,26,104,9,[mask],3,2
+3,27,104,9,[mask],3,2
+3,28,104,9,[mask],3,2
+3,29,104,9,[mask],3,2
+3,30,104,9,[mask],3,2
+3,31,112,9,[other],1,[mask]
+3,32,104,9,[mask],1,[mask]
+4,3,104,9,1,[unknown],2
+4,4,104,7,[mask],[unknown],0
+4,5,[other],9,1,[unknown],1
+4,6,104,9,1,[unknown],0
+4,7,112,9,1,[unknown],1
+4,8,104,7,1,[unknown],2
+4,9,112,9,1,1,[mask]
+4,10,129,7,1,1,[mask]
+4,11,112,7,1,1,[other]
+4,12,112,7,1,5,2
+4,13,108,7,1,[unknown],2
+4,14,129,7,1,5,2
+4,15,112,7,[mask],1,[other]
+4,16,112,7,1,5,[mask]
+4,17,118,7,[mask],5,1
+4,18,104,7,[mask],[unknown],[mask]
+4,19,104,7,[mask],8,1
+4,20,104,0,[mask],8,[mask]
+4,21,104,0,[mask],5,[other]
+4,22,112,7,[mask],3,[mask]
+4,23,104,2,[mask],8,[mask]
+4,24,112,9,[mask],3,1
+4,25,104,2,[mask],3,[mask]
+4,26,107,9,[other],6,[unknown]
+4,27,104,2,[mask],3,2
+4,28,107,9,2,6,[unknown]
+4,29,104,2,[mask],3,1
+4,30,104,9,[mask],3,1
+4,31,107,9,[mask],3,1
+4,32,104,3,[mask],3,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv
new file mode 100644
index 00000000..bfc352bc
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+5,3,102,7,[mask],9,1
+5,4,113,7,[mask],[unknown],1
+5,5,104,7,[mask],[unknown],1
+5,6,104,7,[mask],[unknown],1
+5,7,104,7,[mask],[unknown],1
+5,8,104,7,[mask],[unknown],1
+5,9,104,7,[mask],3,1
+5,10,104,9,[mask],3,1
+5,11,104,9,[mask],3,2
+5,12,104,9,[mask],3,2
+5,13,112,9,[mask],1,2
+5,14,104,5,9,1,2
+5,15,104,9,[other],1,0
+5,16,111,[unknown],[other],3,[mask]
+5,17,112,[unknown],[mask],3,2
+5,18,104,2,[mask],1,[mask]
+5,19,110,9,[other],1,[mask]
+5,20,110,2,[other],7,[mask]
+5,21,104,2,[mask],8,[mask]
+5,22,[other],9,[other],8,[mask]
+5,23,104,2,[other],8,[mask]
+5,24,129,9,[other],8,2
+5,25,104,9,[other],1,[mask]
+5,26,107,9,[other],1,[mask]
+5,27,104,2,[other],1,[mask]
+5,28,[other],9,[other],1,[other]
+5,29,107,1,[other],1,[mask]
+5,30,118,1,[other],[unknown],[mask]
+5,31,[other],2,1,[unknown],[mask]
+5,32,[other],9,[other],5,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv
new file mode 100644
index 00000000..13756617
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+6,4,129,9,1,5,[other]
+6,5,129,9,1,5,[mask]
+6,6,129,9,1,5,[other]
+6,7,112,9,1,5,[other]
+6,8,129,7,1,5,[mask]
+6,9,129,2,1,8,[mask]
+6,10,129,7,1,5,2
+6,11,112,7,1,8,2
+6,12,104,7,[mask],[unknown],2
+6,13,104,7,[mask],[unknown],1
+6,14,104,7,[mask],[unknown],1
+6,15,104,7,[mask],[unknown],1
+6,16,104,9,[mask],1,1
+6,17,104,9,[mask],[unknown],[mask]
+6,18,112,9,1,3,[mask]
+6,19,112,9,1,1,[other]
+6,20,112,9,1,1,[mask]
+6,21,111,9,1,5,2
+6,22,112,9,1,1,[mask]
+6,23,118,7,1,5,2
+6,24,112,7,1,1,[other]
+6,25,112,0,1,1,[mask]
+6,26,118,7,[mask],[unknown],0
+6,27,129,7,1,8,[mask]
+6,28,129,0,[mask],5,[other]
+6,29,112,7,[mask],5,[mask]
+6,30,104,7,[mask],5,[mask]
+6,31,104,7,[mask],8,2
+6,32,129,9,1,8,2
+6,33,104,9,1,5,2
+7,3,129,2,1,5,[mask]
+7,4,129,2,1,5,1
+7,5,112,2,1,5,1
+7,6,107,2,1,5,1
+7,7,107,2,[mask],[unknown],1
+7,8,107,7,[mask],[unknown],1
+7,9,107,7,[mask],[unknown],1
+7,10,107,7,[mask],[unknown],1
+7,11,107,7,[mask],[unknown],1
+7,12,107,7,[mask],[unknown],1
+7,13,104,7,[mask],[unknown],1
+7,14,104,7,[mask],[unknown],1
+7,15,104,9,[mask],[unknown],2
+7,16,104,9,[mask],3,2
+7,17,104,9,[mask],3,2
+7,18,104,9,[mask],3,2
+7,19,104,9,[mask],3,2
+7,20,104,9,[other],1,[mask]
+7,21,112,9,[other],1,[mask]
+7,22,104,9,[other],1,[mask]
+7,23,112,9,[other],1,[mask]
+7,24,104,2,6,1,[mask]
+7,25,110,9,[other],1,[other]
+7,26,110,2,[mask],7,[mask]
+7,27,[other],2,[other],8,[mask]
+7,28,[other],1,[other],8,[other]
+7,29,121,1,[mask],7,[mask]
+7,30,[other],2,1,8,[mask]
+7,31,[other],9,1,5,[other]
+7,32,129,9,1,5,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv
new file mode 100644
index 00000000..ca8bff47
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+8,4,104,2,1,5,[mask]
+8,5,129,9,1,5,1
+8,6,104,2,1,5,[mask]
+8,7,129,7,1,5,1
+8,8,112,2,1,5,1
+8,9,107,7,[mask],[unknown],1
+8,10,107,7,[mask],[unknown],1
+8,11,107,7,[mask],[unknown],1
+8,12,107,7,[mask],[unknown],1
+8,13,123,7,[mask],[unknown],1
+8,14,107,7,[mask],[unknown],1
+8,15,118,7,[mask],[unknown],1
+8,16,104,7,[mask],[unknown],2
+8,17,104,9,[mask],1,2
+8,18,104,9,[mask],1,2
+8,19,104,9,[mask],1,2
+8,20,104,9,[mask],1,[mask]
+8,21,104,9,[other],7,[mask]
+8,22,104,9,[mask],7,[mask]
+8,23,104,9,[other],[unknown],[mask]
+8,24,112,9,[other],1,[mask]
+8,25,109,2,[other],1,[mask]
+8,26,104,9,[other],1,[mask]
+8,27,107,9,[other],1,[mask]
+8,28,107,2,[other],1,[mask]
+8,29,[other],9,[other],8,[other]
+8,30,[other],1,3,8,[other]
+8,31,[other],1,1,8,[other]
+8,32,[other],1,1,8,[other]
+8,33,[other],2,1,5,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv
new file mode 100644
index 00000000..a6cb20c5
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,104,7,[mask],0,1
+9,4,104,7,[mask],0,1
+9,5,104,7,9,0,1
+9,6,104,7,9,3,0
+9,7,104,0,9,3,1
+9,8,104,[mask],[mask],3,0
+9,9,104,0,9,3,1
+9,10,112,[mask],[mask],3,0
+9,11,[unknown],5,9,3,[other]
+9,12,112,0,[mask],3,0
+9,13,[unknown],7,[mask],[unknown],[other]
+9,14,[unknown],7,[mask],[unknown],0
+9,15,104,7,[mask],[unknown],1
+9,16,104,9,[mask],[unknown],2
+9,17,104,9,[mask],[unknown],0
+9,18,104,9,[other],4,[mask]
+9,19,104,9,[other],4,2
+9,20,104,9,[other],1,[mask]
+9,21,112,9,[other],1,[mask]
+9,22,112,2,[other],1,[mask]
+9,23,[other],2,6,[unknown],[mask]
+9,24,[other],9,[other],1,[other]
+9,25,107,1,[other],1,[mask]
+9,26,118,2,6,[unknown],[other]
+9,27,110,1,1,[unknown],[other]
+9,28,110,2,1,[unknown],[other]
+9,29,129,1,1,8,[other]
+9,30,129,2,1,5,[other]
+9,31,129,2,1,5,[other]
+9,32,129,2,1,5,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv
similarity index 74%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv
index 29653eb6..830dae7f 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv
@@ -1,17 +1,17 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 0,4,119,9,2,7,[other]
-0,5,111,9,2,7,[other]
-0,6,111,9,2,7,[other]
+0,5,104,9,2,7,[other]
+0,6,111,9,9,7,[other]
 0,7,111,9,9,7,[other]
 0,8,111,9,9,7,[other]
-0,9,107,1,9,7,[other]
-0,10,104,1,9,7,[other]
-0,11,116,1,9,7,[other]
-0,12,104,6,9,7,[other]
+0,9,104,9,9,7,[other]
+0,10,116,1,9,7,[other]
+0,11,104,6,9,7,[other]
+0,12,104,1,9,7,[other]
 0,13,104,1,9,7,[other]
-0,14,104,1,9,[unknown],[other]
+0,14,102,1,9,[unknown],[other]
 0,15,104,6,9,[unknown],[other]
-0,16,104,6,9,[unknown],[other]
+0,16,104,6,9,7,[other]
 0,17,104,6,9,7,[other]
 0,18,104,6,9,7,[other]
 0,19,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv
similarity index 86%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv
index ec1ff900..8f038e88 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,111,9,2,9,[other]
-1,5,111,1,2,7,[other]
+1,4,111,9,2,7,[other]
+1,5,111,1,9,7,[other]
 1,6,111,1,9,7,[other]
 1,7,111,1,9,7,[other]
 1,8,111,1,9,7,[other]
@@ -9,10 +9,10 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 1,11,129,1,9,[unknown],[other]
 1,12,102,1,9,[unknown],[other]
 1,13,104,6,9,[unknown],[other]
-1,14,104,1,9,[unknown],[other]
+1,14,104,6,9,[unknown],[other]
 1,15,104,6,9,[unknown],[other]
 1,16,104,6,9,[unknown],[other]
-1,17,104,6,9,[unknown],[other]
+1,17,104,6,9,7,[other]
 1,18,104,6,9,7,[other]
 1,19,104,6,9,7,[other]
 1,20,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv
new file mode 100644
index 00000000..a3056096
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+2,3,115,9,2,2,[unknown]
+2,4,118,9,2,7,[mask]
+2,5,115,9,9,7,[mask]
+2,6,115,9,9,7,[mask]
+2,7,115,9,9,7,[mask]
+2,8,112,9,9,7,[other]
+2,9,115,9,9,7,[other]
+2,10,104,9,9,7,[other]
+2,11,112,6,9,7,[other]
+2,12,112,9,9,7,[other]
+2,13,112,9,9,7,[other]
+2,14,112,9,9,7,[other]
+2,15,112,9,9,7,[other]
+2,16,112,9,9,7,[other]
+2,17,112,9,9,7,[other]
+2,18,112,9,9,8,[other]
+2,19,112,9,9,7,[other]
+2,20,112,9,9,7,[other]
+2,21,112,9,9,7,[other]
+2,22,112,9,9,7,[other]
+2,23,112,9,9,7,[other]
+2,24,112,9,9,7,[other]
+2,25,112,9,9,7,[other]
+2,26,112,9,9,8,[other]
+2,27,112,9,9,7,[other]
+2,28,112,9,9,7,[other]
+2,29,112,9,9,7,[other]
+2,30,112,9,9,7,[other]
+2,31,112,9,9,7,[other]
+2,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv
new file mode 100644
index 00000000..861690e2
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+3,3,112,9,2,2,[mask]
+3,4,112,9,2,[mask],[mask]
+3,5,115,9,2,9,[other]
+3,6,112,9,2,2,[other]
+3,7,112,9,2,2,[mask]
+3,8,112,9,9,2,[other]
+3,9,112,9,2,2,[other]
+3,10,112,9,9,7,[other]
+3,11,112,9,9,7,[other]
+3,12,112,9,9,7,[other]
+3,13,112,9,9,7,[other]
+3,14,112,9,9,7,[other]
+3,15,112,9,9,7,[other]
+3,16,112,9,9,7,[other]
+3,17,112,9,9,7,[other]
+3,18,112,9,9,8,[other]
+3,19,112,9,9,7,[other]
+3,20,112,9,9,7,[other]
+3,21,112,9,9,7,[other]
+3,22,112,9,9,7,[other]
+3,23,112,9,9,7,[other]
+3,24,112,9,9,7,[other]
+3,25,112,9,9,7,[other]
+3,26,112,9,9,8,[other]
+3,27,112,9,9,7,[other]
+3,28,112,9,9,7,[other]
+3,29,112,9,9,7,[other]
+3,30,112,9,9,7,[other]
+3,31,112,9,9,7,[other]
+3,32,112,9,9,7,[other]
+4,3,119,9,2,7,[other]
+4,4,112,9,9,7,[other]
+4,5,112,9,9,7,[other]
+4,6,112,9,9,7,[other]
+4,7,112,9,9,7,[other]
+4,8,112,9,9,7,[other]
+4,9,112,9,9,7,[other]
+4,10,112,9,9,7,[other]
+4,11,112,9,9,7,[other]
+4,12,112,9,9,8,[other]
+4,13,112,9,9,7,[other]
+4,14,112,9,9,7,[other]
+4,15,112,9,9,7,[other]
+4,16,112,9,9,7,[other]
+4,17,112,9,9,7,[other]
+4,18,112,9,9,7,[other]
+4,19,112,9,9,7,[other]
+4,20,112,9,9,8,[other]
+4,21,112,9,9,7,[other]
+4,22,112,9,9,7,[other]
+4,23,112,9,9,7,[other]
+4,24,112,9,9,7,[other]
+4,25,112,9,9,7,[other]
+4,26,112,9,9,7,[other]
+4,27,112,9,9,7,[other]
+4,28,112,9,9,8,[other]
+4,29,112,9,9,7,[other]
+4,30,112,9,9,7,[other]
+4,31,112,9,9,7,[other]
+4,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv
similarity index 79%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv
index cf194871..f36537ee 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv
@@ -4,17 +4,17 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 5,5,111,1,2,7,[other]
 5,6,111,1,2,7,[other]
 5,7,111,1,9,7,[other]
-5,8,111,1,9,7,[other]
+5,8,111,3,9,7,[other]
 5,9,102,1,9,7,[other]
-5,10,129,1,9,[unknown],[other]
+5,10,102,1,9,[unknown],[other]
 5,11,102,1,9,[unknown],[other]
 5,12,102,6,9,[unknown],[other]
 5,13,104,6,9,[unknown],[other]
-5,14,104,6,9,[unknown],[other]
-5,15,104,6,9,[unknown],[other]
-5,16,104,6,9,7,[other]
+5,14,104,6,9,5,[other]
+5,15,104,6,9,5,[other]
+5,16,104,9,9,5,[other]
 5,17,104,9,9,7,[other]
-5,18,104,9,9,7,[other]
+5,18,104,6,9,7,[other]
 5,19,104,6,9,7,[other]
 5,20,104,6,9,7,[other]
 5,21,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv
similarity index 55%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv
index 4cae85f8..ef0be269 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv
@@ -1,22 +1,22 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,4,111,9,2,7,[other]
-6,5,[unknown],9,2,7,[other]
-6,6,111,9,9,7,[other]
+6,5,111,9,2,7,[other]
+6,6,111,9,2,7,[other]
 6,7,111,1,9,7,[other]
 6,8,111,1,9,7,[other]
-6,9,111,1,9,7,[other]
+6,9,102,1,9,7,[other]
 6,10,102,1,9,[unknown],[other]
-6,11,129,1,9,[unknown],[other]
-6,12,102,1,9,[unknown],[other]
-6,13,104,6,9,[unknown],[other]
-6,14,104,1,9,[unknown],[other]
-6,15,104,6,9,[unknown],[other]
-6,16,104,6,9,[unknown],[other]
-6,17,104,6,9,[unknown],[other]
+6,11,104,6,9,[unknown],[other]
+6,12,104,1,9,[unknown],[other]
+6,13,102,6,9,[unknown],[other]
+6,14,104,6,9,[unknown],[other]
+6,15,104,6,9,7,[other]
+6,16,104,6,9,7,[other]
+6,17,104,6,9,7,[other]
 6,18,104,6,9,7,[other]
 6,19,104,6,9,7,[other]
 6,20,104,6,9,7,[other]
-6,21,104,6,9,7,[other]
+6,21,104,9,9,7,[other]
 6,22,104,6,9,7,[other]
 6,23,104,6,9,7,[other]
 6,24,104,6,9,7,[other]
@@ -29,23 +29,23 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,31,104,6,9,7,[other]
 6,32,104,6,9,7,[other]
 6,33,104,6,9,7,[other]
-7,3,111,2,9,7,[other]
-7,4,111,5,9,7,[other]
-7,5,117,1,9,7,[other]
-7,6,129,6,9,[unknown],[other]
-7,7,129,6,9,7,[other]
-7,8,104,9,9,7,[other]
-7,9,129,6,9,[unknown],[other]
-7,10,104,9,9,7,[other]
-7,11,104,[mask],9,7,[other]
-7,12,104,[mask],9,7,[other]
-7,13,104,[mask],9,7,[other]
-7,14,107,[mask],9,7,[other]
-7,15,107,1,9,7,[other]
+7,3,119,2,9,7,[other]
+7,4,104,6,9,7,[other]
+7,5,104,6,9,7,[other]
+7,6,104,6,9,7,[other]
+7,7,104,6,9,7,[other]
+7,8,104,6,9,7,[other]
+7,9,104,6,9,7,[other]
+7,10,104,6,9,7,[other]
+7,11,104,9,9,7,[other]
+7,12,104,6,9,7,[other]
+7,13,104,6,9,7,[other]
+7,14,104,6,9,7,[other]
+7,15,104,6,9,7,[other]
 7,16,104,6,9,7,[other]
 7,17,104,6,9,7,[other]
-7,18,104,1,9,7,[other]
-7,19,104,1,9,7,[other]
+7,18,104,6,9,7,[other]
+7,19,104,6,9,7,[other]
 7,20,104,6,9,7,[other]
 7,21,104,6,9,7,[other]
 7,22,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv
similarity index 92%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv
index b1dc8274..1b61ff20 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-predictions/sequifier-test-hp-search-custom-eval-run-2-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv
@@ -9,10 +9,10 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 8,11,129,1,9,[unknown],[other]
 8,12,102,1,9,[unknown],[other]
 8,13,104,6,9,[unknown],[other]
-8,14,104,1,9,[unknown],[other]
+8,14,104,6,9,[unknown],[other]
 8,15,104,6,9,[unknown],[other]
 8,16,104,6,9,[unknown],[other]
-8,17,104,6,9,[unknown],[other]
+8,17,104,6,9,7,[other]
 8,18,104,6,9,7,[other]
 8,19,104,6,9,7,[other]
 8,20,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv
new file mode 100644
index 00000000..4c3bf37e
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,113,3,2,[other],[other]
+9,4,119,9,2,7,[mask]
+9,5,129,9,9,[other],[other]
+9,6,119,9,9,7,[mask]
+9,7,119,9,9,7,[other]
+9,8,104,9,9,7,[other]
+9,9,129,9,9,2,[other]
+9,10,112,9,9,2,[other]
+9,11,112,9,9,2,[other]
+9,12,112,9,9,7,[other]
+9,13,112,9,9,7,[other]
+9,14,112,9,9,7,[other]
+9,15,112,9,9,7,[other]
+9,16,112,9,9,7,[other]
+9,17,112,9,9,8,[other]
+9,18,112,9,9,7,[other]
+9,19,112,9,9,7,[other]
+9,20,112,9,9,7,[other]
+9,21,112,9,9,7,[other]
+9,22,112,9,9,7,[other]
+9,23,112,9,9,7,[other]
+9,24,112,9,9,7,[other]
+9,25,112,9,9,8,[other]
+9,26,112,9,9,7,[other]
+9,27,112,9,9,7,[other]
+9,28,112,9,9,7,[other]
+9,29,112,9,9,7,[other]
+9,30,112,9,9,7,[other]
+9,31,112,9,9,7,[other]
+9,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv
new file mode 100644
index 00000000..32cbdbc6
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+0,4,121,1,1,2,1
+0,5,113,0,[other],8,2
+0,6,113,1,1,8,0
+0,7,105,3,[other],8,1
+0,8,113,1,1,8,0
+0,9,105,3,1,1,1
+0,10,128,1,8,8,0
+0,11,113,1,0,7,1
+0,12,105,8,9,1,1
+0,13,113,1,8,0,1
+0,14,105,1,9,7,1
+0,15,105,1,7,1,1
+0,16,113,1,9,1,1
+0,17,113,1,[other],1,2
+0,18,113,1,[other],7,1
+0,19,105,8,[other],2,1
+0,20,113,1,[other],8,1
+0,21,113,1,[other],7,1
+0,22,105,8,9,2,1
+0,23,113,1,8,0,1
+0,24,113,1,9,1,1
+0,25,113,1,[other],1,1
+0,26,113,1,[other],7,1
+0,27,105,8,[other],2,1
+0,28,113,1,[other],2,1
+0,29,113,1,[other],7,1
+0,30,105,8,[other],2,1
+0,31,113,1,8,2,1
+0,32,113,3,9,7,1
+0,33,105,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv
new file mode 100644
index 00000000..9d999e02
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+1,4,121,5,[other],9,1
+1,5,121,5,[other],7,1
+1,6,121,5,[other],7,1
+1,7,121,5,[other],7,1
+1,8,121,5,[other],0,1
+1,9,122,1,0,0,1
+1,10,113,1,7,8,1
+1,11,113,1,7,0,1
+1,12,105,1,7,2,1
+1,13,113,1,7,2,1
+1,14,113,[mask],7,1,1
+1,15,113,1,1,2,1
+1,16,113,[mask],[other],1,1
+1,17,113,1,1,2,1
+1,18,113,3,1,7,1
+1,19,105,1,[other],2,1
+1,20,113,1,1,7,1
+1,21,105,8,9,1,1
+1,22,113,1,8,0,1
+1,23,113,1,9,7,1
+1,24,105,1,[other],1,1
+1,25,113,1,[other],7,2
+1,26,105,8,[other],2,1
+1,27,113,1,[other],2,1
+1,28,113,1,[other],7,1
+1,29,105,8,[other],2,1
+1,30,113,1,[other],2,1
+1,31,113,1,[other],7,1
+1,32,105,8,9,2,1
+1,33,113,1,1,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv
new file mode 100644
index 00000000..c3a6b4fb
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+2,3,128,1,8,[unknown],2
+2,4,105,0,[other],7,2
+2,5,105,8,9,0,1
+2,6,128,1,8,0,2
+2,7,105,1,[other],6,1
+2,8,113,1,9,2,1
+2,9,113,1,[other],0,2
+2,10,113,1,[other],2,1
+2,11,113,1,[other],8,1
+2,12,113,1,[other],8,1
+2,13,113,1,[other],7,1
+2,14,105,8,[other],2,1
+2,15,113,1,8,2,1
+2,16,113,3,[other],7,1
+2,17,105,8,[other],2,1
+2,18,113,1,1,2,1
+2,19,113,3,[other],7,1
+2,20,105,8,[other],2,1
+2,21,113,1,[other],2,1
+2,22,113,1,[other],7,1
+2,23,105,8,[other],2,1
+2,24,113,1,[other],2,1
+2,25,113,1,[other],7,1
+2,26,105,8,[other],2,1
+2,27,113,1,[other],2,1
+2,28,113,1,9,7,1
+2,29,105,1,[other],7,1
+2,30,105,8,[other],2,1
+2,31,113,1,[other],2,1
+2,32,113,[mask],9,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv
new file mode 100644
index 00000000..5cb291b4
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+3,3,102,5,1,1,0
+3,4,128,0,1,5,[unknown]
+3,5,105,0,8,5,0
+3,6,127,[other],8,[mask],2
+3,7,109,0,8,[unknown],1
+3,8,127,0,8,[mask],[unknown]
+3,9,102,4,[other],[mask],1
+3,10,124,1,7,8,[unknown]
+3,11,113,1,7,0,2
+3,12,113,1,[other],6,1
+3,13,113,1,[other],8,1
+3,14,113,1,[other],2,1
+3,15,113,1,[other],2,1
+3,16,113,8,[other],1,1
+3,17,113,1,[other],2,1
+3,18,113,3,1,1,1
+3,19,103,1,[other],1,1
+3,20,103,1,9,7,1
+3,21,105,5,8,0,1
+3,22,105,1,8,2,1
+3,23,113,1,9,7,1
+3,24,105,1,9,1,1
+3,25,113,1,9,1,2
+3,26,113,1,[other],1,2
+3,27,113,1,[other],1,1
+3,28,113,1,[other],7,1
+3,29,105,8,[other],2,1
+3,30,113,1,8,2,1
+3,31,113,3,[other],1,1
+3,32,113,1,9,1,1
+4,3,127,0,[other],[unknown],1
+4,4,127,2,1,2,1
+4,5,127,3,[other],8,1
+4,6,127,1,1,0,1
+4,7,127,0,[other],0,1
+4,8,113,1,1,0,1
+4,9,113,3,[other],8,1
+4,10,113,1,[other],0,1
+4,11,113,1,[other],0,1
+4,12,113,1,9,0,1
+4,13,113,1,[other],1,1
+4,14,113,1,[other],1,1
+4,15,113,1,[other],1,1
+4,16,113,1,[other],1,1
+4,17,113,1,9,1,1
+4,18,105,0,[other],1,1
+4,19,113,0,8,0,1
+4,20,105,0,8,2,1
+4,21,113,0,8,0,1
+4,22,113,0,8,2,1
+4,23,113,3,[other],8,1
+4,24,113,1,[other],8,1
+4,25,113,1,[other],8,1
+4,26,113,1,[other],8,1
+4,27,113,1,[other],8,1
+4,28,113,1,[other],7,1
+4,29,105,1,[other],7,1
+4,30,105,1,9,0,1
+4,31,113,1,5,0,1
+4,32,113,1,5,1,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv
new file mode 100644
index 00000000..79f882a9
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+5,3,128,1,[other],4,1
+5,4,105,1,[other],7,1
+5,5,105,1,5,0,1
+5,6,113,1,[other],0,1
+5,7,113,1,[other],2,1
+5,8,113,1,5,2,1
+5,9,113,1,[other],1,1
+5,10,113,1,[other],2,1
+5,11,113,8,9,1,1
+5,12,113,1,[other],7,1
+5,13,105,8,9,1,1
+5,14,103,1,8,0,1
+5,15,113,1,8,7,1
+5,16,105,5,9,1,1
+5,17,113,1,8,0,1
+5,18,113,1,[other],7,1
+5,19,105,8,[other],2,1
+5,20,113,1,[other],8,1
+5,21,113,1,[other],7,1
+5,22,105,8,[other],2,1
+5,23,113,1,[other],2,1
+5,24,113,1,9,7,1
+5,25,105,1,[other],7,1
+5,26,105,8,5,2,1
+5,27,113,1,[other],2,1
+5,28,113,[mask],9,7,1
+5,29,113,1,[other],2,1
+5,30,113,1,[other],7,1
+5,31,105,8,[other],7,1
+5,32,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv
new file mode 100644
index 00000000..359be249
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+6,4,122,4,7,8,1
+6,5,113,1,0,8,1
+6,6,113,1,7,8,1
+6,7,113,1,7,8,1
+6,8,113,1,1,1,1
+6,9,120,0,9,1,1
+6,10,102,0,8,7,1
+6,11,105,0,7,8,1
+6,12,113,1,1,0,1
+6,13,113,1,[other],7,1
+6,14,105,8,5,7,1
+6,15,105,1,[other],8,1
+6,16,113,1,7,0,1
+6,17,113,1,[other],7,1
+6,18,113,8,[other],2,1
+6,19,113,1,[other],8,1
+6,20,113,1,[other],0,1
+6,21,113,1,[other],2,1
+6,22,113,1,[other],7,1
+6,23,105,8,5,2,1
+6,24,113,1,6,8,1
+6,25,113,1,[other],1,1
+6,26,113,1,1,7,1
+6,27,105,8,9,1,1
+6,28,113,1,8,0,1
+6,29,113,1,9,7,1
+6,30,105,1,[other],1,1
+6,31,113,1,9,7,2
+6,32,105,[mask],9,1,1
+6,33,113,1,[other],0,1
+7,3,105,0,9,8,0
+7,4,105,0,1,1,2
+7,5,105,0,9,6,[unknown]
+7,6,113,0,9,1,1
+7,7,113,0,9,1,2
+7,8,113,0,9,6,2
+7,9,113,9,8,1,1
+7,10,113,1,[other],8,1
+7,11,113,1,[other],8,1
+7,12,113,1,[other],8,1
+7,13,113,1,[other],7,1
+7,14,108,1,9,0,1
+7,15,113,1,0,7,1
+7,16,108,1,[other],7,1
+7,17,105,8,[other],7,1
+7,18,105,1,5,2,1
+7,19,113,1,[other],8,1
+7,20,113,1,[other],0,1
+7,21,113,8,[other],2,1
+7,22,113,1,[other],8,1
+7,23,113,1,[other],2,1
+7,24,113,8,[other],7,1
+7,25,105,1,[other],2,1
+7,26,113,1,9,7,1
+7,27,105,1,5,1,1
+7,28,113,1,9,8,1
+7,29,113,1,1,7,1
+7,30,105,8,9,1,1
+7,31,113,1,8,8,1
+7,32,113,1,[other],7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv
new file mode 100644
index 00000000..67187a34
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+8,4,124,1,[other],7,2
+8,5,121,8,[other],6,1
+8,6,114,1,[other],8,2
+8,7,113,1,[other],0,1
+8,8,113,1,[other],0,1
+8,9,113,8,[other],2,1
+8,10,113,1,[other],2,1
+8,11,113,1,[other],0,1
+8,12,113,1,7,2,1
+8,13,113,1,9,1,1
+8,14,113,1,[other],1,1
+8,15,113,1,[other],1,1
+8,16,113,1,9,1,1
+8,17,103,1,9,1,1
+8,18,103,0,8,1,1
+8,19,113,0,8,7,1
+8,20,105,0,9,2,1
+8,21,113,0,8,0,1
+8,22,113,0,[other],8,1
+8,23,113,0,[other],8,1
+8,24,113,1,[other],8,1
+8,25,113,1,[other],8,1
+8,26,113,1,[other],8,1
+8,27,113,1,[other],8,1
+8,28,113,1,[other],7,1
+8,29,108,8,[other],7,1
+8,30,108,1,[other],7,1
+8,31,108,8,5,7,1
+8,32,105,1,[other],0,1
+8,33,113,1,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv
new file mode 100644
index 00000000..6793444f
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,108,1,[other],8,1
+9,4,113,1,[other],7,1
+9,5,105,8,[other],7,1
+9,6,105,5,[other],2,1
+9,7,113,1,7,7,1
+9,8,113,8,[other],1,1
+9,9,113,1,[other],2,1
+9,10,113,1,[other],2,1
+9,11,113,8,5,2,1
+9,12,113,1,[other],8,1
+9,13,113,1,1,7,1
+9,14,105,8,9,7,1
+9,15,105,1,8,0,1
+9,16,113,1,8,7,1
+9,17,105,8,9,7,1
+9,18,105,1,8,2,1
+9,19,113,1,[other],7,2
+9,20,113,8,[other],7,1
+9,21,113,1,[other],2,1
+9,22,113,1,[other],8,2
+9,23,113,1,[other],7,1
+9,24,105,8,[other],2,1
+9,25,113,1,[other],2,1
+9,26,113,1,[other],2,1
+9,27,113,1,[other],0,1
+9,28,113,1,[other],2,1
+9,29,113,3,1,1,1
+9,30,105,1,[other],[mask],1
+9,31,113,1,9,7,1
+9,32,105,1,9,1,1

From edb9ffdda27c3cded740a35ec463f8bae0b37485 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 15 Jun 2026 11:04:27 +0200
Subject: [PATCH 58/81] Fix class_share calculation

---
 .../config/hyperparameter_search_config.py    |   2 +
 src/sequifier/config/train_config.py          |  22 ++
 src/sequifier/hyperparameter_search.py        |  30 +-
 src/sequifier/preprocess.py                   |   2 +-
 src/sequifier/train.py                        | 148 ++++++--
 .../hyperparameter-search-custom-eval.yaml    |   1 +
 tests/integration/test_training.py            |   2 +-
 .../unit/test_hyperparameter_search_config.py |  24 ++
 tests/unit/test_train.py                      | 347 +++++++++++++++++-
 9 files changed, 527 insertions(+), 51 deletions(-)

diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 94c20f00..a36ace88 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -670,6 +670,7 @@ class HyperparameterSearchConfig(BaseModel):
         metadata_config_path: The path to the data-driven configuration file.
         hp_search_name: The name for the hyperparameter search.
         search_strategy: The search strategy, either "sample" or "grid".
+        seed: Optional random seed passed to the Optuna sampler.
         n_samples: The number of samples to draw for the search.
         model_config_write_path: The path to write the model configurations to.
         training_data_path: The path to the training data.
@@ -700,6 +701,7 @@ class HyperparameterSearchConfig(BaseModel):
     metadata_config_path: str
     hp_search_name: str
     search_strategy: str = "bayesian"
+    seed: Optional[int] = None
     n_trials: Optional[int] = Field(None, alias="n_samples")
     prune_trials: Optional[bool] = True
     model_config_write_path: str
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index b8886d71..54efab88 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -35,6 +35,26 @@
 AnyType = str | int | float
 
 
+def _validate_class_share_log_columns(config_values: dict[str, Any]) -> None:
+    training_spec = config_values.get("training_spec", {})
+
+    for col in training_spec.get("class_share_log_columns", []):
+        if col not in config_values["target_columns"]:
+            raise ValueError(f"Class-share column {col!r} must be a target column.")
+        if config_values["target_column_types"].get(col) != "categorical":
+            raise ValueError(
+                f"Class-share column {col!r} must be a categorical target column."
+            )
+        if col not in config_values["n_classes"]:
+            raise ValueError(
+                f"Class-share column {col!r} has no configured class count."
+            )
+        if col not in config_values["id_maps"]:
+            raise ValueError(
+                f"Class-share column {col!r} has no index map for logging."
+            )
+
+
 @beartype
 def load_train_config(
     config_path: str, args_config: dict[str, Any], skip_metadata: bool
@@ -146,6 +166,8 @@ def load_train_config(
             source=f"metadata config '{metadata_config_path}'",
         )
 
+        _validate_class_share_log_columns(config_values)
+
     return try_catch_excess_keys(config_path, TrainModel, config_values)
 
 
diff --git a/src/sequifier/hyperparameter_search.py b/src/sequifier/hyperparameter_search.py
index 0fe83532..a481d938 100644
--- a/src/sequifier/hyperparameter_search.py
+++ b/src/sequifier/hyperparameter_search.py
@@ -6,7 +6,7 @@
 import sys
 import time
 import warnings
-from typing import Union
+from typing import Any, Union
 
 import optuna
 import torch._dynamo
@@ -24,6 +24,18 @@
 from sequifier.io.yaml import TrainModelDumper  # noqa: E402
 
 
+def create_sampler(config: Any) -> optuna.samplers.BaseSampler:
+    strategy = getattr(config, "search_strategy", "bayesian")
+    seed = getattr(config, "seed", None)
+    if strategy in ["sample"]:
+        return optuna.samplers.RandomSampler(seed=seed)
+    if strategy == "grid":
+        if hasattr(optuna.samplers, "BruteForceSampler"):
+            return optuna.samplers.BruteForceSampler(seed=seed)
+        raise RuntimeError("Grid search requires Optuna >= 3.1 for BruteForceSampler.")
+    return optuna.samplers.TPESampler(seed=seed)
+
+
 def set_pdeathsig():
     """Binds child process lifecycle to the parent orchestrator via Linux prctl."""
     if sys.platform.startswith("linux"):
@@ -216,7 +228,8 @@ def consume_metrics(last_read_pos: int, best_val_loss: float) -> tuple[int, floa
                 raise KeyError(
                     f"Metric '{metric}' missing in {eval_json_path}. Found keys: {list(eval_results.keys())}"
                 )
-            metrics.append(float(eval_results[metric]))
+            value = eval_results[metric]
+            metrics.append(float("nan") if value is None else float(value))
 
         if len(metrics) == 1:
             return metrics[0]
@@ -244,18 +257,7 @@ def hyperparameter_search(config_path: str, skip_metadata: bool) -> None:
     config = load_hyperparameter_search_config(config_path, skip_metadata)
 
     os.makedirs(os.path.join(config.project_root, "state", "optuna"), exist_ok=True)
-    strategy = getattr(config, "search_strategy", "bayesian")
-    if strategy in ["sample"]:
-        sampler = optuna.samplers.RandomSampler()
-    elif strategy == "grid":
-        if hasattr(optuna.samplers, "BruteForceSampler"):
-            sampler = optuna.samplers.BruteForceSampler()
-        else:
-            raise RuntimeError(
-                "Grid search requires Optuna >= 3.1 for BruteForceSampler."
-            )
-    else:  # "bayesian"
-        sampler = optuna.samplers.TPESampler()
+    sampler = create_sampler(config)
 
     storage_path = os.path.join(
         config.project_root, "state", "optuna", f"{config.hp_search_name}.db"
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index cde9834d..9c2d74fc 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -1291,7 +1291,7 @@ def load_precomputed_id_maps(
                     min_val = min(user_values)
                     if min_val == 2:
                         raise ValueError(
-                            f"Precomputed map {file} uses legacy user IDs starting at 2; shifting user IDs by 1 to reserve the BERT mask ID."
+                            f"Precomputed map {file} uses legacy user IDs starting at 2"
                         )
                     if min_val != SPECIAL_TOKEN_IDS.user_start:
                         raise ValueError(
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index a9c41f7a..8f294da7 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -14,7 +14,6 @@
 from typing import Any, Optional, Union  # noqa: E402
 
 import numpy as np  # noqa: E402
-import polars as pl  # noqa: E402
 import torch  # noqa: E402
 import torch._dynamo  # noqa: E402
 import torch.distributed as dist  # noqa: E402
@@ -52,6 +51,8 @@
 
 torch._dynamo.config.suppress_errors = True
 
+ClassCounts = dict[str, Tensor]
+
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
 from sequifier.helpers import (  # noqa: E402
@@ -520,6 +521,49 @@ def _get_evaluation_loss_mask(metadata: dict[str, Tensor]) -> Tensor:
     return valid_mask
 
 
+@beartype
+def accumulate_class_counts(
+    counts: ClassCounts,
+    output: dict[str, Tensor],
+    valid_mask: Tensor,
+    n_classes: dict[str, int],
+) -> None:
+    """Accumulates predicted class counts over valid evaluation tokens."""
+    flattened_mask = valid_mask.bool().T.contiguous().reshape(-1)
+
+    for col, running_counts in counts.items():
+        if col not in output:
+            raise RuntimeError(f"Output is missing class-share column {col!r}.")
+
+        predicted_ids = output[col].argmax(dim=-1).contiguous().reshape(-1)
+
+        if predicted_ids.numel() != flattened_mask.numel():
+            raise RuntimeError(
+                f"Prediction/mask size mismatch for {col!r}: "
+                f"{predicted_ids.numel()} predictions versus "
+                f"{flattened_mask.numel()} mask entries."
+            )
+
+        valid_predictions = predicted_ids[flattened_mask]
+
+        if valid_predictions.numel() == 0:
+            continue
+
+        batch_counts = torch.bincount(
+            valid_predictions.to(torch.int64),
+            minlength=n_classes[col],
+        )
+
+        if batch_counts.numel() != running_counts.numel():
+            raise RuntimeError(
+                f"Class-count size mismatch for {col!r}: "
+                f"{batch_counts.numel()} counts versus "
+                f"{running_counts.numel()} expected classes."
+            )
+
+        running_counts.add_(batch_counts)
+
+
 class TransformerEmbeddingModel(nn.Module):
     """A wrapper around the TransformerModel to expose the embedding functionality."""
 
@@ -1369,13 +1413,13 @@ def train_model(
                 self.start_epoch == 1
                 and self.hparams.training_spec.calculate_validation_loss_on_initialization
             ):
-                total_loss, total_losses, output = self._evaluate(
+                total_loss, total_losses, class_counts = self._evaluate(
                     valid_loader, ddp_model
                 )
                 elapsed = 0.0
 
                 self._log_epoch_results(
-                    0, 0, elapsed, total_loss, total_losses, output, 0
+                    0, 0, elapsed, total_loss, total_losses, class_counts, 0
                 )
             for epoch in range(self.start_epoch, self.hparams.training_spec.epochs + 1):
                 if (
@@ -1393,7 +1437,7 @@ def train_model(
 
                     self._train_epoch(train_loader, valid_loader, epoch, ddp_model)
 
-                    total_loss, total_losses, output = self._evaluate(
+                    total_loss, total_losses, class_counts = self._evaluate(
                         valid_loader, ddp_model
                     )
                     elapsed = time.time() - epoch_start_time
@@ -1405,7 +1449,7 @@ def train_model(
                         elapsed,
                         total_loss,
                         total_losses,
-                        output,
+                        class_counts,
                         total_expected_batches,
                     )
 
@@ -1673,7 +1717,7 @@ def _train_epoch(
 
                 if should_save_batch.item() == 1:
                     if self.save_batch_interval_minutes_val_loss:
-                        val_loss, val_losses, output = self._evaluate(
+                        val_loss, val_losses, class_counts = self._evaluate(
                             valid_loader, ddp_model
                         )
 
@@ -1687,7 +1731,7 @@ def _train_epoch(
                                 (current_time - self.last_batch_save_time),
                                 val_loss,
                                 val_losses,
-                                output,
+                                class_counts,
                                 current_global_step,
                             )
                             val_loss_batch[0] = float(val_loss)
@@ -1906,14 +1950,16 @@ def _transform_val(self, col: str, val: Tensor) -> Tensor:
     @beartype
     def _evaluate(
         self, valid_loader: DataLoader, ddp_model: Optional[nn.Module] = None
-    ) -> tuple[np.float32, dict[str, np.float32], dict[str, Tensor]]:
+    ) -> tuple[np.float32, dict[str, np.float32], ClassCounts]:
         """Evaluates the model on the validation set.
 
         Iterates through the validation data, calculates the total loss,
         and aggregates results across all processes if in distributed mode.
         Also calculates a one-time baseline loss on the first call. The
         DataLoader is expected to yield SequifierBatch objects. IDs are
-        currently unused during evaluation.
+        currently unused during evaluation. Class shares are counted over
+        the same valid token population as validation loss; for BERT this
+        means masked evaluation positions, not all sequence positions.
 
         Args:
             valid_loader: DataLoader for the validation dataset.
@@ -1923,12 +1969,33 @@ def _evaluate(
             A tuple containing:
             - The total aggregated validation loss (float).
             - A dictionary of aggregated losses for each target column (dict[str, float]).
-            - The output tensor dictionary from the last batch (used for class share logging).
+            - A dictionary of globally aggregated class-count tensors.
         """
 
         model_to_call = ddp_model if ddp_model is not None else self
         target_names = list(self.target_columns)
-        output = {}  # for type checking
+        class_count_columns = list(dict.fromkeys(self.class_share_log_columns))
+
+        for col in class_count_columns:
+            missing_class_ids = [
+                class_id
+                for class_id in range(self.n_classes[col])
+                if class_id not in self.index_maps[col]
+            ]
+            if missing_class_ids:
+                raise ValueError(
+                    f"Class-share column {col!r} is missing index-map entries "
+                    f"for class IDs {missing_class_ids}."
+                )
+
+        local_class_counts: ClassCounts = {
+            col: torch.zeros(
+                self.n_classes[col],
+                dtype=torch.int64,
+                device=self.device,
+            )
+            for col in class_count_columns
+        }
 
         def new_accumulators() -> tuple[dict[str, Tensor], dict[str, Tensor]]:
             return (
@@ -2056,11 +2123,24 @@ def finalize_components(
                         loss_sums,
                         token_counts,
                     )
+                    accumulate_class_counts(
+                        local_class_counts,
+                        output,
+                        valid_mask,
+                        self.n_classes,
+                    )
 
             total_loss_global, total_losses_global = finalize_components(
                 total_loss_sums, total_loss_counts, "validation"
             )
 
+            if self.hparams.training_spec.distributed:
+                for col in class_count_columns:
+                    dist.all_reduce(
+                        local_class_counts[col],
+                        op=dist.ReduceOp.SUM,
+                    )
+
             # Handle one-time baseline loss calculation with the same aggregation semantics.
             if not hasattr(self, "baseline_loss"):
                 baseline_loss_sums, baseline_loss_counts = new_accumulators()
@@ -2161,7 +2241,10 @@ def finalize_components(
                     k: np.float32(v.detach().cpu().item())
                     for k, v in total_losses_global.items()
                 },
-                output,
+                {
+                    col: counts.detach().cpu()
+                    for col, counts in local_class_counts.items()
+                },
             )
         finally:
             model_to_call.train(was_training)
@@ -2490,7 +2573,7 @@ def _log_epoch_results(
         elapsed: float,
         total_loss: np.float32,
         total_losses: dict[str, np.float32],
-        output: dict[str, Tensor],
+        class_counts: ClassCounts,
         global_step: int,
     ) -> None:
         """Logs the results of an epoch.
@@ -2504,8 +2587,8 @@ class share statistics (if configured) to the log file.
             elapsed: Time taken for the epoch (in seconds).
             total_loss: The total aggregated validation loss.
             total_losses: A dictionary of aggregated losses for each target.
-            output: The output tensor dictionary from the last validation batch,
-                    used for class share logging.
+            class_counts: Globally aggregated class-count tensors used for
+                class share logging.
             batch: Current batch number.
         """
         if self.rank == 0:
@@ -2543,31 +2626,28 @@ class share statistics (if configured) to the log file.
                 self.logger.info("[INFO]  - " + ", ".join(loss_strs))
 
             for categorical_column in self.class_share_log_columns:
-                output_values = (
-                    output[categorical_column]
-                    .argmax(dim=-1)  # class dimension
-                    .reshape(-1)  # scalar category ID per token
-                    .detach()
-                    .cpu()
-                    .numpy()
-                )
+                counts = class_counts[categorical_column].to(torch.int64)
+                total = counts.sum()
 
-                output_counts_df = (
-                    pl.Series("values", output_values)
-                    .value_counts()
-                    .with_columns(
-                        (pl.col("count") / pl.col("count").sum()).alias("share")
+                if total.item() == 0:
+                    self.logger.warning(
+                        "[WARNING] No valid predictions available for "
+                        f"class-share column {categorical_column!r}."
                     )
-                    .sort("values")
-                )
+                    continue
+
+                shares = counts.to(torch.float64) / total
 
                 value_shares = " | ".join(
-                    f"{self.index_maps[categorical_column][int(row['values'])]}: "
-                    f"{row['share']:5.5f}"
-                    for row in output_counts_df.iter_rows(named=True)
+                    f"{self.index_maps[categorical_column][class_id]}: "
+                    f"{shares[class_id].item():5.5f}"
+                    for class_id in range(counts.numel())
+                    if counts[class_id].item() > 0
                 )
 
-                self.logger.info(f"[INFO] {categorical_column}: {value_shares}")
+                self.logger.info(
+                    f"[INFO] {categorical_column} (n={total.item()}): {value_shares}"
+                )
 
             self.logger.info("-" * 89)
 
diff --git a/tests/configs/hyperparameter-search-custom-eval.yaml b/tests/configs/hyperparameter-search-custom-eval.yaml
index abb6b043..d93d0792 100644
--- a/tests/configs/hyperparameter-search-custom-eval.yaml
+++ b/tests/configs/hyperparameter-search-custom-eval.yaml
@@ -21,6 +21,7 @@ inference_batch_size: 10
 
 # Search Strategy
 search_strategy: bayesian
+seed: 101
 n_samples: 4
 
 # Configuration Loading Overrides (set to null to use values from metadata)
diff --git a/tests/integration/test_training.py b/tests/integration/test_training.py
index 65128d30..d90c32ef 100644
--- a/tests/integration/test_training.py
+++ b/tests/integration/test_training.py
@@ -148,7 +148,7 @@ def test_model_files_exists(run_training, run_training_from_checkpoint, project_
                 for suffix in ["best", "last"]
             ]
             + [
-                f"sequifier-test-hp-search-custom-eval-run-{i}-{suffix}-1.onnx"
+                f"sequifier-test-hp-search-custom-eval-run-{i}-{suffix}-3.onnx"
                 for i in range(4)
                 for suffix in ["best", "last"]
             ]
diff --git a/tests/unit/test_hyperparameter_search_config.py b/tests/unit/test_hyperparameter_search_config.py
index 77ea8639..16cfbf7e 100644
--- a/tests/unit/test_hyperparameter_search_config.py
+++ b/tests/unit/test_hyperparameter_search_config.py
@@ -1,3 +1,4 @@
+import optuna
 import pytest
 import yaml
 from pydantic import ValidationError
@@ -5,6 +6,7 @@
 from sequifier.config.hyperparameter_search_config import (
     TrainingSpecHyperparameterSampling,
 )
+from sequifier.hyperparameter_search import create_sampler
 from sequifier.io.yaml import TrainModelDumper
 
 
@@ -152,3 +154,25 @@ def test_bert_training_spec_dumps_to_plain_yaml():
         "type": "GeometricDistribution",
         "p": 1.0,
     }
+
+
+@pytest.mark.parametrize(
+    ("strategy", "sampler_name"),
+    [
+        ("sample", "RandomSampler"),
+        ("grid", "BruteForceSampler"),
+        ("bayesian", "TPESampler"),
+    ],
+)
+def test_create_sampler_passes_config_seed(monkeypatch, strategy, sampler_name):
+    recorded = {}
+
+    def fake_sampler(*, seed=None):
+        recorded["seed"] = seed
+        return object()
+
+    monkeypatch.setattr(optuna.samplers, sampler_name, fake_sampler)
+
+    create_sampler(type("Config", (), {"search_strategy": strategy, "seed": 123})())
+
+    assert recorded["seed"] == 123
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index e2527ab6..8418b669 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -1,10 +1,13 @@
 import copy
 import json
+from types import SimpleNamespace
 
+import numpy as np
 import pytest
 import torch
 import yaml
 from pydantic import ValidationError
+from torch.utils.data import DataLoader
 
 from sequifier.config.probabilities import PoissonDistributionFloor
 from sequifier.config.train_config import (
@@ -16,8 +19,13 @@
     load_train_config,
 )
 from sequifier.helpers import ModelWindowView, StoredWindowLayout
+from sequifier.io.batch import SequifierBatch
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS
-from sequifier.train import TransformerModel, _get_evaluation_loss_mask
+from sequifier.train import (
+    TransformerModel,
+    _get_evaluation_loss_mask,
+    accumulate_class_counts,
+)
 
 
 def _training_spec_kwargs(**overrides):
@@ -324,6 +332,76 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
     assert config.special_token_ids == SPECIAL_TOKEN_IDS.ids_by_label
 
 
+@pytest.mark.parametrize(
+    ("class_share_column", "metadata_overrides", "config_overrides", "match"),
+    [
+        (
+            "missing_col",
+            {},
+            {},
+            "Class-share column 'missing_col' must be a target column",
+        ),
+        (
+            "real_col",
+            {},
+            {},
+            "Class-share column 'real_col' must be a categorical target column",
+        ),
+        (
+            "cat_col",
+            {"n_classes": {}},
+            {"drop_n_classes": True},
+            "Class-share column 'cat_col' has no configured class count",
+        ),
+        (
+            "cat_col",
+            {"id_maps": {}},
+            {},
+            "Class-share column 'cat_col' has no index map for logging",
+        ),
+    ],
+)
+def test_load_train_config_validates_class_share_columns_after_metadata_load(
+    tmp_path,
+    model_config,
+    class_share_column,
+    metadata_overrides,
+    config_overrides,
+    match,
+):
+    config_path = tmp_path / "train.yaml"
+    metadata_path = tmp_path / "metadata.json"
+    config_values = model_config.model_dump()
+    config_values["project_root"] = str(tmp_path)
+    config_values["metadata_config_path"] = metadata_path.name
+    config_values["training_spec"]["class_share_log_columns"] = [class_share_column]
+
+    storage_layout = config_values.pop("storage_layout")
+    config_values.pop("window_view")
+    config_values["context_length"] = model_config.window_view.context_length
+
+    if config_overrides.get("drop_n_classes"):
+        config_values.pop("n_classes")
+
+    metadata = {
+        "split_paths": ["data/train.pt", "data/val.pt"],
+        "stored_context_width": storage_layout["stored_context_width"],
+        "max_target_offset": storage_layout["max_target_offset"],
+        "stored_window_layout_version": storage_layout["version"],
+        "column_types": config_values["column_types"],
+        "n_classes": model_config.n_classes,
+        "id_maps": model_config.id_maps,
+        "special_token_ids": SPECIAL_TOKEN_IDS.ids_by_label,
+    }
+    metadata.update(metadata_overrides)
+
+    config_path.write_text(yaml.safe_dump(config_values))
+    metadata_path.write_text(json.dumps(metadata))
+
+    with pytest.raises(ValueError, match=match):
+        load_train_config(str(config_path), {}, skip_metadata=False)
+
+
 def test_forward_train_shapes(model, model_config):
     """Tests the output shapes of the forward_train method."""
     batch_size = model_config.training_spec.batch_size
@@ -509,6 +587,165 @@ def test_evaluation_loss_mask_intersects_target_bert_and_sample_masks():
     )
 
 
+def _categorical_logits_from_batch_predictions(batch_predictions, n_classes):
+    prediction_ids = torch.tensor(batch_predictions, dtype=torch.long).T.contiguous()
+    return torch.nn.functional.one_hot(prediction_ids, num_classes=n_classes).float()
+
+
+def test_accumulate_class_counts_all_samples_valid():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [[0, 1], [1, 2]],
+            n_classes=3,
+        )
+    }
+    valid_mask = torch.ones(2, 2, dtype=torch.bool)
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([1, 2, 1]))
+
+
+def test_accumulate_class_counts_excludes_synthetic_samples():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [
+                [0, 1],
+                [2, 2],
+            ],
+            n_classes=3,
+        )
+    }
+    valid_mask = _get_evaluation_loss_mask(
+        {
+            "target_valid_mask": torch.ones(2, 2, dtype=torch.bool),
+            "sample_valid_mask": torch.tensor([True, False]),
+        }
+    )
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([1, 1, 0]))
+
+
+def test_accumulate_class_counts_excludes_target_padding():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [[0, 2]],
+            n_classes=3,
+        )
+    }
+    valid_mask = torch.tensor([[True, False]])
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([1, 0, 0]))
+
+
+def test_accumulate_class_counts_combines_sample_and_token_masks():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [
+                [0, 1],
+                [2, 2],
+            ],
+            n_classes=3,
+        )
+    }
+    valid_mask = _get_evaluation_loss_mask(
+        {
+            "target_valid_mask": torch.tensor(
+                [
+                    [True, False],
+                    [True, True],
+                ]
+            ),
+            "sample_valid_mask": torch.tensor([True, False]),
+        }
+    )
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([1, 0, 0]))
+
+
+def test_accumulate_class_counts_respects_bert_mask():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [
+                [0, 1],
+                [2, 2],
+            ],
+            n_classes=3,
+        )
+    }
+    valid_mask = _get_evaluation_loss_mask(
+        {
+            "target_valid_mask": torch.ones(2, 2, dtype=torch.bool),
+            "bert_mask": torch.tensor(
+                [
+                    [False, True],
+                    [True, True],
+                ]
+            ),
+            "sample_valid_mask": torch.tensor([True, False]),
+        }
+    )
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([0, 1, 0]))
+
+
+def test_accumulate_class_counts_retains_missing_class_slots():
+    counts = {"cat_col": torch.zeros(5, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [[3, 3]],
+            n_classes=5,
+        )
+    }
+    valid_mask = torch.ones(1, 2, dtype=torch.bool)
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 5})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([0, 0, 0, 2, 0]))
+
+
+def test_accumulate_class_counts_all_zero_when_no_positions_are_valid():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [[0, 1]],
+            n_classes=3,
+        )
+    }
+    valid_mask = torch.zeros(1, 2, dtype=torch.bool)
+
+    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+    assert torch.equal(counts["cat_col"], torch.tensor([0, 0, 0]))
+
+
+def test_accumulate_class_counts_raises_on_shape_mismatch():
+    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
+    output = {
+        "cat_col": _categorical_logits_from_batch_predictions(
+            [[0, 1, 2]],
+            n_classes=3,
+        )
+    }
+    valid_mask = torch.ones(1, 2, dtype=torch.bool)
+
+    with pytest.raises(RuntimeError, match="Prediction/mask size mismatch"):
+        accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
+
+
 def test_calculate_loss_components_aggregate_by_token_count():
     model = TransformerModel.__new__(TransformerModel)
     model.target_column_types = {"real_col": "real"}
@@ -535,6 +772,114 @@ def test_calculate_loss_components_aggregate_by_token_count():
     assert torch.isclose(aggregate, torch.tensor(200.0 / 101.0, dtype=torch.float64))
 
 
+class _QueuedEvalModel(torch.nn.Module):
+    def __init__(self, outputs):
+        super().__init__()
+        self.outputs = iter(outputs)
+
+    def forward(self, data, metadata=None, return_logits=False):
+        return next(self.outputs)
+
+
+def _evaluation_shell_model():
+    model = TransformerModel.__new__(TransformerModel)
+    model.input_columns = ["cat_col"]
+    model.target_columns = ["cat_col"]
+    model.target_column_types = {"cat_col": "categorical"}
+    model.n_classes = {"cat_col": 3}
+    model.loss_weights = None
+    model.criterion = {"cat_col": torch.nn.CrossEntropyLoss(reduction="none")}
+    model.device = "cpu"
+    model.class_share_log_columns = ["cat_col"]
+    model.index_maps = {  # type: ignore
+        "cat_col": {
+            0: "[unknown]",
+            1: "a",
+            2: "b",
+        }
+    }  # type: ignore
+    model.decoder = {"cat_col": torch.nn.Linear(1, 1)}
+    model.hparams = SimpleNamespace(
+        seed=42,
+        training_spec=SimpleNamespace(
+            distributed=False,
+            layer_autocast=False,
+            data_parallelism="none",
+            training_objective="causal",
+        ),
+    )
+    return model
+
+
+def _validation_batch(predictions, sample_valid_mask=None):
+    batch_size = len(predictions)
+    seq_len = len(predictions[0])
+    inputs = {
+        "cat_col": torch.zeros(batch_size, seq_len, dtype=torch.long),
+    }
+    targets = {
+        "cat_col": torch.tensor(predictions, dtype=torch.long),
+    }
+    valid_mask = torch.ones(batch_size, seq_len, dtype=torch.bool)
+    metadata = {
+        "attention_valid_mask": valid_mask.clone(),
+        "target_valid_mask": valid_mask,
+    }
+    if sample_valid_mask is not None:
+        metadata["sample_valid_mask"] = torch.tensor(
+            sample_valid_mask,
+            dtype=torch.bool,
+        )
+    return SequifierBatch(inputs=inputs, targets=targets, metadata=metadata)
+
+
+def test_evaluate_returns_class_counts_across_all_validation_batches():
+    model = _evaluation_shell_model()
+    batches = [
+        _validation_batch([[0, 0, 1]]),
+        _validation_batch([[2, 2]]),
+    ]
+    outputs = [
+        {"cat_col": _categorical_logits_from_batch_predictions([[0, 0, 1]], 3)},
+        {"cat_col": _categorical_logits_from_batch_predictions([[2, 2]], 3)},
+    ]
+    eval_model = _QueuedEvalModel(outputs)
+    valid_loader = DataLoader(batches, batch_size=None)
+
+    total_loss, total_losses, class_counts = TransformerModel._evaluate(
+        model,
+        valid_loader,
+        eval_model,
+    )
+
+    assert np.isfinite(total_loss)
+    assert np.isfinite(total_losses["cat_col"])
+    assert torch.equal(class_counts["cat_col"], torch.tensor([2, 1, 2]))
+    assert eval_model.training
+    assert model.baseline_loss >= 0
+
+
+def test_evaluate_excludes_synthetic_final_batch_from_class_counts():
+    model = _evaluation_shell_model()
+    batches = [
+        _validation_batch([[0, 1]]),
+        _validation_batch([[2, 2]], sample_valid_mask=[False]),
+    ]
+    outputs = [
+        {"cat_col": _categorical_logits_from_batch_predictions([[0, 1]], 3)},
+        {"cat_col": _categorical_logits_from_batch_predictions([[2, 2]], 3)},
+    ]
+    valid_loader = DataLoader(batches, batch_size=None)
+
+    _, _, class_counts = TransformerModel._evaluate(
+        model,
+        valid_loader,
+        _QueuedEvalModel(outputs),
+    )
+
+    assert torch.equal(class_counts["cat_col"], torch.tensor([1, 1, 0]))
+
+
 def test_calculate_loss_zero_token_training_batch_is_differentiable():
     model = TransformerModel.__new__(TransformerModel)
     model.target_column_types = {"real_col": "real"}

From 9b8245d9aebba075088a3c4036b1ef3e948b0e65 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 15 Jun 2026 11:52:02 +0200
Subject: [PATCH 59/81] shorten docstrings & clean up

---
 hooks/consolidate_docs.py                     |   8 +-
 hooks/correct_docs_version.py                 |  25 +-
 hooks/validate_docs_deps.py                   |  28 +-
 .../config/hyperparameter_search_config.py    | 175 +----
 src/sequifier/config/infer_config.py          |  41 +-
 src/sequifier/config/preprocess_config.py     |  37 +-
 src/sequifier/config/train_config.py          |  90 +--
 src/sequifier/distributed/env.py              |   8 +-
 src/sequifier/helpers.py                      | 222 +-----
 src/sequifier/hyperparameter_search.py        |  44 +-
 src/sequifier/infer.py                        | 498 +-----------
 .../io/sequifier_dataset_from_file.py         |  22 +-
 .../sequifier_dataset_from_folder_parquet.py  |  17 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |  51 +-
 .../io/sequifier_dataset_from_folder_pt.py    |  18 +-
 .../sequifier_dataset_from_folder_pt_lazy.py  |  46 +-
 src/sequifier/io/yaml.py                      |  49 +-
 src/sequifier/make.py                         |   6 +-
 src/sequifier/model/layers.py                 |   7 -
 src/sequifier/optimizers/ademamix.py          |  59 +-
 src/sequifier/optimizers/optimizers.py        |   9 +-
 src/sequifier/preprocess.py                   | 708 +-----------------
 src/sequifier/sequifier.py                    |  17 +-
 src/sequifier/train.py                        | 473 ++----------
 src/sequifier/visualize_training.py           |  67 +-
 tests/integration/test_inference.py           |  12 +-
 tests/integration/test_preprocessing.py       |   7 +-
 tests/integration/test_visualize_training.py  |   3 -
 .../source_scripts/hp_search_eval_script.py   |   4 -
 ...uifier_dataset_from_folder_parquet_lazy.py |  12 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  16 +-
 tests/unit/test_helpers.py                    |  33 +-
 tests/unit/test_infer.py                      |  46 +-
 tests/unit/test_preprocess.py                 |  32 +-
 tests/unit/test_train.py                      |  10 +-
 tools/analysis.py                             |  11 +-
 tools/convert_to_pt.py                        |   6 -
 tools/resize_pt_files.py                      |  56 +-
 tools/stream_folder_via_scp.py                |  26 +-
 39 files changed, 259 insertions(+), 2740 deletions(-)

diff --git a/hooks/consolidate_docs.py b/hooks/consolidate_docs.py
index 324b8426..8426f6f5 100644
--- a/hooks/consolidate_docs.py
+++ b/hooks/consolidate_docs.py
@@ -2,7 +2,6 @@
 import sys
 from pathlib import Path
 
-# 1. Define the exact order of files to read
 FILES_TO_READ = [
     "README.md",
     "documentation/configs/preprocess.md",
@@ -19,7 +18,6 @@
 def main():
     consolidated_content = []
 
-    # 2. Read contents
     for filepath in FILES_TO_READ:
         path = Path(filepath)
         if not path.is_file():
@@ -28,23 +26,19 @@ def main():
 
         consolidated_content.append(path.read_text(encoding="utf-8"))
 
-    # 3. Join with newlines to preserve markdown structure between files
     final_content = "\n\n".join(consolidated_content)
 
     output_path = Path(OUTPUT_FILE)
 
-    # 4. Check current content to avoid failing the hook if nothing changed
     current_content = ""
     if output_path.is_file():
         current_content = output_path.read_text(encoding="utf-8")
 
-    # 5. Write and exit with status code 1 if an update was needed
     if current_content != final_content:
-        # Ensure the output directory exists
         output_path.parent.mkdir(parents=True, exist_ok=True)
         output_path.write_text(final_content, encoding="utf-8")
         print(f"Hook updated {OUTPUT_FILE}")
-        sys.exit(1)  # pre-commit requires a non-zero exit code when a file is modified
+        sys.exit(1)
 
 
 if __name__ == "__main__":
diff --git a/hooks/correct_docs_version.py b/hooks/correct_docs_version.py
index 655e3328..2edd6763 100755
--- a/hooks/correct_docs_version.py
+++ b/hooks/correct_docs_version.py
@@ -1,24 +1,15 @@
 #!/usr/bin/env python
 
-"""
-Compares the project version in:
-1. pyproject.toml (project.version or tool.poetry.version)
-2. docs/source/conf.py (release = "...")
-3. README.md (BibTeX citation version = {...})
-
-Exits with a non-zero status code if they do not ALL match exactly.
-"""
+"""Require pyproject, docs conf, and README citation versions to match."""
 
 import re
 import sys
 from pathlib import Path
 
-# --- Configuration ---
 ROOT_DIR = Path(__file__).parent.parent
 PYPROJECT_PATH = ROOT_DIR / "pyproject.toml"
 CONF_PATH = ROOT_DIR / "docs/source/conf.py"
 README_PATH = ROOT_DIR / "README.md"
-# ---------------------
 
 try:
     try:
@@ -31,17 +22,15 @@
 
 
 def get_toml_version() -> str | None:
-    """Fetches version from pyproject.toml."""
+    """Read pyproject version."""
     if not PYPROJECT_PATH.exists():
         return None
     try:
         data = tomllib.loads(PYPROJECT_PATH.read_text(encoding="utf-8"))
 
-        # Check standard PEP 621
         val = None
         if "project" in data and "version" in data["project"]:
             val = data["project"]["version"]
-        # Check Poetry
         elif (
             "tool" in data
             and "poetry" in data["tool"]
@@ -55,12 +44,11 @@ def get_toml_version() -> str | None:
 
 
 def get_conf_version() -> str | None:
-    """Fetches release from docs/source/conf.py."""
+    """Read docs release version."""
     if not CONF_PATH.exists():
         return None
     try:
         content = CONF_PATH.read_text(encoding="utf-8")
-        # Regex for: release = "..." or release = '...'
         match = re.search(r"^release\s*=\s*['\"]([^'\"]+)['\"]", content, re.MULTILINE)
         return match.group(1).strip() if match else None
     except Exception:
@@ -68,13 +56,11 @@ def get_conf_version() -> str | None:
 
 
 def get_readme_version() -> str | None:
-    """Fetches version from README.md BibTeX citation."""
+    """Read README BibTeX version."""
     if not README_PATH.exists():
         return None
     try:
         content = README_PATH.read_text(encoding="utf-8")
-        # Regex for BibTeX: version = {1.0.0} or version = "1.0.0"
-        # Captures the content inside the first { } or " " encountered after 'version ='
         match = re.search(r"version\s*=\s*[{\"]([^}\"]+)[}\"]", content)
         return match.group(1).strip() if match else None
     except Exception:
@@ -88,12 +74,10 @@ def main():
     v_conf = get_conf_version()
     v_readme = get_readme_version()
 
-    # 1. Print detected versions for transparency
     print(f"  pyproject.toml: '{v_toml}'")
     print(f"  docs/conf.py:   '{v_conf}'")
     print(f"  README.md:      '{v_readme}'")
 
-    # 2. Check for missing versions
     if v_toml is None or v_conf is None or v_readme is None:
         print(
             "\n❌ Error: Could not extract version from one or more files.",
@@ -101,7 +85,6 @@ def main():
         )
         sys.exit(1)
 
-    # 3. Check for exact equality
     if v_toml == v_conf == v_readme:
         print("\n✅ All versions match.")
         sys.exit(0)
diff --git a/hooks/validate_docs_deps.py b/hooks/validate_docs_deps.py
index 46c434ed..44866309 100755
--- a/hooks/validate_docs_deps.py
+++ b/hooks/validate_docs_deps.py
@@ -2,30 +2,20 @@
 import sys
 from pathlib import Path
 
-# --- Configuration ---
 PYPROJECT_PATH = Path("./pyproject.toml")
 WORKFLOW_PATH = Path("./.github/workflows/docs.yml")
 
 
 def normalize_dep(dep_string):
-    """
-    Removes whitespace, quotes, and trailing commas for consistent comparison.
-    Example: ' "polars>= 1.0" ' -> 'polars>=1.0'
-    """
-    # Remove single and double quotes
+    """Normalize dependency strings for comparison."""
     s = dep_string.replace('"', "").replace("'", "")
-    # Remove all whitespace
     s = "".join(s.split())
-    # Remove trailing commas if present
     s = s.rstrip(",")
     return s
 
 
 def parse_pyproject_toml(file_path):
-    """
-    Extracts dependencies from the [project] dependencies list.
-    We use a simple state parser to avoid requiring 'tomli' on Python < 3.11.
-    """
+    """Parse project dependency strings without tomli."""
     deps = set()
     if not file_path.exists():
         print(f"Error: {file_path} not found.")
@@ -39,18 +29,15 @@ def parse_pyproject_toml(file_path):
     for line in lines:
         stripped = line.strip()
 
-        # Detect start of dependencies block
         if stripped.startswith("dependencies = ["):
             in_dependencies = True
             continue
 
-        # Detect end of block
         if in_dependencies and stripped.startswith("]"):
             in_dependencies = False
             break
 
         if in_dependencies:
-            # Ignore empty lines or comments inside the block
             if not stripped or stripped.startswith("#"):
                 continue
 
@@ -62,10 +49,7 @@ def parse_pyproject_toml(file_path):
 
 
 def parse_workflow_yml(file_path):
-    """
-    Extracts dependencies specifically from the 'pip install \' block
-    in the Install dependencies step.
-    """
+    """Parse docs workflow pip-install dependency strings."""
     deps = set()
     if not file_path.exists():
         print(f"Error: {file_path} not found.")
@@ -79,20 +63,16 @@ def parse_workflow_yml(file_path):
     for line in lines:
         stripped = line.strip()
 
-        # Logic to find the specific pip install block that uses line continuation
-        # We look for 'pip install \' but exclude 'pip install -r' lines
         if "pip install \\" in line and "-r " not in line:
             in_pip_block = True
             continue
 
         if in_pip_block:
-            # If the line contains a quoted string, it's a dependency
             if '"' in line or "'" in line:
                 normalized = normalize_dep(stripped.rstrip("\\"))
                 if normalized:
                     deps.add(normalized)
 
-            # If the line does not end with a backslash, the multiline command is over
             if not line.rstrip().endswith("\\"):
                 in_pip_block = False
 
@@ -105,10 +85,8 @@ def main():
     toml_deps = parse_pyproject_toml(PYPROJECT_PATH)
     yml_deps = parse_workflow_yml(WORKFLOW_PATH)
 
-    # 1. Check for items in TOML but missing in YAML
     missing_in_yml = toml_deps - yml_deps
 
-    # 2. Check for items in YAML but missing in TOML
     missing_in_toml = yml_deps - toml_deps
 
     if not missing_in_yml and not missing_in_toml:
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index a36ace88..0ca0ccfe 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -29,13 +29,7 @@
 
 
 class FloatDistribution(BaseModel):
-    """Pydantic model representing a floating-point hyperparameter distribution for Optuna.
-
-    Attributes:
-        low (float): The lower bound of the distribution.
-        high (float): The upper bound of the distribution.
-        log (bool): If True, sample from the distribution in the log domain. Defaults to False.
-    """
+    """Optuna float range with optional step/log sampling."""
 
     low: float
     high: float
@@ -53,14 +47,7 @@ def validate_step_and_log(self):
 
 
 class IntDistribution(BaseModel):
-    """Pydantic model representing an integer hyperparameter distribution for Optuna.
-
-    Attributes:
-        low (int): The lower bound of the distribution.
-        high (int): The upper bound of the distribution.
-        step (int): The spacing between valid integer values. Defaults to 1.
-        log (bool): If True, sample from the distribution in the log domain. Defaults to False.
-    """
+    """Optuna integer range with step/log sampling."""
 
     low: int
     high: int
@@ -100,12 +87,7 @@ def sample_param(
 
 
 class BERTSpecHyperparameterSampling(BaseModel):
-    """Pydantic model for BERT objective hyperparameter sampling.
-
-    Each BERT spec field is sampled independently so Optuna can track the
-    conditional search space without receiving nested objects as categorical
-    values.
-    """
+    """Search space for BERT objective masking parameters."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
@@ -145,22 +127,7 @@ def sample_trial(self, trial: Any) -> BERTSpecModel:
 def load_hyperparameter_search_config(
     config_path: str, skip_metadata: bool
 ) -> "HyperparameterSearchConfig":
-    """Load a hyperparameter search configuration from a YAML file.
-
-    This function reads a YAML configuration file, processes it to include
-    data-driven configurations if needed, and returns a HyperparameterSearchConfig
-    object.
-
-    Args:
-        config_path: The path to the hyperparameter search configuration file.
-        skip_metadata: A boolean flag indicating whether the configuration is
-            for unprocessed data. If False, it will load and integrate
-            data-driven configurations.
-
-    Returns:
-        An instance of the HyperparameterSearchConfig class, populated with the
-        configuration from the file.
-    """
+    """Load hyperparameter-search YAML plus optional metadata-derived fields."""
     with open(config_path, "r") as f:
         config_values = yaml.safe_load(f)
 
@@ -238,40 +205,7 @@ def load_hyperparameter_search_config(
 
 
 class TrainingSpecHyperparameterSampling(BaseModel):
-    """Pydantic model for training specification hyperparameter sampling.
-
-    Attributes:
-        device: The device to train on (e.g., 'cuda', 'cpu').
-        epochs: A list of possible numbers of epochs to train for.
-        log_interval: The interval in batches for logging.
-        class_share_log_columns: Columns for which to log class share.
-        early_stopping_epochs: Number of epochs for early stopping.
-        save_interval_epochs: Interval in epochs for saving model checkpoints.
-        save_latest_interval_minutes: the time interval in which a checkpoint is written to the "latest" checkpoint path
-        save_batch_interval_minutes: the time interval in which a checkpoint is written to a unique checkpoint path
-        save_batch_interval_minutes_val_loss: calculate val loss at the moment of batch interval saving
-        calculate_validation_loss_on_initialization: calculate val loss on weight initialization
-        training_objective: Training objective choices, either 'causal' or 'bert'.
-        batch_size: A list of possible batch sizes.
-        learning_rate: A list of possible learning rates.
-        bert_spec: Optional BERT hyperparameter search space. Required if 'bert' can be sampled.
-        criterion: A dictionary mapping target columns to loss functions.
-        class_weights: Optional dictionary mapping columns to class weights.
-        accumulation_steps: A list of possible gradient accumulation steps.
-        dropout: A list of possible dropout rates.
-        loss_weights: Optional dictionary mapping columns to loss weights.
-        optimizer: A list of possible optimizer configurations.
-        scheduler: A list of possible scheduler configurations.
-        continue_training: Flag to continue training from a checkpoint.
-        layer_type_dtypes: Dictionary mapping layer types (linear, embedding, norm) to dtypes (bfloat16, float8_e4m3fn).
-        layer_autocast: Whether to use autocast
-        sampling_strategy: data sampling in distributed training: 'exact', 'oversampling' or 'undersampling'
-        data_parallelism: 'DDP' or 'FSDP'
-        fsdp_cpu_offload: fsdp cpu offload
-        torch_compile: compile entire model ('outer') or transformer layers ('inner') with torch.compile, alternatively 'none'
-        float32_matmul_precision: precision level of float32 computations. One of 'highest', 'high' and 'medium'
-
-    """
+    """Training-spec search space with paired LR/scheduler candidates."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
@@ -322,17 +256,7 @@ class TrainingSpecHyperparameterSampling(BaseModel):
     float32_matmul_precision: str = "highest"
 
     def __init__(self, **kwargs):
-        """Initialize the TrainingSpecHyperparameterSampling instance.
-
-        This method initializes the Pydantic BaseModel and then processes the
-        optimizer and scheduler configurations from the provided keyword
-        arguments, converting them into DotDict objects.
-
-        Args:
-            **kwargs: Keyword arguments that correspond to the attributes of this
-                class. The 'optimizer' and 'scheduler' arguments are expected
-                to be lists of dictionaries.
-        """
+        """Normalize optimizer/scheduler dicts after Pydantic validation."""
         super().__init__(
             **{k: v for k, v in kwargs.items() if k not in ["optimizer", "scheduler"]}
         )
@@ -433,18 +357,7 @@ def validate_scheduler_config(cls, v, info_dict):
         return v
 
     def sample_trial(self, trial: Any) -> TrainingSpecModel:
-        """Samples training hyperparameters using an Optuna trial.
-
-        This method leverages the provided Optuna trial to suggest values for
-        hyperparameters like batch size, dropout, and learning rate based on the
-        defined search spaces (categorical lists or distributions).
-
-        Args:
-            trial (Any): The Optuna trial object used for suggesting hyperparameters.
-
-        Returns:
-            TrainingSpecModel: A populated training specification model with the sampled hyperparameters.
-        """
+        """Sample training hyperparameters for one Optuna trial."""
         lr_sched_index = trial.suggest_categorical(
             "lr_sched_index", list(range(len(self.learning_rate)))
         )
@@ -519,17 +432,7 @@ def sample_trial(self, trial: Any) -> TrainingSpecModel:
 
 
 class ModelSpecHyperparameterSampling(BaseModel):
-    """Pydantic model for model specification hyperparameter sampling.
-
-    Attributes:
-        initial_embedding_dim: A list of possible sizes for the initial input embedding.
-        feature_embedding_dims: A list of possible dictionaries defining embedding dimensions for each input column.
-        joint_embedding_dim: A list of possible sizes for the joint embedding layer projection.
-        dim_model: A list of possible numbers of expected features in the input (d_model).
-        n_head: A list of possible numbers of heads in the multi-head attention models.
-        dim_feedforward: A list of possible dimensions of the feedforward network model.
-        num_layers: A list of possible numbers of layers in the transformer model.
-    """
+    """Model-architecture search space with paired width choices."""
 
     initial_embedding_dim: list[int]
     joint_embedding_dim: list[Optional[int]]
@@ -584,19 +487,7 @@ def validate_model_spec(cls, v, info):
         return v
 
     def sample_trial(self, trial: Any) -> ModelSpecModel:
-        """Samples model architecture hyperparameters using an Optuna trial.
-
-        This method uses the Optuna trial to suggest structural parameters such as
-        the number of layers, feedforward dimensions, and attention heads. It ensures
-        that dependent dimensions (like `n_head` and `dim_model`) stay correctly paired
-        and that invalid key-value head combinations are filtered out.
-
-        Args:
-            trial (Any): The Optuna trial object used for suggesting hyperparameters.
-
-        Returns:
-            ModelSpecModel: A populated model specification model with the sampled architecture parameters.
-        """
+        """Sample architecture hyperparameters for one Optuna trial."""
         dim_model_idx = trial.suggest_categorical(
             "dim_model_idx", list(range(len(self.dim_model)))
         )
@@ -663,39 +554,7 @@ def sample_trial(self, trial: Any) -> ModelSpecModel:
 
 
 class HyperparameterSearchConfig(BaseModel):
-    """Pydantic model for hyperparameter search configuration.
-
-    Attributes:
-        project_root: The path to the sequifier project directory.
-        metadata_config_path: The path to the data-driven configuration file.
-        hp_search_name: The name for the hyperparameter search.
-        search_strategy: The search strategy, either "sample" or "grid".
-        seed: Optional random seed passed to the Optuna sampler.
-        n_samples: The number of samples to draw for the search.
-        model_config_write_path: The path to write the model configurations to.
-        training_data_path: The path to the training data.
-        validation_data_path: The path to the validation data.
-        read_format: The file format of the input data.
-        input_columns: A list of lists of columns to be used for training.
-        column_types: A list of dictionaries mapping columns to their types.
-        categorical_columns: A list of lists of categorical columns.
-        real_columns: A list of lists of real-valued columns.
-        target_columns: The list of target columns for model training.
-        target_column_types: A dictionary mapping target columns to their types.
-        id_maps: A dictionary mapping categorical values to their indexed representation.
-        context_length: A list of possible sequence lengths.
-        n_classes: The number of classes for each categorical column.
-        inference_batch_size: The batch size for inference.
-        export_onnx: If True, exports the model in ONNX format.
-        export_pt: If True, exports the model using torch.save.
-        export_with_dropout: If True, exports the model with dropout enabled.
-        model_hyperparameter_sampling: The sampling configuration for model hyperparameters.
-        training_hyperparameter_sampling: The sampling configuration for training hyperparameters.
-        evaluation_inference_config: The inference config to infer on for hyperparameter search optimization
-        evaluation_script: The script that outputs the evaluation metrics, typically from the inference output
-        evaluation_metrics: The evaluation metrics to optimize during hyperparameter search
-        evaluation_metric_directions: The direction to optimize evaluation_metrics in. Only 'minimize' and 'maximize' are allowed
-    """
+    """Top-level Optuna search config."""
 
     project_root: str
     metadata_config_path: str
@@ -843,19 +702,7 @@ def validate_search_strategy(cls, v: str) -> str:
         return v
 
     def sample_trial(self, trial: Any, run_index: int) -> TrainModel:
-        """Generates a complete training configuration using an Optuna trial.
-
-        This method orchestrates the sampling of both model and training specifications,
-        as well as data sequence parameters, combining them into a final configuration
-        ready for model execution.
-
-        Args:
-            trial (Any): The Optuna trial object used for suggesting hyperparameters.
-            run_index (int): The current run/trial index, used to assign a unique name to the model.
-
-        Returns:
-            TrainModel: A fully populated configuration instance for the current trial.
-        """
+        """Sample a concrete TrainModel for one trial/run index."""
         model_spec = self.model_hyperparameter_sampling.sample_trial(trial)
 
         input_columns_index = trial.suggest_categorical(
diff --git a/src/sequifier/config/infer_config.py b/src/sequifier/config/infer_config.py
index 84fa6e8e..14b5f35b 100644
--- a/src/sequifier/config/infer_config.py
+++ b/src/sequifier/config/infer_config.py
@@ -29,17 +29,7 @@
 def load_inferer_config(
     config_path: str, args_config: dict, skip_metadata: bool
 ) -> "InfererModel":
-    """
-    Load inferer configuration from a YAML file and update it with args_config.
-
-    Args:
-        config_path: Path to the YAML configuration file.
-        args_config: Dictionary containing additional configuration arguments.
-        skip_metadata: Flag indicating whether to process the configuration or not.
-
-    Returns:
-        InfererModel instance with loaded configuration.
-    """
+    """Load inference YAML plus CLI overrides and optional metadata fields."""
     with open(config_path, "r") as f:
         config_values = yaml.safe_load(f)
     config_values.update(args_config)
@@ -123,34 +113,7 @@ def load_inferer_config(
 
 
 class InfererModel(BaseModel):
-    """Pydantic model for inference configuration.
-
-    Attributes:
-        project_root: The path to the sequifier project directory.
-        metadata_config_path: The path to the data-driven configuration file.
-        model_path: The path to the trained model file(s).
-        model_type: The type of model, either 'embedding' or 'generative'.
-        data_path: The path to the data to be used for inference.
-        training_config_path: The path to the training configuration file.
-        read_format: The file format of the input data (e.g., 'csv', 'parquet').
-        write_format: The file format for the inference output.
-        input_columns: The list of input columns used for inference.
-        categorical_columns: A list of columns that are categorical.
-        real_columns: A list of columns that are real-valued.
-        target_columns: The list of target columns for inference.
-        column_types: A dictionary mapping each column to its numeric type ('int64' or 'float64').
-        target_column_types: A dictionary mapping target columns to their types ('categorical' or 'real').
-        output_probabilities: If True, outputs the probability distributions for categorical target columns.
-        map_to_id: If True, maps categorical output values back to their original IDs.
-        seed: The random seed for reproducibility.
-        device: The device to run inference on (e.g., 'cuda', 'cpu', 'mps').
-        context_length: The sequence length of the model's input.
-        inference_batch_size: The batch size for inference.
-        sample_from_distribution_columns: A list of columns from which to sample from the distribution.
-        infer_with_dropout: If True, applies dropout during inference.
-        autoregression: If True, performs autoregressive inference.
-        autoregression_total_steps: The number of total steps for autoregressive inference.
-    """
+    """Top-level inference config."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index 5745ead0..bcff237b 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -21,16 +21,7 @@
 def load_preprocessor_config(
     config_path: str, args_config: dict
 ) -> "PreprocessorModel":
-    """
-    Load preprocessor configuration from a YAML file and update it with args_config.
-
-    Args:
-        config_path: Path to the YAML configuration file.
-        args_config: Dictionary containing additional configuration arguments.
-
-    Returns:
-        PreprocessorModel instance with loaded configuration.
-    """
+    """Load preprocessing YAML plus CLI overrides."""
     with open(config_path, "r") as f:
         config_values = yaml.safe_load(f)
 
@@ -42,31 +33,7 @@ def load_preprocessor_config(
 
 
 class PreprocessorModel(BaseModel):
-    """
-    Pydantic model for preprocessor configuration.
-
-    Attributes:
-        project_root: The path to the sequifier project directory.
-        data_path: The path to the input data file.
-        read_format: The file type of the input data. Can be 'csv' or 'parquet'.
-        write_format: The file type for the preprocessed output data.
-        merge_output: If True, combines all preprocessed data into a single file.
-        selected_columns: A list of columns to be included in the preprocessing. If None, all columns are used.
-        split_ratios: A list of floats that define the relative sizes of data splits (e.g., for train, validation, test).
-                           The sum of proportions must be 1.0.
-        stored_context_width: The physical serialized window width.
-        max_target_offset: The number of future items retained after the input window.
-        stride_by_split: A list of step sizes for creating subsequences within each data split.
-        max_rows: The maximum number of input rows to process. If None, all rows are processed.
-        seed: A random seed for reproducibility.
-        n_cores: The number of CPU cores to use for parallel processing. If None, it uses the available CPU cores.
-        batches_per_file: The number of batches to process per file.
-        process_by_file: A flag to indicate if processing should be done file by file.
-        continue_preprocessing: Continue preprocessing job that was interrupted while writing to temp folder.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-        mask_column: Optional input column used to mask all model input columns.
-                              Requires metadata_config_path when set.
-    """
+    """Top-level preprocessing config."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 54efab88..f855a972 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -59,17 +59,7 @@ def _validate_class_share_log_columns(config_values: dict[str, Any]) -> None:
 def load_train_config(
     config_path: str, args_config: dict[str, Any], skip_metadata: bool
 ) -> "TrainModel":
-    """
-    Load training configuration from a YAML file and update it with args_config.
-
-    Args:
-        config_path: Path to the YAML configuration file.
-        args_config: Dictionary containing additional configuration arguments.
-        skip_metadata: Flag indicating whether to process the configuration or not.
-
-    Returns:
-        TrainModel instance with loaded configuration.
-    """
+    """Load train YAML plus CLI overrides and optional metadata-derived fields."""
     with open(config_path, "r") as f:
         config_values = yaml.safe_load(f)
 
@@ -210,42 +200,7 @@ class BERTSpecModel(BaseModel):
 
 
 class TrainingSpecModel(BaseModel):
-    """Pydantic model for training specifications.
-
-    Attributes:
-        device: The torch.device to train the model on (e.g., 'cuda', 'cpu', 'mps').
-        device_max_concat_length: Maximum sequence length for concatenation on device.
-        epochs: The total number of epochs to train for.
-        log_interval: The interval in batches for logging.
-        class_share_log_columns: A list of column names for which to log the class share of predictions.
-        early_stopping_epochs: Number of epochs to wait for validation loss improvement before stopping.
-        save_interval_epochs: The interval in epochs for checkpointing the model.
-        save_latest_interval_minutes: the time interval in which a checkpoint is written to the "latest" checkpoint path
-        save_batch_interval_minutes: the time interval in which a checkpoint is written to a unique checkpoint path
-        save_batch_interval_minutes_val_loss: calculate val loss at the moment of batch interval saving
-        calculate_validation_loss_on_initialization: calculate val loss on weight initialization
-        batch_size: The training batch size.
-        learning_rate: The learning rate.
-        criterion: A dictionary mapping each target column to a loss function.
-        class_weights: A dictionary mapping categorical target columns to a list of class weights.
-        accumulation_steps: The number of gradient accumulation steps.
-        dropout: The dropout value for the transformer model.
-        loss_weights: A dictionary mapping columns to specific loss weights.
-        optimizer: The optimizer configuration.
-        scheduler: The learning rate scheduler configuration.
-        scheduler_step_on: The time of the .step() call on the scheduler, either 'epoch' or 'batch'
-        continue_training: If True, continue training from the latest checkpoint.
-        distributed: If True, enables distributed training.
-        load_full_data_to_ram: If True, loads the entire dataset into RAM.
-        world_size: The number of processes for distributed training.
-        num_workers: The number of worker threads for data loading.
-        backend: The distributed training backend (e.g., 'nccl').
-        layer_type_dtypes: Dictionary mapping layer types (linear, embedding, norm) to dtypes (bfloat16, float8_e4m3fn).
-        layer_autocast: Whether to use autocast
-        sampling_strategy: how to equalize data between GPUs
-        torch_compile: compile entire model ('outer') or transformer layers ('inner') with torch.compile, alternatively 'none'
-        float32_matmul_precision: precision level of float32 computations. One of 'highest', 'high' and 'medium'
-    """
+    """Training loop, optimization, precision, and distribution settings."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
@@ -437,16 +392,7 @@ def validate_data_parallelism(cls, v):
 
 
 class ModelSpecModel(BaseModel):
-    """Pydantic model for model specifications.
-
-    Attributes:
-        initial_embedding_dim: The size of the input embedding. Must be equal to dim_model if joint_embedding_dim is None.
-        feature_embedding_dims: The embedding dimensions for each input column. Must sum to initial_embedding_dim.
-        joint_embedding_dim: Joint embedding layer after initial embedding. Must be equal to dim_model if specified.
-        n_head: The number of heads in the multi-head attention models.
-        dim_feedforward: The dimension of the feedforward network model.
-        num_layers: The number of layers in the transformer model.
-    """
+    """Transformer architecture settings."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
@@ -569,35 +515,7 @@ def validate_n_kv_heads(cls, v, info):
 
 
 class TrainModel(BaseModel):
-    """Pydantic model for training configuration.
-
-    Attributes:
-        project_root: The path to the sequifier project directory.
-        metadata_config_path: The path to the data-driven configuration file.
-        model_name: The name of the model being trained.
-        training_data_path: The path to the training data.
-        validation_data_path: The path to the validation data.
-        read_format: The file format of the input data (e.g., 'csv', 'parquet').
-        input_columns: The list of input columns to be used for training.
-        column_types: A dictionary mapping each column to its numeric type ('int64' or 'float64').
-        categorical_columns: A list of columns that are categorical.
-        real_columns: A list of columns that are real-valued.
-        target_columns: The list of target columns for model training.
-        target_column_types: A dictionary mapping target columns to their types ('categorical' or 'real').
-        id_maps: For each categorical column, a map from distinct values to their indexed representation.
-        context_length: The sequence length of the model's input.
-        n_classes: The number of classes for each categorical column.
-        inference_batch_size: The batch size to be used for inference after model export.
-        seed: The random seed for numpy and PyTorch.
-        model_type: Causal or BERT model type
-        export_generative_model: If True, exports the generative model.
-        export_embedding_model: If True, exports the embedding model.
-        export_onnx: If True, exports the model in ONNX format.
-        export_pt: If True, exports the model using torch.save.
-        export_with_dropout: If True, exports the model with dropout enabled.
-        model_spec: The specification of the transformer model architecture.
-        training_spec: The specification of the training run configuration.
-    """
+    """Top-level training config."""
 
     model_config = ConfigDict(arbitrary_types_allowed=True, extra="forbid")
 
diff --git a/src/sequifier/distributed/env.py b/src/sequifier/distributed/env.py
index 51a853f2..89e24ecd 100644
--- a/src/sequifier/distributed/env.py
+++ b/src/sequifier/distributed/env.py
@@ -10,13 +10,7 @@
 def setup_distributed_env(
     rank: int, local_rank: int, world_size: int, backend: str = "nccl"
 ):
-    """Sets up the distributed training environment.
-
-    Args:
-        rank: The rank of the current process.
-        world_size: The total number of processes.
-        backend: The distributed backend to use.
-    """
+    """Initialize torch.distributed with env defaults."""
     os.environ["MASTER_ADDR"] = os.getenv("MASTER_ADDR", "localhost")
     os.environ["MASTER_PORT"] = os.getenv("MASTER_PORT", "12355")
 
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index a7c4f1e4..4769da24 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -195,36 +195,7 @@ def construct_index_maps(
     target_columns_index_map: list[str],
     map_to_id: Optional[bool],
 ) -> dict[str, dict[int, Union[str, int]]]:
-    """Constructs reverse index maps (int index to original ID).
-
-    This function creates reverse mappings from the integer indices back to
-    the original string or integer identifiers. It only performs this
-    operation if `map_to_id` is True and `id_maps` is provided.
-
-    Special reserved IDs are always decoded as their sentinel labels:
-    0 maps to "[unknown]", 1 maps to "[other]", and 2 maps to "[mask]".
-
-    Args:
-        id_maps: A nested dictionary mapping column names to their
-            respective ID-to-index maps (e.g.,
-            `{'col_name': {'original_id': 1, ...}}`). Expected to be
-            provided if `map_to_id` is True.
-        target_columns_index_map: A list of column names for which to
-            construct the reverse maps.
-        map_to_id: A boolean flag. If True, the reverse maps are
-            constructed. If False or None, an empty dictionary is returned.
-
-    Returns:
-        A dictionary where keys are column names from
-        `target_columns_index_map` and values are the reverse maps
-        (index-to-original-ID). Returns an empty dict if `map_to_id`
-        is not True.
-
-    Raises:
-        AssertionError: If `map_to_id` is True but `id_maps` is None.
-        AssertionError: If the values of a map are not consistently
-            string or integer (excluding the added '0' key).
-    """
+    """Build index-to-ID maps, including reserved token labels."""
     index_map = {}
     if map_to_id is not None and map_to_id:
         if id_maps is None:
@@ -245,21 +216,7 @@ def construct_index_maps(
 def read_data(
     path: str, read_format: str, columns: Optional[list[str]] = None
 ) -> pl.DataFrame:
-    """Reads data from a CSV or Parquet file into a Polars DataFrame.
-
-    Args:
-        path: The file path to read from.
-        read_format: The format of the file. Supported formats are
-            "csv" and "parquet".
-        columns: An optional list of column names to read. This argument
-            is only used when `read_format` is "parquet".
-
-    Returns:
-        A Polars DataFrame containing the data from the file.
-
-    Raises:
-        ValueError: If `read_format` is not "csv" or "parquet".
-    """
+    """Read CSV/Parquet into Polars."""
     if read_format == "csv":
         return pl.read_csv(path, separator=",")
     if read_format == "parquet":
@@ -269,32 +226,7 @@ def read_data(
 
 @beartype
 def write_data(data: pl.DataFrame, path: str, write_format: str, **kwargs) -> None:
-    """Writes a Polars (or Pandas) DataFrame to a CSV or Parquet file.
-
-    This function detects the type of the input DataFrame.
-    - For Polars DataFrames, it uses `.write_csv()` or `.write_parquet()`.
-    - For other DataFrame types (presumably Pandas), it uses `.to_csv()`
-      or `.to_parquet()`.
-
-    Note: The type hint specifies `pl.DataFrame`, but the implementation
-    includes a fallback path that suggests compatibility with Pandas
-    DataFrames.
-
-    Args:
-        data: The Polars (or Pandas) DataFrame to write.
-        path: The destination file path.
-        write_format: The format to write. Supported formats are
-            "csv" and "parquet".
-        **kwargs: Additional keyword arguments passed to the underlying
-            write function (e.g., `write_csv` for Polars, `to_csv` for
-            Pandas).
-
-    Returns:
-        None.
-
-    Raises:
-        ValueError: If `write_format` is not "csv" or "parquet".
-    """
+    """Write Polars/Pandas data as CSV or Parquet."""
     if isinstance(data, pl.DataFrame):
         if write_format == "csv":
             data.write_csv(path, **kwargs)
@@ -318,27 +250,7 @@ def write_data(data: pl.DataFrame, path: str, write_format: str, **kwargs) -> No
 def subset_to_input_columns(
     data: Union[pl.DataFrame, pl.LazyFrame], input_columns: list[str]
 ) -> Union[pl.DataFrame, pl.LazyFrame]:
-    """Filters a DataFrame to rows where 'inputCol' is in a list of column_names.
-
-    This function supports both Polars (DataFrame, LazyFrame) and Pandas
-    DataFrames, dispatching to the appropriate filtering method.
-
-    - For Polars objects, it uses `data.filter(pl.col("inputCol").is_in(...))`.
-    - For other objects (presumably Pandas), it builds a numpy boolean
-      mask and filters using `data.loc[...]`.
-
-    Note: The type hint only specifies Polars objects, but the
-    implementation includes a fallback path for Pandas-like objects.
-
-    Args:
-        data: The Polars (DataFrame, LazyFrame) or Pandas DataFrame to
-            filter. It must contain a column named "inputCol".
-        input_columns: A list of values. Rows will be kept if their
-            value in "inputCol" is present in this list.
-
-    Returns:
-        A filtered DataFrame or LazyFrame of the same type as the input.
-    """
+    """Keep long-format rows whose inputCol is selected."""
     if isinstance(data, (pl.DataFrame, pl.LazyFrame)):
         return data.filter(pl.col("inputCol").is_in(input_columns))
 
@@ -356,40 +268,7 @@ def numpy_to_pytorch(
     all_columns: list[str],
     resolved_view: ResolvedWindowView,
 ) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
-    """Converts a long-format Polars DataFrame to a dict of sequence tensors.
-
-    This function assumes the input DataFrame `data` is in a long format
-    where each row represents a sequence for a specific feature. It expects
-    a column named "inputCol" that contains the feature name (e.g.,
-    'price', 'volume') and other columns representing time steps (e.g.,
-    "0", "1", ..., "L").
-
-    It generates two tensors for each column in `all_columns`:
-    1.  An "input" tensor (from time steps L down to 1).
-    2.  A "target" tensor (from time steps L-1 down to 0).
-
-    Example:
-        For `context_length = 3` and `all_columns = ['price']`, it will create:
-        - 'price': Tensor from columns ["3", "2", "1"]
-        - 'price_target': Tensor from columns ["2", "1", "0"]
-
-    Args:
-        data: The long-format Polars DataFrame. Must contain "inputCol"
-            and columns named as strings of integers for time steps.
-        column_types: A dictionary mapping feature names (from "inputCol")
-            to their desired `torch.dtype`.
-        all_columns: A list of all feature names (from "inputCol") to
-            be processed and converted into tensors.
-        context_length: The total sequence length (L). This determines the
-            column names for time steps (e.g., "0" to "L").
-
-    Returns:
-        A tuple of:
-        - a dictionary mapping feature names to their corresponding PyTorch
-          tensors. Target tensors are stored with a `_target` suffix
-          (e.g., `{'price': <tensor>, 'price_target': <tensor>}`).
-        - a metadata dictionary containing any explicit masks.
-    """
+    """Convert long-format Polars windows to tensors plus masks."""
     input_seq_cols = columns_from_slice(
         resolved_view.input_slice, resolved_view.storage.stored_context_width
     )
@@ -397,11 +276,9 @@ def numpy_to_pytorch(
         resolved_view.target_slice, resolved_view.storage.stored_context_width
     )
 
-    # We will create a unified dictionary
     unified_tensors = {}
 
     for col_name in all_columns:
-        # Create the input sequence tensor (e.g., from t=1 to t=L)
         input_tensor = torch.tensor(
             data.filter(pl.col("inputCol") == col_name)
             .select(input_seq_cols)
@@ -410,8 +287,6 @@ def numpy_to_pytorch(
         )
         unified_tensors[col_name] = input_tensor
 
-        # Create the target sequence tensor (e.g., from t=0 to t=L-1)
-        # We'll store it with a "_target" suffix to distinguish it
         target_tensor = torch.tensor(
             data.filter(pl.col("inputCol") == col_name)
             .select(target_seq_cols)
@@ -432,7 +307,7 @@ def build_valid_mask(
     full_length: int,
     view_slice: slice,
 ) -> Tensor:
-    """Builds a boolean validity mask from explicit left-padding metadata."""
+    """Boolean mask from left-padding metadata."""
 
     full_positions = torch.arange(
         full_length, device=left_pad_lengths.device, dtype=left_pad_lengths.dtype
@@ -454,7 +329,7 @@ def columns_from_slice(view_slice: slice, stored_context_width: int) -> list[str
 
 @beartype
 def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Tensor:
-    """Extracts one leftPadLength value per long-format subsequence."""
+    """One leftPadLength per long-format subsequence."""
     if "leftPadLength" not in data.columns:
         raise ValueError(
             "Dataset layout v1 does not contain explicit padding metadata. "
@@ -474,24 +349,7 @@ def get_left_pad_lengths_from_preprocessed_data(data: pl.DataFrame) -> Tensor:
 
 @beartype
 def normalize_path(path: str, project_root: str) -> str:
-    """Normalizes a path to be relative to a project path, then joins them.
-
-    This function ensures that a given `path` is correctly expressed as
-    an absolute path rooted at `project_root`. It does this by first
-    removing the `project_root` prefix from `path` (if it exists)
-    and then joining the result back to `project_root`.
-
-    This is useful for handling paths that might be provided as either
-    relative (e.g., "data/file.txt") or absolute
-    (e.g., "/abs/path/to/project/data/file.txt").
-
-    Args:
-        path: The path to normalize.
-        project_root: The absolute path to the project's root directory.
-
-    Returns:
-        A normalized, absolute path.
-    """
+    """Return path rooted under project_root."""
     project_root_normalized = (project_root + os.sep).replace(os.sep + os.sep, os.sep)
     path2 = os.path.join(project_root, path.replace(project_root_normalized, ""))
     return path2
@@ -499,18 +357,9 @@ def normalize_path(path: str, project_root: str) -> str:
 
 @beartype
 def configure_logger(project_root: str, model_name: str, rank: Optional[int] = 0):
-    """Configures Loguru to replicate the legacy LogFile behavior.
-
-    Legacy Behavior Mapping:
-    1. Console: Only Rank 0 prints high-level info.
-    2. File 2 (Detailed): Captures ALL logs (equivalent to old level 2).
-    3. File 3 (Summary): Captures only HIGH importance logs (equivalent to old level 3).
-    4. Formatting: Files contain raw messages only (no timestamp prefix).
-    """
-    # Clear default handler
+    """Configure console plus rank-scoped debug/info log files."""
     logger.remove()
 
-    # 1. Console Handler (Rank 0 only, INFO/Level 3 and up)
     if rank == 0 or rank is None:
         logger.add(
             sys.stderr,
@@ -518,15 +367,11 @@ def configure_logger(project_root: str, model_name: str, rank: Optional[int] = 0
             level="INFO",
         )
 
-    # Determine paths
     log_dir = os.path.join(project_root, "logs")
     os.makedirs(log_dir, exist_ok=True)
 
     rank_str = f"rank{rank}" if rank is not None else "rank0"
 
-    # 2. File 2 (Detailed/Debug) - Equivalent to old 'level=2'
-    # Captures everything from DEBUG up.
-    # Format is just {message} to match f.write(f"{string}\n")
     file_2_path = os.path.join(log_dir, f"sequifier-{model_name}-{rank_str}-2.txt")
     logger.add(
         file_2_path,
@@ -536,7 +381,6 @@ def configure_logger(project_root: str, model_name: str, rank: Optional[int] = 0
         mode="a",
     )
 
-    # 3. File 3 (Summary/Info)
     file_3_path = os.path.join(log_dir, f"sequifier-{model_name}-{rank_str}-3.txt")
     logger.add(
         file_3_path,
@@ -551,28 +395,22 @@ def configure_logger(project_root: str, model_name: str, rank: Optional[int] = 0
 @beartype
 def configure_determinism(seed: int, strict: bool = False) -> None:
     """Enforces deterministic execution for reproducibility."""
-    # 1. Set standard seeds
     random.seed(seed)
     np.random.seed(seed)
     torch.manual_seed(seed)
     torch.cuda.manual_seed_all(seed)
 
-    # 2. Ensure deterministic behavior in CUDA/CuDNN
     torch.backends.cudnn.benchmark = False
     torch.backends.cudnn.deterministic = True
     if strict:
-        # 3. Enforce deterministic algorithms in PyTorch (crucial for SDPA/FlashAttention)
-        # This forces PyTorch to error out if a non-deterministic operation is used,
-        # or select the deterministic version of a kernel (e.g. for Flash Attn).
         torch.use_deterministic_algorithms(True, warn_only=True)
 
-        # 4. Set CuBLAS workspace (Required for deterministic algorithms with CUDA >= 10.2)
         os.environ["CUBLAS_WORKSPACE_CONFIG"] = ":4096:8"
 
 
 @beartype
 def get_torch_dtype(dtype_str: str) -> torch.dtype:
-    """Converts a string to a torch dtype, supporting bfloat16 and fp8."""
+    """String-to-torch dtype mapping."""
     dtype_map = {
         "float32": torch.float32,
         "float16": torch.float16,
@@ -597,22 +435,7 @@ def get_torch_dtype(dtype_str: str) -> torch.dtype:
 def get_best_model_path(
     project_root: str, run_name: str, model_type: str
 ) -> tuple[str, int]:
-    """
-    Searches for the exported 'best' model file for a given run and returns its path and epoch.
-
-    Args:
-        project_root: The root directory of the project.
-        run_name: The unique identifier for the hyperparameter search run.
-        model_type: The extension of the exported model (e.g., 'onnx' or 'pt').
-
-    Returns:
-        A tuple containing:
-            - The file path to the best model (str).
-            - The actual epoch at which this model was saved (int).
-
-    Raises:
-        FileNotFoundError: If no matching model files are found.
-    """
+    """Return the highest-epoch exported best model path."""
     search_pattern = os.path.join(
         project_root, "models", f"sequifier-{run_name}-best-*.{model_type}"
     )
@@ -624,7 +447,6 @@ def get_best_model_path(
             f"Could not find an exported 'best' model matching: {search_pattern}"
         )
 
-    # Find the file with the highest epoch number in its name
     best_model_path = max(
         matching_models,
         key=lambda p: int(os.path.splitext(os.path.basename(p))[0].split("-")[-1]),
@@ -636,11 +458,7 @@ def get_best_model_path(
 def get_last_training_batch_timedelta(
     model_name: str, rank: int, project_root: str = "."
 ) -> float:
-    """
-    Reads the level 2 log file, finds the last two mid-epoch training logs,
-    and returns the timedelta between them in seconds.
-    """
-    # Construct the path to the level 2 log file based on configure_logger()
+    """Return seconds between the last two mid-epoch train log entries."""
     log_path = os.path.join(
         project_root, "logs", f"sequifier-{model_name}-rank{rank}-2.txt"
     )
@@ -648,8 +466,6 @@ def get_last_training_batch_timedelta(
     if not os.path.exists(log_path):
         raise FileNotFoundError(f"Log file not found: {log_path}")
 
-    # Regex to capture the timestamp of mid-epoch training batch logs
-    # Matches lines like: "2026-05-26 15:15:39 | INFO | [INFO] Epoch   1 | Batch   10/... | Loss: ..."
     train_log_pattern = re.compile(
         r"^(\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2})\s+\|.*?\[INFO\] Epoch.*?Batch"
     )
@@ -669,7 +485,6 @@ def get_last_training_batch_timedelta(
             "Not enough mid-epoch training logs found in the file to calculate a timedelta."
         )
 
-    # Get the last two chronologically recorded batch timestamps
     t1, t2 = timestamps[-2], timestamps[-1]
 
     return (t2 - t1).total_seconds()
@@ -682,10 +497,7 @@ def apply_bert_masking(
     config: Any,  # TrainConfig
     eval_seed: Optional[int] = None,
 ) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
-    """
-    Applies BERT-style span corruption to the input data using custom distributions.
-    Explicitly passes the boolean prediction mask via the metadata dictionary.
-    """
+    """Apply BERT span corruption and attach prediction masks."""
     data_batch = {k: tensor.clone() for k, tensor in data_batch.items()}
     targets_batch = {k: tensor.clone().detach() for k, tensor in targets_batch.items()}
     metadata_batch = (
@@ -711,14 +523,11 @@ def apply_bert_masking(
             gpu_rng_state = torch.cuda.get_rng_state(device)
         torch.manual_seed(eval_seed)
 
-    # Calculate exact number of tokens to mask per sequence based on valid length
     masking_prob = config.training_spec.bert_spec.masking_probability
     budgets = (valid_mask.sum(dim=1) * masking_prob).long()
 
     bert_mask = torch.zeros((batch_size, seq_len), dtype=torch.bool, device=device)
 
-    # 2. Pre-sample lengths and start positions to avoid massive GPU-CPU sync overhead
-    # We sample more spans than needed to account for overlaps
     max_spans = int(seq_len * masking_prob) + 10
 
     sampled_lengths = config.training_spec.bert_spec.span_masking.sample(
@@ -726,7 +535,6 @@ def apply_bert_masking(
     )
 
     sampled_starts_pct = torch.rand((batch_size, max_spans), device=device)
-    # 3. Span Masking Loop
     for i in range(batch_size):
         budget = budgets[i].item()
         valid_positions = valid_mask[i].nonzero(as_tuple=True)[0]
@@ -742,7 +550,6 @@ def apply_bert_masking(
         current_masked = 0
         span_idx = 0
 
-        # Apply spans until budget is hit
         while current_masked < budget and span_idx < max_spans:
             span_len = sampled_lengths_list[span_idx]
             start_idx = sampled_starts[span_idx]
@@ -770,7 +577,6 @@ def apply_bert_masking(
                 idx = torch.randperm(len(unmasked_valid), device=device)[:remaining]
                 bert_mask[i, unmasked_valid[idx]] = True
 
-    # 4. Create the replacement split masks (e.g., 80% Mask, 10% Random, 10% Identical)
     replacement_probs = torch.rand((batch_size, seq_len), device=device)
 
     p_masked = config.training_spec.bert_spec.replacement_distribution.masked
@@ -783,7 +589,6 @@ def apply_bert_masking(
         & (replacement_probs < (p_masked + p_random))
     )
 
-    # 5. Apply corruption to data_batch
     for col, tensor in data_batch.items():
         if col in config.categorical_columns:
             random_tokens = torch.randint(
@@ -806,7 +611,6 @@ def apply_bert_masking(
             tensor[mask_token_mask] = mask_val
             tensor[random_token_mask] = random_noise[random_token_mask]
 
-    # 6. Append explicit prediction and attention masks to metadata.
     metadata_batch["bert_mask"] = bert_mask
     metadata_batch["attention_valid_mask"] = valid_mask.detach()
 
diff --git a/src/sequifier/hyperparameter_search.py b/src/sequifier/hyperparameter_search.py
index a481d938..b21eea08 100644
--- a/src/sequifier/hyperparameter_search.py
+++ b/src/sequifier/hyperparameter_search.py
@@ -37,35 +37,17 @@ def create_sampler(config: Any) -> optuna.samplers.BaseSampler:
 
 
 def set_pdeathsig():
-    """Binds child process lifecycle to the parent orchestrator via Linux prctl."""
+    """Ask Linux to SIGTERM children when this parent dies."""
     if sys.platform.startswith("linux"):
         libc = ctypes.CDLL("libc.so.6")
         libc.prctl(1, signal.SIGTERM)  # PR_SET_PDEATHSIG = 1
 
 
 def objective(trial: optuna.Trial, config) -> Union[float, tuple[float, ...]]:
-    """The central objective engine bridging Optuna to pure CLI execution.
-
-    This function handles generating the YAML configuration for the specific
-    trial, dynamically allocating a port for distributed training, launching the
-    training subprocess, asynchronously polling the validation metrics, and reporting
-    them back to Optuna for potential pruning.
-
-    Args:
-        trial (optuna.Trial): The Optuna trial object managing the current hyperparameter combination.
-        config (HyperparameterSearchConfig): The parsed hyperparameter search configuration.
-
-    Returns:
-        float: The best validation loss achieved during the trial.
-
-    Raises:
-        optuna.TrialPruned: If the trial is pruned by the Optuna orchestrator.
-        RuntimeError: If the training subprocess fails or is externally preempted.
-    """
+    """Run one Optuna trial through the CLI trainer and metrics JSONL."""
     run_config = config.sample_trial(trial, trial.number)
     run_name = run_config.model_name
 
-    # 1. YAML Generation
     config_path = os.path.join(
         config.project_root, config.model_config_write_path, f"{run_name}.yaml"
     )
@@ -102,7 +84,7 @@ def objective(trial: optuna.Trial, config) -> Union[float, tuple[float, ...]]:
     best_val_loss = float("inf")
 
     def consume_metrics(last_read_pos: int, best_val_loss: float) -> tuple[int, float]:
-        """Helper closure to read written metrics and evaluate pruning."""
+        """Read complete metric lines; report/prune single-objective trials."""
         if os.path.exists(metrics_path):
             with open(metrics_path, "r") as f:
                 f.seek(last_read_pos)
@@ -116,7 +98,6 @@ def consume_metrics(last_read_pos: int, best_val_loss: float) -> tuple[int, floa
                         global_step = data.get("global_step")
 
                         if global_step is not None and val_loss is not None:
-                            # 5. Cooperative Pruning Evaluation
                             is_multi_objective = (
                                 config.evaluation_metrics is not None
                                 and len(config.evaluation_metrics) > 1
@@ -136,11 +117,11 @@ def consume_metrics(last_read_pos: int, best_val_loss: float) -> tuple[int, floa
                                             )
                                             timeout_val = (timedelta * 2) + 30
                                         except (ValueError, FileNotFoundError):
-                                            timeout_val = 60.0  # Safe default fallback
+                                            timeout_val = 60.0
 
                                         process.wait(timeout=timeout_val)
                                     except subprocess.TimeoutExpired:
-                                        process.kill()  # Escalation
+                                        process.kill()
                                     raise optuna.TrialPruned()
 
                         last_read_pos = f.tell()
@@ -149,7 +130,6 @@ def consume_metrics(last_read_pos: int, best_val_loss: float) -> tuple[int, floa
                         break
         return last_read_pos, best_val_loss
 
-    # 4. Asynchronous Polling & Caching Mitigation
     while process.poll() is None:
         last_read_pos, best_val_loss = consume_metrics(last_read_pos, best_val_loss)
         time.sleep(2)
@@ -241,19 +221,7 @@ def consume_metrics(last_read_pos: int, best_val_loss: float) -> tuple[int, floa
 
 @beartype
 def hyperparameter_search(config_path: str, skip_metadata: bool) -> None:
-    """Main function for initiating an Optuna-based hyperparameter search process.
-
-    This function loads the configuration, initializes the Optuna study with a
-    minimization direction, and kicks off the optimization loop. Once the configured
-    number of trials is complete, it prints out the best trial's value and hyperparameters.
-
-    Args:
-        config_path (str): Path to the hyperparameter search YAML configuration file.
-        skip_metadata (bool): Flag indicating whether to skip loading/processing data metadata.
-
-    Raises:
-        ValueError: If `n_trials` is not defined in the configuration.
-    """
+    """Load config, create Optuna study, and optimize trials."""
     config = load_hyperparameter_search_config(config_path, skip_metadata)
 
     os.makedirs(os.path.join(config.project_root, "state", "optuna"), exist_ok=True)
diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index b3ed690f..0baed6ef 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -34,21 +34,7 @@
 
 @beartype
 def infer(args: Any, args_config: dict[str, Any]) -> None:
-    """Runs the main inference pipeline.
-
-    This function orchestrates the inference process. It loads the main
-    inference configuration, retrieves necessary metadata like ID maps and
-    column statistics from a `metadata_config` file (if required for mapping or
-    normalization), and then delegates the core work to the `infer_worker`
-    function.
-
-    Args:
-        args: Command-line arguments, typically from `argparse`. Expected
-            to have attributes like `config_path` and `skip_metadata`.
-        args_config: A dictionary of configuration overrides, often
-            passed from the command line, that will be merged into the
-            loaded configuration file.
-    """
+    """Load inference config and dispatch the worker."""
     logger.info("--- Starting Inference ---")
     config_path = (
         args.config_path if args.config_path is not None else "configs/infer.yaml"
@@ -86,33 +72,13 @@ def infer(args: Any, args_config: dict[str, Any]) -> None:
 
 @beartype
 def load_pt_dataset(data_path: str, start_pct: float, end_pct: float) -> Iterator[Any]:
-    """Lazily loads and yields data from .pt files in a directory.
-
-    This function scans a directory for `.pt` files, sorts them, and then
-    yields the contents of a specific slice of those files defined by a
-    start and end percentage. This allows for processing large datasets
-    in chunks without loading everything into memory.
-
-    Args:
-        data_path: The path to the folder containing the `.pt` files.
-        start_pct: The starting percentage (0.0 to 100.0) of the file list
-            to begin loading from.
-        end_pct: The ending percentage (0.0 to 100.0) of the file list
-            to stop loading at.
-
-    Yields:
-        Iterator: An iterator where each item is the data loaded from a
-        single `.pt` file (e.g., using `torch.load`).
-    """
-    # Get all .pt files in the directory (not nested)
+    """Yield a percentage slice of sorted top-level PT files."""
     pt_files = sorted(Path(data_path).glob("*.pt"))
 
-    # Calculate slice indices
     total = len(pt_files)
     start_idx = int(total * start_pct / 100)
     end_idx = int(total * end_pct / 100)
 
-    # Lazily load and yield data from files in range
     for pt_file in pt_files[start_idx:end_idx]:
         yield torch.load(pt_file, weights_only=False)
 
@@ -121,24 +87,7 @@ def load_pt_dataset(data_path: str, start_pct: float, end_pct: float) -> Iterato
 def load_parquet_folder_dataset(
     data_path: str, start_pct: float, end_pct: float
 ) -> Iterator[Any]:
-    """Lazily loads and yields data from long-format .parquet chunk files in a directory.
-
-    This function scans a directory for `.parquet` files, sorts them, and then
-    yields the contents of a specific slice of those files defined by a
-    start and end percentage. This allows for processing large datasets
-    in chunks without loading everything into memory.
-
-    Args:
-        data_path: The path to the folder containing the `.parquet` files.
-        start_pct: The starting percentage (0.0 to 100.0) of the file list
-            to begin loading from.
-        end_pct: The ending percentage (0.0 to 100.0) of the file list
-            to stop loading at.
-
-    Yields:
-        Iterator: An iterator where each item is a Polars DataFrame loaded from a
-        single `.parquet` file.
-    """
+    """Yield a percentage slice of sorted top-level Parquet files."""
     parquet_files = sorted(Path(data_path).glob("*.parquet"))
 
     total = len(parquet_files)
@@ -157,35 +106,12 @@ def infer_worker(
     selected_columns_statistics: dict[str, dict[str, float]],
     percentage_limits: Optional[tuple[float, float]],
 ):
-    """Core worker function that performs inference.
-
-    This function handles the main workflow:
-    1. Loads the dataset based on `config.read_format` (parquet, csv, or pt).
-    2. Iterates over one or more model paths specified in the config.
-    3. For each model, initializes an `Inferer` object with all necessary
-       configurations, mappings, and statistics.
-    4. Calls the appropriate inference function (`infer_generative` or
-       `infer_embedding`) based on the `config.model_type`.
-    5. Manages the data iterators and passes data chunks to the
-       inference functions.
-
-    Args:
-        config: The fully resolved `InfererModel` configuration object.
-        args_config: A dictionary of command-line arguments, passed to the
-            `Inferer` for potential model loading overrides.
-        id_maps: A nested dictionary mapping categorical column names to
-            their value-to-index maps. `None` if `map_to_id` is False.
-        selected_columns_statistics: A nested dictionary containing 'mean'
-            and 'std' for real-valued columns used for normalization.
-        percentage_limits: A tuple (start_pct, end_pct) used only when
-            `config.read_format == "pt"` to slice the dataset.
-    """
+    """Load data, instantiate models, and run the configured inference mode."""
     logger.info(f"[INFO] Reading data from '{config.data_path}'...")
 
     is_folder_input = os.path.isdir(
         normalize_path(config.data_path, config.project_root)
     )
-    # Step 1: Use Polars for data ingestion
     dataset = None
     if not is_folder_input:
         # Standalone Single-File Path Execution
@@ -275,18 +201,7 @@ def calculate_item_positions(
     prediction_length: int,
     training_objective: str,
 ) -> np.ndarray:
-    """
-    Calculates absolute item positions for inference outputs based on the training objective.
-
-    Args:
-        start_positions: 1D array of base start positions for each sequence in the batch.
-        context_length: The length of the input sequence window.
-        prediction_length: The total number of predicted tokens per sequence.
-        training_objective: Either "causal" or "bert".
-
-    Returns:
-        A 1D array of absolute item positions mapped to every flattened prediction row.
-    """
+    """Return flattened absolute item positions for inference outputs."""
     if training_objective == "bert":
         # Anchor positions to the start of the input sequence and tile forwards
         base_positions = start_positions
@@ -395,31 +310,11 @@ def infer_embedding(
     dataset: Union[list[Any], Iterator[Any]],
     column_types: dict[str, torch.dtype],
 ) -> None:
-    """Performs inference with an embedding model and saves the results.
-
-    This function iterates through the provided dataset (which can be a list
-    of DataFrames or an iterator of tensors). For each data chunk, it
-    calls the appropriate function (`get_embeddings` or `get_embeddings_pt`)
-    to generate embeddings. It then formats these embeddings into a
-    Polars DataFrame, associating them with their `sequenceId`, `subsequenceId`,
-    and absolute `itemPosition`, and writes the resulting DataFrame to the
-    configured output path.
-
-    Args:
-        config: The `InfererModel` configuration object.
-        inferer: The initialized `Inferer` instance.
-        model_id: A string identifier for the model, used for naming
-            output files.
-        dataset: A list containing a Polars DataFrame (for parquet/csv) or
-            an iterator of loaded PyTorch data (for .pt files).
-        column_types: A dictionary mapping column names to their
-            `torch.dtype`.
-    """
+    """Write embeddings for each dataset chunk."""
     for data_id, data in enumerate(dataset):
         prediction_length = inferer.prediction_length
         valid_prediction_mask = None
 
-        # Step 1: Get embeddings and base position/ID data
         is_folder_input = os.path.isdir(
             normalize_path(config.data_path, config.project_root)
         )
@@ -589,31 +484,8 @@ def infer_generative(
     dataset: Union[list[Any], Iterator[Any]],
     column_types: dict[str, torch.dtype],
 ):
-    """Executes the generative inference pipeline and exports results to disk.
-
-    This function processes the input dataset in chunks to accommodate large data
-    volumes. It handles various input formats (standalone CSV/Parquet, folder-based
-    Parquet, or PyTorch tensors) and routes the data to the appropriate inference
-    logic (standard sequence prediction or step-by-step autoregression). After
-    obtaining raw model outputs, it calculates aligned sequence IDs and absolute
-    item positions, applies necessary post-processing (such as reverse-mapping
-    categorical IDs and denormalizing real values), and writes the final
-    probabilities and predictions to the configured output directory.
-
-    Args:
-        config: The inference configuration object dictating I/O paths,
-            autoregression settings, and output formats.
-        inferer: The initialized `Inferer` instance responsible for executing
-            the underlying model logic.
-        model_id: A string identifier for the current model, used to construct
-            the names of the generated output files and directories.
-        dataset: A list or iterator yielding data chunks, typically containing
-            either Polars DataFrames or PyTorch tensor dictionaries.
-        column_types: A dictionary mapping input column names to their expected
-            `torch.dtype`.
-    """
+    """Write generative predictions/probabilities for each dataset chunk."""
     for data_id, data in enumerate(dataset):
-        # Step 1: Adapt Data Subsetting (now works on Polars DF)
         is_folder_input = os.path.isdir(
             normalize_path(config.data_path, config.project_root)
         )
@@ -924,24 +796,7 @@ def get_embeddings_pt(
     data: dict[str, torch.Tensor],
     metadata: dict[str, torch.Tensor],
 ) -> np.ndarray:
-    """Generates embeddings from a batch of PyTorch tensor data.
-
-    This function serves as a wrapper for `Inferer.infer_embedding` when
-    the input data is already in PyTorch tensor format (from loading `.pt`
-    files which contain sequences, targets, sequence_ids, subsequence_ids,
-    and start_positions). It converts the tensor dictionary to a NumPy array
-    dictionary before passing it to the inferer.
-
-    Args:
-        config: The `InfererModel` configuration object (unused, but
-            kept for consistent function signature).
-        inferer: The initialized `Inferer` instance.
-        data: A dictionary mapping column/feature names to `torch.Tensor`s
-              (the sequences part loaded from the .pt file).
-
-    Returns:
-        A NumPy array containing the computed embeddings for the batch.
-    """
+    """Infer embeddings from PT tensors."""
     resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     for tensor in data.values():
         validate_stored_window_width(tensor, config.storage_layout.stored_context_width)
@@ -963,42 +818,10 @@ def get_probs_preds_from_dict(
     metadata: dict[str, torch.Tensor],
     total_steps: int = 1,
 ) -> tuple[Optional[dict[str, np.ndarray]], dict[str, np.ndarray]]:
-    """Generates predictions from PyTorch tensor data, supporting autoregression.
-
-    This function performs generative inference on a batch of PyTorch tensor
-    data loaded from `.pt` files (which contain sequences, targets,
-    sequence_ids, subsequence_ids, and start_positions). It implements an
-    autoregressive loop:
-    1. Runs inference on the initial data `X` (sequences).
-    2. For each subsequent step:
-       a. Creates the next input `X_next` by shifting the previous input
-          `X` and appending the prediction from the last step.
-       b. Runs inference on `X_next`.
-    3. Collects and reshapes all predictions and probabilities from all
-       steps into a single flat batch, ordered by original sample index, then by step.
-
-    Args:
-        config: The `InfererModel` configuration object, used to check
-            `output_probabilities` and `input_columns`.
-        inferer: The initialized `Inferer` instance.
-        data: A dictionary mapping column/feature names to `torch.Tensor`s
-              (the sequences part loaded from the .pt file).
-        total_steps: The number of total autoregressive steps to
-            perform. A value of 1 means simple, non-autoregressive
-            inference.
-
-    Returns:
-        A tuple `(probs, preds)`:
-            - `probs`: A dictionary mapping target columns to NumPy arrays
-              of probabilities, ordered by sample index then step,
-              or `None` if `config.output_probabilities` is False.
-            - `preds`: A dictionary mapping target columns to NumPy arrays
-              of final predictions, ordered by sample index then step.
-    """
+    """Infer PT predictions, flattened sample-major across autoregressive steps."""
 
     target_cols = inferer.target_columns
 
-    # 2. Initialize input and containers for storing results from all steps
     X = {
         key: tensor.numpy()
         for key, tensor in data.items()
@@ -1008,7 +831,6 @@ def get_probs_preds_from_dict(
     all_probs_list = {col: [] for col in target_cols}
     all_preds_list = {col: [] for col in target_cols}
 
-    # 3. Autoregressive loop
     metadata_for_step = metadata_np
     for i in range(total_steps):
         if config.output_probabilities:
@@ -1079,22 +901,7 @@ def get_embeddings(
     data: pl.DataFrame,
     column_types: dict[str, torch.dtype],
 ) -> np.ndarray:
-    """Generates embeddings from a Polars DataFrame.
-
-    This function converts a Polars DataFrame into the NumPy array dictionary
-    format expected by the `Inferer`. It uses `numpy_to_pytorch` for the
-    main conversion, then transforms the tensors to NumPy arrays before
-    passing them to `inferer.infer_embedding`.
-
-    Args:
-        config: The `InfererModel` configuration object.
-        inferer: The initialized `Inferer` instance.
-        data: The input Polars DataFrame chunk.
-        column_types: A dictionary mapping column names to `torch.dtype`.
-
-    Returns:
-        A NumPy array containing the computed embeddings for the batch.
-    """
+    """Infer embeddings from a Polars chunk."""
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
     resolved_view = resolve_window_view(config.storage_layout, config.window_view)
     X, metadata = numpy_to_pytorch(
@@ -1119,28 +926,7 @@ def get_probs_preds_from_df(
     data: pl.DataFrame,
     column_types: dict[str, torch.dtype],
 ) -> tuple[Optional[dict[str, np.ndarray]], dict[str, np.ndarray]]:
-    """Generates predictions from a Polars DataFrame (non-autoregressive).
-
-    This function converts a Polars DataFrame into the NumPy array dictionary
-    format expected by the `Inferer`. It's used for standard,
-    non-autoregressive generative inference.
-    It calls `inferer.infer_generative` once and returns the
-    probabilities (if requested) and predictions.
-
-    Args:
-        config: The `InfererModel` configuration object.
-        inferer: The initialized `Inferer` instance.
-        data: The input Polars DataFrame chunk.
-        column_types: A dictionary mapping column names to `torch.dtype`.
-
-    Returns:
-        A tuple `(probs, preds)`:
-            - `probs`: A dictionary mapping target columns to NumPy arrays
-              of probabilities, or `None` if `config.output_probabilities`
-              is False.
-            - `preds`: A dictionary mapping target columns to NumPy arrays
-              of final predictions.
-    """
+    """Infer non-autoregressive predictions from a Polars chunk."""
     all_columns = sorted(list(set(config.input_columns + config.target_columns)))
 
     resolved_view = resolve_window_view(config.storage_layout, config.window_view)
@@ -1166,46 +952,20 @@ def get_probs_preds_from_df(
 
 @beartype
 def fill_number(number: Union[int, float], max_length: int) -> str:
-    """Pads a number with leading zeros to a specified string length.
-
-    Used for creating sortable string keys (e.g., "001-001", "001-002").
-
-    Args:
-        number: The integer or float to format.
-        max_length: The total desired length of the output string.
-
-    Returns:
-        A string representation of the number, padded with leading zeros.
-    """
+    """Left-pad a number for sortable string keys."""
     number_str = str(number)
     return f"{'0' * (max_length - len(number_str))}{number_str}"
 
 
 @beartype
 def verify_variable_order(data: pl.DataFrame) -> None:
-    """Verifies that the DataFrame is correctly sorted for autoregression.
-
-    Checks two conditions:
-    1. `sequenceId` is globally sorted in ascending order.
-    2. `subsequenceId` is sorted in ascending order *within* each
-       `sequenceId` group.
-
-    Args:
-        data: The Polars DataFrame to check.
-
-    Raises:
-        AssertionError: If `sequenceId` is not globally sorted or if
-            `subsequenceId` is not sorted within `sequenceId` groups.
-    """
-    # Check if the entire 'sequenceId' column is sorted. This is a global property.
+    """Require sequenceId order and in-sequence subsequenceId order."""
     is_globally_sorted = data.select(
         (pl.col("sequenceId").diff().fill_null(0) >= 0).all()
     ).item()
     if not is_globally_sorted:
         raise ValueError("sequenceId must be in ascending order for autoregression")
 
-    # Check if 'subsequenceId' is sorted within each 'sequenceId' group.
-    # This results in a boolean Series, on which we can call .all() directly.
     is_group_sorted = (
         data.select(
             (pl.col("subsequenceId").diff().fill_null(0) >= 0)
@@ -1235,25 +995,7 @@ def get_probs_preds_autoregression(
     np.ndarray,
     np.ndarray,
 ]:
-    """Generates autoregressive predictions and aligns them with sequence IDs and positions.
-
-    Extracts the initial sequence context from the sorted input DataFrame, maps it
-    to PyTorch tensors, and executes step-by-step autoregressive inference.
-
-    Args:
-        config: Inference configuration object.
-        inferer: Initialized `Inferer` instance.
-        data: Input DataFrame, sorted globally by `sequenceId` and locally by `subsequenceId`.
-        column_types: Mapping of input column names to their `torch.dtype`.
-        context_length: Length of the input sequence context.
-
-    Returns:
-        A tuple containing:
-            - probs: Dict of probability arrays per target column (None if disabled).
-            - preds: Dict of final prediction arrays per target column.
-            - sequence_ids_for_preds: 1D array of sequence IDs matching the output shape.
-            - item_positions_for_preds: 1D array of absolute item positions for each step.
-    """
+    """Infer autoregressive predictions with sequence IDs, positions, and mask."""
     verify_variable_order(data)
 
     distinct_cols = len(np.unique(data["inputCol"].to_numpy()))
@@ -1279,7 +1021,6 @@ def get_probs_preds_autoregression(
         resolved_view,
     )
 
-    # Run the autoregressive PyTorch inference
     probs, preds = get_probs_preds_from_dict(
         config,
         inferer,
@@ -1288,7 +1029,6 @@ def get_probs_preds_autoregression(
         config.autoregression_total_steps,
     )
 
-    # 4. Generate the final output arrays using the perfectly aligned bases
     item_positions_for_preds = np.concatenate(
         [
             np.arange(start_pos, start_pos + config.autoregression_total_steps)
@@ -1314,26 +1054,7 @@ def get_probs_preds_autoregression(
 
 
 class Inferer:
-    """A class for performing inference with a trained sequifier model.
-
-    This class encapsulates the model (either ONNX session or PyTorch model),
-    normalization statistics, ID mappings, and all configuration needed
-    to run inference. It provides methods to handle batching, model-specific
-    inference calls (PyTorch vs. ONNX), and post-processing
-    (like inverting normalization).
-
-    Attributes:
-        model_type: 'generative' or 'embedding'.
-        map_to_id: Whether to map integer predictions back to original IDs.
-        selected_columns_statistics: Dict of 'mean' and 'std' for real columns.
-        index_map: The inverse of `id_maps`, for mapping indices back to values.
-        device: The device ('cuda' or 'cpu') for inference.
-        target_columns: List of columns the model predicts.
-        target_column_types: Dict mapping target columns to 'categorical' or 'real'.
-        inference_model_type: 'onnx' or 'pt'.
-        ort_session: `onnxruntime.InferenceSession` if using ONNX.
-        inference_model: The loaded PyTorch model if using 'pt'.
-    """
+    """Inference runtime for PT/ONNX sequifier models."""
 
     @beartype
     def __init__(
@@ -1357,27 +1078,7 @@ def __init__(
         args_config: dict[str, Any],
         training_config_path: str,
     ):
-        """Initializes the Inferer.
-
-        Args:
-            model_type: The type of model to use for inference.
-            model_path: The path to the trained model.
-            project_root: The path to the sequifier project directory.
-            id_maps: A dictionary of id maps for categorical columns.
-            selected_columns_statistics: A dictionary of statistics for numerical columns.
-            map_to_id: Whether to map the output to the original ids.
-            categorical_columns: A list of categorical columns.
-            real_columns: A list of real columns.
-            selected_columns: A list of selected columns.
-            target_columns: A list of target columns.
-            target_column_types: A dictionary of target column types.
-            sample_from_distribution_columns: A list of columns to sample from the distribution.
-            infer_with_dropout: Whether to use dropout during inference.
-            inference_batch_size: The batch size for inference.
-            device: The device to use for inference.
-            args_config: The command-line arguments.
-            training_config_path: The path to the training configuration file.
-        """
+        """Load a PT or ONNX backend and postprocessing state."""
         self.model_type = model_type
         self.map_to_id = map_to_id
         self.selected_columns_statistics = selected_columns_statistics
@@ -1434,19 +1135,7 @@ def __init__(
     def invert_normalization(
         self, values: np.ndarray, target_column: str
     ) -> np.ndarray:
-        """Inverts Z-score normalization for a given target column.
-
-        Uses the 'mean' and 'std' stored in `self.selected_columns_statistics`
-        to transform normalized values back to their original scale.
-
-        Args:
-            values: A NumPy array of normalized values.
-            target_column: The name of the column whose statistics should be
-                used for the inverse transformation.
-
-        Returns:
-            A NumPy array of values in their original scale.
-        """
+        """Invert target-column Z-score normalization."""
         std = self.selected_columns_statistics[target_column]["std"]
         mean = self.selected_columns_statistics[target_column]["mean"]
         return (values * (std - 1e-9)) + mean
@@ -1457,19 +1146,7 @@ def infer_embedding(
         x: dict[str, np.ndarray],
         metadata: dict[str, np.ndarray],
     ) -> np.ndarray:
-        """Performs inference with an embedding model.
-
-        This is a high-level wrapper that calls
-        `adjust_and_infer_embedding` to handle batching and model-specific
-        logic.
-
-        Args:
-            x: A dictionary mapping feature names to NumPy arrays. All arrays
-               must have the same first dimension (batch size).
-
-        Returns:
-            A 2D NumPy array of the resulting embeddings.
-        """
+        """Return embeddings for a feature-array batch."""
         assert x is not None
         size = x[list(x.keys())[0]].shape[0]
         embedding = self.adjust_and_infer_embedding(x, size, metadata)
@@ -1484,36 +1161,7 @@ def infer_generative(
         probs: Optional[dict[str, np.ndarray]] = None,
         return_probs: bool = False,
     ) -> dict[str, np.ndarray]:
-        """Performs generative inference, returning probabilities or predictions.
-
-        This function orchestrates the generative inference process.
-        1. If `probs` are not provided, it calls `adjust_and_infer_generative`
-           to get the raw model output (logits or real values) using `x`.
-        2. If `return_probs` is True:
-           - It normalizes the logits for categorical columns to get
-             probabilities (using `softmax`, implemented in `normalize`).
-           - It returns a dictionary of probabilities (for categorical) and
-             raw predicted values (for real).
-        3. If `return_probs` is False (default):
-           - It converts the model outputs (either from `x` or `probs`) into
-             final predictions.
-           - For categorical columns, it either takes the `argmax` or samples
-             from the distribution (`sample_with_cumsum`).
-           - For real columns, it returns the value as-is.
-
-        Args:
-            x: A dictionary mapping feature names to NumPy arrays. Required
-               if `probs` is not provided.
-            probs: An optional dictionary of probabilities/logits. If provided,
-                   this skips the model inference step.
-            return_probs: If True, returns normalized probabilities for
-                categorical targets. If False, returns final class
-                predictions (via argmax or sampling).
-
-        Returns:
-            A dictionary mapping target column names to NumPy arrays. The
-            content of the arrays depends on `return_probs`.
-        """
+        """Return target probabilities or decoded predictions."""
         if probs is None or (
             x is not None and len(set(x.keys()).difference(set(probs.keys()))) > 0
         ):  # type: ignore
@@ -1572,20 +1220,7 @@ def adjust_and_infer_embedding(
         size: int,
         metadata: dict[str, np.ndarray],
     ):
-        """Handles batching and backend-specific calls for embedding inference.
-
-        This function prepares the input data `x` into batches using
-        `prepare_inference_batches` and then calls the correct inference
-        backend based on `self.inference_model_type` (.pt or .onnx).
-
-        Args:
-            x: The complete dictionary of input features (NumPy arrays).
-            size: The total number of samples in `x`, used to truncate
-                any padding added for batching.
-
-        Returns:
-            A NumPy array of embeddings, concatenated from all batches.
-        """
+        """Batch embedding inference across the active backend."""
         if self.inference_model_type == "onnx":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=True)
@@ -1623,22 +1258,7 @@ def adjust_and_infer_generative(
         size: int,
         metadata: dict[str, np.ndarray],
     ):
-        """Handles batching and backend-specific calls for generative inference.
-
-        This function prepares the input data `x` into batches using
-        `prepare_inference_batches` and then calls the correct inference
-        backend based on `self.inference_model_type` (.pt or .onnx).
-        It aggregates the results from all batches.
-
-        Args:
-            x: The complete dictionary of input features (NumPy arrays).
-            size: The total number of samples in `x`, used to truncate
-                any padding added for batching.
-
-        Returns:
-            A dictionary mapping target column names to NumPy arrays of raw
-            model outputs (logits or real values).
-        """
+        """Batch generative inference across the active backend."""
         if self.inference_model_type == "onnx":
             assert x is not None
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=True)
@@ -1679,23 +1299,7 @@ def adjust_and_infer_generative(
     def prepare_inference_batches(
         self, x: dict[str, np.ndarray], pad_to_batch_size: bool
     ) -> list[dict[str, np.ndarray]]:
-        """Splits input data into batches for inference.
-
-        This function takes a large dictionary of feature arrays and splits
-        them into a list of smaller dictionaries (batches) of size
-        `self.inference_batch_size`.
-
-        Args:
-            x: A dictionary of feature arrays.
-            pad_to_batch_size: If True (for ONNX), the last batch will be
-                padded up to `self.inference_batch_size` by repeating
-                samples. If False (for PyTorch), the last batch may be
-                smaller.
-
-        Returns:
-            A list of dictionaries, where each dictionary is a single batch
-            ready for inference.
-        """
+        """Split feature arrays into backend-sized batches."""
         size = x[list(x.keys())[0]].shape[0]
         if size == self.inference_batch_size:
             return [x]
@@ -1726,20 +1330,7 @@ def infer_pure(
         x: dict[str, np.ndarray],
         metadata: dict[str, np.ndarray],
     ) -> list[np.ndarray]:
-        """Performs a single inference pass using the ONNX session.
-
-        This function assumes `x` is already a single, correctly-sized
-        batch. It formats the input dictionary to match the ONNX model's
-        input names and executes `self.ort_session.run()`.
-
-        Args:
-            x: A dictionary of feature arrays for a single batch. This
-               batch *must* be of size `self.inference_batch_size`.
-
-        Returns:
-            A list of NumPy arrays, representing the raw outputs from the
-            ONNX model.
-        """
+        """Run one ONNX batch and flatten sequence-major outputs."""
         metadata = metadata or {}
         ort_inputs = {}
         for session_input in self.ort_session.get_inputs():
@@ -1766,18 +1357,7 @@ def infer_pure(
 
     @beartype
     def expand_to_batch_size(self, x: np.ndarray) -> np.ndarray:
-        """Pads a NumPy array to match `self.inference_batch_size`.
-
-        Repeats samples from `x` until the array's first dimension
-        is equal to `self.inference_batch_size`.
-
-        Args:
-            x: The input NumPy array to pad.
-
-        Returns:
-            A new NumPy array of size `self.inference_batch_size` in the
-            first dimension.
-        """
+        """Repeat leading samples until the ONNX batch size is met."""
         repetitions = self.inference_batch_size // x.shape[0]
         filler = self.inference_batch_size % x.shape[0]
         return np.concatenate(([x] * repetitions) + [x[0:filler, :]], axis=0)
@@ -1785,19 +1365,7 @@ def expand_to_batch_size(self, x: np.ndarray) -> np.ndarray:
 
 @beartype
 def normalize(outs: dict[str, np.ndarray]) -> dict[str, np.ndarray]:
-    """Applies the softmax function to a dictionary of logits.
-
-    Converts raw model logits for categorical columns into probabilities
-    that sum to 1.
-
-    Args:
-        outs: A dictionary mapping target column names to NumPy arrays
-              of logits.
-
-    Returns:
-        A dictionary mapping the same target column names to NumPy arrays
-        of probabilities.
-    """
+    """Softmax logits by target column."""
     normalizer = {
         target_column: np.repeat(
             np.sum(np.exp(target_values), axis=1), target_values.shape[1]
@@ -1813,21 +1381,7 @@ def normalize(outs: dict[str, np.ndarray]) -> dict[str, np.ndarray]:
 
 @beartype
 def sample_with_cumsum(probs: np.ndarray, is_log_probs: bool = True) -> np.ndarray:
-    """Samples from a probability distribution using the inverse CDF method.
-
-    Takes an array of logits, computes the cumulative probability
-    distribution, draws a random number `r` from [0, 1), and returns
-    the index of the first class `i` where `cumsum[i] > r`.
-
-    Args:
-        probs: A 2D NumPy array of logits or normalized probabilities.
-               Shape is (batch_size, num_classes).
-        is_log_probs: Boolean flag indicating if the passed array are logits or
-               probabilities
-    Returns:
-        A 1D NumPy array of shape (batch_size,) containing the sampled
-        class indices.
-    """
+    """Sample class indices from logits or probabilities."""
     if is_log_probs:
         cumulative_probs = np.cumsum(np.exp(probs), axis=1)
     else:
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index 6dc03b16..da4bc2b0 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -17,14 +17,7 @@
 
 
 class SequifierDatasetFromFile(IterableDataset):
-    """
-    An iterable-style dataset that pre-loads all data into CPU RAM and yields
-    pre-collated batches.
-
-    This is the idiomatic PyTorch solution for implementing custom 'en block'
-    batching. The __iter__ method handles shuffling and batch slicing, ensuring
-    maximum performance.
-    """
+    """Eager single-file dataset yielding pre-collated batches."""
 
     def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         super().__init__()
@@ -33,7 +26,6 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.shuffle = shuffle
         self.epoch = 0
 
-        # Create a unified list of all columns the model might need
         all_columns = sorted(list(set(config.input_columns + config.target_columns)))
 
         logger.info(
@@ -46,7 +38,6 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
             for col in config.column_types
         }
 
-        # self.all_tensors now holds both inputs and targets
         resolved_view = resolve_window_view(config.storage_layout, config.window_view)
         all_tensors, metadata_tensors = numpy_to_pytorch(
             data=data_df,
@@ -78,24 +69,15 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         logger.info(f"[INFO] Dataset loaded with {self.n_samples} samples.")
 
     def set_epoch(self, epoch: int):
-        """Allows the training loop to set the epoch for deterministic shuffling."""
+        """Set the shuffle epoch."""
         self.epoch = epoch
 
     def __len__(self) -> int:
-        """Returns the total number of samples in the dataset."""
         return math.ceil(self.n_samples / self.batch_size)
 
     def __iter__(
         self,
     ) -> Iterator[SequifierBatch]:
-        """Yields batches of data.
-
-        Handles shuffling (if enabled) and slicing data based on distributed
-        rank and worker ID.
-
-        Yields:
-            An iterator where each item is a SequifierBatch.
-        """
         worker_info = torch.utils.data.get_worker_info()
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 17414dcf..93035822 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -22,13 +22,7 @@
 
 
 class SequifierDatasetFromFolderParquet(IterableDataset):
-    """
-    An efficient PyTorch IterableDataset that pre-loads a folder of chunked
-    Parquet files entirely into CPU RAM at initialization.
-
-    Yields full, pre-collated batches natively. Fully supports DDP/FSDP distributed
-    environments using customizable sampling strategies.
-    """
+    """Eager Parquet-folder dataset yielding rank/worker-aligned batches."""
 
     def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         super().__init__()
@@ -147,11 +141,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
         return total_batches
 
     def set_epoch(self, epoch: int):
-        """Allows the training loop to synchronize seed steps for shuffling."""
+        """Set the shuffle epoch."""
         self.epoch = epoch
 
     def _get_target_samples(self) -> int:
-        """Calculates precise sample counts per rank to manage FSDP layer allocations."""
+        """Return per-rank sample count under the configured sampling strategy."""
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
 
@@ -180,18 +174,15 @@ def __iter__(
         worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
 
-        # 1. Coordinate global shuffling masks
         indices = torch.arange(self.n_samples)
         if self.shuffle:
             g = torch.Generator()
             g.manual_seed(self.config.seed + self.epoch)
             indices = indices[torch.randperm(self.n_samples, generator=g)]
 
-        # 2. Slice metrics based on GPU distribution metrics
         indices_for_rank = indices[rank::world_size].tolist()
         sample_is_real = [True] * len(indices_for_rank)
 
-        # 3. Synchronize cross-device oversampling/undersampling rules
         if self.sampling_strategy == "oversampling":
             real_count = len(indices_for_rank)
             if real_count == 0:
@@ -211,11 +202,9 @@ def __iter__(
             indices_for_rank = indices_for_rank[: self.target_samples]
             sample_is_real = sample_is_real[: self.target_samples]
 
-        # 4. Map worker task splits
         indices_for_worker = indices_for_rank[worker_id::num_workers]
         sample_is_real_for_worker = sample_is_real[worker_id::num_workers]
 
-        # 5. Extract and pass unified data frames
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
             batch_sample_is_real = sample_is_real_for_worker[i : i + self.batch_size]
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index 2d03ffc4..b45ed0cc 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -23,29 +23,7 @@
 
 
 class SequifierDatasetFromFolderParquetLazy(IterableDataset):
-    """
-    An efficient, memory-safe PyTorch IterableDataset for out-of-core training.
-
-    Streams long-format Parquet files sequentially using cross-file buffering to yield
-    exact batches, eliminating CPU cloning bottlenecks. Fully supports DDP/FSDP by
-    precisely calculating and distributing sample boundaries across GPU ranks and workers.
-
-    Args:
-        data_path (str): Path to the directory containing `.parquet` chunks and `metadata.json`.
-        config (TrainModel): Training configuration (batch size, workers, sequence length, etc.).
-        shuffle (bool, optional): If True, deterministically shuffles file order and
-            sample indices per epoch. Defaults to True.
-
-    Yields:
-        SequifierBatch:
-            A named batch containing sequence dictionaries, target dictionaries,
-            metadata dictionaries, and optional identifiers.
-
-    Raises:
-        FileNotFoundError: If `metadata.json` is missing.
-        Exception: If sample counts are uneven across ranks using the 'exact' sampling
-            strategy, or if a GPU rank is assigned no files.
-    """
+    """Streams long-format Parquet chunks into rank/worker-aligned batches."""
 
     def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         super().__init__()
@@ -96,11 +74,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
         return total_batches
 
     def set_epoch(self, epoch: int):
-        """Allows the training loop to set the epoch for deterministic file shuffling."""
+        """Set the shuffle epoch."""
         self.epoch = epoch
 
     def _get_target_samples(self) -> int:
-        """Calculates exact sample count per rank to ensure FSDP syncs properly."""
+        """Return per-rank sample count under the configured sampling strategy."""
         world_size = dist.get_world_size() if dist.is_initialized() else 1
 
         num_files = len(self.batch_files_info)
@@ -169,7 +147,6 @@ def __iter__(
         worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
 
-        # 1. Distribute files among ranks
         num_files = len(self.batch_files_info)
         original_files_for_this_rank = list(range(rank, num_files, world_size))
         rank_real_samples = sum(
@@ -183,11 +160,9 @@ def __iter__(
             else:
                 raise Exception(f"No file found for GPU rank {rank}.")
 
-        # 2. Assign exact sample quotas and boundaries to this specific worker thread
         base_samples_per_worker = self.target_samples // num_workers
         remainder = self.target_samples % num_workers
 
-        # Calculate exactly where this worker's data starts and ends in the global stream
         worker_start_sample = 0
         for i in range(worker_id):
             worker_start_sample += base_samples_per_worker + (1 if i < remainder else 0)
@@ -197,7 +172,6 @@ def __iter__(
         )
         worker_end_sample = worker_start_sample + worker_target_samples
 
-        # 3. Shuffle files deterministically
         g = torch.Generator()
         g.manual_seed(self.config.seed + self.epoch)
 
@@ -207,7 +181,6 @@ def __iter__(
         else:
             ordered_files = files_for_this_rank.copy()
 
-        # 4. Extend files based on exact target requirements
         extended_files = []
         current_samples = 0
         file_idx = 0
@@ -217,7 +190,6 @@ def __iter__(
             current_samples += self.batch_files_info[f_id]["samples"]
             file_idx += 1
 
-        # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
         global_file_start_sample = 0
 
@@ -228,7 +200,6 @@ def __iter__(
             self.resolved_view.target_slice, self.folder_layout.stored_context_width
         )
 
-        # Initialize cross-file buffers
         seq_buffer: Dict[str, torch.Tensor] = {}
         tgt_buffer: Dict[str, torch.Tensor] = {}
         meta_buffer: Dict[str, torch.Tensor] = {}
@@ -243,23 +214,19 @@ def __iter__(
             file_end = global_file_start_sample + file_samples
             global_file_start_sample += file_samples
 
-            # Skip this file if it belongs entirely to other workers
             if file_end <= worker_start_sample or file_start >= worker_end_sample:
                 continue
 
-            # This file overlaps with our worker's assigned boundary. Load it.
             file_path = os.path.join(self.data_dir, self.batch_files_info[f_id]["path"])
             df = pl.read_parquet(file_path)
             left_pad_lengths = get_left_pad_lengths_from_preprocessed_data(df)
 
-            # Generate indices for the whole file using torch (matching pt_lazy)
             indices = torch.arange(file_samples)
             if self.shuffle:
                 g_file = torch.Generator()
                 g_file.manual_seed(self.config.seed + self.epoch + f_id + rank)
                 indices = indices[torch.randperm(file_samples, generator=g_file)]
 
-            # Slice the indices to extract ONLY the portion belonging to this worker
             worker_file_start_idx = max(0, worker_start_sample - file_start)
             worker_file_end_idx = min(file_samples, worker_end_sample - file_start)
 
@@ -279,19 +246,13 @@ def __iter__(
 
             worker_indices_np = worker_indices.numpy()
 
-            # 1. Single-pass partition: Groups data natively in C++/Rust, eliminating O(F * N) scans
             feature_partitions = {
                 frame.item(0, "inputCol"): frame
                 for frame in df.partition_by("inputCol")
             }
 
-            # Dynamic feature names fallback to gracefully handle MagicMock objects in unit tests
-
-            # Process Long format data structures into PyTorch Tensors
             new_seq, new_tgt = {}, {}
 
-            # 2. Iterate over the expected config columns, not the dynamically found ones
-            # Process inputs
             for col_name in self.config.input_columns:
                 if col_name in feature_partitions:
                     feature_chunk = feature_partitions[col_name][worker_indices_np]
@@ -302,7 +263,6 @@ def __iter__(
                 else:
                     raise ValueError(f"Column not found in input data: {col_name}")
 
-            # Process targets
             for col_name in self.config.target_columns:
                 if col_name in feature_partitions:
                     feature_chunk = feature_partitions[col_name][worker_indices_np]
@@ -320,7 +280,6 @@ def __iter__(
 
             del df
 
-            # Append the new slice to the cross-file buffer
             if buffer_len == 0:
                 seq_buffer = new_seq
                 tgt_buffer = new_tgt
@@ -343,12 +302,10 @@ def __iter__(
 
             buffer_len += num_new_samples
 
-            # Yield batches as long as the buffer contains at least `batch_size` samples
             while buffer_len >= self.batch_size:
                 if yielded_samples >= worker_target_samples:
                     break
 
-                # Slice out a perfect batch from the top of the buffer
                 batch_seq = {k: v[: self.batch_size] for k, v in seq_buffer.items()}
                 batch_tgt = {k: v[: self.batch_size] for k, v in tgt_buffer.items()}
                 batch_meta = {k: v[: self.batch_size] for k, v in meta_buffer.items()}
@@ -360,13 +317,11 @@ def __iter__(
                 )
                 yielded_samples += self.batch_size
 
-                # Keep the remainder in the buffer for the next loop/file
                 seq_buffer = {k: v[self.batch_size :] for k, v in seq_buffer.items()}
                 tgt_buffer = {k: v[self.batch_size :] for k, v in tgt_buffer.items()}
                 meta_buffer = {k: v[self.batch_size :] for k, v in meta_buffer.items()}
                 buffer_len -= self.batch_size
 
-        # 6. Yield the final partial batch from the buffer if any remains
         if buffer_len > 0 and yielded_samples < worker_target_samples:
             remaining_needed = worker_target_samples - yielded_samples
             final_yield_size = min(buffer_len, remaining_needed)
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index aff09a89..7327fe93 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -19,13 +19,7 @@
 
 
 class SequifierDatasetFromFolderPt(IterableDataset):
-    """
-    An efficient PyTorch IterableDataset that pre-loads all data into RAM.
-
-    This strategy pays a one-time I/O cost at initialization, after which
-    all data access during training is extremely fast. It yields full,
-    pre-collated batches natively.
-    """
+    """Eager PT-folder dataset yielding rank/worker-aligned batches."""
 
     def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         super().__init__()
@@ -60,7 +54,6 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         }
         all_left_pad_lengths: list[torch.Tensor] = []
 
-        # Load all data files into RAM
         for file_info in metadata["batch_files"]:
             file_path = os.path.join(self.data_dir, file_info["path"])
             (
@@ -107,11 +100,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
         return total_batches
 
     def set_epoch(self, epoch: int):
-        """Allows the training loop to set the epoch for deterministic shuffling."""
+        """Set the shuffle epoch."""
         self.epoch = epoch
 
     def _get_target_samples(self) -> int:
-        """Calculates exact sample count per rank to ensure FSDP syncs properly."""
+        """Return per-rank sample count under the configured sampling strategy."""
         world_size = dist.get_world_size() if dist.is_initialized() else 1
         rank = dist.get_rank() if dist.is_initialized() else 0
 
@@ -140,18 +133,15 @@ def __iter__(
         worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
 
-        # 1. Global Shuffling
         indices = torch.arange(self.n_samples)
         if self.shuffle:
             g = torch.Generator()
             g.manual_seed(self.config.seed + self.epoch)
             indices = indices[torch.randperm(self.n_samples, generator=g)]
 
-        # 2. Distribute across GPU ranks
         indices_for_rank = indices[rank::world_size].tolist()
         sample_is_real = [True] * len(indices_for_rank)
 
-        # 3. Handle FSDP oversampling/undersampling sync requirements
         if self.sampling_strategy == "oversampling":
             real_count = len(indices_for_rank)
             if real_count == 0:
@@ -171,11 +161,9 @@ def __iter__(
             indices_for_rank = indices_for_rank[: self.target_samples]
             sample_is_real = sample_is_real[: self.target_samples]
 
-        # 4. Distribute among CPU workers for this GPU
         indices_for_worker = indices_for_rank[worker_id::num_workers]
         sample_is_real_for_worker = sample_is_real[worker_id::num_workers]
 
-        # 5. Yield full batches
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
             batch_sample_is_real = sample_is_real_for_worker[i : i + self.batch_size]
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 8193b793..506d12b7 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -20,29 +20,7 @@
 
 
 class SequifierDatasetFromFolderPtLazy(IterableDataset):
-    """
-    An efficient, memory-safe PyTorch IterableDataset for out-of-core training.
-
-    Streams pre-processed chunked files sequentially using cross-file buffering to yield
-    exact batches, eliminating CPU cloning bottlenecks. Fully supports DDP/FSDP by
-    precisely calculating and distributing sample boundaries across GPU ranks and workers.
-
-    Args:
-        data_path (str): Path to the directory containing `.pt` chunks and `metadata.json`.
-        config (TrainModel): Training configuration (batch size, workers, sequence length, etc.).
-        shuffle (bool, optional): If True, deterministically shuffles file order and
-            sample indices per epoch. Defaults to True.
-
-    Yields:
-        SequifierBatch:
-            A named batch containing sequence dictionaries, target dictionaries,
-            metadata dictionaries, and optional identifiers.
-
-    Raises:
-        FileNotFoundError: If `metadata.json` is missing.
-        Exception: If sample counts are uneven across ranks using the 'exact' sampling
-            strategy, or if a GPU rank is assigned no files.
-    """
+    """Streams PT chunks into rank/worker-aligned batches."""
 
     def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         super().__init__()
@@ -88,11 +66,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
         return total_batches
 
     def set_epoch(self, epoch: int):
-        """Allows the training loop to set the epoch for deterministic file shuffling."""
+        """Set the shuffle epoch."""
         self.epoch = epoch
 
     def _get_target_samples(self) -> int:
-        """Calculates exact sample count per rank to ensure FSDP syncs properly."""
+        """Return per-rank sample count under the configured sampling strategy."""
         world_size = dist.get_world_size() if dist.is_initialized() else 1
 
         num_files = len(self.batch_files_info)
@@ -161,7 +139,6 @@ def __iter__(
         worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
 
-        # 1. Distribute files among ranks
         num_files = len(self.batch_files_info)
         original_files_for_this_rank = list(range(rank, num_files, world_size))
         rank_real_samples = sum(
@@ -175,11 +152,9 @@ def __iter__(
             else:
                 raise Exception(f"No file found for GPU rank {rank}.")
 
-        # 2. Assign exact sample quotas and boundaries to this specific worker thread
         base_samples_per_worker = self.target_samples // num_workers
         remainder = self.target_samples % num_workers
 
-        # Calculate exactly where this worker's data starts and ends in the global stream
         worker_start_sample = 0
         for i in range(worker_id):
             worker_start_sample += base_samples_per_worker + (1 if i < remainder else 0)
@@ -189,7 +164,6 @@ def __iter__(
         )
         worker_end_sample = worker_start_sample + worker_target_samples
 
-        # 3. Shuffle files deterministically
         g = torch.Generator()
         g.manual_seed(self.config.seed + self.epoch)
 
@@ -199,7 +173,6 @@ def __iter__(
         else:
             ordered_files = files_for_this_rank.copy()
 
-        # 4. Extend files based on exact target requirements
         extended_files = []
         current_samples = 0
         file_idx = 0
@@ -209,11 +182,9 @@ def __iter__(
             current_samples += self.batch_files_info[f_id]["samples"]
             file_idx += 1
 
-        # 5. Stream data using precise global boundaries and a CROSS-FILE BUFFER
         yielded_samples = 0
         global_file_start_sample = 0
 
-        # Initialize cross-file buffers
         seq_buffer: Dict[str, torch.Tensor] = {}
         tgt_buffer: Dict[str, torch.Tensor] = {}
         meta_buffer: Dict[str, torch.Tensor] = {}
@@ -228,11 +199,9 @@ def __iter__(
             file_end = global_file_start_sample + file_samples
             global_file_start_sample += file_samples
 
-            # Skip this file if it belongs entirely to other workers
             if file_end <= worker_start_sample or file_start >= worker_end_sample:
                 continue
 
-            # This file overlaps with our worker's assigned boundary. Load it.
             file_path = os.path.join(self.data_dir, self.batch_files_info[f_id]["path"])
             (
                 sequences_batch,
@@ -246,14 +215,12 @@ def __iter__(
                     tensor, self.folder_layout.stored_context_width
                 )
 
-            # Generate indices for the whole file
             indices = torch.arange(file_samples)
             if self.shuffle:
                 g_file = torch.Generator()
                 g_file.manual_seed(self.config.seed + self.epoch + f_id + rank)
                 indices = indices[torch.randperm(file_samples, generator=g_file)]
 
-            # Slice the indices to extract ONLY the portion belonging to this worker
             worker_file_start_idx = max(0, worker_start_sample - file_start)
             worker_file_end_idx = min(file_samples, worker_end_sample - file_start)
 
@@ -270,7 +237,6 @@ def __iter__(
                 del sequences_batch
                 continue
 
-            # Extract the data subset for this worker (Advanced indexing copies the data)
             new_seq = {
                 k: v[worker_indices, self.resolved_view.input_slice]
                 for k, v in sequences_batch.items()
@@ -287,10 +253,8 @@ def __iter__(
             )
             new_meta["sample_valid_mask"] = sample_is_real
 
-            # Free the large file immediately to keep RAM down
             del sequences_batch, left_pad_lengths_batch
 
-            # Append the new slice to the cross-file buffer
             if buffer_len == 0:
                 seq_buffer = new_seq
                 tgt_buffer = new_tgt
@@ -313,12 +277,10 @@ def __iter__(
 
             buffer_len += num_new_samples
 
-            # Yield batches as long as the buffer contains at least `batch_size` samples
             while buffer_len >= self.batch_size:
                 if yielded_samples >= worker_target_samples:
                     break
 
-                # Slice out a perfect batch from the top of the buffer
                 batch_seq = {k: v[: self.batch_size] for k, v in seq_buffer.items()}
                 batch_tgt = {k: v[: self.batch_size] for k, v in tgt_buffer.items()}
                 batch_meta = {k: v[: self.batch_size] for k, v in meta_buffer.items()}
@@ -330,13 +292,11 @@ def __iter__(
                 )
                 yielded_samples += self.batch_size
 
-                # Keep the remainder in the buffer for the next loop/file
                 seq_buffer = {k: v[self.batch_size :] for k, v in seq_buffer.items()}
                 tgt_buffer = {k: v[self.batch_size :] for k, v in tgt_buffer.items()}
                 meta_buffer = {k: v[self.batch_size :] for k, v in meta_buffer.items()}
                 buffer_len -= self.batch_size
 
-        # 6. Yield the final partial batch from the buffer if any remains
         if buffer_len > 0 and yielded_samples < worker_target_samples:
             remaining_needed = worker_target_samples - yielded_samples
             final_yield_size = min(buffer_len, remaining_needed)
diff --git a/src/sequifier/io/yaml.py b/src/sequifier/io/yaml.py
index 934f40bf..1b8aa1c2 100644
--- a/src/sequifier/io/yaml.py
+++ b/src/sequifier/io/yaml.py
@@ -12,67 +12,30 @@
 
 
 def represent_sequifier_object(dumper, data):
-    """
-    Represents objects from 'sequifier.config.train_config' (like TrainModel,
-    ModelSpecModel, TrainingSpecModel) as a simple YAML mapping,
-    using the object's __dict__. This effectively removes the
-    !!python/object tag and the explicit '__dict__:', '__fields_set__:' keys.
-    """
-    # We assume the object's __dict__ contains the attributes to be serialized.
-    # If these objects are Pydantic models, using data.model_dump(mode='python')
-    # would be more robust if available.
+    """Represent sequifier config objects as plain YAML mappings."""
     return dumper.represent_dict(data.__dict__)
 
 
 def represent_dot_dict(dumper, data):
-    """
-    Represents DotDict objects as a simple YAML mapping.
-    The original output showed a 'dictitems' attribute. If your DotDict
-    is essentially a dictionary, this will work.
-    """
-    # If DotDict has a specific attribute like 'dictitems' that holds the actual dict:
-    # return dumper.represent_dict(data.dictitems)
-    # If DotDict is a subclass of dict or dict-like:
+    """Represent DotDict as a plain YAML mapping."""
     return dumper.represent_dict(dict(data))
 
 
 def represent_numpy_float(dumper, data):
-    """
-    Represents numpy.float64 (and similar numpy floats) as standard YAML floats.
-    """
+    """Represent NumPy floats as YAML floats."""
     return dumper.represent_float(float(data))
 
 
 def represent_numpy_int(dumper, data):
-    """
-    Represents numpy.int64 (and similar numpy integers) as standard YAML integers.
-    """
+    """Represent NumPy integers as YAML integers."""
     return dumper.represent_int(int(data))
 
 
 class TrainModelDumper(yaml.Dumper):
-    """A custom YAML dumper for TrainModel objects.
+    """YAML dumper for sequifier config objects."""
 
-    This dumper extends the base yaml.Dumper to provide custom serialization
-    for TrainModel and related objects, ensuring a clean and readable YAML
-    output. It also modifies the indentation behavior for better formatting.
-    """
-
-    # You can add more customizations here if needed, like indent width.
     def increase_indent(self, flow=False, indentless=False):
-        """Increase the indentation level for the YAML output.
-
-        This method overrides the default behavior to force indentation for all
-        block-style collections, improving the readability of the output YAML.
-
-        Args:
-            flow: Whether the context is a flow-style collection.
-            indentless: Whether the context is an indentless sequence.
-
-        Returns:
-            The result of the parent class's increase_indent method, with flow
-            forced to False.
-        """
+        """Indent block sequences."""
         return super(TrainModelDumper, self).increase_indent(flow, False)
 
 
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index c5bab86f..51d43a5b 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -100,11 +100,7 @@
 
 
 def make(args):
-    """Creates a new sequifier project.
-
-    Args:
-        args: The command-line arguments.
-    """
+    """Create a sequifier project scaffold."""
     project_name = args.project_name
 
     if not (project_name and len(project_name) > 0):
diff --git a/src/sequifier/model/layers.py b/src/sequifier/model/layers.py
index 4e485d4c..ddd2c839 100644
--- a/src/sequifier/model/layers.py
+++ b/src/sequifier/model/layers.py
@@ -10,17 +10,10 @@ def __init__(self, dim: int, eps: float = 1e-6):
         self.weight = nn.Parameter(torch.ones(dim))
 
     def forward(self, x):
-        # 1. Cast input to float32 once for stability
         x_fp32 = x.to(torch.float32)
-
-        # 2. Calculate variance
         var = torch.mean(x_fp32.pow(2), dim=-1, keepdim=True)
-
-        # 3. Normalize
         x_normed = x_fp32 * torch.rsqrt(var + self.eps)
 
-        # 4. Cast back to the *input tensor's* dtype (traceable),
-        #    rather than self.weight.dtype (not traceable in Cast ops)
         return (self.weight.to(x_normed.dtype) * x_normed).to(x.dtype)
 
 
diff --git a/src/sequifier/optimizers/ademamix.py b/src/sequifier/optimizers/ademamix.py
index 9eeced8e..a69ad875 100644
--- a/src/sequifier/optimizers/ademamix.py
+++ b/src/sequifier/optimizers/ademamix.py
@@ -9,26 +9,7 @@
 
 
 class AdEMAMix(Optimizer):
-    """Implements the AdEMAMix optimizer.
-
-    This optimizer is based on the paper "AdEMAMix: A Novel Adaptive Optimizer for
-    Deep Learning". It combines the advantages of Adam and EMA, and introduces a
-    mixing term to further improve performance.
-
-    Args:
-        params (iterable): Iterable of parameters to optimize or dicts defining
-            parameter groups.
-        learning_rate (float, optional): Learning rate (default: 1e-3).
-        betas (Tuple[float, float, float], optional): Coefficients used for
-            computing running averages of gradient and its square
-            (default: (0.9, 0.999, 0.9999)).
-        eps (float, optional): Term added to the denominator to improve
-            numerical stability (default: 1e-8).
-        weight_decay (float, optional): Weight decay (L2 penalty) (default: 0).
-        alpha (float, optional): Mixing coefficient (default: 5.0).
-        T_alpha_beta3 (int, optional): Time period for alpha and beta3 scheduling
-            (default: None).
-    """
+    """AdEMAMix optimizer."""
 
     def __init__(
         self,
@@ -62,24 +43,11 @@ def __init__(
         super(AdEMAMix, self).__init__(params, defaults)
 
     def __setstate__(self, state):
-        """Set the state of the optimizer.
-
-        Args:
-            state (dict): The state of the optimizer.
-        """
         super(AdEMAMix, self).__setstate__(state)
 
     @torch.no_grad()
     def step(self, closure=None):
-        """Perform a single optimization step.
-
-        Args:
-            closure (callable, optional): A closure that reevaluates the model
-                and returns the loss. (default: None)
-
-        Returns:
-            The loss, if the closure is provided. Otherwise, returns None.
-        """
+        """Run one optimizer step."""
         loss = None
         if closure is not None:
             with torch.enable_grad():
@@ -163,27 +131,7 @@ def _update_adamemix(
         weight_decay,
         eps,
     ):
-        """Perform the AdEMAMix update for a single parameter group.
-
-        Args:
-            params (list[torch.Tensor]): List of parameters to update.
-            grads (list[torch.Tensor]): List of gradients for each parameter.
-            exp_avgs (list[torch.Tensor]): List of exponential moving averages of
-                gradients.
-            exp_avg_sqs (list[torch.Tensor]): List of exponential moving averages
-                of squared gradients.
-            exp_avg_slow (list[torch.Tensor]): List of slow exponential moving
-                averages of gradients.
-            state_steps (list[int]): List of steps for each parameter.
-            beta1 (float): Coefficient for the first moment estimate.
-            beta2 (float): Coefficient for the second moment estimate.
-            beta3 (float): Coefficient for the slow moment estimate.
-            alpha (float): Mixing coefficient.
-            T_alpha_beta3 (int): Time period for alpha and beta3 scheduling.
-            learning_rate (float): Learning rate.
-            weight_decay (float): Weight decay.
-            eps (float): Epsilon term for numerical stability.
-        """
+        """Update one parameter group in place."""
         for i, param in enumerate(params):
             grad = grads[i]
             exp_avg = exp_avgs[i]
@@ -211,7 +159,6 @@ def _update_adamemix(
                 alpha_t = alpha
                 beta3_t = beta3
 
-            # Decay the first and second moment running average coefficient
             exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
             exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
             exp_avg_slow_i.mul_(beta3_t).add_(grad, alpha=1 - beta3_t)
diff --git a/src/sequifier/optimizers/optimizers.py b/src/sequifier/optimizers/optimizers.py
index 2930433e..dbcef9cd 100644
--- a/src/sequifier/optimizers/optimizers.py
+++ b/src/sequifier/optimizers/optimizers.py
@@ -7,14 +7,7 @@
 
 
 def get_optimizer_class(optimizer_name: str) -> torch.optim.Optimizer:
-    """Gets the optimizer class from a string.
-
-    Args:
-        optimizer_name: The name of the optimizer.
-
-    Returns:
-        The optimizer class.
-    """
+    """Resolve a custom, torch-optimizer, or torch optimizer class."""
     if optimizer_name in CUSTOM_OPTIMIZERS:
         return CUSTOM_OPTIMIZERS[optimizer_name]
     elif hasattr(torch_optimizer, optimizer_name):
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 9c2d74fc..9a2f4938 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -61,20 +61,7 @@ def _stable_json_digest(value: Any) -> str:
 
 @beartype
 def preprocess(args: Any, args_config: dict[str, Any]) -> None:
-    """Runs the main data preprocessing pipeline.
-
-    This function loads the preprocessing configuration, initializes the
-    `Preprocessor` class, and executes the preprocessing steps based on the
-    loaded configuration.
-
-    Args:
-        args: An object containing command-line arguments. Expected to have
-            a `config_path` attribute specifying the path to the YAML
-            configuration file.
-        args_config: A dictionary containing additional configuration parameters
-            that may override or supplement the settings loaded from the
-            config file.
-    """
+    """Load preprocessing config and run preprocessing."""
     logger.info("--- Starting Preprocessing ---")
     config_path = args.config_path or "configs/preprocess.yaml"
     config = load_preprocessor_config(config_path, args_config)
@@ -83,22 +70,7 @@ def preprocess(args: Any, args_config: dict[str, Any]) -> None:
 
 
 class Preprocessor:
-    """A class for preprocessing data for the sequifier model.
-
-    This class handles loading, preprocessing, and saving data. It supports
-    single-file and multi-file processing, and can handle large datasets by
-    processing them in batches.
-
-    Attributes:
-        project_root (str): The path to the sequifier project directory.
-        batches_per_file (int): The number of batches to process per file.
-        data_name_root (str): The root name of the data file.
-        merge_output (bool): Whether to combine the output into a single file.
-        target_dir (str): The target directory for temporary files.
-        seed (int): The random seed for reproducibility.
-        n_cores (int): The number of cores to use for parallel processing.
-        split_paths (list[str]): The paths to the output split files.
-    """
+    """Stateful preprocessing pipeline for single-file or folder inputs."""
 
     @beartype
     def __init__(
@@ -124,28 +96,7 @@ def __init__(
         max_target_offset: int = 1,
         mask_column: Optional[str] = None,
     ):
-        """Initializes the Preprocessor with the given parameters.
-
-        Args:
-            project_root: The path to the sequifier project directory.
-            data_path: The path to the input data file.
-            read_format: The file type of the input data.
-            write_format: The file type for the preprocessed output data.
-            merge_output: Whether to combine the output into a single file.
-            selected_columns: A list of columns to be included in the preprocessing.
-            split_ratios: A list of floats that define the relative sizes of data splits.
-            stored_context_width: The physical serialized window width.
-            max_target_offset: The number of future items retained after each input window.
-            stride_by_split: A list of step sizes for creating subsequences.
-            max_rows: The maximum number of input rows to process.
-            seed: A random seed for reproducibility.
-            n_cores: The number of CPU cores to use for parallel processing.
-            batches_per_file: The number of batches to process per file.
-            process_by_file: A flag to indicate if processing should be done file by file.
-            use_precomputed_maps: An optional list of columns for which to enforce precomputed maps
-            metadata_config_path: Optional path to a precomputed metadata config
-            mask_column: Optional input column used to mask all data columns
-        """
+        """Initialize and run preprocessing from validated config fields."""
         self.project_root = project_root
         self.batches_per_file = batches_per_file
         self.data_path = data_path
@@ -195,12 +146,9 @@ def __init__(
         self._setup_split_paths(write_format, len(split_ratios))
 
         if self.continue_preprocessing:
-            # 1. Determine what paths indicate "completion"
             if self.merge_output:
-                # If merging, check for the final files (e.g., "data/mydata-split0.pt")
                 paths_to_check = self.split_paths
             else:
-                # If not merging, check for the output folders (e.g., "data/mydata-split0")
                 paths_to_check = [
                     os.path.join(
                         self.project_root, "data", f"{self.data_name_root}-split{i}"
@@ -208,7 +156,6 @@ def __init__(
                     for i in range(len(split_ratios))
                 ]
 
-            # 2. If any target exists, skip processing and jump to cleanup
             if any(os.path.exists(p) for p in paths_to_check):
                 logger.info(
                     "Existing split paths found with continue_preprocessing=True. "
@@ -420,23 +367,7 @@ def __init__(
     def _create_schema(
         self, col_types: dict[str, str], stored_context_width: int
     ) -> dict[str, Any]:
-        """Creates the Polars schema for the intermediate sequence DataFrame.
-
-        This schema defines the structure of the DataFrame after sequence
-        extraction, which includes sequence identifiers, the start item position
-        within the original sequence, the input column name, and columns for
-        each stored item in the serialized window.
-
-        Args:
-            col_types: A dictionary mapping data column names to their Polars
-                string representations (e.g., "Int64", "Float64").
-            stored_context_width: The number of items stored for each extracted window.
-
-        Returns:
-            A dictionary defining the Polars schema. Keys are column names
-            (e.g., "sequenceId", "subsequenceId", "startItemPosition", "inputCol", "0", "1", ...)
-            and values are Polars data types (e.g., `pl.Int64`).
-        """
+        """Build the long-format extracted-window schema."""
         schema = {
             "sequenceId": pl.Int64,
             "subsequenceId": pl.Int64,
@@ -483,39 +414,7 @@ def _get_column_metadata_across_files(
         dict[str, str],
         list[str],
     ]:
-        """Scans multiple data files to compute combined column metadata.
-
-        This method iterates through all files in `data_path` matching `read_format`,
-        loading each one to incrementally build up metadata. It computes:
-        1.  ID maps for categorical/string columns.
-        2.  Mean and standard deviation for numerical (float) columns.
-        3.  The total number of unique classes for mapped columns.
-        4.  The data types of all columns.
-
-        This is used when the dataset is split into multiple files to get a
-        consistent global view of the data.
-
-        Args:
-            data_path: The path to the root data directory.
-            read_format: The file extension (e.g., "csv", "parquet") to read.
-            max_rows: The maximum total number of rows to process across all
-                files. If `None`, all rows are processed.
-            selected_columns: A list of columns to include. If `None`, all
-                columns (except "sequenceId" and "itemPosition") are used.
-
-        Returns:
-            A tuple containing:
-                - n_classes (dict[str, int]): Map of column name to its
-                  number of unique classes (including padding/unknown).
-                - id_maps (dict[str, dict[Union[str, int], int]]): Nested map
-                  from column name to its value-to-integer-ID map.
-                - selected_columns_statistics (dict[str, dict[str, float]]):
-                  Nested map from numerical column name to its 'mean' and 'std'.
-                - col_types (dict[str, str]): Map of column name to its
-                  Polars string data type.
-                - data_columns (list[str]): List of all processed data
-                  column names.
-        """
+        """Accumulate metadata/statistics over a folder input."""
 
         n_rows_running_count = 0
         id_maps, selected_columns_statistics = {}, {}
@@ -549,7 +448,6 @@ def _get_column_metadata_across_files(
                         self.mask_column,
                     )
 
-                    # Get columns for the current file (excluding metadata cols)
                     current_file_cols = _get_data_columns(data, self.mask_column)
 
                     if col_types is None:
@@ -615,13 +513,7 @@ def _get_column_metadata_across_files(
 
     @beartype
     def _setup_directories(self) -> None:
-        """Sets up the output directories for preprocessed data.
-
-        This method creates the base `data/` directory within the `project_root`
-        if it doesn't exist. It also creates a temporary directory (defined by
-        `self.target_dir`) for storing intermediate batch files, removing it
-        first if it already exists to ensure a clean run.
-        """
+        """Prepare or validate preprocessing temp directories."""
 
         temp_path = os.path.join(self.project_root, "data", self.target_dir)
 
@@ -636,16 +528,7 @@ def _setup_directories(self) -> None:
 
     @beartype
     def _setup_split_paths(self, write_format: str, n_splits: int) -> None:
-        """Sets up the final output paths for the data splits.
-
-        This method constructs the full file paths for each data split
-        (e.g., train, validation, test) based on the `data_name_root` and
-        `write_format`. The paths are stored in the `self.split_paths` attribute.
-
-        Args:
-            write_format: The file extension for the output files (e.g., "pt", "parquet").
-            n_splits: The number of splits to create (e.g., 3 for train/val/test).
-        """
+        """Set final split output paths."""
         split_paths = [
             os.path.join(
                 self.project_root,
@@ -679,28 +562,7 @@ def _process_batches_multiple_files(
         subsequence_start_mode: str = "distribute",
         mask_column: Optional[str] = None,
     ) -> None:
-        """Processes batches of data from multiple files.
-
-        Args:
-            file_paths: A list of file paths to process.
-            read_format: The file format to read.
-            selected_columns: A list of columns to be included in the preprocessing.
-            max_rows: The maximum number of rows to process.
-            schema: The schema for the preprocessed data.
-            n_cores: The number of cores to use for parallel processing.
-            layout: The sequence and serialized-window layout.
-            stride_by_split: A list of step sizes for creating subsequences.
-            data_columns: A list of data columns.
-            n_classes: A dictionary containing the number of classes for each categorical column.
-            id_maps: A dictionary containing the id maps for each categorical column.
-            selected_columns_statistics: A dictionary containing the statistics for each numerical column.
-            col_types: A dictionary containing the column types.
-            split_ratios: A list of floats that define the relative sizes of data splits.
-            write_format: The file format for the output files.
-            process_by_file: A flag to indicate if processing should be done file by file.
-            subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-            mask_column: Optional input column used to mask all data columns.
-        """
+        """Dispatch folder preprocessing by file or by process shard."""
         if mask_column is None:
             mask_column = self.mask_column
 
@@ -815,21 +677,7 @@ def _process_batches_multiple_files(
 
     @beartype
     def _cleanup(self, write_format: str) -> None:
-        """Finalizes output files and removes temporary directories.
-
-        If `write_format` is 'pt' and `merge_output` is False,
-        this method moves the processed .pt batch files from the temporary
-        `target_dir` into their final split-specific subfolders (e.g.,
-        'data_name_root-split0/'). It also generates a 'metadata.json' file
-        in each of these subfolders, which is required by
-        `SequifierDatasetFromFolderPt`.
-
-        Finally, it removes the temporary `target_dir` if it's empty or
-        if `target_dir` is "temp" (implying `merge_output` was True).
-
-        Args:
-            write_format: The file format of the output files (e.g., "pt").
-        """
+        """Move split outputs, write folder metadata, and remove temp files."""
 
         logger.info("Start cleanup")
         temp_output_path = os.path.join(self.project_root, "data", self.target_dir)
@@ -943,23 +791,7 @@ def _export_metadata(
         col_types: dict[str, str],
         selected_columns_statistics: dict[str, dict[str, float]],
     ) -> None:
-        """Exports the computed data metadata to a JSON file.
-
-        Saves metadata such as class counts, ID mappings, split paths,
-        column types, and numerical statistics to a JSON file. This file is
-        saved in the `configs/metadata_configs/` directory, named after the
-        `data_name_root`. This metadata is essential for initializing the
-        model and data loaders during training.
-
-        Args:
-            id_maps: A dictionary containing the id maps for each
-                categorical column.
-            n_classes: A dictionary containing the number of classes for
-                each categorical column.
-            col_types: A dictionary containing the column types.
-            selected_columns_statistics: A dictionary containing the
-                statistics for each numerical column.
-        """
+        """Write metadata config JSON for training/inference."""
         data_driven_config = {
             "n_classes": n_classes,
             "id_maps": id_maps,
@@ -993,19 +825,7 @@ def _export_metadata(
 
     @beartype
     def _create_metadata_for_folder(self, folder_path: str, write_format: str) -> None:
-        """Scans a directory for batch files, counts samples, and writes metadata.json.
-
-        This method is used when `merge_output` is False. It iterates over all
-        files in the given `folder_path` matching the `write_format`, loads each
-        one to count the number of samples (sequences), and writes a `metadata.json`
-        file in that same folder. This JSON file contains the total sample count and a
-        list of all batch files with their respective sample counts.
-
-        Args:
-            folder_path: The path to the directory containing the batch files
-                for a specific data split.
-            write_format: The file format of the output files (e.g., 'pt', 'parquet').
-        """
+        """Write metadata.json for an unmerged split folder."""
         logger.info(f"Creating metadata for folder '{folder_path}'")
         batch_files_metadata = []
         total_samples = 0
@@ -1166,37 +986,7 @@ def _apply_column_statistics(
     n_classes: Optional[dict[str, int]] = None,
     col_types: Optional[dict[str, str]] = None,
 ) -> tuple[pl.DataFrame, dict[str, int], dict[str, str]]:
-    """Applies pre-computed statistics to transform the data.
-
-    This function performs two main transformations on the DataFrame:
-    1.  **Categorical Mapping**: Replaces original values in categorical
-        columns with their corresponding integer IDs using `id_maps`.
-    2.  **Numerical Standardization**: Standardizes numerical columns
-        (those in `selected_columns_statistics`) using the Z-score
-        formula: (value - mean) / (std + 1e-9).
-
-    It also computes `n_classes` and `col_types` if they are not provided.
-
-    Args:
-        data: The input Polars DataFrame to transform.
-        data_columns: A list of column names to process.
-        id_maps: A nested dictionary mapping column names to their
-            value-to-integer-ID maps.
-        selected_columns_statistics: A nested dictionary mapping
-            column names to their 'mean' and 'std' statistics.
-        n_classes: An optional dictionary mapping column names to
-            their total number of unique classes. If `None`, it's computed
-            from `id_maps`.
-        col_types: An optional dictionary mapping column names to
-            their string data types. If `None`, it's inferred from the
-            `data` schema.
-
-    Returns:
-        A tuple `(data, n_classes, col_types)` where:
-            - `data`: The transformed Polars DataFrame.
-            - `n_classes`: The (potentially computed) class count dictionary.
-            - `col_types`: The (potentially computed) column type dictionary.
-    """
+    """Apply categorical maps and numeric standardization."""
     if n_classes is None:
         n_classes = {col: max(id_maps[col].values()) + 1 for col in id_maps}
 
@@ -1235,17 +1025,7 @@ def load_precomputed_id_maps(
     data_columns: Optional[list[str]],
     required_maps: Optional[list[str]] = None,
 ) -> dict[str, dict[Union[str, int], int]]:
-    """Loads custom ID maps from configs/id_maps if the folder exists.
-
-    Args:
-        project_root: The path to the project root directory.
-        data_columns: Optional list of columns present in the data to validate
-            against the found map files.
-        required_maps: Optional list of columns for which a precomputed id_map is required
-
-    Returns:
-        A dictionary mapping column names to their ID maps.
-    """
+    """Load and validate precomputed ID maps."""
     custom_maps = {}
     path = os.path.join(project_root, "configs", "id_maps")
 
@@ -1264,7 +1044,6 @@ def load_precomputed_id_maps(
                     )
 
                 with open(os.path.join(path, file), "r") as f:
-                    # Load and ensure values are integers
                     m = {k: int(v) for k, v in json.load(f).items()}
 
                     if not len(m) > 0:
@@ -1328,38 +1107,7 @@ def _get_column_statistics(
     dict[str, dict[Union[str, int], int]],
     dict[str, dict[str, float]],
 ]:
-    """Computes or updates column statistics from a data chunk.
-
-    This function iterates over the `data_columns` and updates the
-    `id_maps` and `selected_columns_statistics` dictionaries based on the
-    provided `data` chunk.
-
-    - For string/integer columns: It creates an ID map of unique values and
-      merges it with any existing map in `id_maps`.
-    - For float columns: It computes the mean and standard deviation of the
-      chunk and combines them with the existing statistics in
-      `selected_columns_statistics` using a numerically stable parallel
-      algorithm, weighted by `n_rows_running_count`.
-
-    Args:
-        data: The Polars DataFrame chunk to process.
-        data_columns: A list of column names in `data` to analyze.
-        id_maps: The dictionary of existing ID maps to be updated.
-        selected_columns_statistics: The dictionary of existing numerical
-            statistics to be updated.
-        n_rows_running_count: The total number of rows processed *before*
-            this chunk, used for weighting statistics.
-        precomputed_id_maps: A dictionary of pre-loaded ID maps that should
-            be applied and not re-computed.
-
-    Returns:
-        A tuple `(id_maps, selected_columns_statistics)` containing the
-        updated dictionaries.
-
-    Raises:
-        ValueError: If a column has an unsupported data type (neither
-            string, integer, nor float).
-    """
+    """Update ID maps and numeric statistics from one chunk."""
     if mask_column is not None and mask_column in data.columns:
         mask_expr = _validate_and_create_mask_column_expr(
             data[mask_column], data.schema[mask_column], mask_column
@@ -1435,26 +1183,7 @@ def _load_and_preprocess_data(
     max_rows: Optional[int],
     mask_column: Optional[str] = None,
 ) -> pl.DataFrame:
-    """Loads data from a file and performs initial preparation.
-
-    This function reads a data file using the specified `read_format`.
-    It then performs the following steps:
-    1.  Asserts that the data contains no null or NaN values.
-    2.  Selects only the `selected_columns` if provided.
-    3.  Slices the DataFrame to `max_rows` if provided.
-
-    Args:
-        data_path: The path to the data file.
-        read_format: The file format to read (e.g., "csv", "parquet").
-        selected_columns: A list of columns to load. If `None`, all
-            columns are loaded.
-        max_rows: The maximum number of rows to load. If `None`, all
-            rows are loaded.
-
-    Returns:
-        A Polars DataFrame containing the loaded and initially
-        prepared data.
-    """
+    """Read, validate, column-filter, and row-limit one input file."""
     logger.info(f"Reading data from '{data_path}'...")
     columns_to_read = _selected_columns_with_optional_mask(
         data_path, read_format, selected_columns, mask_column
@@ -1560,34 +1289,7 @@ def _process_batches_multiple_files_inner(
     subsequence_start_mode: str,
     mask_column: Optional[str],
 ):
-    """Inner function for processing batches of data from multiple files.
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        data_name_root: The root name of the data file.
-        process_id: The id of the process.
-        file_paths: A list of file paths to process.
-        read_format: The file format to read.
-        selected_columns: A list of columns to be included in the preprocessing.
-        max_rows: The maximum number of rows to process.
-        schema: The schema for the preprocessed data.
-        n_cores: The number of cores to use for parallel processing.
-        layout: The sequence and serialized-window layout.
-        stride_by_split: A list of step sizes for creating subsequences.
-        data_columns: A list of data columns.
-        n_classes: A dictionary containing the number of classes for each categorical column.
-        id_maps: A dictionary containing the id maps for each categorical column.
-        selected_columns_statistics: A dictionary containing the statistics for each numerical column.
-        col_types: A dictionary containing the column types.
-        split_ratios: A list of floats that define the relative sizes of data splits.
-        write_format: The file format for the output files.
-        split_paths: The paths to the output split files.
-        target_dir: The target directory for temporary files.
-        batches_per_file: The number of batches to process per file.
-        merge_output: Whether to combine the output into a single file.
-        continue_preprocessing: Continue preprocessing job that was interrupted while writing to temp folder.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-    """
+    """Process this worker's file shard."""
     n_files = len(file_paths)
     if n_files <= 0:
         raise ValueError("No files found to process.")
@@ -1713,29 +1415,7 @@ def _process_batches_single_file(
     subsequence_start_mode: str,
     merge_output: bool,
 ) -> int:
-    """Processes batches of data from a single file.
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        data_name_root: The root name of the data file.
-        data: The data to process.
-        schema: The schema for the preprocessed data.
-        n_cores: The number of cores to use for parallel processing.
-        layout: The sequence and serialized-window layout.
-        stride_by_split: A list of step sizes for creating subsequences.
-        data_columns: A list of data columns.
-        col_types: A dictionary containing the column types.
-        split_ratios: A list of floats that define the relative sizes of data splits.
-        write_format: The file format for the output files.
-        split_paths: The paths to the output split files.
-        target_dir: The target directory for temporary files.
-        batches_per_file: The number of batches to process per file.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-        merge_output: merge output
-
-    Returns:
-        The number of batches processed.
-    """
+    """Split one file into worker batches and preprocess them."""
     n_cores = n_cores or multiprocessing.cpu_count()
     batch_limits = get_batch_limits(data, n_cores)
     valid_batch_limits = [(s, e) for s, e in batch_limits if (e - s) > 0]
@@ -1774,41 +1454,20 @@ def _process_batches_single_file(
 def get_combined_statistics(
     n1: int, mean1: float, std1: float, n2: int, mean2: float, std2: float
 ) -> tuple[float, float]:
-    """Calculates the combined mean and standard deviation of two data subsets.
-
-    Uses a stable parallel algorithm (related to Welford's algorithm) to
-    combine statistics from two subsets without needing the original data.
-
-    Args:
-        n1: Number of samples in subset 1.
-        mean1: Mean of subset 1.
-        std1: Standard deviation of subset 1.
-        n2: Number of samples in subset 2.
-        mean2: Mean of subset 2.
-        std2: Standard deviation of subset 2.
-
-    Returns:
-        A tuple `(combined_mean, combined_std)` containing the combined
-        mean and standard deviation of the two subsets.
-    """
+    """Combine two mean/std summaries."""
     if n1 == 0:
         return mean2, std2
     if n2 == 0:
         return mean1, std1
 
-    # Step 1: Calculate the combined mean.
     combined_mean = (n1 * mean1 + n2 * mean2) / (n1 + n2)
 
     if n1 + n2 <= 1:
         return combined_mean, 0.0
 
-    # Step 2: Calculate the pooled sum of squared differences.
-    # This includes the internal variance of each subset and the variance
-    # between the subset mean and the combined mean.
     sum_of_squares1 = (n1 - 1) * std1**2 + n1 * (mean1 - combined_mean) ** 2
     sum_of_squares2 = (n2 - 1) * std2**2 + n2 * (mean2 - combined_mean) ** 2
 
-    # Step 3: Calculate the combined standard deviation.
     combined_std = math.sqrt((sum_of_squares1 + sum_of_squares2) / (n1 + n2 - 1))
 
     return combined_mean, combined_std
@@ -1816,20 +1475,7 @@ def get_combined_statistics(
 
 @beartype
 def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]:
-    """Creates a map from unique values in a column to integer indices.
-
-    Finds all unique values in the specified `column` of the `data`
-    DataFrame, sorts them, and creates a dictionary mapping each unique
-    value to a 1-based integer index.
-
-    Args:
-        data: The Polars DataFrame containing the column.
-        column: The name of the column to map.
-
-    Returns:
-        A dictionary mapping unique values (str or int) to an integer
-        index (starting from 1).
-    """
+    """Map sorted user values to IDs after reserved tokens."""
     ids = sorted(
         [int(x) if not isinstance(x, str) else x for x in np.unique(data[column])]
     )  # type: ignore
@@ -1871,21 +1517,7 @@ def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]
 
 @beartype
 def get_batch_limits(data: pl.DataFrame, n_batches: int) -> list[tuple[int, int]]:
-    """Calculates row indices to split a DataFrame into batches.
-
-    This function divides the DataFrame into `n_batches` roughly equal
-    chunks. Crucially, it ensures that no `sequenceId` is split across
-    two different batches. It does this by finding the ideal split points
-    and then adjusting them to the nearest `sequenceId` boundary.
-
-    Args:
-        data: The DataFrame to split. Must be sorted by "sequenceId".
-        n_batches: The desired number of batches.
-
-    Returns:
-        A list of `(start_index, end_index)` tuples, where each tuple
-        defines the row indices for a batch.
-    """
+    """Split rows into batches without crossing sequenceId boundaries."""
     sequence_ids = data.get_column("sequenceId").to_numpy()
     new_sequence_id_indices = np.concatenate(
         [
@@ -1918,20 +1550,7 @@ def get_batch_limits(data: pl.DataFrame, n_batches: int) -> list[tuple[int, int]
 def combine_maps(
     map1: dict[Union[str, int], int], map2: dict[Union[str, int], int]
 ) -> dict[Union[str, int], int]:
-    """Combines two ID maps into a new, consolidated map.
-
-    Takes all unique keys from both `map1` and `map2`, sorts them,
-    and creates a new, single map where keys are mapped to 1-based
-    indices based on the sorted order. This ensures a consistent
-    mapping across different data chunks.
-
-    Args:
-        map1: The first ID map.
-        map2: The second ID map.
-
-    Returns:
-        A new, combined, and re-indexed ID map.
-    """
+    """Merge maps and reassign user IDs after reserved tokens."""
     keys1 = {k for k in map1.keys() if k not in SPECIAL_TOKEN_LABELS}
     keys2 = {k for k in map2.keys() if k not in SPECIAL_TOKEN_LABELS}
 
@@ -1940,7 +1559,6 @@ def combine_maps(
         id_: i + SPECIAL_TOKEN_IDS.user_start for i, id_ in enumerate(combined_keys)
     }
 
-    # Re-apply special mapping rules if these are string keys
     if combined_keys and isinstance(combined_keys[0], str):
         id_map[SPECIAL_TOKEN_IDS.labels_by_id[SPECIAL_TOKEN_IDS.unknown]] = (
             SPECIAL_TOKEN_IDS.unknown
@@ -1954,22 +1572,7 @@ def combine_maps(
 
 @beartype
 def get_group_bounds(data_subset: pl.DataFrame, split_ratios: list[float]):
-    """Calculates row indices for splitting a sequence into groups.
-
-    This function takes a DataFrame `data_subset` (which typically
-    contains all items for a single `sequenceId`) and calculates the
-    row indices to split it into multiple groups (e.g., train, val, test)
-    based on the provided `split_ratios`.
-
-    Args:
-        data_subset: The DataFrame (for a single sequence) to split.
-        split_ratios: A list of floats (e.g., [0.8, 0.1, 0.1]) that
-            sum to 1.0, defining the relative sizes of the splits.
-
-    Returns:
-        A list of `(start_index, end_index)` tuples, one for each
-        proportion, defining the row slices for each group.
-    """
+    """Return per-split row bounds for one sequence."""
     n = data_subset.shape[0]
     upper_bounds = list((np.cumsum(split_ratios) * n).astype(int))
     lower_bounds = [0] + list(upper_bounds[:-1])
@@ -1984,26 +1587,7 @@ def process_and_write_data_pt(
     path: str,
     column_types: dict[str, str],
 ):
-    """Processes the sequence DataFrame and writes it to a .pt file.
-
-    This function takes the long-format sequence DataFrame (`data`),
-    aggregates it by `sequenceId` and `subsequenceId`, and pivots it
-    so that each `inputCol` becomes its own column containing a list
-    of sequence items. It also extracts the `startItemPosition`.
-
-    It then converts these lists into NumPy arrays and stores one full
-    sequence tensor per feature. Each tensor has shape
-    `(batch_size, stored_context_width)`. The final five-element tuple
-    `(sequences_dict, sequence_ids_tensor, subsequence_ids_tensor, start_item_positions_tensor, left_pad_lengths_tensor)`
-    is saved to a .pt file using `torch.save`.
-
-    Args:
-        data: The long-format Polars DataFrame of extracted sequences.
-        stored_context_width: The stored serialized window width.
-        path: The output file path (e.g., "data/batch_0.pt").
-        column_types: A dictionary mapping column names to their
-            string data types, used to determine the correct torch dtype.
-    """
+    """Write long-format sequences as packed PT tensors."""
     if data.is_empty():
         return
 
@@ -2079,23 +1663,7 @@ def _write_accumulated_sequences(
     layout: StoredWindowLayout,
     col_types: dict[str, str],
 ):
-    """Helper to write a batch of accumulated sequences to a single .pt file.
-
-    This function concatenates a list of sequence DataFrames and writes
-    the combined DataFrame to a single .pt file using
-    `process_and_write_data_pt`. This is used to batch multiple sequences
-    into fewer, larger files when `write_format` is 'pt'.
-
-    Args:
-        sequences_to_write: A list of Polars DataFrames to combine and write.
-        split_path: The base path for the split (e.g., "data/split0.pt").
-        write_format: The file format (e.g., "pt").
-        process_id: The ID of the parent multiprocessing process.
-        file_index: The index of this file batch for this split and process.
-        target_dir: The temporary directory to write the file into.
-        layout: The sequence and serialized-window layout.
-        col_types: A dictionary mapping column names to their string types.
-    """
+    """Write one accumulated sequence shard."""
 
     if not sequences_to_write:
         return
@@ -2133,29 +1701,10 @@ def preprocess_batch(
     subsequence_start_mode: str,
     merge_output: bool,
 ) -> None:
-    """Processes a batch of data.
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        data_name_root: The root name of the data file.
-        process_id: The id of the process.
-        batch: The batch of data to process.
-        schema: The schema for the preprocessed data.
-        split_paths: The paths to the output split files.
-        layout: The sequence and serialized-window layout.
-        stride_by_split: A list of step sizes for creating subsequences.
-        data_columns: A list of data columns.
-        col_types: A dictionary containing the column types.
-        split_ratios: A list of floats that define the relative sizes of data splits.
-        target_dir: The target directory for temporary files.
-        write_format: The file format for the output files.
-        batches_per_file: The number of batches to process per file.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-    """
+    """Extract and write all split windows for one batch."""
     sequence_ids = sorted(batch.get_column("sequenceId").unique().to_list())
 
     if not merge_output:
-        # New logic for batching sequences into files for .pt format
         sequences_by_split = {i: [] for i in range(len(split_paths))}
         file_indices = {i: 0 for i in range(len(split_paths))}
 
@@ -2266,30 +1815,7 @@ def extract_sequences(
     columns: list[str],
     subsequence_start_mode: str,
 ) -> pl.DataFrame:
-    """Extracts subsequences from a DataFrame of full sequences.
-
-    This function takes a DataFrame where each row contains all items
-    for a single `sequenceId`. It iterates through each `sequenceId`,
-    extracts all possible serialized windows using the
-    specified `stride_for_split`, calculates the starting position of each
-    subsequence within the original sequence, and formats them into a new,
-    long-format DataFrame that conforms to the provided `schema`.
-
-    Args:
-        data: The input Polars DataFrame, grouped by "sequenceId".
-        schema: The schema for the output long-format DataFrame.
-        layout: The sequence and serialized-window layout.
-        stride_for_split: The step size to use when sliding the window
-            to create subsequences.
-        columns: A list of the data column names (features) to extract.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-
-    Returns:
-        A new, long-format Polars DataFrame containing the extracted
-        subsequences, matching the provided `schema`. Includes columns
-        for `sequenceId`, `subsequenceId`, `startItemPosition`, `inputCol`,
-        and the sequence items ('0', '1', ...).
-    """
+    """Extract long-format windows from grouped sequences."""
     if data.is_empty():
         return pl.DataFrame(schema=schema)
 
@@ -2341,23 +1867,7 @@ def get_subsequence_starts(
     stride_for_split: int,
     subsequence_start_mode: str,
 ) -> np.ndarray:
-    """Calculates the start indices for extracting subsequences.
-
-    This function determines the starting indices for sliding a window of
-    `stored_context_width` over an input sequence of `in_context_length`. It aims to
-    use `stride_for_split`, but adjusts the step size slightly to ensure
-    that the windows are distributed as evenly as possible and cover the
-    full sequence from the beginning to the end.
-
-    Args:
-        in_context_length: The length of the original input sequence.
-        stored_context_width: The stored window length to extract.
-        stride_for_split: The *desired* step size between subsequences.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-
-    Returns:
-        A numpy array of integer start indices for each subsequence.
-    """
+    """Return window start indices for distribute/exact modes."""
     if subsequence_start_mode not in ["distribute", "exact"]:
         raise ValueError(
             f"subsequence_start_mode must be 'distribute' or 'exact', got '{subsequence_start_mode}'"
@@ -2392,27 +1902,7 @@ def extract_subsequences(
     columns: list[str],
     subsequence_start_mode: str,
 ) -> tuple[dict[str, list[list[Union[float, int]]]], list[int], np.ndarray]:
-    """Extracts subsequences from a dictionary of sequence lists.
-
-    This function takes a dictionary `in_seq` where keys are column
-    names and values are lists of items for a single full sequence.
-    It first pads the sequences with 0s at the beginning if they are
-    shorter than `stored_context_width`. Then, it calculates the subsequence
-    start indices using `get_subsequence_starts` and extracts all
-    subsequences.
-
-    Args:
-        in_seq: A dictionary mapping column names to lists of items
-            (e.g., `{'col_A': [1, 2, 3, 4, 5], 'col_B': [6, 7, 8, 9, 10]}`).
-        stored_context_width: The stored window length to extract.
-        stride_for_split: The desired step size between subsequences.
-        columns: A list of the column names (keys in `in_seq`) to process.
-        subsequence_start_mode: "distribute" to minimize max subsequence overlap, or "exact".
-
-    Returns:
-        A dictionary mapping column names to a *list of lists*, where
-        each inner list is a subsequence.
-    """
+    """Extract padded windows plus left-pad lengths from one sequence."""
     in_seq_len = len(in_seq[columns[0]])
     pad_len = 0
     if in_seq_len < stored_context_width:
@@ -2445,18 +1935,7 @@ def extract_subsequences(
 
 @beartype
 def insert_top_folder(path: str, folder_name: str) -> str:
-    """Inserts a directory name into a file path, just before the filename.
-
-    Example:
-        `insert_top_folder("a/b/c.txt", "temp")` returns `"a/b/temp/c.txt"`
-
-    Args:
-        path: The original file path.
-        folder_name: The name of the folder to insert.
-
-    Returns:
-        The new path string with the folder inserted.
-    """
+    """Insert folder_name before the basename."""
     components = os.path.split(path)
     new_components = list(components[:-1]) + [folder_name] + [components[-1]]
     return os.path.join(*new_components)
@@ -2464,30 +1943,14 @@ def insert_top_folder(path: str, folder_name: str) -> str:
 
 @beartype
 def cast_columns_to_string(data: pl.DataFrame) -> pl.DataFrame:
-    """Casts the column names of a Polars DataFrame to strings.
-
-    This is often necessary because Polars schemas may use integers as
-    column names (e.g., '0', '1', '2'...) which need to be strings for
-    some operations.
-
-    Args:
-        data: The Polars DataFrame.
-
-    Returns:
-        The same DataFrame with its `columns` attribute modified.
-    """
+    """Cast Polars column names to strings."""
     data.columns = [str(col) for col in data.columns]
     return data
 
 
 @beartype
 def delete_files(files: Union[list[str], dict[int, list[str]]]) -> None:
-    """Deletes a list of files from the filesystem.
-
-    Args:
-        files: A list of file paths to delete, or a dictionary
-            whose values are lists of file paths to delete.
-    """
+    """Delete paths from a list or split-indexed dict."""
     if isinstance(files, dict):
         files = [x for y in list(files.values()) for x in y]
     for file in files:
@@ -2505,29 +1968,7 @@ def create_file_paths_for_multiple_files1(
     dataset_name: str,
     write_format: str,
 ) -> dict[int, list[str]]:
-    """Creates a dictionary of temporary file paths for a specific data file.
-
-    This is used in the multi-file, `merge_output=True`
-    workflow. It generates file path names for intermediate batches
-    *before* they are combined.
-
-    The naming pattern is:
-    `{dataset_name}-{process_id}-{file_index_str}-split{split}-{batch_id}.{write_format}`
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        target_dir: The temporary directory to place files in.
-        n_splits: The number of data splits.
-        n_batches: The number of batches created by the process.
-        process_id: The ID of the multiprocessing worker.
-        file_index_str: The index of the file being processed by this worker.
-        dataset_name: The root name of the dataset.
-        write_format: The file extension.
-
-    Returns:
-        A dictionary mapping a split index (int) to a list of file paths
-        (str) for all batches in that split.
-    """
+    """Return per-split temp paths for one multi-file shard."""
     files = {}
     for split in range(n_splits):
         files_for_split = [
@@ -2552,27 +1993,7 @@ def create_file_paths_for_single_file(
     dataset_name: str,
     write_format: str,
 ) -> dict[int, list[str]]:
-    """Creates a dictionary of temporary file paths for a single-file run.
-
-    This is used in the single-file, `merge_output=True`
-    workflow. It generates file path names for intermediate batches
-    created by different processes *before* they are combined.
-
-    The naming pattern is:
-    `{dataset_name}-split{split}-{core_id}.{write_format}`
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        target_dir: The temporary directory to place files in.
-        n_splits: The number of data splits.
-        n_batches: The number of processes (batches) running in parallel.
-        dataset_name: The root name of the dataset.
-        write_format: The file extension.
-
-    Returns:
-        A dictionary mapping a split index (int) to a list of file paths
-        (str) for all batches in that split.
-    """
+    """Return per-split temp paths for one single-file run."""
     files = {}
     for split in range(n_splits):
         files_for_split = [
@@ -2598,29 +2019,7 @@ def create_file_paths_for_multiple_files2(
     dataset_name: str,
     write_format: str,
 ) -> dict[int, list[str]]:
-    """Creates a dictionary of intermediate file paths for a multi-file run.
-
-    This is used in the multi-file, `merge_output=True`
-    workflow. It generates the file paths for the *combined* files
-    from each process, which are the *inputs* to the final combination step.
-
-    The naming pattern is:
-    `{dataset_name}-{process_id}-{file_index}-split{split}.{write_format}`
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        target_dir: The temporary directory where files are located.
-        n_splits: The number of data splits.
-        n_processes: The total number of multiprocessing workers.
-        n_files: A dictionary mapping `process_id` to the number of
-            files that process handled.
-        dataset_name: The root name of the dataset.
-        write_format: The file extension.
-
-    Returns:
-        A dictionary mapping a split index (int) to a list of all
-        intermediate combined file paths (str) for that split.
-    """
+    """Return per-split intermediate paths for multi-file merge."""
     files = {}
     n_files_max = max(n_files.values()) if n_files else 1
     pad_width = len(str(n_files_max - 1))
@@ -2652,31 +2051,7 @@ def combine_multiprocessing_outputs(
     pre_split_str: Optional[str] = None,
     post_split_str: Optional[str] = None,
 ) -> None:
-    """Combines multiple intermediate batch files into final split files.
-
-    This function iterates through each split and combines all the
-    intermediate files listed in `input_files[split]` into a single
-    final output file for that split.
-
-    - For "csv" format, it uses the `csvstack` command-line utility.
-    - For "parquet" format, it uses `pyarrow.parquet.ParquetWriter`
-      to concatenate the files efficiently.
-
-    Args:
-        project_root: The path to the sequifier project directory.
-        target_dir: The temporary directory containing intermediate files.
-        n_splits: The number of data splits.
-        input_files: A dictionary mapping split index (int) to a list
-            of input file paths (str) for that split.
-        dataset_name: The root name for the final output files.
-        write_format: The file format ("csv" or "parquet").
-        in_target_dir: If True, the final combined file is written
-            inside `target_dir`. If False, it's written to `data/`.
-        pre_split_str: An optional string to insert into the filename
-            before the "-split{i}" part.
-        post_split_str: An optional string to insert into the filename
-            after the "-split{i}" part.
-    """
+    """Combine per-split intermediate files."""
     for split in range(n_splits):
         split_file_path = create_split_file_path(
             project_root,
@@ -2732,18 +2107,7 @@ def create_split_file_path(
 
 @beartype
 def combine_parquet_files(files: list[str], out_path: str) -> None:
-    """Combines multiple Parquet files into a single Parquet file.
-
-    This function reads the schema from the first file and uses it to
-    initialize a `ParquetWriter`. It then iterates through all files in
-    the list, reading each one as a table and writing it to the new
-    combined file. This is more memory-efficient than reading all files
-    into one large table first.
-
-    Args:
-        files: A list of paths to the Parquet files to combine.
-        out_path: The path for the combined output Parquet file.
-    """
+    """Stream-concatenate Parquet files with the first file schema."""
     schema = pq.ParquetFile(files[0]).schema_arrow
     with pq.ParquetWriter(out_path, schema=schema, compression="snappy") as writer:
         for file in files:
diff --git a/src/sequifier/sequifier.py b/src/sequifier/sequifier.py
index 54168070..ad62ec32 100644
--- a/src/sequifier/sequifier.py
+++ b/src/sequifier/sequifier.py
@@ -14,15 +14,7 @@
 
 @beartype
 def build_args_config(args: Any) -> dict[str, Any]:
-    """
-    Build configuration dictionary from command-line arguments.
-
-    Args:
-        args: Parsed command-line arguments.
-
-    Returns:
-        Dictionary containing configuration options.
-    """
+    """Build config overrides from parsed CLI args."""
 
     args_config = {
         k: v
@@ -55,12 +47,7 @@ def build_args_config(args: Any) -> dict[str, Any]:
 
 @beartype
 def setup_parser() -> ArgumentParser:
-    """
-    Set up the argument parser for the command-line interface.
-
-    Returns:
-        Configured ArgumentParser object.
-    """
+    """Build the sequifier CLI parser."""
     parser = ArgumentParser()
     subparsers = parser.add_subparsers(dest="command", help="Available commands")
 
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 8f294da7..3a68058c 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -86,7 +86,7 @@
 
 
 def cleanup():
-    """Cleans up the distributed training environment."""
+    """Destroy the active distributed process group."""
     dist.destroy_process_group()
 
 
@@ -122,16 +122,7 @@ def train_worker(
     global_rank: int,
     torch_compile: str,
 ):
-    """The worker function for distributed training.
-
-    Args:
-        rank: The rank of the current process.
-        world_size: The total number of processes.
-        config: The training configuration.
-        from_folder: Whether to load data from a folder (e.g., preprocessed .pt files)
-                     or a single file (e.g., .parquet).
-        global_rank: The global rank
-    """
+    """Run one local distributed-training worker."""
     logger = configure_logger(config.project_root, config.model_name, global_rank)
 
     if config.training_spec.distributed:
@@ -141,7 +132,6 @@ def train_worker(
             global_rank, local_rank, world_size, config.training_spec.backend
         )
 
-    # 1. Create Datasets and DataLoaders with DistributedSampler
     if from_folder:
         if config.read_format == "pt":
             if config.training_spec.load_full_data_to_ram:
@@ -425,12 +415,7 @@ def _mp_train_worker_wrapper(
 
 @beartype
 def train(args: Any, args_config: dict[str, Any]) -> None:
-    """The main training function.
-
-    Args:
-        args: The command-line arguments.
-        args_config: The configuration dictionary.
-    """
+    """Load train config and launch local or distributed training."""
     config_path = args.config_path or "configs/train.yaml"
     config = load_train_config(config_path, args_config, args.skip_metadata)
 
@@ -481,14 +466,7 @@ def train(args: Any, args_config: dict[str, Any]) -> None:
 
 @beartype
 def format_number(number: Union[int, float, np.float32]) -> str:
-    """Format a number for display.
-
-    Args:
-        number: The number to format.
-
-    Returns:
-        A formatted string representation of the number.
-    """
+    """Format finite values, zero, and NaN for logs."""
     if np.isnan(number):
         return "NaN"
     elif number == 0:
@@ -565,29 +543,16 @@ def accumulate_class_counts(
 
 
 class TransformerEmbeddingModel(nn.Module):
-    """A wrapper around the TransformerModel to expose the embedding functionality."""
+    """Embedding-only wrapper for TransformerModel."""
 
     def __init__(self, transformer_model: "TransformerModel"):
-        """Initializes the TransformerEmbeddingModel.
-
-        Args:
-            transformer_model: The TransformerModel to wrap.
-        """
         super().__init__()
         self.transformer_model = transformer_model
         self.logger = self.transformer_model.logger
 
     @beartype
     def _copy_model(self):
-        """Copies the model.
-
-        This creates a deep copy of the model, typically for saving the
-        "best model". It temporarily removes the `log_file` attribute
-        before copying to avoid errors, then re-initializes it.
-
-        Returns:
-            A deep copy of the current TransformerModel instance.
-        """
+        """Deep-copy without copying the logger handle."""
         logger_ref = self.transformer_model.logger
         del self.transformer_model.logger
         del self.logger
@@ -599,14 +564,7 @@ def _copy_model(self):
 
     @conditional_beartype
     def forward(self, src: dict[str, Tensor], metadata: dict[str, Tensor]):
-        """Forward pass for the embedding model.
-
-        Args:
-            src: The input data.
-
-        Returns:
-            The embedded output.
-        """
+        """Return embedding output from the wrapped model."""
         return self.transformer_model.forward_embed(src, metadata=metadata)
 
 
@@ -627,30 +585,13 @@ def forward(self, *inputs: Tensor):
 
 
 class TransformerModel(nn.Module):
-    """The main Transformer model for the sequifier.
-
-    This class implements the Transformer model, including the training and
-    evaluation loops, as well as the export functionality.
-    """
+    """Sequifier transformer plus train/eval/export routines."""
 
     @beartype
     def __init__(
         self, hparams: Any, rank: Optional[int] = None, local_rank: Optional[int] = None
     ):
-        """Initializes the TransformerModel.
-
-        Based on the hyperparameters, this initializes:
-        - Embeddings for categorical and real features (self.encoder)
-        - Positional encoders (self.pos_encoder)
-        - The main TransformerEncoder (self.transformer_encoder)
-        - Output decoders for each target column (self.decoder)
-        - Loss functions (self.criterion)
-        - Optimizer (self.optimizer) and scheduler (self.scheduler)
-
-        Args:
-            hparams: The hyperparameters for the model (e.g., from TrainModel config).
-            rank: The rank of the current process (for distributed training).
-        """
+        """Build model modules and training state from config."""
         super().__init__()
         self.project_root = hparams.project_root
         self.model_type = "Transformer"
@@ -852,7 +793,7 @@ def __init__(
 
     @beartype
     def initialize_optimizer(self, params: Any = None) -> None:
-        """Initializes the optimizer and scheduler."""
+        """Create optimizer and scheduler from training config."""
         if params is None:
             params = self.parameters()
 
@@ -867,7 +808,7 @@ def initialize_optimizer(self, params: Any = None) -> None:
 
     @beartype
     def _apply_layer_dtypes(self) -> None:
-        """Casts specific layer types to configured dtypes (e.g., bfloat16, float8)."""
+        """Cast configured layer classes to requested dtypes."""
         layer_config = self.hparams.training_spec.layer_type_dtypes
 
         if not layer_config:
@@ -875,9 +816,7 @@ def _apply_layer_dtypes(self) -> None:
 
         self.logger.info(f"[INFO] Applying custom layer dtypes: {layer_config}")
 
-        # Iterate over all sub-modules and cast based on type
         for name, module in self.named_modules():
-            # Linear Layers
             if isinstance(module, nn.Linear):
                 is_decoder = any(module is m for m in self.decoder.values())
                 if is_decoder and "decoder" in layer_config:
@@ -885,12 +824,10 @@ def _apply_layer_dtypes(self) -> None:
                 elif "linear" in layer_config:
                     module.to(dtype=get_torch_dtype(layer_config["linear"]))
 
-            # Embeddings
             elif isinstance(module, nn.Embedding) and "embedding" in layer_config:
                 target_dtype = get_torch_dtype(layer_config["embedding"])
                 module.to(dtype=target_dtype)
 
-            # Normalization (RMSNorm, LayerNorm)
             elif isinstance(module, (nn.LayerNorm, RMSNorm)) and "norm" in layer_config:
                 target_dtype = get_torch_dtype(layer_config["norm"])
                 module.to(dtype=target_dtype)
@@ -903,15 +840,7 @@ def _apply_layer_dtypes(self) -> None:
 
     @beartype
     def _init_criterion(self, hparams: Any) -> ModuleDict:
-        """Initializes the criterion (loss function) for each target column.
-
-        Args:
-            hparams: The hyperparameters for the model, used to find criterion names
-                and class weights.
-
-        Returns:
-            A dictionary mapping target column names to their loss function instances.
-        """
+        """Build unreduced per-target loss modules."""
         criterion = ModuleDict()
         for target_column in self.target_columns:
             criterion_name = hparams.training_spec.criterion[target_column]
@@ -941,19 +870,7 @@ def _get_feature_embedding_dims(
         categorical_columns: list[str],
         real_columns: list[str],
     ) -> dict[str, int]:
-        """Calculates the embedding dimension for each column.
-
-        This attempts to distribute the total `embedding_size` across all
-        input columns.
-
-        Args:
-            embedding_size: The total embedding dimension (initial_embedding_dim).
-            categorical_columns: List of categorical column names.
-            real_columns: List of real-valued column names.
-
-        Returns:
-            A dictionary mapping column names to their calculated embedding dimension.
-        """
+        """Allocate embedding dimensions across homogeneous input columns."""
         if not (len(categorical_columns) + len(real_columns)) > 0:
             raise ValueError("No columns found")
 
@@ -1000,35 +917,17 @@ def _get_feature_embedding_dims(
 
     @staticmethod
     def _generate_square_subsequent_mask(sz: int) -> Tensor:
-        """Generates an upper-triangular matrix of -inf, with zeros on diag.
-
-        This is used as a mask to prevent attention to future tokens in the
-        transformer.
-
-        Args:
-            sz: The size of the square mask (sequence length).
-
-        Returns:
-            A square tensor of shape (sz, sz) with -inf in the upper triangle.
-        """
+        """Return a causal attention mask."""
         return torch.triu(torch.ones(sz, sz) * float("-inf"), diagonal=1)
 
     @staticmethod
     def _filter_key(dict_: dict[str, Any], key: str) -> dict[str, Any]:
-        """Filters a key from a dictionary.
-
-        Args:
-            dict_: The dictionary to filter.
-            key: The key to remove.
-
-        Returns:
-            A new dictionary without the specified key.
-        """
+        """Return a copy without key."""
         return {k: v for k, v in dict_.items() if k != key}
 
     @beartype
     def _init_weights(self) -> None:
-        """Initializes the weights of the model."""
+        """Initialize trainable weights with the model default."""
         init_std = 0.02
         for col in self.categorical_columns:
             self.encoder[col].weight.data.normal_(mean=0.0, std=init_std)
@@ -1048,18 +947,7 @@ def _init_weights(self) -> None:
 
     @conditional_beartype
     def _recursive_concat(self, srcs: list[Tensor]):
-        """Recursively concatenates a list of tensors.
-
-        This is used to avoid device-specific limits on the number of tensors
-        that can be concatenated at once by breaking the operation into
-        smaller, recursive chunks.
-
-        Args:
-            srcs: A list of tensors to concatenate along dimension 2.
-
-        Returns:
-            A single tensor resulting from the recursive concatenation.
-        """
+        """Concatenate tensors in chunks to avoid device concat limits."""
         if len(srcs) <= self.device_max_concat_length:
             return torch.cat(srcs, 2)
         else:
@@ -1106,31 +994,14 @@ def _build_attention_mask(self, valid_mask: Tensor, dtype: torch.dtype) -> Tenso
 
     @conditional_beartype
     def _zero_padding_positions(self, x: Tensor, valid_mask: Tensor) -> Tensor:
-        """
-        x shape:
-            (batch_size, context_length, dim_model)
-
-        Zeroes padded query positions so padded rows do not keep evolving.
-        """
+        """Zero padded query positions after attention/FFN layers."""
         return x * valid_mask[:, :, None].to(dtype=x.dtype)
 
     @conditional_beartype
     def forward_inner(
         self, src: dict[str, Tensor], metadata: dict[str, Tensor]
     ) -> Tensor:
-        """The inner forward pass of the model.
-
-        This handles embedding lookup, positional encoding, and passing the
-        combined tensor through the transformer encoder.
-
-        Args:
-            src: A dictionary mapping column names to input tensors
-                 (batch_size, context_length).
-
-        Returns:
-            The raw output tensor from the TransformerEncoder
-            (context_length, batch_size, dim_model).
-        """
+        """Encode inputs into contextual hidden states."""
         srcs = []
         for col in self.categorical_columns:
             src_t = self.encoder[col](src[col].T) * math.sqrt(
@@ -1210,37 +1081,14 @@ def forward_inner(
     def forward_embed(
         self, src: dict[str, Tensor], metadata: dict[str, Tensor]
     ) -> Tensor:
-        """Forward pass for the embedding model.
-
-        This returns only the embedding from the *last* token in the sequence.
-
-        Args:
-            src: A dictionary mapping column names to input tensors
-                 (batch_size, context_length).
-
-        Returns:
-            The embedding tensor for the last token
-            (batch_size, dim_model).
-        """
+        """Return final-step embeddings."""
         return self.forward_inner(src, metadata)[-self.prediction_length :, :, :]
 
     @conditional_beartype
     def forward_train(
         self, src: dict[str, Tensor], metadata: dict[str, Tensor]
     ) -> dict[str, Tensor]:
-        """Forward pass for training.
-
-        This runs the inner forward pass and then applies the appropriate
-        decoder for each target column.
-
-        Args:
-            src: A dictionary mapping column names to input tensors
-                 (batch_size, context_length).
-
-        Returns:
-            A dictionary mapping target column names to their raw output
-            (logit) tensors (context_length, batch_size, n_classes/1).
-        """
+        """Return raw decoded outputs for all target columns."""
         output = self.forward_inner(src, metadata)
         output = {
             target_column: self.decode(target_column, output)
@@ -1251,19 +1099,7 @@ def forward_train(
 
     @conditional_beartype
     def decode(self, target_column: str, output: Tensor) -> Tensor:
-        """Decodes the output of the transformer encoder.
-
-        Applies the appropriate final linear layer for a given target column.
-
-        Args:
-            target_column: The name of the target column to decode.
-            output: The raw output tensor from the TransformerEncoder
-                    (context_length, batch_size, dim_model).
-
-        Returns:
-            The decoded output (logits or real value) for the target column
-            (context_length, batch_size, n_classes/1).
-        """
+        """Project hidden states through one target decoder."""
 
         target_dtype = self.decoder[target_column].weight.dtype
         decoded = self.decoder[target_column](output.to(target_dtype)).to(torch.float32)
@@ -1272,18 +1108,7 @@ def decode(self, target_column: str, output: Tensor) -> Tensor:
 
     @conditional_beartype
     def apply_softmax(self, target_column: str, output: Tensor) -> Tensor:
-        """Applies softmax to the output of the decoder.
-
-        If the target is real, it returns the output unchanged.
-        If the target is categorical, it applies LogSoftmax.
-
-        Args:
-            target_column: The name of the target column.
-            output: The decoded output tensor (logits or real value).
-
-        Returns:
-            The output tensor, with LogSoftmax applied if categorical.
-        """
+        """Apply LogSoftmax only for categorical targets."""
         if target_column in self.real_columns:
             return output
         else:
@@ -1296,21 +1121,7 @@ def forward(
         metadata: dict[str, Tensor],
         return_logits: Union[bool, Tensor] = False,
     ) -> dict[str, Tensor]:
-        """The main forward pass of the model.
-
-        This is typically used for inference/evaluation, returning the
-        probabilities/values for the *last* token in the sequence.
-
-        Args:
-            src: A dictionary mapping column names to input tensors
-                 (batch_size, context_length).
-            return_logits: Return logits
-
-        Returns:
-            A dictionary mapping target column names to their final
-            output (LogSoftmax probabilities or real values) for the
-            last token (batch_size, n_classes/1).
-        """
+        """Return final-step logits or predictions for inference/eval."""
         output = self.forward_train(src, metadata)
         if return_logits:
             return output
@@ -1345,14 +1156,7 @@ def _get_full_state_dict(
 
     @beartype
     def _check_and_terminate(self):
-        """Checks for an external pruning signal and terminates the process if required.
-
-        This method looks for a specific `.prune` file generated by the Optuna orchestrator.
-        If running in a distributed setting, the rank 0 process checks for the file and
-        broadcasts a termination signal to all other ranks. If the signal is received,
-        the process cleans up its distributed process group, clears the GPU cache, and
-        gracefully exits with code 143 (SIGTERM) to allow Optuna to prune the trial.
-        """
+        """Exit 143 when rank 0 broadcasts an Optuna prune sentinel."""
         if os.getenv("SEQUIFIER_HYPERPARAMETER_SEARCH_RUN") is not None:
             should_prune = 0
             if self.rank == 0:
@@ -1388,16 +1192,7 @@ def train_model(
         valid_loader: DataLoader,
         ddp_model: Optional[nn.Module] = None,
     ) -> None:
-        """Trains the model.
-
-        This method contains the main training loop, including epoch iteration,
-        validation, early stopping logic, and model saving/exporting.
-
-        Args:
-            train_loader: DataLoader for the training dataset.
-            valid_loader: DataLoader for the validation dataset.
-            ddp_model: ddp model
-        """
+        """Run epochs, validation, checkpointing, export, and interruption cleanup."""
         self.logger.info(f"--- Starting Training for model: {self.model_name} ---")
 
         best_val_loss = float("inf")
@@ -1485,10 +1280,8 @@ def train_model(
             if self.hparams.training_spec.distributed:
                 dist.barrier()
 
-            # 1. Use a list to hold the answer so it can be broadcasted across ranks
             answer_list = ["n"]
 
-            # 2. Only Rank 0 prompts the user
             if self.rank == 0:
                 try:
                     answer = (
@@ -1503,11 +1296,9 @@ def train_model(
                 except EOFError:  # Handle non-interactive environments
                     answer_list[0] = "n"
 
-            # 3. Broadcast the decision to all GPUs so they stay in sync
             if self.hparams.training_spec.distributed:
                 dist.broadcast_object_list(answer_list, src=0)
 
-            # 4. If the decision is 'y', ALL ranks must participate in state dict extraction
             if answer_list[0] == "y":
                 if self.rank == 0:
                     self.logger.info("[INFO] User opted to export models.")
@@ -1518,10 +1309,9 @@ def train_model(
                             f"[INFO] Exporting 'last' model from epoch {last_epoch}..."
                         )
 
-                    # ALL RANKS MUST EXECUTE THIS to prevent FSDP all_gather deadlocks
+                    # FSDP state extraction is collective; only rank 0 writes the result.
                     last_model_state = self._get_full_state_dict(ddp_model)
 
-                    # ONLY Rank 0 executes the file I/O
                     if self.rank == 0:
                         self._export(last_model_state, "last", last_epoch)
 
@@ -1551,8 +1341,6 @@ def train_model(
                 )
             best_model_state = last_model_state
 
-        # 2. Restrict the export saving to Rank 0 inside the _export method (which you already do)
-        # or guard the I/O specifically:
         if self.rank == 0:
             self._export(last_model_state, "last", last_epoch, clean=True)  # type: ignore
             self._export(best_model_state, "best", last_epoch, clean=True)  # type: ignore
@@ -1569,16 +1357,7 @@ def _train_epoch(
         epoch: int,
         ddp_model: Optional[nn.Module] = None,
     ) -> None:
-        """Trains the model for one epoch.
-
-        Iterates through the training DataLoader, computes loss, performs
-        backpropagation, and updates model parameters. The DataLoader is expected
-        to yield SequifierBatch objects. IDs are currently unused in this loop.
-
-        Args:
-            train_loader: DataLoader for the training dataset.
-            epoch: The current epoch number (used for logging).
-        """
+        """Run one train epoch with optional mid-epoch saves."""
         total_loss = 0.0
         batches_aggregated = 0
 
@@ -1772,24 +1551,7 @@ def _calculate_loss(
         targets: dict[str, Tensor],
         metadata: dict[str, Tensor],
     ) -> tuple[Tensor, dict[str, Tensor]]:
-        """Calculates the loss for the given output and targets.
-
-        Compares the model's output (from `forward_train`) with the target
-        values, applying the appropriate criterion for each target column
-        and combining them using loss weights.
-
-        Args:
-            output: A dictionary of output tensors from the model
-                    (context_length, batch_size, n_classes/1).
-            targets: A dictionary of target tensors
-                     (batch_size, context_length).
-
-        Returns:
-            A tuple containing:
-            - The total combined (weighted) loss as a single Tensor.
-            - A dictionary of individual (unweighted) loss Tensors for each
-              target column.
-        """
+        """Return weighted mean loss and per-target components over valid tokens."""
         target_names = [k for k in targets.keys() if k in self.target_column_types]
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
@@ -1903,15 +1665,7 @@ def _calculate_loss_components(
 
     @beartype
     def _copy_model(self):
-        """Copies the model.
-
-        This creates a deep copy of the model, typically for saving the
-        "best model". It temporarily removes the `log_file` attribute
-        before copying to avoid errors, then re-initializes it.
-
-        Returns:
-            A deep copy of the current TransformerModel instance.
-        """
+        """Deep-copy without copying the logger handle."""
         logger_ref = self.logger
         del self.logger
         model_copy = copy.deepcopy(self)
@@ -1921,20 +1675,7 @@ def _copy_model(self):
 
     @beartype
     def _transform_val(self, col: str, val: Tensor) -> Tensor:
-        """ "Transforms input data to match the format of model output.
-
-        This is used *only* for calculating the baseline loss, where
-        the input (e.g., categorical indices) needs to be one-hot encoded
-        to be comparable to the model's (logit) output.
-
-        Args:
-            col: The name of the column being transformed.
-            val: The input tensor (categorical indices).
-
-        Returns:
-            A tensor transformed to be compatible with the loss function
-            (e.g., one-hot encoded).
-        """
+        """Transform targets into baseline-loss output shape."""
         if self.target_column_types[col] == "categorical":
             target_dtype = self.decoder[col].weight.dtype
             return (
@@ -1951,26 +1692,7 @@ def _transform_val(self, col: str, val: Tensor) -> Tensor:
     def _evaluate(
         self, valid_loader: DataLoader, ddp_model: Optional[nn.Module] = None
     ) -> tuple[np.float32, dict[str, np.float32], ClassCounts]:
-        """Evaluates the model on the validation set.
-
-        Iterates through the validation data, calculates the total loss,
-        and aggregates results across all processes if in distributed mode.
-        Also calculates a one-time baseline loss on the first call. The
-        DataLoader is expected to yield SequifierBatch objects. IDs are
-        currently unused during evaluation. Class shares are counted over
-        the same valid token population as validation loss; for BERT this
-        means masked evaluation positions, not all sequence positions.
-
-        Args:
-            valid_loader: DataLoader for the validation dataset.
-            ddp_model: DDP model
-
-        Returns:
-            A tuple containing:
-            - The total aggregated validation loss (float).
-            - A dictionary of aggregated losses for each target column (dict[str, float]).
-            - A dictionary of globally aggregated class-count tensors.
-        """
+        """Evaluate validation loss and optional class-share counts."""
 
         model_to_call = ddp_model if ddp_model is not None else self
         target_names = list(self.target_columns)
@@ -2258,16 +1980,7 @@ def _export(
         epoch: int,
         clean: bool = False,
     ) -> None:
-        """Exports the model.
-
-        This is a wrapper function that handles exporting the model (and
-        optionally the embedding-only model) on rank 0 only.
-
-        Args:
-            state_dict: The state dict of the model instance to export (e.g., best model or last model).
-            suffix: A string suffix to append to the model filename (e.g., "best", "last").
-            epoch: The current epoch number, included in the filename.
-        """
+        """Export configured model variants from rank 0."""
         if self.rank != 0:
             return
 
@@ -2297,16 +2010,7 @@ def _export_model(
         suffix: str,
         epoch: int,
     ) -> None:
-        """Exports the model to ONNX and/or PyTorch format.
-
-        Saves the model weights as a .pt file and/or exports the model
-        graph and weights as an .onnx file based on the config.
-
-        Args:
-            model: The model instance (TransformerModel or TransformerEmbeddingModel).
-            suffix: A string suffix for the filename (e.g., "best", "last-embedding").
-            epoch: The current epoch number, included in the filename.
-        """
+        """Write one model as ONNX and/or PT."""
         os.makedirs(os.path.join(self.project_root, "models"), exist_ok=True)
 
         if self.export_onnx:
@@ -2388,7 +2092,6 @@ def _export_model(
                 logging.getLogger("torch.onnx").setLevel(logging.ERROR)
             except (ImportError, AttributeError):
                 torch.onnx.disable_log()  # Fallback for older PyTorch versions
-            # 2. Catch and ignore standard Python warnings temporarily
             with warnings.catch_warnings(), open(
                 os.devnull, "w"
             ) as fnull, contextlib.redirect_stdout(fnull), contextlib.redirect_stderr(
@@ -2437,16 +2140,7 @@ def _save(
         ddp_model: Optional[nn.Module] = None,
         suffix: Optional[str] = None,
     ) -> None:
-        """Saves the model checkpoint.
-
-        Saves the model state, optimizer state, and epoch number to a .pt
-        file in the checkpoints directory. Only runs on rank 0.
-
-        Args:
-            val_loss: The validation loss at the current epoch.
-            ddp_model: DDP model
-            suffix: Checkpoint file suffix.
-        """
+        """Save rank-0 checkpoint state."""
         model_to_extract = ddp_model if ddp_model is not None else self
 
         if self.hparams.training_spec.data_parallelism == "FSDP":
@@ -2498,18 +2192,7 @@ def _save(
 
     @beartype
     def _get_optimizer(self, params: Any, **kwargs):
-        """Gets the optimizer.
-
-        Initializes the optimizer specified in the hyperparameters.
-
-        Args:
-            params: params
-            **kwargs: Additional arguments to pass to the optimizer constructor
-                      (e.g., weight_decay).
-
-        Returns:
-            An initialized torch.optim.Optimizer instance.
-        """
+        """Instantiate the configured optimizer."""
         optimizer_class = get_optimizer_class(self.hparams.training_spec.optimizer.name)
         return optimizer_class(
             params, lr=self.hparams.training_spec.learning_rate, **kwargs
@@ -2517,17 +2200,7 @@ def _get_optimizer(self, params: Any, **kwargs):
 
     @beartype
     def _get_scheduler(self, **kwargs):
-        """Gets the scheduler.
-
-        Initializes the learning rate scheduler specified in the hyperparameters.
-
-        Args:
-            **kwargs: Additional arguments to pass to the scheduler constructor
-                      (e.g., step_size).
-
-        Returns:
-            An initialized torch.optim.lr_scheduler._LRScheduler instance.
-        """
+        """Instantiate the configured LR scheduler."""
         scheduler_name = self.hparams.training_spec.scheduler.name
         if hasattr(torch.optim.lr_scheduler, scheduler_name):
             scheduler_class = getattr(torch.optim.lr_scheduler, scheduler_name)
@@ -2539,21 +2212,12 @@ def _get_scheduler(self, **kwargs):
 
     @beartype
     def _initialize_log_file(self):
-        """Initializes the log file."""
-        # Replaces old LogFile class instantiation
+        """Attach the configured logger."""
         self.logger = configure_logger(self.project_root, self.model_name, self.rank)
 
     @beartype
     def _get_latest_model_name(self) -> Optional[str]:
-        """Gets the name of the latest model checkpoint.
-
-        Scans the checkpoints directory for files matching the current
-        `model_name` and returns the path to the most recently modified one.
-
-        Returns:
-            The file path (str) to the latest checkpoint, or None if no
-            checkpoint is found.
-        """
+        """Return the newest checkpoint path for this model name."""
         checkpoint_path = os.path.join(self.project_root, "checkpoints", "*")
 
         files = glob.glob(checkpoint_path)
@@ -2576,21 +2240,7 @@ def _log_epoch_results(
         class_counts: ClassCounts,
         global_step: int,
     ) -> None:
-        """Logs the results of an epoch.
-
-        Writes validation loss, individual losses, learning rate, and
-        class share statistics (if configured) to the log file.
-        Only runs on rank 0.
-
-        Args:
-            epoch: Current epoch number.
-            elapsed: Time taken for the epoch (in seconds).
-            total_loss: The total aggregated validation loss.
-            total_losses: A dictionary of aggregated losses for each target.
-            class_counts: Globally aggregated class-count tensors used for
-                class share logging.
-            batch: Current batch number.
-        """
+        """Log validation metrics and class shares from rank 0."""
         if self.rank == 0:
             learning_rate = self.optimizer.state_dict()["param_groups"][0]["lr"]
 
@@ -2661,21 +2311,7 @@ def load_inference_model(
     device: str,
     infer_with_dropout: bool,
 ) -> torch.nn.Module:
-    """Loads a trained model for inference.
-
-    Args:
-        model_type: "generative" or "embedding".
-        model_path: Path to the saved .pt model file.
-        training_config_path: Path to the .yaml config file used for training.
-        args_config: A dictionary of override configurations.
-        device: The device to load the model onto (e.g., "cuda", "cpu").
-        infer_with_dropout: Whether to force dropout layers to be active
-                          during inference.
-
-    Returns:
-        The loaded and compiled torch.nn.Module (TransformerModel or
-        TransformerEmbeddingModel) in evaluation mode.
-    """
+    """Load a PT checkpoint as a generative or embedding inference module."""
     skip_metadata = args_config.get("skip_metadata", False)
     args_config_subset = {
         k: v for k, v in args_config.items() if k not in ["model_path", "data_path"]
@@ -2730,18 +2366,7 @@ def infer_with_embedding_model(
     target_columns: list[str],
     metadata: list[dict[str, np.ndarray]],
 ) -> np.ndarray:
-    """Performs inference with an embedding model.
-
-    Args:
-        model: The loaded TransformerEmbeddingModel.
-        x: A list of input data dictionaries (batched).
-        device: The device to run inference on.
-        size: The total number of samples (unused in this function).
-        target_columns: List of target column names (unused in this function).
-
-    Returns:
-        A NumPy array containing the concatenated embeddings from all batches.
-    """
+    """Run batched embedding inference and concatenate CPU outputs."""
     outs0 = []
 
     categorical_cols = set(model.transformer_model.categorical_columns)
@@ -2790,19 +2415,7 @@ def infer_with_generative_model(
     target_columns: list[str],
     metadata: list[dict[str, np.ndarray]],
 ) -> dict[str, np.ndarray]:
-    """Performs inference with a generative model.
-
-    Args:
-        model: The loaded TransformerModel.
-        x: A list of input data dictionaries (batched).
-        device: The device to run inference on.
-        size: The total number of samples to trim the final output to.
-        target_columns: List of target column names to extract from the output.
-
-    Returns:
-        A dictionary mapping target column names to their concatenated
-        output NumPy arrays, trimmed to `size`.
-    """
+    """Run batched generative inference and trim CPU outputs."""
     outs0 = []
 
     categorical_cols = set(model.categorical_columns)
diff --git a/src/sequifier/visualize_training.py b/src/sequifier/visualize_training.py
index fb459ced..c888a0f3 100644
--- a/src/sequifier/visualize_training.py
+++ b/src/sequifier/visualize_training.py
@@ -6,19 +6,14 @@
 from typing import Any, Optional
 
 import numpy as np
-import plotly.colors as pc  # Added to fetch consistent colors
+import plotly.colors as pc
 import plotly.graph_objects as go
 from beartype import beartype
-
-# Import Loguru and your custom logger config
 from loguru import logger
 from plotly.subplots import make_subplots
 
 from sequifier.helpers import configure_logger
 
-# -------------------------------------------------------------------------
-# Configuration & Setup
-# -------------------------------------------------------------------------
 VAL_PATTERN = re.compile(
     r"\[INFO\] Validation\s+\|\s*Epoch:\s*(\d+)\s+\|\s*Batch:\s*(\d+)\s+\|\s*Loss:\s*([^\s\|]+)\s+\|\s*Baseline Loss:\s*([^\s\|]+)"
 )
@@ -28,24 +23,21 @@
 )
 
 
-# -------------------------------------------------------------------------
-# Custom Exceptions & Dataclasses
-# -------------------------------------------------------------------------
 class LogParsingError(Exception):
-    """Raised when a log line does not conform to the expected regex pattern."""
+    """Malformed training log line."""
 
     pass
 
 
 class DataContinuityError(Exception):
-    """Raised when training batches or epochs violate chronological order."""
+    """Non-monotonic training batch or epoch sequence."""
 
     pass
 
 
 @dataclass
 class TrainingMetrics:
-    """Encapsulates all extracted metrics to avoid returning massive generic tuples."""
+    """Parsed validation, baseline, variable, and training losses."""
 
     val_losses: dict[float, float] = field(default_factory=dict)
     baseline_losses: dict[float, float] = field(default_factory=dict)
@@ -55,18 +47,15 @@ class TrainingMetrics:
     )
 
     def clear_state(self) -> None:
-        """Clears all metrics; used when a sequence run restarts."""
+        """Clear parsed metrics after a detected run restart."""
         self.val_losses.clear()
         self.baseline_losses.clear()
         self.var_losses.clear()
         self.train_losses.clear()
 
 
-# -------------------------------------------------------------------------
-# Core Parsing Logic
-# -------------------------------------------------------------------------
 class LogParser:
-    """Handles line-by-line log parsing, encapsulating state to reduce complexity."""
+    """Stateful parser for sequifier training logs."""
 
     def __init__(self, model_name: str):
         self.model = model_name
@@ -90,7 +79,7 @@ def parse_file(self, log_file: str) -> TrainingMetrics:
 
     @beartype
     def _process_line(self, line: str) -> None:
-        """Routes the line to the appropriate sub-parser based on strict string matching."""
+        """Dispatch one log line to the matching parser."""
         if "[INFO] Validation | Epoch:" in line:
             self._process_validation(line)
         elif self.pending_var_loss_epoch is not None and "[INFO]  - " in line:
@@ -111,7 +100,6 @@ def _process_validation(self, line: str) -> None:
         val_loss = parse_number(match.group(3))
         baseline = parse_number(match.group(4))
 
-        # Check for restart (epoch 0, batch 0) or standard jump back
         if (epoch == 0 and batch == 0) or (
             self.current_epoch is not None and epoch < self.current_epoch and epoch != 0
         ):
@@ -120,17 +108,14 @@ def _process_validation(self, line: str) -> None:
             self.current_batch = None
             self.expected_num_batches = None
 
-        # Determine precise x-coordinate for plotting
         if (
             epoch == 0
             and batch > 0
             and self.current_epoch is not None
             and self.expected_num_batches is not None
         ):
-            # Mid-epoch validation saves pass epoch as 0 in train.py
             calc_epoch = self.current_epoch - 1 + (batch / self.expected_num_batches)
         elif self.expected_num_batches is not None and batch > 0:
-            # End-of-epoch validation
             calc_epoch = epoch - 1 + (batch / self.expected_num_batches)
         else:
             calc_epoch = float(epoch)
@@ -215,7 +200,7 @@ def _validate_chronology(self, epoch: int, batch: int, num_batches: int) -> None
                 )
 
     def _handle_epoch_1_restart(self) -> None:
-        """Handles edge cases where a sequence restarts at Epoch 1 skipping Epoch 0."""
+        """Handle restarts that resume at epoch 1 without epoch 0 logs."""
         if 0.0 not in self.metrics.val_losses:
             self.metrics.clear_state()
         else:
@@ -243,19 +228,16 @@ def _validate_final_metrics(self) -> None:
             raise DataContinuityError(f"[{self.model}]: No baseline loss data found.")
 
 
-# -------------------------------------------------------------------------
-# Utility Functions
-# -------------------------------------------------------------------------
 @beartype
 def parse_number(val: str) -> float:
-    """Strictly parse numbers, explicitly handling the 'NaN' strings."""
+    """Parse finite floats and literal NaN."""
     val = val.strip()
     return np.nan if val == "NaN" else float(val)
 
 
 @beartype
 def parse_args_to_models(args: argparse.Namespace) -> list[str]:
-    """Extracts the list of models from a file or comma-separated string."""
+    """Read model names from a file or comma-separated argument."""
     if os.path.isfile(args.models) and args.models.endswith(".txt"):
         with open(args.models, "r") as f:
             content = f.read()
@@ -266,7 +248,7 @@ def parse_args_to_models(args: argparse.Namespace) -> list[str]:
 
 @beartype
 def get_log_filepath(args: argparse.Namespace, model: str) -> str:
-    """Finds the appropriate log file for a given model."""
+    """Return the rank-0 log path for a model."""
     log_pattern = os.path.join(
         args.project_root, "logs", f"sequifier-{model}-rank0-3.txt"
     )
@@ -290,7 +272,7 @@ def get_log_filepath(args: argparse.Namespace, model: str) -> str:
 def format_plot_data(
     metrics: TrainingMetrics, bucket_batches: Optional[int], model: str
 ) -> dict[str, Any]:
-    """Formats raw parsed dataclass metrics into chronological arrays for Plotly."""
+    """Convert parsed metrics into Plotly-ready arrays."""
     val_x = sorted(list(metrics.val_losses.keys()))
     val_y = [metrics.val_losses[e] for e in val_x]
     base_y = [metrics.baseline_losses[e] for e in val_x]
@@ -347,14 +329,11 @@ def format_plot_data(
     }
 
 
-# -------------------------------------------------------------------------
-# Plotting & Reporting
-# -------------------------------------------------------------------------
 @beartype
 def _generate_single_model_plot(
     model: str, data: dict[str, Any], yaxis_type: str, out_path: str
 ) -> None:
-    """Handles subplot logic specifically for a single model."""
+    """Write a single-model training report."""
     has_var_losses = bool(data.get("var_losses"))
     subplot_titles = (
         ("Global Losses", "Normalized Variable Validation Losses")
@@ -446,18 +425,17 @@ def _generate_single_model_plot(
 def _generate_multi_model_plot(
     models: list[str], all_data: dict[str, Any], yaxis_type: str, out_path: str
 ) -> None:
-    """Handles subplot logic for comparing multiple models side-by-side."""
+    """Write a multi-model training report."""
     fig = make_subplots(
         rows=1, cols=2, subplot_titles=("Validation Losses", "Training Losses")
     )
     baseline_val = None
-    colors = pc.qualitative.Plotly  # Load Plotly's default distinct color array
+    colors = pc.qualitative.Plotly
 
     for i, model in enumerate(models):
         data = all_data[model]
-        color = colors[i % len(colors)]  # Cycle colors if models exceed palette limit
+        color = colors[i % len(colors)]
 
-        # Validation trace
         fig.add_trace(
             go.Scatter(
                 x=data["val_x"],
@@ -466,14 +444,13 @@ def _generate_multi_model_plot(
                 name=model,
                 legendgroup=model,
                 line=dict(color=color),
-                showlegend=True,  # Only show validation trace in legend to prevent duplicates
+                showlegend=True,
                 hovertemplate=f"<b>{model}</b><br>Val Loss: %{{y}}<br>Epoch: %{{x}}<extra></extra>",
             ),
             row=1,
             col=1,
         )
 
-        # Training trace
         fig.add_trace(
             go.Scatter(
                 x=data["train_x"],
@@ -482,7 +459,7 @@ def _generate_multi_model_plot(
                 name=model,
                 legendgroup=model,
                 line=dict(color=color),
-                showlegend=False,  # Hidden from legend, but linked via legendgroup
+                showlegend=False,
                 hovertemplate=f"<b>{model}</b><br>Train Loss: %{{y}}<br>Epoch: %{{x}}<extra></extra>",
             ),
             row=1,
@@ -504,7 +481,6 @@ def _generate_multi_model_plot(
                 )
 
     if baseline_val is not None:
-        # Plot baseline on the Validation subplot (col=1)
         max_val_x = max(
             [max(all_data[m]["val_x"]) for m in models if all_data[m]["val_x"]] + [0]
         )
@@ -534,7 +510,7 @@ def _generate_multi_model_plot(
 def generate_html_report(
     all_data: dict[str, Any], models: list[str], args: argparse.Namespace
 ) -> None:
-    """Router function to generate the appropriate HTML report based on model count."""
+    """Write the model-count-appropriate HTML report."""
     output_dir = os.path.join(args.project_root, "outputs", "visualization")
     os.makedirs(output_dir, exist_ok=True)
 
@@ -549,12 +525,9 @@ def generate_html_report(
         _generate_multi_model_plot(models, all_data, yaxis_type, out_path)
 
 
-# -------------------------------------------------------------------------
-# Orchestrator
-# -------------------------------------------------------------------------
 @beartype
 def visualize_training(args: argparse.Namespace) -> None:
-    """Main orchestrator function."""
+    """Parse logs and write training visualization HTML."""
     models = parse_args_to_models(args)
     if not models:
         raise ValueError("No models provided to visualize.")
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index bce3bddd..5cc4ec89 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -133,13 +133,11 @@ def test_predictions_cat(predictions):
                 baseline_preds = predictions["model-categorical-1-best-3"]
                 n_baseline_rows = baseline_preds.height
 
-                # 3. Assert the number of rows is scaled by prediction_length
                 assert n_test_rows == n_baseline_rows * prediction_length, (
                     f"Expected {n_baseline_rows * prediction_length} rows for prediction_length={prediction_length}, "
                     f"but found {n_test_rows} rows."
                 )
 
-                # 4. Assert correct number of predictions per sequence
                 baseline_rows_per_seq = (
                     baseline_preds.group_by("sequenceId").len().height
                 )
@@ -291,12 +289,8 @@ def test_predictions_item_position(predictions):
 
 
 def test_embeddings_subsequence_id(embeddings):
-    """
-    Checks if subsequenceId increments correctly within each sequenceId
-    and starts at 0 for each new sequence.
-    """
+    """Check subsequenceId increments within each sequence."""
     for model_name, embeds_df in embeddings.items():
-        # Ensure correct sorting
         embeds_df_sorted = embeds_df.sort("sequenceId", "subsequenceId")
 
         shift_val = 0
@@ -315,17 +309,15 @@ def test_embeddings_subsequence_id(embeddings):
             ),
         )
 
-        # 1. Check if subsequenceId starts at 0 for each new sequence
         first_subsequences = embeds_with_diffs.filter(
             pl.col("same_seq") == False  # noqa: E712
-        )  # First row of each sequence
+        )
         incorrect_starts = first_subsequences.filter(pl.col("subsequenceId") != 0)
         assert incorrect_starts.height == 0, (
             f"Model '{model_name}': Found sequences where subsequenceId does not start at 0:\n"
             f"{incorrect_starts}"
         )
 
-        # 2. Check if subsequenceId increments by 1 within sequences
         within_sequence_diffs = embeds_with_diffs.filter(pl.col("same_seq"))
         incorrect_increments = within_sequence_diffs.filter(pl.col("subseq_diff") != 1)
         assert incorrect_increments.height == 0, (
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
index 708dc24d..a214e7ef 100644
--- a/tests/integration/test_preprocessing.py
+++ b/tests/integration/test_preprocessing.py
@@ -75,7 +75,7 @@ def test_metadata_config(metadata_configs):
 
 
 def load_parquet_folder_outputs(path):
-    """Reads a directory of Parquet chunks into a single sorted Polars DataFrame."""
+    """Sorted frame from Parquet chunks."""
     # Polars natively supports reading all matching files via glob patterns
     data = pl.read_parquet(os.path.join(path, "*.parquet"))
 
@@ -433,10 +433,7 @@ def test_preprocessing_from_precomputed_stats(
     preprocessing_config_path_cat_precomputed_stats,
     preprocessing_config_path_cat_precomputed_stats_negative,
 ):
-    """
-    Tests that providing a preexisting metadata config produces identical
-    outputs to the standard run, bypassing metadata computation.
-    """
+    """Compare precomputed-stat and standard preprocessing outputs."""
 
     run_and_log(
         f"sequifier preprocess --config-path {preprocessing_config_path_cat_precomputed_stats}"
diff --git a/tests/integration/test_visualize_training.py b/tests/integration/test_visualize_training.py
index 5190d7a2..a619748a 100644
--- a/tests/integration/test_visualize_training.py
+++ b/tests/integration/test_visualize_training.py
@@ -25,8 +25,6 @@ def test_visualize_training(
     hp_models_grid = sorted([m for m in models if "hp-search-grid-run" in m])
     assert len(hp_models_grid) > 0, "No hp grid models found in logs directory"
 
-    # 1. Test running visualize-training for each model individually
-    # model_outputs = {}
     for model in single_models:
         # visualize_training.py expects to find "logs/" relative to the working directory
         command = f"sequifier visualize-training {model} --project-root {project_root}"
@@ -42,7 +40,6 @@ def test_visualize_training(
             output_path
         ), f"Visualization output not found for {model}"
 
-    # 2. Test running visualize-training for all models jointly
     models_str = ",".join(hp_models_grid)
     command_joint = f"sequifier visualize-training {models_str} --project-root {project_root} --log-scale --bucket-training-batches 5"
     run_and_log(command_joint)
diff --git a/tests/resources/source_scripts/hp_search_eval_script.py b/tests/resources/source_scripts/hp_search_eval_script.py
index 9381766b..899783bd 100644
--- a/tests/resources/source_scripts/hp_search_eval_script.py
+++ b/tests/resources/source_scripts/hp_search_eval_script.py
@@ -12,16 +12,13 @@ def main():
 
     run_name = sys.argv[1]
 
-    # 3. Load predictions and calculate mean & variance
     preds_path = f"outputs/predictions/sequifier-{run_name}-predictions"
 
     dfs = []
     for root, dir, files in os.walk(preds_path):
         for file in sorted(list(files)):
-            # 1. Read everything as strings to avoid read-time schema crashes
             df = pl.read_csv(os.path.join(preds_path, file), infer_schema_length=0)
 
-            # 2. Cast to Int64 (strict=False turns bad strings to null) & fill nulls with -1
             df = df.with_columns(pl.all().cast(pl.Int64, strict=False).fill_null(-1))
             dfs.append(df)
     df = pl.concat(dfs)
@@ -29,7 +26,6 @@ def main():
     max_freqs = df["itemId"].value_counts()["count"].max()
     stdev_freqs = df["itemId"].value_counts()["count"].std()
 
-    # 4. Save metrics back for Optuna to ingest
     eval_dir = "outputs/evaluations"
     os.makedirs(eval_dir, exist_ok=True)
     eval_json_path = os.path.join(eval_dir, f"{run_name}.json")
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 6fb7b0b5..3d61a5ed 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -85,7 +85,7 @@ def dataset_path(tmp_path):
 
 
 def test_initialization(mock_config, dataset_path):
-    """Tests that metadata is read correctly and __len__ calculates batches."""
+    """Metadata-backed batch length."""
     dataset = SequifierDatasetFromFolderParquetLazy(dataset_path, mock_config)
 
     # 40 total samples / batch size of 5 = 8 batches
@@ -95,7 +95,7 @@ def test_initialization(mock_config, dataset_path):
 
 
 def test_iteration_yields_correct_batches(mock_config, dataset_path):
-    """Tests that the dataset iterates over files and yields structured tensors."""
+    """Structured tensors from file iteration."""
     dataset = SequifierDatasetFromFolderParquetLazy(
         dataset_path, mock_config, shuffle=False
     )
@@ -180,7 +180,7 @@ def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
 @patch("torch.distributed.get_rank", return_value=0)
 @patch("torch.distributed.get_world_size", return_value=2)
 def test_distributed_sharding(mock_ws, mock_rank, mock_init, mock_config, dataset_path):
-    """Tests that the dataset correctly shards files across distributed GPU ranks."""
+    """Distributed file sharding."""
     dataset = SequifierDatasetFromFolderParquetLazy(
         dataset_path, mock_config, shuffle=False
     )
@@ -196,7 +196,7 @@ def test_distributed_sharding(mock_ws, mock_rank, mock_init, mock_config, datase
 
 
 def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
-    """Tests that FSDP validation raises an Exception when file distribution is uneven."""
+    """FSDP exact mode rejects uneven ranks."""
     data_dir = tmp_path / "uneven_parquet_data"
     data_dir.mkdir()
 
@@ -219,7 +219,7 @@ def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
 
 
 def test_oversampling_strategy(mock_config, tmp_path):
-    """Tests that shorter ranks oversample to equal the maximal rank count."""
+    """Oversampling pads short ranks."""
     data_dir = tmp_path / "oversample_parquet_data"
     data_dir.mkdir()
 
@@ -264,7 +264,7 @@ def test_oversampling_strategy(mock_config, tmp_path):
 
 
 def test_undersampling_strategy(mock_config, tmp_path):
-    """Tests that longer ranks truncate samples down to match the minimal rank count."""
+    """Undersampling truncates long ranks."""
     data_dir = tmp_path / "undersample_parquet_data"
     data_dir.mkdir()
 
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 215ddc99..992bcf74 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -27,7 +27,7 @@ def _folder_metadata(total_samples, batch_files):
 # --- Fixtures ---
 @pytest.fixture
 def mock_config(tmp_path):
-    """Creates a mock configuration object."""
+    """Mock dataset config."""
     config = MagicMock()
     config.project_root = str(tmp_path)
     config.training_spec.batch_size = 5
@@ -95,7 +95,7 @@ def side_effect(path, map_location, weights_only):
 
 
 def test_initialization(mock_config, dataset_path):
-    """Tests that metadata is read correctly and __len__ calculates batches."""
+    """Metadata-backed batch length."""
     dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config)
 
     # 40 total samples / batch size of 5 = 8 batches
@@ -105,7 +105,7 @@ def test_initialization(mock_config, dataset_path):
 
 
 def test_iteration_yields_correct_batches(mock_config, dataset_path, mock_torch_load):
-    """Tests that the dataset iterates over files and yields correct tensor slices."""
+    """Tensor slices from file iteration."""
     dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config, shuffle=False)
 
     # Consume the generator
@@ -186,7 +186,7 @@ def side_effect(path, map_location, weights_only):
 def test_distributed_sharding(
     mock_rank, mock_ws, mock_init, mock_config, dataset_path, mock_torch_load
 ):
-    """Tests that the dataset correctly shards files across distributed GPUs."""
+    """Distributed file sharding."""
     dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config, shuffle=False)
 
     # World size = 2, Total files = 4
@@ -210,7 +210,7 @@ def test_distributed_sharding(
 def test_dataloader_worker_sharding_continuous_boundaries(
     mock_worker_info, mock_config, dataset_path, mock_torch_load
 ):
-    """Tests that continuous sample streaming assigns correct file boundaries to workers."""
+    """Worker file boundaries for continuous streaming."""
 
     # Simulate being DataLoader worker ID 1 out of 2 total workers
     mock_info = MagicMock()
@@ -244,7 +244,7 @@ def test_dataloader_worker_sharding_continuous_boundaries(
 def test_exact_strategy_uneven_files_exception(
     mock_rank, mock_ws, mock_init, mock_config, tmp_path
 ):
-    """Tests that 'exact' strategy raises the detailed exception on uneven rank samples."""
+    """Exact mode rejects uneven rank samples."""
 
     data_dir = tmp_path / "data_uneven"
     data_dir.mkdir()
@@ -281,7 +281,7 @@ def test_exact_strategy_uneven_files_exception(
 def test_oversampling_strategy(
     mock_rank, mock_ws, mock_init, mock_config, tmp_path, mock_torch_load
 ):
-    """Tests that 'oversampling' pads a rank with fewer samples by looping its files."""
+    """Oversampling loops short ranks."""
     data_dir = tmp_path / "data_oversample"
     data_dir.mkdir()
 
@@ -327,7 +327,7 @@ def test_oversampling_strategy(
 def test_undersampling_strategy(
     mock_rank, mock_ws, mock_init, mock_config, tmp_path, mock_torch_load
 ):
-    """Tests that 'undersampling' truncates a rank with more samples to match the lightest rank."""
+    """Undersampling truncates heavy ranks."""
     data_dir = tmp_path / "data_undersample"
     data_dir.mkdir()
 
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index 46bea614..53856f92 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -13,13 +13,9 @@
     resolve_window_view,
 )
 
-# ==========================================
-# 1. Test Index Map Construction
-# ==========================================
-
 
 def test_construct_index_maps_string():
-    """Tests reversing a string-to-int map, ensuring 0 maps to '[unknown]'."""
+    """Reverse string ID map."""
     id_maps: dict[str, dict[str | int, int]] = {"itemId": {"apple": 3, "banana": 4}}
     target_cols = ["itemId"]
 
@@ -36,7 +32,7 @@ def test_construct_index_maps_string():
 
 
 def test_construct_index_maps_integer():
-    """Tests reversing an int-to-int map with special IDs as sentinels."""
+    """Reverse int ID map with sentinels."""
     id_maps: dict[str, dict[str | int, int]] = {"storeId": {100: 3, 101: 4}}
     target_cols = ["storeId"]
 
@@ -51,29 +47,15 @@ def test_construct_index_maps_integer():
 
 
 def test_construct_index_maps_flag_false():
-    """Tests that nothing is returned if map_to_id is False."""
+    """Empty maps when map_to_id is false."""
     id_maps: dict[str, dict[str | int, int]] = {"itemId": {"a": 1}}
     result = construct_index_maps(id_maps, ["itemId"], map_to_id=False)
     assert result == {}
 
 
-# ==========================================
-# 2. Test Numpy to PyTorch Conversion
-# ==========================================
-
-
 def test_numpy_to_pytorch_shapes_and_shifting():
-    """
-    Tests that DataFrames are correctly converted to input and target Tensors.
-
-    Logic to test:
-    If context_length = 3:
-    - Input cols:  ['3', '2', '1']
-    - Target cols: ['2', '1', '0']
-    """
+    """Check causal input and shifted target tensors."""
 
-    # Setup: 2 sequences for feature "A"
-    # Seq 1: [10, 20, 30, 40] (where 40 is t=0, 30 is t=1, etc.)
     data = pl.DataFrame(
         {
             "sequenceId": [1, 2],
@@ -102,7 +84,6 @@ def test_numpy_to_pytorch_shapes_and_shifting():
         resolved_view,
     )
 
-    # 1. Check Keys
     assert "A" in tensors
     assert "A_target" in tensors
     assert torch.equal(
@@ -114,19 +95,15 @@ def test_numpy_to_pytorch_shapes_and_shifting():
         torch.tensor([[True, True, True], [True, True, True]]),
     )
 
-    # 2. Check Input Tensor (Cols 3, 2, 1)
-    # Row 0: [10, 20, 30]
     expected_input = torch.tensor([[10, 20, 30], [11, 21, 31]], dtype=torch.float32)
     assert torch.equal(tensors["A"], expected_input)
 
-    # 3. Check Target Tensor (Cols 2, 1, 0) -> Shifted by 1 step into future
-    # Row 0: [20, 30, 40]
     expected_target = torch.tensor([[20, 30, 40], [21, 31, 41]], dtype=torch.float32)
     assert torch.equal(tensors["A_target"], expected_target)
 
 
 def test_numpy_to_pytorch_dtypes():
-    """Tests that column_types dict is respected for tensor dtypes."""
+    """Check column_types controls tensor dtypes."""
 
     # Note: numpy_to_pytorch filters by inputCol, so we need to handle how
     # it selects rows. It assumes all rows matching "inputCol" have valid data
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 9afb97ea..cb15f9db 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -21,7 +21,7 @@
 
 
 def test_normalize():
-    """Tests the softmax normalization of raw logits."""
+    """Softmax-normalized logits."""
     # Create raw logits
     # Row 0: exp(0)=1, exp(0)=1 -> probs: 0.5, 0.5
     # Row 1: exp(1)=e, exp(2)=e^2 -> probs: e/(e+e^2), e^2/(e+e^2)
@@ -47,7 +47,7 @@ def test_normalize():
 
 @patch("numpy.random.rand")
 def test_sample_with_cumsum(mock_rand):
-    """Tests inverse CDF sampling with both raw logits and pure probabilities."""
+    """Inverse-CDF sampling for logits and probabilities."""
     # Mock the random thresholds to strictly control the sampling outcome.
     mock_rand.return_value = np.array([[0.05], [0.90]])
 
@@ -92,7 +92,7 @@ def mock_inferer():
 
 
 def test_inferer_invert_normalization(mock_inferer):
-    """Tests that normalized real outputs are scaled back correctly."""
+    """Real outputs denormalize correctly."""
     # Normalized values
     values = np.array([-1.0, 0.0, 1.0])
 
@@ -105,7 +105,7 @@ def test_inferer_invert_normalization(mock_inferer):
 
 
 def test_inferer_expand_to_batch_size(mock_inferer):
-    """Tests the array padding logic for strictly sized batches (e.g., ONNX)."""
+    """Strict-batch array padding."""
     mock_inferer.inference_batch_size = 5
 
     # Input has 2 samples, batch size needs to be 5
@@ -119,7 +119,7 @@ def test_inferer_expand_to_batch_size(mock_inferer):
 
 
 def test_inferer_prepare_inference_batches_pad(mock_inferer):
-    """Tests chunking data when padding is requested."""
+    """Chunking with padding."""
     mock_inferer.inference_batch_size = 4
     x = {"cat_col": np.array([[1], [2], [3]])}  # 3 total samples
 
@@ -132,7 +132,7 @@ def test_inferer_prepare_inference_batches_pad(mock_inferer):
 
 
 def test_inferer_prepare_inference_batches_no_pad(mock_inferer):
-    """Tests chunking data without padding."""
+    """Chunking without padding."""
     mock_inferer.inference_batch_size = 4
     x = {"cat_col": np.array([[1], [2], [3]])}  # 3 total samples
 
@@ -144,7 +144,7 @@ def test_inferer_prepare_inference_batches_no_pad(mock_inferer):
 
 
 def test_inferer_prepare_inference_batches_split(mock_inferer):
-    """Tests chunking data into multiple separated batches."""
+    """Chunking into multiple batches."""
     mock_inferer.inference_batch_size = 2
     x = {"cat_col": np.array([[1], [2], [3], [4], [5]])}  # 5 total samples
 
@@ -641,17 +641,11 @@ def ar_inferer(ar_config):
 
 
 def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
-    """
-    Tests that the autoregressive loop calls the model the correct number of times
-    and accurately shifts the input tensor to append the latest prediction.
-    """
+    """Check autoregressive loop count and tensor shifting."""
     initial_data = {"target_col": torch.tensor([[1.0, 2.0, 3.0], [10.0, 20.0, 30.0]])}
     metadata = _dummy_attention_metadata(batch_size=2, context_length=3)
 
-    # Patch the method on the actual instance
     with patch.object(ar_inferer, "infer_generative") as mock_infer:
-        # Step 0: Predicts 4.0 for row 0, 40.0 for row 1
-        # Step 1: Predicts 5.0 for row 0, 50.0 for row 1
         mock_infer.side_effect = [
             {"target_col": np.array([[4.0], [40.0]])},
             {"target_col": np.array([[5.0], [50.0]])},
@@ -662,10 +656,8 @@ def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
             ar_config, ar_inferer, initial_data, metadata, total_steps=total_steps
         )
 
-        # 1. Verify Loop Count
         assert mock_infer.call_count == 2
 
-        # 2. Verify Tensor Shifting
         expected_shifted_x = {
             "target_col": np.array([[2.0, 3.0, 4.0], [20.0, 30.0, 40.0]])
         }
@@ -675,7 +667,6 @@ def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
             second_call_args["target_col"], expected_shifted_x["target_col"]
         )
 
-        # 3. Verify Output Reshaping
         assert "target_col" in preds
         np.testing.assert_array_equal(preds["target_col"], [4.0, 5.0, 40.0, 50.0])
 
@@ -684,31 +675,21 @@ def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
 def test_get_probs_preds_from_dict_with_probabilities(
     mock_sample, ar_config, ar_inferer
 ):
-    """
-    Tests the probability branching logic. When output_probabilities=True,
-    `infer_generative` normalizes outputs, and passes them as probabilities
-    to the second call which triggers sampling with logits=False.
-    """
-    # 1. Override the config and inferer to treat the column as categorical
+    """Check probability output drives non-logit sampling."""
     ar_config.target_column_types["target_col"] = "categorical"
     ar_inferer.target_column_types["target_col"] = "categorical"
 
-    # 2. Force it to route to the sampling branch
     ar_config.output_probabilities = True
     ar_config.sample_from_distribution_columns = ["target_col"]
     ar_inferer.sample_from_distribution_columns = ["target_col"]
 
     initial_data = {"target_col": torch.tensor([[1.0, 2.0]])}
 
-    # Raw model output (Logits)
     dummy_logits = {"target_col": np.array([[np.log(0.2), np.log(0.8)]])}
-
-    # What we want our mock sample_with_cumsum to return
     mock_sample.return_value = np.array([1])
 
     metadata = _dummy_attention_metadata(batch_size=1, context_length=2)
 
-    # Mock the *inner* backend call, allowing infer_generative to execute its actual logic
     with patch.object(
         ar_inferer, "adjust_and_infer_generative", return_value=dummy_logits
     ) as mock_adjust:
@@ -718,27 +699,20 @@ def test_get_probs_preds_from_dict_with_probabilities(
 
         assert mock_adjust.call_count == 1
 
-        # Verify sample_with_cumsum was called correctly during the second pass
         mock_sample.assert_called_once()
         args, kwargs = mock_sample.call_args
 
-        # 1. Assert it was passed the normalized probabilities, NOT the raw logits
         np.testing.assert_allclose(args[0], [[0.2, 0.8]])
 
-        # 2. Assert the new flag was toggled correctly based on (probs is None)
         assert kwargs.get("is_log_probs") is False
 
-        # Verify final outputs
         assert probs is not None
         np.testing.assert_allclose(probs["target_col"], [[0.2, 0.8]])
         np.testing.assert_array_equal(preds["target_col"], [1])
 
 
 def test_get_probs_preds_from_dict_ignores_unselected_columns(ar_config, ar_inferer):
-    """
-    Tests that columns not explicitly defined in `config.input_columns`
-    are filtered out before inference.
-    """
+    """Check inference ignores columns outside config.input_columns."""
     initial_data = {
         "target_col": torch.tensor([[1.0, 2.0]]),
         "ignored_col": torch.tensor([[9.0, 9.0]]),
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index 9ac84b5a..cb222176 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -24,14 +24,11 @@
     process_and_write_data_pt,
 )
 
-# ==========================================
-# 1. Test Sequence Extraction (Sliding Window)
-# ==========================================
 RESERVED_MASK_COLUMN = "[mask]"
 
 
 def test_extract_subsequences_basic():
-    """Tests basic sliding window extraction with sufficient length."""
+    """Basic sliding-window extraction."""
     input_data = {"col1": [10, 11, 12, 13, 14, 15]}
     context_length = 3
     stride = 1
@@ -73,7 +70,7 @@ def test_extract_subsequences_bert_width():
 
 
 def test_extract_subsequences_padding():
-    """Tests that sequences shorter than context_length are padded with 0s."""
+    """Short sequences receive zero padding."""
     input_data = {"col1": [1, 2]}  # Length 2
     stored_context_width = 5
     stride = 1
@@ -568,7 +565,7 @@ def test_preprocessor_config_requires_metadata_config_for_mask_column(
 
 @pytest.mark.parametrize("mode", ["distribute", "exact"])
 def test_extract_subsequences_modes(mode):
-    """Tests logic for 'distribute' vs 'exact' modes."""
+    """Distribute vs exact stride modes."""
     # Length 10. Seq_len 2 (window 3).
     # distribute: adjusts stride to cover data evenly.
     # exact: strictly adheres to stride, throws error if misalignment.
@@ -602,13 +599,8 @@ def test_extract_subsequences_modes(mode):
         assert len(result["col1"]) > 0
 
 
-# ==========================================
-# 2. Test Batch Limits (Index Math)
-# ==========================================
-
-
 def test_get_batch_limits_perfect_split():
-    """Tests splitting where batch boundaries align perfectly with sequences."""
+    """Batch boundaries align with sequence boundaries."""
     # 3 sequences, each length 10. Total 30 rows.
     # If we want 3 batches, we should get 3 chunks of 10.
     data = pl.DataFrame({"sequenceId": np.repeat([1, 2, 3], 10), "val": np.arange(30)})
@@ -620,7 +612,7 @@ def test_get_batch_limits_perfect_split():
 
 
 def test_get_batch_limits_uneven_split():
-    """Tests that a batch never splits a sequenceId in half."""
+    """Batches never split a sequenceId."""
     # Seq 1: 5 rows
     # Seq 2: 15 rows
     # Seq 3: 5 rows
@@ -642,13 +634,8 @@ def test_get_batch_limits_uneven_split():
             assert prev_id != curr_id, f"Batch split at {start} broke a sequence"
 
 
-# ==========================================
-# 3. Test Statistics (Mathematical Logic)
-# ==========================================
-
-
 def test_get_combined_statistics_logic():
-    """Tests Welford's algorithm logic for merging stats."""
+    """Merged Welford stats."""
     # Create two chunks of data
     chunk1 = np.random.normal(loc=10, scale=2, size=100)
     chunk2 = np.random.normal(loc=20, scale=5, size=50)
@@ -672,7 +659,7 @@ def test_get_combined_statistics_logic():
 
 
 def test_get_column_statistics_state_accumulation():
-    """Tests that processing data in chunks yields same stats as processing at once."""
+    """Chunked and full-pass stats match."""
     data_full = pl.DataFrame(
         {
             "cat_col": ["a", "b", "a", "c", "b", "d"],
@@ -699,8 +686,6 @@ def test_get_column_statistics_state_accumulation():
         chunk2, ["cat_col", "num_col"], id_maps, stats, running_count, {}
     )
 
-    # Validations
-    # 1. Categorical: Should contain all unique keys a,b,c,d
     assert len(id_maps["cat_col"]) == 6
     assert set(id_maps["cat_col"].keys()) == {
         "[unknown]",
@@ -711,7 +696,6 @@ def test_get_column_statistics_state_accumulation():
         "d",
     }
 
-    # 2. Numerical: Should match full dataset calculation
     expected_mean = data_full["num_col"].mean()
     expected_std = data_full["num_col"].std()
 
@@ -750,7 +734,7 @@ def test_apply_column_statistics_errors_when_all_rows_masked():
 
 
 def test_create_id_map():
-    """Tests basic ID mapping creation."""
+    """Basic ID map creation."""
     df = pl.DataFrame({"A": ["z", "x", "y", "x"]})
     mapping = create_id_map(df, "A")
 
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 8418b669..c26d8a24 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -99,7 +99,7 @@ def test_poisson_span_masking_samples_at_least_one_token():
 
 @pytest.fixture
 def model_config(tmp_path):
-    """Creates a valid TrainModel configuration for testing."""
+    """Valid TrainModel config."""
     project_root = str(tmp_path)
 
     # Ensure necessary directories exist to avoid init errors (logging)
@@ -206,7 +206,7 @@ def _all_valid_metadata(batch_size, seq_len, device=None):
 
 
 def test_transformer_model_initialization(model, model_config):
-    """Tests that the model initializes with the correct layers."""
+    """Expected model layers."""
     # Check if encoder dicts were created
     assert "cat_col" in model.encoder
     assert model.pos_encoder is None or "cat_col" in model.pos_encoder
@@ -403,7 +403,7 @@ def test_load_train_config_validates_class_share_columns_after_metadata_load(
 
 
 def test_forward_train_shapes(model, model_config):
-    """Tests the output shapes of the forward_train method."""
+    """forward_train output shapes."""
     batch_size = model_config.training_spec.batch_size
     seq_len = model_config.window_view.context_length
 
@@ -434,7 +434,7 @@ def test_forward_train_shapes(model, model_config):
 
 
 def test_forward_inference_shapes(model, model_config):
-    """Tests the output shapes of the forward (inference) method."""
+    """Inference output shapes."""
     batch_size = model_config.training_spec.batch_size
     seq_len = model_config.window_view.context_length
     prediction_length = model_config.model_spec.prediction_length  # 1
@@ -467,7 +467,7 @@ def test_forward_inference_shapes(model, model_config):
 
 
 def test_calculate_loss(model, model_config):
-    """Tests that loss calculation returns a scalar tensor."""
+    """Scalar loss tensor."""
     batch_size = model_config.training_spec.batch_size
     seq_len = model_config.window_view.context_length
 
diff --git a/tools/analysis.py b/tools/analysis.py
index 9e13e322..fd642571 100644
--- a/tools/analysis.py
+++ b/tools/analysis.py
@@ -4,16 +4,7 @@
 
 
 def invert_normalization(values, target_column, selected_columns_statistics):
-    """
-    Invert the normalization of values for a target column.
-
-    Args:
-        values: Normalized values.
-        target_column: Target column name.
-
-    Returns:
-        Denormalized values.
-    """
+    """Invert target-column normalization."""
     std = selected_columns_statistics[target_column]["std"]
     mean = selected_columns_statistics[target_column]["mean"]
     return (values * (std - 1e-9)) + mean
diff --git a/tools/convert_to_pt.py b/tools/convert_to_pt.py
index 060b5af6..bb42f2ee 100644
--- a/tools/convert_to_pt.py
+++ b/tools/convert_to_pt.py
@@ -18,7 +18,6 @@ def main():
 
     assert os.path.exists(in_path), f"Input file not found: {in_path}"
 
-    # 1. Read Data
     if in_path.endswith(".csv"):
         df = pl.read_csv(in_path)
     elif in_path.endswith(".parquet"):
@@ -26,7 +25,6 @@ def main():
     else:
         raise ValueError("Input must be .csv or .parquet")
 
-    # 2. Validate Schema
     required_meta = {"sequenceId", "subsequenceId", "startItemPosition", "inputCol"}
     assert required_meta.issubset(
         df.columns
@@ -35,7 +33,6 @@ def main():
     seq_cols = sorted([c for c in df.columns if c.isdigit()], key=int)
     assert len(seq_cols) > 1, "No numbered sequence columns found"
 
-    # 3. Aggregate
     feature_names = df["inputCol"].unique().to_list()
     aggs = [
         pl.concat_list(seq_cols).filter(pl.col("inputCol") == f).flatten().alias(f)
@@ -48,7 +45,6 @@ def main():
     )
     grouped = grouped.sort(["sequenceId", "subsequenceId"])
 
-    # 4. Create Tensors
     seq_ids = torch.from_numpy(grouped["sequenceId"].to_numpy().astype(np.int64))
     sub_ids = torch.from_numpy(grouped["subsequenceId"].to_numpy().astype(np.int64))
     start_pos = torch.from_numpy(
@@ -70,7 +66,6 @@ def main():
 
     total_rows = len(seq_ids)
 
-    # 5. Save
     if chunk_size is not None and total_rows > chunk_size:
         assert not os.path.isfile(
             out_path
@@ -83,7 +78,6 @@ def main():
         for i in range(0, total_rows, chunk_size):
             end = i + chunk_size
 
-            # Slice everything
             s_ids_chunk = seq_ids[i:end]
             sub_ids_chunk = sub_ids[i:end]
             start_pos_chunk = start_pos[i:end]
diff --git a/tools/resize_pt_files.py b/tools/resize_pt_files.py
index c3d1475a..ae7ff89e 100644
--- a/tools/resize_pt_files.py
+++ b/tools/resize_pt_files.py
@@ -2,16 +2,13 @@
 import json
 import os
 import sys
-from typing import Any, Dict, List, Tuple  # Added List
+from typing import Any, Dict, List, Tuple
 
 import torch
 
 
 def unpack_dataset_tuple(data_tuple: Tuple) -> Dict[str, Any]:
-    """
-    Unpacks the standard Sequifier 5-element tuple into a structured dictionary
-    that is easier to handle programmatically.
-    """
+    """Unpack the standard Sequifier PT tuple."""
     return {
         "sequences": data_tuple[0],
         "targets": data_tuple[1],
@@ -22,9 +19,7 @@ def unpack_dataset_tuple(data_tuple: Tuple) -> Dict[str, Any]:
 
 
 def pack_dataset_tuple(data_dict: Dict[str, Any]) -> Tuple:
-    """
-    Repacks the dictionary back into the 5-element tuple format expected by Sequifier.
-    """
+    """Pack a dataset dict into the standard Sequifier PT tuple."""
     return (
         data_dict["sequences"],
         data_dict["targets"],
@@ -35,33 +30,26 @@ def pack_dataset_tuple(data_dict: Dict[str, Any]) -> Tuple:
 
 
 def concat_dataset_list(data_list: List[Dict[str, Any]]) -> Dict[str, Any]:
-    """
-    Concatenates a list of dataset dictionaries along dim 0.
-    OPTIMIZED: Uses torch.cat on a list of tensors (O(N)) instead of iterative pairwise concat (O(N^2)).
-    """
+    """Concatenate dataset dictionaries along dim 0."""
     if not data_list:
         return {}
 
-    # If only one item, just return it (avoid copy)
     if len(data_list) == 1:
         return data_list[0]
 
     combined = {}
     first = data_list[0]
 
-    # Concatenate sequence dicts
     combined["sequences"] = {
         k: torch.cat([d["sequences"][k] for d in data_list], dim=0)
         for k in first["sequences"]
     }
 
-    # Concatenate target dicts
     combined["targets"] = {
         k: torch.cat([d["targets"][k] for d in data_list], dim=0)
         for k in first["targets"]
     }
 
-    # Concatenate metadata tensors
     combined["seq_ids"] = torch.cat([d["seq_ids"] for d in data_list], dim=0)
     combined["sub_ids"] = torch.cat([d["sub_ids"] for d in data_list], dim=0)
     combined["start_pos"] = torch.cat([d["start_pos"] for d in data_list], dim=0)
@@ -69,12 +57,8 @@ def concat_dataset_list(data_list: List[Dict[str, Any]]) -> Dict[str, Any]:
     return combined
 
 
-# ... [slice_dataset, clone_dataset, get_row_size_bytes remain unchanged] ...
 def slice_dataset(data: Dict[str, Any], start: int, end: int) -> Dict[str, Any]:
-    """
-    Slices a dataset dictionary from start to end index.
-    Returns a view (fast), not a copy.
-    """
+    """Slice a dataset dictionary without cloning."""
     sliced = {}
     sliced["sequences"] = {k: v[start:end] for k, v in data["sequences"].items()}
     sliced["targets"] = {k: v[start:end] for k, v in data["targets"].items()}
@@ -85,10 +69,7 @@ def slice_dataset(data: Dict[str, Any], start: int, end: int) -> Dict[str, Any]:
 
 
 def clone_dataset(data: Dict[str, Any]) -> Dict[str, Any]:
-    """
-    Deep clones a dataset. Used for the remainder to allow the
-    large original tensor to be freed from memory.
-    """
+    """Clone all tensors in a dataset dictionary."""
     cloned = {}
     cloned["sequences"] = {k: v.clone() for k, v in data["sequences"].items()}
     cloned["targets"] = {k: v.clone() for k, v in data["targets"].items()}
@@ -99,14 +80,10 @@ def clone_dataset(data: Dict[str, Any]) -> Dict[str, Any]:
 
 
 def get_row_size_bytes(data: Dict[str, Any]) -> float:
-    """
-    Calculates the exact size in bytes of a single row in the dataset.
-    """
+    """Return bytes per row across sequence/target tensors."""
     total_bytes = 0
-    # Add size of one row for every tensor in sequences
     for t in data["sequences"].values():
         total_bytes += t.element_size() * t.shape[1]
-    # Add size of one row for every tensor in targets
     for t in data["targets"].values():
         total_bytes += t.element_size() * t.shape[1]
 
@@ -127,7 +104,6 @@ def process_split(
 ):
     os.makedirs(output_dir, exist_ok=True)
 
-    # 1. Determine File Order via metadata
     meta_path = os.path.join(input_dir, "metadata.json")
     if os.path.exists(meta_path):
         with open(meta_path, "r") as f:
@@ -141,7 +117,6 @@ def process_split(
             f"Warning: No metadata.json found in {input_dir}. Using alphabetical sort."
         )
 
-    # State variables
     data_buffer: List[Dict[str, Any]] = []
     buffer_row_count = 0
 
@@ -150,18 +125,16 @@ def process_split(
     total_samples_processed = 0
 
     target_bytes = target_size_mb * 1024 * 1024
-    target_rows = None  # Calculated dynamically on first load
+    target_rows = None
 
     print(f"Processing {input_dir} -> {output_dir}")
 
-    # 2. Iterate Files
     for file_name in input_files:
         file_path = os.path.join(input_dir, file_name)
         if not os.path.exists(file_path):
             continue
 
         try:
-            # Load file
             loaded_tuple = torch.load(file_path, map_location="cpu", weights_only=False)
             current_data = unpack_dataset_tuple(loaded_tuple)
 
@@ -169,27 +142,21 @@ def process_split(
             if current_rows == 0:
                 continue
 
-            # Calculate target_rows if not yet set (once per split)
             if target_rows is None:
                 bytes_per_row = get_row_size_bytes(current_data)
                 target_rows = max(1, int(target_bytes / bytes_per_row))
 
-            # Add to buffer
             data_buffer.append(current_data)
             buffer_row_count += current_rows
 
-            # 3. Flush Buffer if we have enough data
             if buffer_row_count >= target_rows:
-                # O(N) Merge
                 full_data = concat_dataset_list(data_buffer)
 
-                # Clear buffer references immediately
                 data_buffer = []
                 buffer_row_count = 0
 
                 num_rows = full_data["seq_ids"].shape[0]
 
-                # Slice and Save Loop
                 start_idx = 0
                 while start_idx + target_rows <= num_rows:
                     end_idx = start_idx + target_rows
@@ -209,24 +176,19 @@ def process_split(
 
                     start_idx = end_idx
 
-                # Handle Remainder
                 if start_idx < num_rows:
-                    # Clone remainder to free the massive full_data tensor
                     remainder_data = clone_dataset(
                         slice_dataset(full_data, start_idx, num_rows)
                     )
-                    # Start new buffer with remainder
                     data_buffer = [remainder_data]
                     buffer_row_count = remainder_data["seq_ids"].shape[0]
 
-                # Explicitly free full_data
                 del full_data
 
         except Exception as e:
             print(f"Error processing {file_path}: {e}")
             sys.exit(1)
 
-    # 4. Flush final remainder
     if buffer_row_count > 0:
         full_data = concat_dataset_list(data_buffer)
         fname = f"{dataset_name}-{split_suffix}-{output_batch_idx}.pt"
@@ -237,7 +199,6 @@ def process_split(
         new_batch_files_metadata.append({"path": fname, "samples": chunk_len})
         total_samples_processed += chunk_len
 
-    # 5. Write New Metadata
     new_metadata = {
         "total_samples": total_samples_processed,
         "batch_files": new_batch_files_metadata,
@@ -245,7 +206,6 @@ def process_split(
     with open(os.path.join(output_dir, "metadata.json"), "w") as f:
         json.dump(new_metadata, f, indent=4)
 
-    # 6. Validation
     if (
         expected_total_samples is not None
         and total_samples_processed != expected_total_samples
diff --git a/tools/stream_folder_via_scp.py b/tools/stream_folder_via_scp.py
index 7c116fc1..f5bcf543 100644
--- a/tools/stream_folder_via_scp.py
+++ b/tools/stream_folder_via_scp.py
@@ -10,14 +10,10 @@
 
 
 def upload_worker(sftp, upload_queue):
-    """
-    Worker thread that reads (local_path, remote_path) from queue,
-    uploads via SFTP, and deletes the local temp file.
-    """
+    """Upload queued zip batches and delete local temp files."""
     while True:
         task = upload_queue.get()
         if task is None:
-            # Sentinel value received, stop processing
             upload_queue.task_done()
             break
 
@@ -29,23 +25,20 @@ def upload_worker(sftp, upload_queue):
         except Exception as e:
             print(f"    [ERR] Failed to upload Batch {batch_num}: {e}")
         finally:
-            # Clean up local temp file after upload (or failure)
             if os.path.exists(local_path):
                 os.remove(local_path)
             upload_queue.task_done()
 
 
 def create_zip_locally(files, source_root, batch_num):
-    """Zips files to a temp file and returns the path."""
+    """Zip a batch of files into a temp archive."""
     print(f"[+] Zipping Batch {batch_num} ({len(files)} files)...")
 
-    # Create temp file
     fd, tmp_path = tempfile.mkstemp(suffix=".zip")
-    os.close(fd)  # Close file descriptor, we only need path
+    os.close(fd)
 
     with zipfile.ZipFile(tmp_path, "w", zipfile.ZIP_DEFLATED) as zf:
         for file_path in files:
-            # Preserve folder structure relative to the parent of source_root
             rel_path = os.path.relpath(file_path, os.path.dirname(source_root))
             zf.write(file_path, arcname=rel_path)
 
@@ -74,7 +67,6 @@ def main():
     password = getpass.getpass(f"Password for {args.remote_user}@{args.remote_host}: ")
     chunk_size = args.chunk_mb * 1024 * 1024
 
-    # Connect SSH/SFTP
     ssh = paramiko.SSHClient()
     ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
 
@@ -82,7 +74,6 @@ def main():
         ssh.connect(args.remote_host, username=args.remote_user, password=password)
         sftp = ssh.open_sftp()
 
-        # Setup Remote Directory
         folder_name = os.path.basename(source_path.rstrip(os.sep))
         target_dir = f"{args.remote_path.rstrip('/')}/{folder_name}-zips"
 
@@ -91,14 +82,12 @@ def main():
         except FileNotFoundError:
             sftp.mkdir(target_dir)
 
-        # Start Upload Worker Thread
         upload_q = queue.Queue()
         worker = threading.Thread(
             target=upload_worker, args=(sftp, upload_q), daemon=True
         )
         worker.start()
 
-        # Main Loop: Walk and Zip
         batch_files = []
         current_batch_size = 0
         batch_count = 1
@@ -109,14 +98,11 @@ def main():
                 size = os.path.getsize(full_path)
 
                 if current_batch_size + size > chunk_size and batch_files:
-                    # 1. Create Zip (Blocking Main Thread)
                     zip_path = create_zip_locally(batch_files, source_path, batch_count)
 
-                    # 2. Queue for Upload (Non-Blocking)
                     remote_zip_path = f"{target_dir}/batch_{batch_count:03d}.zip"
                     upload_q.put((zip_path, remote_zip_path, batch_count))
 
-                    # 3. Reset for next batch immediately
                     batch_files = []
                     current_batch_size = 0
                     batch_count += 1
@@ -124,16 +110,14 @@ def main():
                 batch_files.append(full_path)
                 current_batch_size += size
 
-        # Process final batch
         if batch_files:
             zip_path = create_zip_locally(batch_files, source_path, batch_count)
             remote_zip_path = f"{target_dir}/batch_{batch_count:03d}.zip"
             upload_q.put((zip_path, remote_zip_path, batch_count))
 
-        # Wait for uploads to finish
         print("[*] Local processing done. Waiting for remaining uploads...")
-        upload_q.put(None)  # Send sentinel to stop worker
-        worker.join()  # Wait for worker to finish
+        upload_q.put(None)
+        worker.join()
         print("[+] Streaming complete.")
 
     except paramiko.AuthenticationException:

From 99a244098df2bc91721498a69a4c385285971aa9 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 15 Jun 2026 12:45:10 +0200
Subject: [PATCH 60/81] Add pin_memory to SequifierBatch

---
 src/sequifier/io/batch.py | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/src/sequifier/io/batch.py b/src/sequifier/io/batch.py
index 597a4db7..8f9e7fe4 100644
--- a/src/sequifier/io/batch.py
+++ b/src/sequifier/io/batch.py
@@ -22,3 +22,20 @@ def __post_init__(self) -> None:
             raise ValueError(
                 f"SequifierBatch metadata is missing required keys: {missing}"
             )
+
+    def pin_memory(self) -> "SequifierBatch":
+        return SequifierBatch(
+            inputs={k: v.pin_memory() for k, v in self.inputs.items()},
+            targets={k: v.pin_memory() for k, v in self.targets.items()},
+            metadata={k: v.pin_memory() for k, v in self.metadata.items()},
+            sequence_ids=(
+                self.sequence_ids.pin_memory()
+                if self.sequence_ids is not None
+                else None
+            ),
+            subsequence_ids=(
+                self.subsequence_ids.pin_memory()
+                if self.subsequence_ids is not None
+                else None
+            ),
+        )

From d450171961b2d0a9ca76b51803bc5230a41cb9d4 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 15 Jun 2026 18:50:06 +0200
Subject: [PATCH 61/81] Token level loss scaling for distributed training

---
 src/sequifier/infer.py                    |   4 +-
 src/sequifier/train.py                    | 387 ++++++++++-----
 tests/unit/test_infer.py                  |  52 +-
 tests/unit/test_train.py                  | 425 +++++++++++++++-
 tests/unit/test_train_distributed_loss.py | 566 ++++++++++++++++++++++
 tests/unit/test_train_fsdp_loss.py        | 553 +++++++++++++++++++++
 6 files changed, 1858 insertions(+), 129 deletions(-)
 create mode 100644 tests/unit/test_train_distributed_loss.py
 create mode 100644 tests/unit/test_train_fsdp_loss.py

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 0baed6ef..57b9093f 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -1232,7 +1232,9 @@ def adjust_and_infer_embedding(
                 self.infer_pure(x_sub, metadata_sub)[0]
                 for x_sub, metadata_sub in zip(x_adjusted, metadata_adjusted)
             ]
-            embeddings = np.concatenate(inference_batch_embeddings, axis=0)[:size]
+            embeddings = np.concatenate(inference_batch_embeddings, axis=0)[
+                : size * self.prediction_length
+            ]
         elif self.inference_model_type == "pt":
             x_adjusted = self.prepare_inference_batches(x, pad_to_batch_size=False)
             metadata_adjusted = self.prepare_inference_batches(
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 3a68058c..71a0363d 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -240,6 +240,7 @@ def train_worker(
         mesh = init_device_mesh(
             "cuda", (world_size,)
         )  # 1D mesh for standard ZeRO-3 full sharding
+        model._data_parallel_group = mesh.get_group()
 
         mp_policy = None
         if config.training_spec.layer_autocast:
@@ -1358,7 +1359,12 @@ def _train_epoch(
         ddp_model: Optional[nn.Module] = None,
     ) -> None:
         """Run one train epoch with optional mid-epoch saves."""
-        total_loss = 0.0
+        target_names = self._loss_target_names()
+        train_loss_sums, train_token_count = self._new_loss_accumulators(target_names)
+        empty_global_batches = 0
+        accumulated_global_token_count = torch.zeros(
+            (), device=self.device, dtype=torch.int64
+        )
         batches_aggregated = 0
 
         start_time = time.time()
@@ -1416,17 +1422,43 @@ def _train_epoch(
                         output = model_to_call(
                             data, metadata=metadata, return_logits=True
                         )
-                        loss, losses = self._calculate_loss(output, targets, metadata)
+                        (
+                            loss,
+                            backward_components,
+                            local_loss_sums,
+                            local_token_count,
+                            global_token_count,
+                        ) = self._calculate_training_loss(output, targets, metadata)
                 else:
                     output = model_to_call(data, metadata=metadata, return_logits=True)
-                    loss, losses = self._calculate_loss(output, targets, metadata)
+                    (
+                        loss,
+                        backward_components,
+                        local_loss_sums,
+                        local_token_count,
+                        global_token_count,
+                    ) = self._calculate_training_loss(output, targets, metadata)
 
                 self.scaler.scale(loss).backward()
+                self._accumulate_loss_components(
+                    train_loss_sums,
+                    train_token_count,
+                    local_loss_sums,
+                    local_token_count,
+                )
+                accumulated_global_token_count += global_token_count.detach()
+                if global_token_count.detach().cpu().item() == 0:
+                    empty_global_batches += 1
 
-                if (
+                optimizer_step_due = (
                     self.accumulation_steps is None
                     or (batch_count + 1) % self.accumulation_steps == 0
                     or (batch_count + 1) == num_batches
+                )
+                optimizer_step_performed = False
+                if (
+                    optimizer_step_due
+                    and accumulated_global_token_count.detach().cpu().item() > 0
                 ):
                     self.scaler.unscale_(self.optimizer)
 
@@ -1435,28 +1467,49 @@ def _train_epoch(
                     self.scaler.step(self.optimizer)
                     self.scaler.update()
                     self.optimizer.zero_grad()
+                    optimizer_step_performed = True
+
+                if optimizer_step_due:
+                    if not optimizer_step_performed:
+                        self.optimizer.zero_grad()
+                    accumulated_global_token_count.zero_()
 
-                total_loss += loss.item()
                 batches_aggregated += 1
                 if (batch_count + 1) % self.log_interval == 0:
+                    avg_train_loss, _ = self._finalize_loss_components(
+                        train_loss_sums,
+                        train_token_count,
+                        target_names,
+                        "training",
+                        raise_on_empty=False,
+                    )
                     if self.rank == 0:
                         learning_rate = self.scheduler.get_last_lr()[0]
                         s_per_batch = (time.time() - start_time) / max(
                             1, batches_aggregated
                         )
-                        avg_train_loss = total_loss / max(1, batches_aggregated)
                         self.logger.info(
-                            f"[INFO] Epoch {epoch:3d} | Batch {(batch_count+1):5d}/{num_batches:5d} | Loss: {format_number(avg_train_loss)} | LR: {format_number(learning_rate)} | S/Batch {format_number(s_per_batch)}"
+                            f"[INFO] Epoch {epoch:3d} | Batch {(batch_count+1):5d}/{num_batches:5d} | Loss: {format_number(avg_train_loss.detach().cpu().item())} | LR: {format_number(learning_rate)} | S/Batch {format_number(s_per_batch)}"
                         )
-                        total_loss = 0.0
+                        if empty_global_batches > 0:
+                            self.logger.warning(
+                                "[WARNING] "
+                                f"{empty_global_batches} training microbatch(es) "
+                                "in this logging window contained no selected tokens."
+                            )
+                    train_loss_sums, train_token_count = self._new_loss_accumulators(
+                        target_names
+                    )
+                    empty_global_batches = 0
+                    if self.rank == 0:
                         batches_aggregated = 0
                         self.start_batch = 0
                         start_time = time.time()
                     self._check_and_terminate()
 
-                del data, targets, output, loss, losses
+                del data, targets, output, loss, backward_components
 
-                if self.scheduler_step_on == "batch":
+                if self.scheduler_step_on == "batch" and optimizer_step_performed:
                     if (
                         not hasattr(self.scheduler, "total_steps")
                         or self.scheduler.last_epoch < self.scheduler.total_steps
@@ -1551,8 +1604,21 @@ def _calculate_loss(
         targets: dict[str, Tensor],
         metadata: dict[str, Tensor],
     ) -> tuple[Tensor, dict[str, Tensor]]:
-        """Return weighted mean loss and per-target components over valid tokens."""
-        target_names = [k for k in targets.keys() if k in self.target_column_types]
+        """Return backward-scaled loss and components for the current rank."""
+        loss, backward_components, _, _, _ = self._calculate_training_loss(
+            output, targets, metadata
+        )
+        return loss, backward_components
+
+    @beartype
+    def _calculate_training_loss(
+        self,
+        output: dict[str, Tensor],
+        targets: dict[str, Tensor],
+        metadata: dict[str, Tensor],
+    ) -> tuple[Tensor, dict[str, Tensor], dict[str, Tensor], Tensor, Tensor]:
+        """Return the normalized backward loss plus local metric primitives."""
+        target_names = self._loss_target_names(targets)
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
@@ -1560,66 +1626,51 @@ def _calculate_loss(
         if "bert_mask" in metadata:
             valid_mask = valid_mask & metadata["bert_mask"].bool()
 
-        mask = valid_mask.T.contiguous().reshape(-1)
-        token_count = mask.sum(dtype=torch.int64)
-
-        losses = {}
-        for target_column in target_names:
-            target_column_type = self.target_column_types[target_column]
-            if target_column_type == "categorical":
-                output_tensor = (
-                    output[target_column]
-                    .float()
-                    .reshape(-1, self.n_classes[target_column])
-                )
-            elif target_column_type == "real":
-                output_tensor = (
-                    output[target_column].to(dtype=torch.float32).reshape(-1)
-                )
-            else:
-                raise ValueError(
-                    f"Target column type {target_column_type} not in ['categorical', 'real']"
-                )
-
-            target_tensor = targets[target_column].T.contiguous().reshape(-1)
-
-            if self.target_column_types[target_column] == "real":
-                target_tensor = target_tensor.to(dtype=output_tensor.dtype)
-
-            raw_loss = self.criterion[target_column](output_tensor, target_tensor)
-
-            current_mask = mask.to(dtype=raw_loss.dtype)
-            denominator = token_count.clamp_min(1).to(dtype=raw_loss.dtype)
-
-            losses[target_column] = (raw_loss * current_mask).sum() / denominator
+        local_sums, local_count = self._calculate_local_loss_components(
+            output, targets, valid_mask
+        )
+        global_count = local_count.detach().clone()
+        gradient_average_factor = self._gradient_reduction_factor()
+
+        if gradient_average_factor > 1:
+            dist.all_reduce(
+                global_count,
+                op=dist.ReduceOp.SUM,
+                group=self._data_parallel_process_group(),
+            )
 
         loss = None
+        backward_components = {}
+        denominator = global_count.clamp_min(1)
         for target_column in target_names:
-            losses[target_column] = losses[target_column] * (
-                self.loss_weights[target_column]
-                if self.loss_weights is not None
-                else 1.0
+            denominator_for_sum = denominator.to(dtype=local_sums[target_column].dtype)
+            backward_components[target_column] = (
+                local_sums[target_column]
+                * self._loss_weight(target_column)
+                * gradient_average_factor
+                / denominator_for_sum
             )
             if loss is None:
-                loss = losses[target_column].clone()
+                loss = backward_components[target_column].clone()
             else:
-                loss += losses[target_column]
+                loss += backward_components[target_column]
 
         if loss is None:
             raise RuntimeError(
                 "Loss calculation failed; no loss tensors were generated."
             )
 
-        return loss, losses
+        return loss, backward_components, local_sums, local_count, global_count
 
     @beartype
-    def _calculate_loss_components(
+    def _calculate_local_loss_components(
         self,
         output: dict[str, Tensor],
         targets: dict[str, Tensor],
         valid_mask: Tensor,
-    ) -> tuple[dict[str, Tensor], dict[str, Tensor]]:
-        target_names = [k for k in targets.keys() if k in self.target_column_types]
+    ) -> tuple[dict[str, Tensor], Tensor]:
+        """Return unweighted, unnormalized local loss sums and one token count."""
+        target_names = self._loss_target_names(targets)
         if not target_names:
             raise RuntimeError("Loss calculation failed; no target columns were found.")
 
@@ -1627,7 +1678,6 @@ def _calculate_loss_components(
         token_count = mask.sum(dtype=torch.int64)
 
         loss_sums = {}
-        token_counts = {}
         for target_column in target_names:
             target_column_type = self.target_column_types[target_column]
             if target_column_type == "categorical":
@@ -1651,17 +1701,148 @@ def _calculate_loss_components(
                 target_tensor = target_tensor.to(dtype=output_tensor.dtype)
 
             raw_loss = self.criterion[target_column](output_tensor, target_tensor)
+            if raw_loss.numel() != mask.numel():
+                raise RuntimeError(
+                    "Loss/mask size mismatch for target column "
+                    f"{target_column!r}: loss has {raw_loss.numel()} elements "
+                    f"but mask has {mask.numel()}."
+                )
             current_mask = mask.to(dtype=raw_loss.dtype)
 
-            loss_sum = (raw_loss * current_mask).sum(dtype=torch.float64)
-            loss_sums[target_column] = loss_sum * (
-                self.loss_weights[target_column]
-                if self.loss_weights is not None
-                else 1.0
+            loss_sums[target_column] = (raw_loss * current_mask).sum()
+
+        return loss_sums, token_count
+
+    @beartype
+    def _calculate_loss_components(
+        self,
+        output: dict[str, Tensor],
+        targets: dict[str, Tensor],
+        valid_mask: Tensor,
+    ) -> tuple[dict[str, Tensor], Tensor]:
+        """Return detached local loss sums and one shared token count for metrics."""
+        loss_sums, token_count = self._calculate_local_loss_components(
+            output, targets, valid_mask
+        )
+        return (
+            {col: loss_sum.detach().double() for col, loss_sum in loss_sums.items()},
+            token_count.detach(),
+        )
+
+    @beartype
+    def _loss_target_names(
+        self, targets: Optional[dict[str, Tensor]] = None
+    ) -> list[str]:
+        """Return configured target columns in stable training-config order."""
+        configured_targets = getattr(
+            self, "target_columns", list(self.target_column_types.keys())
+        )
+        if targets is not None:
+            missing_targets = [
+                col
+                for col in configured_targets
+                if col in self.target_column_types and col not in targets
+            ]
+            if missing_targets:
+                raise RuntimeError(f"Missing target columns: {sorted(missing_targets)}")
+
+        return [col for col in configured_targets if col in self.target_column_types]
+
+    @beartype
+    def _loss_weight(self, target_column: str) -> float:
+        """Return the configured scalar loss weight for a target column."""
+        if self.loss_weights is None:
+            return 1.0
+        return float(self.loss_weights[target_column])
+
+    @beartype
+    def _distributed_is_initialized(self) -> bool:
+        """Return whether torch.distributed collectives are currently usable."""
+        return dist.is_available() and dist.is_initialized()
+
+    @beartype
+    def _data_parallel_process_group(self) -> Optional[dist.ProcessGroup]:
+        """Return the process group used by the data-parallel reducer."""
+        return getattr(self, "_data_parallel_group", None)
+
+    @beartype
+    def _gradient_reduction_factor(self) -> int:
+        """Return the gradient multiplier needed before averaged reducers run."""
+        if not self._distributed_is_initialized():
+            return 1
+
+        training_spec = getattr(getattr(self, "hparams", None), "training_spec", None)
+        data_parallelism = getattr(training_spec, "data_parallelism", None)
+        if data_parallelism in {"DDP", "FSDP"}:
+            return dist.get_world_size(group=self._data_parallel_process_group())
+
+        return 1
+
+    @beartype
+    def _new_loss_accumulators(
+        self, target_names: list[str]
+    ) -> tuple[dict[str, Tensor], Tensor]:
+        """Create detached sum/count accumulators for logging or validation."""
+        return (
+            {
+                col: torch.zeros((), device=self.device, dtype=torch.float64)
+                for col in target_names
+            },
+            torch.zeros((), device=self.device, dtype=torch.float64),
+        )
+
+    @beartype
+    def _accumulate_loss_components(
+        self,
+        sums: dict[str, Tensor],
+        count: Tensor,
+        batch_sums: dict[str, Tensor],
+        batch_count: Tensor,
+    ) -> None:
+        """Accumulate detached local unweighted loss sums and token counts."""
+        for col in batch_sums:
+            sums[col] = sums[col] + batch_sums[col].detach().double()
+        count += batch_count.detach().double()
+
+    @beartype
+    def _finalize_loss_components(
+        self,
+        sums: dict[str, Tensor],
+        count: Tensor,
+        target_names: list[str],
+        label: str,
+        raise_on_empty: bool = True,
+    ) -> tuple[Tensor, dict[str, Tensor]]:
+        """Reduce local loss sums/counts and return weighted token means."""
+        packed = torch.stack([sums[col] for col in target_names] + [count])
+
+        if self._distributed_is_initialized():
+            dist.all_reduce(
+                packed,
+                op=dist.ReduceOp.SUM,
+                group=self._data_parallel_process_group(),
             )
-            token_counts[target_column] = token_count
 
-        return loss_sums, token_counts
+        n_targets = len(target_names)
+        reduced_sums = dict(zip(target_names, packed[:n_targets]))
+        reduced_count = packed[n_targets]
+
+        if reduced_count.detach().cpu().item() == 0:
+            if raise_on_empty:
+                raise RuntimeError(f"No valid {label} tokens found.")
+
+            losses = {
+                col: torch.zeros((), device=self.device, dtype=torch.float64)
+                for col in target_names
+            }
+            return torch.zeros((), device=self.device, dtype=torch.float64), losses
+
+        losses = {}
+        total = torch.zeros((), device=self.device, dtype=torch.float64)
+        for col in target_names:
+            losses[col] = reduced_sums[col] / reduced_count * self._loss_weight(col)
+            total = total + losses[col]
+        return total, losses
 
     @beartype
     def _copy_model(self):
@@ -1695,7 +1876,7 @@ def _evaluate(
         """Evaluate validation loss and optional class-share counts."""
 
         model_to_call = ddp_model if ddp_model is not None else self
-        target_names = list(self.target_columns)
+        target_names = self._loss_target_names()
         class_count_columns = list(dict.fromkeys(self.class_share_log_columns))
 
         for col in class_count_columns:
@@ -1719,61 +1900,13 @@ def _evaluate(
             for col in class_count_columns
         }
 
-        def new_accumulators() -> tuple[dict[str, Tensor], dict[str, Tensor]]:
-            return (
-                {
-                    col: torch.zeros((), device=self.device, dtype=torch.float64)
-                    for col in target_names
-                },
-                {
-                    col: torch.zeros((), device=self.device, dtype=torch.float64)
-                    for col in target_names
-                },
-            )
-
-        def accumulate_components(
-            sums: dict[str, Tensor],
-            counts: dict[str, Tensor],
-            batch_sums: dict[str, Tensor],
-            batch_counts: dict[str, Tensor],
-        ) -> None:
-            for col in batch_sums:
-                sums[col] = sums[col] + batch_sums[col].detach().double()
-                counts[col] = counts[col] + batch_counts[col].detach().double()
-
-        def finalize_components(
-            sums: dict[str, Tensor],
-            counts: dict[str, Tensor],
-            label: str,
-        ) -> tuple[Tensor, dict[str, Tensor]]:
-            packed = torch.stack(
-                [sums[col] for col in target_names]
-                + [counts[col] for col in target_names]
-            )
-
-            if self.hparams.training_spec.distributed:
-                dist.all_reduce(packed, op=dist.ReduceOp.SUM)
-
-            n_targets = len(target_names)
-            reduced_sums = dict(zip(target_names, packed[:n_targets]))
-            reduced_counts = dict(zip(target_names, packed[n_targets:]))
-
-            losses = {}
-            total = torch.zeros((), device=self.device, dtype=torch.float64)
-            for col in target_names:
-                if reduced_counts[col].detach().cpu().item() == 0:
-                    raise RuntimeError(
-                        f"No valid {label} tokens found for target column {col!r}."
-                    )
-                losses[col] = reduced_sums[col] / reduced_counts[col]
-                total = total + losses[col]
-            return total, losses
-
         was_training = model_to_call.training
         model_to_call.eval()
 
         try:
-            total_loss_sums, total_loss_counts = new_accumulators()
+            total_loss_sums, total_loss_count = self._new_loss_accumulators(
+                target_names
+            )
 
             with torch.no_grad():
                 for batch_idx, batch in enumerate(valid_loader):
@@ -1839,9 +1972,9 @@ def finalize_components(
                             output, targets, valid_mask
                         )
 
-                    accumulate_components(
+                    self._accumulate_loss_components(
                         total_loss_sums,
-                        total_loss_counts,
+                        total_loss_count,
                         loss_sums,
                         token_counts,
                     )
@@ -1852,20 +1985,23 @@ def finalize_components(
                         self.n_classes,
                     )
 
-            total_loss_global, total_losses_global = finalize_components(
-                total_loss_sums, total_loss_counts, "validation"
+            total_loss_global, total_losses_global = self._finalize_loss_components(
+                total_loss_sums, total_loss_count, target_names, "validation"
             )
 
-            if self.hparams.training_spec.distributed:
+            if self._distributed_is_initialized():
                 for col in class_count_columns:
                     dist.all_reduce(
                         local_class_counts[col],
                         op=dist.ReduceOp.SUM,
+                        group=self._data_parallel_process_group(),
                     )
 
             # Handle one-time baseline loss calculation with the same aggregation semantics.
             if not hasattr(self, "baseline_loss"):
-                baseline_loss_sums, baseline_loss_counts = new_accumulators()
+                baseline_loss_sums, baseline_loss_count = self._new_loss_accumulators(
+                    target_names
+                )
 
                 with torch.no_grad():
                     for batch_idx, batch in enumerate(valid_loader):
@@ -1941,15 +2077,18 @@ def finalize_components(
                                 targets_for_baseline,
                                 valid_mask,
                             )
-                            accumulate_components(
+                            self._accumulate_loss_components(
                                 baseline_loss_sums,
-                                baseline_loss_counts,
+                                baseline_loss_count,
                                 loss_sums,
                                 token_counts,
                             )
 
-                baseline_loss, baseline_losses = finalize_components(
-                    baseline_loss_sums, baseline_loss_counts, "baseline validation"
+                baseline_loss, baseline_losses = self._finalize_loss_components(
+                    baseline_loss_sums,
+                    baseline_loss_count,
+                    target_names,
+                    "baseline validation",
                 )
                 self.baseline_loss = baseline_loss.detach().cpu().item()
                 self.baseline_losses = {
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index cb15f9db..acd000d8 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -1,5 +1,5 @@
 import json
-from unittest.mock import patch
+from unittest.mock import MagicMock, patch
 
 import numpy as np
 import polars as pl
@@ -730,3 +730,53 @@ def test_get_probs_preds_from_dict_ignores_unselected_columns(ar_config, ar_infe
 
         assert "target_col" in first_call_args
         assert "ignored_col" not in first_call_args
+
+
+def test_adjust_and_infer_embedding_onnx_truncation():
+    """
+    Catches the truncation bug where ONNX embeddings are sliced by `[:size]`
+    instead of `[:size * self.prediction_length]`.
+    """
+    # 1. Mock the Inferer instance to isolate the target method
+    #    We avoid actual __init__ to bypass ONNX runtime initialization.
+    inferer_mock = MagicMock(spec=Inferer)
+    inferer_mock.inference_model_type = "onnx"
+    inferer_mock.inference_batch_size = 2
+
+    # Set prediction_length > 1 to expose the bug (e.g., BERT objective)
+    inferer_mock.prediction_length = 5
+    embedding_dim = 16
+
+    # 2. Mock `prepare_inference_batches` to pass our dummy inputs straight through
+    inferer_mock.prepare_inference_batches.side_effect = (
+        lambda data, pad_to_batch_size: [data]
+    )
+
+    # 3. Mock `infer_pure` to return the shape an ONNX model would actually return.
+    #    `infer_pure` flattens the output to (batch_size * prediction_length, dim).
+    #    For a batch size of 2 and prediction length of 5, it returns (10, 16).
+    flattened_sequence_length = (
+        inferer_mock.inference_batch_size * inferer_mock.prediction_length
+    )
+    dummy_onnx_output = np.ones((flattened_sequence_length, embedding_dim))
+    inferer_mock.infer_pure.return_value = [dummy_onnx_output]
+
+    # 4. Construct dummy inputs
+    size = 2
+    x = {"feature_col": np.zeros((size, 10))}  # 2 sequences, context_length = 10
+    metadata = {"attention_valid_mask": np.ones((size, 10))}
+
+    # 5. Execute the method under test
+    # Passing inferer_mock as `self` to the unbound class method.
+    embeddings = Inferer.adjust_and_infer_embedding(inferer_mock, x, size, metadata)
+
+    # 6. Assert the bug is fixed
+    expected_rows = size * inferer_mock.prediction_length
+
+    assert embeddings.shape[0] == expected_rows, (
+        f"ONNX Embedding truncation bug detected! "
+        f"Expected {expected_rows} rows (size * prediction_length), "
+        f"but got {embeddings.shape[0]} rows. "
+        f"Check the slice at the end of np.concatenate in adjust_and_infer_embedding."
+    )
+    assert embeddings.shape[1] == embedding_dim
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index c26d8a24..3600d9c7 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -205,6 +205,41 @@ def _all_valid_metadata(batch_size, seq_len, device=None):
     }
 
 
+def _loss_shell_model(
+    target_column_types,
+    *,
+    loss_weights=None,
+    class_weights=None,
+):
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_columns = list(target_column_types)
+    model.target_column_types = dict(target_column_types)
+    model.loss_weights = loss_weights
+    model.device = "cpu"
+    model.criterion = {}
+    model.n_classes = {}
+
+    for col, col_type in target_column_types.items():
+        if col_type == "categorical":
+            weight = None
+            n_classes = 3
+            if class_weights is not None and col in class_weights:
+                col_class_weights = class_weights[col]
+                weight = torch.tensor(col_class_weights, dtype=torch.float32)
+                n_classes = len(col_class_weights)
+            model.criterion[col] = torch.nn.CrossEntropyLoss(
+                reduction="none",
+                weight=weight,
+            )
+            model.n_classes[col] = n_classes
+        elif col_type == "real":
+            model.criterion[col] = torch.nn.MSELoss(reduction="none")
+        else:
+            raise ValueError(col_type)
+
+    return model
+
+
 def test_transformer_model_initialization(model, model_config):
     """Expected model layers."""
     # Check if encoder dicts were created
@@ -557,6 +592,290 @@ def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
     assert torch.isclose(component_losses["real_target"], torch.tensor(1.0))
 
 
+def test_calculate_local_loss_components_categorical_target_with_padding():
+    model = _loss_shell_model({"cat_col": "categorical"})
+    model.n_classes["cat_col"] = 3
+    logits = torch.tensor(
+        [
+            [[2.0, 0.0, 0.0]],
+            [[0.0, 2.0, 0.0]],
+            [[0.0, 0.0, 2.0]],
+        ]
+    )
+    targets = {"cat_col": torch.tensor([[0, 1, 2]])}
+    valid_mask = torch.tensor([[True, False, True]])
+
+    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
+        model,
+        {"cat_col": logits},
+        targets,
+        valid_mask,
+    )
+    raw_loss = torch.nn.functional.cross_entropy(
+        logits.reshape(-1, 3),
+        targets["cat_col"].T.reshape(-1),
+        reduction="none",
+    )
+
+    assert torch.isclose(loss_sums["cat_col"], raw_loss[[0, 2]].sum())
+    assert torch.equal(token_count, torch.tensor(2))
+
+
+def test_calculate_local_loss_components_real_target_with_padding():
+    model = _loss_shell_model({"real_col": "real"})
+    output = {"real_col": torch.tensor([[[1.0]], [[2.0]], [[4.0]]])}
+    targets = {"real_col": torch.tensor([[0.0, 10.0, 1.0]])}
+    valid_mask = torch.tensor([[True, False, True]])
+
+    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
+        model, output, targets, valid_mask
+    )
+
+    assert torch.isclose(loss_sums["real_col"], torch.tensor(10.0))
+    assert torch.equal(token_count, torch.tensor(2))
+
+
+def test_calculate_training_loss_intersects_target_and_bert_masks():
+    model = _loss_shell_model({"real_col": "real"})
+    output = {"real_col": torch.tensor([[[1.0]], [[2.0]], [[4.0]]])}
+    targets = {"real_col": torch.tensor([[0.0, 10.0, 1.0]])}
+    metadata = {
+        "target_valid_mask": torch.tensor([[True, True, True]]),
+        "bert_mask": torch.tensor([[True, False, True]]),
+    }
+
+    total_loss, component_losses = TransformerModel._calculate_loss(
+        model, output, targets, metadata
+    )
+
+    assert torch.isclose(total_loss, torch.tensor(5.0))
+    assert torch.isclose(component_losses["real_col"], torch.tensor(5.0))
+
+
+def test_calculate_training_loss_applies_unequal_target_loss_weights():
+    model = _loss_shell_model(
+        {"cat_col": "categorical", "real_col": "real"},
+        loss_weights={"cat_col": 0.25, "real_col": 2.0},
+    )
+    model.n_classes["cat_col"] = 3
+    output = {
+        "cat_col": torch.tensor(
+            [
+                [[2.0, 0.0, 0.0]],
+                [[0.0, 2.0, 0.0]],
+            ]
+        ),
+        "real_col": torch.tensor([[[1.0]], [[3.0]]]),
+    }
+    targets = {
+        "cat_col": torch.tensor([[0, 1]]),
+        "real_col": torch.tensor([[0.0, 1.0]]),
+    }
+    metadata = {"target_valid_mask": torch.ones(1, 2, dtype=torch.bool)}
+
+    total_loss, component_losses = TransformerModel._calculate_loss(
+        model, output, targets, metadata
+    )
+    cat_mean = torch.nn.functional.cross_entropy(
+        output["cat_col"].reshape(-1, 3),
+        targets["cat_col"].T.reshape(-1),
+        reduction="mean",
+    )
+    real_mean = torch.nn.functional.mse_loss(
+        output["real_col"].reshape(-1),
+        targets["real_col"].T.reshape(-1),
+        reduction="mean",
+    )
+
+    assert torch.isclose(component_losses["cat_col"], cat_mean * 0.25)
+    assert torch.isclose(component_losses["real_col"], real_mean * 2.0)
+    assert torch.isclose(total_loss, cat_mean * 0.25 + real_mean * 2.0)
+
+
+def test_calculate_local_loss_components_keeps_class_weights_inside_criterion():
+    class_weights = {"cat_col": [1.0, 4.0, 1.0]}
+    model = _loss_shell_model({"cat_col": "categorical"}, class_weights=class_weights)
+    logits = torch.tensor(
+        [
+            [[2.0, 0.0, 0.0]],
+            [[2.0, 0.0, 0.0]],
+        ]
+    )
+    targets = {"cat_col": torch.tensor([[0, 1]])}
+    valid_mask = torch.ones(1, 2, dtype=torch.bool)
+
+    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
+        model,
+        {"cat_col": logits},
+        targets,
+        valid_mask,
+    )
+    expected = torch.nn.functional.cross_entropy(
+        logits.reshape(-1, 3),
+        targets["cat_col"].T.reshape(-1),
+        reduction="none",
+        weight=torch.tensor(class_weights["cat_col"]),
+    ).sum()
+
+    assert torch.isclose(loss_sums["cat_col"], expected)
+    assert torch.equal(token_count, torch.tensor(2))
+
+
+def test_calculate_local_loss_components_zero_selected_tokens_stays_connected():
+    model = _loss_shell_model({"real_col": "real"})
+    output = {"real_col": torch.ones(3, 1, 1, requires_grad=True)}
+    targets = {"real_col": torch.zeros(1, 3)}
+    valid_mask = torch.zeros(1, 3, dtype=torch.bool)
+
+    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
+        model,
+        output,
+        targets,
+        valid_mask,
+    )
+    loss_sums["real_col"].backward()
+
+    assert torch.equal(token_count, torch.tensor(0))
+    assert torch.equal(loss_sums["real_col"].detach(), torch.tensor(0.0))
+    assert torch.equal(output["real_col"].grad, torch.zeros_like(output["real_col"]))
+
+
+def test_calculate_local_loss_components_raises_on_output_mask_mismatch():
+    model = _loss_shell_model({"real_col": "real"})
+    output = {"real_col": torch.zeros(3, 1, 1)}
+    targets = {"real_col": torch.zeros(1, 3)}
+    valid_mask = torch.ones(1, 2, dtype=torch.bool)
+
+    with pytest.raises(RuntimeError, match="Loss/mask size mismatch"):
+        TransformerModel._calculate_local_loss_components(
+            model,
+            output,
+            targets,
+            valid_mask,
+        )
+
+
+def test_training_and_validation_loss_finalization_share_weighted_sum_count_semantics():
+    model = _loss_shell_model(
+        {"cat_col": "categorical", "real_col": "real"},
+        loss_weights={"cat_col": 0.5, "real_col": 3.0},
+    )
+    model.n_classes["cat_col"] = 3
+    output = {
+        "cat_col": torch.tensor(
+            [
+                [[1.0, 0.0, 0.0]],
+                [[0.0, 1.0, 0.0]],
+                [[0.0, 0.0, 1.0]],
+            ]
+        ),
+        "real_col": torch.tensor([[[1.0]], [[2.0]], [[3.0]]]),
+    }
+    targets = {
+        "cat_col": torch.tensor([[0, 1, 2]]),
+        "real_col": torch.tensor([[0.0, 0.0, 0.0]]),
+    }
+    valid_mask = torch.tensor([[True, False, True]])
+    metadata = {"target_valid_mask": valid_mask}
+
+    training_loss, training_components = TransformerModel._calculate_loss(
+        model, output, targets, metadata
+    )
+    sums, count = TransformerModel._calculate_loss_components(
+        model, output, targets, valid_mask
+    )
+    finalized_loss, finalized_components = TransformerModel._finalize_loss_components(
+        model,
+        sums,
+        count.double(),
+        ["cat_col", "real_col"],
+        "test",
+    )
+
+    assert torch.isclose(training_loss.double(), finalized_loss)
+    assert torch.isclose(
+        training_components["cat_col"].double(), finalized_components["cat_col"]
+    )
+    assert torch.isclose(
+        training_components["real_col"].double(), finalized_components["real_col"]
+    )
+
+
+@pytest.mark.parametrize(
+    ("data_parallelism", "expected"),
+    [
+        ("DDP", 7),
+        ("FSDP", 7),
+    ],
+)
+def test_gradient_reduction_factor_uses_active_world_size(
+    data_parallelism,
+    expected,
+    monkeypatch,
+):
+    model = _loss_shell_model({"real_col": "real"})
+    model.hparams = SimpleNamespace(
+        training_spec=SimpleNamespace(data_parallelism=data_parallelism)
+    )
+    group = object()
+    model._data_parallel_process_group = lambda: group
+
+    monkeypatch.setattr("sequifier.train.dist.is_available", lambda: True)
+    monkeypatch.setattr("sequifier.train.dist.is_initialized", lambda: True)
+
+    def fake_get_world_size(group=None):
+        assert group is model._data_parallel_process_group()
+        return 7
+
+    monkeypatch.setattr("sequifier.train.dist.get_world_size", fake_get_world_size)
+
+    assert TransformerModel._gradient_reduction_factor(model) == expected
+
+
+def test_calculate_loss_requires_all_configured_targets():
+    model = _loss_shell_model(
+        {"cat_col": "categorical", "real_col": "real"},
+        loss_weights={"cat_col": 1.0, "real_col": 1.0},
+    )
+    output = {
+        "cat_col": torch.zeros(1, 1, 3),
+        "real_col": torch.zeros(1, 1, 1),
+    }
+    targets = {"cat_col": torch.zeros(1, 1, dtype=torch.long)}
+    metadata = {"target_valid_mask": torch.ones(1, 1, dtype=torch.bool)}
+
+    with pytest.raises(RuntimeError, match="Missing target columns: \\['real_col'\\]"):
+        TransformerModel._calculate_loss(model, output, targets, metadata)
+
+
+def test_training_loss_enables_fsdp_token_weighting(
+    monkeypatch,
+):
+    model = _loss_shell_model({"real_col": "real"})
+    model.hparams = SimpleNamespace(
+        training_spec=SimpleNamespace(data_parallelism="FSDP")
+    )
+    output = {"real_col": torch.zeros(1, 1, 1)}
+    targets = {"real_col": torch.ones(1, 1)}
+    metadata = {"target_valid_mask": torch.ones(1, 1, dtype=torch.bool)}
+
+    monkeypatch.setattr("sequifier.train.dist.is_available", lambda: True)
+    monkeypatch.setattr("sequifier.train.dist.is_initialized", lambda: True)
+    monkeypatch.setattr(
+        "sequifier.train.dist.get_world_size",
+        lambda group=None: 2,
+    )
+
+    def fake_all_reduce(tensor, op=None, group=None):
+        tensor.fill_(2)
+
+    monkeypatch.setattr("sequifier.train.dist.all_reduce", fake_all_reduce)
+
+    loss, _ = TransformerModel._calculate_loss(model, output, targets, metadata)
+
+    assert torch.isclose(loss, torch.tensor(1.0))
+
+
 def test_evaluation_loss_mask_intersects_target_bert_and_sample_masks():
     metadata = {
         "target_valid_mask": torch.tensor(
@@ -752,13 +1071,13 @@ def test_calculate_loss_components_aggregate_by_token_count():
     model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
     model.loss_weights = {"real_col": 1.0}
 
-    first_sums, first_counts = TransformerModel._calculate_loss_components(
+    first_sums, first_count = TransformerModel._calculate_loss_components(
         model,
         {"real_col": torch.zeros(1, 1, 1)},
         {"real_col": torch.tensor([[10.0]])},
         torch.ones(1, 1, dtype=torch.bool),
     )
-    second_sums, second_counts = TransformerModel._calculate_loss_components(
+    second_sums, second_count = TransformerModel._calculate_loss_components(
         model,
         {"real_col": torch.zeros(100, 1, 1)},
         {"real_col": torch.ones(1, 100)},
@@ -766,7 +1085,7 @@ def test_calculate_loss_components_aggregate_by_token_count():
     )
 
     aggregate = (first_sums["real_col"] + second_sums["real_col"]) / (
-        first_counts["real_col"] + second_counts["real_col"]
+        first_count + second_count
     )
 
     assert torch.isclose(aggregate, torch.tensor(200.0 / 101.0, dtype=torch.float64))
@@ -903,6 +1222,106 @@ def test_calculate_loss_zero_token_training_batch_is_differentiable():
     assert torch.equal(output["real_col"].grad, torch.zeros_like(output["real_col"]))
 
 
+def _train_epoch_test_batch(model, target_valid_mask):
+    seq_len = model.window_view.context_length
+    return SequifierBatch(
+        inputs={
+            "cat_col": torch.ones(1, seq_len, dtype=torch.long),
+            "real_col": torch.ones(1, seq_len, dtype=torch.float32),
+        },
+        targets={
+            "cat_col": torch.ones(1, seq_len, dtype=torch.long),
+            "real_col": torch.ones(1, seq_len, dtype=torch.float32),
+        },
+        metadata={
+            "attention_valid_mask": torch.ones(1, seq_len, dtype=torch.bool),
+            "target_valid_mask": target_valid_mask,
+        },
+    )
+
+
+def test_train_epoch_skips_optimizer_and_batch_scheduler_for_empty_accumulation_window(
+    model,
+):
+    model.rank = 0
+    model.log_interval = 2
+    model.accumulation_steps = 2
+    model.scheduler_step_on = "batch"
+    model.start_batch = 0
+    model.optimizer = torch.optim.AdamW(
+        model.parameters(),
+        lr=0.01,
+        weight_decay=0.1,
+    )
+    model.scheduler = torch.optim.lr_scheduler.StepLR(
+        model.optimizer,
+        step_size=1,
+        gamma=0.1,
+    )
+    seq_len = model.window_view.context_length
+    empty_batch = _train_epoch_test_batch(
+        model,
+        torch.zeros(1, seq_len, dtype=torch.bool),
+    )
+    before = {name: param.detach().clone() for name, param in model.named_parameters()}
+    lr_before = model.scheduler.get_last_lr()[0]
+
+    TransformerModel._train_epoch(
+        model,
+        DataLoader([empty_batch, empty_batch], batch_size=None),
+        DataLoader([], batch_size=None),
+        epoch=1,
+    )
+
+    after = dict(model.named_parameters())
+    assert all(torch.equal(before[name], after[name]) for name in before)
+    assert len(model.optimizer.state) == 0
+    assert model.scheduler.get_last_lr()[0] == lr_before
+
+
+def test_train_epoch_steps_once_for_mixed_empty_and_nonempty_accumulation_window(
+    model,
+):
+    model.rank = 0
+    model.log_interval = 2
+    model.accumulation_steps = 2
+    model.scheduler_step_on = "batch"
+    model.start_batch = 0
+    model.optimizer = torch.optim.AdamW(
+        model.parameters(),
+        lr=0.01,
+        weight_decay=0.1,
+    )
+    model.scheduler = torch.optim.lr_scheduler.StepLR(
+        model.optimizer,
+        step_size=1,
+        gamma=0.1,
+    )
+    seq_len = model.window_view.context_length
+    empty_batch = _train_epoch_test_batch(
+        model,
+        torch.zeros(1, seq_len, dtype=torch.bool),
+    )
+    nonempty_batch = _train_epoch_test_batch(
+        model,
+        torch.ones(1, seq_len, dtype=torch.bool),
+    )
+    before = {name: param.detach().clone() for name, param in model.named_parameters()}
+
+    TransformerModel._train_epoch(
+        model,
+        DataLoader([empty_batch, nonempty_batch], batch_size=None),
+        DataLoader([], batch_size=None),
+        epoch=1,
+    )
+
+    after = dict(model.named_parameters())
+    assert any(not torch.equal(before[name], after[name]) for name in before)
+    assert len(model.optimizer.state) > 0
+    assert model.scheduler.last_epoch == 1
+    assert model.scheduler.get_last_lr()[0] == pytest.approx(0.001)
+
+
 def test_padding_keys_are_masked(bert_model):
     seq_len = bert_model.window_view.context_length
 
diff --git a/tests/unit/test_train_distributed_loss.py b/tests/unit/test_train_distributed_loss.py
new file mode 100644
index 00000000..51423489
--- /dev/null
+++ b/tests/unit/test_train_distributed_loss.py
@@ -0,0 +1,566 @@
+import datetime
+import socket
+from types import SimpleNamespace
+
+import pytest
+import torch
+import torch.distributed as dist
+import torch.multiprocessing as mp
+from torch.nn.parallel import DistributedDataParallel as DDP
+
+from sequifier.train import TransformerModel
+
+
+class _TinyRealModel(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.param = torch.nn.Parameter(torch.tensor(0.0))
+
+    def forward(self, seq_len):
+        return {"real_col": self.param.expand(seq_len, 1, 1)}
+
+
+class _TinyMultiTargetModel(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.cat_logits = torch.nn.Parameter(torch.tensor([0.2, -0.1, 0.0]))
+        self.real_param = torch.nn.Parameter(torch.tensor(0.5))
+
+    def forward(self, seq_len):
+        return {
+            "cat_col": self.cat_logits.reshape(1, 1, 3).expand(seq_len, 1, 3),
+            "real_col": self.real_param.expand(seq_len, 1, 1),
+        }
+
+
+def _free_port():
+    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as sock:
+        sock.bind(("127.0.0.1", 0))
+        return sock.getsockname()[1]
+
+
+def _loss_shell(target_column_types, *, data_parallelism=None, loss_weights=None):
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_columns = list(target_column_types)
+    model.target_column_types = dict(target_column_types)
+    model.loss_weights = loss_weights
+    model.device = "cpu"
+    model.hparams = SimpleNamespace(
+        training_spec=SimpleNamespace(data_parallelism=data_parallelism)
+    )
+    model.criterion = {}
+    model.n_classes = {}
+
+    for col, col_type in target_column_types.items():
+        if col_type == "real":
+            model.criterion[col] = torch.nn.MSELoss(reduction="none")
+        elif col_type == "categorical":
+            model.criterion[col] = torch.nn.CrossEntropyLoss(reduction="none")
+            model.n_classes[col] = 3
+        else:
+            raise ValueError(col_type)
+
+    return model
+
+
+def _real_batch(seq_len, selected_count, target_value):
+    targets = {"real_col": torch.full((1, seq_len), float(target_value))}
+    mask = torch.zeros(1, seq_len, dtype=torch.bool)
+    mask[:, :selected_count] = True
+    metadata = {"target_valid_mask": mask}
+    return targets, metadata
+
+
+def _concat_real_batch(seq_lens, selected_counts, target_values):
+    target_parts = []
+    mask_parts = []
+    for seq_len, selected_count, target_value in zip(
+        seq_lens, selected_counts, target_values
+    ):
+        target_parts.append(torch.full((1, seq_len), float(target_value)))
+        mask = torch.zeros(1, seq_len, dtype=torch.bool)
+        mask[:, :selected_count] = True
+        mask_parts.append(mask)
+    return (
+        {"real_col": torch.cat(target_parts, dim=1)},
+        {"target_valid_mask": torch.cat(mask_parts, dim=1)},
+    )
+
+
+def _single_process_real_grad(seq_lens, selected_counts, target_values):
+    model = _TinyRealModel()
+    shell = _loss_shell({"real_col": "real"})
+    targets, metadata = _concat_real_batch(seq_lens, selected_counts, target_values)
+    loss, _ = TransformerModel._calculate_loss(
+        shell,
+        model(sum(seq_lens)),
+        targets,
+        metadata,
+    )
+    loss.backward()
+    return model.param.grad.detach().clone()
+
+
+def _single_process_real_update(seq_lens, selected_counts, target_values, lr):
+    model = _TinyRealModel()
+    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
+    shell = _loss_shell({"real_col": "real"})
+    targets, metadata = _concat_real_batch(seq_lens, selected_counts, target_values)
+    loss, _ = TransformerModel._calculate_loss(
+        shell,
+        model(sum(seq_lens)),
+        targets,
+        metadata,
+    )
+    loss.backward()
+    optimizer.step()
+    return model.param.detach().clone()
+
+
+def _single_process_accumulated_real_update(microbatches, lr):
+    model = _TinyRealModel()
+    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
+    shell = _loss_shell({"real_col": "real"})
+    for microbatch in microbatches:
+        targets, metadata = _concat_real_batch(
+            microbatch["seq_lens"],
+            microbatch["selected_counts"],
+            microbatch["target_values"],
+        )
+        loss, _ = TransformerModel._calculate_loss(
+            shell,
+            model(sum(microbatch["seq_lens"])),
+            targets,
+            metadata,
+        )
+        loss.backward()
+    optimizer.step()
+    return model.param.detach().clone()
+
+
+def _single_process_multi_target_grad(
+    seq_lens, cat_targets, real_targets, loss_weights
+):
+    model = _TinyMultiTargetModel()
+    shell = _loss_shell(
+        {"cat_col": "categorical", "real_col": "real"},
+        loss_weights=loss_weights,
+    )
+    targets = {
+        "cat_col": torch.tensor([sum(cat_targets, [])], dtype=torch.long),
+        "real_col": torch.tensor([sum(real_targets, [])], dtype=torch.float32),
+    }
+    metadata = {"target_valid_mask": torch.ones(1, sum(seq_lens), dtype=torch.bool)}
+    loss, _ = TransformerModel._calculate_loss(
+        shell,
+        model(sum(seq_lens)),
+        targets,
+        metadata,
+    )
+    loss.backward()
+    return torch.cat(
+        [
+            model.cat_logits.grad.detach(),
+            model.real_param.grad.detach().reshape(1),
+        ]
+    )
+
+
+def _ddp_case_worker(rank, world_size, init_method, case, queue):
+    torch.set_num_threads(1)
+    dist.init_process_group(
+        "gloo",
+        rank=rank,
+        world_size=world_size,
+        init_method=init_method,
+        timeout=datetime.timedelta(seconds=30),
+    )
+    try:
+
+        def put_result(tensor):
+            detached = tensor.detach().cpu()
+            if detached.numel() == 1:
+                queue.put(float(detached.item()))
+            else:
+                queue.put(detached.tolist())
+
+        if case["kind"] in {
+            "real_grad",
+            "real_update",
+            "empty_grad",
+            "accumulation_update",
+            "empty_window_state",
+        }:
+            model = _TinyRealModel()
+            ddp_model = DDP(model)
+            shell = _loss_shell({"real_col": "real"}, data_parallelism="DDP")
+
+            if case["kind"] == "empty_window_state":
+                optimizer = torch.optim.AdamW(
+                    model.parameters(),
+                    lr=case["lr"],
+                    weight_decay=case["weight_decay"],
+                )
+                scheduler = torch.optim.lr_scheduler.StepLR(
+                    optimizer,
+                    step_size=1,
+                    gamma=0.1,
+                )
+                before = model.param.detach().clone()
+                accumulated_global_token_count = torch.zeros((), dtype=torch.int64)
+
+                for batch_idx, microbatch in enumerate(case["microbatches"]):
+                    seq_len = microbatch["seq_lens"][rank]
+                    targets, metadata = _real_batch(
+                        seq_len,
+                        microbatch["selected_counts"][rank],
+                        microbatch["target_values"][rank],
+                    )
+                    loss, _, _, _, global_count = (
+                        TransformerModel._calculate_training_loss(
+                            shell,
+                            ddp_model(seq_len),
+                            targets,
+                            metadata,
+                        )
+                    )
+                    loss.backward()
+                    accumulated_global_token_count += global_count.detach()
+                    optimizer_step_due = (batch_idx + 1) % case[
+                        "accumulation_steps"
+                    ] == 0 or (batch_idx + 1) == len(case["microbatches"])
+                    optimizer_step_performed = False
+                    if (
+                        optimizer_step_due
+                        and accumulated_global_token_count.detach().cpu().item() > 0
+                    ):
+                        optimizer.step()
+                        optimizer.zero_grad()
+                        optimizer_step_performed = True
+
+                    if optimizer_step_due:
+                        if not optimizer_step_performed:
+                            optimizer.zero_grad()
+                        accumulated_global_token_count.zero_()
+
+                    if optimizer_step_performed:
+                        scheduler.step()
+
+                if rank == 0:
+                    queue.put(
+                        [
+                            float((model.param.detach() - before).abs().item()),
+                            float(scheduler.get_last_lr()[0]),
+                            float(scheduler.last_epoch),
+                            float(len(optimizer.state)),
+                        ]
+                    )
+                return
+
+            if case["kind"] == "accumulation_update":
+                optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
+                for microbatch in case["microbatches"]:
+                    seq_len = microbatch["seq_lens"][rank]
+                    targets, metadata = _real_batch(
+                        seq_len,
+                        microbatch["selected_counts"][rank],
+                        microbatch["target_values"][rank],
+                    )
+                    loss, _, _, _, _ = TransformerModel._calculate_training_loss(
+                        shell,
+                        ddp_model(seq_len),
+                        targets,
+                        metadata,
+                    )
+                    loss.backward()
+                optimizer.step()
+                if rank == 0:
+                    put_result(model.param)
+                return
+
+            seq_len = case["seq_lens"][rank]
+            targets, metadata = _real_batch(
+                seq_len,
+                case["selected_counts"][rank],
+                case["target_values"][rank],
+            )
+            loss, _, _, _, _ = TransformerModel._calculate_training_loss(
+                shell,
+                ddp_model(seq_len),
+                targets,
+                metadata,
+            )
+            loss.backward()
+
+            if case["kind"] == "real_update":
+                optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
+                optimizer.step()
+                result = model.param.detach().clone()
+            else:
+                result = model.param.grad.detach().clone()
+
+            if rank == 0:
+                put_result(result)
+            return
+
+        if case["kind"] == "metrics":
+            shell = _loss_shell(
+                {"cat_col": "categorical", "real_col": "real"},
+                data_parallelism="DDP",
+                loss_weights=case["loss_weights"],
+            )
+            sums = {
+                "cat_col": torch.tensor(case["cat_sums"][rank], dtype=torch.float64),
+                "real_col": torch.tensor(case["real_sums"][rank], dtype=torch.float64),
+            }
+            count = torch.tensor(case["counts"][rank], dtype=torch.float64)
+            total, losses = TransformerModel._finalize_loss_components(
+                shell,
+                sums,
+                count,
+                ["cat_col", "real_col"],
+                "training",
+            )
+            if rank == 0:
+                queue.put(
+                    [
+                        float(total.detach().cpu().item()),
+                        float(losses["cat_col"].detach().cpu().item()),
+                        float(losses["real_col"].detach().cpu().item()),
+                    ]
+                )
+            return
+
+        if case["kind"] == "multi_target_grad":
+            model = _TinyMultiTargetModel()
+            ddp_model = DDP(model)
+            shell = _loss_shell(
+                {"cat_col": "categorical", "real_col": "real"},
+                data_parallelism="DDP",
+                loss_weights=case["loss_weights"],
+            )
+            seq_len = case["seq_lens"][rank]
+            targets = {
+                "cat_col": torch.tensor([case["cat_targets"][rank]], dtype=torch.long),
+                "real_col": torch.tensor(
+                    [case["real_targets"][rank]], dtype=torch.float32
+                ),
+            }
+            metadata = {"target_valid_mask": torch.ones(1, seq_len, dtype=torch.bool)}
+            loss, _, _, _, _ = TransformerModel._calculate_training_loss(
+                shell,
+                ddp_model(seq_len),
+                targets,
+                metadata,
+            )
+            loss.backward()
+            if rank == 0:
+                put_result(
+                    torch.cat(
+                        [
+                            model.cat_logits.grad.detach(),
+                            model.real_param.grad.detach().reshape(1),
+                        ]
+                    )
+                )
+            return
+
+        raise ValueError(case["kind"])
+    finally:
+        dist.destroy_process_group()
+
+
+def _run_ddp_case(case):
+    ctx = mp.get_context("spawn")
+    queue = ctx.SimpleQueue()
+    init_method = f"tcp://127.0.0.1:{_free_port()}"
+    mp.spawn(
+        _ddp_case_worker,
+        args=(2, init_method, case, queue),
+        nprocs=2,
+        join=True,
+    )
+    return torch.tensor(queue.get())
+
+
+pytestmark = pytest.mark.skipif(
+    not dist.is_available(),
+    reason="torch.distributed is not available",
+)
+
+
+def test_ddp_unequal_token_counts_match_single_process_reference():
+    case = {
+        "kind": "real_grad",
+        "seq_lens": [100, 10],
+        "selected_counts": [100, 10],
+        "target_values": [2.0, 8.0],
+    }
+
+    ddp_grad = _run_ddp_case(case)
+    reference_grad = _single_process_real_grad(
+        case["seq_lens"], case["selected_counts"], case["target_values"]
+    )
+    old_equal_rank_mean = torch.tensor((-2.0 * 2.0 + -2.0 * 8.0) / 2.0)
+
+    assert torch.allclose(ddp_grad, reference_grad, atol=1e-6)
+    assert not torch.allclose(ddp_grad, old_equal_rank_mean, atol=1e-3)
+
+
+def test_ddp_zero_token_rank_does_not_attenuate_populated_rank():
+    case = {
+        "kind": "real_grad",
+        "seq_lens": [5, 5],
+        "selected_counts": [5, 0],
+        "target_values": [3.0, 100.0],
+    }
+
+    ddp_grad = _run_ddp_case(case)
+    reference_grad = _single_process_real_grad(
+        case["seq_lens"], case["selected_counts"], case["target_values"]
+    )
+
+    assert torch.allclose(ddp_grad, reference_grad, atol=1e-6)
+    assert torch.allclose(ddp_grad, torch.tensor(-6.0), atol=1e-6)
+
+
+def test_ddp_equal_token_counts_retain_equal_rank_mean_behavior():
+    case = {
+        "kind": "real_grad",
+        "seq_lens": [4, 4],
+        "selected_counts": [4, 4],
+        "target_values": [1.0, 5.0],
+    }
+
+    ddp_grad = _run_ddp_case(case)
+    equal_rank_mean = torch.tensor((-2.0 * 1.0 + -2.0 * 5.0) / 2.0)
+
+    assert torch.allclose(ddp_grad, equal_rank_mean, atol=1e-6)
+
+
+def test_ddp_multi_target_weighting_matches_single_process_reference():
+    case = {
+        "kind": "multi_target_grad",
+        "seq_lens": [3, 1],
+        "cat_targets": [[0, 1, 2], [1]],
+        "real_targets": [[1.0, 2.0, 3.0], [4.0]],
+        "loss_weights": {"cat_col": 0.25, "real_col": 2.0},
+    }
+
+    ddp_grad = _run_ddp_case(case)
+    reference_grad = _single_process_multi_target_grad(
+        case["seq_lens"],
+        case["cat_targets"],
+        case["real_targets"],
+        case["loss_weights"],
+    )
+
+    assert torch.allclose(ddp_grad, reference_grad, atol=1e-6)
+
+
+def test_ddp_world_size_two_optimizer_update_matches_single_process_reference():
+    case = {
+        "kind": "real_update",
+        "seq_lens": [6, 4],
+        "selected_counts": [6, 4],
+        "target_values": [1.0, 3.0],
+        "lr": 0.1,
+    }
+
+    ddp_param = _run_ddp_case(case)
+    reference_param = _single_process_real_update(
+        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
+    )
+
+    assert torch.allclose(ddp_param, reference_param, atol=1e-6)
+
+
+def test_ddp_globally_empty_batch_gradients_are_finite_and_zero():
+    case = {
+        "kind": "empty_grad",
+        "seq_lens": [3, 4],
+        "selected_counts": [0, 0],
+        "target_values": [1.0, 3.0],
+    }
+
+    ddp_grad = _run_ddp_case(case)
+
+    assert torch.isfinite(ddp_grad)
+    assert torch.equal(ddp_grad, torch.tensor(0.0))
+
+
+def test_ddp_globally_empty_accumulation_window_leaves_state_unchanged():
+    case = {
+        "kind": "empty_window_state",
+        "microbatches": [
+            {
+                "seq_lens": [3, 4],
+                "selected_counts": [0, 0],
+                "target_values": [1.0, 3.0],
+            },
+            {
+                "seq_lens": [2, 5],
+                "selected_counts": [0, 0],
+                "target_values": [2.0, 4.0],
+            },
+        ],
+        "accumulation_steps": 2,
+        "lr": 0.01,
+        "weight_decay": 0.1,
+    }
+
+    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_ddp_case(case)
+
+    assert param_delta == 0
+    assert lr == pytest.approx(case["lr"])
+    assert scheduler_epoch == 0
+    assert optimizer_state_len == 0
+
+
+def test_ddp_gradient_accumulation_preserves_per_microbatch_weighting():
+    microbatches = [
+        {
+            "seq_lens": [4, 2],
+            "selected_counts": [4, 2],
+            "target_values": [1.0, 3.0],
+        },
+        {
+            "seq_lens": [1, 5],
+            "selected_counts": [1, 5],
+            "target_values": [10.0, 2.0],
+        },
+    ]
+    case = {
+        "kind": "accumulation_update",
+        "microbatches": microbatches,
+        "lr": 0.05,
+    }
+
+    ddp_param = _run_ddp_case(case)
+    reference_param = _single_process_accumulated_real_update(microbatches, case["lr"])
+
+    assert torch.allclose(ddp_param, reference_param, atol=1e-6)
+
+
+def test_ddp_global_training_metric_finalization_matches_manual_sum_count():
+    case = {
+        "kind": "metrics",
+        "cat_sums": [10.0, 5.0],
+        "real_sums": [2.0, 8.0],
+        "counts": [4.0, 6.0],
+        "loss_weights": {"cat_col": 0.5, "real_col": 2.0},
+    }
+
+    total, cat_loss, real_loss = _run_ddp_case(case)
+
+    global_count = sum(case["counts"])
+    expected_cat = (
+        sum(case["cat_sums"]) / global_count * case["loss_weights"]["cat_col"]
+    )
+    expected_real = (
+        sum(case["real_sums"]) / global_count * case["loss_weights"]["real_col"]
+    )
+
+    assert cat_loss == pytest.approx(expected_cat)
+    assert real_loss == pytest.approx(expected_real)
+    assert total == pytest.approx(expected_cat + expected_real)
diff --git a/tests/unit/test_train_fsdp_loss.py b/tests/unit/test_train_fsdp_loss.py
new file mode 100644
index 00000000..b4f4a6b9
--- /dev/null
+++ b/tests/unit/test_train_fsdp_loss.py
@@ -0,0 +1,553 @@
+import datetime
+import importlib
+import socket
+from types import SimpleNamespace
+
+import pytest
+import torch
+import torch.distributed as dist
+import torch.multiprocessing as mp
+from torch.distributed.checkpoint.state_dict import (
+    StateDictOptions,
+    get_model_state_dict,
+)
+
+from sequifier.train import TransformerModel
+
+_train_module = importlib.import_module("sequifier.train")
+MixedPrecisionPolicy = getattr(_train_module, "MixedPrecisionPolicy")
+fully_shard = getattr(_train_module, "fully_shard")
+init_device_mesh = getattr(_train_module, "init_device_mesh")
+
+pytestmark = pytest.mark.skipif(
+    not dist.is_available()
+    or not dist.is_nccl_available()
+    or not torch.cuda.is_available()
+    or torch.cuda.device_count() < 2,
+    reason="FSDP2 loss tests require two CUDA devices and NCCL",
+)
+
+
+class _TinyRealModel(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.param = torch.nn.Parameter(torch.zeros(2))
+
+    def forward(self, seq_len):
+        return {"real_col": self.param[0].expand(seq_len, 1, 1)}
+
+
+class _TinyMultiTargetModel(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.cat_logits = torch.nn.Parameter(torch.tensor([0.2, -0.1, 0.0]))
+        self.real_param = torch.nn.Parameter(torch.tensor([0.5, 0.0]))
+
+    def forward(self, seq_len):
+        return {
+            "cat_col": self.cat_logits.reshape(1, 1, 3).expand(seq_len, 1, 3),
+            "real_col": self.real_param[0].expand(seq_len, 1, 1),
+        }
+
+
+def _free_port():
+    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as sock:
+        sock.bind(("127.0.0.1", 0))
+        return sock.getsockname()[1]
+
+
+def _loss_shell(target_column_types, *, device, loss_weights=None):
+    model = TransformerModel.__new__(TransformerModel)
+    model.target_columns = list(target_column_types)
+    model.target_column_types = dict(target_column_types)
+    model.loss_weights = loss_weights
+    model.device = device
+    model.hparams = SimpleNamespace(
+        training_spec=SimpleNamespace(data_parallelism="FSDP")
+    )
+    model.criterion = {}
+    model.n_classes = {}
+
+    for col, col_type in target_column_types.items():
+        if col_type == "real":
+            model.criterion[col] = torch.nn.MSELoss(reduction="none")
+        elif col_type == "categorical":
+            model.criterion[col] = torch.nn.CrossEntropyLoss(reduction="none")
+            model.n_classes[col] = 3
+        else:
+            raise ValueError(col_type)
+
+    return model
+
+
+def _real_batch(seq_len, selected_count, target_value, device):
+    targets = {"real_col": torch.full((1, seq_len), float(target_value), device=device)}
+    mask = torch.zeros(1, seq_len, dtype=torch.bool, device=device)
+    mask[:, :selected_count] = True
+    metadata = {"target_valid_mask": mask}
+    return targets, metadata
+
+
+def _concat_real_batch(seq_lens, selected_counts, target_values):
+    target_parts = []
+    mask_parts = []
+    for seq_len, selected_count, target_value in zip(
+        seq_lens, selected_counts, target_values
+    ):
+        target_parts.append(torch.full((1, seq_len), float(target_value)))
+        mask = torch.zeros(1, seq_len, dtype=torch.bool)
+        mask[:, :selected_count] = True
+        mask_parts.append(mask)
+    return (
+        {"real_col": torch.cat(target_parts, dim=1)},
+        {"target_valid_mask": torch.cat(mask_parts, dim=1)},
+    )
+
+
+def _single_process_real_update(seq_lens, selected_counts, target_values, lr):
+    model = _TinyRealModel()
+    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
+    shell = _loss_shell({"real_col": "real"}, device="cpu")
+    targets, metadata = _concat_real_batch(seq_lens, selected_counts, target_values)
+    loss, _ = TransformerModel._calculate_loss(
+        shell,
+        model(sum(seq_lens)),
+        targets,
+        metadata,
+    )
+    loss.backward()
+    optimizer.step()
+    return model.param.detach().clone()
+
+
+def _single_process_accumulated_real_update(microbatches, lr):
+    model = _TinyRealModel()
+    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
+    shell = _loss_shell({"real_col": "real"}, device="cpu")
+    for microbatch in microbatches:
+        targets, metadata = _concat_real_batch(
+            microbatch["seq_lens"],
+            microbatch["selected_counts"],
+            microbatch["target_values"],
+        )
+        loss, _ = TransformerModel._calculate_loss(
+            shell,
+            model(sum(microbatch["seq_lens"])),
+            targets,
+            metadata,
+        )
+        loss.backward()
+    optimizer.step()
+    return model.param.detach().clone()
+
+
+def _single_process_multi_target_update(
+    seq_lens,
+    cat_targets,
+    real_targets,
+    loss_weights,
+    lr,
+):
+    model = _TinyMultiTargetModel()
+    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
+    shell = _loss_shell(
+        {"cat_col": "categorical", "real_col": "real"},
+        device="cpu",
+        loss_weights=loss_weights,
+    )
+    targets = {
+        "cat_col": torch.tensor([sum(cat_targets, [])], dtype=torch.long),
+        "real_col": torch.tensor([sum(real_targets, [])], dtype=torch.float32),
+    }
+    metadata = {"target_valid_mask": torch.ones(1, sum(seq_lens), dtype=torch.bool)}
+    loss, _ = TransformerModel._calculate_loss(
+        shell,
+        model(sum(seq_lens)),
+        targets,
+        metadata,
+    )
+    loss.backward()
+    optimizer.step()
+    return torch.cat(
+        [
+            model.cat_logits.detach().reshape(-1),
+            model.real_param.detach().reshape(-1),
+        ]
+    )
+
+
+def _state_vector(model, keys, rank):
+    options = StateDictOptions(full_state_dict=True, cpu_offload=True)
+    state = get_model_state_dict(model, options=options)
+    if rank != 0:
+        return None
+
+    values = []
+    for key in keys:
+        values.append(state[key].reshape(-1).to(dtype=torch.float32))
+    return torch.cat(values)
+
+
+def _mixed_precision_policy(case):
+    reduce_dtype = case.get("reduce_dtype")
+    if reduce_dtype is None:
+        return None
+
+    return MixedPrecisionPolicy(
+        param_dtype=torch.bfloat16,
+        reduce_dtype=reduce_dtype,
+        output_dtype=torch.bfloat16,
+    )
+
+
+def _fsdp_case_worker(rank, world_size, init_method, case, queue):
+    torch.cuda.set_device(rank)
+    device = torch.device(f"cuda:{rank}")
+    dist.init_process_group(
+        backend="nccl",
+        rank=rank,
+        world_size=world_size,
+        init_method=init_method,
+        timeout=datetime.timedelta(seconds=60),
+    )
+    try:
+        mesh = init_device_mesh("cuda", (world_size,))
+        mp_policy = _mixed_precision_policy(case)
+
+        if case["kind"] in {
+            "real_update",
+            "accumulation_update",
+            "empty_window_state",
+        }:
+            model = _TinyRealModel().to(device)
+            fully_shard(model, mesh=mesh, mp_policy=mp_policy)
+            shell = _loss_shell({"real_col": "real"}, device=str(device))
+            shell._data_parallel_group = mesh.get_group()
+
+            if case["kind"] == "empty_window_state":
+                optimizer = torch.optim.AdamW(
+                    model.parameters(),
+                    lr=case["lr"],
+                    weight_decay=case["weight_decay"],
+                )
+                scheduler = torch.optim.lr_scheduler.StepLR(
+                    optimizer,
+                    step_size=1,
+                    gamma=0.1,
+                )
+                before = _state_vector(model, ["param"], rank)
+                accumulated_global_token_count = torch.zeros(
+                    (), dtype=torch.int64, device=device
+                )
+
+                for batch_idx, microbatch in enumerate(case["microbatches"]):
+                    seq_len = microbatch["seq_lens"][rank]
+                    targets, metadata = _real_batch(
+                        seq_len,
+                        microbatch["selected_counts"][rank],
+                        microbatch["target_values"][rank],
+                        device,
+                    )
+                    loss, _, _, _, global_count = (
+                        TransformerModel._calculate_training_loss(
+                            shell,
+                            model(seq_len),
+                            targets,
+                            metadata,
+                        )
+                    )
+                    loss.backward()
+                    accumulated_global_token_count += global_count.detach()
+                    optimizer_step_due = (batch_idx + 1) % case[
+                        "accumulation_steps"
+                    ] == 0 or (batch_idx + 1) == len(case["microbatches"])
+                    optimizer_step_performed = False
+                    if (
+                        optimizer_step_due
+                        and accumulated_global_token_count.detach().cpu().item() > 0
+                    ):
+                        optimizer.step()
+                        optimizer.zero_grad()
+                        optimizer_step_performed = True
+
+                    if optimizer_step_due:
+                        if not optimizer_step_performed:
+                            optimizer.zero_grad()
+                        accumulated_global_token_count.zero_()
+
+                    if optimizer_step_performed:
+                        scheduler.step()
+
+                after = _state_vector(model, ["param"], rank)
+                if rank == 0:
+                    assert before is not None
+                    assert after is not None
+                    queue.put(
+                        [
+                            float((after - before).abs().max().item()),
+                            float(scheduler.get_last_lr()[0]),
+                            float(scheduler.last_epoch),
+                            float(len(optimizer.state)),
+                        ]
+                    )
+                return
+
+            optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
+            if case["kind"] == "accumulation_update":
+                for microbatch in case["microbatches"]:
+                    seq_len = microbatch["seq_lens"][rank]
+                    targets, metadata = _real_batch(
+                        seq_len,
+                        microbatch["selected_counts"][rank],
+                        microbatch["target_values"][rank],
+                        device,
+                    )
+                    loss, _ = TransformerModel._calculate_loss(
+                        shell,
+                        model(seq_len),
+                        targets,
+                        metadata,
+                    )
+                    loss.backward()
+            else:
+                seq_len = case["seq_lens"][rank]
+                targets, metadata = _real_batch(
+                    seq_len,
+                    case["selected_counts"][rank],
+                    case["target_values"][rank],
+                    device,
+                )
+                loss, _ = TransformerModel._calculate_loss(
+                    shell,
+                    model(seq_len),
+                    targets,
+                    metadata,
+                )
+                loss.backward()
+
+            optimizer.step()
+            state = _state_vector(model, ["param"], rank)
+            if rank == 0:
+                assert state is not None
+                queue.put(state.tolist())
+            return
+
+        if case["kind"] == "multi_target_update":
+            model = _TinyMultiTargetModel().to(device)
+            fully_shard(model, mesh=mesh, mp_policy=mp_policy)
+            shell = _loss_shell(
+                {"cat_col": "categorical", "real_col": "real"},
+                device=str(device),
+                loss_weights=case["loss_weights"],
+            )
+            shell._data_parallel_group = mesh.get_group()
+            optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
+            seq_len = case["seq_lens"][rank]
+            targets = {
+                "cat_col": torch.tensor(
+                    [case["cat_targets"][rank]], dtype=torch.long, device=device
+                ),
+                "real_col": torch.tensor(
+                    [case["real_targets"][rank]], dtype=torch.float32, device=device
+                ),
+            }
+            metadata = {
+                "target_valid_mask": torch.ones(
+                    1, seq_len, dtype=torch.bool, device=device
+                )
+            }
+            loss, _ = TransformerModel._calculate_loss(
+                shell,
+                model(seq_len),
+                targets,
+                metadata,
+            )
+            loss.backward()
+            optimizer.step()
+            state = _state_vector(model, ["cat_logits", "real_param"], rank)
+            if rank == 0:
+                assert state is not None
+                queue.put(state.tolist())
+            return
+
+        raise ValueError(case["kind"])
+    finally:
+        dist.destroy_process_group()
+        torch.cuda.empty_cache()
+
+
+def _run_fsdp_case(case):
+    ctx = mp.get_context("spawn")
+    queue = ctx.SimpleQueue()
+    init_method = f"tcp://127.0.0.1:{_free_port()}"
+    mp.spawn(
+        _fsdp_case_worker,
+        args=(2, init_method, case, queue),
+        nprocs=2,
+        join=True,
+    )
+    return torch.tensor(queue.get(), dtype=torch.float32)
+
+
+def test_fsdp_unequal_token_counts_match_single_process_reference():
+    case = {
+        "kind": "real_update",
+        "seq_lens": [100, 10],
+        "selected_counts": [100, 10],
+        "target_values": [2.0, 8.0],
+        "lr": 0.1,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_real_update(
+        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
+    )
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+
+
+@pytest.mark.parametrize("reduce_dtype", [torch.bfloat16, torch.float32])
+def test_fsdp_mixed_precision_unequal_token_counts_match_reference(reduce_dtype):
+    case = {
+        "kind": "real_update",
+        "seq_lens": [100, 10],
+        "selected_counts": [100, 10],
+        "target_values": [2.0, 8.0],
+        "lr": 0.1,
+        "reduce_dtype": reduce_dtype,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_real_update(
+        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
+    )
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-2, rtol=1e-2)
+
+
+def test_fsdp_zero_token_rank_does_not_attenuate_populated_rank():
+    case = {
+        "kind": "real_update",
+        "seq_lens": [5, 5],
+        "selected_counts": [5, 0],
+        "target_values": [3.0, 100.0],
+        "lr": 0.1,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_real_update(
+        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
+    )
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+
+
+@pytest.mark.parametrize("reduce_dtype", [torch.bfloat16, torch.float32])
+def test_fsdp_mixed_precision_zero_token_rank_is_not_attenuated(reduce_dtype):
+    case = {
+        "kind": "real_update",
+        "seq_lens": [5, 5],
+        "selected_counts": [5, 0],
+        "target_values": [3.0, 100.0],
+        "lr": 0.1,
+        "reduce_dtype": reduce_dtype,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_real_update(
+        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
+    )
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-2, rtol=1e-2)
+
+
+def test_fsdp_multi_target_weighting_matches_single_process_reference():
+    case = {
+        "kind": "multi_target_update",
+        "seq_lens": [3, 1],
+        "cat_targets": [[0, 1, 2], [1]],
+        "real_targets": [[1.0, 2.0, 3.0], [4.0]],
+        "loss_weights": {"cat_col": 0.25, "real_col": 2.0},
+        "lr": 0.1,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_multi_target_update(
+        case["seq_lens"],
+        case["cat_targets"],
+        case["real_targets"],
+        case["loss_weights"],
+        case["lr"],
+    )
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+
+
+def test_fsdp_world_size_two_optimizer_update_matches_single_process_reference():
+    case = {
+        "kind": "real_update",
+        "seq_lens": [6, 4],
+        "selected_counts": [6, 4],
+        "target_values": [1.0, 3.0],
+        "lr": 0.1,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_real_update(
+        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
+    )
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+
+
+def test_fsdp_gradient_accumulation_preserves_per_microbatch_weighting():
+    microbatches = [
+        {
+            "seq_lens": [4, 2],
+            "selected_counts": [4, 2],
+            "target_values": [1.0, 3.0],
+        },
+        {
+            "seq_lens": [1, 5],
+            "selected_counts": [1, 5],
+            "target_values": [10.0, 2.0],
+        },
+    ]
+    case = {
+        "kind": "accumulation_update",
+        "microbatches": microbatches,
+        "lr": 0.05,
+    }
+
+    fsdp_param = _run_fsdp_case(case)
+    reference_param = _single_process_accumulated_real_update(microbatches, case["lr"])
+
+    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+
+
+def test_fsdp_globally_empty_accumulation_window_leaves_state_unchanged():
+    case = {
+        "kind": "empty_window_state",
+        "microbatches": [
+            {
+                "seq_lens": [3, 4],
+                "selected_counts": [0, 0],
+                "target_values": [1.0, 3.0],
+            },
+            {
+                "seq_lens": [2, 5],
+                "selected_counts": [0, 0],
+                "target_values": [2.0, 4.0],
+            },
+        ],
+        "accumulation_steps": 2,
+        "lr": 0.01,
+        "weight_decay": 0.1,
+    }
+
+    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_fsdp_case(case)
+
+    assert param_delta == 0
+    assert lr == pytest.approx(case["lr"])
+    assert scheduler_epoch == 0
+    assert optimizer_state_len == 0

From 214661173b2d107d3dc412de66319b662b0ec406 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 15 Jun 2026 19:18:18 +0200
Subject: [PATCH 62/81] Vectorize bert masking

---
 src/sequifier/config/probabilities.py |  65 +++--
 src/sequifier/helpers.py              | 249 +++++++++++++-------
 tests/unit/test_helpers.py            | 327 +++++++++++++++++++++++++-
 tests/unit/test_probabilities.py      |  51 ++++
 4 files changed, 583 insertions(+), 109 deletions(-)
 create mode 100644 tests/unit/test_probabilities.py

diff --git a/src/sequifier/config/probabilities.py b/src/sequifier/config/probabilities.py
index 231dd6a1..ddb92ae1 100644
--- a/src/sequifier/config/probabilities.py
+++ b/src/sequifier/config/probabilities.py
@@ -1,5 +1,6 @@
+import math
 from abc import ABC, abstractmethod
-from typing import Annotated, Literal, Union
+from typing import Annotated, Literal, Optional, Union
 
 import torch
 from pydantic import BaseModel, Field
@@ -11,7 +12,12 @@ class ProbabilityDistributionBaseClass(ABC):
     """
 
     @abstractmethod
-    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
+    def sample(
+        self,
+        shape: tuple[int, ...],
+        device: torch.device,
+        generator: Optional[torch.Generator] = None,
+    ) -> torch.Tensor:
         pass
 
 
@@ -19,14 +25,19 @@ class GeometricDistribution(BaseModel, ProbabilityDistributionBaseClass):
     type: Literal["GeometricDistribution"] = "GeometricDistribution"
     p: float = Field(..., gt=0.0, le=1.0)
 
-    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
+    def sample(
+        self,
+        shape: tuple[int, ...],
+        device: torch.device,
+        generator: Optional[torch.Generator] = None,
+    ) -> torch.Tensor:
         if self.p == 1.0:
             return torch.ones(shape, device=device, dtype=torch.long)
 
-        # torch.distributions.Geometric models the number of failures before the first success.
-        # Adding 1 converts it to the total number of trials (span length).
-        m = torch.distributions.Geometric(probs=torch.tensor([self.p], device=device))
-        return (m.sample(shape).squeeze(-1) + 1).long()
+        uniform = torch.rand(shape, device=device, generator=generator)
+        return (
+            torch.floor(torch.log1p(-uniform) / math.log1p(-self.p)).to(torch.long) + 1
+        )
 
 
 class NormalDistributionDiscretizedFloor(BaseModel, ProbabilityDistributionBaseClass):
@@ -36,9 +47,16 @@ class NormalDistributionDiscretizedFloor(BaseModel, ProbabilityDistributionBaseC
     mean: float
     standard_deviation: float = Field(..., gt=0.0)
 
-    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
-        val = torch.normal(
-            mean=self.mean, std=self.standard_deviation, size=shape, device=device
+    def sample(
+        self,
+        shape: tuple[int, ...],
+        device: torch.device,
+        generator: Optional[torch.Generator] = None,
+    ) -> torch.Tensor:
+        val = (
+            torch.randn(shape, device=device, generator=generator)
+            * self.standard_deviation
+            + self.mean
         )
         return torch.clamp(torch.round(val), min=0).long() + 1
 
@@ -52,12 +70,18 @@ class LogNormalDistributionDiscretizedFloor(
     mean: float
     standard_deviation: float = Field(..., gt=0.0)
 
-    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
-        m = torch.distributions.LogNormal(
-            loc=torch.tensor([self.mean], device=device),
-            scale=torch.tensor([self.standard_deviation], device=device),
+    def sample(
+        self,
+        shape: tuple[int, ...],
+        device: torch.device,
+        generator: Optional[torch.Generator] = None,
+    ) -> torch.Tensor:
+        normal = (
+            torch.randn(shape, device=device, generator=generator)
+            * self.standard_deviation
+            + self.mean
         )
-        val = m.sample(shape).squeeze(-1)
+        val = torch.exp(normal)
         return torch.round(val).long() + 1
 
 
@@ -65,9 +89,14 @@ class PoissonDistributionFloor(BaseModel, ProbabilityDistributionBaseClass):
     type: Literal["PoissonDistributionFloor"] = "PoissonDistributionFloor"
     rate: float = Field(..., gt=0.0)
 
-    def sample(self, shape: tuple, device: torch.device) -> torch.Tensor:
-        m = torch.distributions.Poisson(rate=torch.tensor([self.rate], device=device))
-        return m.sample(shape).squeeze(-1).long() + 1
+    def sample(
+        self,
+        shape: tuple[int, ...],
+        device: torch.device,
+        generator: Optional[torch.Generator] = None,
+    ) -> torch.Tensor:
+        rate = torch.full(shape, self.rate, device=device)
+        return torch.poisson(rate, generator=generator).long() + 1
 
 
 ProbabilityDistribution = Annotated[
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 4769da24..4faee46b 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -1,4 +1,5 @@
 import glob
+import math
 import os
 import random
 import re
@@ -490,6 +491,102 @@ def get_last_training_batch_timedelta(
     return (t2 - t1).total_seconds()
 
 
+def _build_bert_span_mask(
+    valid_mask: torch.Tensor,
+    masking_probability: float,
+    span_distribution: Any,
+    *,
+    generator: Optional[torch.Generator] = None,
+) -> torch.Tensor:
+    """Construct exact-budget, non-overlapping BERT span masks."""
+    valid_mask = valid_mask.bool()
+    batch_size, seq_len = valid_mask.shape
+    device = valid_mask.device
+
+    valid_lengths = valid_mask.sum(dim=1, dtype=torch.long)
+    budgets = (valid_lengths.to(torch.float32) * masking_probability).to(torch.long)
+
+    max_spans = max(1, math.floor(seq_len * masking_probability) + 10)
+    sampled_lengths = span_distribution.sample(
+        (batch_size, max_spans),
+        device=device,
+        generator=generator,
+    )
+    sampled_lengths = sampled_lengths.to(torch.long).clamp_min_(1)
+
+    used_before = sampled_lengths.cumsum(dim=1) - sampled_lengths
+    remaining = (budgets[:, None] - used_before).clamp_min(0)
+    span_lengths = torch.minimum(sampled_lengths, remaining)
+
+    n_spans = (span_lengths > 0).sum(dim=1)
+    total_gap_length = valid_lengths - budgets
+
+    gap_slot = torch.arange(max_spans + 1, device=device)
+    active_gap_slot = gap_slot[None, :] <= n_spans[:, None]
+
+    uniform = torch.rand(
+        (batch_size, max_spans + 1),
+        device=device,
+        dtype=torch.float32,
+        generator=generator,
+    )
+    uniform = uniform.clamp_min(torch.finfo(uniform.dtype).tiny)
+
+    gap_weights = torch.where(
+        active_gap_slot,
+        -torch.log(uniform),
+        torch.zeros_like(uniform),
+    )
+    cumulative_weights = gap_weights.cumsum(dim=1)
+    weight_totals = cumulative_weights[:, -1:].clamp_min(
+        torch.finfo(gap_weights.dtype).tiny
+    )
+
+    gap_edges = torch.floor(
+        cumulative_weights
+        / weight_totals
+        * total_gap_length[:, None].to(gap_weights.dtype)
+    ).to(torch.long)
+    gap_edges = torch.where(
+        gap_slot[None, :] >= n_spans[:, None],
+        total_gap_length[:, None],
+        gap_edges,
+    )
+
+    gaps = torch.diff(
+        torch.cat(
+            [
+                torch.zeros((batch_size, 1), dtype=torch.long, device=device),
+                gap_edges,
+            ],
+            dim=1,
+        ),
+        dim=1,
+    )
+
+    lengths_before = span_lengths.cumsum(dim=1) - span_lengths
+    gaps_through_current = gaps[:, :max_spans].cumsum(dim=1)
+
+    span_starts = lengths_before + gaps_through_current
+    span_ends = span_starts + span_lengths
+
+    compact_position = valid_mask.to(torch.long).cumsum(dim=1) - 1
+    compact_position = compact_position.clamp_min(0)
+
+    started_spans = torch.searchsorted(
+        span_starts.contiguous(),
+        compact_position.contiguous(),
+        right=True,
+    )
+    ended_spans = torch.searchsorted(
+        span_ends.contiguous(),
+        compact_position.contiguous(),
+        right=True,
+    )
+
+    return valid_mask & (started_spans > ended_spans)
+
+
 def apply_bert_masking(
     data_batch: Dict[str, torch.Tensor],
     targets_batch: Dict[str, torch.Tensor],
@@ -498,18 +595,13 @@ def apply_bert_masking(
     eval_seed: Optional[int] = None,
 ) -> tuple[Dict[str, torch.Tensor], Dict[str, torch.Tensor], Dict[str, torch.Tensor]]:
     """Apply BERT span corruption and attach prediction masks."""
-    data_batch = {k: tensor.clone() for k, tensor in data_batch.items()}
-    targets_batch = {k: tensor.clone().detach() for k, tensor in targets_batch.items()}
-    metadata_batch = (
-        {k: tensor.clone().detach() for k, tensor in metadata_batch.items()}
-        if metadata_batch
-        else {}
-    )
+    if not metadata_batch or "attention_valid_mask" not in metadata_batch:
+        raise ValueError("BERT masking requires metadata['attention_valid_mask']")
 
     valid_mask = metadata_batch["attention_valid_mask"].bool()
-
     batch_size, seq_len = valid_mask.shape
     device = valid_mask.device
+
     for target_name, target in targets_batch.items():
         if target.shape != valid_mask.shape:
             raise ValueError(
@@ -517,70 +609,32 @@ def apply_bert_masking(
                 f"expected {valid_mask.shape}"
             )
 
+    generator: Optional[torch.Generator] = None
     if eval_seed is not None:
-        cpu_rng_state = torch.get_rng_state()
-        if device.type == "cuda":
-            gpu_rng_state = torch.cuda.get_rng_state(device)
-        torch.manual_seed(eval_seed)
-
-    masking_prob = config.training_spec.bert_spec.masking_probability
-    budgets = (valid_mask.sum(dim=1) * masking_prob).long()
-
-    bert_mask = torch.zeros((batch_size, seq_len), dtype=torch.bool, device=device)
-
-    max_spans = int(seq_len * masking_prob) + 10
-
-    sampled_lengths = config.training_spec.bert_spec.span_masking.sample(
-        (batch_size, max_spans), device=device
+        seeded_generator = torch.Generator(device=device)
+        seeded_generator.manual_seed(eval_seed)
+        generator = seeded_generator
+
+    bert_spec = config.training_spec.bert_spec
+    if bert_spec is None:
+        raise ValueError("bert_spec must be configured for BERT training")
+
+    bert_mask = _build_bert_span_mask(
+        valid_mask,
+        bert_spec.masking_probability,
+        bert_spec.span_masking,
+        generator=generator,
     )
 
-    sampled_starts_pct = torch.rand((batch_size, max_spans), device=device)
-    for i in range(batch_size):
-        budget = budgets[i].item()
-        valid_positions = valid_mask[i].nonzero(as_tuple=True)[0]
-        valid_len = len(valid_positions)
-
-        if budget < 1 or valid_len < 1:
-            continue
-
-        sampled_starts = (sampled_starts_pct[i] * valid_len).long().tolist()
-
-        sampled_lengths_list = sampled_lengths[i].tolist()
-
-        current_masked = 0
-        span_idx = 0
-
-        while current_masked < budget and span_idx < max_spans:
-            span_len = sampled_lengths_list[span_idx]
-            start_idx = sampled_starts[span_idx]
-
-            end_idx = min(start_idx + span_len, valid_len)
-            span_positions = valid_positions[start_idx:end_idx]
-            if len(span_positions) == 0:
-                span_idx += 1
-                continue
+    replacement = bert_spec.replacement_distribution
+    p_masked = replacement.masked
+    p_random = replacement.random
 
-            unmasked_positions = span_positions[~bert_mask[i, span_positions]]
-            allowance = budget - current_masked
-            if len(unmasked_positions) > allowance:
-                unmasked_positions = unmasked_positions[:allowance]
-
-            bert_mask[i, unmasked_positions] = True
-            current_masked = bert_mask[i].sum().item()
-            span_idx += 1
-
-        # Fallback: if heavy overlaps exhausted our max_spans, uniformly mask the remainder
-        if current_masked < budget:
-            remaining = budget - current_masked
-            unmasked_valid = (valid_mask[i] & ~bert_mask[i]).nonzero(as_tuple=True)[0]
-            if len(unmasked_valid) > 0:
-                idx = torch.randperm(len(unmasked_valid), device=device)[:remaining]
-                bert_mask[i, unmasked_valid[idx]] = True
-
-    replacement_probs = torch.rand((batch_size, seq_len), device=device)
-
-    p_masked = config.training_spec.bert_spec.replacement_distribution.masked
-    p_random = config.training_spec.bert_spec.replacement_distribution.random
+    replacement_probs = torch.rand(
+        (batch_size, seq_len),
+        device=device,
+        generator=generator,
+    )
 
     mask_token_mask = bert_mask & (replacement_probs < p_masked)
     random_token_mask = (
@@ -589,33 +643,48 @@ def apply_bert_masking(
         & (replacement_probs < (p_masked + p_random))
     )
 
+    masked_data = dict(data_batch)
+
     for col, tensor in data_batch.items():
         if col in config.categorical_columns:
-            random_tokens = torch.randint(
-                low=SPECIAL_TOKEN_IDS.user_start,
-                high=config.n_classes[col],
-                size=(batch_size, seq_len),
-                device=device,
-                dtype=tensor.dtype,
-            )
+            output = tensor.clone()
+
+            if p_masked > 0.0:
+                output.masked_fill_(mask_token_mask, SPECIAL_TOKEN_IDS.mask)
+
+            if p_random > 0.0:
+                random_tokens = torch.randint(
+                    low=SPECIAL_TOKEN_IDS.user_start,
+                    high=config.n_classes[col],
+                    size=tensor.shape,
+                    device=device,
+                    dtype=tensor.dtype,
+                    generator=generator,
+                )
+                output[random_token_mask] = random_tokens[random_token_mask]
 
-            tensor[mask_token_mask] = SPECIAL_TOKEN_IDS.mask
-            tensor[random_token_mask] = random_tokens[random_token_mask]
+            masked_data[col] = output
 
         elif col in config.real_columns:
-            mask_val = 0.0
-            random_noise = torch.randn(
-                (batch_size, seq_len), device=device, dtype=tensor.dtype
-            )
+            output = tensor.clone()
+
+            if p_masked > 0.0:
+                output.masked_fill_(mask_token_mask, 0.0)
 
-            tensor[mask_token_mask] = mask_val
-            tensor[random_token_mask] = random_noise[random_token_mask]
+            if p_random > 0.0:
+                random_noise = torch.randn(
+                    tensor.shape,
+                    device=device,
+                    dtype=tensor.dtype,
+                    generator=generator,
+                )
+                output[random_token_mask] = random_noise[random_token_mask]
 
-    metadata_batch["bert_mask"] = bert_mask
-    metadata_batch["attention_valid_mask"] = valid_mask.detach()
+            masked_data[col] = output
 
-    if eval_seed is not None:
-        torch.set_rng_state(cpu_rng_state)  # type: ignore
-        if device.type == "cuda":
-            torch.cuda.set_rng_state(gpu_rng_state, device)  # type: ignore
-    return data_batch, targets_batch, metadata_batch
+    detached_targets = {col: tensor.detach() for col, tensor in targets_batch.items()}
+    output_metadata = {key: tensor.detach() for key, tensor in metadata_batch.items()}
+    output_metadata["bert_mask"] = bert_mask
+    output_metadata["attention_valid_mask"] = valid_mask
+
+    return masked_data, detached_targets, output_metadata
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
index 53856f92..360f04df 100644
--- a/tests/unit/test_helpers.py
+++ b/tests/unit/test_helpers.py
@@ -1,6 +1,7 @@
 from types import SimpleNamespace
 
 import polars as pl
+import pytest
 import torch
 
 from sequifier.helpers import (
@@ -12,6 +13,7 @@
     numpy_to_pytorch,
     resolve_window_view,
 )
+from sequifier.special_tokens import SPECIAL_TOKEN_IDS
 
 
 def test_construct_index_maps_string():
@@ -248,10 +250,333 @@ def test_numpy_to_pytorch_includes_explicit_padding_masks():
 
 
 class _OnesSpanMasking:
-    def sample(self, shape, device):
+    def sample(self, shape, device, generator=None):
         return torch.ones(shape, dtype=torch.long, device=device)
 
 
+class _VariableSpanMasking:
+    def sample(self, shape, device, generator=None):
+        lengths = torch.tensor([3, 2, 4, 1, 5], dtype=torch.long, device=device)
+        repeats = (shape[1] + lengths.numel() - 1) // lengths.numel()
+        return lengths.repeat(repeats)[: shape[1]].repeat(shape[0], 1)
+
+
+def _bert_masking_config(
+    masking_probability=0.5,
+    *,
+    categorical_columns=None,
+    real_columns=None,
+    n_classes=None,
+    replacement_distribution=None,
+    span_masking=None,
+):
+    return SimpleNamespace(
+        categorical_columns=categorical_columns or [],
+        real_columns=real_columns or [],
+        n_classes=n_classes or {},
+        context_length=6,
+        training_spec=SimpleNamespace(
+            batch_size=4,
+            bert_spec=SimpleNamespace(
+                masking_probability=masking_probability,
+                span_masking=span_masking or _OnesSpanMasking(),
+                replacement_distribution=replacement_distribution
+                or SimpleNamespace(masked=0.0, random=0.0, identical=1.0),
+            ),
+        ),
+    )
+
+
+def test_apply_bert_masking_builds_exact_valid_masks():
+    valid_mask = torch.tensor(
+        [
+            [True, True, True, True, True, True],
+            [False, False, True, True, True, True],
+            [False, False, False, False, False, False],
+            [False, True, False, True, True, True],
+        ]
+    )
+    config = _bert_masking_config(masking_probability=0.5)
+    data_batch = {"passthrough": torch.arange(24).reshape(4, 6)}
+    targets_batch = {"passthrough": torch.arange(24).reshape(4, 6)}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    _, _, masked_metadata = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=11,
+    )
+
+    bert_mask = masked_metadata["bert_mask"]
+    expected_budget = (valid_mask.sum(dim=1) * 0.5).long()
+
+    assert bert_mask.dtype == torch.bool
+    assert bert_mask.device == valid_mask.device
+    assert torch.equal(bert_mask.sum(dim=1), expected_budget)
+    assert not torch.any(bert_mask & ~valid_mask)
+
+
+def test_apply_bert_masking_eval_seed_is_reproducible_without_global_rng_change():
+    valid_mask = torch.tensor(
+        [
+            [True, True, True, True, True, True],
+            [False, True, True, True, True, True],
+        ]
+    )
+    config = _bert_masking_config(masking_probability=0.5)
+    data_batch = {"passthrough": torch.arange(12).reshape(2, 6)}
+    targets_batch = {"passthrough": torch.arange(12).reshape(2, 6)}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    torch.manual_seed(123)
+    rng_state = torch.get_rng_state().clone()
+    _, _, first_metadata = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=99,
+    )
+    after_first = torch.get_rng_state().clone()
+
+    _, _, second_metadata = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=99,
+    )
+    after_second = torch.get_rng_state().clone()
+
+    assert torch.equal(first_metadata["bert_mask"], second_metadata["bert_mask"])
+    assert torch.equal(after_first, rng_state)
+    assert torch.equal(after_second, rng_state)
+
+
+def test_apply_bert_masking_handles_variable_length_spans_and_sparse_masks():
+    valid_mask = torch.tensor(
+        [
+            [True, True, True, True, True, True, True, True],
+            [False, True, False, True, True, False, True, True],
+        ]
+    )
+    config = _bert_masking_config(
+        masking_probability=0.75,
+        span_masking=_VariableSpanMasking(),
+    )
+    data_batch = {"passthrough": torch.arange(16).reshape(2, 8)}
+    targets_batch = {"passthrough": torch.arange(16).reshape(2, 8)}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    _, _, masked_metadata = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=99,
+    )
+
+    assert torch.equal(
+        masked_metadata["bert_mask"],
+        torch.tensor(
+            [
+                [False, True, True, True, True, True, True, False],
+                [False, True, False, True, True, False, False, False],
+            ]
+        ),
+    )
+
+
+@pytest.mark.skipif(not torch.cuda.is_available(), reason="CUDA required")
+def test_apply_bert_masking_cuda_stays_on_device_and_preserves_rng():
+    device = torch.device("cuda")
+    valid_mask = torch.tensor(
+        [
+            [True, True, True, True, True, True],
+            [False, True, True, True, True, True],
+        ],
+        device=device,
+    )
+    config = _bert_masking_config(masking_probability=0.5)
+    data_batch = {"passthrough": torch.arange(12, device=device).reshape(2, 6)}
+    targets_batch = {"passthrough": torch.arange(12, device=device).reshape(2, 6)}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    torch.manual_seed(123)
+    cpu_state = torch.get_rng_state().clone()
+    cuda_state = torch.cuda.get_rng_state(device).clone()
+
+    _, _, masked_metadata = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=99,
+    )
+
+    assert masked_metadata["bert_mask"].device == device
+    assert torch.equal(torch.get_rng_state(), cpu_state)
+    assert torch.equal(torch.cuda.get_rng_state(device), cuda_state)
+
+
+def test_apply_bert_masking_replaces_categorical_with_mask_token_without_mutation():
+    valid_mask = torch.tensor([[True, False, True, True]])
+    config = _bert_masking_config(
+        masking_probability=1.0,
+        categorical_columns=["cat_col"],
+        n_classes={"cat_col": 8},
+        replacement_distribution=SimpleNamespace(
+            masked=1.0,
+            random=0.0,
+            identical=0.0,
+        ),
+    )
+    original = torch.tensor([[3, 4, 5, 6]])
+    data_batch = {"cat_col": original.clone()}
+    targets_batch = {"cat_col": original.clone()}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    masked_data, _, _ = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=7,
+    )
+
+    assert torch.equal(
+        masked_data["cat_col"],
+        torch.tensor(
+            [
+                [
+                    SPECIAL_TOKEN_IDS.mask,
+                    4,
+                    SPECIAL_TOKEN_IDS.mask,
+                    SPECIAL_TOKEN_IDS.mask,
+                ]
+            ]
+        ),
+    )
+    assert torch.equal(data_batch["cat_col"], original)
+
+
+def test_apply_bert_masking_replaces_categorical_with_random_tokens_without_mutation():
+    valid_mask = torch.ones((1, 5), dtype=torch.bool)
+    config = _bert_masking_config(
+        masking_probability=1.0,
+        categorical_columns=["cat_col"],
+        n_classes={"cat_col": 12},
+        replacement_distribution=SimpleNamespace(
+            masked=0.0,
+            random=1.0,
+            identical=0.0,
+        ),
+    )
+    original = torch.tensor([[3, 4, 5, 6, 7]])
+    data_batch = {"cat_col": original.clone()}
+    targets_batch = {"cat_col": original.clone()}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    masked_data, _, _ = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=7,
+    )
+
+    assert torch.all(masked_data["cat_col"] >= SPECIAL_TOKEN_IDS.user_start)
+    assert torch.all(masked_data["cat_col"] < config.n_classes["cat_col"])
+    assert not torch.equal(masked_data["cat_col"], original)
+    assert torch.equal(data_batch["cat_col"], original)
+
+
+def test_apply_bert_masking_replaces_real_with_zero_without_mutation():
+    valid_mask = torch.tensor([[True, False, True, True]])
+    config = _bert_masking_config(
+        masking_probability=1.0,
+        real_columns=["real_col"],
+        replacement_distribution=SimpleNamespace(
+            masked=1.0,
+            random=0.0,
+            identical=0.0,
+        ),
+    )
+    original = torch.tensor([[1.5, 2.5, 3.5, 4.5]])
+    data_batch = {"real_col": original.clone()}
+    targets_batch = {"real_col": original.clone()}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    masked_data, _, _ = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=7,
+    )
+
+    assert torch.equal(
+        masked_data["real_col"],
+        torch.tensor([[0.0, 2.5, 0.0, 0.0]]),
+    )
+    assert torch.equal(data_batch["real_col"], original)
+
+
+def test_apply_bert_masking_replaces_real_with_random_noise_without_mutation():
+    valid_mask = torch.tensor([[True, True, False, True]])
+    config = _bert_masking_config(
+        masking_probability=1.0,
+        real_columns=["real_col"],
+        replacement_distribution=SimpleNamespace(
+            masked=0.0,
+            random=1.0,
+            identical=0.0,
+        ),
+    )
+    original = torch.tensor([[10.0, 11.0, 12.0, 13.0]])
+    data_batch = {"real_col": original.clone()}
+    targets_batch = {"real_col": original.clone()}
+    metadata = {
+        "attention_valid_mask": valid_mask,
+        "target_valid_mask": valid_mask.clone(),
+    }
+
+    masked_data, _, _ = apply_bert_masking(
+        data_batch,
+        targets_batch,
+        metadata,
+        config,
+        eval_seed=7,
+    )
+
+    assert torch.equal(masked_data["real_col"][~valid_mask], original[~valid_mask])
+    assert torch.all(masked_data["real_col"][valid_mask] != original[valid_mask])
+    assert torch.equal(data_batch["real_col"], original)
+
+
 def test_apply_bert_masking_uses_explicit_valid_mask_for_zero_values():
     config = SimpleNamespace(
         categorical_columns=[],
diff --git a/tests/unit/test_probabilities.py b/tests/unit/test_probabilities.py
new file mode 100644
index 00000000..d413d841
--- /dev/null
+++ b/tests/unit/test_probabilities.py
@@ -0,0 +1,51 @@
+import pytest
+import torch
+
+from sequifier.config.probabilities import (
+    GeometricDistribution,
+    LogNormalDistributionDiscretizedFloor,
+    NormalDistributionDiscretizedFloor,
+    PoissonDistributionFloor,
+)
+
+
+@pytest.mark.parametrize(
+    "distribution",
+    [
+        GeometricDistribution(p=0.35),
+        NormalDistributionDiscretizedFloor(mean=2.0, standard_deviation=0.5),
+        LogNormalDistributionDiscretizedFloor(mean=0.5, standard_deviation=0.4),
+        PoissonDistributionFloor(rate=2.0),
+    ],
+)
+def test_probability_distributions_use_local_generator_without_global_rng(
+    distribution,
+):
+    device = torch.device("cpu")
+    torch.manual_seed(123)
+    rng_state = torch.get_rng_state().clone()
+
+    first_generator = torch.Generator(device=device)
+    first_generator.manual_seed(99)
+    first_samples = distribution.sample(
+        (64,),
+        device=device,
+        generator=first_generator,
+    )
+    after_first = torch.get_rng_state().clone()
+
+    second_generator = torch.Generator(device=device)
+    second_generator.manual_seed(99)
+    second_samples = distribution.sample(
+        (64,),
+        device=device,
+        generator=second_generator,
+    )
+    after_second = torch.get_rng_state().clone()
+
+    assert torch.equal(first_samples, second_samples)
+    assert first_samples.device == device
+    assert first_samples.dtype == torch.long
+    assert torch.all(first_samples >= 1)
+    assert torch.equal(after_first, rng_state)
+    assert torch.equal(after_second, rng_state)

From 5e620f3f2fc8ad3335f4acab44f8f5f0087c944b Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 15 Jun 2026 22:04:50 +0200
Subject: [PATCH 63/81] improve tests

---
 src/sequifier/infer.py                        |  32 +-
 src/sequifier/preprocess.py                   |  65 ++--
 tests/integration/test_inference.py           | 320 +++++++++++++++---
 ...uifier_dataset_from_folder_parquet_lazy.py |  77 ++++-
 tests/unit/test_infer.py                      |  72 +++-
 tests/unit/test_preprocess.py                 |  21 ++
 tests/unit/test_train.py                      |   2 +-
 tests/unit/test_train_distributed_loss.py     | 185 +++++++++-
 tests/unit/test_train_fsdp_loss.py            | 189 ++++++++++-
 9 files changed, 844 insertions(+), 119 deletions(-)

diff --git a/src/sequifier/infer.py b/src/sequifier/infer.py
index 57b9093f..df26023b 100644
--- a/src/sequifier/infer.py
+++ b/src/sequifier/infer.py
@@ -1368,26 +1368,40 @@ def expand_to_batch_size(self, x: np.ndarray) -> np.ndarray:
 @beartype
 def normalize(outs: dict[str, np.ndarray]) -> dict[str, np.ndarray]:
     """Softmax logits by target column."""
-    normalizer = {
-        target_column: np.repeat(
-            np.sum(np.exp(target_values), axis=1), target_values.shape[1]
-        ).reshape(target_values.shape)
+    shifted_values = {
+        target_column: target_values - np.max(target_values, axis=1, keepdims=True)
         for target_column, target_values in outs.items()
     }
+    exp_values = {
+        target_column: np.exp(target_values)
+        for target_column, target_values in shifted_values.items()
+    }
     probs = {
-        target_column: np.exp(target_values) / normalizer[target_column]
-        for target_column, target_values in outs.items()
+        target_column: target_values / np.sum(target_values, axis=1, keepdims=True)
+        for target_column, target_values in exp_values.items()
     }
     return probs
 
 
 @beartype
 def sample_with_cumsum(probs: np.ndarray, is_log_probs: bool = True) -> np.ndarray:
-    """Sample class indices from logits or probabilities."""
+    """Sample class indices from log-probabilities or probabilities."""
     if is_log_probs:
-        cumulative_probs = np.cumsum(np.exp(probs), axis=1)
+        sampling_probs = np.exp(probs)
     else:
-        cumulative_probs = np.cumsum(probs, axis=1)
+        sampling_probs = probs
+
+    if not np.isfinite(sampling_probs).all():
+        raise ValueError("Sampling probabilities must be finite.")
+    if np.any(sampling_probs < 0):
+        raise ValueError("Sampling probabilities must be non-negative.")
+
+    row_sums = sampling_probs.sum(axis=1)
+    if not np.allclose(row_sums, 1.0):
+        raise ValueError("Sampling probabilities must sum to 1.0 for each row.")
+
+    cumulative_probs = np.cumsum(sampling_probs, axis=1)
+    cumulative_probs[:, -1] = 1.0
     random_threshold = np.random.rand(cumulative_probs.shape[0], 1)
     random_threshold = np.repeat(random_threshold, probs.shape[1], axis=1)
     return (random_threshold < cumulative_probs).argmax(axis=1)
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 9a2f4938..cd5bd024 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -1518,32 +1518,51 @@ def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]
 @beartype
 def get_batch_limits(data: pl.DataFrame, n_batches: int) -> list[tuple[int, int]]:
     """Split rows into batches without crossing sequenceId boundaries."""
+    if n_batches <= 0:
+        raise ValueError("n_batches must be positive.")
+    if data.is_empty():
+        raise ValueError("Cannot split an empty dataset into batches.")
+
     sequence_ids = data.get_column("sequenceId").to_numpy()
-    new_sequence_id_indices = np.concatenate(
-        [
-            [0],
-            np.where(
-                np.concatenate([[False], sequence_ids[1:] != sequence_ids[:-1]], axis=0)
-            )[0],
-        ]
+    sequence_start_indices = np.concatenate(
+        [[0], np.where(sequence_ids[1:] != sequence_ids[:-1])[0] + 1]
     )
+    sequence_boundaries = np.concatenate([sequence_start_indices, [data.shape[0]]])
+    sequence_count = len(sequence_start_indices)
 
-    ideal_step = math.ceil(data.shape[0] / n_batches)
-    ideal_limits = np.array(
-        [ideal_step * m for m in range(n_batches)] + [data.shape[0]]
-    )
-    distances = [
-        np.abs(new_sequence_id_indices - ideal_limit)
-        for ideal_limit in ideal_limits[:-1]
-    ]
-    actual_limit_indices = [
-        np.where(distance == np.min(distance))[0] for distance in distances
-    ]
-    actual_limits = [
-        int(new_sequence_id_indices[limit_index[0]])
-        for limit_index in actual_limit_indices
-    ] + [data.shape[0]]
-    return list(zip(actual_limits[:-1], actual_limits[1:]))
+    if n_batches > sequence_count:
+        raise ValueError(
+            "Cannot create more non-empty batches than there are sequences without "
+            "splitting a sequence."
+        )
+
+    interior_boundaries = sequence_boundaries[1:-1]
+    ideal_limits = np.linspace(0, data.shape[0], n_batches + 1)[1:-1]
+
+    selected_boundaries: list[int] = []
+    previous_boundary = 0
+    for batch_index, ideal_limit in enumerate(ideal_limits):
+        remaining_boundaries_needed = len(ideal_limits) - batch_index - 1
+        candidates = [
+            int(boundary)
+            for boundary in interior_boundaries
+            if boundary > previous_boundary
+            and (data.shape[0] - boundary) >= remaining_boundaries_needed
+        ]
+        if not candidates:
+            raise ValueError(
+                "Cannot create requested non-empty batches without splitting a sequence."
+            )
+
+        selected_boundary = min(
+            candidates,
+            key=lambda boundary: abs(boundary - ideal_limit),
+        )
+        selected_boundaries.append(selected_boundary)
+        previous_boundary = selected_boundary
+
+    limits = [0, *selected_boundaries, data.shape[0]]
+    return list(zip(limits[:-1], limits[1:]))
 
 
 @beartype
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
index 5cc4ec89..1f9d9887 100644
--- a/tests/integration/test_inference.py
+++ b/tests/integration/test_inference.py
@@ -1,14 +1,254 @@
 import json
 import os
+import re
+from collections import Counter
 
 import numpy as np
 import polars as pl
 import torch
 
+from sequifier.helpers import (
+    ModelWindowView,
+    StoredWindowLayout,
+    resolve_window_view,
+    stored_window_layout_from_metadata,
+)
+
 TARGET_VARIABLE_DICT = {"categorical": "itemId", "real": "itemValue"}
 BERT_SEQ_LENGTH = 8
 BERT_EMBEDDING_DIM = 16
 BERT_DATA_NAME = "test-data-categorical-1-lookahead-0"
+CAUSAL_CONTEXT_LENGTH = 8
+
+
+def _project_path(project_root, path):
+    if os.path.isabs(path) or path.startswith(project_root):
+        return path
+    return os.path.join(project_root, path)
+
+
+def _causal_storage_layout(data_path):
+    metadata_path = (
+        os.path.join(data_path, "metadata.json") if os.path.isdir(data_path) else None
+    )
+    if metadata_path is not None and os.path.exists(metadata_path):
+        with open(metadata_path, "r") as f:
+            return stored_window_layout_from_metadata(json.load(f))
+
+    return StoredWindowLayout(
+        stored_context_width=CAUSAL_CONTEXT_LENGTH + 1,
+        max_target_offset=1,
+        version=2,
+    )
+
+
+def _target_valid_mask(left_pad_lengths, storage_layout, prediction_length):
+    resolved_view = resolve_window_view(
+        storage_layout,
+        ModelWindowView(
+            context_length=CAUSAL_CONTEXT_LENGTH,
+            objective="causal",
+            target_offset=1,
+        ),
+    )
+    masks = resolved_view.build_masks(torch.tensor(left_pad_lengths, dtype=torch.int64))
+    return masks["target_valid_mask"][:, -prediction_length:].reshape(-1).numpy()
+
+
+def _read_pt_window_metadata(data_path):
+    contents = []
+    for root, _, files in os.walk(data_path):
+        for file in sorted(files):
+            if not file.endswith(".pt"):
+                continue
+
+            loaded = torch.load(os.path.join(root, file), weights_only=False)
+            if len(loaded) == 5:
+                _, sequence_ids, _, start_positions, left_pad_lengths = loaded
+            else:
+                _, sequence_ids, _, start_positions = loaded
+                left_pad_lengths = torch.zeros_like(sequence_ids)
+
+            contents.append(
+                pl.DataFrame(
+                    {
+                        "sequenceId": sequence_ids.detach().cpu().numpy(),
+                        "startItemPosition": start_positions.detach().cpu().numpy(),
+                        "leftPadLength": left_pad_lengths.detach().cpu().numpy(),
+                    }
+                )
+            )
+
+    assert len(contents) > 0, f"no files found for {data_path}"
+    return pl.concat(contents, how="vertical")
+
+
+def _read_long_window_metadata(data_path):
+    paths = []
+    if os.path.isdir(data_path):
+        for root, _, files in os.walk(data_path):
+            paths.extend(
+                os.path.join(root, file)
+                for file in sorted(files)
+                if file.endswith((".csv", ".parquet"))
+            )
+    else:
+        paths = [data_path]
+
+    contents = []
+    for path in paths:
+        data = pl.read_csv(path) if path.endswith(".csv") else pl.read_parquet(path)
+        if "leftPadLength" not in data.columns:
+            data = data.with_columns(pl.lit(0).alias("leftPadLength"))
+        contents.append(
+            data.group_by(["sequenceId", "subsequenceId"], maintain_order=True).agg(
+                pl.col("startItemPosition").first(),
+                pl.col("leftPadLength").first(),
+            )
+        )
+
+    assert len(contents) > 0, f"no files found for {data_path}"
+    return pl.concat(contents, how="vertical")
+
+
+def _prediction_source_spec(project_root, model_name):
+    if model_name == "model-categorical-multitarget-5-best-3":
+        return {
+            "path": _project_path(
+                project_root, "data/test-data-categorical-multitarget-5-split2"
+            ),
+            "format": "parquet",
+            "prediction_length": 1,
+            "autoregression_total_steps": None,
+        }
+    if model_name == "model-categorical-multitarget-5-last-3":
+        return {
+            "path": _project_path(
+                project_root, "data/test-data-categorical-multitarget-5-split2"
+            ),
+            "format": "parquet",
+            "prediction_length": 1,
+            "autoregression_total_steps": None,
+        }
+    if model_name == "model-categorical-distributed-best-3":
+        return {
+            "path": _project_path(project_root, "data/test-data-categorical-3-split2"),
+            "format": "pt",
+            "prediction_length": 1,
+            "autoregression_total_steps": None,
+        }
+    if model_name == "model-categorical-lazy-best-3":
+        return {
+            "path": _project_path(project_root, "data/test-data-categorical-3-split2"),
+            "format": "pt",
+            "prediction_length": 1,
+            "autoregression_total_steps": None,
+        }
+    if model_name == "model-categorical-1-best-3-autoregression":
+        return {
+            "path": _project_path(project_root, "data/test-data-categorical-1-split2"),
+            "format": "pt",
+            "prediction_length": 1,
+            "autoregression_total_steps": 20,
+        }
+    if model_name == "model-real-1-best-3-autoregression":
+        return {
+            "path": _project_path(
+                project_root, "data/test-data-real-1-split1-autoregression.csv"
+            ),
+            "format": "csv",
+            "prediction_length": 1,
+            "autoregression_total_steps": 20,
+            "csv_autoregression": True,
+        }
+
+    match = re.fullmatch(r"model-categorical-(\d+)-inf-size-best-3", model_name)
+    if match is not None:
+        data_number = int(match.group(1))
+        return {
+            "path": _project_path(
+                project_root, f"data/test-data-categorical-{data_number}-split2"
+            ),
+            "format": "pt",
+            "prediction_length": 3,
+            "autoregression_total_steps": None,
+        }
+
+    match = re.fullmatch(r"model-categorical-(\d+)-best-3", model_name)
+    if match is not None:
+        data_number = int(match.group(1))
+        return {
+            "path": _project_path(
+                project_root, f"data/test-data-categorical-{data_number}-split2"
+            ),
+            "format": "pt",
+            "prediction_length": 1,
+            "autoregression_total_steps": None,
+        }
+
+    match = re.fullmatch(r"model-real-(\d+)-best-3", model_name)
+    if match is not None:
+        data_number = int(match.group(1))
+        return {
+            "path": _project_path(
+                project_root, f"data/test-data-real-{data_number}-split1.parquet"
+            ),
+            "format": "parquet",
+            "prediction_length": 1,
+            "autoregression_total_steps": None,
+        }
+
+    raise AssertionError(f"No source metadata mapping for {model_name}")
+
+
+def _expected_prediction_positions(project_root, model_name):
+    spec = _prediction_source_spec(project_root, model_name)
+    metadata = (
+        _read_pt_window_metadata(spec["path"])
+        if spec["format"] == "pt"
+        else _read_long_window_metadata(spec["path"])
+    )
+    storage_layout = _causal_storage_layout(spec["path"])
+    prediction_length = spec["prediction_length"]
+
+    if spec["autoregression_total_steps"] is not None:
+        total_steps = spec["autoregression_total_steps"]
+        if spec.get("csv_autoregression", False):
+            metadata = metadata.group_by("sequenceId", maintain_order=True).head(1)
+
+        sequence_ids = np.repeat(metadata["sequenceId"].to_numpy(), total_steps)
+        item_positions = np.concatenate(
+            [
+                np.arange(start, start + total_steps)
+                for start in (
+                    metadata["startItemPosition"].to_numpy() + CAUSAL_CONTEXT_LENGTH
+                )
+            ],
+            axis=0,
+        )
+        valid_mask = np.repeat(
+            _target_valid_mask(
+                metadata["leftPadLength"].to_numpy(), storage_layout, prediction_length
+            ),
+            total_steps,
+        )
+    else:
+        starts = metadata["startItemPosition"].to_numpy()
+        offsets = np.arange(-prediction_length + 1, 1)
+        sequence_ids = np.repeat(metadata["sequenceId"].to_numpy(), prediction_length)
+        item_positions = np.repeat(
+            starts + CAUSAL_CONTEXT_LENGTH, prediction_length
+        ) + np.tile(offsets, len(starts))
+        valid_mask = _target_valid_mask(
+            metadata["leftPadLength"].to_numpy(), storage_layout, prediction_length
+        )
+
+    return pl.DataFrame(
+        {
+            "sequenceId": sequence_ids[valid_mask],
+            "itemPosition": item_positions[valid_mask],
+        }
+    )
 
 
 def _categorical_metadata(project_root):
@@ -260,67 +500,37 @@ def test_bert_embeddings(bert_predictions, bert_embeddings, project_root):
     )
 
 
-def test_predictions_item_position(predictions):
-    """
-    Checks if itemPosition increments correctly within each sequenceId.
-    """
+def test_predictions_item_position(predictions, project_root):
+    """Check itemPosition values against source preprocessing metadata."""
     for model_name, preds_df in predictions.items():
-        # Ensure correct sorting for comparison
-        preds_df_sorted = preds_df.sort("sequenceId", "itemPosition")
+        expected_positions = _expected_prediction_positions(project_root, model_name)
+        actual_pairs = list(preds_df.select(["sequenceId", "itemPosition"]).iter_rows())
+        expected_pairs = list(expected_positions.iter_rows())
 
-        # Calculate differences and sequence changes
-        preds_with_diffs = preds_df_sorted.with_columns(
-            (pl.col("itemPosition") - pl.col("itemPosition").shift(1)).alias(
-                "pos_diff"
-            ),
-            (pl.col("sequenceId") == pl.col("sequenceId").shift(1)).alias("same_seq"),
+        duplicate_pairs = [
+            pair for pair, count in Counter(actual_pairs).items() if count > 1
+        ]
+        assert duplicate_pairs == [], (
+            f"Model '{model_name}': Found duplicate sequence-position pairs:\n"
+            f"{duplicate_pairs[:10]}"
         )
 
-        # Filter for rows within the same sequence (excluding the first row of each sequence)
-        within_sequence_diffs = preds_with_diffs.filter(pl.col("same_seq"))
-
-        # Check if all position differences within sequences are 1
-        incorrect_increments = within_sequence_diffs.filter(pl.col("pos_diff") != 1)
-
-        assert incorrect_increments.height == 0, (
-            f"Model '{model_name}': Found incorrect itemPosition increments within sequences:\n"
-            f"{incorrect_increments}"
-        )
+        assert Counter(actual_pairs) == Counter(
+            expected_pairs
+        ), f"Model '{model_name}': itemPosition values do not match source metadata."
 
 
 def test_embeddings_subsequence_id(embeddings):
     """Check subsequenceId increments within each sequence."""
     for model_name, embeds_df in embeddings.items():
-        embeds_df_sorted = embeds_df.sort("sequenceId", "subsequenceId")
-
-        shift_val = 0
-        if "categorical-1" in model_name:
-            shift_val = 1
-        if "categorical-3" in model_name:
-            shift_val = 3
-
-        # Calculate differences and sequence changes
-        embeds_with_diffs = embeds_df_sorted.with_columns(
-            (pl.col("subsequenceId") - pl.col("subsequenceId").shift(shift_val)).alias(
-                "subseq_diff"
-            ),
-            (pl.col("sequenceId") == pl.col("sequenceId").shift(shift_val)).alias(
-                "same_seq"
-            ),
-        )
-
-        first_subsequences = embeds_with_diffs.filter(
-            pl.col("same_seq") == False  # noqa: E712
-        )
-        incorrect_starts = first_subsequences.filter(pl.col("subsequenceId") != 0)
-        assert incorrect_starts.height == 0, (
-            f"Model '{model_name}': Found sequences where subsequenceId does not start at 0:\n"
-            f"{incorrect_starts}"
-        )
-
-        within_sequence_diffs = embeds_with_diffs.filter(pl.col("same_seq"))
-        incorrect_increments = within_sequence_diffs.filter(pl.col("subseq_diff") != 1)
-        assert incorrect_increments.height == 0, (
-            f"Model '{model_name}': Found incorrect subsequenceId increments within sequences:\n"
-            f"{incorrect_increments}"
-        )
+        for sequence_id, sequence_df in embeds_df.group_by(
+            "sequenceId", maintain_order=True
+        ):
+            subsequence_ids = sorted(
+                sequence_df.get_column("subsequenceId").unique().to_list()
+            )
+            expected_ids = list(range(len(subsequence_ids)))
+            assert subsequence_ids == expected_ids, (
+                f"Model '{model_name}', sequenceId {sequence_id}: expected "
+                f"subsequenceIds {expected_ids}, found {subsequence_ids}."
+            )
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index 3d61a5ed..baa87136 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -64,13 +64,25 @@ def dataset_path(tmp_path):
         "0": pl.Float64,
     }
 
-    # Populate 4 files with 10 rows (10 sequences) each
+    # Populate 4 files with distinct row markers so sharding can be observed.
     batch_files = []
     for i in range(1, 5):
         filename = f"file_{i}.parquet"
         rows = []
         for s in range(10):
-            rows.append((s, 0, s * 2, 0, "item", float(s), float(s + 1), float(s + 2)))
+            marker = i * 100 + s
+            rows.append(
+                (
+                    s,
+                    0,
+                    s * 2,
+                    0,
+                    "item",
+                    float(marker),
+                    float(marker + 1),
+                    float(marker + 2),
+                )
+            )
 
         df = pl.DataFrame(rows, schema=schema, orient="row")
         df.write_parquet(data_dir / filename)
@@ -84,6 +96,16 @@ def dataset_path(tmp_path):
     return str(data_dir)
 
 
+def _input_markers(batches):
+    return torch.cat([batch.inputs["item"][:, 0] for batch in batches]).tolist()
+
+
+def _sample_valid_mask(batches):
+    return torch.cat(
+        [batch.metadata["sample_valid_mask"] for batch in batches]
+    ).tolist()
+
+
 def test_initialization(mock_config, dataset_path):
     """Metadata-backed batch length."""
     dataset = SequifierDatasetFromFolderParquetLazy(dataset_path, mock_config)
@@ -194,6 +216,12 @@ def test_distributed_sharding(mock_ws, mock_rank, mock_init, mock_config, datase
     for batch in batches:
         assert batch.inputs["item"].shape[0] == 5
 
+    assert _input_markers(batches) == [
+        *[float(100 + i) for i in range(10)],
+        *[float(300 + i) for i in range(10)],
+    ]
+    assert _sample_valid_mask(batches) == [True] * 20
+
 
 def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
     """FSDP exact mode rejects uneven ranks."""
@@ -218,7 +246,8 @@ def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
         assert "different number of samples per rank/GPU" in str(exc_info.value)
 
 
-def test_oversampling_strategy(mock_config, tmp_path):
+@patch("torch.distributed.get_rank", return_value=1)
+def test_oversampling_strategy(mock_rank, mock_config, tmp_path):
     """Oversampling pads short ranks."""
     data_dir = tmp_path / "oversample_parquet_data"
     data_dir.mkdir()
@@ -227,6 +256,7 @@ def test_oversampling_strategy(mock_config, tmp_path):
         "sequenceId": pl.Int64,
         "subsequenceId": pl.Int64,
         "startItemPosition": pl.Int64,
+        "leftPadLength": pl.Int64,
         "inputCol": pl.String,
         "2": pl.Float64,
         "1": pl.Float64,
@@ -236,7 +266,16 @@ def test_oversampling_strategy(mock_config, tmp_path):
     # File 1 has 15 rows, File 2 has 10 rows
     for i, num_rows in [(1, 15), (2, 10)]:
         rows = [
-            (s, 0, s * 2, "item", float(s), float(s + 1), float(s + 2))
+            (
+                s,
+                0,
+                s * 2,
+                0,
+                "item",
+                float(i * 100 + s),
+                float(i * 100 + s + 1),
+                float(i * 100 + s + 2),
+            )
             for s in range(num_rows)
         ]
         pl.DataFrame(rows, schema=schema, orient="row").write_parquet(
@@ -261,9 +300,19 @@ def test_oversampling_strategy(mock_config, tmp_path):
         )
         # Should match max(15, 10)
         assert dataset.target_samples == 15
+        assert len(dataset) == 3
+
+        batches = list(dataset)
+        assert len(batches) == 3
+        assert _input_markers(batches) == [
+            *[float(200 + i) for i in range(10)],
+            *[float(200 + i) for i in range(5)],
+        ]
+        assert _sample_valid_mask(batches) == [True] * 10 + [False] * 5
 
 
-def test_undersampling_strategy(mock_config, tmp_path):
+@patch("torch.distributed.get_rank", return_value=0)
+def test_undersampling_strategy(mock_rank, mock_config, tmp_path):
     """Undersampling truncates long ranks."""
     data_dir = tmp_path / "undersample_parquet_data"
     data_dir.mkdir()
@@ -272,6 +321,7 @@ def test_undersampling_strategy(mock_config, tmp_path):
         "sequenceId": pl.Int64,
         "subsequenceId": pl.Int64,
         "startItemPosition": pl.Int64,
+        "leftPadLength": pl.Int64,
         "inputCol": pl.String,
         "2": pl.Float64,
         "1": pl.Float64,
@@ -280,7 +330,16 @@ def test_undersampling_strategy(mock_config, tmp_path):
 
     for i, num_rows in [(1, 15), (2, 10)]:
         rows = [
-            (s, 0, s * 2, "item", float(s), float(s + 1), float(s + 2))
+            (
+                s,
+                0,
+                s * 2,
+                0,
+                "item",
+                float(i * 100 + s),
+                float(i * 100 + s + 1),
+                float(i * 100 + s + 2),
+            )
             for s in range(num_rows)
         ]
         pl.DataFrame(rows, schema=schema, orient="row").write_parquet(
@@ -305,3 +364,9 @@ def test_undersampling_strategy(mock_config, tmp_path):
         )
         # Should match min(15, 10)
         assert dataset.target_samples == 10
+        assert len(dataset) == 2
+
+        batches = list(dataset)
+        assert len(batches) == 2
+        assert _input_markers(batches) == [float(100 + i) for i in range(10)]
+        assert _sample_valid_mask(batches) == [True] * 10
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index acd000d8..2c9d8c78 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -45,16 +45,34 @@ def test_normalize():
     np.testing.assert_allclose(probs.sum(axis=1), [1.0, 1.0, 1.0])
 
 
+def test_normalize_is_stable_for_large_logits():
+    """Softmax remains finite for large-magnitude logits."""
+    outs = {
+        "target_col": np.array(
+            [[1000.0, 1001.0], [-1001.0, -1000.0], [1000.0, -1000.0]]
+        )
+    }
+
+    probs = normalize(outs)["target_col"]
+
+    assert np.isfinite(probs).all()
+    np.testing.assert_allclose(probs.sum(axis=1), [1.0, 1.0, 1.0])
+    np.testing.assert_allclose(
+        probs[0],
+        [1.0 / (1.0 + np.e), np.e / (1.0 + np.e)],
+    )
+
+
 @patch("numpy.random.rand")
 def test_sample_with_cumsum(mock_rand):
-    """Inverse-CDF sampling for logits and probabilities."""
+    """Inverse-CDF sampling for log-probabilities and probabilities."""
     # Mock the random thresholds to strictly control the sampling outcome.
     mock_rand.return_value = np.array([[0.05], [0.90]])
 
-    # Path 1: Test with logits=True (default)
-    raw_logits = np.array([[np.log(0.1), np.log(0.9)], [np.log(0.8), np.log(0.2)]])
-    sampled_from_logits = sample_with_cumsum(raw_logits, is_log_probs=True)
-    np.testing.assert_array_equal(sampled_from_logits, [0, 1])
+    # Path 1: Test with log-probabilities=True (default)
+    log_probs = np.array([[np.log(0.1), np.log(0.9)], [np.log(0.8), np.log(0.2)]])
+    sampled_from_log_probs = sample_with_cumsum(log_probs, is_log_probs=True)
+    np.testing.assert_array_equal(sampled_from_log_probs, [0, 1])
 
     # Path 2: Test with logits=False (pre-normalized probabilities)
     pure_probs = np.array([[0.1, 0.9], [0.8, 0.2]])
@@ -62,6 +80,32 @@ def test_sample_with_cumsum(mock_rand):
     np.testing.assert_array_equal(sampled_from_probs, [0, 1])
 
 
+@patch("numpy.random.rand")
+def test_sample_with_cumsum_clamps_final_cumulative_probability(mock_rand):
+    """Tiny floating-point deficits still allow the final class to be sampled."""
+    mock_rand.return_value = np.array([[0.9999999999999999]])
+    probs = np.array([[0.1, 0.8999999999999998]])
+
+    sampled = sample_with_cumsum(probs, is_log_probs=False)
+
+    np.testing.assert_array_equal(sampled, [1])
+
+
+def test_sample_with_cumsum_rejects_invalid_probability_mass():
+    """Invalid probability rows fail instead of falling through to class zero."""
+    with pytest.raises(ValueError, match="sum to 1.0"):
+        sample_with_cumsum(np.array([[0.4, 0.4]]), is_log_probs=False)
+
+
+@pytest.fixture
+def empty_parquet_path(tmp_path):
+    path = tmp_path / "empty.parquet"
+    pl.DataFrame({"target_col": []}, schema={"target_col": pl.Float64}).write_parquet(
+        path
+    )
+    return str(path)
+
+
 @pytest.fixture
 def mock_inferer():
     """Sets up an Inferer instance with mocked heavy dependencies (ONNX/PyTorch)."""
@@ -157,14 +201,16 @@ def test_inferer_prepare_inference_batches_split(mock_inferer):
     np.testing.assert_array_equal(batches[2]["cat_col"], [[5]])
 
 
-def test_infer_config_defaults_bert_prediction_length_to_context_length():
+def test_infer_config_defaults_bert_prediction_length_to_context_length(
+    empty_parquet_path,
+):
     config = InfererModel(
         project_root=".",
         metadata_config_path="dummy.json",
         model_path="dummy.onnx",
         model_type="generative",
         training_objective="bert",
-        data_path="tests/unit/data/empty.parquet",
+        data_path=empty_parquet_path,
         input_columns=["target_col"],
         categorical_columns=[],
         real_columns=["target_col"],
@@ -189,14 +235,14 @@ def test_infer_config_defaults_bert_prediction_length_to_context_length():
     assert config.prediction_length == config.window_view.context_length
 
 
-def test_infer_config_defaults_causal_prediction_length_to_one():
+def test_infer_config_defaults_causal_prediction_length_to_one(empty_parquet_path):
     config = InfererModel(
         project_root=".",
         metadata_config_path="dummy.json",
         model_path="dummy.onnx",
         model_type="generative",
         training_objective="causal",
-        data_path="tests/unit/data/empty.parquet",
+        data_path=empty_parquet_path,
         input_columns=["target_col"],
         categorical_columns=[],
         real_columns=["target_col"],
@@ -221,7 +267,7 @@ def test_infer_config_defaults_causal_prediction_length_to_one():
     assert config.prediction_length == 1
 
 
-def test_infer_config_rejects_bert_prediction_length_mismatch():
+def test_infer_config_rejects_bert_prediction_length_mismatch(empty_parquet_path):
     with pytest.raises(ValueError, match="prediction_length must be equal"):
         InfererModel(
             project_root=".",
@@ -229,7 +275,7 @@ def test_infer_config_rejects_bert_prediction_length_mismatch():
             model_path="dummy.onnx",
             model_type="generative",
             training_objective="bert",
-            data_path="tests/unit/data/empty.parquet",
+            data_path=empty_parquet_path,
             input_columns=["target_col"],
             categorical_columns=[],
             real_columns=["target_col"],
@@ -580,7 +626,7 @@ def capture_write(dataframe, path, write_format):
 
 
 @pytest.fixture
-def ar_config():
+def ar_config(empty_parquet_path):
     """Provides an actual InfererModel configuration for autoregressive inference."""
     return InfererModel(
         project_root=".",
@@ -588,7 +634,7 @@ def ar_config():
         model_path="dummy.onnx",
         model_type="generative",
         training_objective="causal",
-        data_path="tests/unit/data/empty.parquet",
+        data_path=empty_parquet_path,
         input_columns=["target_col"],
         categorical_columns=[],
         real_columns=["target_col"],
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
index cb222176..6fb2a5c0 100644
--- a/tests/unit/test_preprocess.py
+++ b/tests/unit/test_preprocess.py
@@ -625,6 +625,15 @@ def test_get_batch_limits_uneven_split():
 
     limits = get_batch_limits(data, n_batches=2)
 
+    assert len(limits) == 2
+    assert limits[0][0] == 0
+    assert limits[-1][1] == data.height
+    assert all(start < end for start, end in limits)
+    assert all(
+        left_end == right_start
+        for (_, left_end), (right_start, _) in zip(limits, limits[1:])
+    )
+
     # Check that split points are valid sequence boundaries
     for start, end in limits:
         # Start of batch must match start of a sequence (unless 0)
@@ -632,6 +641,18 @@ def test_get_batch_limits_uneven_split():
             prev_id = data["sequenceId"][start - 1]
             curr_id = data["sequenceId"][start]
             assert prev_id != curr_id, f"Batch split at {start} broke a sequence"
+        if end != data.height:
+            prev_id = data["sequenceId"][end - 1]
+            curr_id = data["sequenceId"][end]
+            assert prev_id != curr_id, f"Batch split at {end} broke a sequence"
+
+
+def test_get_batch_limits_rejects_too_many_nonempty_batches():
+    """Too many requested chunks fail instead of producing empty batches."""
+    data = pl.DataFrame({"sequenceId": [1, 2, 3, 3], "val": np.arange(4)})
+
+    with pytest.raises(ValueError, match="more non-empty batches"):
+        get_batch_limits(data, n_batches=4)
 
 
 def test_get_combined_statistics_logic():
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 3600d9c7..837337ff 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -561,7 +561,7 @@ def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
     assert torch.isclose(component_losses["real_col"], torch.tensor(2.0))
 
 
-def test_calculate_loss_uses_target_columns_for_fallback_mask_inference():
+def test_calculate_loss_uses_explicit_target_mask_for_target_columns():
     model = TransformerModel.__new__(TransformerModel)
     model.target_column_types = {"real_target": "real"}
     model.criterion = {"real_target": torch.nn.MSELoss(reduction="none")}
diff --git a/tests/unit/test_train_distributed_loss.py b/tests/unit/test_train_distributed_loss.py
index 51423489..99c06f0d 100644
--- a/tests/unit/test_train_distributed_loss.py
+++ b/tests/unit/test_train_distributed_loss.py
@@ -7,7 +7,9 @@
 import torch.distributed as dist
 import torch.multiprocessing as mp
 from torch.nn.parallel import DistributedDataParallel as DDP
+from torch.utils.data import DataLoader
 
+from sequifier.io.batch import SequifierBatch
 from sequifier.train import TransformerModel
 
 
@@ -20,6 +22,16 @@ def forward(self, seq_len):
         return {"real_col": self.param.expand(seq_len, 1, 1)}
 
 
+class _TinyTrainEpochRealModel(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.param = torch.nn.Parameter(torch.tensor(0.0))
+
+    def forward(self, data, metadata=None, return_logits=False):
+        seq_len = data["real_col"].shape[1]
+        return {"real_col": self.param.expand(seq_len, 1, 1)}
+
+
 class _TinyMultiTargetModel(torch.nn.Module):
     def __init__(self):
         super().__init__()
@@ -46,7 +58,13 @@ def _loss_shell(target_column_types, *, data_parallelism=None, loss_weights=None
     model.loss_weights = loss_weights
     model.device = "cpu"
     model.hparams = SimpleNamespace(
-        training_spec=SimpleNamespace(data_parallelism=data_parallelism)
+        training_spec=SimpleNamespace(
+            data_parallelism=data_parallelism,
+            distributed=data_parallelism is not None,
+            training_objective="causal",
+            layer_autocast=False,
+            layer_type_dtypes=None,
+        )
     )
     model.criterion = {}
     model.n_classes = {}
@@ -63,6 +81,28 @@ def _loss_shell(target_column_types, *, data_parallelism=None, loss_weights=None
     return model
 
 
+class _IdentityScaler:
+    def scale(self, loss):
+        return loss
+
+    def unscale_(self, optimizer):
+        return None
+
+    def step(self, optimizer):
+        optimizer.step()
+
+    def update(self):
+        return None
+
+
+class _NoopLogger:
+    def info(self, *args, **kwargs):
+        return None
+
+    def warning(self, *args, **kwargs):
+        return None
+
+
 def _real_batch(seq_len, selected_count, target_value):
     targets = {"real_col": torch.full((1, seq_len), float(target_value))}
     mask = torch.zeros(1, seq_len, dtype=torch.bool)
@@ -71,6 +111,18 @@ def _real_batch(seq_len, selected_count, target_value):
     return targets, metadata
 
 
+def _real_epoch_batch(seq_len, selected_count, target_value):
+    targets, metadata = _real_batch(seq_len, selected_count, target_value)
+    return SequifierBatch(
+        inputs={"real_col": torch.ones(1, seq_len, dtype=torch.float32)},
+        targets=targets,
+        metadata={
+            "attention_valid_mask": torch.ones(1, seq_len, dtype=torch.bool),
+            **metadata,
+        },
+    )
+
+
 def _concat_real_batch(seq_lens, selected_counts, target_values):
     target_parts = []
     mask_parts = []
@@ -138,6 +190,37 @@ def _single_process_accumulated_real_update(microbatches, lr):
     return model.param.detach().clone()
 
 
+def _analytical_accumulated_real_update(microbatches, lr):
+    accumulated_grad = 0.0
+    for microbatch in microbatches:
+        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
+        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
+        token_count = selected_counts.sum().item()
+        if token_count == 0:
+            continue
+
+        weighted_target_mean = (
+            selected_counts * target_values
+        ).sum().item() / token_count
+        accumulated_grad += -2.0 * weighted_target_mean
+
+    return torch.tensor(-lr * accumulated_grad)
+
+
+def _analytical_combined_real_update(microbatches, lr):
+    weighted_target_sum = 0.0
+    token_count = 0.0
+    for microbatch in microbatches:
+        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
+        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
+        weighted_target_sum += (selected_counts * target_values).sum().item()
+        token_count += selected_counts.sum().item()
+
+    if token_count == 0:
+        return torch.tensor(0.0)
+    return torch.tensor(lr * 2.0 * weighted_target_sum / token_count)
+
+
 def _single_process_multi_target_grad(
     seq_lens, cat_targets, real_targets, loss_weights
 ):
@@ -190,7 +273,67 @@ def put_result(tensor):
             "empty_grad",
             "accumulation_update",
             "empty_window_state",
+            "train_epoch_empty_window_state",
         }:
+            if case["kind"] == "train_epoch_empty_window_state":
+                model = _TinyTrainEpochRealModel()
+                ddp_model = DDP(model)
+                shell = _loss_shell({"real_col": "real"}, data_parallelism="DDP")
+                shell.input_columns = ["real_col"]
+                shell.rank = rank
+                shell.device = "cpu"
+                shell.accumulation_steps = case["accumulation_steps"]
+                shell.scheduler_step_on = "batch"
+                shell.start_batch = 0
+                shell.log_interval = case["accumulation_steps"]
+                shell.scaler = _IdentityScaler()
+                shell.logger = _NoopLogger()
+                shell.save_latest_interval_minutes = None
+                shell.save_batch_interval_minutes = None
+                shell.save_batch_interval_minutes_val_loss = False
+                shell.last_latest_save_time = 0.0
+                shell.last_batch_save_time = 0.0
+                shell.parameters = model.parameters
+                shell.optimizer = torch.optim.AdamW(
+                    model.parameters(),
+                    lr=case["lr"],
+                    weight_decay=case["weight_decay"],
+                )
+                shell.scheduler = torch.optim.lr_scheduler.StepLR(
+                    shell.optimizer,
+                    step_size=1,
+                    gamma=0.1,
+                )
+
+                train_loader = [
+                    _real_epoch_batch(
+                        microbatch["seq_lens"][rank],
+                        microbatch["selected_counts"][rank],
+                        microbatch["target_values"][rank],
+                    )
+                    for microbatch in case["microbatches"]
+                ]
+                before = model.param.detach().clone()
+
+                TransformerModel._train_epoch(
+                    shell,
+                    DataLoader(train_loader, batch_size=None),
+                    DataLoader([], batch_size=None),
+                    epoch=1,
+                    ddp_model=ddp_model,
+                )
+
+                if rank == 0:
+                    queue.put(
+                        [
+                            float((model.param.detach() - before).abs().item()),
+                            float(shell.scheduler.get_last_lr()[0]),
+                            float(shell.scheduler.last_epoch),
+                            float(len(shell.optimizer.state)),
+                        ]
+                    )
+                return
+
             model = _TinyRealModel()
             ddp_model = DDP(model)
             shell = _loss_shell({"real_col": "real"}, data_parallelism="DDP")
@@ -517,16 +660,44 @@ def test_ddp_globally_empty_accumulation_window_leaves_state_unchanged():
     assert optimizer_state_len == 0
 
 
+def test_ddp_train_epoch_globally_empty_accumulation_window_leaves_state_unchanged():
+    case = {
+        "kind": "train_epoch_empty_window_state",
+        "microbatches": [
+            {
+                "seq_lens": [3, 4],
+                "selected_counts": [0, 0],
+                "target_values": [1.0, 3.0],
+            },
+            {
+                "seq_lens": [2, 5],
+                "selected_counts": [0, 0],
+                "target_values": [2.0, 4.0],
+            },
+        ],
+        "accumulation_steps": 2,
+        "lr": 0.01,
+        "weight_decay": 0.1,
+    }
+
+    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_ddp_case(case)
+
+    assert param_delta == 0
+    assert lr == pytest.approx(case["lr"])
+    assert scheduler_epoch == 0
+    assert optimizer_state_len == 0
+
+
 def test_ddp_gradient_accumulation_preserves_per_microbatch_weighting():
     microbatches = [
         {
-            "seq_lens": [4, 2],
-            "selected_counts": [4, 2],
+            "seq_lens": [1, 1],
+            "selected_counts": [1, 1],
             "target_values": [1.0, 3.0],
         },
         {
-            "seq_lens": [1, 5],
-            "selected_counts": [1, 5],
+            "seq_lens": [5, 5],
+            "selected_counts": [5, 5],
             "target_values": [10.0, 2.0],
         },
     ]
@@ -537,9 +708,11 @@ def test_ddp_gradient_accumulation_preserves_per_microbatch_weighting():
     }
 
     ddp_param = _run_ddp_case(case)
-    reference_param = _single_process_accumulated_real_update(microbatches, case["lr"])
+    reference_param = _analytical_accumulated_real_update(microbatches, case["lr"])
+    combined_batch_param = _analytical_combined_real_update(microbatches, case["lr"])
 
     assert torch.allclose(ddp_param, reference_param, atol=1e-6)
+    assert not torch.allclose(ddp_param, combined_batch_param, atol=1e-3)
 
 
 def test_ddp_global_training_metric_finalization_matches_manual_sum_count():
diff --git a/tests/unit/test_train_fsdp_loss.py b/tests/unit/test_train_fsdp_loss.py
index b4f4a6b9..dbdd09cc 100644
--- a/tests/unit/test_train_fsdp_loss.py
+++ b/tests/unit/test_train_fsdp_loss.py
@@ -11,7 +11,9 @@
     StateDictOptions,
     get_model_state_dict,
 )
+from torch.utils.data import DataLoader
 
+from sequifier.io.batch import SequifierBatch
 from sequifier.train import TransformerModel
 
 _train_module = importlib.import_module("sequifier.train")
@@ -37,6 +39,16 @@ def forward(self, seq_len):
         return {"real_col": self.param[0].expand(seq_len, 1, 1)}
 
 
+class _TinyTrainEpochRealModel(torch.nn.Module):
+    def __init__(self):
+        super().__init__()
+        self.param = torch.nn.Parameter(torch.zeros(2))
+
+    def forward(self, data, metadata=None, return_logits=False):
+        seq_len = data["real_col"].shape[1]
+        return {"real_col": self.param[0].expand(seq_len, 1, 1)}
+
+
 class _TinyMultiTargetModel(torch.nn.Module):
     def __init__(self):
         super().__init__()
@@ -63,7 +75,13 @@ def _loss_shell(target_column_types, *, device, loss_weights=None):
     model.loss_weights = loss_weights
     model.device = device
     model.hparams = SimpleNamespace(
-        training_spec=SimpleNamespace(data_parallelism="FSDP")
+        training_spec=SimpleNamespace(
+            data_parallelism="FSDP",
+            distributed=True,
+            training_objective="causal",
+            layer_autocast=False,
+            layer_type_dtypes=None,
+        )
     )
     model.criterion = {}
     model.n_classes = {}
@@ -80,6 +98,28 @@ def _loss_shell(target_column_types, *, device, loss_weights=None):
     return model
 
 
+class _IdentityScaler:
+    def scale(self, loss):
+        return loss
+
+    def unscale_(self, optimizer):
+        return None
+
+    def step(self, optimizer):
+        optimizer.step()
+
+    def update(self):
+        return None
+
+
+class _NoopLogger:
+    def info(self, *args, **kwargs):
+        return None
+
+    def warning(self, *args, **kwargs):
+        return None
+
+
 def _real_batch(seq_len, selected_count, target_value, device):
     targets = {"real_col": torch.full((1, seq_len), float(target_value), device=device)}
     mask = torch.zeros(1, seq_len, dtype=torch.bool, device=device)
@@ -88,6 +128,20 @@ def _real_batch(seq_len, selected_count, target_value, device):
     return targets, metadata
 
 
+def _real_epoch_batch(seq_len, selected_count, target_value):
+    targets = {"real_col": torch.full((1, seq_len), float(target_value))}
+    mask = torch.zeros(1, seq_len, dtype=torch.bool)
+    mask[:, :selected_count] = True
+    return SequifierBatch(
+        inputs={"real_col": torch.ones(1, seq_len, dtype=torch.float32)},
+        targets=targets,
+        metadata={
+            "attention_valid_mask": torch.ones(1, seq_len, dtype=torch.bool),
+            "target_valid_mask": mask,
+        },
+    )
+
+
 def _concat_real_batch(seq_lens, selected_counts, target_values):
     target_parts = []
     mask_parts = []
@@ -141,6 +195,37 @@ def _single_process_accumulated_real_update(microbatches, lr):
     return model.param.detach().clone()
 
 
+def _analytical_accumulated_real_update(microbatches, lr):
+    accumulated_grad = 0.0
+    for microbatch in microbatches:
+        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
+        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
+        token_count = selected_counts.sum().item()
+        if token_count == 0:
+            continue
+
+        weighted_target_mean = (
+            selected_counts * target_values
+        ).sum().item() / token_count
+        accumulated_grad += -2.0 * weighted_target_mean
+
+    return torch.tensor([-lr * accumulated_grad, 0.0])
+
+
+def _analytical_combined_real_update(microbatches, lr):
+    weighted_target_sum = 0.0
+    token_count = 0.0
+    for microbatch in microbatches:
+        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
+        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
+        weighted_target_sum += (selected_counts * target_values).sum().item()
+        token_count += selected_counts.sum().item()
+
+    if token_count == 0:
+        return torch.tensor([0.0, 0.0])
+    return torch.tensor([lr * 2.0 * weighted_target_sum / token_count, 0.0])
+
+
 def _single_process_multi_target_update(
     seq_lens,
     cat_targets,
@@ -218,7 +303,69 @@ def _fsdp_case_worker(rank, world_size, init_method, case, queue):
             "real_update",
             "accumulation_update",
             "empty_window_state",
+            "train_epoch_empty_window_state",
         }:
+            if case["kind"] == "train_epoch_empty_window_state":
+                model = _TinyTrainEpochRealModel().to(device)
+                fully_shard(model, mesh=mesh, mp_policy=mp_policy)
+                shell = _loss_shell({"real_col": "real"}, device=str(device))
+                shell._data_parallel_group = mesh.get_group()
+                shell.input_columns = ["real_col"]
+                shell.rank = rank
+                shell.accumulation_steps = case["accumulation_steps"]
+                shell.scheduler_step_on = "batch"
+                shell.start_batch = 0
+                shell.log_interval = case["accumulation_steps"]
+                shell.scaler = _IdentityScaler()
+                shell.logger = _NoopLogger()
+                shell.save_latest_interval_minutes = None
+                shell.save_batch_interval_minutes = None
+                shell.save_batch_interval_minutes_val_loss = False
+                shell.last_latest_save_time = 0.0
+                shell.last_batch_save_time = 0.0
+                shell.parameters = model.parameters
+                shell.optimizer = torch.optim.AdamW(
+                    model.parameters(),
+                    lr=case["lr"],
+                    weight_decay=case["weight_decay"],
+                )
+                shell.scheduler = torch.optim.lr_scheduler.StepLR(
+                    shell.optimizer,
+                    step_size=1,
+                    gamma=0.1,
+                )
+                before = _state_vector(model, ["param"], rank)
+                train_loader = [
+                    _real_epoch_batch(
+                        microbatch["seq_lens"][rank],
+                        microbatch["selected_counts"][rank],
+                        microbatch["target_values"][rank],
+                    )
+                    for microbatch in case["microbatches"]
+                ]
+
+                TransformerModel._train_epoch(
+                    shell,
+                    DataLoader(train_loader, batch_size=None),
+                    DataLoader([], batch_size=None),
+                    epoch=1,
+                    ddp_model=model,
+                )
+
+                after = _state_vector(model, ["param"], rank)
+                if rank == 0:
+                    assert before is not None
+                    assert after is not None
+                    queue.put(
+                        [
+                            float((after - before).abs().max().item()),
+                            float(shell.scheduler.get_last_lr()[0]),
+                            float(shell.scheduler.last_epoch),
+                            float(len(shell.optimizer.state)),
+                        ]
+                    )
+                return
+
             model = _TinyRealModel().to(device)
             fully_shard(model, mesh=mesh, mp_policy=mp_policy)
             shell = _loss_shell({"real_col": "real"}, device=str(device))
@@ -503,13 +650,13 @@ def test_fsdp_world_size_two_optimizer_update_matches_single_process_reference()
 def test_fsdp_gradient_accumulation_preserves_per_microbatch_weighting():
     microbatches = [
         {
-            "seq_lens": [4, 2],
-            "selected_counts": [4, 2],
+            "seq_lens": [1, 1],
+            "selected_counts": [1, 1],
             "target_values": [1.0, 3.0],
         },
         {
-            "seq_lens": [1, 5],
-            "selected_counts": [1, 5],
+            "seq_lens": [5, 5],
+            "selected_counts": [5, 5],
             "target_values": [10.0, 2.0],
         },
     ]
@@ -520,9 +667,11 @@ def test_fsdp_gradient_accumulation_preserves_per_microbatch_weighting():
     }
 
     fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_accumulated_real_update(microbatches, case["lr"])
+    reference_param = _analytical_accumulated_real_update(microbatches, case["lr"])
+    combined_batch_param = _analytical_combined_real_update(microbatches, case["lr"])
 
     assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+    assert not torch.allclose(fsdp_param, combined_batch_param, atol=1e-3)
 
 
 def test_fsdp_globally_empty_accumulation_window_leaves_state_unchanged():
@@ -551,3 +700,31 @@ def test_fsdp_globally_empty_accumulation_window_leaves_state_unchanged():
     assert lr == pytest.approx(case["lr"])
     assert scheduler_epoch == 0
     assert optimizer_state_len == 0
+
+
+def test_fsdp_train_epoch_globally_empty_accumulation_window_leaves_state_unchanged():
+    case = {
+        "kind": "train_epoch_empty_window_state",
+        "microbatches": [
+            {
+                "seq_lens": [3, 4],
+                "selected_counts": [0, 0],
+                "target_values": [1.0, 3.0],
+            },
+            {
+                "seq_lens": [2, 5],
+                "selected_counts": [0, 0],
+                "target_values": [2.0, 4.0],
+            },
+        ],
+        "accumulation_steps": 2,
+        "lr": 0.01,
+        "weight_decay": 0.1,
+    }
+
+    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_fsdp_case(case)
+
+    assert param_delta == 0
+    assert lr == pytest.approx(case["lr"])
+    assert scheduler_epoch == 0
+    assert optimizer_state_len == 0

From a303826c2e0d64c6374a3f28174d877a033cefe4 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Tue, 16 Jun 2026 11:42:16 +0200
Subject: [PATCH 64/81] Improve tests

---
 src/sequifier/config/train_config.py          |   5 +-
 src/sequifier/train.py                        |  16 +-
 tests/integration/conftest.py                 |   9 +-
 .../integration/test_hyperparameter_search.py |   3 -
 tests/integration/test_identities.py          |  56 ++-
 tests/integration/test_make.py                |   7 +-
 tests/integration/test_onnx_export.py         |  58 ++-
 tests/integration/test_training.py            | 282 ++++++------
 ...orical-1-best-embedding-3-0-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-1-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-2-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-3-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-4-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-5-embeddings.csv |   4 +-
 ...orical-1-best-embedding-3-6-embeddings.csv |   2 +-
 ...orical-1-best-embedding-3-7-embeddings.csv |   2 +-
 ...inf-size-best-embedding-3-0-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-1-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-2-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-3-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-4-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-5-embeddings.csv |  12 +-
 ...inf-size-best-embedding-3-6-embeddings.csv |   6 +-
 ...inf-size-best-embedding-3-7-embeddings.csv |   6 +-
 ...cal-bert-best-embedding-1-0-embeddings.csv |  10 +-
 ...cal-bert-best-embedding-1-1-embeddings.csv |  10 +-
 ...cal-bert-best-embedding-1-2-embeddings.csv |   8 +-
 ...cal-bert-best-embedding-1-3-embeddings.csv |  16 +-
 ...cal-bert-best-embedding-1-4-embeddings.csv |   8 +-
 ...cal-bert-best-embedding-1-5-embeddings.csv |  18 +-
 ...cal-bert-best-embedding-1-6-embeddings.csv |  10 +-
 ...cal-bert-best-embedding-1-7-embeddings.csv |   8 +-
 ...-categorical-bert-best-1-0-predictions.csv |   6 +-
 ...-categorical-bert-best-1-1-predictions.csv |   4 +-
 ...-categorical-bert-best-1-2-predictions.csv |   4 +-
 ...-categorical-bert-best-1-3-predictions.csv |   2 +-
 ...-categorical-bert-best-1-4-predictions.csv |   2 +-
 ...-categorical-bert-best-1-5-predictions.csv |  10 +-
 ...-categorical-bert-best-1-6-predictions.csv |   4 +-
 ...-categorical-bert-best-1-7-predictions.csv |   2 +-
 ...rical-distributed-best-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-best-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-best-3-7-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-0-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-1-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-2-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-3-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-4-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-5-predictions.csv |   4 +-
 ...cal-multitarget-5-last-3-6-predictions.csv |   2 +-
 ...cal-multitarget-5-last-3-7-predictions.csv |   2 +-
 ...al-1-best-3-autoregression-predictions.csv | 400 +++++++++---------
 ...uifier-model-real-1-best-3-predictions.csv |  48 +--
 ...uifier-model-real-3-best-3-predictions.csv |  48 +--
 ...uifier-model-real-5-best-3-predictions.csv |  48 +--
 ...ifier-model-real-50-best-3-predictions.csv |  48 +--
 ...ustom-eval-run-0-best-3-0-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-1-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-2-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-3-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-4-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-5-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-6-predictions.csv} |   0
 ...ustom-eval-run-0-best-3-7-predictions.csv} |   0
 ...ustom-eval-run-1-best-3-0-predictions.csv} |   6 +-
 ...ustom-eval-run-1-best-3-1-predictions.csv} |   6 +-
 ...ustom-eval-run-1-best-3-2-predictions.csv} |  32 +-
 ...ustom-eval-run-1-best-3-3-predictions.csv} |  60 +--
 ...ustom-eval-run-1-best-3-4-predictions.csv} |  14 +-
 ...ustom-eval-run-1-best-3-5-predictions.csv} |  34 +-
 ...ustom-eval-run-1-best-3-6-predictions.csv} |   8 +-
 ...custom-eval-run-1-best-3-7-predictions.csv |  31 ++
 ...custom-eval-run-2-best-3-0-predictions.csv |  31 ++
 ...custom-eval-run-2-best-3-1-predictions.csv |  31 ++
 ...custom-eval-run-2-best-3-2-predictions.csv |  31 ++
 ...custom-eval-run-2-best-3-3-predictions.csv |  61 +++
 ...custom-eval-run-2-best-3-4-predictions.csv |  31 ++
 ...custom-eval-run-2-best-3-5-predictions.csv |  61 +++
 ...custom-eval-run-2-best-3-6-predictions.csv |  31 ++
 ...custom-eval-run-2-best-3-7-predictions.csv |  31 ++
 ...custom-eval-run-3-best-3-0-predictions.csv |  31 ++
 ...custom-eval-run-3-best-3-1-predictions.csv |  31 ++
 ...custom-eval-run-3-best-3-2-predictions.csv |  31 ++
 ...custom-eval-run-3-best-3-3-predictions.csv |  61 +++
 ...custom-eval-run-3-best-3-4-predictions.csv |  31 ++
 ...custom-eval-run-3-best-3-5-predictions.csv |  61 +++
 ...custom-eval-run-3-best-3-6-predictions.csv |  31 ++
 ...custom-eval-run-3-best-3-7-predictions.csv |  31 ++
 ...custom-eval-run-4-best-3-0-predictions.csv |  31 --
 ...custom-eval-run-4-best-3-1-predictions.csv |  31 --
 ...custom-eval-run-4-best-3-2-predictions.csv |  31 --
 ...custom-eval-run-4-best-3-3-predictions.csv |  61 ---
 ...custom-eval-run-4-best-3-4-predictions.csv |  31 --
 ...custom-eval-run-4-best-3-5-predictions.csv |  61 ---
 ...custom-eval-run-4-best-3-6-predictions.csv |  31 --
 ...custom-eval-run-4-best-3-7-predictions.csv |  31 --
 ...custom-eval-run-5-best-3-0-predictions.csv |  31 --
 ...custom-eval-run-5-best-3-1-predictions.csv |  31 --
 ...custom-eval-run-5-best-3-2-predictions.csv |  31 --
 ...custom-eval-run-5-best-3-3-predictions.csv |  61 ---
 ...custom-eval-run-5-best-3-4-predictions.csv |  31 --
 ...custom-eval-run-5-best-3-5-predictions.csv |  61 ---
 ...custom-eval-run-5-best-3-6-predictions.csv |  31 --
 ...custom-eval-run-5-best-3-7-predictions.csv |  31 --
 ...custom-eval-run-6-best-3-7-predictions.csv |  31 --
 ...l-categorical-1-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-3-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-5-probabilities.csv |   4 +-
 ...l-categorical-1-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-1-best-3-7-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-3-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-5-probabilities.csv |   4 +-
 ...l-categorical-3-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-3-best-3-7-probabilities.csv |   2 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-0-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-1-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-2-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-3-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-4-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-5-probabilities.csv |  12 +-
 ...ical-3-inf-size-best-3-6-probabilities.csv |   6 +-
 ...ical-3-inf-size-best-3-7-probabilities.csv |   6 +-
 ...l-categorical-5-best-3-0-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-1-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-2-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-3-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-4-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-5-probabilities.csv |   4 +-
 ...l-categorical-5-best-3-6-probabilities.csv |   2 +-
 ...l-categorical-5-best-3-7-probabilities.csv |   2 +-
 ...-categorical-50-best-3-0-probabilities.csv |   2 +-
 ...-categorical-50-best-3-1-probabilities.csv |   2 +-
 ...-categorical-50-best-3-2-probabilities.csv |   2 +-
 ...-categorical-50-best-3-3-probabilities.csv |   4 +-
 ...-categorical-50-best-3-4-probabilities.csv |   2 +-
 ...-categorical-50-best-3-5-probabilities.csv |   4 +-
 ...-categorical-50-best-3-6-probabilities.csv |   2 +-
 ...-categorical-50-best-3-7-probabilities.csv |   2 +-
 ...ategorical-bert-best-1-0-probabilities.csv |  10 +-
 ...ategorical-bert-best-1-1-probabilities.csv |  10 +-
 ...ategorical-bert-best-1-2-probabilities.csv |   8 +-
 ...ategorical-bert-best-1-3-probabilities.csv |  16 +-
 ...ategorical-bert-best-1-4-probabilities.csv |   8 +-
 ...ategorical-bert-best-1-5-probabilities.csv |  18 +-
 ...ategorical-bert-best-1-6-probabilities.csv |  10 +-
 ...ategorical-bert-best-1-7-probabilities.csv |   8 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-0-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-1-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-2-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-3-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-4-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-5-probabilities.csv |   4 +-
 ...l-multitarget-5-best-3-6-probabilities.csv |   2 +-
 ...l-multitarget-5-best-3-7-probabilities.csv |   2 +-
 ...uifier_dataset_from_folder_parquet_lazy.py |   2 +-
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  78 +++-
 tests/unit/test_infer.py                      |  37 +-
 tests/unit/test_train.py                      | 113 ++++-
 tests/unit/test_train_distributed_loss.py     |  42 +-
 tests/unit/test_train_fsdp_loss.py            |  38 +-
 197 files changed, 1845 insertions(+), 1579 deletions(-)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv => sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv} (100%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv} (88%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv} (88%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv} (51%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv} (52%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv} (76%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv} (71%)
 rename tests/resources/target_outputs/predictions/{sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv => sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv} (84%)
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv
 create mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv

diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index f855a972..21d2457d 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -152,7 +152,10 @@ def load_train_config(
 
         config_values["id_maps"] = metadata_config["id_maps"]
         config_values["special_token_ids"] = validate_special_token_ids(
-            metadata_config["special_token_ids"],
+            metadata_config.get(
+                "special_token_ids",
+                SPECIAL_TOKEN_IDS.ids_by_label,
+            ),
             source=f"metadata config '{metadata_config_path}'",
         )
 
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 71a0363d..b0b28bed 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1439,7 +1439,19 @@ def _train_epoch(
                         global_token_count,
                     ) = self._calculate_training_loss(output, targets, metadata)
 
-                self.scaler.scale(loss).backward()
+                if self.accumulation_steps is None:
+                    accumulation_divisor = 1
+                else:
+                    window_start = (
+                        batch_count // self.accumulation_steps
+                    ) * self.accumulation_steps
+                    accumulation_divisor = min(
+                        self.accumulation_steps,
+                        num_batches - window_start,
+                    )
+
+                backward_loss = loss / accumulation_divisor
+                self.scaler.scale(backward_loss).backward()
                 self._accumulate_loss_components(
                     train_loss_sums,
                     train_token_count,
@@ -1507,7 +1519,7 @@ def _train_epoch(
                         start_time = time.time()
                     self._check_and_terminate()
 
-                del data, targets, output, loss, backward_components
+                del data, targets, output, loss, backward_loss, backward_components
 
                 if self.scheduler_step_on == "batch" and optimizer_step_performed:
                     if (
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
index a8575c49..f96c0f61 100644
--- a/tests/integration/conftest.py
+++ b/tests/integration/conftest.py
@@ -2,6 +2,7 @@
 import os
 import shutil
 import subprocess
+import sys
 import time
 
 import polars as pl
@@ -326,8 +327,6 @@ def format_configs_locally(
     inference_config_path_lazy,
     hp_search_configs,
 ):
-    from sys import platform
-
     config_paths = [
         preprocessing_config_path_cat,
         preprocessing_config_path_cat_multitarget,
@@ -371,7 +370,7 @@ def format_configs_locally(
     cuda_available = torch.cuda.is_available()
 
     for config_path in config_paths:
-        if str(platform).startswith("win") or cuda_available:
+        if sys.platform.startswith("win") or cuda_available:
             with open(config_path, "r") as f:
                 config = yaml.safe_load(f)
 
@@ -380,7 +379,7 @@ def format_configs_locally(
 
             assert config is not None, config_path
 
-            if platform == "windows":
+            if sys.platform.startswith("win"):
                 config = {
                     attr: reformat_parameter(attr, param, "linux->local")
                     for attr, param in config.items()
@@ -397,7 +396,7 @@ def format_configs_locally(
 
     yield
 
-    if str(platform).startswith("win") or cuda_available:
+    if sys.platform.startswith("win") or cuda_available:
         for config_path in config_paths:
             with open(config_path, "w") as f:
                 yaml.dump(
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
index ebe52dd2..17c83460 100644
--- a/tests/integration/test_hyperparameter_search.py
+++ b/tests/integration/test_hyperparameter_search.py
@@ -37,9 +37,6 @@ def test_hp_search_bert_outputs(run_hp_search, project_root):
     with open(generated_configs[0], "r") as f:
         generated_config = yaml.safe_load(f)
 
-    assert generated_config["metadata_config_path"].endswith(
-        "test-data-categorical-1-lookahead-0.json"
-    )
     assert generated_config["metadata_config_path"].endswith(
         "test-data-categorical-1-lookahead-0.json"
     )
diff --git a/tests/integration/test_identities.py b/tests/integration/test_identities.py
index 0495a56e..f2ecee46 100644
--- a/tests/integration/test_identities.py
+++ b/tests/integration/test_identities.py
@@ -2,22 +2,48 @@
 import pytest
 
 
+def _frame_mismatches(actual, expected, *, rtol=1e-5, atol=1e-6):
+    mismatches = []
+    for column in actual.columns:
+        actual_values = actual[column].to_numpy()
+        expected_values = expected[column].to_numpy()
+        if np.issubdtype(actual_values.dtype, np.floating):
+            try:
+                np.testing.assert_allclose(
+                    actual_values,
+                    expected_values,
+                    rtol=rtol,
+                    atol=atol,
+                )
+            except AssertionError as exc:
+                mismatches.append((column, str(exc)))
+        elif not np.array_equal(actual_values, expected_values):
+            mismatches.append(
+                (
+                    column,
+                    f"{actual_values = } != {expected_values = }",
+                )
+            )
+    return mismatches
+
+
 @pytest.mark.optional
 def test_identities(targets, predictions, probabilities, embeddings):
     failed_models = []
     for model_name, preds in predictions.items():
-        equal = preds.to_numpy() == targets["preds"][model_name].to_numpy()
-        mean_equal = np.mean(equal.astype(int))
+        mismatches = _frame_mismatches(preds, targets["preds"][model_name])
         if model_name != "model-real-1-best-3-autoregression":
-            if not mean_equal == 1.0:
+            if mismatches:
                 failed_models.append(
                     (
                         model_name,
-                        equal,
-                        f"{model_name} preds are not identical to target: {preds.to_numpy() = } != {targets['preds'][model_name].to_numpy() = }: {equal = }, {mean_equal = }",
+                        mismatches,
+                        f"{model_name} preds differ from target: {mismatches = }",
                     )
                 )
         else:
+            equal = preds.to_numpy() == targets["preds"][model_name].to_numpy()
+            mean_equal = np.mean(equal.astype(int))
             if mean_equal == 1.0:
                 failed_models.append(
                     (
@@ -28,22 +54,24 @@ def test_identities(targets, predictions, probabilities, embeddings):
                 )
 
     for model_name, probs in probabilities.items():
-        equal = probs.to_numpy() == targets["probs"][model_name].to_numpy()
-        all_equal = np.all(equal)
-        if not all_equal:
+        mismatches = _frame_mismatches(probs, targets["probs"][model_name])
+        if mismatches:
             failed_models.append(
-                (model_name, equal, f"{model_name} probs are not identical to target")
+                (
+                    model_name,
+                    mismatches,
+                    f"{model_name} probs differ from target: {mismatches = }",
+                )
             )
 
     for model_name, embeds in embeddings.items():
-        equal = embeds.to_numpy() == targets["embeds"][model_name].to_numpy()
-        all_equal = np.all(equal)
-        if not all_equal:
+        mismatches = _frame_mismatches(embeds, targets["embeds"][model_name])
+        if mismatches:
             failed_models.append(
                 (
                     model_name,
-                    equal,
-                    f"{model_name} embeddings are not identical to target",
+                    mismatches,
+                    f"{model_name} embeddings differ from target: {mismatches = }",
                 )
             )
 
diff --git a/tests/integration/test_make.py b/tests/integration/test_make.py
index 5b09d088..88d1f366 100644
--- a/tests/integration/test_make.py
+++ b/tests/integration/test_make.py
@@ -1,5 +1,6 @@
 import os
 import shutil
+import sys
 import time
 
 import pytest
@@ -125,9 +126,7 @@ def adapt_configs(config_strings):
     with open(infer_config_path, "w") as f:
         f.write(infer_config_string)
 
-    from sys import platform
-
-    if platform == "windows":
+    if sys.platform.startswith("win"):
         for config_path in [
             preprocess_config_path,
             train_config_path,
@@ -165,7 +164,7 @@ def test_make(adapt_configs):
         else:
             assert False, f"Training for 'sequifier train --config-path {test_project_name}/configs/train.yaml' was unsuccessful"
     else:
-        assert False, f"Preprocessing for 'sequifier preprocess --config-path {test_project_name}/configs/train.yaml' was unsuccessful"
+        assert False, f"Preprocessing for 'sequifier preprocess --config-path {test_project_name}/configs/preprocess.yaml' was unsuccessful"
 
     # clean up, only if tests didn't fail
     shutil.rmtree(test_project_name)
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
index 4a4be821..c7f2bd72 100644
--- a/tests/integration/test_onnx_export.py
+++ b/tests/integration/test_onnx_export.py
@@ -2,6 +2,7 @@
 
 import numpy as np
 import onnxruntime
+import torch
 
 from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
 from sequifier.helpers import ModelWindowView, StoredWindowLayout
@@ -84,6 +85,7 @@ def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
         ),
     )
     model = TransformerModel(config)
+    model.eval()
 
     model._export_model(model, "best", 1)
     export_path = os.path.join(
@@ -195,15 +197,51 @@ def test_onnx_export_preserves_feature_name_order(tmp_path):
 
     assert input_names == ["cat2_in", "cat10_in", "attention_valid_mask"]
 
-    outputs = session.run(
-        None,
-        {
-            "cat2_in": np.array([[3, 4, 5, 6], [6, 5, 4, 3]], dtype=np.int64),
-            "cat10_in": np.array([[3, 4, 5, 3], [5, 4, 3, 5]], dtype=np.int64),
-            "attention_valid_mask": np.ones(
-                (inference_batch_size, context_length), dtype=np.bool_
-            ),
-        },
-    )
+    ort_inputs = {
+        "cat2_in": np.array([[3, 4, 5, 6], [6, 5, 4, 3]], dtype=np.int64),
+        "cat10_in": np.array([[3, 4, 5, 3], [5, 4, 3, 5]], dtype=np.int64),
+        "attention_valid_mask": np.ones(
+            (inference_batch_size, context_length), dtype=np.bool_
+        ),
+    }
+    outputs = session.run(None, ort_inputs)
 
     assert outputs[0].shape == (1, inference_batch_size, config.n_classes["cat2"])
+
+    def assert_onnx_matches_torch(onnx_inputs):
+        torch_inputs_for_case = {
+            "cat2": torch.tensor(onnx_inputs["cat2_in"], dtype=torch.long),
+            "cat10": torch.tensor(onnx_inputs["cat10_in"], dtype=torch.long),
+        }
+        torch_metadata_for_case = {
+            "attention_valid_mask": torch.tensor(
+                onnx_inputs["attention_valid_mask"], dtype=torch.bool
+            )
+        }
+        with torch.no_grad():
+            torch_outputs = model(torch_inputs_for_case, torch_metadata_for_case)
+
+        onnx_outputs = session.run(None, onnx_inputs)
+        np.testing.assert_allclose(
+            onnx_outputs[0],
+            torch_outputs["cat2"].detach().numpy(),
+            rtol=1e-4,
+            atol=1e-5,
+        )
+        return onnx_outputs[0]
+
+    base_outputs = assert_onnx_matches_torch(ort_inputs)
+
+    cat2_changed_inputs = {
+        **ort_inputs,
+        "cat2_in": np.array([[6, 6, 5, 4], [3, 3, 4, 5]], dtype=np.int64),
+    }
+    cat2_changed_outputs = assert_onnx_matches_torch(cat2_changed_inputs)
+    assert not np.allclose(base_outputs, cat2_changed_outputs, rtol=1e-5, atol=1e-6)
+
+    cat10_changed_inputs = {
+        **ort_inputs,
+        "cat10_in": np.array([[5, 5, 4, 3], [3, 3, 5, 4]], dtype=np.int64),
+    }
+    cat10_changed_outputs = assert_onnx_matches_torch(cat10_changed_inputs)
+    assert not np.allclose(base_outputs, cat10_changed_outputs, rtol=1e-5, atol=1e-6)
diff --git a/tests/integration/test_training.py b/tests/integration/test_training.py
index d90c32ef..ca8a4ab6 100644
--- a/tests/integration/test_training.py
+++ b/tests/integration/test_training.py
@@ -1,162 +1,148 @@
 import os
 
-import numpy as np
-
 
 def test_checkpoint_files_exists(
-    run_training, run_training_from_checkpoint, project_root
+    run_training, run_training_from_checkpoint, run_hp_search, project_root
 ):
-    found_items = np.array(
-        sorted(list(os.listdir(os.path.join(project_root, "checkpoints"))))
-    )
-    expected_items = np.array(
-        sorted(
-            [
-                f"model-{model_type}-{j}-epoch-{i}.pt"
-                for model_type in ["categorical", "real"]
-                for j in [1, 3, 5, 50]
-                for i in range(1, 4)
-            ]
-            + [
-                f"model-real-{j}-epoch-{i}-batch-1.pt"
-                for j in [1, 3, 5, 50]
-                for i in range(1, 4)
-            ]
-            + [f"model-real-{j}-latest.pt" for j in [1, 3, 5, 50]]
-            + [f"model-real-1-from-epoch-checkpoint-epoch-{i}.pt" for i in range(1, 4)]
-            + [
-                f"model-categorical-3-from-mid-epoch-checkpoint-epoch-{i}.pt"
-                for i in range(1, 4)
-            ]
-            + [
-                f"model-categorical-{j}-inf-size-epoch-{i}.pt"
-                for j in [1, 3]
-                for i in range(1, 4)
-            ]
-            + ["model-categorical-bert-epoch-1.pt"]
-            + ["model-real-bert-epoch-1.pt"]
-            + [f"model-categorical-multitarget-5-epoch-{i}.pt" for i in range(1, 4)]
-            + [
-                f"model-categorical-multitarget-5-eager-epoch-{i}.pt"
-                for i in range(1, 4)
-            ]
-            + [f"model-categorical-distributed-epoch-{i}.pt" for i in range(1, 4)]
-            + [
-                f"model-categorical-distributed-lazy-parquet-epoch-{i}.pt"
-                for i in range(1, 4)
-            ]
-            + [f"model-categorical-lazy-epoch-{i}.pt" for i in range(1, 4)]
-        )
+    found_items = sorted(os.listdir(os.path.join(project_root, "checkpoints")))
+    expected_items = sorted(
+        [
+            f"model-{model_type}-{j}-epoch-{i}.pt"
+            for model_type in ["categorical", "real"]
+            for j in [1, 3, 5, 50]
+            for i in range(1, 4)
+        ]
+        + [
+            f"model-real-{j}-epoch-{i}-batch-1.pt"
+            for j in [1, 3, 5, 50]
+            for i in range(1, 4)
+        ]
+        + [f"model-real-{j}-latest.pt" for j in [1, 3, 5, 50]]
+        + [f"model-real-1-from-epoch-checkpoint-epoch-{i}.pt" for i in range(1, 4)]
+        + [
+            f"model-categorical-3-from-mid-epoch-checkpoint-epoch-{i}.pt"
+            for i in range(1, 4)
+        ]
+        + [
+            f"model-categorical-{j}-inf-size-epoch-{i}.pt"
+            for j in [1, 3]
+            for i in range(1, 4)
+        ]
+        + ["model-categorical-bert-epoch-1.pt"]
+        + ["model-real-bert-epoch-1.pt"]
+        + [f"model-categorical-multitarget-5-epoch-{i}.pt" for i in range(1, 4)]
+        + [f"model-categorical-multitarget-5-eager-epoch-{i}.pt" for i in range(1, 4)]
+        + [f"model-categorical-distributed-epoch-{i}.pt" for i in range(1, 4)]
+        + [
+            f"model-categorical-distributed-lazy-parquet-epoch-{i}.pt"
+            for i in range(1, 4)
+        ]
+        + [f"model-categorical-lazy-epoch-{i}.pt" for i in range(1, 4)]
     )
 
-    print(f"{expected_items = }")
-    print(f"{found_items = }")
+    assert set(found_items) == set(expected_items), {
+        "missing": sorted(set(expected_items) - set(found_items)),
+        "unexpected": sorted(set(found_items) - set(expected_items)),
+    }
 
-    assert np.all(
-        found_items == expected_items
-    ), f"{found_items = } != {expected_items = }"
 
-
-def test_model_files_exists(run_training, run_training_from_checkpoint, project_root):
+def test_model_files_exists(
+    run_training, run_hp_search, run_training_from_checkpoint, project_root
+):
     model_type_formats = {"categorical": ["onnx", "pt"], "real": ["onnx", "pt"]}
-    found_items = np.array(
-        sorted(
-            [
-                f
-                for f in os.listdir(os.path.join(project_root, "models"))
-                if not f.endswith(".onnx.data")
-            ]
-        )
+    found_items = sorted(
+        [
+            f
+            for f in os.listdir(os.path.join(project_root, "models"))
+            if not f.endswith(".onnx.data")
+        ]
     )
 
-    expected_items = np.array(
-        sorted(
-            [
-                f"sequifier-model-{model_type2}-{j}-{kind}{model_type}-3.{model_type_format}"
-                for model_type2 in ["categorical", "real"]
-                for model_type in ["", "-embedding"]
-                for model_type_format in model_type_formats[model_type2]
-                for j in [1, 3, 5, 50]
-                for kind in ["best", "last"]
-            ]
-            + [
-                f"sequifier-model-categorical-3-from-mid-epoch-checkpoint-{kind}{model_type}-3.{model_type_format}"
-                for model_type in ["", "-embedding"]
-                for kind in ["best", "last"]
-                for model_type_format in ["pt", "onnx"]
-            ]
-            + [
-                f"sequifier-model-real-1-from-epoch-checkpoint-{kind}-embedding-3.{model_type_format}"
-                for kind in ["best", "last"]
-                for model_type_format in ["pt", "onnx"]
-            ]
-            + [
-                "sequifier-model-categorical-multitarget-5-best-3.onnx",
-                "sequifier-model-categorical-multitarget-5-last-3.onnx",
-                "sequifier-model-categorical-multitarget-5-best-embedding-3.onnx",
-                "sequifier-model-categorical-multitarget-5-last-embedding-3.onnx",
-                "sequifier-model-categorical-multitarget-5-eager-best-3.onnx",
-                "sequifier-model-categorical-multitarget-5-eager-last-3.onnx",
-                "sequifier-model-categorical-multitarget-5-eager-best-embedding-3.onnx",
-                "sequifier-model-categorical-multitarget-5-eager-last-embedding-3.onnx",
-                "sequifier-model-real-1-best-3-autoregression.pt",
-                "sequifier-model-categorical-1-best-3-autoregression.onnx",
-                "sequifier-model-categorical-1-inf-size-best-3.onnx",
-                "sequifier-model-categorical-1-inf-size-last-3.onnx",
-                "sequifier-model-categorical-1-inf-size-best-embedding-3.onnx",
-                "sequifier-model-categorical-1-inf-size-last-embedding-3.onnx",
-                "sequifier-model-categorical-3-inf-size-best-3.pt",
-                "sequifier-model-categorical-3-inf-size-last-3.pt",
-                "sequifier-model-categorical-3-inf-size-best-embedding-3.pt",
-                "sequifier-model-categorical-3-inf-size-last-embedding-3.pt",
-                "sequifier-model-categorical-bert-best-1.pt",
-                "sequifier-model-categorical-bert-last-1.pt",
-                "sequifier-model-categorical-bert-best-embedding-1.pt",
-                "sequifier-model-categorical-bert-last-embedding-1.pt",
-                "sequifier-model-real-bert-best-1.pt",
-                "sequifier-model-real-bert-last-1.pt",
-                "sequifier-model-categorical-distributed-best-3.pt",
-                "sequifier-model-categorical-distributed-best-3.onnx",
-                "sequifier-model-categorical-distributed-last-3.pt",
-                "sequifier-model-categorical-distributed-last-3.onnx",
-                "sequifier-model-categorical-distributed-lazy-parquet-best-3.pt",
-                "sequifier-model-categorical-distributed-lazy-parquet-best-3.onnx",
-                "sequifier-model-categorical-distributed-lazy-parquet-last-3.pt",
-                "sequifier-model-categorical-distributed-lazy-parquet-last-3.onnx",
-                "sequifier-model-categorical-lazy-best-3.pt",
-                "sequifier-model-categorical-lazy-best-3.onnx",
-                "sequifier-model-categorical-lazy-last-3.pt",
-                "sequifier-model-categorical-lazy-last-3.onnx",
-            ]
-            + [
-                f"sequifier-test-hp-search-sample-run-{i}-{suffix}-1.pt"
-                for i in range(4)
-                for suffix in ["best", "last"]
-            ]
-            + [
-                f"sequifier-test-hp-search-bert-run-0-{suffix}-1.pt"
-                for suffix in ["best", "last"]
-            ]
-            + [
-                f"sequifier-test-hp-search-grid-run-{i}-{suffix}-2.pt"
-                for i in range(4)
-                for suffix in ["best", "last"]
-            ]
-            + [
-                f"sequifier-test-hp-search-bayesian-run-{i}-{suffix}-1.pt"
-                for i in range(4)
-                for suffix in ["best", "last"]
-            ]
-            + [
-                f"sequifier-test-hp-search-custom-eval-run-{i}-{suffix}-3.onnx"
-                for i in range(4)
-                for suffix in ["best", "last"]
-            ]
-        )
+    expected_items = sorted(
+        [
+            f"sequifier-model-{model_type2}-{j}-{kind}{model_type}-3.{model_type_format}"
+            for model_type2 in ["categorical", "real"]
+            for model_type in ["", "-embedding"]
+            for model_type_format in model_type_formats[model_type2]
+            for j in [1, 3, 5, 50]
+            for kind in ["best", "last"]
+        ]
+        + [
+            f"sequifier-model-categorical-3-from-mid-epoch-checkpoint-{kind}{model_type}-3.{model_type_format}"
+            for model_type in ["", "-embedding"]
+            for kind in ["best", "last"]
+            for model_type_format in ["pt", "onnx"]
+        ]
+        + [
+            f"sequifier-model-real-1-from-epoch-checkpoint-{kind}-embedding-3.{model_type_format}"
+            for kind in ["best", "last"]
+            for model_type_format in ["pt", "onnx"]
+        ]
+        + [
+            "sequifier-model-categorical-multitarget-5-best-3.onnx",
+            "sequifier-model-categorical-multitarget-5-last-3.onnx",
+            "sequifier-model-categorical-multitarget-5-best-embedding-3.onnx",
+            "sequifier-model-categorical-multitarget-5-last-embedding-3.onnx",
+            "sequifier-model-categorical-multitarget-5-eager-best-3.onnx",
+            "sequifier-model-categorical-multitarget-5-eager-last-3.onnx",
+            "sequifier-model-categorical-multitarget-5-eager-best-embedding-3.onnx",
+            "sequifier-model-categorical-multitarget-5-eager-last-embedding-3.onnx",
+            "sequifier-model-real-1-best-3-autoregression.pt",
+            "sequifier-model-categorical-1-best-3-autoregression.onnx",
+            "sequifier-model-categorical-1-inf-size-best-3.onnx",
+            "sequifier-model-categorical-1-inf-size-last-3.onnx",
+            "sequifier-model-categorical-1-inf-size-best-embedding-3.onnx",
+            "sequifier-model-categorical-1-inf-size-last-embedding-3.onnx",
+            "sequifier-model-categorical-3-inf-size-best-3.pt",
+            "sequifier-model-categorical-3-inf-size-last-3.pt",
+            "sequifier-model-categorical-3-inf-size-best-embedding-3.pt",
+            "sequifier-model-categorical-3-inf-size-last-embedding-3.pt",
+            "sequifier-model-categorical-bert-best-1.pt",
+            "sequifier-model-categorical-bert-last-1.pt",
+            "sequifier-model-categorical-bert-best-embedding-1.pt",
+            "sequifier-model-categorical-bert-last-embedding-1.pt",
+            "sequifier-model-real-bert-best-1.pt",
+            "sequifier-model-real-bert-last-1.pt",
+            "sequifier-model-categorical-distributed-best-3.pt",
+            "sequifier-model-categorical-distributed-best-3.onnx",
+            "sequifier-model-categorical-distributed-last-3.pt",
+            "sequifier-model-categorical-distributed-last-3.onnx",
+            "sequifier-model-categorical-distributed-lazy-parquet-best-3.pt",
+            "sequifier-model-categorical-distributed-lazy-parquet-best-3.onnx",
+            "sequifier-model-categorical-distributed-lazy-parquet-last-3.pt",
+            "sequifier-model-categorical-distributed-lazy-parquet-last-3.onnx",
+            "sequifier-model-categorical-lazy-best-3.pt",
+            "sequifier-model-categorical-lazy-best-3.onnx",
+            "sequifier-model-categorical-lazy-last-3.pt",
+            "sequifier-model-categorical-lazy-last-3.onnx",
+        ]
+        + [
+            f"sequifier-test-hp-search-sample-run-{i}-{suffix}-1.pt"
+            for i in range(4)
+            for suffix in ["best", "last"]
+        ]
+        + [
+            f"sequifier-test-hp-search-bert-run-0-{suffix}-1.pt"
+            for suffix in ["best", "last"]
+        ]
+        + [
+            f"sequifier-test-hp-search-grid-run-{i}-{suffix}-2.pt"
+            for i in range(4)
+            for suffix in ["best", "last"]
+        ]
+        + [
+            f"sequifier-test-hp-search-bayesian-run-{i}-{suffix}-1.pt"
+            for i in range(4)
+            for suffix in ["best", "last"]
+        ]
+        + [
+            f"sequifier-test-hp-search-custom-eval-run-{i}-{suffix}-3.onnx"
+            for i in range(4)
+            for suffix in ["best", "last"]
+        ]
     )
 
-    print(f"{expected_items = }")
-    print(f"{found_items = }")
-    assert np.all(
-        found_items == expected_items
-    ), f"{found_items = } != {expected_items = }"
+    assert set(found_items) == set(expected_items), {
+        "missing": sorted(set(expected_items) - set(found_items)),
+        "unexpected": sorted(set(found_items) - set(expected_items)),
+    }
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
index b6602347..ef886f9d 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,3,0.129906,-0.5716,0.18384615,0.48382017,-0.95800894,-0.06879192,-1.5230852,0.32024476,-0.032266557,-0.2324689,1.2026423,1.2646347,-1.6631967,0.8923384,1.0081543,2.146518,1.260503,-0.024123436,-0.4261092,0.71811444,0.6665925,-0.83603954,0.7274643,0.081591725,-0.46056688,-0.7607759,-0.5974563,0.5428709,0.04133952,0.6946457,0.54585093,-1.3706633,0.9921805,-1.1147716,-1.1846336,-0.6535198,1.6928166,-0.6169712,1.1777917,-0.33404595,-1.2148309,0.16786538,1.4557567,-0.7030484,0.40415078,-2.0602548,-0.44774872,-0.9443908,-0.37846982,-1.4340259,-1.0351791,-0.54817986,-1.3410672,1.057497,0.6667623,1.8221252,-1.484842,1.5296896,0.2569867,-2.118917,0.27047974,-0.802363,-0.34508198,-0.7845012,0.29995131,-1.5465921,0.3353841,-0.85972494,0.9633959,0.1057641,0.77337104,1.4520825,-2.3267972,0.8137759,-0.9976677,0.25013983,-0.27935335,0.8346441,-0.33092296,-1.0873228,0.49370176,-0.4727684,-0.7608272,-0.8022482,1.3239352,-2.2216249,0.55208087,-1.1424916,1.1881708,-0.4595066,-0.09527821,0.9202227,0.57056504,-2.0961475,1.643413,-1.046783,-0.116270036,-0.3729071,-0.61354154,-0.8894295,-1.1429787,-0.78733057,0.57443416,1.4698808,0.43613648,-0.075808406,-0.14003257,-2.9505506,-0.016151994,0.6177055,0.68450046,0.5607075,0.87257403,-1.3993292,0.35658643,-0.0063092983,-0.25653496,-0.31633425,-3.5138404,0.88777286,1.6519107,0.40932092,-0.027439872,-0.79843307,-1.1049063,0.07724614,0.56875324,-1.0056511,0.48330793,-0.157245,-1.6178644,0.1934022,-1.9760464,0.0589674,0.5110818,0.089602806,-0.968479,-1.0943018,-0.7229412,1.4727569,0.47599703,-0.6288432,-2.055541,-1.011629,0.9921429,0.5582884,0.50531685,-0.097561255,0.731814,0.6783084,0.3472921,-1.7636684,0.27717423,0.28878227,0.5271399,1.5502092,0.1620326,-0.50224847,1.5970831,-0.6694094,-0.53928703,-1.924573,-0.013257886,-0.10813448,-2.0357018,-1.6700076,0.2103319,1.2394764,-1.1461264,0.036453854,0.40960938,-0.83274007,-0.008255889,-0.30468202,0.5700774,0.6194162,-1.6036439,0.2115395,0.6522949,-0.39073604,0.21115232,-0.46797746,-0.018942848,0.09517513,-0.61526686,0.67094487,-0.45691407,-0.42524102,-0.8716745,0.07637033,-0.72497505,-1.2465461,-0.4429114,1.5433344,-0.8214027,1.8114822,-0.90004414,-0.6661404,0.8775225,0.91526055
+0,0,3,0.12990583,-0.57160014,0.1838459,0.48382023,-0.95800894,-0.06879183,-1.5230852,0.32024524,-0.032266203,-0.23246907,1.2026421,1.264635,-1.6631967,0.8923387,1.008154,2.1465178,1.2605029,-0.024123702,-0.4261095,0.7181141,0.6665921,-0.8360398,0.7274644,0.08159177,-0.46056688,-0.7607759,-0.59745675,0.5428708,0.041339252,0.6946458,0.54585075,-1.3706634,0.99218065,-1.1147714,-1.1846339,-0.65351975,1.6928166,-0.61697114,1.1777916,-0.33404547,-1.2148314,0.16786565,1.4557565,-0.7030481,0.40415096,-2.0602548,-0.44774878,-0.9443907,-0.3784696,-1.4340262,-1.0351793,-0.54817986,-1.341067,1.0574965,0.66676235,1.8221258,-1.4848418,1.5296896,0.25698647,-2.118917,0.2704801,-0.8023632,-0.34508163,-0.7845013,0.29995131,-1.5465924,0.3353845,-0.8597248,0.96339536,0.10576437,0.7733713,1.4520824,-2.3267972,0.81377566,-0.99766785,0.25014007,-0.27935317,0.8346441,-0.33092317,-1.087323,0.4937019,-0.47276843,-0.7608273,-0.80224836,1.3239352,-2.2216246,0.5520812,-1.1424915,1.1881708,-0.45950648,-0.09527827,0.9202227,0.5705651,-2.0961478,1.6434134,-1.0467834,-0.116270155,-0.37290683,-0.6135416,-0.88942945,-1.1429784,-0.78733057,0.57443386,1.4698808,0.43613678,-0.075808465,-0.14003229,-2.95055,-0.016151935,0.6177054,0.6845007,0.56070817,0.87257427,-1.3993288,0.35658634,-0.0063095656,-0.25653526,-0.3163344,-3.5138404,0.887773,1.6519105,0.40932044,-0.027439797,-0.7984332,-1.1049061,0.07724614,0.5687538,-1.0056508,0.48330805,-0.15724464,-1.6178644,0.19340196,-1.9760467,0.058967575,0.5110821,0.08960242,-0.9684786,-1.094302,-0.72294104,1.4727569,0.47599727,-0.628843,-2.055541,-1.0116285,0.992143,0.55828834,0.50531685,-0.09756176,0.7318141,0.67830867,0.34729233,-1.7636684,0.27717403,0.28878212,0.52713984,1.5502089,0.16203246,-0.50224864,1.5970831,-0.6694091,-0.53928715,-1.9245727,-0.013257689,-0.10813395,-2.0357018,-1.6700076,0.21033196,1.2394766,-1.1461272,0.036453612,0.4096094,-0.8327397,-0.008255711,-0.30468205,0.5700773,0.61941594,-1.6036438,0.21153934,0.6522951,-0.39073592,0.21115246,-0.46797752,-0.018942907,0.09517552,-0.61526686,0.6709447,-0.45691368,-0.42524138,-0.8716745,0.07637033,-0.72497517,-1.246546,-0.44291162,1.5433346,-0.82140267,1.8114822,-0.9000441,-0.6661404,0.8775227,0.9152606
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
index 03e75139..91bd4f31 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,3,0.6310362,0.334674,0.16728853,0.36186692,-0.9478092,-0.3926634,-1.3003972,-0.044596065,-0.20526434,-0.026398035,0.19380486,0.8118285,-2.0392873,1.2787219,1.2542659,1.4919833,1.088667,-0.010551525,-0.8448257,0.43219802,0.56950766,-0.8026797,0.74732757,-0.42725757,-1.0895562,-0.21274805,-0.7626764,0.026918262,-0.5983657,0.28117272,0.73443246,-0.9515915,0.81785864,-1.6458837,-0.71132445,0.23738682,1.6034894,0.019240104,1.3241774,-0.20347075,-0.7043628,0.38013214,1.2755842,-1.4004871,0.46275398,-1.9880224,-0.9198009,-1.0473716,-0.65675557,-0.84835905,-1.3224105,-0.70712143,-0.7341193,1.1134466,0.6568125,1.3994768,-0.954168,1.3808511,0.251891,-1.9545913,0.2308793,-0.26487717,-0.64339316,-0.2622317,0.603964,-1.4150456,0.6588127,0.08482099,1.1321739,-0.099473916,0.010979951,1.831232,-1.8722236,0.88451487,-0.76108027,0.080021404,-0.06698817,1.07159,-0.9351867,-1.0435191,0.026436348,-0.41959018,-0.52513933,-0.5412578,2.0294697,-2.7480896,0.53006375,-1.0863824,0.33880177,-0.14623298,0.6506472,1.0939751,-0.48882413,-1.810358,1.3499824,-0.89643645,-0.008765998,-0.06027395,-0.48046795,-1.2652203,-1.4560691,-0.946743,-0.27704963,1.9105397,0.20772824,-0.3049073,-0.34610513,-3.3094442,-0.46867853,0.65010065,0.4332434,0.5593448,0.8796511,-0.8775561,0.56195873,0.2979048,0.54411083,-0.49651542,-2.4436908,1.2963794,1.9646289,0.41595295,0.077845775,-0.8488377,-1.2093579,-0.2594095,0.12893897,-0.5105848,0.64491534,-0.017900532,-2.1271238,0.27657434,-1.7817888,-0.6613088,0.2365224,-0.07435218,-1.0712526,-0.80908316,-0.77087396,1.1399101,0.7127807,-0.94809335,-2.0319622,-0.7065952,1.1998774,0.56364894,0.62384087,-0.25050914,-0.12355562,0.011005514,0.4171477,-2.2401168,1.0053111,0.10580753,0.5745275,1.6763166,0.3952029,-0.3734366,2.0440457,-0.5067878,-0.59976774,-2.560222,0.28748652,-0.068856396,-1.7209245,-1.5318912,0.7257471,0.8677974,-0.9443491,0.32327485,0.09680152,-1.0634658,0.2343228,-0.73152024,0.6149891,0.50479555,-1.6433326,0.36874586,1.5886227,-0.48504025,-0.5036868,-0.24184056,-0.20931806,-0.046914995,-0.6832404,0.3294942,0.5522635,-0.77527386,-1.4231143,0.22487257,-0.16668221,-1.7301935,0.025722886,1.6323115,-1.0889417,1.2082901,-0.3673339,0.19397202,0.32195246,1.1962756
+1,0,3,0.6310361,0.33467388,0.16728823,0.36186728,-0.94780934,-0.39266327,-1.3003972,-0.044595696,-0.20526414,-0.026398215,0.1938047,0.81182855,-2.039287,1.2787219,1.2542659,1.4919837,1.0886672,-0.010551577,-0.8448256,0.43219796,0.5695074,-0.80267984,0.7473274,-0.42725766,-1.0895563,-0.21274823,-0.7626765,0.026918093,-0.5983658,0.28117266,0.73443246,-0.9515915,0.8178587,-1.6458836,-0.71132463,0.23738661,1.6034894,0.019240078,1.3241775,-0.20347083,-0.7043624,0.38013235,1.2755847,-1.4004868,0.46275395,-1.9880222,-0.91980124,-1.0473715,-0.65675545,-0.84835905,-1.3224108,-0.70712143,-0.7341194,1.1134468,0.6568125,1.3994766,-0.95416814,1.3808515,0.25189096,-1.9545913,0.23087955,-0.26487717,-0.64339334,-0.2622316,0.6039643,-1.415046,0.6588125,0.08482099,1.1321739,-0.09947382,0.01097981,1.8312315,-1.8722234,0.88451487,-0.7610804,0.08002127,-0.06698804,1.0715898,-0.93518686,-1.0435194,0.026436012,-0.41959038,-0.5251393,-0.54125756,2.0294697,-2.7480886,0.5300638,-1.0863824,0.3388017,-0.14623287,0.6506474,1.0939751,-0.4888243,-1.8103579,1.3499825,-0.8964366,-0.008765843,-0.060274,-0.4804681,-1.2652204,-1.4560695,-0.9467429,-0.2770497,1.9105395,0.20772837,-0.30490705,-0.34610507,-3.309444,-0.46867862,0.6501004,0.43324322,0.559345,0.87965095,-0.87755597,0.5619585,0.29790485,0.5441109,-0.49651536,-2.4436905,1.2963796,1.9646293,0.41595292,0.07784585,-0.8488378,-1.2093579,-0.25940973,0.12893897,-0.51058507,0.64491534,-0.017900245,-2.127124,0.27657425,-1.7817887,-0.66130847,0.2365224,-0.07435237,-1.0712522,-0.8090829,-0.7708737,1.1399105,0.71278065,-0.9480938,-2.0319617,-0.7065952,1.1998773,0.56364936,0.62384117,-0.25050887,-0.123555444,0.011005579,0.41714785,-2.2401168,1.005311,0.10580762,0.57452756,1.6763165,0.39520282,-0.37343678,2.0440457,-0.50678766,-0.59976757,-2.5602221,0.28748658,-0.06885617,-1.7209247,-1.531891,0.72574735,0.8677975,-0.944349,0.32327467,0.09680159,-1.0634657,0.23432283,-0.73151994,0.61498916,0.50479555,-1.6433319,0.36874565,1.5886225,-0.4850403,-0.5036866,-0.24184072,-0.20931797,-0.046914864,-0.6832405,0.32949412,0.5522636,-0.77527416,-1.4231143,0.22487274,-0.16668202,-1.7301934,0.025722574,1.6323113,-1.0889417,1.2082905,-0.36733398,0.19397217,0.32195246,1.1962752
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
index ae9e5907..6fefa7ce 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,2,0.14896812,-0.7527326,0.098616265,1.5360665,-0.83212125,-0.8661572,0.87993956,0.7707245,0.8990906,0.35979807,1.5049816,0.9981651,-2.2045355,0.46189427,1.3522851,0.81136656,-0.3787586,-0.41562492,-1.6569216,0.9181141,0.8893558,-0.6614972,0.35581687,1.0705252,0.61813873,-1.6933419,-1.6523572,0.4721115,-0.6415743,-0.8052874,-1.1905364,-1.3483986,0.7520653,-1.4446422,0.032763932,-0.77299255,0.8698881,-0.9012206,-0.105312526,-0.03295452,-1.7241203,0.40647608,1.3450831,0.00073365227,0.23433386,-0.9771891,-0.1076481,1.2662536,-0.7311288,-0.6664411,0.006385311,-0.31623417,0.108964175,0.43619278,0.284846,0.34632137,0.13046862,1.7500304,-0.31573698,-2.1822793,-0.073760144,-0.63938767,-2.4171443,-1.3825034,1.4659554,0.061223067,-2.0762868,-1.1751351,-0.081808135,-1.6416469,0.61133504,1.4318693,-1.5806884,1.4651628,-0.42543516,0.27195463,-1.9691073,0.5124442,0.7785031,-1.2194265,-1.2140163,-0.21277447,-0.64433247,-0.34592688,1.480311,-1.6828288,0.5201631,-0.45541105,-0.17800705,-0.29835147,0.06779409,0.17683165,-0.5690362,-0.4795994,0.24317117,0.7038561,0.47528592,0.38569337,-0.92215085,0.12652376,-0.5222174,0.79115415,0.1062228,1.2142566,0.57600933,0.3640244,1.4705232,-1.8675983,-0.151888,0.84477943,0.5522926,0.25622973,0.50753284,-0.87316334,-1.7905984,-0.8826149,-0.80747306,1.2090178,-2.0752993,-0.18883346,0.35716313,0.12954345,-0.5863821,-0.3407679,-0.8663318,0.8249107,1.2127439,-1.0448643,-0.53810304,-0.6317458,-0.50353926,0.4781922,-0.91658026,-0.21077591,0.49530765,0.66140443,-0.061057545,-0.011890421,-1.4830853,-0.5170973,1.2860987,-0.48179677,-1.2840593,-0.12392463,-0.5612262,0.5283732,1.6894128,-0.015182488,1.6194154,0.6165103,0.846802,-1.259089,-0.19867773,0.05881237,0.87158996,1.7141718,-2.1852305,-0.17253563,1.7481165,0.7287171,-0.9333205,-0.5228718,-1.1484395,-1.0014697,-1.5860218,-0.9670557,0.5662043,-0.6358826,0.6285491,1.4611297,-1.5053117,-0.73541224,-1.1782357,0.4759477,0.56338716,0.29015008,-0.5925466,0.4408311,-0.23582521,-1.1164836,1.0048937,-0.39400414,-0.7058568,0.99132824,-1.3044466,2.0163782,-1.8661258,0.013690879,-0.98518753,-0.2840235,-1.4385029,-0.70052123,-1.5944667,1.7985247,0.9840921,1.857999,-0.74875927,1.5471251,0.73758185,0.14302142
+2,0,2,0.14896777,-0.7527335,0.09861636,1.5360669,-0.8321218,-0.8661579,0.8799402,0.7707242,0.89909035,0.3597984,1.5049816,0.99816555,-2.204535,0.4618939,1.3522851,0.811366,-0.37875888,-0.4156246,-1.6569219,0.918114,0.88935566,-0.6614972,0.3558165,1.0705256,0.6181391,-1.6933414,-1.6523575,0.47211194,-0.64157414,-0.8052875,-1.1905366,-1.3483986,0.75206506,-1.4446423,0.032763835,-0.77299273,0.8698887,-0.90122116,-0.105313264,-0.032954592,-1.7241205,0.4064759,1.3450825,0.00073461805,0.2343334,-0.9771892,-0.10764814,1.2662542,-0.7311291,-0.6664417,0.0063853106,-0.31623444,0.10896421,0.43619284,0.28484625,0.34632155,0.1304682,1.7500312,-0.3157372,-2.1822789,-0.07376069,-0.63938814,-2.4171436,-1.382503,1.4659556,0.061222862,-2.076287,-1.1751353,-0.08180853,-1.6416469,0.61133516,1.4318693,-1.5806879,1.4651626,-0.42543525,0.27195483,-1.9691069,0.51244324,0.77850366,-1.2194266,-1.2140155,-0.2127739,-0.64433223,-0.3459279,1.4803112,-1.6828287,0.5201632,-0.4554113,-0.17800674,-0.29835144,0.06779454,0.17683099,-0.5690358,-0.47959915,0.24317142,0.70385575,0.47528574,0.38569346,-0.92215097,0.12652385,-0.52221704,0.79115444,0.106221735,1.2142566,0.5760093,0.36402392,1.4705237,-1.8675978,-0.151888,0.84478027,0.55229247,0.25622958,0.5075333,-0.8731629,-1.790598,-0.88261557,-0.80747306,1.2090179,-2.0752997,-0.18883318,0.35716256,0.12954298,-0.58638185,-0.34076798,-0.8663316,0.82491094,1.212744,-1.0448644,-0.53810245,-0.63174623,-0.5035392,0.4781915,-0.9165807,-0.21077545,0.49530777,0.6614037,-0.06105711,-0.011890445,-1.4830858,-0.5170972,1.2860987,-0.48179644,-1.2840594,-0.123924315,-0.561227,0.5283733,1.6894126,-0.015183194,1.619416,0.61651087,0.84680295,-1.2590885,-0.19867837,0.058812138,0.8715905,1.7141719,-2.1852307,-0.17253566,1.748117,0.72871757,-0.93332094,-0.5228718,-1.1484399,-1.0014701,-1.5860217,-0.9670559,0.5662045,-0.63588244,0.62854993,1.4611295,-1.5053114,-0.7354125,-1.1782357,0.4759481,0.56338674,0.29015055,-0.5925469,0.44083118,-0.23582503,-1.1164836,1.0048939,-0.39400432,-0.705857,0.9913283,-1.3044467,2.0163794,-1.8661263,0.013690455,-0.9851868,-0.28402337,-1.4385033,-0.7005213,-1.5944669,1.7985251,0.98409235,1.8579992,-0.7487591,1.5471251,0.7375818,0.14302124
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
index 26935d23..baf0de12 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,2,0.3853767,-0.7623611,0.5445376,1.1489445,-0.93219095,-1.0841124,-0.27961808,0.19042599,0.06699983,0.6020505,0.7307122,0.9754021,-2.318475,0.97368747,1.6034338,0.94340336,1.2762036,-0.30378067,-0.97008324,0.62717986,1.3900626,-0.9464742,0.4837369,-0.030570975,-0.4290407,-1.1276956,-0.7923079,0.563961,-0.40978292,-0.23345281,-0.27136502,-0.8576837,0.69629353,-1.6003033,0.11132374,-0.5791753,1.9399358,-0.92576635,0.53490555,-0.49527624,-1.3456327,-0.176038,1.349483,-1.1282389,0.20037875,-0.955993,-0.8822724,-0.07519414,-0.77170074,-0.9350284,-0.5532876,-0.94019765,-0.10448791,1.1707093,0.6575927,1.3083346,-0.6112351,1.6713382,0.16979486,-2.0898654,0.00984007,-0.44183394,-1.1921827,-1.2657421,1.6432866,-1.130562,-0.46437404,-0.8430221,0.7453206,-0.65720564,0.35852155,1.6237744,-2.0383017,1.1427602,-0.8917945,0.20761068,-1.3489536,0.62074554,-0.44713894,-1.8621219,-0.5244267,-0.45581624,-0.39755735,-0.43458134,1.8005577,-2.0100255,0.3588848,-1.2393866,0.95991385,-0.65878445,0.741961,0.43640068,-0.37811747,-1.3537853,1.0124421,0.12569562,0.45669925,-0.062139705,-0.6899904,-0.7770923,-1.2683299,-0.42139253,-0.4123218,1.7224118,0.38184428,-0.5022623,0.1969878,-2.748319,-0.18802647,1.3159317,0.4502178,-0.26388726,0.5703611,-0.85277075,-0.46127146,0.23827219,-0.4298961,0.41151482,-2.6187694,0.5646321,1.1139392,-0.032928437,0.1021941,-0.6601091,-1.3054987,0.09395422,0.43201578,-1.4673975,0.08933542,-0.71540993,-1.6057761,0.4075759,-1.6277498,0.05748504,-0.18920064,0.1056629,-0.8918468,-0.17913246,-1.4436157,0.077475116,0.6109399,-0.83478075,-1.3056538,-0.87130284,0.28275952,0.87232274,0.8705629,-0.14861424,0.91572547,0.6586051,0.6749446,-1.7701945,0.1629751,0.4728509,1.0638493,1.7032028,-0.6542733,-0.017882407,2.1696196,-0.047487322,-0.77344674,-2.259672,-1.1815661,-0.5823015,-1.638453,-1.8893389,0.9176421,0.32815483,0.4039307,0.17879207,-0.32735565,-1.1835169,-1.0626798,-0.12413608,-0.13082603,0.5998312,-1.7939893,0.5362146,1.2261001,-1.4384091,0.4796257,-0.45061567,-0.100991376,0.51506233,-1.2538948,1.3633755,-0.58895797,-0.21253212,-1.3333261,0.50792605,-0.97876984,-1.5163478,-0.6666502,1.8272516,-1.4814882,2.3199954,-0.6350161,0.1867251,0.83364874,0.9136821
-4,0,2,1.8327771,0.9652215,-0.25314748,0.21621384,1.9031243,0.7036974,-0.930624,-0.8883423,0.55273986,0.5570372,1.0132973,-0.96464247,-0.51950616,-0.05579906,-0.37479147,0.44072837,0.16671647,0.5790387,0.41375738,0.80011296,-1.5359617,-0.22593148,1.7854865,-0.71542114,-1.0921632,-0.041372035,-0.21795599,-0.3369114,-1.2671196,-0.7473329,0.18677177,-1.8774186,2.0318227,0.2733341,-0.9697609,-0.58609706,-0.36415994,0.22731349,0.4721085,-1.3540337,-0.25599056,0.65862226,1.5556433,-0.81543595,1.168724,-2.2140863,0.42116505,0.47876942,-0.6478898,0.44233415,0.20755047,-0.5501464,-0.5956206,-0.60861623,-1.3033564,-0.9219007,1.9484233,-0.02818083,0.9555862,-0.6107661,1.3185904,0.4262676,-1.1561952,-0.5161904,-0.8352084,0.22098036,-0.32627368,-1.1956955,1.7369198,-1.1766452,0.07778335,-1.4228516,-2.164931,0.61946803,-0.4467436,-0.01088463,-0.43634295,1.737324,0.6609657,-0.20005353,-0.56503403,0.29499987,0.11585043,-0.05404717,1.7766248,-1.624335,0.4819137,1.0377663,-1.2305183,0.8004889,-1.9219416,1.6361125,0.13642022,-1.2401025,0.56233984,-0.15820779,1.5966985,-0.25904042,-0.5514964,0.43703973,0.4001095,-0.12492117,1.6866364,2.0976322,-0.49593046,0.8581702,0.2179227,-0.6346643,-1.7050065,-1.6924847,-0.08375938,-0.38175428,-0.32693657,-0.9640046,-0.9026502,-0.5941344,0.19496141,-0.8319495,-0.55289805,-0.80228716,1.650307,1.2991756,-1.0835283,0.39504835,0.29286614,-0.13108593,-0.38219267,-0.59767234,-1.3788447,-0.26507112,0.34214637,1.0462471,1.5181249,-0.59406966,0.58952206,2.7725644,-1.0286809,1.18978,0.3369609,-1.0014182,-0.02982089,-0.5898692,-0.99967504,0.01042568,1.4441023,-0.87508506,0.57642156,2.5007522,-1.5390786,-1.6458069,-1.7197639,-0.74723464,1.1642295,-0.31345525,-0.0055794707,1.5905838,-0.9592406,-0.4174035,-0.6339656,0.8530592,1.2996758,-0.11439483,0.3914311,0.46125937,-3.0907335,-0.27514952,0.20107323,-1.4204425,-1.7638282,-0.48746726,-1.2546072,-0.2341572,-0.10110028,-0.24766739,1.7387944,-0.7589158,0.75030494,0.5836463,0.10834081,0.7989048,-0.4846295,-0.62584215,0.5027704,1.1866003,-0.28132582,-0.80156577,0.44897035,1.1347668,0.07279313,-0.61313444,0.8892234,0.71706593,0.65837866,0.33229896,1.174505,-0.7435528,0.5451965,0.36238024,1.4138215,0.37747207
+3,0,2,0.38537654,-0.7623611,0.5445376,1.1489445,-0.93219095,-1.0841125,-0.27961814,0.19042616,0.06699963,0.6020508,0.7307117,0.975402,-2.3184752,0.97368747,1.6034342,0.94340354,1.2762038,-0.30378082,-0.97008324,0.62717956,1.3900623,-0.9464736,0.48373678,-0.030571124,-0.42904106,-1.1276956,-0.7923078,0.5639612,-0.40978295,-0.23345235,-0.27136517,-0.8576837,0.69629353,-1.600303,0.11132369,-0.5791752,1.9399357,-0.92576593,0.5349059,-0.49527636,-1.3456329,-0.17603783,1.3494829,-1.1282393,0.20037884,-0.955993,-0.8822727,-0.0751942,-0.7717005,-0.93502855,-0.5532879,-0.9401976,-0.10448796,1.1707094,0.6575927,1.3083345,-0.611235,1.6713382,0.16979468,-2.0898654,0.0098398905,-0.44183353,-1.1921822,-1.2657422,1.6432862,-1.1305616,-0.46437377,-0.8430222,0.7453205,-0.6572058,0.3585215,1.6237743,-2.0383017,1.1427608,-0.8917951,0.20761041,-1.3489532,0.62074584,-0.447139,-1.8621218,-0.52442676,-0.45581636,-0.39755735,-0.43458155,1.8005574,-2.0100255,0.35888514,-1.2393867,0.95991373,-0.65878403,0.74196094,0.43640032,-0.37811738,-1.3537853,1.012442,0.12569553,0.45669976,-0.0621398,-0.68999076,-0.7770921,-1.2683296,-0.42139223,-0.41232187,1.7224118,0.3818442,-0.5022627,0.1969878,-2.7483199,-0.18802676,1.315932,0.45021772,-0.2638872,0.5703608,-0.85277075,-0.4612713,0.23827244,-0.42989627,0.41151485,-2.6187692,0.564632,1.1139395,-0.03292864,0.102194116,-0.6601094,-1.3054988,0.09395442,0.43201578,-1.4673973,0.08933548,-0.71541005,-1.6057762,0.40757605,-1.6277497,0.057484966,-0.18920068,0.10566316,-0.89184695,-0.17913231,-1.4436156,0.07747503,0.6109399,-0.83478105,-1.3056535,-0.87130284,0.28275958,0.87232304,0.8705628,-0.1486141,0.91572547,0.6586048,0.6749444,-1.7701942,0.16297537,0.47285134,1.0638492,1.7032028,-0.65427315,-0.017882247,2.1696198,-0.047487043,-0.773447,-2.2596717,-1.181566,-0.58230144,-1.6384531,-1.8893386,0.9176424,0.32815474,0.4039305,0.17879185,-0.32735547,-1.1835169,-1.0626794,-0.124136135,-0.13082623,0.5998316,-1.793989,0.5362147,1.2261003,-1.4384094,0.4796257,-0.45061588,-0.100991115,0.5150622,-1.253895,1.3633755,-0.5889575,-0.21253206,-1.333326,0.5079265,-0.97876984,-1.5163475,-0.6666502,1.8272516,-1.481488,2.3199954,-0.6350159,0.1867249,0.83364856,0.9136824
+4,0,2,1.8327769,0.9652212,-0.25314742,0.2162136,1.9031242,0.70369756,-0.93062437,-0.8883426,0.5527405,0.5570374,1.0132976,-0.9646425,-0.5195063,-0.05579895,-0.37479118,0.44072798,0.16671617,0.5790385,0.41375798,0.80011284,-1.535962,-0.22593139,1.7854862,-0.7154213,-1.0921632,-0.041371867,-0.21795613,-0.33691156,-1.2671199,-0.74733275,0.18677123,-1.8774186,2.0318227,0.27333388,-0.9697611,-0.5860972,-0.36415997,0.22731379,0.47210854,-1.3540338,-0.25599036,0.6586218,1.5556432,-0.8154355,1.1687236,-2.214086,0.42116502,0.47876915,-0.64788955,0.44233435,0.20754997,-0.55014664,-0.5956206,-0.6086165,-1.3033571,-0.9219002,1.9484233,-0.028180521,0.9555862,-0.6107659,1.3185904,0.42626792,-1.1561952,-0.51619065,-0.83520895,0.2209806,-0.3262739,-1.1956955,1.7369201,-1.1766452,0.07778294,-1.4228518,-2.164931,0.61946815,-0.44674355,-0.0108849695,-0.4363428,1.737323,0.6609657,-0.20005423,-0.5650333,0.29500026,0.11585065,-0.054047324,1.7766253,-1.624335,0.48191375,1.0377663,-1.2305183,0.8004891,-1.9219418,1.6361127,0.13642037,-1.2401029,0.56233984,-0.158208,1.5966983,-0.25904053,-0.55149657,0.43704,0.40011016,-0.12492106,1.6866363,2.0976322,-0.49593005,0.85817003,0.21792285,-0.6346643,-1.7050064,-1.6924843,-0.083759315,-0.38175458,-0.32693613,-0.96400416,-0.9026496,-0.594135,0.19496153,-0.8319494,-0.55289817,-0.802287,1.6503073,1.2991755,-1.083528,0.3950478,0.29286572,-0.13108596,-0.38219288,-0.59767187,-1.3788453,-0.26507127,0.3421463,1.0462469,1.5181253,-0.59406966,0.58952194,2.7725646,-1.0286808,1.1897799,0.33696112,-1.0014181,-0.029820807,-0.5898688,-0.99967515,0.010425864,1.4441029,-0.87508553,0.5764213,2.5007524,-1.539079,-1.6458069,-1.7197638,-0.74723464,1.1642295,-0.313455,-0.005579108,1.5905834,-0.9592405,-0.4174034,-0.6339656,0.85305965,1.299676,-0.114394456,0.39143074,0.46125907,-3.0907335,-0.27514967,0.20107278,-1.4204426,-1.7638285,-0.48746756,-1.254607,-0.23415706,-0.101100445,-0.24766745,1.738794,-0.75891614,0.7503048,0.58364636,0.10834089,0.79890466,-0.48462954,-0.62584233,0.50277036,1.1866002,-0.28132573,-0.8015658,0.44897038,1.1347669,0.072793104,-0.6131346,0.88922364,0.71706647,0.6583793,0.3322992,1.1745046,-0.7435528,0.5451967,0.36237976,1.4138212,0.3774721
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
index 0777c948..a053e062 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,2,0.8678487,-1.1492083,-0.2957783,0.9786945,-1.9542453,-0.07664241,-0.039730284,-0.09718514,-0.465507,0.088307,0.014937709,1.1039451,-1.9457144,0.7994327,1.8773037,1.5537523,0.39788526,-0.4953133,-1.4370357,1.0003625,0.84108806,-0.99132866,0.15745479,0.4423289,-1.5118519,-1.9748083,-1.3705087,0.11752797,0.04718379,0.7227247,0.15624402,-1.3321223,0.8274153,-0.5020909,-0.8579634,-0.8208209,1.370573,-0.32510355,0.9213625,-0.45679018,-2.0167282,0.77819085,1.3592292,-1.6879606,0.35116172,-1.6433353,-0.8796014,0.47101834,-0.49356297,-1.540319,-0.56203103,-0.74793,0.114850305,0.5929445,1.0806793,0.8520335,-0.28449142,2.3426988,-0.71767795,-2.077216,0.4469343,-0.51118195,-1.8134437,-0.31995004,0.5275096,-0.7924808,-0.90600455,-0.5945146,0.62720716,-0.76678073,1.4406855,0.8350828,-1.3691337,0.43675637,-0.93423057,-0.58701897,-0.53982204,1.3742961,-0.3684079,-1.3804277,-1.3538327,-1.2029366,-0.6361641,-0.2061667,1.1179498,-0.8696951,0.46374846,-0.9316529,0.36905092,0.18406926,0.5726551,-0.097014,-0.3084513,-1.0851082,1.2294934,-0.43448213,0.4714254,0.29762262,-0.66758883,-0.2411063,-0.8684391,-0.15433899,0.3217168,1.7343627,0.6421659,-0.39611214,1.2057977,-3.0774305,0.8041049,1.0713868,0.3196744,0.885012,0.5413774,-1.8904048,-0.6433149,-0.7131405,-0.89828914,0.49162254,-1.7892698,1.266022,1.2143357,0.16972247,-0.4222612,-0.9209112,-1.2874361,-0.30553934,0.78019506,-1.3130422,-0.54638046,-0.06269859,-1.2766076,0.3067186,-1.1979859,-0.5041537,0.24134871,0.9985403,-0.20058915,-0.19003288,-0.9156221,0.73883057,0.60628974,-0.39632013,-1.401826,0.26617125,1.2859261,1.7502289,1.1291437,-0.2593595,1.014177,-0.002415313,-0.0126089025,-1.6622946,0.84452736,0.40896028,0.24172997,0.849172,-1.2459072,0.4409478,2.3070042,-0.20285252,-0.9055849,-1.9905902,-0.6291581,-0.70495003,-1.5236276,-1.6411536,0.28731304,0.23528825,-0.85808456,0.8726922,-0.8224612,-1.3814898,-0.8996288,0.0075938655,1.4490063,0.62249285,-1.3583359,-0.39259228,0.55505204,-1.1764418,1.2858897,-0.5120127,-0.3620197,0.028119635,-1.0583314,0.6261959,-0.93444824,-0.68522364,-0.5574861,0.12772661,-1.1170981,-1.7487799,-0.62091166,1.1936655,-0.55242705,1.6852115,-1.5539129,0.93077433,0.54523623,-0.32935345
+5,0,2,0.8678485,-1.1492082,-0.29577842,0.9786945,-1.9542456,-0.0766423,-0.039730456,-0.097185545,-0.46550742,0.0883074,0.014938131,1.1039449,-1.9457139,0.79943234,1.8773037,1.5537525,0.39788535,-0.49531272,-1.4370352,1.0003623,0.8410879,-0.9913289,0.15745485,0.44232905,-1.5118521,-1.9748081,-1.3705083,0.11752842,0.047184132,0.72272474,0.1562442,-1.3321223,0.82741535,-0.50209045,-0.85796386,-0.8208208,1.3705729,-0.32510337,0.92136246,-0.45679024,-2.0167284,0.77819073,1.3592293,-1.6879604,0.35116142,-1.6433353,-0.8796013,0.4710182,-0.49356297,-1.540319,-0.56203073,-0.74792975,0.11485019,0.59294385,1.0806789,0.85203356,-0.2844916,2.3426995,-0.7176778,-2.077216,0.44693428,-0.5111819,-1.8134439,-0.31994978,0.52750957,-0.79248077,-0.9060044,-0.5945146,0.62720716,-0.7667804,1.4406855,0.8350828,-1.3691337,0.436756,-0.9342305,-0.58701885,-0.5398224,1.3742958,-0.36840814,-1.3804276,-1.3538328,-1.2029362,-0.6361641,-0.20616648,1.1179501,-0.8696952,0.463748,-0.9316529,0.36905104,0.18406911,0.57265526,-0.09701394,-0.30845135,-1.0851074,1.229493,-0.4344818,0.47142524,0.29762283,-0.6675888,-0.24110633,-0.8684386,-0.1543391,0.32171702,1.7343627,0.6421661,-0.39611164,1.2057976,-3.07743,0.80410504,1.0713863,0.3196743,0.8850115,0.54137725,-1.8904045,-0.6433146,-0.7131403,-0.89828914,0.4916221,-1.7892698,1.2660217,1.2143362,0.16972245,-0.42226088,-0.9209111,-1.2874365,-0.30553946,0.78019506,-1.3130425,-0.54638034,-0.06269859,-1.2766078,0.3067184,-1.197986,-0.50415367,0.24134831,0.99854076,-0.20058848,-0.19003294,-0.91562223,0.7388306,0.6062894,-0.39632004,-1.4018264,0.2661708,1.2859259,1.7502286,1.1291435,-0.25935975,1.0141768,-0.0024153837,-0.012609187,-1.6622946,0.84452695,0.40896052,0.24173,0.8491717,-1.2459071,0.44094783,2.3070037,-0.20285298,-0.9055851,-1.99059,-0.629158,-0.7049506,-1.5236279,-1.641153,0.2873126,0.23528822,-0.8580843,0.8726923,-0.8224612,-1.3814898,-0.8996295,0.007593808,1.4490062,0.6224926,-1.3583354,-0.39259246,0.5550518,-1.1764413,1.2858899,-0.5120128,-0.36201936,0.028119579,-1.0583311,0.62619597,-0.93444824,-0.6852235,-0.5574866,0.12772661,-1.117098,-1.7487797,-0.62091154,1.1936653,-0.5524274,1.6852114,-1.5539129,0.9307747,0.54523605,-0.32935366
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
index 79e33b1c..22cbffa0 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
@@ -1,3 +1,3 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,3,0.47420236,1.2638448,-0.2942875,0.6063466,0.32506454,-0.011651637,-1.3833891,-1.2344038,-0.45170528,1.0377885,1.8840404,-0.33119497,-1.0696628,0.9726884,0.80814546,0.9354395,0.7167119,0.42845112,-0.015223959,0.81312424,-1.2162951,-1.1050558,0.6366257,-1.6387163,-0.9819105,-0.4953311,0.12158322,-0.67469686,-1.7174512,-0.2108714,1.0931984,-1.6311551,1.6655928,-1.1295462,-0.82908773,-0.27834365,0.35710472,0.7679168,1.2862792,-1.5655477,-0.47575817,0.38371342,0.8257985,-0.6541342,-0.06021277,-1.688667,0.31571636,-1.3439896,-1.3430765,0.0015745121,-0.87187445,-1.2474327,-0.82622105,0.53904265,0.4325758,0.7165552,-0.013134802,1.0078027,1.2916025,-1.6230769,0.1835245,1.4117191,-0.9933948,-0.17660141,-0.85399544,0.1709921,1.0725634,-0.23263937,1.7940853,0.49845877,-0.58278996,0.9100562,-2.1336787,-0.0805108,-0.15371813,0.5364138,0.036808733,0.27225935,0.43129516,-0.5169288,0.8617008,0.37777528,-0.3340564,-0.50667363,2.3349266,-3.2664785,0.23907632,0.13636121,-0.1356525,-0.2975334,-0.56251806,1.5495608,-0.25081053,-2.012842,1.4139545,-0.16621536,0.7844478,0.5935697,-0.85539633,-0.5690532,-0.5415988,-0.8971024,1.8330035,2.7651486,1.039496,-0.23426563,-0.57880294,-1.180363,-1.095461,-0.30837286,0.20563056,-0.45784765,0.799281,-0.585805,0.84253716,0.11244704,0.7201907,-1.2291319,-1.666961,-0.44462907,2.2034056,1.0664692,0.81338817,-0.4559743,-1.2628946,-0.22252318,-1.3892785,-0.74601096,0.09792413,-0.9173109,-1.1128796,0.4023642,-0.7147697,-0.77360827,0.39257395,2.3405876,-0.2756536,0.4053474,-0.8130045,-0.097918354,0.619593,-1.450402,-1.562418,-1.1680082,1.2857003,-1.2578353,0.41585723,1.0423979,-0.53722227,-0.38153183,-0.51949257,-1.5859616,1.3256049,-0.097610995,0.5610497,2.6576302,1.24299,-0.7566359,0.91003966,0.73991305,-0.10965724,-2.0155516,-0.009600935,-0.68587035,-1.7821821,-1.0589889,-0.13436046,0.4220983,-1.1379086,0.48460454,-0.61625963,-0.3349504,0.53863937,0.18685772,0.26678517,-0.20175351,-0.7204558,0.21726036,1.1120167,0.79319596,-0.36998117,-0.6713484,0.72676253,-0.18856229,0.46767554,-1.068567,0.32332298,-0.06708165,-1.2248515,-0.7845498,0.5460865,-0.61680746,1.510683,1.7673895,-0.68418354,0.44412625,0.7936944,-0.8888349,0.03981725,1.6578218
-7,0,2,-0.087797716,-0.72064584,0.3919503,0.30277744,-1.0405561,-0.8779837,-0.7976403,0.071234636,-0.29967022,0.2773496,-0.10566512,0.77108765,-1.7671878,0.824929,1.5373951,1.5010269,1.4846264,-0.7896349,-0.88700837,0.6919234,0.6787521,-0.4330691,0.6862605,-0.2904699,-0.9619546,-0.36133736,-0.9279964,0.18862396,-0.13149612,0.33348706,0.35246968,-0.86275643,0.7647198,-1.5118189,-0.24293631,-0.62797725,1.9434764,-0.15230915,1.2640054,-0.53051674,-1.571733,-0.23882791,1.2569784,-1.5155659,0.31609696,-1.4623228,-1.5276749,-0.63017756,-0.15530244,-1.2590177,-1.1136601,-0.8283698,-0.33160418,1.6386752,0.7145629,1.4172399,-0.9240341,1.155421,-0.20543078,-2.2690709,0.032772336,-0.6074186,-0.8408421,-0.43909267,1.1674483,-1.5343087,0.394567,-0.6319556,0.69497925,-0.17052399,0.66387993,1.6770401,-1.6856111,0.8501759,-1.0100725,-0.06956196,-0.49753228,0.45815924,-0.8491857,-1.5388991,-0.21994229,-0.61590695,-0.43685135,-0.54839456,1.844041,-1.9056602,0.23943688,-0.8901972,1.0178715,-0.4182901,0.47147426,0.28248206,-0.42557982,-1.3693498,1.378879,-0.832486,0.2925802,0.03857666,-0.99315274,-0.88979,-1.3541446,-0.38844082,-0.56667507,2.0802116,0.20410404,-0.5688037,-0.25536713,-2.8470826,0.22910759,1.339881,0.41405326,0.3723613,1.0213392,-0.61211425,0.20765173,0.22666255,-0.052260146,0.20032822,-2.3570492,1.4302993,1.4743407,-0.2883624,0.44256923,-0.7141606,-0.87645143,-0.20459794,0.5313736,-1.1576004,0.2882262,-0.3185675,-2.041501,0.55573565,-1.3157725,0.2728311,0.18704554,-0.34522107,-0.7096739,-0.43224367,-1.601502,1.0732054,0.91898465,-0.68341035,-1.5872612,-0.59699357,0.90725124,1.1939424,0.6279298,-0.49085775,0.9603635,0.668255,0.66679686,-1.698892,0.26019427,0.57732755,0.5378443,1.2200284,0.32803506,-0.15348187,2.141097,-0.014183752,-1.0378197,-2.6077065,-0.3066731,-0.06714382,-1.3892245,-1.7354015,0.53556174,0.9041617,-0.32855996,0.12410693,-0.40722385,-1.3432457,-0.73200125,-0.25400352,0.17272082,1.1034756,-2.2850537,0.6256181,1.7550697,-1.3452938,-0.017986335,-0.54053277,0.05288191,0.28705782,-0.98898697,0.7631376,0.15638618,-0.6830087,-1.3387133,0.62448514,-0.7181843,-1.9862368,0.070950225,1.8409837,-1.5815738,1.639628,-0.8540574,0.19265085,0.4421182,0.67777854
+6,0,3,0.47420266,1.2638454,-0.29428756,0.60634655,0.32506496,-0.011651864,-1.3833889,-1.2344038,-0.45170534,1.037789,1.8840408,-0.33119506,-1.0696626,0.9726885,0.8081458,0.93543977,0.71671206,0.42845142,-0.015223262,0.8131241,-1.2162951,-1.1050558,0.6366258,-1.6387162,-0.981911,-0.4953308,0.12158294,-0.6746963,-1.7174507,-0.21087167,1.0931985,-1.6311551,1.6655928,-1.1295464,-0.82908833,-0.2783438,0.35710457,0.7679168,1.2862791,-1.565548,-0.47575852,0.38371316,0.8257988,-0.6541346,-0.06021297,-1.6886668,0.3157164,-1.34399,-1.3430762,0.0015743595,-0.87187505,-1.2474334,-0.82622164,0.5390428,0.43257558,0.7165556,-0.0131348735,1.0078024,1.2916025,-1.6230769,0.18352437,1.4117194,-0.9933953,-0.17660153,-0.85399514,0.17099203,1.0725635,-0.23263934,1.7940859,0.49845862,-0.5827904,0.9100559,-2.1336787,-0.08051071,-0.15371832,0.5364136,0.036808893,0.2722593,0.4312951,-0.5169293,0.86170137,0.37777555,-0.33405718,-0.50667346,2.3349266,-3.2664793,0.23907636,0.13636166,-0.13565257,-0.2975336,-0.56251824,1.5495615,-0.25081053,-2.0128424,1.4139546,-0.1662151,0.78444827,0.5935687,-0.8553964,-0.56905305,-0.5415987,-0.8971025,1.8330035,2.7651489,1.0394961,-0.234266,-0.5788035,-1.1803628,-1.0954615,-0.30837297,0.20563039,-0.45784846,0.7992813,-0.58580524,0.8425376,0.11244689,0.720191,-1.2291318,-1.6669607,-0.4446297,2.2034054,1.0664696,0.81338793,-0.45597473,-1.262895,-0.22252297,-1.3892789,-0.7460112,0.09792392,-0.9173114,-1.1128796,0.4023646,-0.7147693,-0.7736084,0.39257377,2.3405874,-0.27565384,0.40534794,-0.8130043,-0.097918786,0.61959296,-1.450402,-1.5624176,-1.1680084,1.2857002,-1.2578357,0.41585746,1.0423982,-0.53722286,-0.38153228,-0.519493,-1.5859611,1.3256052,-0.09761098,0.5610493,2.6576302,1.2429906,-0.7566362,0.9100397,0.73991334,-0.10965712,-2.0155513,-0.009600936,-0.68587065,-1.7821826,-1.0589895,-0.13436031,0.42209822,-1.1379088,0.4846043,-0.61625975,-0.3349506,0.5386397,0.18685767,0.2667846,-0.20175363,-0.72045606,0.21726064,1.1120164,0.79319555,-0.36998132,-0.67134815,0.72676235,-0.18856223,0.4676755,-1.068567,0.32332298,-0.06708162,-1.2248513,-0.78454924,0.5460869,-0.6168076,1.5106833,1.7673894,-0.6841837,0.44412613,0.79369456,-0.8888351,0.039817635,1.6578221
+7,0,2,-0.087797664,-0.72064584,0.39195046,0.3027774,-1.0405561,-0.877984,-0.79764,0.07123449,-0.29967046,0.27734974,-0.10566526,0.7710878,-1.7671878,0.82492906,1.5373951,1.5010266,1.4846263,-0.78963494,-0.8870087,0.6919234,0.6787521,-0.43306896,0.68626046,-0.29046977,-0.96195453,-0.36133754,-0.9279964,0.1886242,-0.13149588,0.33348688,0.3524693,-0.8627565,0.7647197,-1.5118191,-0.24293655,-0.6279771,1.9434764,-0.15230925,1.2640053,-0.5305171,-1.5717331,-0.23882796,1.2569784,-1.515566,0.31609678,-1.4623228,-1.5276749,-0.6301773,-0.15530233,-1.2590178,-1.11366,-0.8283696,-0.3316043,1.6386755,0.71456283,1.4172398,-0.9240341,1.1554213,-0.20543054,-2.2690706,0.03277201,-0.60741836,-0.84084237,-0.43909258,1.1674479,-1.5343083,0.39456686,-0.63195604,0.69497895,-0.17052427,0.66388005,1.6770406,-1.6856107,0.8501753,-1.0100727,-0.069561884,-0.49753237,0.4581591,-0.8491859,-1.5388991,-0.21994227,-0.6159067,-0.4368516,-0.54839456,1.8440411,-1.9056602,0.23943676,-0.8901972,1.0178717,-0.41829002,0.47147444,0.28248173,-0.4255799,-1.3693498,1.3788788,-0.832486,0.2925802,0.038576376,-0.9931526,-0.8897898,-1.3541447,-0.38844064,-0.5666752,2.0802116,0.20410384,-0.56880385,-0.25536695,-2.847082,0.22910748,1.3398811,0.4140531,0.37236091,1.0213392,-0.6121137,0.20765178,0.22666255,-0.052260183,0.2003284,-2.3570495,1.430299,1.474341,-0.28836238,0.44256905,-0.7141607,-0.87645155,-0.20459776,0.5313733,-1.1576004,0.28822646,-0.3185678,-2.0415008,0.5557355,-1.3157727,0.27283114,0.18704557,-0.345221,-0.70967424,-0.43224388,-1.6015024,1.0732054,0.91898453,-0.6834103,-1.5872614,-0.59699357,0.9072513,1.1939424,0.62792987,-0.49085766,0.9603636,0.668255,0.6667969,-1.698892,0.26019427,0.57732755,0.53784436,1.2200284,0.3280351,-0.15348202,2.1410975,-0.0141839655,-1.0378197,-2.6077065,-0.30667314,-0.067144245,-1.3892245,-1.7354017,0.5355618,0.9041617,-0.32855982,0.12410726,-0.40722403,-1.3432459,-0.7320008,-0.25400352,0.17272069,1.1034757,-2.285054,0.6256184,1.7550694,-1.3452939,-0.017986381,-0.5405325,0.05288163,0.28705782,-0.98898715,0.76313794,0.15638609,-0.6830085,-1.3387135,0.62448525,-0.7181843,-1.9862365,0.070949845,1.8409837,-1.5815736,1.6396283,-0.85405725,0.19265085,0.4421183,0.6777789
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
index 902c22b2..3df20b08 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,3,1.6896031,0.2908513,-0.23292185,1.3335884,0.4545921,0.34372616,-0.4165884,-0.42207703,0.60733074,1.2122438,0.8359352,0.5889715,-1.7851549,1.4177665,0.7491191,0.5095494,0.26941696,0.21018945,-0.880237,0.5611264,0.49656537,-0.8069379,1.1553068,-0.0126349665,-0.10190317,-1.7098439,-0.7785349,0.08457157,-1.0872965,-0.46575686,0.17051879,-1.0981392,1.2801948,-0.20283222,-0.37779662,-1.045028,0.8350199,-0.4449086,0.37728134,-1.3267673,-1.3647467,0.49804264,1.072318,-1.5584799,0.62113297,-1.7105299,0.936378,0.96202284,-0.5543043,-0.22153261,0.5113075,-0.5170268,0.11910198,-0.16186874,0.33424348,-0.8031384,1.6559254,0.47857162,0.5240287,-0.85450006,1.1488028,-0.12382944,-2.3143659,-1.7480177,0.34907168,0.3368916,-0.96712863,-1.1980493,1.7140709,-1.4368651,0.66559404,-0.27778557,-1.7404891,1.0585526,-0.3457661,-0.026605887,-1.9635391,2.3253458,0.7187862,-0.6399079,-2.111302,-0.7639529,-0.3719801,-0.0044646794,1.4715577,-1.4121687,0.046070594,0.049432546,-0.67208856,0.77220047,-0.78486633,1.6651013,-0.5048683,-1.0169152,0.19552465,0.6060275,1.0623702,0.15507454,-0.3246709,0.2723275,-0.9548158,-0.060560215,1.6082474,1.8419513,-0.6165263,0.5530547,0.53137124,-1.6408414,-0.083764516,-0.6223381,0.053751897,0.17250697,-0.5837834,-1.9890918,-2.1878116,-0.63507724,-1.3977062,-0.09614903,-0.99611247,-0.8769,0.8914697,1.3250514,-1.3865021,0.070221275,0.14449106,-0.4635786,0.22937016,-1.06633,-1.124088,0.17105861,0.38474694,1.002527,0.062628046,-1.4390727,-0.07946947,2.1103234,-1.7293836,1.1202102,-0.1935348,-1.0497704,-0.1828823,-0.8191538,-0.9199905,-0.0011521083,1.4552091,0.028564112,1.5781827,1.5662959,-0.37651274,-0.75987285,-0.96915853,-1.9078714,1.1331102,-0.114896916,0.53727424,2.130492,-2.0829363,0.2657546,0.7349715,-0.30037367,0.7455469,-1.2468303,-0.94141734,-0.96966463,-2.640029,-1.1223298,0.8763038,-0.96782756,-0.92439544,1.2501682,-1.4089605,-0.8066179,-0.87894547,0.5093231,0.8661666,-1.0657727,-0.44514516,0.16784558,-0.82460415,-0.62026,0.22096872,-0.5744515,-0.05113996,0.9549697,-0.7317686,0.2157382,-0.73088646,1.3564851,-0.56419,-0.3550247,-1.0255395,-0.57220864,-0.9036402,0.36711925,-0.09569928,0.6653445,-0.5859867,1.2525154,1.7926836,0.031228034
+8,0,3,1.6896036,0.2908516,-0.23292227,1.333588,0.4545924,0.34372577,-0.41658887,-0.4220771,0.60733056,1.2122439,0.8359348,0.58897126,-1.7851542,1.4177668,0.74911904,0.5095494,0.26941735,0.21018945,-0.8802367,0.56112623,0.49656504,-0.8069377,1.1553066,-0.012634999,-0.1019033,-1.7098432,-0.7785347,0.08457122,-1.0872964,-0.46575698,0.17051879,-1.0981392,1.2801946,-0.20283177,-0.377797,-1.0450277,0.8350198,-0.44490838,0.37728155,-1.326767,-1.3647465,0.49804258,1.0723178,-1.5584799,0.62113315,-1.7105298,0.9363782,0.962023,-0.5543043,-0.22153258,0.51130754,-0.517027,0.119101755,-0.16186844,0.33424366,-0.80313826,1.655926,0.4785711,0.5240289,-0.8545002,1.1488025,-0.12382955,-2.314366,-1.7480171,0.34907153,0.3368917,-0.9671285,-1.1980494,1.7140709,-1.4368649,0.6655941,-0.27778602,-1.7404888,1.058552,-0.34576625,-0.026606066,-1.9635391,2.325346,0.718786,-0.63990754,-2.1113021,-0.7639529,-0.37198016,-0.00446413,1.4715579,-1.4121689,0.046070803,0.04943272,-0.67208874,0.77220094,-0.7848665,1.6651015,-0.50486845,-1.0169153,0.19552442,0.60602796,1.06237,0.15507467,-0.32467127,0.27232736,-0.9548156,-0.060560405,1.6082474,1.8419516,-0.6165264,0.5530548,0.531371,-1.6408411,-0.08376483,-0.6223387,0.053751606,0.17250703,-0.58378434,-1.9890922,-2.1878114,-0.63507706,-1.3977063,-0.09614923,-0.996112,-0.87689984,0.8914697,1.3250511,-1.3865025,0.07022147,0.14449127,-0.4635789,0.22937027,-1.0663304,-1.1240877,0.1710589,0.38474682,1.002527,0.06262817,-1.4390726,-0.07946953,2.1103234,-1.7293843,1.1202103,-0.19353478,-1.0497701,-0.18288228,-0.8191537,-0.91999054,-0.001152092,1.4552096,0.02856364,1.5781829,1.566296,-0.37651303,-0.75987285,-0.9691587,-1.9078715,1.1331105,-0.11489664,0.5372741,2.1304917,-2.0829363,0.2657544,0.73497146,-0.30037382,0.7455473,-1.2468303,-0.9414173,-0.96966416,-2.6400287,-1.1223292,0.87630415,-0.9678274,-0.9243957,1.250168,-1.4089605,-0.8066182,-0.8789456,0.509323,0.8661667,-1.0657727,-0.44514495,0.1678456,-0.8246042,-0.62025964,0.22096853,-0.5744515,-0.051140007,0.9549694,-0.7317687,0.21573786,-0.73088586,1.3564854,-0.5641902,-0.35502443,-1.0255394,-0.57220846,-0.90364045,0.36711854,-0.095699474,0.6653445,-0.5859865,1.2525156,1.7926835,0.031228162
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
index f9df3070..80e7457d 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
@@ -1,2 +1,2 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,2,-0.16720563,-0.95600957,0.28606522,0.6061068,-0.3845801,-0.20699826,-1.1377771,0.20950346,-0.13692231,-0.39809802,1.1865551,1.4303648,-1.0548229,0.47550696,0.9458502,2.1207294,2.1566582,-0.24435441,-0.64201504,1.7221576,0.93151706,-0.7352646,0.34909317,0.72475654,-0.53412133,-0.38585177,-0.7807762,0.55336505,0.0011219378,1.0769404,0.178836,-1.3198038,1.3214653,-1.3146976,-0.7111258,-0.48764378,1.6924084,-0.51583934,1.3672884,-0.5570951,-0.8911707,-0.7621178,1.6342849,-1.217147,0.21610765,-2.0004432,-0.9046609,-0.66080266,0.099175125,-1.5156087,-1.408444,-1.4207674,-1.6178051,1.1580858,0.932027,1.5196209,-1.0533147,1.2685812,0.0013257095,-1.983875,-0.02037545,-0.64693433,-1.3806919,-1.0592809,0.54943085,-1.3961786,-0.15649292,-0.5929407,0.9686789,-0.40543142,1.0557822,0.863148,-2.2666905,0.6786515,-0.98612607,-0.46339324,-0.5806146,1.359029,-0.97610635,-1.6902065,0.015555903,-1.0601696,-0.68368477,-0.4881073,1.0633672,-2.1536195,0.27614263,-0.94285184,0.8829792,-0.1332724,-0.69135016,0.057943374,0.29546717,-1.8476412,1.6050953,-0.77256185,-0.34552783,-0.39992267,-0.9135076,-1.4197912,-1.4342732,0.05444667,0.38957366,1.6343316,-0.018683603,0.20149979,-0.5184369,-2.7988343,-0.23812824,0.39474156,0.7759951,0.26147765,0.7695459,-0.8783734,0.24983662,0.22944757,-0.50273734,0.17921253,-2.96453,0.8493479,0.8776484,0.17855343,-0.3165259,-0.49537572,-1.4034132,-0.5197445,0.6516528,-1.1936586,0.4376411,-0.31979737,-1.1554341,-0.094374776,-1.4360906,-0.666431,0.19690529,0.22096215,-0.77558017,-0.6480959,-1.0999115,1.1237246,0.32265925,-0.7196827,-2.2545595,-1.4951689,0.5562803,1.1043608,1.1391095,0.12772338,0.39196157,0.68036157,-0.48576477,-1.285276,0.81914747,0.30907387,-0.07296551,1.534054,0.30991095,-0.007137307,1.7125996,-0.8783491,-0.07996723,-2.1059916,-0.3130489,-0.31512117,-1.8248315,-1.5771933,0.51070344,0.7182178,-0.2582489,-0.45623493,-0.2644527,-0.9527642,-0.7283463,0.17178611,0.21193407,0.9493434,-1.3720604,0.15322249,0.701293,-0.551722,-0.15245207,0.06241879,-0.6320952,0.5529451,-0.4287735,1.0122428,-0.07743517,-0.28508037,-1.175873,0.21220748,-1.3312136,-1.4357239,0.21157013,0.5917681,-1.1742185,1.9612663,-1.2883047,0.06013697,1.073086,0.5143654
+9,0,2,-0.1672056,-0.9560093,0.28606495,0.60610676,-0.3845802,-0.20699823,-1.1377773,0.2095035,-0.13692239,-0.3980985,1.1865553,1.430365,-1.054823,0.47550726,0.9458505,2.1207294,2.1566586,-0.24435459,-0.64201486,1.7221576,0.9315173,-0.73526466,0.3490929,0.724757,-0.53412163,-0.38585144,-0.78077614,0.5533652,0.0011221742,1.0769403,0.17883608,-1.3198036,1.3214654,-1.3146971,-0.7111261,-0.4876436,1.6924083,-0.51583886,1.3672888,-0.55709535,-0.89117086,-0.7621175,1.6342846,-1.2171476,0.2161075,-2.0004432,-0.90466124,-0.6608028,0.09917505,-1.515609,-1.4084437,-1.4207675,-1.6178051,1.1580858,0.932027,1.5196207,-1.0533146,1.2685808,0.001325828,-1.9838752,-0.020375686,-0.64693415,-1.3806916,-1.0592812,0.5494306,-1.396179,-0.15649273,-0.5929404,0.96867883,-0.40543112,1.0557821,0.8631483,-2.2666907,0.67865187,-0.9861259,-0.46339324,-0.5806145,1.3590288,-0.9761064,-1.6902065,0.015556082,-1.0601697,-0.6836848,-0.48810717,1.0633674,-2.1536198,0.27614254,-0.9428517,0.88297933,-0.13327223,-0.69135034,0.05794356,0.2954672,-1.8476416,1.6050957,-0.7725617,-0.34552756,-0.399923,-0.91350776,-1.4197913,-1.434273,0.05444653,0.38957354,1.6343312,-0.01868396,0.20149925,-0.51843625,-2.798834,-0.2381288,0.39474154,0.77599514,0.26147756,0.769546,-0.8783735,0.24983686,0.2294478,-0.5027371,0.1792123,-2.9645298,0.849348,0.8776484,0.17855346,-0.31652552,-0.49537584,-1.4034135,-0.5197447,0.6516527,-1.1936587,0.43764108,-0.31979695,-1.1554341,-0.09437473,-1.4360908,-0.666431,0.19690526,0.22096193,-0.7755801,-0.64809597,-1.0999116,1.1237248,0.32265925,-0.71968305,-2.2545598,-1.4951686,0.55628073,1.1043607,1.1391095,0.12772316,0.3919614,0.6803616,-0.48576495,-1.285276,0.8191475,0.3090739,-0.07296532,1.5340542,0.30991074,-0.0071372786,1.7126004,-0.8783499,-0.07996719,-2.1059916,-0.3130492,-0.31512108,-1.8248317,-1.577193,0.5107035,0.71821797,-0.25824943,-0.4562349,-0.26445264,-0.95276433,-0.7283461,0.17178622,0.211934,0.949343,-1.3720604,0.15322232,0.7012931,-0.5517222,-0.15245192,0.062418915,-0.6320953,0.5529452,-0.42877352,1.0122427,-0.0774351,-0.28508034,-1.1758733,0.21220773,-1.3312134,-1.4357238,0.21157022,0.591768,-1.174219,1.9612663,-1.2883047,0.06013686,1.073086,0.5143656
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
index b7a6d23e..8e069e93 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,1,0.25,-0.0029144287,-1.1171875,-0.29296875,1.234375,0.76953125,-0.5703125,1.0390625,0.96875,-0.13964844,0.67578125,2.8125,0.44726562,0.28320312,-0.90234375,-0.04638672,-0.67578125,0.5625,3.265625,1.6875,1.015625,-0.90625,0.640625,0.2578125,0.030883789,0.79296875,-0.56640625,-0.4375,-0.921875,-0.87890625,0.09814453,0.05859375,0.16503906,1.140625,-1.125,0.0069274902,0.2890625,0.23144531,0.19335938,0.017700195,-0.37890625,-0.42578125,-0.25585938,-1.0703125,0.119628906,0.7578125,0.14355469,-1.484375,0.13476562,0.1328125,0.5,-1.5390625,0.068847656,0.119628906,-1.3515625,0.05126953,-0.33789062,-0.44726562,1.2265625,-1.2734375,-0.068359375,-0.36132812,2.1875,-0.40820312,-0.6953125,-1.1640625,0.025756836,-1.3984375,-0.2578125,0.8203125,0.047607422,1.4140625,-0.53125,-0.78515625,1.09375,2.859375,-0.90234375,-0.13671875,-2.34375,0.025390625,0.49414062,-0.068359375,0.54296875,0.31054688,0.04711914,-0.296875,-1.15625,-1.3359375,0.390625,1.65625,1.421875,0.66796875,-1.625,1.640625,-0.013427734,-0.026367188,-0.32617188,0.5390625,-0.765625,-0.25195312,-0.16503906,0.24121094,1.0078125,-1.234375,-1.4609375,0.27148438,-0.640625,0.6953125,-0.45507812,-0.14746094,1.203125,0.5,0.14160156,-0.24804688,1.4375,-1.5390625,-1.0,-0.41992188,0.8203125,0.18652344,-0.21972656,0.39257812,0.47460938,-0.125,-1.484375,2.40625,-0.49804688,-1.2890625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40625,-1.6953125,0.546875,-1.3671875,-0.62109375,-0.08544922,-0.78515625,1.4140625,-1.640625,-0.875,2.03125,-0.671875,-0.375,-0.14160156,0.99609375,-0.76953125,0.38671875,1.1015625,-0.94140625,0.55078125,-0.875,-1.03125,-0.088378906,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.3984375,0.73046875,2.28125,0.068847656,0.079589844,-1.28125,-2.375,1.7734375,0.033935547,0.3828125,-0.79296875,-0.33398438,0.7421875,-0.94921875,-0.19433594,-0.90234375,-1.125,-0.6953125,0.64453125,-0.765625,-0.052490234,0.4140625,1.109375,-1.4453125,0.9375,-0.9609375,-1.2890625,1.796875,-0.9375,-1.4140625,-0.5390625,0.43554688,-0.58203125,-0.20507812,1.3828125,-0.4140625,0.30859375,0.33789062,1.2265625
-0,0,2,0.29882812,0.50390625,0.65625,-0.48828125,0.25976562,-0.027954102,0.671875,-0.6015625,0.59375,-0.66796875,0.68359375,0.09765625,1.609375,-0.73046875,0.0064086914,-0.31640625,-0.19628906,0.00049972534,2.625,0.16210938,-0.15332031,-1.3984375,0.66796875,0.05883789,-0.703125,-1.0078125,0.296875,-0.3125,-0.3828125,1.609375,0.64453125,-1.0703125,1.5078125,-1.40625,0.19335938,0.20996094,0.06640625,-1.046875,-0.84765625,0.8203125,-2.15625,0.29101562,-0.828125,-0.6484375,-0.27539062,-0.51953125,0.3125,-0.97265625,1.3359375,0.67578125,-0.78515625,0.34570312,-0.93359375,1.5390625,0.20117188,-0.30078125,0.13769531,-0.42382812,-1.140625,-0.83203125,1.3515625,0.5078125,-0.05419922,1.2578125,-0.18945312,0.4453125,-0.06201172,-1.8046875,-1.265625,1.546875,-0.66796875,0.5859375,0.64453125,0.061279297,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.015563965,-0.80859375,-0.6640625,0.5625,-0.1796875,-0.27734375,-0.609375,-1.234375,-1.9375,1.484375,-0.5078125,1.1953125,0.94921875,-0.96484375,0.328125,-1.1015625,-0.73828125,-1.921875,-0.75390625,-1.296875,0.042236328,-1.375,1.125,1.796875,-0.018066406,0.44335938,2.609375,-1.0390625,0.59375,0.28710938,-0.62109375,0.703125,1.1484375,0.64453125,-0.075683594,0.46484375,0.31054688,-0.21679688,0.24121094,-0.06591797,1.3046875,0.54296875,-0.61328125,1.1953125,-0.91015625,-1.1171875,1.1171875,-1.8125,-1.3984375,-0.19140625,-0.10058594,1.109375,-0.609375,-0.54296875,-1.359375,0.3828125,0.40234375,0.39453125,-0.40429688,0.21484375,-1.875,2.5,0.5703125,-0.67578125,2.328125,-0.25,-1.5234375,2.21875,-1.0859375,-1.34375,0.82421875,-0.9609375,0.203125,1.03125,-1.4140625,-0.32421875,-0.034423828,1.2578125,-0.82421875,1.09375,0.31835938,-0.39257812,-0.53515625,2.390625,1.78125,-0.953125,-0.7734375,-1.1484375,1.015625,1.1015625,-0.39453125,-0.30859375,-0.5,-1.7265625,-0.9140625,-0.890625,0.265625,0.98046875,-1.4375,0.50390625,0.26171875,-1.921875,-1.015625,1.171875,0.2734375,-1.5390625,-0.41796875,0.43164062,-0.20996094,-0.65234375,-0.05810547,-1.703125,1.78125,1.1796875,-0.53125,-1.0859375,0.90234375,1.796875,-0.3515625,1.3671875,1.0703125
-0,0,3,0.66796875,1.21875,-1.203125,-1.3125,-0.515625,-1.90625,0.31640625,1.1875,-0.52734375,-0.67578125,-0.053466797,-1.203125,-1.828125,-0.33203125,-0.9296875,1.0625,-0.27734375,0.91796875,0.1640625,0.020629883,0.90234375,-0.234375,-0.03466797,1.3359375,-1.71875,-0.96484375,-2.375,0.6640625,1.9921875,0.9765625,0.76953125,1.3984375,-1.5234375,-0.38671875,0.734375,-0.017211914,0.9375,1.3671875,-0.049804688,0.62109375,-2.15625,-0.27539062,0.072753906,1.7578125,0.71484375,2.015625,0.43164062,0.59765625,0.1875,1.25,0.28320312,-0.22949219,-1.3515625,-0.796875,-0.13183594,-0.43359375,-0.98046875,0.012390137,-0.06347656,-0.61328125,-0.1640625,0.53515625,0.34960938,-0.38085938,1.328125,0.23535156,0.16503906,0.06347656,-1.0703125,0.59765625,0.049316406,-0.171875,0.390625,-0.8359375,0.6796875,0.08544922,2.09375,-0.043945312,-0.42773438,0.03857422,1.2734375,0.609375,1.5703125,0.9140625,-0.08544922,0.98828125,-1.328125,0.515625,-1.7421875,0.3515625,0.95703125,2.21875,1.796875,-0.41015625,-0.068847656,-1.40625,-0.328125,0.013549805,0.96875,1.0390625,0.875,-0.18457031,1.03125,-0.17382812,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0077819824,-0.37890625,-0.22558594,2.546875,-0.8828125,-0.4765625,-0.88671875,-0.35351562,-1.0390625,0.98828125,0.68359375,2.046875,-1.875,-0.86328125,-2.953125,-0.21582031,0.18847656,-0.98046875,1.25,0.17382812,-1.9765625,0.90234375,-0.37695312,0.67578125,-1.0703125,-1.8828125,0.58984375,0.53515625,-0.55078125,0.111816406,-2.1875,0.32421875,0.89453125,1.7109375,1.2109375,-0.6171875,-0.49804688,-1.625,1.0703125,-0.37695312,0.13085938,-0.041992188,-1.3515625,-0.7890625,1.4375,0.19433594,-0.7109375,-0.6015625,1.0625,-0.37695312,1.90625,-2.0625,0.9609375,-0.32617188,0.078125,-1.3046875,-0.15039062,-0.26953125,0.29296875,0.46679688,-0.5859375,-0.32421875,0.46875,-1.1015625,0.984375,0.953125,0.25,0.99609375,-1.546875,0.041992188,-0.1015625,1.0,0.53515625,0.62109375,-0.8828125,-0.50390625,-0.32617188,-0.24804688,0.064453125,-0.5390625,-1.0546875,-1.234375,0.31640625,-0.037841797,-0.87890625,1.296875,0.1171875,1.8984375,-0.2265625,0.09667969,0.42578125
+0,0,1,0.24902344,-0.004058838,-1.1171875,-0.29101562,1.21875,0.76953125,-0.56640625,1.03125,0.9609375,-0.14453125,0.66796875,2.828125,0.4453125,0.28320312,-0.91015625,-0.04663086,-0.67578125,0.5625,3.265625,1.6875,1.015625,-0.8984375,0.63671875,0.25585938,0.028808594,0.7890625,-0.56640625,-0.4453125,-0.91796875,-0.875,0.095703125,0.056152344,0.16601562,1.1328125,-1.125,0.00289917,0.2890625,0.2265625,0.19140625,0.016479492,-0.37890625,-0.42578125,-0.25195312,-1.078125,0.11669922,0.76171875,0.14355469,-1.484375,0.13183594,0.13085938,0.49804688,-1.546875,0.0703125,0.119140625,-1.3515625,0.05053711,-0.33789062,-0.44726562,1.234375,-1.2578125,-0.06982422,-0.359375,2.21875,-0.40820312,-0.6953125,-1.15625,0.02746582,-1.4140625,-0.25976562,0.8125,0.044433594,1.40625,-0.53125,-0.77734375,1.0859375,2.875,-0.91015625,-0.13769531,-2.328125,0.02319336,0.4921875,-0.06640625,0.546875,0.3125,0.045654297,-0.296875,-1.15625,-1.34375,0.38671875,1.65625,1.421875,0.671875,-1.625,1.640625,-0.013549805,-0.024658203,-0.32617188,0.546875,-0.76953125,-0.25195312,-0.1640625,0.24316406,1.0078125,-1.2421875,-1.4765625,0.26953125,-0.63671875,0.69140625,-0.45507812,-0.14648438,1.203125,0.49804688,0.13867188,-0.24902344,1.4296875,-1.53125,-1.0,-0.4140625,0.81640625,0.18652344,-0.21289062,0.39257812,0.4765625,-0.12597656,-1.4765625,2.421875,-0.49804688,-1.265625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40625,-1.6875,0.546875,-1.3671875,-0.609375,-0.08544922,-0.78515625,1.4140625,-1.640625,-0.87890625,2.03125,-0.671875,-0.37695312,-0.14160156,1.0,-0.76953125,0.39257812,1.09375,-0.9375,0.54296875,-0.875,-1.0390625,-0.08886719,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.375,0.73046875,2.28125,0.064941406,0.080078125,-1.28125,-2.359375,1.7734375,0.03564453,0.37890625,-0.7890625,-0.3359375,0.7421875,-0.95703125,-0.19726562,-0.90234375,-1.109375,-0.69140625,0.6484375,-0.76953125,-0.053222656,0.41210938,1.109375,-1.4453125,0.9453125,-0.9609375,-1.2890625,1.8046875,-0.93359375,-1.4296875,-0.53515625,0.43359375,-0.578125,-0.20507812,1.375,-0.421875,0.31054688,0.33398438,1.21875
+0,0,2,0.29882812,0.50390625,0.66015625,-0.47851562,0.26171875,-0.026367188,0.6796875,-0.6171875,0.58984375,-0.66796875,0.68359375,0.095703125,1.6015625,-0.73046875,0.00793457,-0.31054688,-0.19433594,-0.004638672,2.625,0.16601562,-0.15136719,-1.3984375,0.6640625,0.05517578,-0.703125,-1.0,0.29492188,-0.3125,-0.37890625,1.609375,0.64453125,-1.0625,1.515625,-1.4140625,0.19726562,0.20996094,0.07080078,-1.046875,-0.84765625,0.82421875,-2.140625,0.29101562,-0.82421875,-0.64453125,-0.27734375,-0.51953125,0.31445312,-0.984375,1.3515625,0.671875,-0.78515625,0.33984375,-0.93359375,1.5390625,0.20019531,-0.30273438,0.140625,-0.421875,-1.1484375,-0.828125,1.3359375,0.50390625,-0.051757812,1.25,-0.1953125,0.44921875,-0.05859375,-1.796875,-1.265625,1.5390625,-0.6640625,0.578125,0.640625,0.06738281,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.012817383,-0.80078125,-0.66796875,0.5625,-0.1796875,-0.27734375,-0.609375,-1.2421875,-1.953125,1.484375,-0.50390625,1.1953125,0.953125,-0.9609375,0.33007812,-1.109375,-0.74609375,-1.8984375,-0.75,-1.296875,0.04248047,-1.375,1.125,1.796875,-0.018920898,0.44335938,2.609375,-1.03125,0.5859375,0.28515625,-0.625,0.69921875,1.15625,0.640625,-0.07714844,0.46484375,0.30859375,-0.21582031,0.24707031,-0.06347656,1.296875,0.5390625,-0.6171875,1.1875,-0.9140625,-1.1171875,1.1171875,-1.8203125,-1.40625,-0.18652344,-0.10205078,1.109375,-0.609375,-0.54296875,-1.3515625,0.38476562,0.40234375,0.40039062,-0.40625,0.21582031,-1.8828125,2.484375,0.57421875,-0.671875,2.34375,-0.25,-1.5234375,2.21875,-1.0703125,-1.3359375,0.82421875,-0.9609375,0.20019531,1.0390625,-1.4140625,-0.32226562,-0.03173828,1.2578125,-0.82421875,1.0703125,0.32421875,-0.38867188,-0.53515625,2.390625,1.7734375,-0.94921875,-0.7734375,-1.15625,1.015625,1.1015625,-0.39453125,-0.30664062,-0.50390625,-1.734375,-0.9140625,-0.88671875,0.265625,0.98046875,-1.4375,0.51171875,0.2578125,-1.9296875,-1.015625,1.171875,0.26953125,-1.5390625,-0.41796875,0.43164062,-0.20898438,-0.65625,-0.05859375,-1.7109375,1.7890625,1.1875,-0.53125,-1.0859375,0.8984375,1.796875,-0.34765625,1.359375,1.0703125
+0,0,3,0.6640625,1.21875,-1.203125,-1.3125,-0.515625,-1.8984375,0.31445312,1.1875,-0.53125,-0.68359375,-0.05419922,-1.1796875,-1.828125,-0.33203125,-0.9375,1.0703125,-0.27929688,0.91796875,0.1640625,0.022338867,0.90234375,-0.23535156,-0.03564453,1.328125,-1.7265625,-0.96484375,-2.390625,0.6640625,1.984375,0.9765625,0.76953125,1.3984375,-1.515625,-0.3828125,0.73828125,-0.01928711,0.9453125,1.359375,-0.044433594,0.62109375,-2.15625,-0.27734375,0.072265625,1.7421875,0.71484375,2.03125,0.43359375,0.58984375,0.1875,1.25,0.28320312,-0.23144531,-1.3515625,-0.80078125,-0.1328125,-0.43359375,-0.9765625,0.013122559,-0.06591797,-0.60546875,-0.1640625,0.53515625,0.34765625,-0.38085938,1.328125,0.23535156,0.16503906,0.0625,-1.0625,0.58984375,0.049560547,-0.16796875,0.390625,-0.828125,0.67578125,0.08935547,2.09375,-0.044677734,-0.42578125,0.03930664,1.2734375,0.609375,1.578125,0.9140625,-0.08691406,0.9765625,-1.328125,0.51953125,-1.75,0.35351562,0.9609375,2.203125,1.7890625,-0.40820312,-0.06640625,-1.421875,-0.33203125,0.014221191,0.97265625,1.03125,0.87890625,-0.18457031,1.0234375,-0.17382812,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0061035156,-0.37890625,-0.22851562,2.546875,-0.8828125,-0.47851562,-0.88671875,-0.35546875,-1.03125,0.9921875,0.6875,2.0625,-1.8828125,-0.86328125,-2.953125,-0.21777344,0.19042969,-0.98046875,1.2421875,0.17285156,-1.9765625,0.90234375,-0.37890625,0.67578125,-1.078125,-1.875,0.58984375,0.53515625,-0.546875,0.11035156,-2.1875,0.32226562,0.890625,1.71875,1.21875,-0.62109375,-0.49414062,-1.625,1.078125,-0.37695312,0.12988281,-0.041259766,-1.359375,-0.7890625,1.4296875,0.19433594,-0.71484375,-0.6015625,1.0546875,-0.37695312,1.90625,-2.0625,0.96875,-0.33007812,0.083496094,-1.296875,-0.14648438,-0.26757812,0.29492188,0.47070312,-0.5859375,-0.32421875,0.47265625,-1.109375,0.9921875,0.953125,0.24804688,0.9921875,-1.546875,0.04248047,-0.103027344,1.0,0.52734375,0.62109375,-0.88671875,-0.50390625,-0.32421875,-0.24609375,0.0625,-0.54296875,-1.046875,-1.2421875,0.31445312,-0.036132812,-0.86328125,1.296875,0.119140625,1.921875,-0.23046875,0.09277344,0.42382812
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
index 9633bd19..4c736c3e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,1,1.0546875,0.89453125,-0.55078125,0.83203125,-0.65234375,-0.37890625,0.77734375,1.6484375,0.5625,-1.6171875,-0.53515625,0.54296875,0.76953125,-0.1015625,-1.4453125,0.828125,-0.05859375,0.58203125,0.34179688,-0.37304688,0.83984375,-0.9375,-0.032226562,-0.9140625,0.15820312,-0.953125,-1.5078125,-0.14160156,-1.1484375,-0.87890625,1.3203125,-1.3359375,1.71875,0.76171875,0.22167969,-1.578125,-0.122558594,-1.7265625,-0.7265625,-1.125,-1.2421875,-1.296875,-1.28125,0.017822266,-1.8828125,0.9296875,0.48046875,1.0078125,1.125,1.546875,-1.3046875,-0.13378906,-0.73046875,1.7421875,-0.671875,0.5078125,1.15625,0.66796875,-0.15234375,-0.23046875,0.05883789,0.51171875,0.08251953,-0.12207031,0.72265625,-0.5859375,-0.90625,-0.40429688,-0.7109375,0.25195312,1.5859375,-1.8125,0.57421875,-1.0390625,2.296875,2.609375,1.5390625,0.82421875,-0.34179688,-1.1328125,-0.38085938,-1.25,-0.42578125,-1.2734375,0.9296875,-1.1875,1.28125,0.59765625,-0.3046875,1.4921875,2.25,-1.3359375,0.16992188,0.734375,-0.45703125,0.57421875,0.19042969,-0.67578125,-0.21484375,0.5859375,-1.5703125,1.15625,1.03125,-0.02734375,1.9921875,-0.55859375,0.40429688,0.57421875,0.87890625,1.59375,-0.31835938,0.0119018555,0.296875,-0.61328125,1.0,-0.2890625,-1.6953125,0.16992188,-0.15234375,-0.28125,-0.66796875,-0.40039062,0.22363281,-0.026977539,-0.57421875,0.5859375,0.5703125,-0.59375,-0.48828125,-0.515625,-0.38476562,0.2109375,-0.25195312,-1.28125,-1.7578125,-0.46484375,-0.609375,0.37695312,-0.49609375,-2.328125,2.15625,-0.66796875,-0.5234375,-1.9453125,-0.30859375,-0.19140625,-0.7265625,-1.09375,0.61328125,3.640625,-0.44921875,1.0625,-1.7890625,0.828125,0.19335938,-0.9921875,0.14355469,1.2109375,0.29492188,-0.8515625,0.2578125,0.36523438,-0.029418945,0.12792969,-0.83984375,0.53125,0.15625,0.8359375,1.9765625,-2.328125,-0.33203125,0.29882812,-0.41601562,-0.25195312,0.27539062,1.1328125,1.0078125,1.203125,-0.42773438,0.38476562,-0.83203125,1.4375,0.875,0.13183594,0.32617188,-1.203125,1.4296875,0.91015625,-1.328125,-0.1953125,-0.41015625,-0.8203125,0.8203125,0.29101562,-1.6875,1.109375,0.049072266,-0.037353516,-0.34960938,-0.3828125
-1,0,2,-0.5546875,1.265625,-1.3671875,-1.5078125,0.30664062,-1.9453125,0.34375,1.6328125,1.609375,-0.58984375,0.21875,1.03125,-0.9453125,0.515625,-1.21875,0.53515625,0.2734375,0.21972656,-0.37109375,-0.07373047,0.33984375,-0.39453125,1.3359375,-0.0390625,-0.7421875,-0.921875,-2.421875,1.1015625,1.578125,-0.4765625,1.3671875,0.21191406,-1.1171875,0.14941406,1.5703125,-1.0859375,0.6640625,0.86328125,-0.7734375,1.046875,-1.546875,0.43164062,-0.22070312,-0.26953125,-0.32421875,2.625,-0.06201172,1.8125,1.140625,1.609375,-1.03125,1.0859375,-0.640625,-0.09033203,-0.82421875,-0.08984375,-0.51953125,0.48828125,-1.109375,1.40625,0.3984375,0.94140625,-0.16015625,0.609375,0.54296875,-1.203125,-0.12695312,0.71875,-1.65625,1.265625,0.52734375,0.7890625,2.046875,-0.53515625,0.55859375,2.09375,0.51171875,0.703125,-0.8359375,-0.40039062,0.48242188,-0.36523438,0.76171875,-0.13378906,-0.119628906,0.94140625,-1.421875,-0.27148438,-1.1640625,-0.6796875,0.5859375,1.53125,0.98046875,-0.25976562,0.7890625,-1.875,-0.53515625,-0.26171875,0.40625,-0.55859375,0.14160156,0.49414062,0.6953125,0.15527344,-0.76953125,0.068847656,-1.046875,-0.40039062,0.041259766,0.56640625,-1.6484375,-0.068359375,1.6640625,-2.3125,0.6484375,-0.14941406,-1.5625,-0.3125,0.91796875,0.08203125,1.171875,-2.34375,0.026367188,-1.359375,0.37890625,0.96484375,-0.26171875,-1.359375,1.0390625,-0.24902344,-1.328125,-0.91796875,-0.5390625,0.578125,-1.8984375,1.5390625,1.1875,-0.1953125,0.91015625,-2.03125,1.296875,0.123535156,0.6484375,0.8359375,-1.2421875,-1.3359375,-0.81640625,-1.3203125,-1.453125,-0.703125,0.60546875,-0.61328125,-0.48242188,0.875,-0.73828125,1.828125,0.5625,0.39453125,0.45898438,0.29101562,-0.27148438,1.1171875,0.73046875,0.36914062,0.026733398,-0.72265625,-0.38671875,0.7578125,-0.49414062,-1.9921875,-1.03125,-0.05883789,-1.578125,-0.6953125,0.8515625,-0.13183594,1.0703125,0.84375,-0.31445312,0.021362305,0.08154297,-1.1171875,0.30664062,0.16894531,0.33203125,0.06640625,-0.82421875,-0.8515625,-1.7734375,-1.0,0.2890625,0.81640625,-0.7265625,0.68359375,-0.81640625,2.640625,2.03125,0.09814453,-0.8984375,-0.48242188
-1,0,3,1.2578125,0.5703125,-0.9921875,0.103515625,-1.359375,-0.11767578,0.609375,1.0078125,-1.1640625,-1.4453125,-0.21582031,-0.37890625,-0.42382812,0.453125,-1.7421875,-0.8203125,-1.0625,0.28320312,-0.953125,-0.53125,0.59375,-0.21972656,-0.57421875,-0.46679688,-0.107421875,-2.171875,-2.390625,-0.4375,-0.008850098,0.19433594,-0.38085938,-1.4453125,0.2890625,0.87890625,-0.328125,-0.546875,-0.09423828,-2.390625,0.45507812,-0.08203125,0.40820312,-0.20996094,-0.859375,1.2421875,0.9375,0.3515625,-1.0234375,0.60546875,-0.025146484,1.1640625,0.16015625,-0.19824219,-1.65625,0.70703125,-0.90625,-0.67578125,-0.63671875,1.25,-1.0625,-1.03125,-0.28125,0.765625,1.265625,0.7734375,1.921875,-0.34960938,-0.28125,-0.5078125,-1.9296875,0.5,1.21875,-0.36132812,1.6640625,-0.6953125,0.8828125,0.828125,2.53125,1.234375,0.609375,0.099609375,-0.41210938,-0.3828125,0.37109375,-1.3359375,-0.27734375,0.99609375,0.31835938,-0.60546875,0.24121094,0.984375,0.5390625,1.34375,0.21972656,0.58203125,-0.95703125,-1.2890625,-0.734375,-0.13476562,1.1484375,1.0859375,0.38085938,1.2578125,-0.296875,-1.2265625,1.2578125,-1.078125,-0.71484375,0.69140625,0.41015625,-0.33398438,-0.96875,1.015625,1.578125,0.26367188,0.119628906,0.46679688,-0.3046875,0.515625,0.122558594,1.921875,0.30664062,-2.109375,-1.1484375,-0.59765625,-0.056152344,-1.25,0.171875,-1.046875,0.001335144,-0.16796875,0.421875,-0.5703125,-0.18945312,-0.022338867,-1.046875,2.4375,-0.9140625,-1.6875,1.40625,-2.3125,0.64453125,-0.671875,-0.28320312,-0.65234375,0.8671875,0.30664062,0.375,0.20214844,2.203125,3.59375,0.41992188,0.26367188,-1.609375,1.015625,-0.0071411133,0.29492188,0.65625,1.1484375,-1.359375,0.62890625,0.97265625,1.75,0.8046875,-0.62109375,-0.51171875,-0.55078125,-0.3515625,1.6796875,0.5234375,-1.7578125,-0.62109375,-0.94140625,0.23242188,-0.55078125,0.703125,1.3984375,-0.3125,1.1171875,1.3125,-0.14257812,0.064941406,0.122558594,1.3671875,-0.64453125,0.49023438,-0.2890625,0.640625,1.4140625,-0.0033111572,-1.515625,0.21582031,-1.6015625,0.5390625,-1.7421875,-0.81640625,1.2109375,0.30273438,-1.09375,-1.234375,-0.36328125
+1,0,1,1.0546875,0.8984375,-0.5546875,0.82421875,-0.65625,-0.375,0.77734375,1.6484375,0.5625,-1.6171875,-0.53515625,0.5390625,0.76171875,-0.09814453,-1.4453125,0.82421875,-0.056884766,0.5859375,0.33984375,-0.37304688,0.8359375,-0.9375,-0.03515625,-0.91796875,0.16015625,-0.9453125,-1.515625,-0.14648438,-1.1484375,-0.8828125,1.328125,-1.328125,1.71875,0.7578125,0.22460938,-1.578125,-0.12792969,-1.7265625,-0.7265625,-1.125,-1.2421875,-1.3046875,-1.28125,0.018188477,-1.875,0.9296875,0.48046875,1.015625,1.125,1.546875,-1.3125,-0.12792969,-0.734375,1.7578125,-0.671875,0.5078125,1.15625,0.66796875,-0.15332031,-0.22460938,0.05493164,0.515625,0.084472656,-0.12011719,0.7265625,-0.5859375,-0.90234375,-0.40625,-0.703125,0.25,1.59375,-1.8125,0.5703125,-1.046875,2.296875,2.578125,1.5390625,0.82421875,-0.34375,-1.125,-0.38671875,-1.25,-0.42578125,-1.265625,0.9375,-1.1875,1.28125,0.59375,-0.30859375,1.4921875,2.234375,-1.34375,0.16894531,0.734375,-0.4609375,0.56640625,0.18457031,-0.67578125,-0.21484375,0.58984375,-1.578125,1.15625,1.03125,-0.028076172,2.0,-0.55859375,0.40234375,0.57421875,0.8828125,1.5859375,-0.31835938,0.015380859,0.296875,-0.61328125,1.0,-0.29101562,-1.6875,0.17382812,-0.15039062,-0.28710938,-0.6640625,-0.40820312,0.22265625,-0.02709961,-0.578125,0.58984375,0.57421875,-0.58984375,-0.49804688,-0.515625,-0.37890625,0.2109375,-0.2578125,-1.28125,-1.75,-0.46875,-0.60546875,0.37890625,-0.49414062,-2.328125,2.171875,-0.6640625,-0.51953125,-1.9375,-0.30664062,-0.19335938,-0.7265625,-1.0859375,0.61328125,3.640625,-0.44921875,1.0546875,-1.8046875,0.828125,0.19824219,-0.984375,0.14550781,1.2109375,0.296875,-0.859375,0.2578125,0.36914062,-0.02758789,0.12451172,-0.8359375,0.53125,0.14941406,0.83203125,1.96875,-2.34375,-0.33398438,0.296875,-0.4140625,-0.25390625,0.27734375,1.15625,1.0078125,1.203125,-0.4296875,0.38085938,-0.828125,1.4375,0.87109375,0.13378906,0.32226562,-1.203125,1.4375,0.91796875,-1.3203125,-0.19238281,-0.40429688,-0.828125,0.81640625,0.29101562,-1.6875,1.109375,0.05078125,-0.03540039,-0.35546875,-0.38085938
+1,0,2,-0.5546875,1.265625,-1.359375,-1.5078125,0.30859375,-1.9609375,0.34375,1.6328125,1.6015625,-0.58203125,0.21777344,1.03125,-0.953125,0.51953125,-1.2109375,0.53515625,0.26953125,0.22363281,-0.37109375,-0.071777344,0.33789062,-0.39257812,1.328125,-0.037109375,-0.73828125,-0.9296875,-2.421875,1.1015625,1.578125,-0.4765625,1.375,0.20800781,-1.125,0.1484375,1.5703125,-1.09375,0.6640625,0.8671875,-0.7734375,1.0546875,-1.5546875,0.43164062,-0.21679688,-0.265625,-0.3203125,2.609375,-0.059814453,1.8125,1.15625,1.609375,-1.0234375,1.0859375,-0.640625,-0.08886719,-0.828125,-0.088378906,-0.51953125,0.484375,-1.109375,1.3984375,0.39453125,0.9453125,-0.15527344,0.609375,0.546875,-1.1953125,-0.12451172,0.71875,-1.6484375,1.2578125,0.53125,0.78125,2.0625,-0.5390625,0.5625,2.078125,0.5078125,0.703125,-0.83203125,-0.40429688,0.48242188,-0.36328125,0.765625,-0.13183594,-0.119628906,0.9375,-1.421875,-0.26953125,-1.15625,-0.68359375,0.58203125,1.5390625,0.99609375,-0.26171875,0.78515625,-1.875,-0.5390625,-0.25976562,0.41015625,-0.5625,0.14746094,0.49609375,0.6953125,0.15820312,-0.76171875,0.06542969,-1.0546875,-0.40039062,0.044677734,0.5703125,-1.6484375,-0.06689453,1.6640625,-2.328125,0.64453125,-0.1484375,-1.578125,-0.31054688,0.921875,0.083496094,1.1796875,-2.359375,0.028686523,-1.3671875,0.37890625,0.97265625,-0.25976562,-1.359375,1.03125,-0.24902344,-1.3203125,-0.91015625,-0.5390625,0.5859375,-1.90625,1.5390625,1.1875,-0.19335938,0.91015625,-2.03125,1.3046875,0.12597656,0.6484375,0.828125,-1.234375,-1.3359375,-0.81640625,-1.3046875,-1.453125,-0.6953125,0.60546875,-0.61328125,-0.48632812,0.87109375,-0.73828125,1.8125,0.5625,0.39648438,0.45898438,0.29492188,-0.26953125,1.1171875,0.7265625,0.36523438,0.02722168,-0.7265625,-0.38671875,0.7578125,-0.49414062,-2.0,-1.03125,-0.059326172,-1.578125,-0.6953125,0.8515625,-0.125,1.0703125,0.83984375,-0.31445312,0.022705078,0.08300781,-1.1171875,0.30664062,0.16894531,0.33203125,0.06591797,-0.83203125,-0.84765625,-1.7734375,-0.99609375,0.29101562,0.8125,-0.7265625,0.6796875,-0.81640625,2.625,2.03125,0.10107422,-0.8984375,-0.48632812
+1,0,3,1.2578125,0.56640625,-0.9921875,0.10546875,-1.359375,-0.11669922,0.61328125,1.0078125,-1.15625,-1.4609375,-0.21484375,-0.38085938,-0.43164062,0.45507812,-1.75,-0.8203125,-1.0625,0.2890625,-0.953125,-0.53125,0.59765625,-0.22167969,-0.57421875,-0.46289062,-0.10205078,-2.1875,-2.390625,-0.43554688,-0.012329102,0.19042969,-0.3828125,-1.453125,0.28710938,0.87890625,-0.32617188,-0.546875,-0.09863281,-2.375,0.453125,-0.08154297,0.41210938,-0.21386719,-0.859375,1.2421875,0.94140625,0.34960938,-1.0234375,0.60546875,-0.019897461,1.171875,0.16308594,-0.19824219,-1.65625,0.703125,-0.90625,-0.671875,-0.640625,1.265625,-1.0625,-1.0234375,-0.28125,0.765625,1.2734375,0.7734375,1.921875,-0.34960938,-0.28125,-0.5078125,-1.9375,0.49804688,1.2109375,-0.36328125,1.671875,-0.69140625,0.890625,0.828125,2.515625,1.2265625,0.60546875,0.10253906,-0.4140625,-0.38867188,0.375,-1.3359375,-0.28125,1.0,0.32421875,-0.60546875,0.24023438,0.984375,0.5390625,1.34375,0.22167969,0.578125,-0.9609375,-1.28125,-0.7421875,-0.13378906,1.1484375,1.078125,0.38671875,1.2421875,-0.30078125,-1.2421875,1.2734375,-1.078125,-0.71484375,0.6875,0.41210938,-0.3359375,-0.9609375,1.0078125,1.578125,0.27148438,0.119628906,0.46875,-0.30664062,0.515625,0.12597656,1.921875,0.30664062,-2.09375,-1.1484375,-0.6015625,-0.055664062,-1.2578125,0.171875,-1.046875,0.002670288,-0.17089844,0.43359375,-0.5703125,-0.19433594,-0.022949219,-1.046875,2.453125,-0.91015625,-1.703125,1.3984375,-2.3125,0.6484375,-0.671875,-0.28320312,-0.65234375,0.875,0.31054688,0.37695312,0.19921875,2.1875,3.578125,0.421875,0.265625,-1.6171875,1.0234375,-0.008728027,0.29492188,0.66015625,1.140625,-1.3515625,0.625,0.9765625,1.734375,0.80859375,-0.62109375,-0.5078125,-0.546875,-0.34765625,1.6796875,0.51953125,-1.7578125,-0.61328125,-0.93359375,0.23632812,-0.55078125,0.703125,1.390625,-0.31445312,1.1171875,1.328125,-0.14257812,0.06738281,0.123535156,1.359375,-0.640625,0.48828125,-0.29101562,0.640625,1.4140625,-0.0029296875,-1.5234375,0.21484375,-1.6015625,0.54296875,-1.7265625,-0.80859375,1.2109375,0.30664062,-1.0859375,-1.2421875,-0.35742188
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
index 2451ec01..4202cade 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,0,1.203125,-0.09814453,-1.3359375,-0.6015625,0.96484375,-1.4765625,-0.8203125,3.140625,-0.5234375,-1.671875,0.023925781,1.5859375,-0.53125,-0.74609375,-0.57421875,1.7421875,-0.26953125,-1.3984375,0.059326172,0.48828125,-1.3671875,-0.6796875,-0.44921875,1.234375,-0.09033203,0.32617188,-2.5,-0.15136719,-0.8828125,-0.06982422,-0.022216797,0.78515625,0.69140625,0.7734375,-0.07470703,-1.0546875,1.2578125,-0.17285156,0.515625,0.6875,-0.0625,-0.008483887,0.11669922,-0.7890625,-0.11816406,0.65625,-0.46875,0.96875,-0.43945312,2.21875,-0.60546875,-2.21875,-1.3359375,-0.04736328,-0.15039062,-0.33203125,0.91796875,0.296875,0.40039062,-1.046875,-0.36523438,1.484375,-0.1328125,0.53515625,-0.265625,-0.03125,-0.002380371,1.4921875,0.16699219,-0.11767578,-0.4140625,-0.29296875,0.8984375,-0.671875,1.625,2.140625,0.625,0.2734375,-0.2421875,-0.75,1.671875,0.75,0.53515625,-0.62890625,-0.44140625,-1.1640625,1.328125,1.828125,0.49414062,1.65625,1.7421875,-0.7734375,-1.046875,1.3671875,-0.28710938,0.03100586,-1.7265625,-0.734375,-1.0859375,-0.21777344,-1.0625,-0.24414062,1.4375,0.94140625,-0.47265625,0.25195312,-1.640625,-0.62109375,-0.29101562,0.0234375,-1.140625,-1.9296875,1.4140625,-2.265625,0.84765625,0.27929688,-2.046875,1.2734375,-0.30078125,-1.21875,1.25,-0.4765625,-0.92578125,-0.36523438,0.4140625,2.75,-0.43164062,-0.40625,0.44335938,-1.8671875,-1.4765625,-1.640625,0.36914062,-1.1015625,-0.890625,-0.012817383,1.328125,0.34570312,0.43554688,-1.3515625,1.46875,-1.8203125,-0.62890625,-0.15527344,0.16113281,0.46875,-0.32421875,0.83203125,-0.73828125,0.83203125,-0.6171875,-0.018188477,0.8125,-0.4453125,0.045898438,-0.7578125,0.3203125,0.015563965,1.65625,1.015625,-1.1484375,1.875,1.2890625,0.8203125,-0.57421875,0.77734375,0.16894531,1.3125,1.2578125,0.76953125,-1.296875,-0.17089844,0.060546875,1.0859375,-1.28125,0.43359375,0.3515625,-0.69140625,-0.059326172,-1.171875,0.45898438,0.38476562,0.30273438,-1.09375,0.22363281,-0.24414062,-1.3984375,0.0134887695,-0.83984375,-1.3515625,0.20996094,1.421875,-0.11621094,-0.43164062,0.12207031,1.1953125,0.49609375,0.6015625,-1.0703125,0.4609375
-2,0,1,0.31445312,-0.33398438,-0.72265625,-0.4453125,0.40429688,-0.546875,-0.006439209,1.5859375,-0.13769531,-0.828125,-0.53515625,-0.375,1.2265625,0.49804688,-1.6484375,0.44726562,-0.5234375,-1.984375,0.42578125,-0.6875,0.5625,-0.32421875,-0.27734375,-0.06640625,-1.328125,-1.8203125,-1.640625,-0.08984375,-2.078125,0.49023438,-0.6875,-1.1171875,0.5625,0.07519531,0.546875,1.6796875,0.62109375,-1.453125,-1.8671875,1.1953125,-1.4765625,0.3515625,-0.54296875,-0.043701172,0.23144531,1.0625,0.5625,-0.546875,0.24023438,1.71875,-2.109375,-1.328125,-1.40625,0.73046875,-1.9921875,0.62109375,0.125,-0.66796875,-0.19140625,-2.140625,1.1796875,-0.16113281,0.42578125,-0.072265625,0.17675781,0.055908203,-1.0859375,0.18847656,-0.671875,2.0,0.90625,-1.6015625,-0.4609375,-0.49609375,0.044189453,1.921875,1.1171875,1.109375,-0.61328125,-0.8828125,1.5546875,0.31835938,0.27734375,-0.5078125,-0.67578125,-0.030639648,0.28710938,0.9609375,-0.66796875,0.67578125,2.125,-0.17285156,0.14941406,0.41796875,-1.390625,-0.19921875,-0.30273438,-0.53125,1.4140625,-0.328125,-2.25,0.14160156,2.25,1.15625,-0.46679688,-0.22851562,1.484375,-0.9296875,0.40039062,0.71484375,0.2734375,0.36328125,0.8515625,-1.4296875,0.30664062,-0.46875,0.18554688,-0.27148438,-0.65234375,-0.65625,1.328125,-0.55078125,0.4140625,-1.1875,-0.24609375,1.2734375,-0.44921875,-2.296875,-0.005645752,-0.041015625,0.671875,-0.061767578,0.19238281,-1.3046875,-1.34375,-0.19921875,-0.35351562,-0.47070312,-0.019042969,-1.8046875,3.0625,-0.76953125,0.39257812,-0.053955078,-0.06933594,1.828125,-1.234375,-0.01977539,-1.9375,1.921875,0.23632812,1.125,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.4375,1.6171875,1.4296875,0.671875,-0.71484375,1.3515625,1.375,0.49804688,-1.3359375,0.53515625,-1.1171875,-0.33984375,2.375,-0.25195312,0.2265625,-0.14550781,-0.66796875,-0.31835938,0.118652344,0.453125,-0.9375,-0.91796875,0.93359375,-0.58984375,-0.18164062,-0.10888672,1.265625,0.48828125,0.6640625,-0.30273438,1.3203125,0.82421875,-0.5390625,-1.0390625,-0.23242188,0.53125,2.15625,0.024658203,0.83203125,-0.15625,1.40625,-0.20996094,1.21875,0.27148438
-2,0,2,0.76953125,0.111328125,-0.36523438,-0.22363281,0.20703125,-0.7578125,-0.2734375,1.4140625,0.5546875,-1.09375,0.5859375,1.4140625,0.4375,0.54296875,-1.6015625,0.8046875,0.19335938,-1.2109375,-0.11230469,0.119628906,0.98828125,-0.5390625,-0.39257812,0.578125,-0.29296875,-0.73046875,-0.51171875,0.06201172,-1.0078125,-0.059814453,-0.68359375,-0.51171875,1.78125,-0.45117188,1.296875,1.59375,0.83203125,-0.16015625,-2.21875,1.7265625,-1.8359375,0.375,-0.59765625,-0.05078125,0.6328125,0.81640625,-0.33984375,-0.1328125,-0.4453125,1.90625,-1.375,-0.11328125,-0.6875,-0.08935547,-0.796875,0.71484375,0.62109375,0.60546875,-0.07861328,-1.2421875,1.609375,-0.21484375,1.8671875,-0.34375,-1.7421875,0.8515625,-1.171875,0.53515625,-0.53125,1.265625,-0.10449219,-1.1640625,-0.09814453,0.6171875,-0.15527344,1.9296875,1.390625,0.38671875,-1.0,-1.46875,1.3828125,-0.20019531,0.59765625,-1.2890625,-1.3671875,-0.3359375,0.19433594,1.40625,-0.81640625,0.36132812,1.9609375,-0.03540039,-0.8203125,0.62109375,-1.546875,0.93359375,-0.17773438,0.890625,0.734375,0.9921875,-2.46875,0.013305664,2.0,1.03125,-1.3984375,0.35546875,0.91015625,-1.6640625,-1.0703125,0.18652344,0.578125,-0.36132812,1.546875,-1.5078125,-0.359375,-1.21875,0.95703125,-0.045410156,-0.3046875,-1.296875,2.125,0.44140625,-0.59765625,-1.2734375,0.60546875,0.41992188,-0.73046875,-2.1875,0.26367188,-0.36523438,0.18261719,-1.1640625,-0.54296875,-0.79296875,-1.84375,1.65625,0.30859375,-0.62109375,-0.53125,-1.34375,1.8203125,-1.6953125,-0.68359375,-0.3828125,-0.17871094,1.3125,-0.58203125,0.51953125,-2.6875,-0.1171875,0.25976562,1.125,0.69921875,-1.0390625,-0.22753906,-0.78515625,0.90234375,0.47851562,1.1171875,0.54296875,-0.41015625,0.3671875,-0.26171875,0.91796875,-1.5703125,1.28125,-1.109375,-0.58203125,1.3984375,0.08935547,-1.6796875,-0.79296875,-0.64453125,-0.34960938,0.16503906,0.36328125,0.80078125,-0.4609375,0.31054688,-0.828125,-0.49804688,-0.03466797,0.640625,-0.48828125,1.234375,-1.5546875,0.47070312,0.61328125,0.24902344,-0.85546875,1.09375,1.546875,0.86328125,-0.7421875,-0.7109375,-0.67578125,2.21875,1.6484375,1.1953125,-0.23925781
+2,0,0,1.2109375,-0.096191406,-1.3359375,-0.6015625,0.9609375,-1.484375,-0.8203125,3.140625,-0.5234375,-1.6640625,0.0234375,1.5859375,-0.53515625,-0.75,-0.5703125,1.75,-0.27148438,-1.40625,0.059570312,0.48632812,-1.375,-0.6796875,-0.44921875,1.2421875,-0.095703125,0.33007812,-2.484375,-0.15527344,-0.890625,-0.06933594,-0.020019531,0.7890625,0.69140625,0.7734375,-0.07324219,-1.0546875,1.25,-0.171875,0.515625,0.6953125,-0.06201172,-0.008239746,0.11279297,-0.78125,-0.115234375,0.65625,-0.46484375,0.96484375,-0.43945312,2.21875,-0.609375,-2.234375,-1.3359375,-0.048095703,-0.15039062,-0.33398438,0.921875,0.29882812,0.40234375,-1.046875,-0.3671875,1.4765625,-0.12988281,0.53515625,-0.26953125,-0.029663086,-0.000705719,1.4765625,0.16796875,-0.11328125,-0.4140625,-0.2890625,0.9140625,-0.67578125,1.6328125,2.140625,0.62890625,0.26953125,-0.24414062,-0.75390625,1.671875,0.7578125,0.53515625,-0.6328125,-0.44140625,-1.1484375,1.328125,1.84375,0.5078125,1.65625,1.734375,-0.7734375,-1.046875,1.375,-0.28320312,0.033691406,-1.7265625,-0.73828125,-1.0859375,-0.21875,-1.0703125,-0.24804688,1.4296875,0.94140625,-0.46875,0.25390625,-1.640625,-0.62109375,-0.28515625,0.016723633,-1.140625,-1.9453125,1.421875,-2.265625,0.84765625,0.27539062,-2.0625,1.2734375,-0.30273438,-1.21875,1.2578125,-0.47265625,-0.92578125,-0.36523438,0.4140625,2.71875,-0.43945312,-0.41210938,0.4453125,-1.8671875,-1.484375,-1.640625,0.3671875,-1.1015625,-0.8984375,-0.012634277,1.3203125,0.3515625,0.43554688,-1.3515625,1.453125,-1.8203125,-0.6328125,-0.15917969,0.16210938,0.47070312,-0.32226562,0.83984375,-0.73828125,0.828125,-0.6171875,-0.019165039,0.80859375,-0.453125,0.046142578,-0.76171875,0.31835938,0.014831543,1.6640625,1.015625,-1.1484375,1.859375,1.2890625,0.8203125,-0.57421875,0.77734375,0.17285156,1.3203125,1.2578125,0.7734375,-1.2890625,-0.16796875,0.06640625,1.09375,-1.28125,0.4296875,0.35351562,-0.6953125,-0.057128906,-1.171875,0.46679688,0.38476562,0.3046875,-1.0859375,0.22363281,-0.24023438,-1.40625,0.01373291,-0.84375,-1.34375,0.21386719,1.4375,-0.11621094,-0.42773438,0.123046875,1.1953125,0.49804688,0.6015625,-1.0703125,0.45703125
+2,0,1,0.31445312,-0.33398438,-0.72265625,-0.44140625,0.40625,-0.546875,-0.0046691895,1.578125,-0.13769531,-0.8359375,-0.5390625,-0.37695312,1.2265625,0.50390625,-1.6484375,0.44140625,-0.5234375,-1.984375,0.42578125,-0.6953125,0.5625,-0.328125,-0.27734375,-0.06689453,-1.328125,-1.8125,-1.625,-0.088378906,-2.0625,0.48828125,-0.6875,-1.1171875,0.55859375,0.076171875,0.546875,1.671875,0.625,-1.453125,-1.8671875,1.1875,-1.4765625,0.34960938,-0.546875,-0.040771484,0.23144531,1.0625,0.5546875,-0.55078125,0.2421875,1.734375,-2.109375,-1.3203125,-1.4140625,0.7265625,-1.96875,0.6171875,0.12695312,-0.671875,-0.19042969,-2.171875,1.1875,-0.16015625,0.42773438,-0.07324219,0.17675781,0.055419922,-1.0859375,0.19042969,-0.66796875,2.0,0.90625,-1.609375,-0.45898438,-0.5,0.04248047,1.921875,1.109375,1.1015625,-0.61328125,-0.8828125,1.5546875,0.32226562,0.27539062,-0.51171875,-0.671875,-0.028320312,0.28515625,0.9609375,-0.6640625,0.6796875,2.109375,-0.17285156,0.15039062,0.41796875,-1.3984375,-0.19824219,-0.30664062,-0.53515625,1.421875,-0.33203125,-2.25,0.13964844,2.21875,1.15625,-0.47070312,-0.23046875,1.484375,-0.9296875,0.40820312,0.7109375,0.27539062,0.35742188,0.859375,-1.4375,0.3046875,-0.47070312,0.18554688,-0.26953125,-0.65625,-0.65625,1.3359375,-0.55078125,0.4140625,-1.171875,-0.24511719,1.2734375,-0.4453125,-2.3125,-0.00579834,-0.04345703,0.6796875,-0.061035156,0.19140625,-1.3046875,-1.328125,-0.20019531,-0.35351562,-0.46875,-0.018432617,-1.8046875,3.09375,-0.7734375,0.38671875,-0.052246094,-0.07080078,1.8359375,-1.21875,-0.017944336,-1.9296875,1.921875,0.23535156,1.140625,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.43359375,1.625,1.4453125,0.66796875,-0.71875,1.34375,1.3671875,0.49609375,-1.3359375,0.53515625,-1.1171875,-0.33789062,2.375,-0.25390625,0.22949219,-0.14746094,-0.66796875,-0.31445312,0.11816406,0.453125,-0.9375,-0.91796875,0.9296875,-0.5859375,-0.17871094,-0.107421875,1.265625,0.48242188,0.6640625,-0.30078125,1.328125,0.8203125,-0.53515625,-1.046875,-0.23632812,0.53515625,2.15625,0.026489258,0.828125,-0.15429688,1.4140625,-0.2109375,1.2265625,0.26757812
+2,0,2,0.7734375,0.110839844,-0.37109375,-0.22460938,0.20996094,-0.76171875,-0.26953125,1.4140625,0.5546875,-1.09375,0.5859375,1.421875,0.44140625,0.55078125,-1.59375,0.796875,0.18847656,-1.203125,-0.11621094,0.11767578,0.9921875,-0.5390625,-0.39648438,0.578125,-0.29492188,-0.72265625,-0.51171875,0.060791016,-1.0078125,-0.061035156,-0.68359375,-0.51171875,1.796875,-0.453125,1.28125,1.59375,0.828125,-0.15917969,-2.21875,1.7265625,-1.8359375,0.375,-0.59375,-0.052001953,0.62890625,0.81640625,-0.3359375,-0.1328125,-0.4453125,1.90625,-1.3671875,-0.10986328,-0.69140625,-0.08984375,-0.796875,0.71484375,0.6171875,0.609375,-0.08105469,-1.25,1.609375,-0.21484375,1.8671875,-0.34765625,-1.7421875,0.8515625,-1.171875,0.53515625,-0.52734375,1.265625,-0.103027344,-1.171875,-0.09863281,0.609375,-0.15820312,1.9375,1.390625,0.38867188,-1.0,-1.46875,1.3828125,-0.20019531,0.59765625,-1.2890625,-1.375,-0.3359375,0.19042969,1.40625,-0.8125,0.359375,1.96875,-0.035888672,-0.8203125,0.6171875,-1.5390625,0.9453125,-0.17675781,0.8828125,0.7421875,0.9921875,-2.46875,0.012023926,2.0,1.03125,-1.3984375,0.35351562,0.90625,-1.6640625,-1.0703125,0.18164062,0.578125,-0.36523438,1.546875,-1.5078125,-0.35742188,-1.21875,0.953125,-0.044677734,-0.3046875,-1.2890625,2.140625,0.44140625,-0.59765625,-1.2578125,0.60546875,0.41992188,-0.73046875,-2.171875,0.26171875,-0.36523438,0.18066406,-1.171875,-0.54296875,-0.78515625,-1.8515625,1.6640625,0.31054688,-0.625,-0.5390625,-1.3515625,1.8125,-1.6953125,-0.68359375,-0.3828125,-0.17675781,1.3125,-0.5859375,0.51953125,-2.6875,-0.1171875,0.2578125,1.1328125,0.69921875,-1.03125,-0.23144531,-0.77734375,0.89453125,0.4765625,1.1171875,0.546875,-0.41015625,0.36523438,-0.265625,0.91796875,-1.5625,1.2734375,-1.1015625,-0.578125,1.3984375,0.08886719,-1.6796875,-0.7890625,-0.640625,-0.34960938,0.1640625,0.36523438,0.80078125,-0.46875,0.30859375,-0.83203125,-0.49804688,-0.03173828,0.640625,-0.49023438,1.234375,-1.5625,0.47070312,0.61328125,0.24804688,-0.86328125,1.09375,1.546875,0.859375,-0.7421875,-0.7109375,-0.66796875,2.21875,1.6484375,1.2109375,-0.2421875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
index d4d5cdcd..e1904552 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,0,1.0859375,-0.07324219,-1.484375,0.24511719,0.85546875,-1.703125,-0.57421875,2.953125,-0.0625,-0.76171875,0.23046875,2.65625,0.26757812,-0.36328125,-0.69921875,1.734375,0.2734375,-1.328125,0.359375,0.92578125,-0.24414062,-0.73828125,-0.65625,0.5390625,0.9375,1.3359375,-2.703125,0.38671875,-0.59375,-1.0234375,0.34375,-0.045166016,1.6953125,1.34375,0.54296875,-0.62890625,0.99609375,-0.021362305,-0.40820312,0.10644531,0.04736328,0.30859375,-0.17773438,-1.6484375,-0.13574219,0.8828125,-1.5625,0.44726562,0.07519531,0.79296875,-0.84765625,-1.421875,-1.171875,-1.046875,-0.19042969,-0.20800781,0.74609375,0.30078125,-0.24511719,-1.0625,0.21191406,0.71484375,1.3203125,0.8125,-1.265625,-0.32226562,0.30859375,0.27929688,-0.57421875,-0.40234375,0.41601562,-0.42382812,0.49609375,-0.47070312,2.28125,3.09375,1.046875,0.02355957,-0.81640625,-0.84765625,1.453125,0.6875,0.82421875,-1.6015625,0.04321289,-0.3984375,0.47070312,1.328125,0.578125,2.203125,1.875,-0.765625,-1.015625,1.359375,-0.61328125,-0.08691406,-1.3984375,0.0048828125,-0.14941406,0.6796875,-0.51171875,0.16015625,0.052734375,1.265625,-1.046875,-0.18261719,-1.125,0.028930664,-1.3828125,-0.35351562,-1.125,-1.53125,1.765625,-1.8984375,0.61328125,-0.03857422,-1.4609375,0.69140625,-0.0063476562,-0.39453125,1.2734375,0.11425781,-0.890625,-0.11328125,-0.04321289,2.421875,-0.75,-0.62109375,1.3828125,-1.5390625,-1.171875,-2.828125,0.103027344,-0.890625,-1.2109375,0.33984375,1.15625,-0.27539062,0.18066406,-1.171875,0.92578125,-1.2734375,-1.3984375,-0.54296875,-0.19824219,0.54296875,-0.82421875,0.765625,-0.34960938,0.026489258,0.41992188,-0.5390625,0.21484375,-1.2109375,0.28125,-0.12695312,0.96875,0.041503906,1.3359375,0.58984375,-0.18261719,2.46875,0.421875,0.94921875,-0.05883789,0.40039062,0.18945312,0.4765625,1.2734375,1.265625,-1.0234375,-0.640625,-0.88671875,0.91015625,-1.2890625,0.29492188,0.14355469,-0.30078125,0.23828125,-0.30078125,-0.5390625,-0.328125,0.49023438,-1.2109375,0.546875,-0.5234375,-1.1015625,-0.91015625,-0.515625,-0.97265625,0.57421875,0.96484375,-0.42773438,-0.8046875,-0.16699219,1.921875,-0.18457031,0.93359375,-1.453125,-0.171875
-3,0,1,-0.31054688,0.4609375,-1.6953125,-1.1796875,-0.61328125,0.16113281,-0.45703125,2.109375,0.5625,-0.8203125,1.734375,1.125,1.453125,-0.36328125,0.07763672,0.9609375,-0.07128906,-0.9375,0.76171875,-0.53515625,0.62109375,-0.012634277,0.8671875,0.96875,-0.055664062,0.13867188,0.2265625,-0.30859375,-1.2890625,-1.7265625,-0.31054688,1.6953125,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.2421875,0.53515625,0.75,-0.20214844,-0.45703125,0.16113281,-1.5,0.27539062,0.84375,0.23828125,0.55078125,0.3984375,-0.30273438,-2.21875,0.80078125,-0.9140625,0.35546875,-0.85546875,-1.015625,-0.24023438,-0.61328125,0.28125,-0.050048828,-0.12060547,0.86328125,0.69140625,1.5546875,1.40625,0.08886719,-1.28125,0.5390625,-0.765625,0.5390625,-0.40625,0.12109375,0.484375,0.09765625,-0.62109375,2.421875,-0.609375,0.017333984,-0.14648438,-0.03149414,-0.88671875,0.09033203,0.99609375,-0.8359375,0.88671875,0.76171875,-0.6171875,-0.4140625,1.0859375,-0.34570312,1.390625,-0.24902344,-0.875,0.8359375,-0.1171875,-1.125,0.6796875,-0.83203125,-1.3984375,-1.2109375,-1.4140625,0.6796875,3.15625,-0.81640625,-1.234375,-1.1640625,-0.6796875,0.30664062,-0.15429688,0.91796875,-0.546875,-0.47070312,0.8828125,-1.25,1.7734375,-0.30273438,-0.5546875,-0.36523438,0.011047363,-0.64453125,-0.0018310547,-0.828125,0.45507812,-0.44140625,-1.03125,1.28125,-0.88671875,-1.6640625,1.6015625,0.53125,-0.93359375,-0.78125,0.37304688,-0.984375,-1.734375,1.5,0.69140625,0.3359375,-0.43359375,-1.96875,2.484375,-1.375,-1.0078125,-0.061523438,-0.18261719,-0.26171875,0.32421875,-1.59375,-1.1953125,0.24121094,-0.068359375,1.59375,1.421875,-2.140625,0.106933594,-0.18359375,0.41210938,-1.0234375,1.6796875,-0.58203125,0.42382812,-0.27734375,0.90234375,-0.71484375,-0.013977051,2.234375,-0.953125,-0.110839844,1.5859375,-0.2578125,-1.6171875,-1.1015625,-2.0625,-1.5546875,-0.734375,0.47460938,-0.030151367,0.001625061,-1.328125,-1.2734375,-0.12890625,-0.48632812,0.7421875,-0.03100586,0.734375,-1.0546875,-0.45507812,0.15820312,-1.09375,1.0703125,1.734375,-0.21289062,0.6953125,-0.46679688,-0.95703125,0.75,1.90625,1.609375,0.66015625,1.421875
-3,0,2,0.578125,0.41992188,-0.9765625,-0.27929688,-1.546875,-1.1796875,0.515625,1.3671875,-0.12890625,-0.91015625,1.6640625,1.390625,0.76171875,0.19921875,0.42773438,0.95703125,-0.8359375,-0.28320312,1.0625,-0.7265625,0.80078125,-0.29101562,2.28125,2.0,-0.28125,0.5546875,-0.17382812,-1.1640625,-0.57421875,-0.32617188,0.35546875,0.6484375,1.078125,2.125,-0.0016326904,-0.47460938,-0.20605469,0.5,0.60546875,-1.2734375,-1.0546875,-1.84375,1.0078125,-0.4765625,0.66796875,0.19921875,0.3515625,-0.19433594,-2.375,-0.13085938,-0.34765625,0.14355469,-1.6640625,0.95703125,1.625,-0.106933594,-0.53515625,-0.020629883,-0.53125,0.026367188,0.2421875,0.25976562,-1.40625,-1.1796875,1.1484375,1.53125,-1.078125,-0.23925781,1.0234375,0.092285156,-1.4296875,0.88671875,-0.88671875,0.49609375,0.9453125,0.65625,1.0703125,0.61328125,-0.67578125,0.39257812,-0.76953125,0.546875,1.7734375,0.037597656,1.0390625,0.057128906,-0.39453125,1.265625,0.06640625,0.828125,1.2265625,-0.90625,-1.3046875,0.52734375,-0.42578125,-0.75390625,1.5546875,-0.31640625,-0.5078125,-0.5078125,-0.921875,-1.0234375,2.40625,-0.26367188,-1.9453125,-1.6953125,0.7109375,-1.5234375,0.86328125,0.64453125,-0.26171875,-1.6171875,0.921875,-1.234375,0.27539062,-0.53125,0.028320312,0.421875,0.3515625,-1.3515625,1.34375,0.4765625,-0.31445312,-2.140625,0.45507812,2.015625,-0.734375,-0.119628906,1.796875,-1.59375,0.453125,1.234375,1.2109375,-2.625,-2.296875,-0.28710938,-1.015625,0.5625,-0.030273438,-1.4765625,0.89453125,-0.09814453,-0.28710938,0.296875,-0.40234375,0.54296875,-0.9921875,-0.30273438,-0.8046875,0.640625,-0.43945312,0.29492188,-0.69140625,-0.15136719,-0.29101562,-2.109375,-0.78125,-0.55859375,0.53515625,1.4609375,-0.80859375,-1.3828125,0.078125,-0.8515625,0.90625,-0.040039062,-2.0,-1.515625,2.046875,-0.59765625,-0.30664062,0.5390625,-1.5234375,-0.47265625,-0.22363281,-0.09326172,1.5390625,0.22265625,-0.34960938,0.11425781,0.4765625,1.7578125,0.47851562,-0.25390625,0.63671875,-0.7109375,-0.37695312,0.31640625,0.52734375,1.546875,0.8359375,0.60546875,-0.21386719,-0.50390625,0.40820312,-1.53125,0.76953125,0.12988281,1.8046875,0.21191406
-4,0,0,2.046875,0.015991211,-1.2265625,1.3203125,1.1015625,-1.015625,-0.7109375,1.546875,-0.13476562,-0.93359375,1.03125,1.0546875,1.0390625,0.23828125,-1.75,0.83984375,-0.37109375,-0.0859375,0.375,0.013183594,0.875,0.49804688,0.625,0.36523438,0.40429688,0.08935547,-1.7578125,0.33007812,-1.6640625,0.45507812,0.16796875,0.79296875,1.34375,1.765625,0.7109375,0.41015625,0.24316406,-1.2421875,0.734375,0.91015625,-0.80078125,0.203125,-1.0546875,-0.734375,0.40429688,-0.106933594,1.2890625,-0.21386719,-0.81640625,-0.78125,-0.4453125,-1.1953125,-2.203125,-1.5390625,-1.6796875,-0.8203125,0.27929688,0.29296875,0.3828125,-0.6953125,0.09716797,0.23046875,0.049072266,0.76171875,-0.21875,-1.2578125,-0.049560547,0.20996094,0.13867188,0.92578125,1.1875,-2.8125,0.037597656,-0.033935547,0.62109375,2.046875,0.40039062,-0.6484375,0.1484375,-1.9375,-0.48046875,0.29492188,0.114746094,-1.640625,1.1328125,0.6484375,0.25195312,-0.3671875,1.2734375,-0.0703125,1.9609375,0.92578125,0.002456665,0.47460938,0.49804688,-1.546875,-0.19433594,-0.63671875,-0.67578125,-0.064941406,-0.90625,1.3125,0.94140625,-0.29296875,-0.6328125,-1.875,-1.2578125,-0.1796875,-0.12109375,1.3984375,-0.37695312,0.5078125,0.38476562,-0.3203125,1.0,1.296875,-0.78125,-0.859375,0.071777344,-0.265625,-0.12597656,-0.7265625,-1.0859375,0.8359375,-1.3828125,1.5234375,-0.8984375,-1.0859375,0.1484375,0.29101562,0.6171875,-0.7890625,-0.3984375,-0.62109375,-0.84375,0.6953125,0.13183594,-1.1640625,0.48632812,-2.28125,1.96875,-1.984375,-1.1015625,0.44726562,0.76953125,0.31054688,-1.3359375,0.671875,1.4296875,2.46875,-0.029663086,0.85546875,0.45703125,0.2890625,1.53125,1.0859375,-0.953125,0.27148438,0.78125,-0.98046875,-0.18945312,1.4921875,1.0,1.15625,-0.6640625,0.88671875,-0.26757812,-0.15527344,1.125,-1.1328125,-1.203125,-1.375,0.09814453,-1.828125,-1.6484375,-0.82421875,0.35546875,-1.8515625,0.5390625,-0.09033203,-0.49023438,-0.88671875,0.5234375,-1.9921875,1.5,0.39648438,0.58984375,-0.5,-0.052001953,-2.0625,1.1171875,-0.055664062,1.15625,-0.47265625,-0.171875,2.65625,0.6171875,0.014770508,0.3046875,1.015625
-4,0,1,0.38867188,1.015625,-0.84375,-1.046875,1.4921875,0.36523438,-1.2578125,1.28125,0.83203125,-0.13476562,0.16894531,0.7109375,-0.65234375,0.14550781,-0.390625,-0.63671875,-1.1953125,0.2109375,1.8671875,1.171875,0.6640625,-0.39257812,0.9375,0.6796875,-0.84375,-0.7734375,-2.078125,-1.0625,-2.015625,-0.031982422,-0.38671875,0.07421875,0.22460938,1.0546875,-0.57421875,0.09472656,0.38085938,-0.24316406,0.546875,0.86328125,-1.6796875,0.47265625,-0.044433594,0.08251953,0.26953125,0.37890625,0.7890625,-0.875,0.703125,0.65234375,0.984375,-1.2734375,0.107910156,0.28320312,-1.9921875,-0.6484375,-0.34570312,-1.0078125,0.62109375,-1.390625,0.2890625,0.76171875,2.03125,-0.73046875,0.47851562,-1.0546875,-0.05053711,-1.2734375,-0.16992188,1.484375,0.49804688,1.4140625,0.044189453,-1.046875,-0.04296875,0.83203125,-0.37695312,0.3046875,-0.71484375,0.064941406,0.51171875,-0.9140625,0.46484375,-0.32617188,-0.37890625,-0.26367188,-0.50390625,-1.3359375,1.5078125,0.38671875,1.3125,1.75,-1.0,1.453125,-0.53515625,-0.7421875,-0.4140625,-0.028686523,-0.98046875,-0.09326172,-0.62890625,1.234375,1.6640625,-1.609375,-1.1015625,0.5859375,-0.9296875,0.11279297,0.31054688,0.021728516,0.35351562,0.3984375,0.106933594,0.38867188,1.0,-0.53125,-0.5703125,-0.3203125,0.67578125,0.07324219,0.25195312,-0.22070312,-1.0,0.22167969,-1.15625,2.078125,-0.1875,-2.203125,1.359375,2.84375,0.62109375,-1.625,-2.03125,0.47851562,-1.6875,-0.69921875,-1.1640625,-1.4765625,0.41015625,-1.6875,2.140625,-2.296875,-0.98828125,1.875,0.796875,-0.75390625,0.7734375,0.22265625,-0.37695312,1.75,0.375,-0.22753906,1.1640625,-0.37695312,0.051513672,0.83203125,1.6875,0.35351562,0.7578125,-2.046875,0.875,1.203125,0.9765625,2.203125,-1.03125,1.03125,-1.6171875,-1.421875,2.03125,-0.81640625,0.39453125,-1.515625,0.11816406,-0.1875,-0.72265625,-1.0390625,-0.49023438,-1.3828125,0.22851562,0.0050354004,-0.8203125,-0.61328125,0.8359375,0.6015625,-0.35351562,1.2265625,-0.06347656,-0.56640625,1.0625,-0.9765625,-1.109375,0.034423828,0.45898438,-0.5859375,-1.03125,1.953125,0.79296875,0.66796875,-0.421875,0.6640625
-4,0,2,1.265625,-0.80859375,-0.62890625,0.19921875,-0.29101562,-0.51171875,-0.76953125,0.8828125,-0.671875,-0.6328125,0.32617188,1.1875,1.0546875,0.5234375,-0.66015625,1.515625,-0.31445312,-0.44921875,2.265625,-0.46289062,1.6953125,0.15039062,-0.20507812,0.05078125,2.1875,-0.5390625,-0.98046875,-1.3125,-0.5546875,0.49023438,-0.29882812,0.54296875,1.03125,-0.008605957,-1.984375,-0.13183594,0.5390625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.86328125,-1.0625,-0.23925781,-0.099609375,0.921875,0.19726562,-0.6796875,-1.2578125,-0.103515625,-0.44140625,-1.0078125,-1.59375,-0.59375,-0.50390625,1.9765625,0.48632812,0.014526367,0.23632812,-0.7109375,1.328125,-0.578125,0.5234375,1.5234375,0.58203125,0.18652344,-0.48632812,-0.4609375,2.265625,-0.023071289,0.17675781,0.029907227,-0.37890625,0.28515625,1.0859375,2.375,-0.25,0.083984375,-1.6484375,-0.34375,0.88671875,-0.51953125,-0.45703125,0.111328125,1.0,-0.15722656,0.31445312,1.578125,0.55859375,1.6015625,1.578125,-0.40429688,0.28710938,0.97265625,-1.7265625,1.5703125,-0.90625,-0.034179688,0.375,-0.22167969,-0.83984375,0.19335938,-0.22460938,0.28710938,-0.92578125,-0.29882812,-0.48828125,-2.6875,-0.64453125,0.328125,0.48046875,-0.953125,0.65234375,-0.6796875,-0.61328125,-0.100097656,-0.79296875,0.96875,-0.24511719,0.21386719,0.77734375,1.15625,-0.25195312,-1.7265625,-1.2109375,0.8984375,-1.421875,-0.94921875,0.022949219,0.23046875,0.97265625,-0.30273438,-0.2890625,-1.28125,-0.60546875,-0.20898438,1.8515625,-0.62890625,-1.6796875,-2.625,2.171875,0.64453125,-1.078125,-0.12597656,-0.83984375,-0.20507812,0.671875,-0.87109375,0.296875,1.9296875,-0.36523438,-0.111816406,1.3125,-2.09375,-0.30859375,-1.6015625,1.609375,1.265625,1.40625,-0.84765625,-0.48242188,-0.8046875,0.37890625,-0.28710938,-1.140625,0.18652344,0.096191406,-1.0859375,1.8359375,-0.390625,0.30859375,-0.28125,-1.2890625,0.4609375,-0.51171875,1.5859375,0.21289062,-0.12695312,0.96484375,0.09326172,-1.328125,0.22460938,-0.020507812,0.5,0.018676758,-0.546875,1.0,-0.44335938,1.5625,-0.89453125,-0.15917969,-0.53515625,1.25,-0.67578125,-0.81640625,0.48632812,0.7109375,2.765625,2.171875,0.546875
+3,0,0,1.09375,-0.07519531,-1.4921875,0.24316406,0.8515625,-1.703125,-0.57421875,2.953125,-0.059814453,-0.76171875,0.22753906,2.6875,0.265625,-0.36328125,-0.703125,1.734375,0.27929688,-1.328125,0.35742188,0.921875,-0.24414062,-0.734375,-0.65234375,0.55078125,0.94140625,1.328125,-2.703125,0.38085938,-0.60546875,-1.0234375,0.34570312,-0.042236328,1.6953125,1.3515625,0.55078125,-0.62890625,0.99609375,-0.015625,-0.40429688,0.107910156,0.044433594,0.30859375,-0.18066406,-1.65625,-0.13964844,0.87890625,-1.5703125,0.44726562,0.072265625,0.7890625,-0.8515625,-1.421875,-1.1796875,-1.046875,-0.18945312,-0.20703125,0.74609375,0.29882812,-0.24804688,-1.0703125,0.21289062,0.71484375,1.3125,0.80859375,-1.265625,-0.32421875,0.3125,0.28320312,-0.57421875,-0.40234375,0.4140625,-0.42578125,0.50390625,-0.46679688,2.265625,3.09375,1.0390625,0.024291992,-0.8125,-0.84375,1.4453125,0.68359375,0.82421875,-1.609375,0.03930664,-0.39648438,0.47265625,1.3125,0.578125,2.203125,1.8671875,-0.765625,-1.0078125,1.359375,-0.6171875,-0.091308594,-1.390625,0.00680542,-0.14648438,0.6796875,-0.51171875,0.16601562,0.055664062,1.265625,-1.0390625,-0.18261719,-1.109375,0.03173828,-1.3828125,-0.359375,-1.1328125,-1.53125,1.765625,-1.8828125,0.61328125,-0.036865234,-1.4609375,0.6953125,-0.0054016113,-0.39257812,1.2734375,0.11328125,-0.8828125,-0.109375,-0.045166016,2.421875,-0.7578125,-0.62890625,1.40625,-1.546875,-1.1640625,-2.8125,0.10205078,-0.89453125,-1.203125,0.33984375,1.1640625,-0.27929688,0.18359375,-1.1875,0.92578125,-1.2734375,-1.3984375,-0.55078125,-0.20117188,0.53515625,-0.8203125,0.765625,-0.34960938,0.02746582,0.41796875,-0.54296875,0.21386719,-1.203125,0.28125,-0.12988281,0.96875,0.044433594,1.328125,0.58984375,-0.18164062,2.46875,0.42578125,0.94921875,-0.064453125,0.3984375,0.18652344,0.48046875,1.2734375,1.2578125,-1.03125,-0.640625,-0.88671875,0.9140625,-1.2890625,0.29492188,0.14257812,-0.29882812,0.23535156,-0.3046875,-0.546875,-0.32421875,0.49023438,-1.203125,0.546875,-0.5234375,-1.1015625,-0.90625,-0.51953125,-0.9921875,0.5703125,0.96875,-0.4296875,-0.80859375,-0.16503906,1.9296875,-0.1875,0.93359375,-1.4609375,-0.16992188
+3,0,1,-0.31054688,0.45898438,-1.703125,-1.1796875,-0.609375,0.15917969,-0.45703125,2.109375,0.5625,-0.81640625,1.7421875,1.1171875,1.453125,-0.3671875,0.08105469,0.95703125,-0.072753906,-0.93359375,0.76171875,-0.52734375,0.625,-0.012329102,0.8671875,0.96484375,-0.055419922,0.13378906,0.22558594,-0.30664062,-1.2890625,-1.7265625,-0.3125,1.6796875,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.24414062,0.52734375,0.74609375,-0.203125,-0.453125,0.16503906,-1.5078125,0.2734375,0.83984375,0.23828125,0.5546875,0.39648438,-0.30273438,-2.21875,0.796875,-0.9140625,0.35546875,-0.8671875,-1.015625,-0.24023438,-0.6171875,0.28320312,-0.047607422,-0.12011719,0.87109375,0.6875,1.5390625,1.4140625,0.08886719,-1.28125,0.54296875,-0.765625,0.54296875,-0.40625,0.119628906,0.48632812,0.099609375,-0.62890625,2.421875,-0.609375,0.018310547,-0.14453125,-0.032226562,-0.88671875,0.092285156,1.0,-0.8359375,0.8828125,0.76171875,-0.61328125,-0.41210938,1.0859375,-0.34570312,1.390625,-0.24707031,-0.875,0.8359375,-0.114746094,-1.1171875,0.66796875,-0.8359375,-1.3984375,-1.21875,-1.4140625,0.6796875,3.140625,-0.8125,-1.2421875,-1.1796875,-0.6796875,0.3046875,-0.15136719,0.9140625,-0.546875,-0.47070312,0.87890625,-1.25,1.765625,-0.3046875,-0.5546875,-0.359375,0.015197754,-0.64453125,-0.0015182495,-0.82421875,0.45898438,-0.43945312,-1.0234375,1.28125,-0.88671875,-1.65625,1.59375,0.52734375,-0.9296875,-0.78125,0.37304688,-0.984375,-1.734375,1.4921875,0.69140625,0.33398438,-0.43359375,-1.96875,2.5,-1.375,-1.0078125,-0.064453125,-0.1796875,-0.26367188,0.328125,-1.59375,-1.1953125,0.2421875,-0.06982422,1.59375,1.4140625,-2.109375,0.10546875,-0.18457031,0.41210938,-1.03125,1.6875,-0.578125,0.42382812,-0.27539062,0.90234375,-0.71484375,-0.01361084,2.265625,-0.9453125,-0.110839844,1.59375,-0.25585938,-1.625,-1.1015625,-2.078125,-1.5546875,-0.734375,0.484375,-0.03149414,0.0038757324,-1.328125,-1.265625,-0.13671875,-0.48632812,0.7421875,-0.029785156,0.73828125,-1.0546875,-0.45703125,0.15429688,-1.1015625,1.0859375,1.734375,-0.21386719,0.69140625,-0.46484375,-0.94921875,0.75,1.90625,1.6015625,0.65625,1.4296875
+3,0,2,0.57421875,0.41992188,-0.9765625,-0.28125,-1.546875,-1.1796875,0.51171875,1.3671875,-0.12890625,-0.9140625,1.6796875,1.40625,0.7578125,0.19824219,0.4296875,0.94921875,-0.8359375,-0.28125,1.0625,-0.7265625,0.796875,-0.2890625,2.296875,1.9921875,-0.28125,0.5546875,-0.17382812,-1.1640625,-0.578125,-0.32226562,0.35546875,0.64453125,1.0703125,2.125,-0.0005455017,-0.4765625,-0.20605469,0.5,0.60546875,-1.2734375,-1.0703125,-1.84375,1.0,-0.4765625,0.66796875,0.20019531,0.3515625,-0.19335938,-2.375,-0.12988281,-0.34960938,0.14355469,-1.65625,0.95703125,1.6328125,-0.11035156,-0.53515625,-0.022827148,-0.53125,0.029052734,0.24609375,0.26171875,-1.40625,-1.1796875,1.140625,1.53125,-1.0703125,-0.23925781,1.0234375,0.08984375,-1.4375,0.89453125,-0.890625,0.5,0.9375,0.65625,1.0703125,0.609375,-0.671875,0.39257812,-0.77734375,0.54296875,1.7578125,0.029663086,1.0390625,0.059326172,-0.39257812,1.2578125,0.06640625,0.82421875,1.2265625,-0.90625,-1.3046875,0.52734375,-0.42578125,-0.7578125,1.5625,-0.31640625,-0.51171875,-0.51171875,-0.91796875,-1.015625,2.421875,-0.26171875,-1.953125,-1.6875,0.71484375,-1.515625,0.859375,0.640625,-0.26171875,-1.609375,0.92578125,-1.2421875,0.2734375,-0.53515625,0.028930664,0.42382812,0.3515625,-1.359375,1.359375,0.47265625,-0.3125,-2.140625,0.45117188,2.03125,-0.73828125,-0.119628906,1.796875,-1.59375,0.44921875,1.25,1.203125,-2.609375,-2.28125,-0.28515625,-1.0078125,0.5625,-0.028442383,-1.4765625,0.89453125,-0.1015625,-0.29101562,0.29296875,-0.40234375,0.54296875,-0.9921875,-0.30078125,-0.80078125,0.64453125,-0.43945312,0.29882812,-0.6953125,-0.15136719,-0.2890625,-2.109375,-0.7734375,-0.55859375,0.53515625,1.46875,-0.80859375,-1.375,0.075683594,-0.84765625,0.8984375,-0.041503906,-2.0,-1.5234375,2.046875,-0.60546875,-0.31054688,0.54296875,-1.5234375,-0.47265625,-0.22558594,-0.09326172,1.546875,0.22070312,-0.3515625,0.11376953,0.47851562,1.765625,0.48046875,-0.25585938,0.63671875,-0.70703125,-0.37695312,0.31445312,0.52734375,1.5390625,0.828125,0.6015625,-0.21679688,-0.5078125,0.4140625,-1.546875,0.765625,0.13378906,1.8046875,0.20996094
+4,0,0,2.046875,0.01586914,-1.2265625,1.3046875,1.1015625,-1.015625,-0.7109375,1.5546875,-0.13476562,-0.9296875,1.03125,1.0625,1.0546875,0.23828125,-1.7578125,0.84765625,-0.37304688,-0.08251953,0.37695312,0.010253906,0.88671875,0.49609375,0.6328125,0.36523438,0.40820312,0.08544922,-1.7734375,0.33007812,-1.6640625,0.45898438,0.16015625,0.79296875,1.34375,1.7578125,0.7109375,0.41015625,0.24414062,-1.2421875,0.73046875,0.9140625,-0.80078125,0.20703125,-1.0546875,-0.73046875,0.40429688,-0.106933594,1.296875,-0.20605469,-0.81640625,-0.78125,-0.4453125,-1.1953125,-2.203125,-1.546875,-1.6796875,-0.8203125,0.28125,0.29492188,0.38085938,-0.6953125,0.099121094,0.23046875,0.046142578,0.75390625,-0.21777344,-1.2578125,-0.05053711,0.20996094,0.13378906,0.921875,1.1953125,-2.828125,0.03955078,-0.037841797,0.6171875,2.03125,0.40039062,-0.65234375,0.14941406,-1.9375,-0.48046875,0.29882812,0.114746094,-1.6328125,1.125,0.64453125,0.25195312,-0.36914062,1.2734375,-0.07128906,1.953125,0.9296875,0.00340271,0.4765625,0.50390625,-1.546875,-0.18945312,-0.6328125,-0.67578125,-0.06640625,-0.8984375,1.3203125,0.9453125,-0.29296875,-0.62890625,-1.859375,-1.25,-0.1796875,-0.119140625,1.390625,-0.37695312,0.51171875,0.38867188,-0.3203125,1.0,1.296875,-0.77734375,-0.8671875,0.07861328,-0.26757812,-0.12597656,-0.72265625,-1.0859375,0.83984375,-1.3671875,1.515625,-0.90625,-1.09375,0.14941406,0.29101562,0.609375,-0.7890625,-0.40039062,-0.6171875,-0.83984375,0.703125,0.13183594,-1.171875,0.484375,-2.28125,1.9765625,-1.984375,-1.1015625,0.4453125,0.76953125,0.30859375,-1.328125,0.66796875,1.421875,2.46875,-0.028198242,0.84765625,0.45898438,0.29101562,1.5234375,1.0859375,-0.953125,0.26953125,0.78515625,-0.97265625,-0.18945312,1.5,1.0078125,1.15625,-0.67578125,0.88671875,-0.26367188,-0.15332031,1.1171875,-1.125,-1.2109375,-1.3828125,0.100097656,-1.828125,-1.6328125,-0.828125,0.35351562,-1.8515625,0.5390625,-0.09033203,-0.4921875,-0.8828125,0.52734375,-2.0,1.5,0.39648438,0.58984375,-0.50390625,-0.052246094,-2.09375,1.125,-0.057861328,1.15625,-0.47070312,-0.16796875,2.65625,0.609375,0.014770508,0.296875,1.0234375
+4,0,1,0.38671875,1.015625,-0.84375,-1.046875,1.4921875,0.36328125,-1.2421875,1.2734375,0.828125,-0.13476562,0.16992188,0.70703125,-0.64453125,0.14550781,-0.390625,-0.62890625,-1.2109375,0.21484375,1.8671875,1.171875,0.6640625,-0.39257812,0.94921875,0.6796875,-0.84765625,-0.78515625,-2.0625,-1.0625,-2.046875,-0.030639648,-0.390625,0.075683594,0.22460938,1.046875,-0.5703125,0.096191406,0.3828125,-0.23925781,0.54296875,0.8671875,-1.6640625,0.47460938,-0.041503906,0.08251953,0.26953125,0.38085938,0.78125,-0.87109375,0.70703125,0.65234375,0.98046875,-1.28125,0.10498047,0.28320312,-2.0,-0.6484375,-0.34179688,-1.0078125,0.625,-1.390625,0.2890625,0.76171875,2.0625,-0.72265625,0.47460938,-1.0625,-0.048583984,-1.265625,-0.171875,1.484375,0.50390625,1.390625,0.048339844,-1.0390625,-0.04321289,0.8359375,-0.37890625,0.30273438,-0.71875,0.064941406,0.515625,-0.921875,0.46289062,-0.32617188,-0.37890625,-0.26367188,-0.5078125,-1.34375,1.5078125,0.38671875,1.3125,1.7421875,-0.99609375,1.4453125,-0.53515625,-0.7421875,-0.41601562,-0.023925781,-0.9765625,-0.09423828,-0.62890625,1.2265625,1.6640625,-1.6015625,-1.1015625,0.5859375,-0.9296875,0.111328125,0.31445312,0.01928711,0.3515625,0.3984375,0.10888672,0.38671875,0.99609375,-0.52734375,-0.5703125,-0.31835938,0.671875,0.07519531,0.25195312,-0.22265625,-1.0,0.22558594,-1.1640625,2.09375,-0.18554688,-2.1875,1.359375,2.84375,0.6171875,-1.6171875,-2.03125,0.49023438,-1.6875,-0.6953125,-1.1640625,-1.46875,0.40625,-1.6875,2.140625,-2.296875,-0.9921875,1.8671875,0.796875,-0.7578125,0.78125,0.22460938,-0.36914062,1.75,0.375,-0.22363281,1.171875,-0.38085938,0.052978516,0.8359375,1.6875,0.35351562,0.7578125,-2.03125,0.875,1.1875,0.97265625,2.203125,-1.03125,1.0390625,-1.6171875,-1.4140625,2.03125,-0.81640625,0.390625,-1.515625,0.11816406,-0.18847656,-0.71875,-1.046875,-0.49023438,-1.375,0.23046875,0.004486084,-0.82421875,-0.609375,0.8359375,0.60546875,-0.35546875,1.21875,-0.064453125,-0.56640625,1.0703125,-0.9765625,-1.109375,0.036865234,0.46289062,-0.5859375,-1.03125,1.9453125,0.7890625,0.66796875,-0.421875,0.66015625
+4,0,2,1.2578125,-0.80078125,-0.62890625,0.20019531,-0.296875,-0.51953125,-0.765625,0.88671875,-0.68359375,-0.62890625,0.32421875,1.1953125,1.0546875,0.52734375,-0.6640625,1.5078125,-0.3125,-0.45117188,2.265625,-0.46289062,1.6953125,0.14453125,-0.20410156,0.04711914,2.1875,-0.53515625,-0.984375,-1.3125,-0.546875,0.49023438,-0.3046875,0.5390625,1.0390625,-0.012756348,-1.984375,-0.13378906,0.5390625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.859375,-1.0625,-0.24023438,-0.10058594,0.91796875,0.19628906,-0.68359375,-1.2578125,-0.10205078,-0.44335938,-1.0078125,-1.5859375,-0.59375,-0.50390625,1.9609375,0.49023438,0.018676758,0.23535156,-0.71875,1.3203125,-0.578125,0.52734375,1.53125,0.58203125,0.18652344,-0.48828125,-0.45898438,2.265625,-0.023071289,0.18457031,0.032470703,-0.38085938,0.28320312,1.0859375,2.375,-0.25390625,0.083984375,-1.6484375,-0.34375,0.88671875,-0.5234375,-0.45703125,0.111816406,0.99609375,-0.15722656,0.31640625,1.5859375,0.5625,1.6015625,1.5859375,-0.40429688,0.28515625,0.97265625,-1.7265625,1.5546875,-0.91015625,-0.028198242,0.375,-0.21972656,-0.83984375,0.19140625,-0.22167969,0.28710938,-0.91796875,-0.29296875,-0.4921875,-2.6875,-0.64453125,0.32617188,0.484375,-0.953125,0.6484375,-0.67578125,-0.61328125,-0.099609375,-0.79296875,0.96484375,-0.24609375,0.21582031,0.77734375,1.171875,-0.25390625,-1.703125,-1.2109375,0.8984375,-1.421875,-0.9453125,0.025268555,0.23046875,0.9765625,-0.30078125,-0.29296875,-1.296875,-0.61328125,-0.20507812,1.8671875,-0.62109375,-1.6875,-2.640625,2.171875,0.640625,-1.0703125,-0.12597656,-0.83984375,-0.20703125,0.6796875,-0.87109375,0.29882812,1.9140625,-0.36328125,-0.111328125,1.3203125,-2.09375,-0.30664062,-1.5859375,1.625,1.2734375,1.3984375,-0.83984375,-0.48242188,-0.80859375,0.36914062,-0.28320312,-1.1328125,0.1875,0.099121094,-1.0859375,1.828125,-0.390625,0.30859375,-0.28125,-1.2890625,0.45703125,-0.515625,1.6015625,0.2109375,-0.123046875,0.96875,0.09033203,-1.3359375,0.22558594,-0.020263672,0.5,0.018676758,-0.5546875,1.0,-0.44335938,1.5546875,-0.89453125,-0.16308594,-0.5390625,1.25,-0.671875,-0.8203125,0.48632812,0.7109375,2.78125,2.171875,0.546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
index 254cb8d4..25531275 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,0,0.58984375,0.99609375,-0.34570312,-1.3515625,-0.484375,-0.91015625,-0.88671875,0.4609375,0.94921875,-1.25,-0.1953125,-0.58203125,-1.09375,1.546875,-1.6640625,0.671875,-0.33007812,0.453125,-0.81640625,0.87109375,1.671875,0.70703125,0.06347656,0.34570312,0.41015625,-0.58984375,-1.8828125,-0.15136719,-1.0,1.2109375,-0.92578125,-0.22558594,0.83984375,0.20800781,-0.22753906,0.91015625,0.40039062,-0.060546875,-0.58984375,0.6171875,-1.8046875,0.38085938,-0.12109375,0.40625,-0.0045166016,1.34375,0.703125,0.13867188,-0.40039062,1.25,-0.41796875,-0.66796875,-1.515625,-0.021972656,-0.045654297,1.8203125,0.018310547,-0.23730469,0.40429688,-1.546875,0.29101562,0.39453125,1.2890625,0.2421875,0.9765625,1.1875,-1.5859375,1.4765625,0.013549805,-0.22753906,-0.80078125,-1.2109375,-0.765625,-0.07080078,-0.58203125,0.013427734,0.4921875,0.55859375,0.3125,0.64453125,1.078125,-1.046875,1.0,0.42382812,-0.53125,0.49804688,0.34375,0.62890625,-0.5859375,-0.50390625,3.578125,0.421875,0.79296875,1.2734375,-0.52734375,1.03125,-0.97265625,0.31640625,-0.578125,0.34179688,-0.82421875,0.17285156,1.375,-1.140625,0.04663086,0.17480469,-0.38671875,-1.859375,-0.4453125,0.828125,-0.15234375,-0.40429688,0.43554688,-0.8125,-0.6171875,-0.609375,-0.34570312,0.6953125,0.42773438,-0.125,0.73046875,-1.0703125,-0.62890625,-1.8359375,-0.82421875,1.6484375,-0.35742188,-1.6328125,-1.1015625,-0.96484375,1.3125,0.25585938,-2.375,-0.58984375,-1.5625,2.0,0.13769531,-0.45507812,-1.3515625,-2.5625,1.1015625,-2.25,0.5859375,-0.092285156,0.64453125,0.35351562,-0.20507812,-0.76171875,-0.087402344,1.5546875,-1.578125,0.59375,0.60546875,-0.10253906,1.2890625,-0.061279297,0.41601562,0.9375,0.796875,0.41992188,-0.87890625,-0.6640625,-0.37109375,0.09082031,-2.6875,2.40625,0.22851562,-0.012207031,1.421875,-0.91796875,-1.9140625,-0.8671875,-1.1171875,-0.92578125,-0.44921875,1.5078125,1.1875,-1.0078125,-0.18554688,-0.91015625,0.44921875,1.3203125,0.50390625,-0.041992188,1.0078125,-1.109375,-0.13476562,1.609375,0.58203125,-1.3125,-0.23535156,1.3671875,1.2265625,0.3828125,-0.55078125,0.44140625,3.265625,-0.100097656,-0.18652344,0.44726562
-5,0,1,-0.43945312,-0.69921875,-0.38476562,-1.4296875,0.18261719,-1.203125,-0.9453125,1.6484375,0.064453125,0.8125,-0.8671875,-0.53125,1.4140625,0.87890625,-2.140625,0.92578125,-0.515625,-0.86328125,0.10058594,0.5078125,2.453125,-0.18164062,-0.84765625,0.23925781,0.484375,-0.37890625,-2.0625,-1.1328125,-0.421875,1.7421875,-0.26367188,0.50390625,-0.98046875,0.546875,-1.2578125,0.49609375,-0.37304688,-0.49609375,-0.27734375,-0.6015625,-0.94921875,0.33203125,-0.6953125,-0.28515625,0.5390625,2.5,-0.10595703,-0.31640625,1.390625,1.0625,-1.0,-2.421875,-2.328125,0.4375,-0.609375,1.671875,-0.28125,-0.80859375,-0.26367188,-1.5546875,-0.87890625,-0.80859375,1.5859375,0.9609375,1.4296875,2.015625,-1.671875,1.6328125,-0.96875,0.62109375,1.296875,-1.296875,-0.69921875,-0.26367188,-0.23632812,0.24609375,0.26367188,-0.33789062,0.7265625,0.73046875,1.5703125,-0.43359375,0.27929688,1.75,-0.84375,0.64453125,-0.8203125,0.049316406,0.13769531,-0.11621094,1.4765625,-0.36132812,1.7734375,-0.103515625,-0.14453125,0.025878906,-1.84375,0.78125,-0.5390625,0.375,-0.28710938,-0.26953125,0.35742188,0.17578125,0.3671875,1.3125,-1.671875,-2.625,-1.0703125,-0.62109375,-0.19824219,0.7734375,-1.0546875,-0.53125,-0.100097656,-2.375,-0.09667969,0.81640625,0.33007812,1.5859375,0.11230469,-0.3828125,-0.15625,-1.65625,-1.265625,0.63671875,-1.359375,-0.77734375,-0.78515625,0.4765625,2.0625,-0.75390625,-1.71875,-0.5546875,-0.28515625,0.83984375,-0.13671875,0.079589844,-0.100097656,-2.046875,-0.328125,1.609375,0.58984375,0.95703125,0.52734375,0.15820312,-0.5234375,0.20507812,-0.59375,1.0390625,0.034423828,-0.90234375,0.54296875,-0.53515625,0.4375,-0.1796875,1.40625,1.0078125,0.55078125,0.359375,-0.19433594,0.875,0.41601562,0.49804688,-1.7890625,1.7265625,0.734375,0.20996094,0.45898438,0.037597656,-0.49609375,0.63671875,-0.039794922,0.18457031,-1.1171875,1.9921875,0.18261719,-1.6640625,-0.5703125,-0.09667969,-0.66015625,0.875,0.91796875,0.3828125,0.28515625,-0.28515625,0.27734375,0.3984375,0.32226562,-0.7890625,-1.4609375,0.7578125,1.5,0.15039062,-0.55078125,0.27734375,2.375,-0.39453125,1.015625,1.109375
-5,0,2,0.90234375,0.56640625,0.828125,0.19921875,-1.140625,-0.07861328,-1.1953125,-0.034179688,2.609375,-1.453125,1.8359375,-0.03930664,0.34960938,1.3671875,-0.625,-0.42578125,-0.46484375,0.18164062,2.265625,1.1484375,0.24023438,-0.51171875,-0.15625,0.578125,-0.48242188,-0.53515625,0.6484375,-0.52734375,0.90625,0.5078125,-0.40625,-0.008117676,0.78125,-0.6796875,0.38085938,0.91796875,0.953125,-0.19042969,-0.21191406,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24707031,-0.234375,0.50390625,-0.29101562,0.87890625,0.6796875,-0.16015625,-0.265625,-0.74609375,0.95703125,0.9375,0.41015625,-0.65234375,-0.49804688,-0.578125,0.7734375,1.1953125,-0.37890625,1.2578125,0.099609375,0.640625,0.51953125,-0.20703125,-1.2578125,-1.46875,0.41210938,-0.111328125,1.3828125,0.78515625,-0.640625,-0.65234375,0.47851562,-0.52734375,0.48046875,-1.09375,1.2578125,-1.8828125,-0.70703125,0.9375,-1.1640625,-0.6953125,0.5078125,-0.36132812,-1.515625,0.48242188,0.28710938,1.5390625,1.2109375,-0.46875,0.87109375,-0.453125,0.16308594,-1.7890625,0.73828125,-0.640625,-0.36914062,-1.6875,0.609375,0.9140625,-1.296875,-0.34960938,2.171875,-0.6328125,-0.7890625,-0.375,0.78515625,-0.47265625,0.25976562,1.2265625,-0.58203125,0.26757812,0.34960938,-0.625,0.453125,-0.41210938,0.55859375,0.6875,-0.0054626465,0.4453125,-1.6796875,-1.9140625,0.55859375,-0.16894531,-0.19921875,-0.34765625,1.140625,1.0703125,0.61328125,-1.5078125,-0.265625,-1.09375,1.578125,0.16503906,-0.734375,-0.41601562,-1.75,1.6875,-0.26953125,-0.62890625,1.5078125,-0.28710938,-0.075683594,1.4609375,-0.73046875,-1.1875,0.22167969,-1.0859375,-0.16503906,2.15625,-0.578125,-0.609375,0.19628906,1.40625,-0.39648438,0.65625,0.390625,-0.75,-0.3046875,1.390625,1.2421875,-1.171875,-0.14550781,-0.6640625,-0.6171875,0.6484375,-1.6953125,-2.109375,-0.25585938,-2.984375,-0.40234375,-0.578125,2.484375,0.80078125,-1.0546875,-0.4765625,0.36132812,-0.33007812,-0.092285156,-0.068359375,-0.65625,-1.15625,0.20800781,1.1171875,0.58984375,-0.38671875,-0.8203125,-1.5234375,0.296875,0.546875,-0.8671875,-2.8125,1.2578125,3.171875,0.20019531,0.34375,0.85546875
+5,0,0,0.58984375,0.99609375,-0.34179688,-1.34375,-0.48828125,-0.9140625,-0.88671875,0.46484375,0.94921875,-1.2421875,-0.19433594,-0.5859375,-1.09375,1.5390625,-1.671875,0.66796875,-0.33789062,0.453125,-0.8125,0.87890625,1.6640625,0.703125,0.063964844,0.34765625,0.40625,-0.58984375,-1.8828125,-0.15234375,-1.0,1.2265625,-0.9296875,-0.2265625,0.84375,0.2109375,-0.22949219,0.90625,0.3984375,-0.060791016,-0.58984375,0.62109375,-1.7890625,0.38085938,-0.12158203,0.41210938,-0.0056152344,1.3359375,0.69921875,0.13671875,-0.40234375,1.2578125,-0.41796875,-0.6796875,-1.5234375,-0.018310547,-0.041503906,1.8125,0.014404297,-0.23632812,0.40429688,-1.546875,0.29296875,0.39648438,1.2890625,0.24316406,0.96875,1.1796875,-1.578125,1.46875,0.012939453,-0.22949219,-0.80078125,-1.203125,-0.76171875,-0.07324219,-0.5859375,0.015197754,0.49609375,0.5546875,0.30664062,0.6484375,1.078125,-1.046875,1.0078125,0.42773438,-0.53125,0.49609375,0.3515625,0.640625,-0.58984375,-0.50390625,3.59375,0.41992188,0.796875,1.2734375,-0.51953125,1.0390625,-0.98828125,0.31835938,-0.58203125,0.34570312,-0.828125,0.17089844,1.3828125,-1.140625,0.045410156,0.17773438,-0.390625,-1.8671875,-0.4453125,0.83203125,-0.15625,-0.40625,0.4296875,-0.80859375,-0.625,-0.609375,-0.34960938,0.6953125,0.4296875,-0.12792969,0.734375,-1.0703125,-0.62890625,-1.8515625,-0.828125,1.65625,-0.35742188,-1.625,-1.1015625,-0.96875,1.3203125,0.2578125,-2.375,-0.5859375,-1.5703125,2.015625,0.14648438,-0.453125,-1.34375,-2.5625,1.1015625,-2.21875,0.5859375,-0.092285156,0.6484375,0.35742188,-0.20410156,-0.76953125,-0.087890625,1.5625,-1.5859375,0.59765625,0.60546875,-0.09814453,1.296875,-0.06542969,0.41015625,0.9375,0.7890625,0.41601562,-0.87890625,-0.66015625,-0.37109375,0.09082031,-2.6875,2.390625,0.22851562,-0.014282227,1.4140625,-0.9140625,-1.8984375,-0.86328125,-1.1171875,-0.9296875,-0.45117188,1.5078125,1.1875,-1.0,-0.18261719,-0.91015625,0.44921875,1.3125,0.50390625,-0.041015625,1.0078125,-1.109375,-0.13085938,1.6015625,0.578125,-1.3125,-0.23730469,1.359375,1.21875,0.38085938,-0.546875,0.43554688,3.265625,-0.09814453,-0.18554688,0.45117188
+5,0,1,-0.44140625,-0.69921875,-0.38476562,-1.4375,0.17578125,-1.203125,-0.9609375,1.640625,0.057373047,0.8203125,-0.8671875,-0.53125,1.3984375,0.875,-2.15625,0.93359375,-0.51953125,-0.85546875,0.1015625,0.5078125,2.46875,-0.18164062,-0.8515625,0.2421875,0.48046875,-0.38671875,-2.078125,-1.125,-0.42578125,1.75,-0.26757812,0.50390625,-0.98046875,0.5546875,-1.25,0.49804688,-0.37695312,-0.49414062,-0.27734375,-0.6015625,-0.9453125,0.33203125,-0.6953125,-0.28710938,0.53515625,2.484375,-0.106933594,-0.3125,1.390625,1.0703125,-1.0078125,-2.421875,-2.34375,0.43554688,-0.609375,1.6640625,-0.28125,-0.80859375,-0.26757812,-1.5546875,-0.87890625,-0.80859375,1.578125,0.96484375,1.4375,2.015625,-1.671875,1.6171875,-0.953125,0.6171875,1.296875,-1.2890625,-0.6953125,-0.265625,-0.234375,0.25195312,0.265625,-0.33984375,0.7265625,0.734375,1.578125,-0.42382812,0.27929688,1.734375,-0.84375,0.64453125,-0.8203125,0.049316406,0.13476562,-0.111816406,1.4765625,-0.36523438,1.78125,-0.09765625,-0.15332031,0.025878906,-1.84375,0.78125,-0.5390625,0.37695312,-0.2890625,-0.27148438,0.36328125,0.17675781,0.36523438,1.3125,-1.671875,-2.609375,-1.0703125,-0.61328125,-0.19921875,0.78125,-1.0546875,-0.53125,-0.099121094,-2.359375,-0.09667969,0.81640625,0.33007812,1.5859375,0.11376953,-0.3828125,-0.15917969,-1.6484375,-1.2578125,0.6328125,-1.359375,-0.77734375,-0.78515625,0.48046875,2.078125,-0.7578125,-1.71875,-0.546875,-0.28515625,0.8359375,-0.13867188,0.078125,-0.09765625,-2.046875,-0.32226562,1.609375,0.58984375,0.95703125,0.52734375,0.16015625,-0.51953125,0.20507812,-0.59375,1.0234375,0.034179688,-0.90625,0.54296875,-0.52734375,0.43554688,-0.17382812,1.4140625,1.0078125,0.55078125,0.35546875,-0.19433594,0.875,0.41601562,0.5,-1.7890625,1.734375,0.73828125,0.20800781,0.46289062,0.03955078,-0.49804688,0.63671875,-0.036865234,0.18261719,-1.1171875,1.984375,0.18164062,-1.671875,-0.57421875,-0.09863281,-0.6640625,0.875,0.91796875,0.3828125,0.28320312,-0.28710938,0.28710938,0.39453125,0.3203125,-0.79296875,-1.453125,0.7578125,1.5078125,0.14746094,-0.546875,0.2734375,2.375,-0.3984375,1.0078125,1.125
+5,0,2,0.91015625,0.5625,0.828125,0.203125,-1.140625,-0.080078125,-1.1953125,-0.03112793,2.609375,-1.46875,1.8359375,-0.040283203,0.35351562,1.3671875,-0.62890625,-0.42578125,-0.46289062,0.18359375,2.265625,1.1484375,0.2421875,-0.5078125,-0.15820312,0.58203125,-0.484375,-0.53515625,0.6484375,-0.52734375,0.90625,0.5078125,-0.40625,-0.008361816,0.78125,-0.66796875,0.37695312,0.90625,0.94921875,-0.19042969,-0.20898438,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24804688,-0.23339844,0.5078125,-0.29296875,0.87890625,0.68359375,-0.15917969,-0.26367188,-0.74609375,0.95703125,0.9296875,0.40820312,-0.65234375,-0.49804688,-0.578125,0.77734375,1.1875,-0.3828125,1.2578125,0.09765625,0.64453125,0.515625,-0.20703125,-1.2578125,-1.453125,0.41210938,-0.110839844,1.3828125,0.78125,-0.63671875,-0.65234375,0.47851562,-0.52734375,0.48046875,-1.09375,1.2578125,-1.8984375,-0.71484375,0.92578125,-1.1640625,-0.703125,0.51171875,-0.36523438,-1.5078125,0.47851562,0.28710938,1.5390625,1.1953125,-0.47070312,0.875,-0.453125,0.1640625,-1.78125,0.73046875,-0.640625,-0.36523438,-1.6953125,0.609375,0.9140625,-1.296875,-0.34570312,2.171875,-0.63671875,-0.7890625,-0.375,0.78125,-0.47460938,0.26171875,1.203125,-0.578125,0.265625,0.34960938,-0.625,0.453125,-0.41015625,0.5546875,0.6875,-0.0032958984,0.4453125,-1.6796875,-1.90625,0.5625,-0.16699219,-0.1953125,-0.34570312,1.140625,1.078125,0.61328125,-1.5078125,-0.265625,-1.1015625,1.578125,0.16601562,-0.73046875,-0.41601562,-1.75,1.671875,-0.26953125,-0.62890625,1.5,-0.28320312,-0.075683594,1.4609375,-0.73046875,-1.1796875,0.22070312,-1.078125,-0.16601562,2.140625,-0.578125,-0.60546875,0.19335938,1.40625,-0.39648438,0.65625,0.390625,-0.75,-0.30273438,1.390625,1.2421875,-1.171875,-0.14550781,-0.6640625,-0.625,0.6484375,-1.703125,-2.109375,-0.25585938,-3.0,-0.40234375,-0.578125,2.5,0.80078125,-1.046875,-0.47460938,0.36328125,-0.33398438,-0.09277344,-0.068847656,-0.6640625,-1.1640625,0.20605469,1.125,0.58984375,-0.38867188,-0.8203125,-1.5234375,0.296875,0.546875,-0.86328125,-2.796875,1.2578125,3.171875,0.20019531,0.34570312,0.859375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
index 76476f56..dc74307e 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
@@ -1,7 +1,7 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,1,0.13964844,0.10888672,0.27539062,-0.88671875,1.390625,0.61328125,0.48828125,1.0234375,-0.4921875,-1.046875,0.12597656,-0.26757812,0.57421875,0.35351562,-1.25,0.48632812,-0.546875,-1.9140625,1.1328125,-0.53125,-0.087402344,-0.6640625,1.28125,0.5234375,-1.234375,-1.5078125,-0.953125,-0.8671875,-1.7890625,0.44335938,-0.64453125,-0.26171875,0.010559082,-0.4609375,-0.5234375,0.953125,0.91796875,-0.73828125,-1.4609375,1.578125,-1.8046875,0.030517578,-0.07519531,0.09814453,0.38671875,0.5234375,0.63671875,-0.69921875,0.38085938,1.8515625,-1.1953125,-1.6015625,-0.98046875,1.1015625,-2.421875,0.091796875,-0.21289062,-0.80859375,1.328125,-1.6953125,1.734375,0.18359375,0.29296875,0.30273438,-0.5078125,-0.06640625,-1.0703125,0.16210938,0.546875,1.84375,0.05517578,-0.31835938,0.57421875,-0.13574219,-0.66796875,1.8515625,0.76953125,1.1796875,-1.7734375,-0.5234375,1.28125,0.89453125,-0.11816406,0.12207031,-0.63671875,-0.20117188,-0.46484375,1.4453125,-0.36132812,0.98828125,1.3828125,0.21386719,-0.921875,0.36523438,-1.4296875,0.22363281,-0.59375,-0.640625,0.625,-0.890625,-2.375,-0.18066406,2.515625,0.78125,-0.79296875,0.0051574707,0.063964844,-1.0703125,0.49609375,0.98828125,0.703125,0.036376953,0.56640625,-0.796875,0.49414062,-0.31640625,-0.47851562,-0.34765625,-0.484375,-1.1796875,0.71484375,-0.78515625,0.60546875,-1.5703125,-0.22460938,1.265625,-1.3203125,-2.40625,0.013122559,0.17382812,-0.18652344,0.044921875,-0.025512695,-1.3671875,-1.03125,-0.13671875,-0.65234375,0.25390625,0.20507812,-1.7265625,3.375,-0.40429688,1.3125,1.109375,-1.234375,1.0703125,-0.6015625,-0.671875,-2.515625,1.1015625,-0.30859375,-0.03564453,1.0625,-0.484375,-1.078125,-1.2109375,0.38085938,1.359375,1.015625,1.4375,-0.26367188,-0.011962891,1.8203125,1.2421875,-1.78125,1.6640625,-1.671875,-0.24511719,2.125,-0.20117188,0.36914062,-0.40234375,0.099609375,0.62109375,0.68359375,0.041992188,-0.86328125,-0.84765625,0.859375,-0.61328125,-0.3046875,-0.890625,0.13964844,0.86328125,0.1484375,0.54296875,0.59375,0.037841797,0.87109375,-1.8046875,0.09863281,1.375,1.2578125,0.28710938,0.43554688,-0.17871094,1.6875,0.15625,1.40625,0.51953125
-6,0,2,0.828125,0.83203125,-0.3984375,0.84375,1.40625,0.62890625,-0.3515625,0.39257812,-0.10253906,-1.4375,1.109375,-0.19335938,0.53125,-0.54296875,-0.99609375,-0.37109375,-0.81640625,-0.39648438,1.1171875,1.25,0.20410156,-1.8984375,0.36132812,0.35742188,-0.90234375,-1.609375,-1.3828125,-0.34960938,0.22753906,0.057861328,-0.23632812,-0.328125,1.71875,0.20703125,-0.33007812,-0.16992188,0.31445312,-0.6015625,-0.26367188,0.78125,-0.6015625,0.71875,-0.26757812,1.2578125,-0.27929688,0.25390625,-0.734375,-1.0234375,0.43554688,1.7109375,1.140625,-1.7734375,-0.49414062,-0.26757812,-1.0859375,-2.140625,0.3828125,1.3046875,-0.11279297,-1.015625,1.609375,1.59375,1.734375,-0.34570312,-1.0390625,-0.71875,0.05029297,-2.078125,-0.30078125,0.85546875,0.38476562,0.17382812,1.3515625,-1.4921875,0.45898438,0.51953125,0.18359375,-0.20214844,-1.71875,0.39648438,-0.2734375,0.71875,-0.016235352,0.32226562,-0.76171875,0.734375,-2.828125,0.45507812,1.578125,1.2109375,1.4296875,1.203125,1.125,0.22558594,-0.6328125,-0.9375,-0.87109375,1.25,-0.22753906,0.123535156,-0.54296875,-0.40039062,0.890625,-0.11279297,-1.0390625,0.78125,-0.98828125,-0.18164062,-0.34765625,-0.328125,0.18261719,1.96875,0.13378906,-0.63671875,-0.57421875,0.6796875,-0.4140625,0.33398438,-0.46484375,0.83984375,0.052490234,-1.09375,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.59765625,-0.24511719,0.765625,-0.04272461,-2.5,-0.93359375,0.73828125,-1.3125,2.546875,-0.31445312,0.013977051,1.8203125,-1.3203125,1.4296875,-0.75,0.12158203,1.171875,0.6640625,-0.20117188,-0.22558594,-0.28320312,-0.56640625,0.203125,0.4296875,-0.32226562,-1.34375,1.0234375,-1.140625,1.2265625,0.69921875,-0.36523438,-0.39257812,1.0546875,0.78515625,1.1484375,1.0078125,3.03125,-2.671875,1.6171875,-2.25,1.1796875,0.16796875,-0.29492188,-1.7734375,-1.84375,0.7265625,-0.6484375,0.65625,0.27734375,-0.66796875,-1.96875,1.40625,-0.35351562,0.09667969,0.15332031,0.83984375,-0.11230469,-1.015625,1.578125,-0.49609375,-0.26757812,0.5078125,-0.36914062,0.65625,1.546875,-0.31835938,-1.046875,-0.8125,-0.46289062,-0.46875,-0.19726562,0.19726562,0.51953125
-6,0,3,-0.9140625,0.041748047,-0.34179688,1.09375,1.3671875,-0.032226562,-0.43945312,1.0625,0.59765625,-0.66015625,0.27539062,1.84375,-0.2890625,-0.65625,-0.97265625,0.22949219,-0.00076293945,-1.046875,1.96875,-0.48046875,-0.123046875,-2.0625,-0.8125,0.53515625,-0.11816406,-0.62890625,0.71484375,0.73828125,0.68359375,0.40820312,0.44140625,0.2734375,1.328125,-0.515625,2.125,0.88671875,1.171875,-0.82421875,-1.015625,0.6640625,-0.40429688,-0.80859375,-1.8671875,-1.875,-0.828125,0.890625,-0.3984375,-0.94140625,0.86328125,1.75,1.140625,-0.5703125,-1.1484375,0.31054688,-0.12109375,-2.03125,0.35546875,1.515625,0.43359375,0.15722656,0.78125,0.83203125,-0.3515625,1.2890625,-1.3203125,0.5546875,-1.3359375,-1.9140625,0.47851562,-0.049316406,0.8046875,1.4453125,1.765625,-1.7734375,0.8203125,0.921875,-0.08251953,-0.69140625,-0.2890625,0.92578125,-1.03125,0.5703125,-0.28710938,-0.91015625,0.04296875,-1.359375,-0.4375,-0.19238281,1.5625,0.859375,1.4453125,-0.13574219,-0.20117188,1.0859375,-0.3203125,-0.27734375,-0.8203125,-0.9609375,-1.0234375,-0.21289062,0.083984375,0.40625,1.390625,1.1875,-1.0703125,1.3671875,0.61328125,0.19824219,-0.049560547,0.5234375,0.74609375,0.84375,2.484375,-1.578125,0.77734375,-0.35546875,-1.53125,0.66796875,-0.83984375,-0.7890625,1.328125,0.4609375,-0.671875,-0.64453125,-1.296875,2.015625,-0.06689453,0.7578125,0.8203125,-1.28125,1.484375,-2.328125,0.15722656,0.19335938,-0.110839844,0.8984375,0.72265625,-0.12988281,0.41210938,-1.0,1.5703125,0.26367188,-1.4453125,1.3828125,-0.36523438,-1.015625,0.49804688,-0.31640625,-0.1875,-1.25,0.7890625,0.22851562,2.671875,-0.921875,-0.8125,0.0390625,0.140625,-1.59375,1.1953125,0.12207031,0.63671875,0.59765625,1.1328125,0.47851562,0.0546875,0.73046875,0.020141602,0.98046875,-0.33398438,-0.005218506,-1.1953125,-1.40625,-2.359375,-1.2890625,-0.41796875,0.99609375,-0.49804688,-1.53125,-1.640625,-0.953125,0.7109375,0.16601562,0.25585938,-1.234375,-0.83984375,-1.21875,-0.25,0.19042969,-1.2578125,1.078125,-1.2109375,-0.030517578,-0.8671875,-1.1875,-1.609375,0.578125,-0.1875,0.21386719,0.203125,1.609375
-7,0,0,0.8671875,1.2109375,-0.828125,-0.984375,-1.28125,0.32617188,-0.5,1.5234375,-0.7890625,-0.71484375,1.625,0.010192871,-0.060791016,-0.28125,-1.15625,1.640625,-0.28515625,-1.5234375,0.15039062,-0.66796875,1.34375,0.84375,0.47851562,1.125,0.88671875,-0.85546875,-0.9296875,-1.9453125,-1.6171875,-0.62890625,-1.2578125,2.453125,1.015625,2.515625,-0.16992188,0.14550781,-0.40234375,-0.59765625,0.86328125,0.5546875,-0.62890625,0.28125,0.640625,0.20019531,1.3046875,0.36328125,0.58203125,0.27539062,0.6484375,-1.1796875,-0.83984375,-0.5078125,-1.890625,-1.609375,-1.03125,0.59375,0.08642578,-0.33789062,0.31445312,-0.75,-0.19042969,-0.515625,1.8515625,1.0,1.1796875,0.7109375,-0.890625,-0.13964844,0.22460938,0.24707031,-0.03857422,-0.36132812,-0.6796875,-0.18847656,-0.47265625,-0.15332031,-0.23339844,-0.13085938,-0.014709473,0.45898438,-0.48046875,-0.5390625,1.1328125,0.4296875,1.109375,1.2890625,-0.97265625,0.53125,1.359375,-1.015625,1.3359375,0.40625,0.546875,1.2109375,0.21875,-1.3125,0.765625,1.0,-1.71875,-0.29882812,-0.64453125,0.328125,2.234375,-1.7734375,-1.2265625,-1.484375,-2.0,-1.7265625,-0.48242188,0.93359375,0.21289062,-1.0078125,0.09082031,0.57421875,0.52734375,0.203125,0.008850098,-0.984375,-0.265625,0.22753906,-0.36914062,-0.31054688,-1.4375,-1.0,-1.46875,1.0625,-1.7578125,-0.55078125,1.0390625,0.98828125,0.90625,-0.484375,-1.140625,-0.64453125,-1.7578125,0.83203125,1.109375,-0.18359375,-1.6640625,-2.421875,1.4765625,-0.55078125,-0.640625,0.59765625,0.4765625,0.18261719,-0.24902344,-0.52734375,0.71875,1.5390625,-0.75390625,0.27539062,1.2734375,-1.2109375,0.67578125,0.3125,0.59375,-0.86328125,0.77734375,-1.1328125,0.796875,-0.40234375,0.64453125,-0.6484375,-1.140625,3.46875,-0.7578125,-1.1953125,0.33007812,-1.0234375,-1.0,-0.47460938,-1.28125,-1.1171875,-0.38085938,1.171875,0.78515625,-0.72265625,1.0390625,-0.49804688,0.39453125,0.3515625,0.73046875,-1.2890625,1.1484375,-0.053710938,0.08935547,0.28515625,0.25,0.3828125,2.03125,-0.14257812,-0.033203125,-0.7265625,-0.011657715,0.84375,2.0625,1.796875,0.84765625,0.66796875
-7,0,1,0.58984375,1.3984375,-0.609375,-0.51171875,-1.9140625,-0.7421875,1.0234375,0.9765625,0.18652344,0.0099487305,0.11669922,0.44335938,-0.01940918,-0.82421875,-0.51953125,1.796875,-1.2578125,-0.82421875,0.34960938,0.27929688,0.45703125,-0.6015625,1.046875,0.3671875,1.2734375,-0.50390625,-1.984375,-0.71875,0.5,-0.6328125,-0.73828125,-0.008544922,2.21875,-0.43359375,-0.62109375,-0.07519531,0.25976562,-1.28125,-0.515625,-0.6875,-0.42773438,0.18847656,0.29882812,-0.09863281,-0.73046875,0.15234375,0.37890625,0.6640625,-0.27734375,-0.20019531,0.89453125,0.546875,-3.765625,-0.15625,1.15625,1.71875,-0.15332031,0.34375,0.47070312,-0.100097656,-1.0546875,-1.1171875,1.0390625,0.328125,1.6015625,-0.375,-0.23828125,-0.34765625,0.04638672,0.35546875,-0.32617188,1.125,-0.045654297,-0.9921875,2.4375,0.83203125,0.88671875,-0.12792969,-1.4453125,0.34179688,0.59765625,-0.27929688,1.5234375,0.123046875,0.37304688,0.003616333,-1.203125,1.125,1.59375,-0.24609375,0.87890625,-0.34570312,1.0,0.52734375,-1.2734375,0.57421875,-0.84375,1.8203125,-1.3984375,0.35742188,0.33789062,0.9765625,-0.5546875,0.71484375,-0.625,-0.21484375,-1.6875,-1.75,-0.29296875,0.33398438,-1.4609375,-2.3125,0.31445312,-0.20410156,-1.4140625,1.15625,-0.66796875,1.0390625,-1.140625,-0.54296875,-0.13085938,0.41210938,-1.203125,-1.328125,0.484375,0.765625,-1.6328125,-0.09716797,-0.18652344,-1.1328125,1.1484375,-0.23730469,-0.110839844,-2.1875,-0.765625,-0.6640625,1.65625,0.048095703,-1.421875,-2.78125,1.453125,0.15136719,-1.3984375,-1.203125,-0.328125,-0.44921875,0.40625,0.77734375,2.21875,1.9765625,-0.5390625,-1.0703125,0.17578125,-0.3203125,-0.23925781,-1.2734375,1.234375,0.08105469,0.99609375,0.09765625,-0.27148438,-0.072753906,0.22949219,-0.40234375,-1.15625,0.62890625,-0.515625,-1.375,0.58203125,0.46875,0.84375,0.61328125,-0.796875,0.80859375,-1.6015625,1.1328125,2.75,-0.35546875,1.6796875,1.90625,-0.25390625,0.65234375,0.625,-1.0234375,0.047607422,-1.03125,0.49023438,-0.087890625,0.84375,-0.04272461,0.18847656,0.96875,-0.0099487305,0.3125,-1.09375,-0.20214844,0.014343262,1.9921875,1.0703125,0.028198242
-7,0,2,0.9609375,0.671875,-1.0234375,-0.95703125,0.057128906,-0.6953125,-0.15332031,1.9921875,0.82421875,0.625,-1.3359375,1.84375,-0.36328125,1.15625,0.025634766,0.765625,0.62890625,-0.35546875,0.87890625,1.4140625,0.22949219,-0.049072266,-0.74609375,-0.33203125,-0.6171875,1.3828125,-0.64453125,0.028442383,0.8984375,-2.203125,0.5703125,1.796875,1.1796875,0.16113281,-0.59375,-0.4140625,0.22558594,1.4609375,0.080078125,-0.107421875,-0.9765625,-0.07910156,-0.53125,0.49023438,0.26171875,0.05053711,-0.24121094,-0.33203125,0.071777344,0.18066406,0.09472656,-1.390625,-0.35742188,-0.73828125,0.07324219,0.49804688,-1.6484375,0.15332031,0.28515625,-0.64453125,-1.1484375,-0.6875,1.6171875,-0.33789062,0.45117188,0.55859375,0.25195312,-0.18261719,-0.29492188,1.046875,0.22167969,2.078125,-0.859375,-0.7421875,0.86328125,1.5859375,0.16601562,-0.81640625,-2.84375,0.70703125,1.6015625,0.578125,1.859375,2.0625,-0.66796875,0.80078125,-0.6796875,-0.65625,-0.55859375,1.375,1.140625,1.0,-0.16992188,1.2890625,0.020141602,0.111816406,-1.640625,1.03125,-0.12890625,0.45507812,0.66796875,-0.50390625,-0.484375,-0.34375,-0.59765625,0.9375,-0.3359375,-0.9140625,-1.71875,-0.875,-0.17871094,-0.47265625,1.59375,-0.83203125,-0.15039062,-1.078125,0.21484375,0.19238281,0.8984375,0.8046875,1.234375,0.06225586,-1.0703125,-0.78515625,-0.46679688,0.73828125,-0.17675781,-0.23925781,1.2265625,-0.53125,0.8984375,-1.859375,0.29492188,-1.1640625,-1.359375,0.765625,-1.578125,0.16601562,0.22460938,-2.046875,0.071777344,-0.13378906,-1.375,1.234375,-0.23828125,0.06689453,-0.53515625,1.7109375,0.21679688,-1.875,1.5859375,-2.109375,0.34375,0.20410156,0.15039062,0.46679688,1.421875,0.68359375,-0.54296875,0.23535156,-1.578125,2.234375,-0.8359375,0.796875,1.1484375,-0.78515625,-0.484375,-2.875,0.23535156,1.25,-0.55859375,0.73828125,-1.1171875,1.3203125,0.091308594,1.703125,0.23144531,-1.4921875,-0.765625,0.82421875,-0.61328125,0.51953125,-0.41601562,0.47070312,-1.7109375,0.29882812,-1.6796875,-2.046875,1.484375,-0.9765625,-1.5703125,-0.92578125,-1.5234375,0.48242188,0.06982422,0.064941406,0.24121094,-0.37890625,-0.859375,1.265625
+6,0,1,0.13867188,0.10498047,0.2734375,-0.88671875,1.3984375,0.6171875,0.49023438,1.0234375,-0.4921875,-1.0390625,0.12792969,-0.26757812,0.58203125,0.35351562,-1.25,0.48828125,-0.546875,-1.90625,1.1328125,-0.5234375,-0.0859375,-0.6640625,1.28125,0.53125,-1.234375,-1.5078125,-0.95703125,-0.86328125,-1.7890625,0.44335938,-0.63671875,-0.26367188,0.011413574,-0.46484375,-0.52734375,0.953125,0.9140625,-0.7421875,-1.4609375,1.578125,-1.7890625,0.029663086,-0.075683594,0.09765625,0.38476562,0.52734375,0.63671875,-0.703125,0.37304688,1.8671875,-1.1953125,-1.6015625,-0.96875,1.109375,-2.421875,0.09277344,-0.21289062,-0.8046875,1.328125,-1.71875,1.75,0.18945312,0.29101562,0.29882812,-0.51171875,-0.064453125,-1.078125,0.16601562,0.546875,1.8359375,0.060791016,-0.31640625,0.57421875,-0.13769531,-0.671875,1.8671875,0.765625,1.1796875,-1.78125,-0.5234375,1.2890625,0.890625,-0.12011719,0.12109375,-0.63671875,-0.20214844,-0.46679688,1.4609375,-0.35742188,0.9921875,1.3828125,0.21289062,-0.921875,0.36914062,-1.421875,0.22460938,-0.59375,-0.640625,0.62109375,-0.890625,-2.375,-0.17675781,2.515625,0.77734375,-0.79296875,0.0059814453,0.06689453,-1.0703125,0.48632812,0.984375,0.703125,0.03515625,0.5703125,-0.796875,0.49023438,-0.31835938,-0.47460938,-0.35351562,-0.48632812,-1.1875,0.7109375,-0.7890625,0.6015625,-1.5703125,-0.22363281,1.2578125,-1.3203125,-2.421875,0.016235352,0.17382812,-0.18554688,0.044921875,-0.022460938,-1.3671875,-1.03125,-0.13769531,-0.65625,0.25390625,0.20605469,-1.734375,3.375,-0.40625,1.3125,1.1015625,-1.2265625,1.0703125,-0.6015625,-0.6640625,-2.515625,1.1015625,-0.31054688,-0.032958984,1.0625,-0.48828125,-1.09375,-1.1953125,0.38085938,1.3515625,1.015625,1.421875,-0.26171875,-0.012084961,1.828125,1.2421875,-1.78125,1.65625,-1.671875,-0.24511719,2.109375,-0.20117188,0.3671875,-0.40234375,0.09765625,0.6171875,0.6875,0.0390625,-0.8671875,-0.84375,0.86328125,-0.609375,-0.3046875,-0.88671875,0.13964844,0.85546875,0.14746094,0.5390625,0.59765625,0.040527344,0.86328125,-1.828125,0.095703125,1.375,1.2578125,0.28710938,0.4375,-0.18261719,1.6796875,0.15722656,1.4140625,0.51953125
+6,0,2,0.82421875,0.82421875,-0.3984375,0.84375,1.390625,0.62890625,-0.35546875,0.39257812,-0.10058594,-1.4375,1.109375,-0.19238281,0.53515625,-0.54296875,-1.0,-0.37109375,-0.81640625,-0.40039062,1.109375,1.2421875,0.20507812,-1.8984375,0.36132812,0.359375,-0.90234375,-1.609375,-1.3828125,-0.35742188,0.22753906,0.05883789,-0.23632812,-0.32617188,1.71875,0.20800781,-0.328125,-0.16894531,0.31640625,-0.6015625,-0.26367188,0.78125,-0.60546875,0.71484375,-0.26953125,1.2734375,-0.27539062,0.25195312,-0.734375,-1.0234375,0.43945312,1.71875,1.140625,-1.765625,-0.4921875,-0.26367188,-1.0859375,-2.125,0.38476562,1.3046875,-0.109375,-1.015625,1.6171875,1.59375,1.734375,-0.34960938,-1.0390625,-0.7109375,0.05029297,-2.078125,-0.30273438,0.85546875,0.3828125,0.16992188,1.34375,-1.4921875,0.4609375,0.515625,0.18457031,-0.20605469,-1.7109375,0.39648438,-0.27539062,0.71875,-0.018310547,0.31835938,-0.76171875,0.7421875,-2.828125,0.453125,1.5859375,1.203125,1.421875,1.1953125,1.125,0.22753906,-0.6328125,-0.9375,-0.875,1.2421875,-0.22949219,0.123535156,-0.54296875,-0.39453125,0.89453125,-0.11621094,-1.0390625,0.77734375,-0.9921875,-0.18164062,-0.34960938,-0.33203125,0.18261719,1.96875,0.13378906,-0.6328125,-0.57421875,0.671875,-0.41601562,0.33203125,-0.46484375,0.83984375,0.05102539,-1.1015625,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.59765625,-0.2421875,0.76171875,-0.04321289,-2.515625,-0.9375,0.7421875,-1.3203125,2.546875,-0.31835938,0.01574707,1.8125,-1.3125,1.421875,-0.75390625,0.119628906,1.171875,0.6640625,-0.20214844,-0.22460938,-0.27929688,-0.56640625,0.20019531,0.43164062,-0.32421875,-1.34375,1.015625,-1.1328125,1.234375,0.6953125,-0.36523438,-0.38867188,1.046875,0.78515625,1.15625,1.0078125,3.046875,-2.671875,1.609375,-2.234375,1.1875,0.16699219,-0.29492188,-1.7734375,-1.828125,0.734375,-0.65234375,0.65625,0.27539062,-0.66796875,-1.984375,1.40625,-0.35742188,0.10107422,0.15136719,0.84375,-0.115722656,-1.0234375,1.578125,-0.48828125,-0.26953125,0.5078125,-0.36523438,0.66015625,1.546875,-0.31640625,-1.0390625,-0.82421875,-0.47070312,-0.46875,-0.20019531,0.19726562,0.51953125
+6,0,3,-0.91015625,0.041748047,-0.34375,1.078125,1.359375,-0.030395508,-0.43945312,1.0546875,0.59765625,-0.6640625,0.27539062,1.84375,-0.29296875,-0.6640625,-0.97265625,0.23046875,0.0031280518,-1.046875,1.9765625,-0.4765625,-0.12402344,-2.0625,-0.8203125,0.54296875,-0.12060547,-0.62890625,0.7109375,0.73828125,0.69140625,0.41015625,0.4375,0.2734375,1.3203125,-0.52734375,2.140625,0.88671875,1.1796875,-0.828125,-1.015625,0.6640625,-0.40429688,-0.80859375,-1.859375,-1.875,-0.82421875,0.8984375,-0.40039062,-0.94140625,0.8515625,1.7578125,1.125,-0.5703125,-1.15625,0.31445312,-0.123046875,-2.03125,0.35351562,1.515625,0.43359375,0.15722656,0.78125,0.828125,-0.3515625,1.28125,-1.3125,0.55078125,-1.3359375,-1.90625,0.47460938,-0.0546875,0.8203125,1.4453125,1.7734375,-1.7734375,0.81640625,0.921875,-0.08251953,-0.69140625,-0.2890625,0.9296875,-1.0390625,0.578125,-0.29101562,-0.9140625,0.041992188,-1.3671875,-0.43554688,-0.19140625,1.5703125,0.859375,1.4453125,-0.13769531,-0.20117188,1.0859375,-0.31835938,-0.27539062,-0.82421875,-0.9609375,-1.03125,-0.21972656,0.083984375,0.40625,1.3984375,1.1796875,-1.0703125,1.359375,0.6171875,0.19921875,-0.049804688,0.51953125,0.75390625,0.83984375,2.5,-1.59375,0.76953125,-0.35351562,-1.53125,0.671875,-0.84765625,-0.7890625,1.3125,0.46289062,-0.671875,-0.6484375,-1.296875,2.03125,-0.06982422,0.76953125,0.8125,-1.2734375,1.4765625,-2.3125,0.15332031,0.19628906,-0.10644531,0.8984375,0.7265625,-0.12988281,0.41015625,-0.99609375,1.5625,0.265625,-1.453125,1.3828125,-0.36523438,-1.0234375,0.49414062,-0.31640625,-0.18554688,-1.2578125,0.78515625,0.22363281,2.671875,-0.92578125,-0.8203125,0.040527344,0.13964844,-1.578125,1.1875,0.119140625,0.63671875,0.59765625,1.125,0.48046875,0.052734375,0.7265625,0.024047852,0.984375,-0.33203125,-0.004211426,-1.1875,-1.3984375,-2.359375,-1.28125,-0.4140625,0.9921875,-0.5,-1.5234375,-1.6328125,-0.953125,0.703125,0.15820312,0.25585938,-1.234375,-0.828125,-1.21875,-0.25,0.18945312,-1.2578125,1.078125,-1.21875,-0.030395508,-0.87109375,-1.1953125,-1.6171875,0.578125,-0.1875,0.21777344,0.20019531,1.609375
+7,0,0,0.86328125,1.2109375,-0.83203125,-0.984375,-1.2890625,0.32617188,-0.49804688,1.515625,-0.7890625,-0.70703125,1.625,0.0032348633,-0.060791016,-0.28125,-1.15625,1.6328125,-0.2890625,-1.5234375,0.15039062,-0.6640625,1.34375,0.84765625,0.47851562,1.125,0.8984375,-0.859375,-0.93359375,-1.9375,-1.6171875,-0.63671875,-1.2578125,2.453125,1.0078125,2.515625,-0.16796875,0.14746094,-0.40234375,-0.59765625,0.8671875,0.5625,-0.625,0.28125,0.63671875,0.20019531,1.3046875,0.36328125,0.578125,0.26953125,0.65625,-1.1796875,-0.84765625,-0.51171875,-1.890625,-1.609375,-1.03125,0.59765625,0.08544922,-0.33984375,0.3125,-0.74609375,-0.19042969,-0.515625,1.8515625,1.0,1.1796875,0.70703125,-0.88671875,-0.13574219,0.22558594,0.25195312,-0.036376953,-0.36132812,-0.671875,-0.18847656,-0.4765625,-0.15820312,-0.234375,-0.13183594,-0.014892578,0.45507812,-0.48242188,-0.546875,1.1328125,0.42382812,1.109375,1.2890625,-0.97265625,0.52734375,1.375,-1.0234375,1.3359375,0.41015625,0.546875,1.2109375,0.21972656,-1.3203125,0.765625,0.99609375,-1.703125,-0.296875,-0.6484375,0.33007812,2.21875,-1.7734375,-1.2265625,-1.4765625,-2.015625,-1.7109375,-0.47851562,0.93359375,0.20996094,-1.0078125,0.088378906,0.5703125,0.53125,0.20117188,0.0043640137,-0.98046875,-0.27148438,0.22851562,-0.3671875,-0.31054688,-1.4375,-1.0,-1.4609375,1.0625,-1.75,-0.55078125,1.0390625,0.984375,0.9140625,-0.484375,-1.1328125,-0.6484375,-1.7734375,0.82421875,1.1171875,-0.18359375,-1.6640625,-2.4375,1.484375,-0.5546875,-0.640625,0.59375,0.4765625,0.18652344,-0.24804688,-0.5234375,0.72265625,1.5390625,-0.75390625,0.2734375,1.2734375,-1.21875,0.6796875,0.3125,0.59375,-0.859375,0.76953125,-1.140625,0.80078125,-0.40234375,0.64453125,-0.64453125,-1.140625,3.453125,-0.76171875,-1.1796875,0.328125,-1.015625,-1.0,-0.47070312,-1.2890625,-1.1171875,-0.38476562,1.1796875,0.7890625,-0.72265625,1.046875,-0.49804688,0.39453125,0.3515625,0.73046875,-1.2890625,1.1484375,-0.053710938,0.09082031,0.28515625,0.24609375,0.37890625,2.03125,-0.14257812,-0.033203125,-0.7265625,-0.009887695,0.8359375,2.0625,1.796875,0.8515625,0.66796875
+7,0,1,0.5859375,1.3984375,-0.61328125,-0.51953125,-1.9140625,-0.7578125,1.0234375,0.9765625,0.18945312,0.010009766,0.115234375,0.43945312,-0.021118164,-0.828125,-0.51171875,1.796875,-1.2578125,-0.82421875,0.34960938,0.28125,0.45703125,-0.6015625,1.046875,0.3671875,1.2890625,-0.515625,-1.984375,-0.72265625,0.5,-0.62890625,-0.73828125,-0.009216309,2.21875,-0.43554688,-0.625,-0.076660156,0.26171875,-1.28125,-0.51953125,-0.68359375,-0.4296875,0.18847656,0.29882812,-0.103027344,-0.73046875,0.15332031,0.37890625,0.65625,-0.28125,-0.20117188,0.89453125,0.546875,-3.75,-0.15527344,1.15625,1.71875,-0.14941406,0.34375,0.47070312,-0.095703125,-1.0546875,-1.1171875,1.0390625,0.33007812,1.59375,-0.375,-0.234375,-0.34765625,0.040771484,0.35742188,-0.32226562,1.125,-0.045410156,-0.984375,2.453125,0.83203125,0.87890625,-0.13183594,-1.4375,0.34179688,0.59765625,-0.28125,1.53125,0.125,0.37304688,0.0048828125,-1.203125,1.1328125,1.6015625,-0.24316406,0.8828125,-0.3515625,1.0,0.52734375,-1.265625,0.57421875,-0.83984375,1.828125,-1.3984375,0.35742188,0.33789062,0.97265625,-0.5625,0.71875,-0.6171875,-0.20898438,-1.6796875,-1.75,-0.29101562,0.33398438,-1.46875,-2.296875,0.30859375,-0.20117188,-1.4140625,1.15625,-0.66796875,1.03125,-1.140625,-0.546875,-0.12988281,0.41210938,-1.203125,-1.3359375,0.49023438,0.76171875,-1.625,-0.10107422,-0.18554688,-1.125,1.140625,-0.23339844,-0.11035156,-2.171875,-0.765625,-0.6640625,1.6484375,0.048339844,-1.4296875,-2.796875,1.4453125,0.14941406,-1.390625,-1.21875,-0.33007812,-0.44921875,0.40039062,0.77734375,2.25,1.9765625,-0.5390625,-1.0703125,0.17675781,-0.31835938,-0.23925781,-1.2734375,1.234375,0.084472656,0.9921875,0.096191406,-0.2734375,-0.07128906,0.22753906,-0.40429688,-1.1484375,0.625,-0.515625,-1.375,0.578125,0.46679688,0.84765625,0.609375,-0.8046875,0.8046875,-1.6015625,1.1328125,2.75,-0.35351562,1.6796875,1.8984375,-0.25195312,0.65234375,0.62109375,-1.0234375,0.043945312,-1.046875,0.4921875,-0.087402344,0.83984375,-0.043701172,0.18652344,0.96875,-0.011230469,0.31445312,-1.09375,-0.20605469,0.013671875,2.0,1.0703125,0.027832031
+7,0,2,0.9609375,0.67578125,-1.0078125,-0.953125,0.055419922,-0.69921875,-0.1484375,1.9921875,0.8203125,0.62109375,-1.328125,1.8359375,-0.36328125,1.15625,0.020263672,0.7734375,0.62109375,-0.35742188,0.875,1.421875,0.234375,-0.05126953,-0.75390625,-0.328125,-0.6171875,1.375,-0.64453125,0.026977539,0.8984375,-2.203125,0.5625,1.7890625,1.1796875,0.15917969,-0.59765625,-0.41601562,0.22753906,1.4609375,0.08642578,-0.104003906,-0.984375,-0.07714844,-0.52734375,0.49414062,0.265625,0.05053711,-0.23925781,-0.33203125,0.06738281,0.17675781,0.09814453,-1.390625,-0.359375,-0.7421875,0.06982422,0.50390625,-1.6484375,0.15429688,0.28320312,-0.63671875,-1.1484375,-0.69140625,1.6171875,-0.34179688,0.453125,0.55859375,0.25390625,-0.17871094,-0.296875,1.0390625,0.22070312,2.09375,-0.86328125,-0.7421875,0.8671875,1.5859375,0.171875,-0.8125,-2.859375,0.7109375,1.6171875,0.578125,1.84375,2.0625,-0.66796875,0.8046875,-0.6796875,-0.65234375,-0.5625,1.375,1.1484375,1.0,-0.16992188,1.2890625,0.021362305,0.11230469,-1.6328125,1.03125,-0.13183594,0.453125,0.66796875,-0.50390625,-0.48828125,-0.34375,-0.59765625,0.94140625,-0.34375,-0.90625,-1.7265625,-0.87109375,-0.17871094,-0.47070312,1.5859375,-0.83203125,-0.15136719,-1.0859375,0.21484375,0.19140625,0.8984375,0.80859375,1.2265625,0.06738281,-1.078125,-0.78125,-0.46875,0.73046875,-0.18164062,-0.23535156,1.21875,-0.54296875,0.8984375,-1.859375,0.29296875,-1.1640625,-1.359375,0.7734375,-1.5703125,0.16601562,0.22167969,-2.0625,0.0703125,-0.13085938,-1.359375,1.2421875,-0.24121094,0.06738281,-0.53515625,1.71875,0.21972656,-1.8828125,1.5703125,-2.09375,0.34765625,0.20410156,0.14941406,0.46679688,1.4140625,0.68359375,-0.54296875,0.23535156,-1.578125,2.25,-0.83203125,0.796875,1.15625,-0.78515625,-0.484375,-2.859375,0.23925781,1.25,-0.55859375,0.7421875,-1.1171875,1.3203125,0.092285156,1.703125,0.22558594,-1.4921875,-0.765625,0.82421875,-0.609375,0.51953125,-0.41992188,0.47070312,-1.7109375,0.29882812,-1.6953125,-2.046875,1.484375,-0.97265625,-1.5703125,-0.92578125,-1.5390625,0.484375,0.072265625,0.06689453,0.23925781,-0.38085938,-0.85546875,1.265625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
index 439e2692..24fc7499 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,1,1.078125,0.25195312,-0.16601562,-0.296875,0.109375,-0.42578125,-1.28125,0.7890625,0.23632812,0.46484375,0.13574219,0.921875,1.2421875,-0.13476562,-0.55859375,-0.13183594,-1.078125,-2.25,0.9375,0.21386719,1.7890625,-1.0234375,-1.2890625,-0.2109375,1.40625,-0.30078125,-0.71484375,-0.013305664,0.2578125,-0.14648438,-0.84765625,-0.61328125,1.9453125,-0.94921875,-0.93359375,-0.41601562,0.09472656,-1.140625,-2.578125,-0.453125,0.44726562,-0.87890625,-0.71484375,-0.29882812,-1.21875,0.5234375,-0.3203125,0.5,1.765625,1.4296875,-0.27148438,0.36328125,-1.6640625,1.2578125,-0.10595703,0.38671875,-0.5,1.8828125,0.71484375,-0.82421875,-0.69140625,-1.9453125,0.76953125,0.80078125,1.328125,1.46875,-1.4609375,-1.875,0.41601562,0.9375,1.421875,1.2265625,-0.41992188,-0.94921875,2.90625,1.78125,-0.3515625,-0.11669922,-0.5,0.26953125,-0.34765625,0.26757812,0.0007362366,2.34375,-0.48046875,-0.58984375,-0.41601562,2.5,1.0546875,0.91015625,0.8125,0.5390625,0.8203125,0.06933594,-0.84765625,-0.13769531,0.5546875,1.0390625,0.5078125,-0.9453125,0.19824219,1.0390625,-0.46484375,0.37109375,-0.16796875,0.76953125,-1.171875,-1.15625,0.375,0.33398438,-1.015625,-0.15820312,-0.55859375,-1.9453125,-1.78125,0.10253906,-0.625,1.2265625,-1.2578125,0.44140625,0.61328125,0.41210938,-2.203125,-1.359375,-0.8046875,0.18847656,-0.5625,0.43945312,0.6640625,-0.62890625,1.7578125,-0.25390625,0.14550781,-1.5078125,-0.65234375,0.5703125,0.55859375,0.19726562,-0.75,-2.5,1.0703125,2.59375,-0.7109375,-1.21875,-1.265625,1.234375,0.31640625,-0.15332031,0.31640625,0.6015625,0.56640625,0.98046875,0.6328125,-0.6171875,-0.77734375,-1.4609375,0.29296875,-0.049316406,1.171875,-0.64453125,0.35546875,0.57421875,0.080078125,-1.0703125,0.5859375,-0.28515625,0.57421875,-0.74609375,-1.2890625,1.2890625,0.14550781,0.65234375,-0.8984375,-0.52734375,-1.6953125,1.84375,-0.17871094,-0.58203125,0.07861328,-0.48632812,1.25,0.40625,0.828125,-0.68359375,-1.0703125,-1.5546875,0.18652344,0.36328125,0.6953125,0.625,-0.30078125,0.36328125,0.796875,-0.93359375,-0.6328125,-1.8828125,-0.6953125,1.25,1.03125,0.72265625
-8,0,2,0.87109375,0.984375,-0.17871094,0.12597656,0.08203125,-0.60546875,0.765625,1.0703125,1.078125,-0.80859375,1.3203125,1.5546875,0.053710938,-1.03125,-0.14355469,0.30078125,-0.67578125,-1.0078125,0.8359375,0.3359375,0.026489258,-0.36328125,-0.24609375,1.640625,-0.18164062,0.14550781,-2.78125,0.21191406,0.3203125,0.7890625,-0.5078125,-0.90625,1.3671875,-0.78125,0.57421875,0.59375,0.53515625,-0.93359375,-0.21972656,1.0859375,-1.2109375,1.109375,-0.34179688,-2.421875,-0.8671875,-0.7890625,0.25976562,-0.22949219,1.3046875,1.6015625,0.5546875,-0.47070312,-0.92578125,0.6796875,0.46875,0.26757812,0.19921875,-0.5078125,-1.0703125,-0.036376953,1.4765625,0.20703125,0.6328125,0.859375,0.50390625,-0.43554688,-0.14160156,0.55859375,-1.3671875,0.66796875,-0.09863281,0.5390625,0.84765625,-0.42578125,1.8359375,0.8046875,1.3984375,1.046875,-0.35546875,0.36914062,-0.0859375,-1.125,0.83984375,-1.4375,-0.45507812,0.3671875,-0.03173828,-0.036132812,1.75,0.1875,1.34375,-0.5390625,-1.03125,1.2421875,-1.09375,-0.59765625,-2.75,0.49023438,-0.6015625,-0.3671875,-1.453125,1.171875,0.38085938,0.19140625,-0.58984375,1.84375,-0.75,0.16601562,-1.1640625,0.30859375,-0.56640625,-0.42578125,0.91796875,-0.26367188,0.36328125,1.765625,-0.5078125,0.5546875,-0.15917969,-0.44335938,0.57421875,-0.62890625,-0.4921875,-0.94140625,-1.328125,2.109375,-1.2578125,-1.0546875,0.71875,-1.265625,1.3203125,-1.2109375,-0.6875,-0.76171875,-1.59375,0.65234375,1.09375,-0.52734375,0.057373047,-2.46875,1.703125,-0.9453125,-1.7578125,1.3671875,0.71484375,-0.04272461,0.875,-0.89453125,-1.171875,0.9140625,-1.296875,-0.70703125,0.68359375,-1.0234375,1.109375,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.07128906,1.65625,2.453125,1.4609375,-1.53125,0.100097656,-0.5390625,0.084472656,0.859375,-0.16113281,-1.015625,-0.296875,-1.6328125,-0.14746094,-1.5390625,1.0390625,0.9140625,-1.1640625,0.29492188,0.046875,-1.2734375,-1.4140625,0.54296875,-1.0,-0.45703125,0.24902344,0.022705078,0.56640625,-1.6171875,-1.625,-0.234375,2.46875,-0.3671875,-0.9609375,-1.546875,1.03125,1.4921875,1.09375,-0.421875,-0.59375
-8,0,3,1.25,1.5546875,0.10205078,-0.26171875,-1.609375,-0.016479492,-1.328125,1.4375,1.40625,-2.640625,1.7578125,-0.69140625,0.03100586,1.1171875,-0.49609375,0.46679688,0.29101562,0.41210938,1.859375,-0.016845703,-0.8046875,-0.034179688,-0.39453125,-0.25195312,-0.8671875,-1.0546875,-0.71484375,-0.22363281,0.09765625,-0.20605469,-0.25195312,0.01928711,1.0546875,0.71875,0.099609375,0.28710938,1.421875,0.14550781,-0.8828125,1.6875,-2.3125,-0.22265625,0.114746094,2.140625,0.26367188,0.921875,-0.87109375,0.71484375,-0.0625,0.87890625,-0.6484375,-0.09033203,-0.34960938,-0.25390625,-0.4140625,-0.45703125,0.19628906,0.8125,-1.1328125,0.16796875,0.546875,0.65234375,1.65625,-0.41601562,1.359375,0.43359375,-0.36328125,-1.2578125,-0.98046875,0.076660156,-0.2734375,-0.234375,0.6171875,0.04272461,-0.9140625,0.734375,0.14160156,0.34375,-1.53125,1.3203125,-0.81640625,-0.66796875,0.95703125,-1.2421875,-0.17773438,0.3359375,0.24609375,-1.3046875,0.10058594,0.51171875,1.953125,1.5859375,-0.35546875,1.0546875,-0.89453125,-0.14648438,-0.6796875,0.0028839111,0.19433594,-0.703125,-2.390625,0.87890625,1.71875,-1.7421875,0.03149414,1.5390625,-1.109375,-0.46484375,0.053466797,0.20800781,-0.08691406,-0.3125,1.328125,-0.7265625,0.06738281,0.080078125,-0.37890625,0.24707031,-0.25390625,0.859375,1.640625,-0.56640625,-0.15332031,-1.796875,-0.51171875,0.20214844,0.39257812,-1.5546875,-0.11767578,1.3828125,-0.890625,1.1171875,-1.3984375,0.6484375,-2.09375,1.515625,0.9140625,-0.59765625,-0.7109375,-2.328125,1.5859375,-1.1640625,-0.765625,0.03930664,-0.18945312,0.203125,0.48046875,-0.5859375,-0.76171875,1.2421875,-1.40625,0.29492188,0.578125,-0.114746094,-0.31835938,0.09765625,1.1953125,0.34179688,0.53125,0.03564453,-0.66796875,0.16796875,0.44140625,0.13085938,-1.2578125,0.84375,-0.6015625,0.67578125,0.671875,-2.015625,-2.109375,-1.328125,-2.171875,-0.49414062,0.34765625,2.28125,0.36914062,-0.5234375,0.036865234,-0.44726562,0.64453125,-0.27734375,-0.25390625,0.21191406,-0.38085938,0.14160156,0.91796875,1.453125,-0.47460938,0.048583984,0.084472656,0.2578125,0.77734375,-1.421875,-3.140625,1.265625,3.359375,0.6484375,-0.44726562,0.4921875
+8,0,1,1.0703125,0.24707031,-0.1640625,-0.29101562,0.11035156,-0.42382812,-1.28125,0.7890625,0.23632812,0.46875,0.13476562,0.91796875,1.2421875,-0.13476562,-0.56640625,-0.13183594,-1.078125,-2.234375,0.93359375,0.21582031,1.78125,-1.03125,-1.2890625,-0.20996094,1.40625,-0.30273438,-0.71875,-0.009643555,0.2578125,-0.14453125,-0.84375,-0.6171875,1.9453125,-0.953125,-0.9296875,-0.41796875,0.09423828,-1.140625,-2.578125,-0.46484375,0.44726562,-0.87890625,-0.70703125,-0.29882812,-1.2265625,0.5234375,-0.31445312,0.50390625,1.7578125,1.4375,-0.27148438,0.359375,-1.65625,1.2578125,-0.103027344,0.3828125,-0.5,1.8828125,0.7109375,-0.82421875,-0.6953125,-1.9453125,0.76953125,0.8046875,1.3359375,1.4765625,-1.4609375,-1.8671875,0.41992188,0.9375,1.421875,1.21875,-0.41992188,-0.94921875,2.90625,1.78125,-0.35546875,-0.11279297,-0.49414062,0.2734375,-0.34375,0.26757812,0.0005950928,2.359375,-0.48046875,-0.58984375,-0.41601562,2.5,1.0625,0.92578125,0.8125,0.5390625,0.81640625,0.06640625,-0.84375,-0.13378906,0.55078125,1.0390625,0.51171875,-0.94140625,0.1953125,1.0390625,-0.47070312,0.36914062,-0.16601562,0.7734375,-1.171875,-1.140625,0.37890625,0.33007812,-1.0234375,-0.15527344,-0.5625,-1.9453125,-1.78125,0.10498047,-0.6328125,1.21875,-1.265625,0.4453125,0.61328125,0.41796875,-2.21875,-1.3671875,-0.8046875,0.18652344,-0.5625,0.44140625,0.6640625,-0.6328125,1.7734375,-0.25,0.14453125,-1.5,-0.65625,0.56640625,0.55078125,0.19335938,-0.74609375,-2.5,1.078125,2.578125,-0.71484375,-1.21875,-1.265625,1.2265625,0.32421875,-0.15625,0.32421875,0.60546875,0.5625,0.98046875,0.6328125,-0.62109375,-0.78125,-1.453125,0.29101562,-0.05419922,1.171875,-0.65625,0.35742188,0.57421875,0.078125,-1.0625,0.5859375,-0.28320312,0.578125,-0.74609375,-1.28125,1.265625,0.14160156,0.65234375,-0.8984375,-0.53125,-1.703125,1.84375,-0.17773438,-0.578125,0.07519531,-0.48242188,1.25,0.40625,0.81640625,-0.68359375,-1.0703125,-1.5625,0.18847656,0.36328125,0.69140625,0.62890625,-0.30078125,0.36132812,0.796875,-0.9375,-0.640625,-1.84375,-0.6953125,1.2578125,1.0234375,0.7265625
+8,0,2,0.875,0.9921875,-0.1796875,0.12597656,0.080566406,-0.609375,0.765625,1.0703125,1.0703125,-0.80078125,1.328125,1.546875,0.05419922,-1.03125,-0.14648438,0.296875,-0.6796875,-1.0078125,0.83984375,0.33398438,0.026733398,-0.36523438,-0.24511719,1.625,-0.18164062,0.14550781,-2.78125,0.21484375,0.3125,0.7890625,-0.5078125,-0.8984375,1.3671875,-0.77734375,0.578125,0.59375,0.53515625,-0.9375,-0.22070312,1.078125,-1.203125,1.1171875,-0.33984375,-2.40625,-0.8671875,-0.78125,0.2578125,-0.23046875,1.296875,1.609375,0.5625,-0.47460938,-0.92578125,0.67578125,0.47265625,0.265625,0.19921875,-0.515625,-1.0625,-0.035888672,1.4765625,0.2109375,0.6328125,0.86328125,0.50390625,-0.4296875,-0.140625,0.5625,-1.359375,0.67578125,-0.10058594,0.54296875,0.84765625,-0.42773438,1.8359375,0.8046875,1.3828125,1.0546875,-0.35351562,0.37109375,-0.08251953,-1.1328125,0.8359375,-1.4453125,-0.453125,0.36523438,-0.03149414,-0.032470703,1.7421875,0.19042969,1.3515625,-0.5390625,-1.03125,1.2421875,-1.1015625,-0.59375,-2.734375,0.49414062,-0.6015625,-0.37109375,-1.4609375,1.1640625,0.38085938,0.19433594,-0.59765625,1.8359375,-0.75390625,0.16699219,-1.15625,0.30664062,-0.56640625,-0.42773438,0.91796875,-0.26171875,0.36328125,1.765625,-0.5078125,0.5546875,-0.15917969,-0.44140625,0.57421875,-0.625,-0.49023438,-0.9375,-1.328125,2.109375,-1.265625,-1.046875,0.71875,-1.2890625,1.3203125,-1.21875,-0.6875,-0.76953125,-1.6015625,0.65234375,1.09375,-0.5234375,0.05810547,-2.46875,1.703125,-0.94921875,-1.765625,1.375,0.71484375,-0.04248047,0.8671875,-0.89453125,-1.1640625,0.9140625,-1.2890625,-0.703125,0.68359375,-1.015625,1.1171875,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.0703125,1.6484375,2.453125,1.453125,-1.5234375,0.09863281,-0.5390625,0.084472656,0.859375,-0.15722656,-1.015625,-0.29492188,-1.640625,-0.1484375,-1.5390625,1.0390625,0.9140625,-1.1640625,0.29492188,0.050048828,-1.2734375,-1.40625,0.54296875,-1.0,-0.45507812,0.24804688,0.021484375,0.56640625,-1.6171875,-1.6328125,-0.23339844,2.46875,-0.3671875,-0.97265625,-1.5625,1.03125,1.4921875,1.09375,-0.41992188,-0.59375
+8,0,3,1.2578125,1.5546875,0.1015625,-0.25585938,-1.5859375,-0.018310547,-1.328125,1.4375,1.4140625,-2.65625,1.7578125,-0.6875,0.03100586,1.1171875,-0.4921875,0.46484375,0.2890625,0.40625,1.859375,-0.014160156,-0.8046875,-0.036132812,-0.390625,-0.25,-0.86328125,-1.0625,-0.71484375,-0.22070312,0.100097656,-0.20605469,-0.25,0.018432617,1.0546875,0.71484375,0.09716797,0.29101562,1.40625,0.1484375,-0.88671875,1.6953125,-2.3125,-0.22363281,0.114746094,2.140625,0.25976562,0.91796875,-0.87109375,0.7109375,-0.061767578,0.87890625,-0.64453125,-0.09033203,-0.3515625,-0.25,-0.41210938,-0.45507812,0.19433594,0.80859375,-1.1328125,0.16601562,0.546875,0.65625,1.65625,-0.40820312,1.3671875,0.43164062,-0.36328125,-1.265625,-0.98046875,0.07763672,-0.27148438,-0.23730469,0.6171875,0.037841797,-0.91015625,0.734375,0.14160156,0.34179688,-1.53125,1.3203125,-0.81640625,-0.66015625,0.953125,-1.2265625,-0.1796875,0.33398438,0.24902344,-1.3046875,0.1015625,0.51171875,1.9453125,1.5859375,-0.35742188,1.0546875,-0.89453125,-0.14648438,-0.68359375,0.004760742,0.19335938,-0.703125,-2.390625,0.87890625,1.7265625,-1.7421875,0.03515625,1.546875,-1.109375,-0.46289062,0.0546875,0.20410156,-0.091308594,-0.3125,1.3359375,-0.73046875,0.063964844,0.080566406,-0.38085938,0.24902344,-0.25390625,0.859375,1.6484375,-0.56640625,-0.15429688,-1.796875,-0.515625,0.20117188,0.39257812,-1.5546875,-0.11816406,1.390625,-0.890625,1.109375,-1.3984375,0.6484375,-2.078125,1.515625,0.90625,-0.6015625,-0.7109375,-2.328125,1.5859375,-1.171875,-0.76171875,0.033691406,-0.19140625,0.203125,0.48046875,-0.59375,-0.7578125,1.234375,-1.3984375,0.29492188,0.58203125,-0.119628906,-0.31835938,0.09667969,1.1953125,0.34570312,0.52734375,0.037353516,-0.67578125,0.16894531,0.4375,0.12695312,-1.265625,0.83984375,-0.6015625,0.6796875,0.671875,-2.015625,-2.109375,-1.3359375,-2.171875,-0.49023438,0.34765625,2.296875,0.3671875,-0.5234375,0.037597656,-0.4453125,0.64453125,-0.27929688,-0.2578125,0.20898438,-0.38476562,0.14257812,0.91796875,1.4375,-0.4765625,0.045410156,0.080566406,0.25976562,0.78125,-1.4140625,-3.125,1.2578125,3.359375,0.6484375,-0.44921875,0.49609375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
index 90369015..c18d2d33 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
@@ -1,4 +1,4 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,0,0.78515625,1.0,-0.059326172,0.65234375,0.359375,-0.50390625,0.3359375,1.0859375,-0.15917969,-1.0546875,0.72265625,-0.24804688,-1.1328125,-0.8515625,-1.046875,0.14648438,-1.0859375,1.3671875,-0.43164062,0.30273438,-0.024780273,0.80078125,-0.33203125,-0.13867188,-0.61328125,-1.4765625,-2.765625,0.3125,-0.6328125,0.5078125,-1.0390625,-1.109375,-0.38671875,0.21972656,-0.63671875,-0.8984375,0.51171875,-1.265625,0.63671875,-0.10107422,-0.7578125,0.43554688,0.40039062,1.5859375,0.29882812,0.6953125,0.059570312,0.296875,-2.15625,0.80859375,1.28125,-1.2890625,-2.3125,-0.14160156,-0.92578125,-0.4453125,0.07861328,1.0078125,-0.27734375,-0.8125,-0.07714844,1.265625,-0.076660156,-0.171875,1.3046875,-0.58984375,0.33789062,-0.0034332275,0.22070312,0.94140625,0.18554688,-2.0,0.8984375,-1.1015625,-0.13476562,1.2734375,1.7578125,-0.1328125,0.22949219,-0.36328125,0.8984375,0.47265625,-0.17578125,-1.0859375,-0.46289062,1.0078125,0.07128906,-1.3203125,0.14355469,0.60546875,0.625,1.6875,0.88671875,0.59765625,-0.28125,-0.20898438,-1.1015625,0.80078125,0.12158203,0.6015625,1.6640625,0.671875,-0.90625,-0.21191406,0.5,-0.34960938,-0.9453125,-0.53125,0.42382812,-0.16308594,-1.59375,2.015625,0.078125,0.50390625,-0.359375,0.16796875,-0.4765625,1.5625,-0.3046875,1.4453125,1.46875,-1.296875,-0.78515625,-1.03125,0.025268555,0.20703125,0.3359375,-1.0546875,-2.203125,0.18457031,0.22167969,-0.78515625,-0.8515625,0.21289062,0.24511719,1.0,-0.44140625,-1.1015625,1.2421875,-1.890625,1.3515625,-0.54296875,-0.4375,-0.12060547,1.1796875,0.33398438,-0.96875,-0.115722656,2.21875,3.484375,-0.27734375,-0.74609375,-2.359375,2.203125,-0.48828125,0.23925781,0.20214844,1.265625,-1.5,0.4453125,-0.95703125,1.0546875,0.671875,1.75,-1.9375,0.234375,-1.2578125,0.828125,1.21875,-1.140625,-0.06640625,-1.0703125,1.8359375,-0.80078125,-0.48046875,0.0024719238,0.038330078,-0.7265625,1.2265625,0.17480469,-0.5625,0.31054688,1.53125,0.30273438,-0.06640625,0.43554688,1.3046875,0.50390625,0.953125,-2.96875,-0.12597656,-0.103027344,0.011413574,-1.2421875,-1.4140625,1.6640625,1.234375,-0.5390625,-0.890625,-0.50390625
-9,0,1,-0.16210938,1.3515625,-0.18554688,-0.9375,0.55859375,0.32421875,0.58203125,-0.81640625,1.578125,-1.453125,2.109375,-1.2890625,-0.4609375,-0.17871094,-0.43554688,-0.48242188,-1.0625,0.029785156,0.9296875,0.63671875,0.30664062,-0.98046875,1.2578125,0.07373047,-1.6171875,-2.0625,-0.64453125,-0.014831543,-0.08935547,1.328125,0.62109375,-0.2734375,0.703125,0.83984375,0.114746094,-0.17382812,0.59375,0.65234375,-0.30859375,0.00042533875,-0.45117188,0.56640625,1.0546875,0.90625,-0.72265625,0.99609375,0.07324219,-1.6484375,-0.703125,2.234375,0.578125,-1.109375,-0.609375,0.84765625,0.34960938,-0.515625,-0.0063476562,1.09375,0.55078125,-0.171875,0.96875,1.671875,0.20996094,0.49804688,0.99609375,0.24511719,-0.8515625,-0.90234375,0.24316406,0.9765625,-1.1015625,0.49609375,0.38476562,-0.53515625,-0.53515625,-0.34375,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.09033203,-0.040039062,2.171875,0.28320312,0.023925781,-2.203125,-0.55078125,0.671875,-0.13183594,2.203125,0.045898438,1.328125,-0.72265625,-0.11621094,-0.049804688,-1.2109375,0.20996094,-0.80859375,-1.5390625,-0.41210938,-0.6171875,1.4921875,-0.45507812,-0.54296875,1.109375,-2.046875,0.33007812,-0.21484375,-0.34179688,-0.4921875,2.109375,-0.43164062,-2.609375,-0.58203125,0.65625,-0.984375,0.84765625,-0.25,0.13378906,0.5625,-1.171875,0.026733398,-2.21875,-1.265625,2.296875,-1.46875,-0.70703125,-0.4140625,-0.38085938,1.0859375,0.07324219,-1.0546875,-0.03540039,-1.078125,2.375,-0.54296875,0.625,0.72265625,-0.8984375,1.421875,0.22167969,0.17089844,2.203125,0.06591797,-1.0703125,-0.053222656,-0.75,-1.265625,0.6484375,-0.49804688,-0.65234375,-0.68359375,0.052490234,-0.86328125,-0.05078125,0.19726562,-0.27148438,-1.2265625,2.296875,-0.69921875,-0.6484375,1.5,1.125,-2.421875,1.90625,-0.78515625,2.03125,-0.029541016,-0.14746094,-1.2421875,-0.02709961,-0.48242188,-1.125,0.6015625,0.3359375,-0.11328125,-2.6875,-0.5,-0.94140625,1.28125,0.578125,0.5078125,0.796875,-0.98828125,1.7734375,-1.28125,-0.25390625,-0.71484375,0.12695312,0.48632812,0.6640625,0.36523438,0.41796875,-0.12597656,-1.734375,0.2109375,-0.034423828,1.0,0.7265625
-9,0,2,-0.31445312,0.34179688,0.55859375,-2.0625,-0.025634766,0.67578125,1.8046875,-0.69921875,1.046875,-0.98046875,1.1171875,-1.359375,1.3671875,-0.6484375,-0.54296875,0.6953125,-0.045166016,-0.62890625,2.09375,-0.7421875,-0.78515625,-1.4140625,1.25,-0.15234375,-2.484375,-1.6171875,0.18164062,-0.7890625,-0.51171875,1.65625,0.52734375,-0.546875,-0.118652344,0.099121094,1.1015625,0.78125,-0.20898438,-0.13964844,-1.8203125,0.953125,-1.546875,0.51953125,0.94921875,0.97265625,0.6015625,1.375,-0.984375,-1.140625,-0.2265625,1.609375,-1.78125,0.25390625,-1.0078125,1.8671875,-0.25,0.625,0.20898438,0.27734375,-0.045166016,-0.34375,0.60546875,-0.045654297,-0.703125,0.52734375,0.6328125,1.4140625,-0.78515625,-1.1015625,-0.984375,2.109375,-1.1015625,-0.5703125,0.083496094,0.87109375,-0.40625,0.76953125,1.5703125,0.19335938,-0.21386719,0.22753906,1.125,0.4609375,-0.16113281,1.25,0.17089844,-0.82421875,-0.66015625,-1.2734375,-0.5703125,-0.328125,0.87109375,0.03540039,-0.83984375,-1.0,-1.546875,0.48242188,-0.5703125,-0.60546875,-0.25585938,-0.625,-1.3828125,0.0625,2.3125,1.1875,0.17480469,1.7890625,-0.92578125,-0.10107422,0.53125,-0.78515625,0.40039062,1.015625,0.90625,-2.578125,0.546875,-0.34179688,0.28710938,0.19042969,-0.76953125,0.5390625,1.75,-0.38671875,1.2421875,-2.0625,0.83203125,0.15625,-2.140625,-1.6015625,-0.19628906,0.24804688,0.7109375,0.6875,0.32421875,-1.5078125,0.578125,0.40429688,-0.25195312,0.6015625,-0.328125,-0.62890625,2.546875,1.671875,-0.25390625,1.71875,-0.875,-0.36132812,0.19042969,-0.15136719,-2.421875,0.66015625,-0.5390625,-0.140625,-0.4140625,-1.2578125,-1.234375,-0.92578125,0.33789062,-0.07519531,-0.055908203,2.09375,-2.078125,-0.67578125,1.1640625,-0.014465332,-1.6875,-0.109375,-0.17285156,1.3515625,0.024536133,0.46679688,0.34960938,0.50390625,-1.375,-0.30078125,0.035888672,-0.14941406,0.765625,-1.2890625,-0.25390625,0.111328125,-0.26953125,-1.140625,1.0546875,0.79296875,-0.640625,-0.17480469,0.3671875,0.0234375,-1.1484375,-0.17480469,-0.54296875,1.203125,1.265625,0.1328125,-0.029907227,-0.45703125,1.3359375,-0.6953125,2.046875,0.58203125
+9,0,0,0.7890625,1.0,-0.060302734,0.65234375,0.36132812,-0.49804688,0.3359375,1.0859375,-0.15917969,-1.0625,0.72265625,-0.25,-1.125,-0.859375,-1.0546875,0.1484375,-1.0859375,1.375,-0.43554688,0.29882812,-0.024902344,0.79296875,-0.33398438,-0.13867188,-0.609375,-1.4765625,-2.765625,0.3125,-0.63671875,0.51171875,-1.046875,-1.1171875,-0.38476562,0.21972656,-0.63671875,-0.8984375,0.515625,-1.2578125,0.63671875,-0.103515625,-0.7578125,0.44140625,0.40039062,1.59375,0.29882812,0.69921875,0.05444336,0.296875,-2.15625,0.8125,1.28125,-1.2890625,-2.296875,-0.14355469,-0.92578125,-0.44726562,0.07861328,1.0,-0.2734375,-0.8125,-0.07714844,1.265625,-0.072753906,-0.17089844,1.3125,-0.5859375,0.33789062,-0.0016403198,0.22070312,0.94140625,0.18261719,-2.015625,0.90234375,-1.1171875,-0.1328125,1.265625,1.765625,-0.13085938,0.23046875,-0.3671875,0.90234375,0.47265625,-0.17480469,-1.09375,-0.46289062,1.0078125,0.072265625,-1.3203125,0.14453125,0.609375,0.62109375,1.6796875,0.88671875,0.59765625,-0.27734375,-0.20898438,-1.09375,0.80078125,0.12109375,0.60546875,1.6796875,0.67578125,-0.90625,-0.2109375,0.50390625,-0.35351562,-0.94921875,-0.52734375,0.42382812,-0.1640625,-1.59375,2.015625,0.07763672,0.5,-0.36328125,0.16796875,-0.4765625,1.5703125,-0.30078125,1.4453125,1.4609375,-1.296875,-0.77734375,-1.03125,0.02331543,0.20605469,0.33789062,-1.0546875,-2.203125,0.18164062,0.22167969,-0.76953125,-0.85546875,0.21289062,0.2421875,1.0,-0.44335938,-1.1015625,1.2421875,-1.90625,1.3515625,-0.5390625,-0.44140625,-0.118652344,1.1796875,0.33789062,-0.97265625,-0.119140625,2.203125,3.46875,-0.27734375,-0.75,-2.375,2.21875,-0.48828125,0.2421875,0.203125,1.2578125,-1.4921875,0.44335938,-0.95703125,1.0625,0.66015625,1.75,-1.9296875,0.23535156,-1.265625,0.828125,1.2109375,-1.1484375,-0.06640625,-1.0703125,1.8359375,-0.80078125,-0.47851562,0.0023651123,0.03930664,-0.7265625,1.21875,0.17675781,-0.5546875,0.31054688,1.546875,0.30664062,-0.068359375,0.44335938,1.296875,0.50390625,0.94921875,-2.96875,-0.125,-0.1015625,0.0075683594,-1.2421875,-1.3984375,1.6640625,1.234375,-0.5390625,-0.88671875,-0.5078125
+9,0,1,-0.16308594,1.359375,-0.18359375,-0.94140625,0.55859375,0.32421875,0.57421875,-0.81640625,1.578125,-1.4609375,2.109375,-1.296875,-0.46289062,-0.18066406,-0.43164062,-0.48242188,-1.0546875,0.03125,0.9296875,0.640625,0.30664062,-0.984375,1.2578125,0.07373047,-1.609375,-2.046875,-0.65234375,-0.01586914,-0.09667969,1.328125,0.62890625,-0.27539062,0.703125,0.83984375,0.115722656,-0.17578125,0.59375,0.64453125,-0.3046875,-0.001373291,-0.45117188,0.5703125,1.0546875,0.90625,-0.7265625,1.0,0.07128906,-1.6484375,-0.70703125,2.25,0.578125,-1.1171875,-0.609375,0.84765625,0.34765625,-0.51171875,-0.008117676,1.09375,0.546875,-0.171875,0.96875,1.6875,0.20800781,0.5,1.0,0.24511719,-0.8515625,-0.8984375,0.24511719,0.9765625,-1.1015625,0.5,0.390625,-0.5390625,-0.53515625,-0.34179688,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.091796875,-0.037109375,2.1875,0.28320312,0.020263672,-2.21875,-0.55078125,0.671875,-0.13183594,2.203125,0.043945312,1.3203125,-0.72265625,-0.11621094,-0.04711914,-1.2109375,0.21582031,-0.81640625,-1.5390625,-0.41210938,-0.625,1.4921875,-0.45117188,-0.54296875,1.1015625,-2.03125,0.33203125,-0.21191406,-0.34179688,-0.4921875,2.109375,-0.43359375,-2.59375,-0.58203125,0.65625,-0.98046875,0.85546875,-0.24902344,0.1328125,0.55859375,-1.1796875,0.024902344,-2.21875,-1.2734375,2.28125,-1.46875,-0.70703125,-0.41796875,-0.37890625,1.0859375,0.07519531,-1.0546875,-0.040039062,-1.078125,2.34375,-0.5390625,0.62890625,0.72265625,-0.89453125,1.4296875,0.22265625,0.17089844,2.203125,0.0625,-1.078125,-0.052490234,-0.75390625,-1.265625,0.6484375,-0.5,-0.6484375,-0.68359375,0.05444336,-0.87109375,-0.05078125,0.20410156,-0.27148438,-1.234375,2.296875,-0.69921875,-0.64453125,1.5,1.140625,-2.4375,1.921875,-0.78515625,2.015625,-0.028442383,-0.14746094,-1.2421875,-0.026123047,-0.48242188,-1.1171875,0.6015625,0.33203125,-0.11328125,-2.6875,-0.49414062,-0.94140625,1.2890625,0.57421875,0.5078125,0.8046875,-1.0078125,1.7734375,-1.2734375,-0.25585938,-0.72265625,0.125,0.48828125,0.66796875,0.36523438,0.41796875,-0.12695312,-1.71875,0.2109375,-0.03466797,1.0078125,0.7265625
+9,0,2,-0.31445312,0.34179688,0.5546875,-2.0625,-0.025634766,0.671875,1.8046875,-0.69921875,1.0390625,-0.98046875,1.125,-1.3671875,1.3671875,-0.6484375,-0.54296875,0.69140625,-0.043701172,-0.62890625,2.109375,-0.734375,-0.78125,-1.4140625,1.25,-0.15429688,-2.46875,-1.6171875,0.18066406,-0.79296875,-0.515625,1.65625,0.53125,-0.546875,-0.12207031,0.099121094,1.109375,0.7734375,-0.20898438,-0.14160156,-1.8046875,0.9609375,-1.5390625,0.51953125,0.95703125,0.96875,0.59765625,1.3671875,-0.98828125,-1.1484375,-0.2265625,1.609375,-1.78125,0.25,-1.0,1.890625,-0.25195312,0.62109375,0.20996094,0.28125,-0.047851562,-0.34375,0.60546875,-0.045654297,-0.703125,0.52734375,0.6328125,1.421875,-0.78125,-1.1015625,-0.98828125,2.109375,-1.109375,-0.57421875,0.080078125,0.859375,-0.4140625,0.76953125,1.5703125,0.1953125,-0.21386719,0.22753906,1.125,0.45898438,-0.16113281,1.2578125,0.16894531,-0.82421875,-0.65625,-1.28125,-0.57421875,-0.32421875,0.8671875,0.037109375,-0.83203125,-1.0078125,-1.5390625,0.484375,-0.5703125,-0.60546875,-0.25195312,-0.625,-1.375,0.06225586,2.3125,1.1796875,0.17871094,1.78125,-0.9296875,-0.10107422,0.53125,-0.79296875,0.40039062,1.015625,0.90234375,-2.59375,0.53515625,-0.34375,0.28515625,0.19140625,-0.76953125,0.53515625,1.75,-0.3828125,1.234375,-2.0625,0.83203125,0.15625,-2.140625,-1.6171875,-0.19238281,0.24902344,0.71875,0.68359375,0.32617188,-1.515625,0.578125,0.40234375,-0.25585938,0.6015625,-0.32617188,-0.62890625,2.546875,1.671875,-0.25585938,1.7109375,-0.87109375,-0.35546875,0.1875,-0.15332031,-2.421875,0.6640625,-0.54296875,-0.13867188,-0.41210938,-1.2578125,-1.234375,-0.93359375,0.3359375,-0.07519531,-0.05419922,2.09375,-2.0625,-0.671875,1.171875,-0.014465332,-1.6875,-0.10888672,-0.17285156,1.34375,0.024536133,0.46679688,0.3515625,0.49804688,-1.375,-0.29492188,0.03564453,-0.15332031,0.765625,-1.2890625,-0.25390625,0.109375,-0.26953125,-1.140625,1.0390625,0.79296875,-0.63671875,-0.17382812,0.3671875,0.02331543,-1.1484375,-0.17089844,-0.5390625,1.203125,1.2578125,0.13671875,-0.03173828,-0.45703125,1.34375,-0.69140625,2.0625,0.5859375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
index e3eb7be7..6b50fe9d 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
@@ -1,6 +1,6 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-0,0,0,-0.5678054,0.30644566,-0.48223504,-0.37569737,-0.3008679,-0.52658165,-0.32838538,-0.8943048,0.95665604,-0.8637694,-0.3740866,0.95319647,1.8752165,-0.35496402,2.3600826,-1.429543
-0,0,1,0.78683025,-0.48157683,2.1831813,0.00026024866,0.19607927,-1.008873,0.06488745,0.6515402,-1.6177354,1.4758716,-0.756111,-0.30458173,0.38618582,0.34799105,-0.17682126,-1.7913008
-0,0,2,-1.0294193,-1.9363327,-0.10089664,-1.256774,-0.57952416,0.6647573,1.0020547,0.60435945,1.4913826,0.47637454,-0.52362984,1.2796938,-1.1521349,0.8179086,0.8369039,-0.61289996
-0,0,3,1.29303,0.5742223,-0.20042029,0.03720812,-1.1154803,0.26275542,-0.0761054,0.65055084,0.6538049,-1.9854333,-0.64701724,-1.4201233,0.422354,2.221934,-0.06508473,-0.6750595
-0,0,4,-0.14699927,-1.5303152,0.3594103,-0.9477121,-0.4332569,-0.068542324,0.5149603,0.34177327,1.5040935,0.8814393,-0.73049664,-0.66137636,-1.0386573,2.0420604,-1.2320555,1.142538
+0,0,0,-0.6006612,0.27331486,-0.34914157,-0.41479835,-0.3665798,-0.5533585,-0.14582494,-0.91373837,0.9844944,-0.95687056,-0.4577148,0.91134745,1.9406763,-0.3906231,2.3219848,-1.3473451
+0,0,1,0.82811767,-0.5268697,2.0814955,-0.004021469,0.5512644,-1.179125,-0.043747347,0.63015777,-1.3691996,1.6033845,-0.75416285,-0.42233095,0.45330113,0.27154657,-0.41820994,-1.7461473
+0,0,2,-0.84707606,-2.0609052,-0.1081782,-1.045911,-0.42114437,0.60394675,0.59907866,0.25174326,1.3924737,0.7319598,-0.9297085,1.5825528,-1.3019253,0.7942725,0.87738246,-0.1327626
+0,0,3,1.0684123,0.57142717,0.075573,0.11559761,-0.86974573,0.1249011,-0.2048351,0.22256361,0.48389286,-1.8543849,-0.93811846,-1.1520655,0.44447225,2.7120762,-0.38763985,-0.48254395
+0,0,4,-0.17874089,-1.1804047,0.6036014,-1.2056729,-0.21152717,-0.11572556,0.7904921,0.64713424,1.58864,0.2770208,-0.6086349,-0.9361127,-1.0290424,2.055118,-1.3742863,0.8467823
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
index b58246ec..19f8ea9b 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
@@ -1,6 +1,6 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-1,0,0,0.6880688,-0.49868053,1.1768215,0.9832487,-1.6379206,-0.87166226,0.13469009,2.644633,-0.3137418,-0.8415682,0.17771643,-0.33885878,-0.06802454,0.34736037,-1.01832,-0.62350124
-1,0,1,1.3108207,-1.3080025,0.34822875,1.4596325,-1.0497857,-0.16749367,0.485262,0.97075754,0.034637336,-1.7957212,1.4068959,0.78538895,-0.31832948,-0.30980903,-1.3081034,-0.5883961
-1,0,2,1.0650882,-0.09216967,0.36322507,0.18116008,-1.3025135,0.54174715,-0.482454,0.9420967,0.39991647,-2.0071194,-0.3020647,-1.1005768,0.43759206,2.1584413,0.28633898,-1.1556921
-1,0,3,1.8563068,-0.7071074,0.1579034,0.8939158,-0.0670585,-0.49880558,-1.4990184,-0.9207246,0.73329836,-0.34481865,0.627262,1.1733168,-1.3812757,-0.22482446,-1.1992402,1.4072857
-1,0,4,0.49123776,-0.65193045,2.2564845,-0.17658377,0.42154688,1.2235798,-0.052261785,-1.0662595,0.90595466,-0.9850903,-1.1610376,0.45742363,-1.7801737,-0.12341051,0.7481433,-0.5458406
+1,0,0,0.1883566,-0.51069915,1.0757822,0.8057133,-1.7141013,-1.0322051,0.8189624,2.455926,-0.086683646,-0.548176,0.2890321,-0.5181182,-0.046274558,0.71348274,-1.3307245,-0.6185085
+1,0,1,1.5297848,-1.2117101,-0.13439633,1.253659,-0.67018974,-0.5551331,0.7888967,1.1364132,-0.023180284,-1.6521875,1.607315,0.5899527,-0.32935065,-0.41081616,-1.3123547,-0.64372915
+1,0,2,0.8035976,-0.1564827,0.7187234,0.23841827,-1.041676,0.36406025,-0.5592999,0.6205595,0.22290729,-1.892453,-0.571009,-0.9368423,0.4768633,2.6517987,-0.04318322,-0.9589635
+1,0,3,1.6627725,-0.8503898,0.30536455,0.9283896,0.027027875,-0.6029172,-1.3817337,-1.0517935,0.88064855,-0.30291262,0.46670607,1.2897946,-1.2118512,-0.20403868,-1.3723617,1.4220868
+1,0,4,0.3392744,-0.80427676,2.273018,-0.6253882,0.75351644,0.86384004,0.31538412,-0.83312535,1.189355,-1.0529702,-1.1474402,0.37553895,-1.7151668,-0.13089456,0.64540154,-0.4849727
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
index 50c647c9..0eb7cf80 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
@@ -1,5 +1,5 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-2,0,0,-0.36413205,-0.6412734,-1.509915,-0.64748913,1.2593658,-1.3857695,0.8524324,0.52342945,-0.109673396,2.4144866,0.8789581,-0.38112083,0.7561693,-0.6839269,-0.28703287,-0.67817223
-2,0,1,1.2825032,-0.4427156,1.9214746,0.46229368,-1.4471735,-0.5868213,-0.8091576,-0.24864331,-1.2884412,1.7199851,0.23889081,0.040785108,-0.05055482,0.22968066,-1.4788846,0.43781894
-2,0,2,0.094167404,0.0859257,-0.5042325,0.7252808,1.3083975,-1.3224137,-0.33199427,0.13993336,0.08184377,-0.40506566,0.29009566,-2.2843225,1.9552679,1.2503077,-0.18841878,-0.93038386
-2,0,3,1.0551077,-1.1094918,0.8473594,-0.12514752,0.4117495,0.18166554,-1.1333376,-0.8610408,0.65473765,-0.5237611,0.96090883,0.9585637,-1.4609623,-0.9971072,-0.89406204,2.0541434
+2,0,0,-0.45247898,-0.6700307,-1.3980862,-0.8169967,1.1174649,-1.5679284,1.3004793,0.55547404,0.17673689,2.2888486,0.61724645,-0.30405724,0.7114118,-0.48643318,-0.40007603,-0.69985884
+2,0,1,0.96580744,-0.7271615,1.9331442,0.35922375,-1.1587405,-0.7792273,-0.84466994,-0.14708672,-1.1819658,1.8822826,0.33108777,-0.21132755,-0.08135607,0.38287959,-1.4923096,0.7527195
+2,0,2,0.3168712,0.017159618,-0.608047,0.8562339,1.3965137,-1.2637002,-0.35587037,0.14391245,0.03870829,-0.47761393,0.3009319,-2.3473392,1.9502603,0.8832216,0.0150891915,-0.9161799
+2,0,3,0.9275735,-1.1297508,1.0120924,-0.20396979,0.64359254,0.12881663,-1.0047234,-0.9111546,0.6870843,-0.58317107,0.87117267,1.0960609,-1.4761081,-0.9214843,-1.0393838,1.9111625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
index 429d9980..05af7ab7 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
@@ -1,9 +1,9 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-3,0,0,-0.40126666,-0.78331727,-1.5756154,-0.31062096,1.1928762,-1.6843582,0.7851598,0.5358782,0.060748313,2.3220642,0.983614,-0.29303667,0.58234143,-0.36522076,-0.30240387,-0.7496143
-3,0,1,-0.7336195,-1.2810793,-0.38829067,-0.4601508,-1.50845,0.5241958,0.8025443,0.6143513,0.056892652,-0.31004176,0.44623825,2.7555754,-0.0637165,0.22621281,0.6022362,-1.314328
-3,0,2,1.3642316,0.12788779,-0.20111935,-1.036736,-1.948152,-0.38286552,1.4817194,-0.9606339,0.7927826,-1.2917897,-0.020328995,0.9448982,1.0943421,0.79177094,0.19575107,-1.0141716
-3,0,3,0.86852545,-0.017761663,0.41102606,-0.18726309,-0.87506944,0.5694903,0.15854332,0.8119648,0.5742181,-2.3636305,-0.37842566,-1.5641385,0.12360405,2.1495688,0.17902005,-0.5240532
-4,0,0,1.1631641,-0.55724025,1.1540875,-1.288715,-0.25030568,0.29519114,-0.26156953,0.73843837,0.3341611,0.36159894,-2.2564068,-1.12792,-0.6742551,1.7842551,-0.06704759,0.6194069
-4,0,1,0.8389171,-0.6922904,2.0705483,0.090488724,-0.24284306,-0.52842766,-0.25121197,0.6185058,-1.8862301,1.0848103,-0.5888317,0.16298746,0.6342511,0.5813959,0.052398324,-1.9904659
-4,0,2,1.2333139,-0.8917692,-1.4689931,-0.6857654,0.11552239,0.31441647,0.30494234,1.6532247,0.5431021,0.16783965,-0.89129585,-1.6974555,0.21965057,-0.051815066,-0.7658714,1.8788624
-4,0,3,-0.026684264,-0.90918845,0.74477476,-0.18694443,-0.103529036,-0.61683804,-0.64630914,-0.77517974,1.4970412,1.2110367,2.242393,0.84123915,-0.07830191,-1.5589653,-0.5940237,-1.0396498
+3,0,0,-0.5061456,-0.7763785,-1.4411104,-0.52552813,1.0513479,-1.8644568,1.2233515,0.5684989,0.3443002,2.1877117,0.73235464,-0.2621535,0.5448747,-0.17720893,-0.4192689,-0.7051907
+3,0,1,-0.66679925,-1.3665861,-0.22060487,-0.44057253,-1.2946736,0.76740754,0.73320085,0.6129007,0.31359667,-0.85040903,0.46061102,2.6040866,-0.12284056,0.15048675,0.7167322,-1.4336061
+3,0,2,1.1156372,-0.22328217,-0.16104864,-1.1401838,-1.9943426,-0.2494216,1.7525011,-0.9386309,0.7549316,-1.3333093,0.040007867,0.77046984,1.3616405,0.5641874,0.22289011,-0.625249
+3,0,3,0.56514716,0.021596653,0.6965276,-0.2659085,-0.47595215,0.41413322,0.15071203,0.4543448,0.3236465,-2.2226622,-0.6223092,-1.3502904,0.022095501,2.6971529,-0.082981095,-0.391887
+4,0,0,1.2965182,-0.5597377,0.87963766,-1.3444073,-0.53000945,0.48690176,-0.06573936,0.76831144,0.30042756,0.42209244,-2.325122,-1.1396146,-0.7086506,1.4786966,0.15237255,0.84046805
+4,0,1,0.81161726,-0.8216032,2.040686,0.17894322,0.19341934,-0.7916967,-0.28501594,0.5449751,-1.6567922,1.2058699,-0.6466604,0.02430841,0.7775131,0.4694338,-0.07763458,-2.0113945
+4,0,2,0.89558095,-0.9144592,-1.5348337,-0.88432336,0.38815454,0.113910615,0.2328737,1.2174358,0.79367036,0.35744292,-0.73602664,-1.4213881,-0.011158632,0.26761565,-1.0480565,2.2604616
+4,0,3,0.1480164,-0.8613011,0.67656535,-0.3105859,0.05321045,-0.9875798,-0.100072786,-0.68129635,1.649064,1.3399781,2.217513,0.4515288,-0.27188015,-1.0483067,-1.1040651,-1.1851294
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
index 92904b29..28a36481 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
@@ -1,5 +1,5 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-5,0,0,-0.0975191,-0.4876295,-0.2364892,0.8899324,1.3293315,-1.4303449,-0.38584486,0.4049477,0.2341479,-0.24590422,0.4131558,-2.2018154,1.7526464,1.2278333,-0.1008542,-1.0976585
-5,0,1,0.96228456,0.5013961,-0.46397877,0.18880926,-0.21479401,0.068225846,-1.6835177,0.3138923,-0.215686,0.8992038,1.2001247,-0.7830776,0.29820076,1.9573467,-1.7968755,-1.2469637
-5,0,2,1.1223437,0.35189787,-0.4220163,-0.08106533,-1.5475112,0.14656012,-0.7290129,0.048741527,0.048301227,0.31566086,1.0508822,1.2633296,0.41796586,-1.0627791,1.3920348,-2.369732
-5,0,3,0.997344,-0.3670925,1.1623502,-1.5849966,-0.5812203,0.69678265,0.33674517,0.5580875,0.17368835,-0.32163247,-1.9439414,-1.2147049,-0.6150653,1.896507,0.09835436,0.66708636
+5,0,0,0.14931662,-0.53661007,-0.4178915,1.0650523,1.4260073,-1.4046125,-0.4454908,0.43748403,0.17220889,-0.3263942,0.46434885,-2.1899922,1.6828609,0.9274285,0.095608875,-1.1405207
+5,0,1,0.9416289,0.41291103,-0.4414328,0.27544495,-0.43491212,0.074826315,-1.4218235,0.46812102,-0.17113322,0.63309056,1.3964156,-0.91853875,0.35744792,1.9670777,-1.7801918,-1.3875116
+5,0,2,1.1062784,0.10729322,-0.4857815,-0.008219642,-1.3734647,0.07514935,-0.89442974,0.041828547,0.040950842,0.41346258,1.0223055,1.4269317,0.4183222,-1.0225612,1.4259869,-2.336927
+5,0,3,1.1709383,-0.3431763,0.86983496,-1.6041002,-0.6716347,0.9073796,0.5058333,0.52458864,0.07157109,-0.31755292,-2.013162,-1.166504,-0.6992124,1.7159256,0.378617,0.6107121
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
index 4552eb37..964edf93 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
@@ -1,10 +1,10 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-6,0,0,0.77065754,-0.76632786,1.8472377,-0.36171773,-0.12794422,-0.51205856,0.23821487,0.19354652,-2.007292,1.1207803,-0.47570035,0.12879921,0.6755318,0.19856073,1.0114113,-1.9820988
-6,0,1,1.5402608,-0.6269262,1.811596,0.5628006,-1.0015095,-1.0100083,-0.8554023,-0.5945589,-0.7188355,2.0016418,0.036225002,-0.0963106,-0.34438652,-0.04934737,-1.2410477,0.5716474
-6,0,2,0.099329084,-0.9905449,0.37024456,-0.092512354,-1.0474453,0.5579103,-1.3993762,0.9833203,0.59650284,1.0684012,0.9790155,-2.0937629,0.30437842,-0.46602613,1.7425917,-0.64842325
-6,0,3,0.39516813,-0.6314403,1.4948617,1.0518291,-1.0769643,-1.1019645,0.72841096,0.8454386,-0.1329276,-0.27656975,-1.1176367,-0.6240037,0.08811874,-0.8709323,-1.0424569,2.2322922
-6,0,4,1.5397077,-1.1038709,0.17057961,-0.70247877,0.91511214,-0.19965753,0.10995354,-0.09201375,-0.5179603,-1.2657489,-0.9364169,0.99344826,-0.5904373,2.378038,-1.0655441,0.35856313
-7,0,0,-0.38164693,-1.3991219,-0.57864404,-0.64194506,-1.0146283,0.77735037,1.0805509,0.20559101,0.14912546,0.24122998,0.23987257,2.7745647,-0.1806378,-0.38417065,0.53943276,-1.4561769
-7,0,1,0.7505255,-1.2829787,-0.04241756,1.1337081,0.5512444,1.0470235,-2.240779,0.38265908,-0.9141051,0.9473515,1.0557507,-1.5054818,-0.0026612426,0.71467644,-0.11884089,-0.48391303
-7,0,2,1.4440962,0.640023,-1.0184404,1.4597166,-1.5533735,0.19580375,-0.73750097,0.17983381,0.6347194,1.3996398,0.19554296,-1.8371892,0.27592623,0.28698903,-0.50887656,-1.1177573
-7,0,3,1.6090622,-0.552633,2.0960214,0.36178717,-1.3726065,-0.61812073,-0.5444809,-0.8581807,-1.0325772,1.5600145,0.2399627,-0.15540515,-0.32274023,0.06386035,-1.1094363,0.61451924
+6,0,0,0.75977767,-0.8941403,1.8381127,-0.2947546,0.2300058,-0.83663684,0.25620785,0.12637803,-1.7378011,1.182791,-0.50745416,0.036903758,0.838191,0.390844,0.6659436,-2.1043644
+6,0,1,1.2344869,-0.87115544,1.6973119,0.45785195,-0.79289585,-1.1952786,-0.7739234,-0.41190162,-0.6169048,2.0888212,0.17353193,-0.35224319,-0.42837173,0.29140207,-1.3883605,0.87579113
+6,0,2,-0.04716062,-1.1540091,0.7190903,0.21688245,-1.0973245,0.6634729,-1.0191848,1.0636467,0.86636645,0.95576775,0.40887043,-2.1949317,0.34050336,-0.5335777,1.6200428,-0.8434403
+6,0,3,0.5315266,-0.75266963,1.6877258,0.7681073,-0.9721382,-1.0216631,0.68959737,1.0582479,0.28320122,-0.731182,-0.9680603,-0.9397336,0.115028106,-0.47067833,-1.2997086,1.9737223
+6,0,4,1.4608288,-0.97320133,0.2406221,-0.8215309,0.9401043,-0.24250522,0.4038392,0.020989973,-0.51842433,-1.293551,-0.90764123,0.74466497,-0.6153507,2.456806,-1.1631211,0.23413561
+7,0,0,-0.16067488,-1.4088781,-0.52943593,-0.53733695,-0.881949,1.020826,0.9473788,0.22936635,0.48022568,-0.3948946,0.26959687,2.6491475,-0.31377748,-0.2948756,0.564145,-1.6736668
+7,0,1,0.7797228,-1.711448,-0.27516288,1.2757864,0.5518668,1.0981587,-2.2466977,0.15571526,-0.89171106,0.9792887,0.612105,-1.1947432,0.38149735,0.55929846,0.13911384,-0.20466185
+7,0,2,1.4645267,0.6052429,-0.6147724,1.2570165,-1.6225443,-0.0040931324,-0.46623838,0.37951505,0.81095076,1.1644993,0.37510827,-2.0000474,-0.28104275,0.7649079,-0.69417703,-1.187673
+7,0,3,1.3395354,-0.7207316,2.0857847,0.21226466,-1.0327022,-0.8066197,-0.42594796,-0.64940375,-1.1074684,1.6851654,0.36267513,-0.4739317,-0.4367329,0.49596614,-1.2793999,0.7298483
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
index 9c9c4f81..a62eae9c 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
@@ -1,6 +1,6 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-8,0,0,-0.32614353,-0.64439976,-0.4606882,0.65944755,1.3359828,-1.0852644,-0.16756222,0.14784002,-0.08113572,-0.6550381,0.7323079,-2.066391,1.9876616,1.0655345,0.63813996,-1.1135746
-8,0,1,0.12641025,-0.56098837,0.40993986,0.107800655,0.32741314,-0.5968289,-0.84562564,-0.6825451,1.2423956,1.8815519,1.8854672,0.5606451,-0.055782493,-1.6092077,-1.2014889,-0.98526716
-8,0,2,-0.41545722,-0.6968768,-1.6478468,-0.6066328,0.9194668,-1.1579491,0.6114348,0.7974771,-0.52014804,2.3043096,1.1645346,-0.0551654,0.95790863,-0.3786852,-0.23132549,-1.0501995
-8,0,3,1.0761181,0.35165632,-0.40121642,-0.17262518,-1.2321544,0.42185432,-0.7928826,-0.034794606,-0.17338482,0.422188,1.1260121,1.2503253,0.30318612,-1.3496894,1.4538991,-2.2986958
-8,0,4,1.1404579,-1.0539529,0.6959456,0.0002799956,0.50083333,0.25582743,-1.1861986,-0.89913344,0.5945992,-0.5446149,0.99352646,1.0013199,-1.5414313,-0.9312056,-0.95023686,1.9430875
+8,0,0,-0.13603266,-0.7307862,-0.52509326,0.74821085,1.3972374,-1.0351514,-0.087998375,0.14008258,-0.16702926,-0.7757859,0.767171,-2.0736523,1.9220414,0.7747719,0.8359873,-1.0989997
+8,0,1,0.2691684,-0.58000475,0.22857264,0.0726761,0.36003938,-0.88419944,-0.48701763,-0.6307923,1.4697176,1.9778647,1.8036274,0.28785348,-0.23820582,-1.2125394,-1.566934,-0.87618756
+8,0,2,-0.5361712,-0.7612994,-1.4328666,-0.8012423,0.8185195,-1.415421,1.1227995,0.81984663,-0.20334002,2.2054589,0.822599,-0.029563703,1.0102043,-0.28044125,-0.27897397,-1.0894494
+8,0,3,1.0546969,0.12370178,-0.39167616,-0.19105366,-1.0928397,0.34877908,-0.921601,0.05444164,-0.1952575,0.56712765,1.1011527,1.3574673,0.31019345,-1.4148287,1.472835,-2.2225912
+8,0,4,1.0152535,-1.0647243,0.87573284,-0.06597459,0.7320688,0.2022493,-1.0524291,-0.9378761,0.6166235,-0.6034941,0.89612216,1.1415883,-1.5616627,-0.86831933,-1.1108083,1.7946959
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
index 8e68ce40..64e66679 100644
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
+++ b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
@@ -1,5 +1,5 @@
 sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-9,0,0,-0.47500592,0.4115684,-0.3617759,-0.28399315,0.3019857,-0.5060671,-0.6323188,-0.9995569,0.9944466,-0.5976193,-0.70640665,0.9123807,1.9063975,-0.45425025,2.1027493,-1.6554946
-9,0,1,-0.086714655,-0.97823036,0.05636277,-0.01605165,-0.9614703,0.6308942,-1.2854741,1.0908369,0.45563138,0.926207,1.0300418,-2.0144224,0.2822436,-0.24596272,1.959498,-0.8820346
-9,0,2,-0.4039917,-0.86205673,-1.1582229,-1.341583,-0.120875224,0.73452944,1.7072109,-0.28952715,1.6449419,0.5135223,-0.9904969,1.0184028,-1.2944711,1.0233638,0.47282502,-0.68130696
-9,0,3,0.4459618,-1.5170839,-0.057393685,0.48917928,1.573972,1.3741968,-1.1912827,-0.21140826,-0.8198331,0.22414537,1.1317799,-2.0740838,-0.63661385,0.7234822,0.49297914,0.047913294
+9,0,0,-0.4572789,0.4083919,-0.31473643,-0.2729684,0.24014777,-0.4612613,-0.54261976,-1.0134654,1.0257438,-0.6418706,-0.78112227,0.8925522,1.8929882,-0.6186886,2.1505845,-1.5657275
+9,0,1,-0.1670796,-1.1557673,0.45290652,0.3810859,-0.8798305,0.950252,-1.0811738,0.9893472,0.62552804,0.80969876,0.35180464,-1.9200974,0.22418688,-0.624056,2.0414052,-1.0304397
+9,0,2,-0.20206067,-0.93332684,-1.3126967,-0.8317151,0.009184291,0.7905838,1.1029824,-0.8864819,1.4854951,0.84356177,-1.2966992,1.5094204,-1.4029254,0.9184137,0.5197256,-0.3381626
+9,0,3,0.54572934,-1.7311612,-0.3354268,0.58204854,1.634565,1.6117706,-1.2522428,-0.40521324,-1.0268775,0.31409565,0.66306615,-1.6557356,-0.32661834,0.54005146,0.6954313,0.15141636
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
index 8cccbef6..c0c4c994 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId
-0,0,[unknown]
+0,0,116
 0,1,125
-0,2,111
+0,2,116
 0,3,102
-0,4,118
+0,4,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
index 4bd96488..6f05ec83 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId
 1,0,100
-1,1,120
+1,1,124
 1,2,102
 1,3,119
-1,4,122
+1,4,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
index 7cd4f7cd..d7879885 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
@@ -1,5 +1,5 @@
 sequenceId,itemPosition,itemId
-2,0,125
-2,1,115
+2,0,123
+2,1,125
 2,2,125
 2,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
index ae179237..2b1e4f45 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId
 3,0,123
-3,1,111
+3,1,118
 3,2,119
 3,3,102
 4,0,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
index c8b395dc..df93b1d1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
@@ -1,5 +1,5 @@
 sequenceId,itemPosition,itemId
 5,0,125
 5,1,100
-5,2,111
+5,2,119
 5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
index d04d1d0e..0ba233fe 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
@@ -1,10 +1,10 @@
 sequenceId,itemPosition,itemId
 6,0,125
-6,1,115
+6,1,103
 6,2,113
-6,3,115
-6,4,114
-7,0,111
+6,3,120
+6,4,102
+7,0,118
 7,1,125
 7,2,129
-7,3,115
+7,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
index 7143ade0..04a51b82 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
@@ -1,6 +1,6 @@
 sequenceId,itemPosition,itemId
 8,0,125
 8,1,113
-8,2,125
-8,3,117
+8,2,123
+8,3,115
 8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
index d379af64..7b4ec0b2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
@@ -1,5 +1,5 @@
 sequenceId,itemPosition,itemId
 9,0,116
 9,1,113
-9,2,111
+9,2,114
 9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
index 59349bc1..8eb4c4a0 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId
-8,4,102
+8,4,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
index ed59d730..10a61d90 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,4,109,[other],0.00014168024
+0,4,109,[other],-0.007879399
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
index 98c7c6b4..6989e535 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,4,102,[unknown],-0.09726027
+1,4,102,[unknown],-0.09629029
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
index 461de44b..f70f3d04 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,3,[mask],[unknown],-0.12791729
+2,3,[mask],[unknown],-0.1264407
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
index 8cb7700d..0b7fcd80 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,3,103,4,0.043316968
-4,3,110,4,0.032937437
+3,3,103,4,0.035191707
+4,3,110,2,0.025370859
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
index 02507d44..4300c9d3 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,3,110,9,-0.075533785
+5,3,110,9,-0.07751563
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
index 93e77202..d82f9c83 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,4,122,7,-0.04820269
-7,3,113,[other],0.057539806
+6,4,122,7,-0.05601125
+7,3,113,[other],0.044996656
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
index c2229265..7ed5dffb 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,4,110,[mask],-0.095947556
+8,4,110,[mask],-0.10512776
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
index ced3fa21..418d29d1 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,3,114,6,-0.0014844015
+9,3,114,6,-0.004389014
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
index 626180f8..c21215f2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-0,4,102,4,0.016973361
+0,4,102,[other],0.006249614
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
index 35dc1d9f..c249c329 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-1,4,100,[unknown],-0.09465362
+1,4,100,[unknown],-0.09367853
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
index c0b5fba4..57aca839 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-2,3,100,[unknown],-0.12798095
+2,3,100,[unknown],-0.12620808
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
index 5fc0d368..fdadb1a4 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-3,3,111,4,0.058097668
-4,3,128,4,0.047720857
+3,3,111,4,0.047561854
+4,3,128,4,0.038341478
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
index 8901171a..3ec68b25 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-5,3,102,6,-0.06665557
+5,3,100,6,-0.069281995
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
index 5c04fea5..c01739fe 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
@@ -1,3 +1,3 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-6,4,110,7,-0.029965512
-7,3,102,4,0.083941795
+6,4,110,7,-0.04092831
+7,3,102,4,0.06709013
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
index 30927882..e00e6b7c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-8,4,128,0,-0.07864175
+8,4,128,[mask],-0.09031482
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
index 69f195e6..7b52a58c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
@@ -1,2 +1,2 @@
 sequenceId,itemPosition,itemId,supCat1,supReal3
-9,3,125,6,0.00074065477
+9,3,125,6,-0.0026362315
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
index fe0245bd..73c62198 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
@@ -1,201 +1,201 @@
 sequenceId,itemPosition,itemValue
-0,21,0.03249278
-0,22,0.1493005
-0,23,0.12733665
-0,24,0.21319532
-0,25,0.16727091
-0,26,0.31602606
-0,27,0.15728734
-0,28,-0.1362295
-0,29,-0.1362295
-0,30,-0.17217033
-0,31,-0.13922456
-0,32,-0.15020649
-0,33,0.022259623
-0,34,-0.0992903
-0,35,-0.11825907
-0,36,-0.08631166
-0,37,0.029248118
-0,38,-0.1661802
-0,39,0.27209836
-0,40,-0.037392184
-1,19,-0.04438068
-1,20,0.23515917
-1,21,0.16327749
-1,22,-0.0032530688
-1,23,0.2411493
-1,24,0.0893991
-1,25,0.12084734
-1,26,-0.1751654
-1,27,0.12134651
-1,28,-0.15719499
-1,29,0.04671936
-1,30,-0.05860726
-1,31,-0.051868357
-1,32,0.06993115
-1,33,0.051711142
-1,34,-0.08631166
-1,35,-0.015865121
-1,36,-0.17117198
-1,37,-0.057359315
-1,38,0.015770305
-2,17,0.04996402
-2,18,-0.11176976
-2,19,0.17425941
-2,20,0.19422655
-2,21,-0.021293685
-2,22,-0.021044096
-2,23,0.1078687
-2,24,0.077918
-2,25,0.02899853
-2,26,-0.13922456
-2,27,-0.1751654
-2,28,-0.021792863
-2,29,0.09838431
-2,30,0.19722162
-2,31,-0.20611446
-2,32,-0.009344604
-2,33,-0.08481413
-2,34,-0.038140953
-2,35,0.078916356
-2,36,-0.124748394
-3,17,0.057451695
-3,18,0.2780885
-3,19,0.21020025
-3,20,0.30005237
-3,21,0.23016739
-3,22,0.028000174
-3,23,0.12084734
-3,24,0.17825285
-3,25,-0.114764825
-3,26,-0.0601048
-3,27,-0.07183549
-3,28,-0.037392184
-3,29,-0.13822621
-3,30,-0.13922456
-3,31,-0.101287015
-3,32,-0.0076598767
-3,33,-0.05935603
-3,34,-0.054364245
-3,35,0.04047963
-3,36,-0.048623696
-4,14,0.20820354
-4,15,0.16227913
-4,16,-0.17716211
-4,17,-0.08032152
-4,18,0.02600346
-4,19,0.14830214
-4,20,0.07242704
-4,21,-0.089306735
-4,22,-0.15020649
-4,23,0.036236614
-4,24,-0.27100763
-4,25,0.100381024
-4,26,-0.024039164
-4,27,0.19322819
-4,28,-0.09879112
-4,29,0.008969001
-4,30,-0.05860726
-4,31,-0.14421634
-4,32,0.036236614
-4,33,0.1403153
-5,14,0.26610824
-5,15,0.41586173
-5,16,0.30803922
-5,17,0.21519203
-5,18,0.2601181
-5,19,0.16128078
-5,20,-0.023040809
-5,21,0.027001817
-5,22,-0.16418348
-5,23,-0.0058503556
-5,24,-0.109273866
-5,25,-0.12574674
-5,26,0.13232844
-5,27,0.028499352
-5,28,0.02600346
-5,29,0.008157835
-5,30,0.042975523
-5,31,-0.07882399
-5,32,-0.25703064
-5,33,-0.2350668
-6,18,0.40987158
-6,19,0.11136295
-6,20,0.17126435
-6,21,0.30204907
-6,22,0.20720518
-6,23,0.07941554
-6,24,-0.028157387
-6,25,0.024006747
-6,26,0.015770305
-6,27,-0.13922456
-6,28,0.02575387
-6,29,-0.0043060225
-6,30,-0.08181906
-6,31,-0.052367534
-6,32,0.051211964
-6,33,-0.12025579
-6,34,0.056702927
-6,35,0.19222984
-6,36,0.017392633
-6,37,-0.1522032
-7,15,0.156289
-7,16,0.29805565
-7,17,-0.03289958
-7,18,0.16128078
-7,19,0.06943197
-7,20,-0.057608906
-7,21,-0.07283385
-7,22,0.1952249
-7,23,-0.032275606
-7,24,-0.06484699
-7,25,0.066936076
-7,26,0.023757158
-7,27,-0.13722785
-7,28,-0.057858497
-7,29,0.036236614
-7,30,-0.096794404
-7,31,-0.03864013
-7,32,-0.14721142
-7,33,0.06444018
-7,34,0.033740725
-8,20,-0.012370872
-8,21,0.0412284
-8,22,0.21419367
-8,23,-0.143218
-8,24,0.060696352
-8,25,-0.04762534
-8,26,0.114358015
-8,27,0.07741882
-8,28,0.09738596
-8,29,0.058699638
-8,30,-0.040137667
-8,31,-0.0878092
-8,32,0.059947584
-8,33,0.04147799
-8,34,0.05245991
-8,35,0.10137938
-8,36,-0.05411466
-8,37,-0.054613836
-8,38,-0.034397114
-8,39,0.03848292
-9,16,0.066936076
-9,17,0.25113288
-9,18,0.28807208
-9,19,-0.19213746
-9,20,0.060696352
-9,21,0.2092019
-9,22,0.29406223
-9,23,-0.12424921
-9,24,-0.062351096
-9,25,0.00085735274
-9,26,-0.11626236
-9,27,-0.065845355
-9,28,0.27010167
-9,29,0.0051627657
-9,30,-0.01287005
-9,31,-0.109273866
-9,32,-0.097293586
-9,33,-0.055861782
-9,34,0.033740725
-9,35,-0.10228537
+0,21,-0.21923348
+0,22,0.26796454
+0,23,0.35781664
+0,24,0.32586923
+0,25,0.18709765
+0,26,0.16812888
+0,27,0.12320283
+0,28,-0.028172985
+0,29,0.014381966
+0,30,0.02785978
+0,31,-0.052008748
+0,32,-0.047266554
+0,33,-0.11041261
+0,34,0.052818693
+0,35,0.25398755
+0,36,-0.13537154
+0,37,-0.14335838
+0,38,0.0024016856
+0,39,-0.069979176
+0,40,-0.10791672
+1,19,0.35182652
+1,20,0.20307136
+1,21,0.3678002
+1,22,0.27595142
+1,23,0.0662965
+1,24,-0.17031401
+1,25,0.13717982
+1,26,0.14117326
+1,27,0.04607979
+1,28,-0.21623841
+1,29,0.12370201
+1,30,-0.033539154
+1,31,-0.09843233
+1,32,-0.044271484
+1,33,-0.10042905
+1,34,0.04008965
+1,35,0.18410258
+1,36,-0.11740111
+1,37,0.06030637
+1,38,-0.0854537
+2,17,0.1561486
+2,18,0.087262
+2,19,0.052818693
+2,20,0.011574087
+2,21,0.030106083
+2,22,0.29392183
+2,23,0.07827678
+2,24,0.030979645
+2,25,0.15115681
+2,26,0.08526528
+2,27,-0.1313781
+2,28,0.02761019
+2,29,-0.030668877
+2,30,0.12270366
+2,31,-0.05175916
+2,32,0.049574036
+2,33,-0.09593645
+2,34,-0.089447126
+2,35,-0.21124664
+2,36,-0.041026827
+3,17,0.087262
+3,18,0.32986265
+3,19,0.13418475
+3,20,-0.038530935
+3,21,0.1401749
+3,22,0.23002699
+3,23,-0.011762499
+3,24,0.08975788
+3,25,0.0343491
+3,26,0.03709458
+3,27,-0.11390686
+3,28,-0.023680381
+3,29,-0.0050235917
+3,30,0.15215518
+3,31,-0.044271484
+3,32,-0.16931565
+3,33,-0.006021948
+3,34,0.08127186
+3,35,-0.088947944
+3,36,-0.15933208
+4,14,0.14316997
+4,15,0.00483518
+4,16,-0.09443891
+4,17,0.008454223
+4,18,-0.041526005
+4,19,-0.032540794
+4,20,0.07727843
+4,21,0.086263634
+4,22,-0.11191015
+4,23,0.05057239
+4,24,0.04258554
+4,25,0.06579733
+4,26,-0.015506336
+4,27,0.10872666
+4,28,0.1721223
+4,29,-0.021933256
+4,30,0.2230385
+4,31,-0.07097753
+4,32,0.039091293
+4,33,-0.1383666
+5,14,-0.14635345
+5,15,0.21105821
+5,16,0.21804671
+5,17,0.31388897
+5,18,0.16912724
+5,19,0.16413546
+5,20,0.12470037
+5,21,0.124201186
+5,22,-0.1084159
+5,23,0.03360033
+5,24,0.033849917
+5,25,-0.018314214
+5,26,-0.029545726
+5,27,-0.0031516731
+5,28,0.13618147
+5,29,0.038841702
+5,30,0.043583896
+5,31,0.16014202
+5,32,-0.050261624
+5,33,-0.022182846
+6,18,0.18310423
+6,19,0.15015846
+6,20,0.19208944
+6,21,0.072785825
+6,22,0.23102535
+6,23,0.04008965
+6,24,-0.07297424
+6,25,-0.048264913
+6,26,0.008267031
+6,27,-0.15134524
+6,28,-0.12339125
+6,29,-0.12039618
+6,30,-0.036284633
+6,31,-0.08794959
+6,32,0.030230876
+6,33,-0.07596931
+6,34,-0.10941426
+6,35,-0.009017018
+6,36,-0.18528937
+6,37,-0.08844877
+7,15,0.17511737
+7,16,0.2090615
+7,17,-0.112908505
+7,18,0.0022612917
+7,19,0.1721223
+7,20,-0.036534224
+7,21,0.13917653
+7,22,-0.15833373
+7,23,0.08776117
+7,24,-0.03453751
+7,25,-0.004649208
+7,26,0.0055839475
+7,27,0.096746385
+7,28,-0.053755872
+7,29,0.057560887
+7,30,-0.062491495
+7,31,0.11321927
+7,32,0.01725224
+7,33,0.18809602
+7,34,-0.042274773
+8,20,0.26397112
+8,21,0.18410258
+8,22,0.14316997
+8,23,0.104234055
+8,24,0.19109108
+8,25,0.2230385
+8,26,-0.110911794
+8,27,-0.09992987
+8,28,0.19708122
+8,29,-0.10941426
+8,30,0.015255528
+8,31,-0.022182846
+8,32,-0.20325978
+8,33,-0.18928279
+8,34,0.08276939
+8,35,-0.032540794
+8,36,-0.11340769
+8,37,-0.028547369
+8,38,0.03634581
+8,39,0.18410258
+9,16,-0.004961194
+9,17,0.0662965
+9,18,0.17711408
+9,19,0.31588566
+9,20,0.35781664
+9,21,0.34383965
+9,22,0.18410258
+9,23,-0.024054764
+9,24,0.14416832
+9,25,-0.063989036
+9,26,-0.0071450993
+9,27,0.057560887
+9,28,0.036096223
+9,29,-0.17031401
+9,30,-0.019062981
+9,31,0.012322854
+9,32,0.058309656
+9,33,-0.09044548
+9,34,0.014756349
+9,35,-0.03378874
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
index 1cf52011..88320a94 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,0.2900688
-0,9,-0.17616376
-0,10,0.3799209
-0,11,0.26411152
-0,12,-0.3009583
-1,8,0.28208193
-1,9,-0.0039628376
-1,10,0.33798993
-2,8,-0.14820977
-2,9,0.11735309
-3,8,0.36993733
-3,9,0.29805565
-4,7,-0.13523114
-5,7,0.14131364
-6,8,0.3359932
-6,9,0.26810494
-6,10,0.31402937
-7,8,-0.15419991
-8,8,0.16826928
-8,9,-0.25703064
-8,10,0.41586173
-8,11,0.40787488
-9,8,0.31402937
-9,9,0.25812137
+0,8,0.2899284
+0,9,-0.1753058
+0,10,0.37778378
+0,11,0.26596785
+0,12,-0.30309543
+1,8,0.28194156
+1,9,-0.00092096993
+1,10,0.33784953
+2,8,-0.15034688
+2,9,0.118211046
+3,8,0.3678002
+3,9,0.29791525
+4,7,-0.13537154
+5,7,0.1401749
+6,8,0.3358528
+6,9,0.26996127
+6,10,0.31388897
+7,8,-0.15533866
+8,8,0.16812888
+8,9,-0.25717103
+8,10,0.41771805
+8,11,0.40773448
+9,8,0.31388897
+9,9,0.257981
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
index 68074514..fbc8a7dc 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.19088916
-0,9,-0.078307025
-0,10,0.020881303
-0,11,0.13500652
-0,12,0.30980512
-1,8,0.0876035
-1,9,-0.026953768
-1,10,-0.2353295
-2,8,0.046619654
-2,9,-0.18891405
-3,8,0.08957863
-3,9,0.17450903
-4,7,0.37695938
-5,7,-0.18002598
-6,8,0.2999295
-6,9,0.33745688
-6,10,0.007867463
-7,8,0.66730285
-8,8,0.20117323
-8,9,0.19031003
-8,10,-0.03213847
-8,11,-0.012880999
-9,8,0.0023336397
-9,9,-0.2531056
+0,8,-0.19301088
+0,9,-0.08141631
+0,10,0.019747147
+0,11,0.13387236
+0,12,0.30965853
+1,8,0.08696313
+1,9,-0.027594142
+1,10,-0.23646365
+2,8,0.047213733
+2,9,-0.18807307
+3,8,0.08844448
+3,9,0.17436244
+4,7,0.3787879
+5,7,-0.17622232
+6,8,0.2997829
+6,9,0.33731028
+6,10,0.009555192
+7,8,0.6651811
+8,8,0.1990515
+8,9,0.19016345
+8,10,-0.03203817
+8,11,-0.012780699
+9,8,0.0013537919
+9,9,-0.2532522
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
index 7c11814b..699aaf44 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.07564883
-0,9,-0.34528795
-0,10,-0.06342901
-0,11,-0.22871205
-0,12,-0.23238373
-1,8,-0.2268762
-1,9,-0.10984136
-1,10,-0.016901925
-2,8,-0.3618105
-2,9,-0.29939193
-3,8,-0.1360021
-3,9,0.23850942
-4,7,0.104493044
-5,7,-0.3471238
-6,8,0.09990344
-6,9,-0.11305408
-6,10,0.15406075
-7,8,0.12652314
-8,8,-0.17822644
-8,9,-0.11075929
-8,10,-0.29755607
-8,11,-0.29204854
-9,8,-0.14472234
-9,9,-0.019196726
+0,8,-0.07336837
+0,9,-0.34461385
+0,10,-0.06304177
+0,11,-0.2298738
+0,12,-0.23170963
+1,8,-0.22712004
+1,9,-0.109626226
+1,10,-0.01875211
+2,8,-0.3611364
+2,9,-0.30055368
+3,8,-0.13670488
+3,9,0.23918352
+4,7,0.103331305
+5,7,-0.3464497
+6,8,0.099659614
+6,9,-0.11192103
+6,10,0.15381691
+7,8,0.12719724
+8,8,-0.1766344
+8,9,-0.10939674
+8,10,-0.29688197
+8,11,-0.29137447
+9,8,-0.14358929
+9,9,-0.018981587
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
index 22ba483f..7436f650 100644
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
@@ -1,25 +1,25 @@
 sequenceId,itemPosition,itemValue
-0,8,-0.08269817
-0,9,-0.22038624
-0,10,-0.060151752
-0,11,-0.066634566
-0,12,-0.0973849
-1,8,-0.055131875
-1,9,-0.02945879
-1,10,-0.12446355
-2,8,-0.25710306
-2,9,-0.24058047
-3,8,0.01506035
-3,9,0.37075448
-4,7,0.3579036
-5,7,-0.12813523
-6,8,0.3395452
-6,9,0.2771266
-6,10,0.31567925
-7,8,0.35423192
-8,8,-0.039154325
-8,9,-0.015919466
-8,10,-0.053353403
-8,11,-0.12629938
-9,8,-0.0007737763
-9,9,0.19084209
+0,8,-0.07899063
+0,9,-0.21977666
+0,10,-0.058853757
+0,11,-0.06705767
+0,12,-0.0972343
+1,8,-0.05463706
+1,9,-0.029308192
+1,10,-0.12614879
+2,8,-0.25741142
+2,9,-0.24088883
+3,8,0.014293022
+3,9,0.37136406
+4,7,0.360349
+5,7,-0.1284436
+6,8,0.3419906
+6,9,0.27773616
+6,10,0.31812465
+7,8,0.35667732
+8,8,-0.038143177
+8,9,-0.012097187
+8,10,-0.054866537
+8,11,-0.12614879
+9,8,0.0005242191
+9,9,0.19145167
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-0-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-1-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-2-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-3-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-4-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-5-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-6-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv
similarity index 100%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-predictions/sequifier-test-hp-search-custom-eval-run-7-best-3-7-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv
similarity index 88%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv
index 830dae7f..f02aeba2 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-0-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv
@@ -8,9 +8,9 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 0,10,116,1,9,7,[other]
 0,11,104,6,9,7,[other]
 0,12,104,1,9,7,[other]
-0,13,104,1,9,7,[other]
-0,14,102,1,9,[unknown],[other]
-0,15,104,6,9,[unknown],[other]
+0,13,102,1,9,7,[other]
+0,14,104,6,9,7,[other]
+0,15,104,6,9,7,[other]
 0,16,104,6,9,7,[other]
 0,17,104,6,9,7,[other]
 0,18,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv
similarity index 88%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv
index 8f038e88..756d6e7c 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-1-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv
@@ -6,12 +6,12 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 1,8,111,1,9,7,[other]
 1,9,111,1,9,7,[other]
 1,10,102,1,9,7,[other]
-1,11,129,1,9,[unknown],[other]
+1,11,129,1,9,7,[other]
 1,12,102,1,9,[unknown],[other]
 1,13,104,6,9,[unknown],[other]
 1,14,104,6,9,[unknown],[other]
-1,15,104,6,9,[unknown],[other]
-1,16,104,6,9,[unknown],[other]
+1,15,104,6,9,7,[other]
+1,16,104,6,9,7,[other]
 1,17,104,6,9,7,[other]
 1,18,104,6,9,7,[other]
 1,19,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv
similarity index 51%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv
index a3056096..50574151 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-2-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv
@@ -1,20 +1,20 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,115,9,2,2,[unknown]
-2,4,118,9,2,7,[mask]
-2,5,115,9,9,7,[mask]
-2,6,115,9,9,7,[mask]
-2,7,115,9,9,7,[mask]
-2,8,112,9,9,7,[other]
-2,9,115,9,9,7,[other]
-2,10,104,9,9,7,[other]
-2,11,112,6,9,7,[other]
+2,3,115,9,2,2,2
+2,4,115,9,2,2,[mask]
+2,5,118,9,2,7,[mask]
+2,6,112,9,9,7,0
+2,7,112,9,9,2,[unknown]
+2,8,115,9,9,7,[other]
+2,9,112,9,9,7,[other]
+2,10,112,9,9,7,1
+2,11,112,9,9,7,[other]
 2,12,112,9,9,7,[other]
 2,13,112,9,9,7,[other]
 2,14,112,9,9,7,[other]
 2,15,112,9,9,7,[other]
-2,16,112,9,9,7,[other]
-2,17,112,9,9,7,[other]
-2,18,112,9,9,8,[other]
+2,16,112,9,9,7,0
+2,17,112,9,9,8,[other]
+2,18,112,9,9,7,[other]
 2,19,112,9,9,7,[other]
 2,20,112,9,9,7,[other]
 2,21,112,9,9,7,[other]
@@ -22,10 +22,10 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 2,23,112,9,9,7,[other]
 2,24,112,9,9,7,[other]
 2,25,112,9,9,7,[other]
-2,26,112,9,9,8,[other]
+2,26,112,9,9,7,0
 2,27,112,9,9,7,[other]
-2,28,112,9,9,7,[other]
+2,28,112,9,9,7,0
 2,29,112,9,9,7,[other]
-2,30,112,9,9,7,[other]
+2,30,112,9,9,7,0
 2,31,112,9,9,7,[other]
-2,32,112,9,9,7,[other]
+2,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv
similarity index 52%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv
index 861690e2..08e5981d 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-3-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv
@@ -1,61 +1,61 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,112,9,2,2,[mask]
-3,4,112,9,2,[mask],[mask]
-3,5,115,9,2,9,[other]
-3,6,112,9,2,2,[other]
-3,7,112,9,2,2,[mask]
-3,8,112,9,9,2,[other]
-3,9,112,9,2,2,[other]
-3,10,112,9,9,7,[other]
+3,3,112,9,2,2,2
+3,4,115,9,2,9,[mask]
+3,5,118,9,2,2,[mask]
+3,6,112,9,2,2,0
+3,7,112,9,9,2,[other]
+3,8,112,9,2,2,[other]
+3,9,112,9,2,7,[mask]
+3,10,112,9,9,7,0
 3,11,112,9,9,7,[other]
-3,12,112,9,9,7,[other]
+3,12,112,9,9,7,0
 3,13,112,9,9,7,[other]
 3,14,112,9,9,7,[other]
 3,15,112,9,9,7,[other]
-3,16,112,9,9,7,[other]
+3,16,112,9,9,7,0
 3,17,112,9,9,7,[other]
-3,18,112,9,9,8,[other]
-3,19,112,9,9,7,[other]
+3,18,112,9,9,7,[other]
+3,19,112,9,9,7,0
 3,20,112,9,9,7,[other]
-3,21,112,9,9,7,[other]
+3,21,112,9,9,7,0
 3,22,112,9,9,7,[other]
 3,23,112,9,9,7,[other]
 3,24,112,9,9,7,[other]
-3,25,112,9,9,7,[other]
-3,26,112,9,9,8,[other]
+3,25,112,9,9,7,0
+3,26,112,9,9,7,[other]
 3,27,112,9,9,7,[other]
-3,28,112,9,9,7,[other]
+3,28,112,9,9,7,0
 3,29,112,9,9,7,[other]
-3,30,112,9,9,7,[other]
+3,30,112,9,9,7,0
 3,31,112,9,9,7,[other]
 3,32,112,9,9,7,[other]
-4,3,119,9,2,7,[other]
+4,3,119,9,9,7,[other]
 4,4,112,9,9,7,[other]
 4,5,112,9,9,7,[other]
 4,6,112,9,9,7,[other]
 4,7,112,9,9,7,[other]
 4,8,112,9,9,7,[other]
-4,9,112,9,9,7,[other]
+4,9,112,9,9,7,0
 4,10,112,9,9,7,[other]
-4,11,112,9,9,7,[other]
-4,12,112,9,9,8,[other]
-4,13,112,9,9,7,[other]
+4,11,112,9,9,7,0
+4,12,112,9,9,7,[other]
+4,13,112,9,9,7,0
 4,14,112,9,9,7,[other]
-4,15,112,9,9,7,[other]
+4,15,112,9,9,7,0
 4,16,112,9,9,7,[other]
 4,17,112,9,9,7,[other]
-4,18,112,9,9,7,[other]
+4,18,112,9,9,7,0
 4,19,112,9,9,7,[other]
-4,20,112,9,9,8,[other]
+4,20,112,9,9,7,[other]
 4,21,112,9,9,7,[other]
-4,22,112,9,9,7,[other]
+4,22,112,9,9,7,0
 4,23,112,9,9,7,[other]
-4,24,112,9,9,7,[other]
+4,24,112,9,9,7,0
 4,25,112,9,9,7,[other]
 4,26,112,9,9,7,[other]
-4,27,112,9,9,7,[other]
-4,28,112,9,9,8,[other]
+4,27,112,9,9,7,0
+4,28,112,9,9,7,[other]
 4,29,112,9,9,7,[other]
 4,30,112,9,9,7,[other]
-4,31,112,9,9,7,[other]
+4,31,112,9,9,7,0
 4,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv
similarity index 76%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv
index f36537ee..5ade6550 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-4-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv
@@ -2,19 +2,19 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 5,3,111,3,2,2,[other]
 5,4,111,3,2,8,[other]
 5,5,111,1,2,7,[other]
-5,6,111,1,2,7,[other]
-5,7,111,1,9,7,[other]
-5,8,111,3,9,7,[other]
+5,6,111,1,9,7,[other]
+5,7,111,3,9,7,[other]
+5,8,111,1,9,7,[other]
 5,9,102,1,9,7,[other]
-5,10,102,1,9,[unknown],[other]
-5,11,102,1,9,[unknown],[other]
+5,10,102,1,9,7,[other]
+5,11,102,4,9,[unknown],[other]
 5,12,102,6,9,[unknown],[other]
-5,13,104,6,9,[unknown],[other]
+5,13,104,6,9,5,[other]
 5,14,104,6,9,5,[other]
 5,15,104,6,9,5,[other]
 5,16,104,9,9,5,[other]
 5,17,104,9,9,7,[other]
-5,18,104,6,9,7,[other]
+5,18,104,9,9,7,[other]
 5,19,104,6,9,7,[other]
 5,20,104,6,9,7,[other]
 5,21,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv
similarity index 71%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv
index ef0be269..7090f23b 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-5-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv
@@ -1,22 +1,22 @@
 sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,111,9,2,7,[other]
+6,4,111,9,9,7,1
 6,5,111,9,2,7,[other]
-6,6,111,9,2,7,[other]
-6,7,111,1,9,7,[other]
-6,8,111,1,9,7,[other]
-6,9,102,1,9,7,[other]
-6,10,102,1,9,[unknown],[other]
-6,11,104,6,9,[unknown],[other]
-6,12,104,1,9,[unknown],[other]
-6,13,102,6,9,[unknown],[other]
-6,14,104,6,9,[unknown],[other]
+6,6,111,1,2,7,[other]
+6,7,102,1,2,7,[other]
+6,8,104,9,9,7,[other]
+6,9,111,1,9,7,[other]
+6,10,102,1,9,7,[other]
+6,11,104,6,9,7,[other]
+6,12,104,1,9,7,[other]
+6,13,104,6,9,7,[other]
+6,14,104,6,9,7,[other]
 6,15,104,6,9,7,[other]
 6,16,104,6,9,7,[other]
 6,17,104,6,9,7,[other]
 6,18,104,6,9,7,[other]
 6,19,104,6,9,7,[other]
 6,20,104,6,9,7,[other]
-6,21,104,9,9,7,[other]
+6,21,104,6,9,7,[other]
 6,22,104,6,9,7,[other]
 6,23,104,6,9,7,[other]
 6,24,104,6,9,7,[other]
@@ -29,15 +29,15 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 6,31,104,6,9,7,[other]
 6,32,104,6,9,7,[other]
 6,33,104,6,9,7,[other]
-7,3,119,2,9,7,[other]
-7,4,104,6,9,7,[other]
-7,5,104,6,9,7,[other]
-7,6,104,6,9,7,[other]
-7,7,104,6,9,7,[other]
+7,3,111,2,9,7,[other]
+7,4,111,5,9,7,[other]
+7,5,111,1,9,7,[other]
+7,6,117,1,9,7,[other]
+7,7,129,6,9,7,[other]
 7,8,104,6,9,7,[other]
 7,9,104,6,9,7,[other]
 7,10,104,6,9,7,[other]
-7,11,104,9,9,7,[other]
+7,11,104,6,9,7,[other]
 7,12,104,6,9,7,[other]
 7,13,104,6,9,7,[other]
 7,14,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv
similarity index 84%
rename from tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv
rename to tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv
index 1b61ff20..200bb342 100644
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-6-predictions.csv
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv
@@ -6,12 +6,12 @@ sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
 8,8,111,1,9,7,[other]
 8,9,111,1,9,7,[other]
 8,10,102,1,9,7,[other]
-8,11,129,1,9,[unknown],[other]
+8,11,102,1,9,7,[other]
 8,12,102,1,9,[unknown],[other]
 8,13,104,6,9,[unknown],[other]
-8,14,104,6,9,[unknown],[other]
-8,15,104,6,9,[unknown],[other]
-8,16,104,6,9,[unknown],[other]
+8,14,104,6,9,7,[other]
+8,15,104,6,9,7,[other]
+8,16,104,6,9,7,[other]
 8,17,104,6,9,7,[other]
 8,18,104,6,9,7,[other]
 8,19,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv
new file mode 100644
index 00000000..8b618b25
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,113,3,2,6,[other]
+9,4,111,4,2,7,[other]
+9,5,117,4,2,7,[mask]
+9,6,117,4,9,7,[mask]
+9,7,104,4,9,7,[other]
+9,8,104,4,9,7,[other]
+9,9,117,4,9,7,[mask]
+9,10,104,9,9,[other],1
+9,11,121,3,9,7,1
+9,12,104,9,9,7,1
+9,13,112,9,9,2,1
+9,14,104,9,9,2,[other]
+9,15,112,9,9,7,[other]
+9,16,119,9,9,8,1
+9,17,112,9,9,7,[other]
+9,18,119,9,9,8,[other]
+9,19,104,9,9,7,[other]
+9,20,129,1,9,7,[other]
+9,21,104,6,9,7,[other]
+9,22,104,6,9,7,[other]
+9,23,104,6,9,7,[other]
+9,24,104,6,9,7,[other]
+9,25,104,6,9,7,[other]
+9,26,104,6,9,7,[other]
+9,27,104,6,9,7,[other]
+9,28,104,6,9,7,[other]
+9,29,104,6,9,7,[other]
+9,30,104,6,9,7,[other]
+9,31,104,6,9,7,[other]
+9,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv
new file mode 100644
index 00000000..f02aeba2
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+0,4,119,9,2,7,[other]
+0,5,104,9,2,7,[other]
+0,6,111,9,9,7,[other]
+0,7,111,9,9,7,[other]
+0,8,111,9,9,7,[other]
+0,9,104,9,9,7,[other]
+0,10,116,1,9,7,[other]
+0,11,104,6,9,7,[other]
+0,12,104,1,9,7,[other]
+0,13,102,1,9,7,[other]
+0,14,104,6,9,7,[other]
+0,15,104,6,9,7,[other]
+0,16,104,6,9,7,[other]
+0,17,104,6,9,7,[other]
+0,18,104,6,9,7,[other]
+0,19,104,6,9,7,[other]
+0,20,104,6,9,7,[other]
+0,21,104,6,9,7,[other]
+0,22,104,6,9,7,[other]
+0,23,104,6,9,7,[other]
+0,24,104,6,9,7,[other]
+0,25,104,6,9,7,[other]
+0,26,104,6,9,7,[other]
+0,27,104,6,9,7,[other]
+0,28,104,6,9,7,[other]
+0,29,104,6,9,7,[other]
+0,30,104,6,9,7,[other]
+0,31,104,6,9,7,[other]
+0,32,104,6,9,7,[other]
+0,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv
new file mode 100644
index 00000000..756d6e7c
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+1,4,111,9,2,7,[other]
+1,5,111,1,9,7,[other]
+1,6,111,1,9,7,[other]
+1,7,111,1,9,7,[other]
+1,8,111,1,9,7,[other]
+1,9,111,1,9,7,[other]
+1,10,102,1,9,7,[other]
+1,11,129,1,9,7,[other]
+1,12,102,1,9,[unknown],[other]
+1,13,104,6,9,[unknown],[other]
+1,14,104,6,9,[unknown],[other]
+1,15,104,6,9,7,[other]
+1,16,104,6,9,7,[other]
+1,17,104,6,9,7,[other]
+1,18,104,6,9,7,[other]
+1,19,104,6,9,7,[other]
+1,20,104,6,9,7,[other]
+1,21,104,6,9,7,[other]
+1,22,104,6,9,7,[other]
+1,23,104,6,9,7,[other]
+1,24,104,6,9,7,[other]
+1,25,104,6,9,7,[other]
+1,26,104,6,9,7,[other]
+1,27,104,6,9,7,[other]
+1,28,104,6,9,7,[other]
+1,29,104,6,9,7,[other]
+1,30,104,6,9,7,[other]
+1,31,104,6,9,7,[other]
+1,32,104,6,9,7,[other]
+1,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv
new file mode 100644
index 00000000..50574151
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+2,3,115,9,2,2,2
+2,4,115,9,2,2,[mask]
+2,5,118,9,2,7,[mask]
+2,6,112,9,9,7,0
+2,7,112,9,9,2,[unknown]
+2,8,115,9,9,7,[other]
+2,9,112,9,9,7,[other]
+2,10,112,9,9,7,1
+2,11,112,9,9,7,[other]
+2,12,112,9,9,7,[other]
+2,13,112,9,9,7,[other]
+2,14,112,9,9,7,[other]
+2,15,112,9,9,7,[other]
+2,16,112,9,9,7,0
+2,17,112,9,9,8,[other]
+2,18,112,9,9,7,[other]
+2,19,112,9,9,7,[other]
+2,20,112,9,9,7,[other]
+2,21,112,9,9,7,[other]
+2,22,112,9,9,7,[other]
+2,23,112,9,9,7,[other]
+2,24,112,9,9,7,[other]
+2,25,112,9,9,7,[other]
+2,26,112,9,9,7,0
+2,27,112,9,9,7,[other]
+2,28,112,9,9,7,0
+2,29,112,9,9,7,[other]
+2,30,112,9,9,7,0
+2,31,112,9,9,7,[other]
+2,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv
new file mode 100644
index 00000000..08e5981d
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+3,3,112,9,2,2,2
+3,4,115,9,2,9,[mask]
+3,5,118,9,2,2,[mask]
+3,6,112,9,2,2,0
+3,7,112,9,9,2,[other]
+3,8,112,9,2,2,[other]
+3,9,112,9,2,7,[mask]
+3,10,112,9,9,7,0
+3,11,112,9,9,7,[other]
+3,12,112,9,9,7,0
+3,13,112,9,9,7,[other]
+3,14,112,9,9,7,[other]
+3,15,112,9,9,7,[other]
+3,16,112,9,9,7,0
+3,17,112,9,9,7,[other]
+3,18,112,9,9,7,[other]
+3,19,112,9,9,7,0
+3,20,112,9,9,7,[other]
+3,21,112,9,9,7,0
+3,22,112,9,9,7,[other]
+3,23,112,9,9,7,[other]
+3,24,112,9,9,7,[other]
+3,25,112,9,9,7,0
+3,26,112,9,9,7,[other]
+3,27,112,9,9,7,[other]
+3,28,112,9,9,7,0
+3,29,112,9,9,7,[other]
+3,30,112,9,9,7,0
+3,31,112,9,9,7,[other]
+3,32,112,9,9,7,[other]
+4,3,119,9,9,7,[other]
+4,4,112,9,9,7,[other]
+4,5,112,9,9,7,[other]
+4,6,112,9,9,7,[other]
+4,7,112,9,9,7,[other]
+4,8,112,9,9,7,[other]
+4,9,112,9,9,7,0
+4,10,112,9,9,7,[other]
+4,11,112,9,9,7,0
+4,12,112,9,9,7,[other]
+4,13,112,9,9,7,0
+4,14,112,9,9,7,[other]
+4,15,112,9,9,7,0
+4,16,112,9,9,7,[other]
+4,17,112,9,9,7,[other]
+4,18,112,9,9,7,0
+4,19,112,9,9,7,[other]
+4,20,112,9,9,7,[other]
+4,21,112,9,9,7,[other]
+4,22,112,9,9,7,0
+4,23,112,9,9,7,[other]
+4,24,112,9,9,7,0
+4,25,112,9,9,7,[other]
+4,26,112,9,9,7,[other]
+4,27,112,9,9,7,0
+4,28,112,9,9,7,[other]
+4,29,112,9,9,7,[other]
+4,30,112,9,9,7,[other]
+4,31,112,9,9,7,0
+4,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv
new file mode 100644
index 00000000..5ade6550
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+5,3,111,3,2,2,[other]
+5,4,111,3,2,8,[other]
+5,5,111,1,2,7,[other]
+5,6,111,1,9,7,[other]
+5,7,111,3,9,7,[other]
+5,8,111,1,9,7,[other]
+5,9,102,1,9,7,[other]
+5,10,102,1,9,7,[other]
+5,11,102,4,9,[unknown],[other]
+5,12,102,6,9,[unknown],[other]
+5,13,104,6,9,5,[other]
+5,14,104,6,9,5,[other]
+5,15,104,6,9,5,[other]
+5,16,104,9,9,5,[other]
+5,17,104,9,9,7,[other]
+5,18,104,9,9,7,[other]
+5,19,104,6,9,7,[other]
+5,20,104,6,9,7,[other]
+5,21,104,6,9,7,[other]
+5,22,104,6,9,7,[other]
+5,23,104,6,9,7,[other]
+5,24,104,6,9,7,[other]
+5,25,104,6,9,7,[other]
+5,26,104,6,9,7,[other]
+5,27,104,6,9,7,[other]
+5,28,104,6,9,7,[other]
+5,29,104,6,9,7,[other]
+5,30,104,6,9,7,[other]
+5,31,104,6,9,7,[other]
+5,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv
new file mode 100644
index 00000000..7090f23b
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+6,4,111,9,9,7,1
+6,5,111,9,2,7,[other]
+6,6,111,1,2,7,[other]
+6,7,102,1,2,7,[other]
+6,8,104,9,9,7,[other]
+6,9,111,1,9,7,[other]
+6,10,102,1,9,7,[other]
+6,11,104,6,9,7,[other]
+6,12,104,1,9,7,[other]
+6,13,104,6,9,7,[other]
+6,14,104,6,9,7,[other]
+6,15,104,6,9,7,[other]
+6,16,104,6,9,7,[other]
+6,17,104,6,9,7,[other]
+6,18,104,6,9,7,[other]
+6,19,104,6,9,7,[other]
+6,20,104,6,9,7,[other]
+6,21,104,6,9,7,[other]
+6,22,104,6,9,7,[other]
+6,23,104,6,9,7,[other]
+6,24,104,6,9,7,[other]
+6,25,104,6,9,7,[other]
+6,26,104,6,9,7,[other]
+6,27,104,6,9,7,[other]
+6,28,104,6,9,7,[other]
+6,29,104,6,9,7,[other]
+6,30,104,6,9,7,[other]
+6,31,104,6,9,7,[other]
+6,32,104,6,9,7,[other]
+6,33,104,6,9,7,[other]
+7,3,111,2,9,7,[other]
+7,4,111,5,9,7,[other]
+7,5,111,1,9,7,[other]
+7,6,117,1,9,7,[other]
+7,7,129,6,9,7,[other]
+7,8,104,6,9,7,[other]
+7,9,104,6,9,7,[other]
+7,10,104,6,9,7,[other]
+7,11,104,6,9,7,[other]
+7,12,104,6,9,7,[other]
+7,13,104,6,9,7,[other]
+7,14,104,6,9,7,[other]
+7,15,104,6,9,7,[other]
+7,16,104,6,9,7,[other]
+7,17,104,6,9,7,[other]
+7,18,104,6,9,7,[other]
+7,19,104,6,9,7,[other]
+7,20,104,6,9,7,[other]
+7,21,104,6,9,7,[other]
+7,22,104,6,9,7,[other]
+7,23,104,6,9,7,[other]
+7,24,104,6,9,7,[other]
+7,25,104,6,9,7,[other]
+7,26,104,6,9,7,[other]
+7,27,104,6,9,7,[other]
+7,28,104,6,9,7,[other]
+7,29,104,6,9,7,[other]
+7,30,104,6,9,7,[other]
+7,31,104,6,9,7,[other]
+7,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv
new file mode 100644
index 00000000..200bb342
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+8,4,111,9,9,7,[other]
+8,5,111,9,9,7,[other]
+8,6,111,9,9,7,[other]
+8,7,111,1,9,7,[other]
+8,8,111,1,9,7,[other]
+8,9,111,1,9,7,[other]
+8,10,102,1,9,7,[other]
+8,11,102,1,9,7,[other]
+8,12,102,1,9,[unknown],[other]
+8,13,104,6,9,[unknown],[other]
+8,14,104,6,9,7,[other]
+8,15,104,6,9,7,[other]
+8,16,104,6,9,7,[other]
+8,17,104,6,9,7,[other]
+8,18,104,6,9,7,[other]
+8,19,104,6,9,7,[other]
+8,20,104,6,9,7,[other]
+8,21,104,6,9,7,[other]
+8,22,104,6,9,7,[other]
+8,23,104,6,9,7,[other]
+8,24,104,6,9,7,[other]
+8,25,104,6,9,7,[other]
+8,26,104,6,9,7,[other]
+8,27,104,6,9,7,[other]
+8,28,104,6,9,7,[other]
+8,29,104,6,9,7,[other]
+8,30,104,6,9,7,[other]
+8,31,104,6,9,7,[other]
+8,32,104,6,9,7,[other]
+8,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv
new file mode 100644
index 00000000..8b618b25
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,113,3,2,6,[other]
+9,4,111,4,2,7,[other]
+9,5,117,4,2,7,[mask]
+9,6,117,4,9,7,[mask]
+9,7,104,4,9,7,[other]
+9,8,104,4,9,7,[other]
+9,9,117,4,9,7,[mask]
+9,10,104,9,9,[other],1
+9,11,121,3,9,7,1
+9,12,104,9,9,7,1
+9,13,112,9,9,2,1
+9,14,104,9,9,2,[other]
+9,15,112,9,9,7,[other]
+9,16,119,9,9,8,1
+9,17,112,9,9,7,[other]
+9,18,119,9,9,8,[other]
+9,19,104,9,9,7,[other]
+9,20,129,1,9,7,[other]
+9,21,104,6,9,7,[other]
+9,22,104,6,9,7,[other]
+9,23,104,6,9,7,[other]
+9,24,104,6,9,7,[other]
+9,25,104,6,9,7,[other]
+9,26,104,6,9,7,[other]
+9,27,104,6,9,7,[other]
+9,28,104,6,9,7,[other]
+9,29,104,6,9,7,[other]
+9,30,104,6,9,7,[other]
+9,31,104,6,9,7,[other]
+9,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv
new file mode 100644
index 00000000..7269a244
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+0,4,102,6,9,5,1
+0,5,104,6,5,5,1
+0,6,104,6,9,5,1
+0,7,104,6,9,8,1
+0,8,104,6,9,5,1
+0,9,104,6,9,8,1
+0,10,104,4,9,5,1
+0,11,104,6,9,8,0
+0,12,103,4,9,5,1
+0,13,102,6,9,5,0
+0,14,103,6,9,5,1
+0,15,102,6,9,5,1
+0,16,102,6,9,5,1
+0,17,102,6,9,5,1
+0,18,102,6,9,8,1
+0,19,104,6,9,5,1
+0,20,102,6,9,8,1
+0,21,104,4,9,5,1
+0,22,102,6,9,8,1
+0,23,102,6,9,5,1
+0,24,102,6,9,5,1
+0,25,102,6,9,5,1
+0,26,102,6,9,5,1
+0,27,102,6,9,5,1
+0,28,102,6,9,5,1
+0,29,102,6,9,5,1
+0,30,102,6,9,5,1
+0,31,102,6,9,5,1
+0,32,102,6,9,5,1
+0,33,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv
new file mode 100644
index 00000000..f57b3e81
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+1,4,104,6,9,5,1
+1,5,104,6,9,8,1
+1,6,104,6,9,5,1
+1,7,104,6,9,8,1
+1,8,104,6,9,5,1
+1,9,102,6,9,8,1
+1,10,104,6,9,5,1
+1,11,112,6,9,5,1
+1,12,102,6,9,5,1
+1,13,102,6,9,5,1
+1,14,102,6,9,5,1
+1,15,102,6,9,5,1
+1,16,102,6,9,5,1
+1,17,102,6,9,5,1
+1,18,102,6,9,5,1
+1,19,102,6,9,5,1
+1,20,102,6,9,5,1
+1,21,102,6,9,5,1
+1,22,102,6,9,5,1
+1,23,102,6,9,5,1
+1,24,102,6,9,5,1
+1,25,102,6,9,5,1
+1,26,102,6,9,5,1
+1,27,102,6,9,5,1
+1,28,102,6,9,5,1
+1,29,102,6,9,5,1
+1,30,102,6,9,5,1
+1,31,102,6,9,5,1
+1,32,102,6,9,5,1
+1,33,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv
new file mode 100644
index 00000000..5d574330
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+2,3,112,6,5,5,1
+2,4,102,6,5,5,1
+2,5,102,6,1,5,1
+2,6,102,6,5,5,1
+2,7,102,6,1,5,1
+2,8,102,6,1,0,1
+2,9,102,6,1,5,1
+2,10,102,6,9,0,1
+2,11,102,6,9,5,1
+2,12,102,6,9,5,1
+2,13,102,6,9,5,1
+2,14,102,6,9,5,1
+2,15,102,6,9,5,1
+2,16,102,6,9,5,1
+2,17,102,6,9,5,1
+2,18,102,6,9,5,1
+2,19,102,6,9,5,1
+2,20,102,6,9,5,1
+2,21,102,6,9,5,1
+2,22,102,6,9,5,1
+2,23,102,6,9,5,1
+2,24,102,6,9,5,1
+2,25,102,6,9,5,1
+2,26,102,6,9,5,1
+2,27,102,6,9,5,1
+2,28,102,6,9,5,1
+2,29,102,6,9,5,1
+2,30,102,6,9,5,1
+2,31,102,6,9,5,1
+2,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv
new file mode 100644
index 00000000..3bd7fa01
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+3,3,104,4,9,8,1
+3,4,104,4,9,8,0
+3,5,104,4,9,8,1
+3,6,104,4,9,8,0
+3,7,104,4,9,8,1
+3,8,104,4,9,8,0
+3,9,104,4,9,5,1
+3,10,112,9,8,8,0
+3,11,104,4,1,5,1
+3,12,112,6,1,8,0
+3,13,112,4,1,5,1
+3,14,112,6,5,5,0
+3,15,112,6,1,5,1
+3,16,112,6,5,5,1
+3,17,102,6,5,5,1
+3,18,102,6,1,5,1
+3,19,102,6,1,0,1
+3,20,102,6,1,5,1
+3,21,102,6,1,0,1
+3,22,102,6,1,5,1
+3,23,102,6,1,0,1
+3,24,102,6,1,0,1
+3,25,102,6,1,0,1
+3,26,102,6,1,0,1
+3,27,102,6,1,0,1
+3,28,102,6,1,0,1
+3,29,102,6,1,0,1
+3,30,102,6,1,0,1
+3,31,102,6,9,0,1
+3,32,102,6,9,5,1
+4,3,112,6,1,5,1
+4,4,102,6,5,0,1
+4,5,104,6,1,5,1
+4,6,102,6,5,0,1
+4,7,104,6,1,5,1
+4,8,102,6,1,0,1
+4,9,112,6,1,5,1
+4,10,102,6,5,5,1
+4,11,102,6,9,0,1
+4,12,102,6,9,5,1
+4,13,102,6,9,0,1
+4,14,102,6,9,5,1
+4,15,102,6,9,5,1
+4,16,102,6,9,5,1
+4,17,102,6,9,5,1
+4,18,102,6,9,5,1
+4,19,102,6,9,5,1
+4,20,102,6,9,5,1
+4,21,102,6,9,5,1
+4,22,102,6,9,5,1
+4,23,102,6,9,5,1
+4,24,102,6,9,5,1
+4,25,102,6,9,5,1
+4,26,102,6,9,5,1
+4,27,102,6,9,5,1
+4,28,102,6,9,5,1
+4,29,102,6,9,5,1
+4,30,102,6,9,5,1
+4,31,102,6,9,5,1
+4,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv
new file mode 100644
index 00000000..083376d3
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+5,3,102,6,9,2,1
+5,4,102,6,9,7,1
+5,5,102,6,9,2,1
+5,6,102,6,9,7,1
+5,7,102,6,9,2,1
+5,8,102,6,9,7,1
+5,9,102,6,9,8,1
+5,10,102,6,9,5,1
+5,11,102,6,9,5,1
+5,12,102,6,9,5,1
+5,13,102,6,9,5,1
+5,14,102,6,9,5,1
+5,15,102,6,9,5,1
+5,16,102,6,9,5,1
+5,17,102,6,9,5,1
+5,18,102,6,9,5,1
+5,19,102,6,9,5,1
+5,20,102,6,9,5,1
+5,21,102,6,9,5,1
+5,22,102,6,9,5,1
+5,23,102,6,9,5,1
+5,24,102,6,9,5,1
+5,25,102,6,9,5,1
+5,26,102,6,9,5,1
+5,27,102,6,9,5,1
+5,28,102,6,9,5,1
+5,29,102,6,9,5,1
+5,30,102,6,9,5,1
+5,31,102,6,9,5,1
+5,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv
new file mode 100644
index 00000000..2015e2b4
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv
@@ -0,0 +1,61 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+6,4,121,6,1,5,0
+6,5,102,6,1,5,1
+6,6,102,6,5,5,1
+6,7,102,6,5,5,1
+6,8,102,6,1,5,1
+6,9,102,6,1,5,1
+6,10,102,6,1,5,1
+6,11,102,6,1,5,1
+6,12,102,6,1,5,1
+6,13,102,6,1,0,1
+6,14,102,6,1,0,1
+6,15,102,6,1,0,1
+6,16,102,6,1,0,1
+6,17,102,6,1,0,1
+6,18,102,6,1,0,1
+6,19,102,6,1,0,1
+6,20,102,6,1,0,1
+6,21,102,6,9,0,1
+6,22,102,6,9,5,1
+6,23,102,6,9,5,1
+6,24,102,6,9,5,1
+6,25,102,6,9,5,1
+6,26,102,6,9,5,1
+6,27,102,6,9,5,1
+6,28,102,6,9,5,1
+6,29,102,6,9,5,1
+6,30,102,6,9,5,1
+6,31,102,6,9,5,1
+6,32,102,6,9,5,1
+6,33,102,6,9,5,1
+7,3,112,4,1,5,1
+7,4,112,6,5,5,1
+7,5,112,6,5,5,1
+7,6,102,6,5,5,1
+7,7,102,6,1,5,1
+7,8,102,6,1,5,1
+7,9,102,6,1,5,1
+7,10,102,6,1,0,1
+7,11,102,6,9,0,1
+7,12,102,6,9,5,1
+7,13,102,6,9,5,1
+7,14,102,6,9,5,1
+7,15,102,6,9,5,1
+7,16,102,6,9,5,1
+7,17,102,6,9,5,1
+7,18,102,6,9,5,1
+7,19,102,6,9,5,1
+7,20,102,6,9,5,1
+7,21,102,6,9,5,1
+7,22,102,6,9,5,1
+7,23,102,6,9,5,1
+7,24,102,6,9,5,1
+7,25,102,6,9,5,1
+7,26,102,6,9,5,1
+7,27,102,6,9,5,1
+7,28,102,6,9,5,1
+7,29,102,6,9,5,1
+7,30,102,6,9,5,1
+7,31,102,6,9,5,1
+7,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv
new file mode 100644
index 00000000..0c53cfc0
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+8,4,104,6,1,5,1
+8,5,112,6,5,5,1
+8,6,104,6,5,5,1
+8,7,104,6,1,5,1
+8,8,112,6,1,5,1
+8,9,102,6,5,5,1
+8,10,102,6,9,5,1
+8,11,102,6,9,0,1
+8,12,102,6,9,5,1
+8,13,102,6,9,5,1
+8,14,102,6,9,0,1
+8,15,102,6,9,5,1
+8,16,102,6,9,5,1
+8,17,102,6,9,5,1
+8,18,102,6,9,5,1
+8,19,102,6,9,5,1
+8,20,102,6,9,5,1
+8,21,102,6,9,5,1
+8,22,102,6,9,5,1
+8,23,102,6,9,5,1
+8,24,102,6,9,5,1
+8,25,102,6,9,5,1
+8,26,102,6,9,5,1
+8,27,102,6,9,5,1
+8,28,102,6,9,5,1
+8,29,102,6,9,5,1
+8,30,102,6,9,5,1
+8,31,102,6,9,5,1
+8,32,102,6,9,5,1
+8,33,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv
new file mode 100644
index 00000000..2d5e6cb9
--- /dev/null
+++ b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv
@@ -0,0 +1,31 @@
+sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
+9,3,102,6,9,5,1
+9,4,102,6,9,5,1
+9,5,102,6,9,5,1
+9,6,102,6,9,5,1
+9,7,102,6,9,5,1
+9,8,102,6,9,5,1
+9,9,102,6,9,5,1
+9,10,102,6,9,5,1
+9,11,102,6,9,5,1
+9,12,102,6,9,5,1
+9,13,102,6,9,5,1
+9,14,102,6,9,5,1
+9,15,102,6,9,5,1
+9,16,102,6,9,5,1
+9,17,102,6,9,5,1
+9,18,102,6,9,5,1
+9,19,102,6,9,5,1
+9,20,102,6,9,5,1
+9,21,102,6,9,5,1
+9,22,102,6,9,5,1
+9,23,102,6,9,5,1
+9,24,102,6,9,5,1
+9,25,102,6,9,5,1
+9,26,102,6,9,5,1
+9,27,102,6,9,5,1
+9,28,102,6,9,5,1
+9,29,102,6,9,5,1
+9,30,102,6,9,5,1
+9,31,102,6,9,5,1
+9,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv
deleted file mode 100644
index 453ac47f..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,102,3,3,5,0
-0,5,102,3,9,0,1
-0,6,102,1,9,0,0
-0,7,102,1,0,0,2
-0,8,111,1,2,0,2
-0,9,102,1,9,0,2
-0,10,102,1,3,0,2
-0,11,102,1,3,0,2
-0,12,111,1,3,0,2
-0,13,111,1,9,0,2
-0,14,105,1,0,0,2
-0,15,105,1,0,0,2
-0,16,105,1,0,0,2
-0,17,105,1,0,0,2
-0,18,105,1,0,0,2
-0,19,105,1,0,0,2
-0,20,105,1,0,8,2
-0,21,105,1,0,5,2
-0,22,105,5,0,2,2
-0,23,105,5,0,8,2
-0,24,105,5,0,8,2
-0,25,105,5,0,8,2
-0,26,105,5,0,8,2
-0,27,105,3,0,8,2
-0,28,105,3,0,5,1
-0,29,105,3,0,2,1
-0,30,105,3,0,2,1
-0,31,105,3,0,2,1
-0,32,110,3,0,2,1
-0,33,110,3,0,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv
deleted file mode 100644
index 6eac25ae..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,102,7,0,5,[other]
-1,5,111,3,0,1,0
-1,6,111,1,0,1,1
-1,7,111,7,0,1,0
-1,8,111,5,0,1,1
-1,9,[unknown],7,0,1,0
-1,10,102,1,0,5,0
-1,11,110,5,0,1,2
-1,12,102,5,0,5,0
-1,13,102,5,0,2,1
-1,14,[unknown],5,0,1,1
-1,15,110,2,0,8,0
-1,16,102,3,0,5,0
-1,17,111,3,0,2,2
-1,18,111,3,3,5,2
-1,19,111,3,3,1,2
-1,20,111,1,3,0,2
-1,21,102,1,2,0,2
-1,22,111,[mask],2,0,2
-1,23,112,[mask],2,5,2
-1,24,109,[mask],0,5,2
-1,25,109,[mask],0,2,2
-1,26,109,[mask],0,[other],2
-1,27,109,3,8,[other],2
-1,28,109,1,8,5,2
-1,29,109,1,8,[other],2
-1,30,104,1,8,[other],2
-1,31,104,1,9,[other],2
-1,32,111,3,9,0,2
-1,33,127,[mask],9,[other],2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv
deleted file mode 100644
index 9b51b012..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,112,2,0,5,1
-2,4,112,7,2,0,[other]
-2,5,112,7,3,5,2
-2,6,104,7,3,0,2
-2,7,104,7,9,0,2
-2,8,104,7,3,0,0
-2,9,104,4,3,[other],0
-2,10,112,7,3,0,0
-2,11,104,7,3,0,0
-2,12,116,4,3,6,0
-2,13,112,4,3,6,0
-2,14,112,7,3,6,0
-2,15,116,7,3,6,0
-2,16,116,4,3,6,0
-2,17,112,4,3,6,0
-2,18,112,8,3,6,0
-2,19,112,8,3,6,0
-2,20,118,4,7,6,0
-2,21,112,4,9,6,1
-2,22,118,8,7,6,0
-2,23,112,4,9,6,0
-2,24,112,8,7,0,1
-2,25,112,4,4,6,2
-2,26,112,4,3,[other],1
-2,27,104,4,4,8,[other]
-2,28,112,4,4,[other],1
-2,29,104,7,4,8,1
-2,30,104,4,4,[other],1
-2,31,104,8,4,[other],1
-2,32,104,8,4,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv
deleted file mode 100644
index 4c107703..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,111,3,3,5,0
-3,4,111,2,3,5,0
-3,5,112,3,3,2,0
-3,6,102,3,3,5,1
-3,7,118,2,3,0,0
-3,8,102,6,3,2,0
-3,9,102,6,3,0,0
-3,10,102,6,3,0,0
-3,11,102,6,3,2,0
-3,12,102,6,3,2,0
-3,13,102,3,3,2,2
-3,14,102,6,3,0,2
-3,15,102,6,3,0,0
-3,16,102,1,3,2,2
-3,17,111,3,3,0,2
-3,18,102,1,3,0,2
-3,19,111,1,3,0,2
-3,20,111,1,3,0,2
-3,21,111,1,2,0,2
-3,22,111,1,2,0,2
-3,23,102,1,2,0,2
-3,24,111,1,2,0,2
-3,25,111,1,2,5,2
-3,26,105,5,9,5,2
-3,27,105,[mask],0,8,2
-3,28,105,[mask],0,8,2
-3,29,105,[mask],0,8,2
-3,30,105,[mask],0,[other],2
-3,31,105,[mask],9,[other],2
-3,32,105,[mask],9,[other],2
-4,3,113,7,9,5,2
-4,4,104,4,0,0,1
-4,5,112,6,2,0,1
-4,6,104,7,9,0,1
-4,7,118,7,9,0,1
-4,8,104,4,9,0,0
-4,9,102,7,0,0,0
-4,10,102,7,3,0,0
-4,11,111,6,3,0,0
-4,12,102,4,3,0,0
-4,13,102,6,3,0,0
-4,14,102,6,3,0,0
-4,15,102,6,3,0,0
-4,16,102,6,3,6,0
-4,17,102,1,3,2,0
-4,18,102,3,3,2,2
-4,19,102,6,3,0,0
-4,20,102,1,3,2,0
-4,21,102,3,3,2,2
-4,22,111,3,3,0,2
-4,23,102,1,3,0,2
-4,24,111,1,3,0,2
-4,25,102,1,3,0,2
-4,26,111,1,3,0,2
-4,27,102,1,2,0,2
-4,28,111,1,2,0,2
-4,29,111,1,2,0,2
-4,30,111,1,2,5,2
-4,31,105,1,9,5,2
-4,32,105,1,0,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv
deleted file mode 100644
index d03238fa..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,104,7,9,8,2
-5,4,104,1,9,0,2
-5,5,104,1,9,0,2
-5,6,104,1,9,0,2
-5,7,104,1,9,0,2
-5,8,104,1,9,0,2
-5,9,104,1,0,0,2
-5,10,104,7,9,0,0
-5,11,122,7,0,7,0
-5,12,122,7,3,7,0
-5,13,119,4,3,7,0
-5,14,119,4,3,0,0
-5,15,112,3,9,6,0
-5,16,122,7,3,0,0
-5,17,102,4,9,6,0
-5,18,102,8,3,0,0
-5,19,102,8,9,6,0
-5,20,102,8,3,0,0
-5,21,102,3,3,6,0
-5,22,102,4,3,0,0
-5,23,102,3,3,6,0
-5,24,102,4,3,6,0
-5,25,102,3,3,6,0
-5,26,102,4,3,6,0
-5,27,102,3,3,6,0
-5,28,102,4,3,2,0
-5,29,102,3,3,6,2
-5,30,111,6,3,0,2
-5,31,102,3,3,6,2
-5,32,102,6,3,6,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv
deleted file mode 100644
index 244d29b4..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,110,1,0,5,1
-6,5,110,2,0,0,1
-6,6,112,1,0,[other],1
-6,7,111,7,[unknown],[other],2
-6,8,111,3,2,5,1
-6,9,112,2,9,5,1
-6,10,108,7,0,1,[other]
-6,11,104,7,0,[other],1
-6,12,119,7,9,0,1
-6,13,104,7,9,0,0
-6,14,122,7,0,0,1
-6,15,111,7,4,[other],0
-6,16,119,7,3,0,0
-6,17,122,7,9,0,0
-6,18,102,7,0,5,0
-6,19,111,7,0,0,0
-6,20,111,8,0,0,0
-6,21,112,8,0,2,0
-6,22,102,7,0,5,0
-6,23,111,3,0,2,0
-6,24,111,3,0,1,0
-6,25,111,1,0,1,0
-6,26,111,1,0,1,0
-6,27,110,1,0,1,0
-6,28,102,1,0,4,0
-6,29,111,1,0,[unknown],2
-6,30,110,1,0,1,1
-6,31,110,1,0,1,1
-6,32,110,1,0,1,1
-6,33,110,1,0,1,1
-7,3,102,3,0,5,0
-7,4,102,3,0,[unknown],0
-7,5,111,3,0,[unknown],0
-7,6,102,8,0,1,[other]
-7,7,102,3,0,2,0
-7,8,111,3,0,2,1
-7,9,112,4,0,1,[other]
-7,10,102,1,0,2,0
-7,11,111,1,8,2,2
-7,12,110,1,[mask],1,2
-7,13,[mask],1,0,1,2
-7,14,111,1,0,1,2
-7,15,111,1,2,1,2
-7,16,110,1,2,1,2
-7,17,[mask],1,0,5,1
-7,18,109,1,0,8,2
-7,19,102,1,0,5,1
-7,20,110,7,0,8,2
-7,21,127,3,0,5,1
-7,22,109,3,0,5,1
-7,23,109,7,0,5,1
-7,24,109,7,0,5,1
-7,25,104,7,0,5,1
-7,26,109,7,0,5,1
-7,27,119,7,0,0,[other]
-7,28,111,7,0,[other],0
-7,29,111,7,0,0,0
-7,30,111,7,9,0,0
-7,31,111,7,0,0,0
-7,32,111,7,0,0,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv
deleted file mode 100644
index 0735ca91..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,111,3,0,6,[other]
-8,5,102,7,9,7,[other]
-8,6,104,7,0,7,0
-8,7,102,8,0,5,1
-8,8,112,3,0,2,0
-8,9,111,7,0,5,2
-8,10,111,1,3,0,2
-8,11,102,6,9,0,2
-8,12,111,1,3,0,2
-8,13,111,1,3,0,2
-8,14,102,1,9,0,2
-8,15,111,1,2,0,2
-8,16,111,1,2,5,2
-8,17,111,5,0,5,2
-8,18,105,5,0,[other],2
-8,19,105,[mask],0,2,2
-8,20,105,[mask],0,8,2
-8,21,105,[mask],0,8,2
-8,22,105,[mask],0,[other],2
-8,23,105,3,0,[other],2
-8,24,109,5,9,0,2
-8,25,105,[mask],0,6,2
-8,26,105,[mask],3,[other],2
-8,27,109,[mask],0,[other],2
-8,28,109,3,0,[other],2
-8,29,109,3,0,2,2
-8,30,109,3,0,6,1
-8,31,109,3,0,2,1
-8,32,109,3,0,[other],[other]
-8,33,109,3,8,2,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv
deleted file mode 100644
index 20d90721..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-predictions/sequifier-test-hp-search-custom-eval-run-4-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,108,4,9,1,1
-9,4,113,7,9,8,1
-9,5,104,4,9,0,1
-9,6,119,8,9,0,1
-9,7,119,7,9,0,2
-9,8,104,7,9,0,0
-9,9,122,7,9,0,0
-9,10,102,7,3,0,0
-9,11,122,7,3,0,0
-9,12,102,4,3,6,0
-9,13,102,7,3,0,0
-9,14,111,6,3,0,0
-9,15,112,3,3,6,0
-9,16,102,7,3,5,0
-9,17,102,8,3,6,0
-9,18,102,3,3,2,0
-9,19,102,4,3,6,0
-9,20,102,3,3,6,0
-9,21,102,6,3,2,0
-9,22,102,3,3,2,1
-9,23,118,3,3,6,2
-9,24,102,6,3,0,0
-9,25,102,3,3,2,2
-9,26,111,3,3,0,2
-9,27,102,3,3,6,2
-9,28,102,6,3,0,2
-9,29,111,1,3,2,2
-9,30,102,1,3,0,2
-9,31,102,1,2,0,2
-9,32,111,1,2,0,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv
deleted file mode 100644
index 0b011a28..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,104,7,[mask],[unknown],1
-0,5,104,7,[mask],[unknown],0
-0,6,104,7,1,3,0
-0,7,104,0,1,3,0
-0,8,112,7,1,[unknown],0
-0,9,108,5,1,[unknown],0
-0,10,112,1,1,5,2
-0,11,129,2,1,5,2
-0,12,129,2,1,5,2
-0,13,112,2,1,5,2
-0,14,107,7,[mask],[unknown],1
-0,15,107,7,[mask],[unknown],1
-0,16,107,7,[mask],[unknown],1
-0,17,107,7,[mask],[unknown],1
-0,18,107,7,[mask],[unknown],1
-0,19,104,7,[mask],[unknown],1
-0,20,104,7,[mask],[unknown],1
-0,21,104,9,[mask],[unknown],1
-0,22,104,9,[mask],3,2
-0,23,104,9,[mask],3,2
-0,24,104,9,[mask],3,2
-0,25,104,9,[mask],3,2
-0,26,104,9,[mask],1,2
-0,27,104,9,[other],1,[mask]
-0,28,104,[unknown],[other],1,[mask]
-0,29,112,[unknown],[other],1,[mask]
-0,30,112,2,[mask],[unknown],[mask]
-0,31,110,2,[mask],[unknown],[mask]
-0,32,110,2,[other],7,[mask]
-0,33,104,2,[other],7,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv
deleted file mode 100644
index 531427bf..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,104,7,[mask],1,1
-1,5,104,7,9,1,1
-1,6,104,7,9,[unknown],1
-1,7,104,7,9,[unknown],1
-1,8,104,7,9,[unknown],1
-1,9,118,7,9,[unknown],[other]
-1,10,112,7,9,[unknown],0
-1,11,118,7,1,[unknown],[other]
-1,12,118,7,1,[unknown],[other]
-1,13,110,7,1,[unknown],[other]
-1,14,118,7,1,[unknown],[other]
-1,15,112,7,1,[unknown],2
-1,16,108,7,[mask],[unknown],2
-1,17,129,1,[mask],8,2
-1,18,129,2,[mask],5,2
-1,19,129,2,1,1,2
-1,20,129,7,1,5,[mask]
-1,21,104,9,1,5,[mask]
-1,22,129,9,1,5,[other]
-1,23,112,9,1,5,[mask]
-1,24,112,9,1,5,[mask]
-1,25,112,0,1,5,[mask]
-1,26,112,6,[mask],[unknown],[mask]
-1,27,123,7,[mask],[unknown],[mask]
-1,28,109,7,[mask],5,1
-1,29,104,7,[mask],1,[mask]
-1,30,104,7,[mask],8,1
-1,31,104,7,[mask],8,1
-1,32,104,0,[mask],8,1
-1,33,104,7,[mask],8,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv
deleted file mode 100644
index 4350e88b..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,112,2,[mask],5,[mask]
-2,4,104,7,[mask],[unknown],[mask]
-2,5,104,7,1,8,1
-2,6,104,7,[mask],[unknown],0
-2,7,129,7,1,8,1
-2,8,104,7,[mask],1,0
-2,9,112,7,1,1,1
-2,10,107,7,[mask],[unknown],2
-2,11,129,9,1,1,2
-2,12,112,9,[other],1,2
-2,13,104,7,1,1,2
-2,14,118,7,1,1,[other]
-2,15,112,7,1,5,2
-2,16,129,7,1,1,[other]
-2,17,112,7,1,5,[other]
-2,18,112,0,1,5,[other]
-2,19,112,7,1,5,[other]
-2,20,112,7,1,[unknown],[other]
-2,21,112,7,1,[unknown],0
-2,22,108,7,1,[unknown],2
-2,23,119,7,[mask],[unknown],0
-2,24,129,0,1,5,2
-2,25,112,7,[mask],8,2
-2,26,104,7,[mask],5,2
-2,27,129,7,1,8,2
-2,28,104,9,1,5,2
-2,29,112,9,1,8,[mask]
-2,30,112,9,1,5,[mask]
-2,31,104,9,1,5,[mask]
-2,32,112,0,[mask],8,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv
deleted file mode 100644
index b312c008..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,104,0,[mask],8,[mask]
-3,4,104,0,[mask],8,[other]
-3,5,104,[mask],[mask],6,0
-3,6,104,0,[mask],8,[unknown]
-3,7,104,0,[mask],8,0
-3,8,104,0,[mask],6,[mask]
-3,9,[other],1,[mask],8,[unknown]
-3,10,129,2,1,6,[unknown]
-3,11,129,2,1,6,[unknown]
-3,12,129,2,1,6,[unknown]
-3,13,129,2,1,5,2
-3,14,129,2,1,5,2
-3,15,129,2,1,5,2
-3,16,129,2,1,5,2
-3,17,112,2,1,5,2
-3,18,107,7,[mask],5,1
-3,19,107,7,[mask],[unknown],1
-3,20,107,7,[mask],[unknown],1
-3,21,107,7,[mask],[unknown],1
-3,22,107,7,[mask],[unknown],1
-3,23,104,7,[mask],[unknown],1
-3,24,104,7,[mask],[unknown],1
-3,25,104,[mask],[mask],[unknown],1
-3,26,104,9,[mask],3,2
-3,27,104,9,[mask],3,2
-3,28,104,9,[mask],3,2
-3,29,104,9,[mask],3,2
-3,30,104,9,[mask],3,2
-3,31,112,9,[other],1,[mask]
-3,32,104,9,[mask],1,[mask]
-4,3,104,9,1,[unknown],2
-4,4,104,7,[mask],[unknown],0
-4,5,[other],9,1,[unknown],1
-4,6,104,9,1,[unknown],0
-4,7,112,9,1,[unknown],1
-4,8,104,7,1,[unknown],2
-4,9,112,9,1,1,[mask]
-4,10,129,7,1,1,[mask]
-4,11,112,7,1,1,[other]
-4,12,112,7,1,5,2
-4,13,108,7,1,[unknown],2
-4,14,129,7,1,5,2
-4,15,112,7,[mask],1,[other]
-4,16,112,7,1,5,[mask]
-4,17,118,7,[mask],5,1
-4,18,104,7,[mask],[unknown],[mask]
-4,19,104,7,[mask],8,1
-4,20,104,0,[mask],8,[mask]
-4,21,104,0,[mask],5,[other]
-4,22,112,7,[mask],3,[mask]
-4,23,104,2,[mask],8,[mask]
-4,24,112,9,[mask],3,1
-4,25,104,2,[mask],3,[mask]
-4,26,107,9,[other],6,[unknown]
-4,27,104,2,[mask],3,2
-4,28,107,9,2,6,[unknown]
-4,29,104,2,[mask],3,1
-4,30,104,9,[mask],3,1
-4,31,107,9,[mask],3,1
-4,32,104,3,[mask],3,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv
deleted file mode 100644
index bfc352bc..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,102,7,[mask],9,1
-5,4,113,7,[mask],[unknown],1
-5,5,104,7,[mask],[unknown],1
-5,6,104,7,[mask],[unknown],1
-5,7,104,7,[mask],[unknown],1
-5,8,104,7,[mask],[unknown],1
-5,9,104,7,[mask],3,1
-5,10,104,9,[mask],3,1
-5,11,104,9,[mask],3,2
-5,12,104,9,[mask],3,2
-5,13,112,9,[mask],1,2
-5,14,104,5,9,1,2
-5,15,104,9,[other],1,0
-5,16,111,[unknown],[other],3,[mask]
-5,17,112,[unknown],[mask],3,2
-5,18,104,2,[mask],1,[mask]
-5,19,110,9,[other],1,[mask]
-5,20,110,2,[other],7,[mask]
-5,21,104,2,[mask],8,[mask]
-5,22,[other],9,[other],8,[mask]
-5,23,104,2,[other],8,[mask]
-5,24,129,9,[other],8,2
-5,25,104,9,[other],1,[mask]
-5,26,107,9,[other],1,[mask]
-5,27,104,2,[other],1,[mask]
-5,28,[other],9,[other],1,[other]
-5,29,107,1,[other],1,[mask]
-5,30,118,1,[other],[unknown],[mask]
-5,31,[other],2,1,[unknown],[mask]
-5,32,[other],9,[other],5,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv
deleted file mode 100644
index 13756617..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,129,9,1,5,[other]
-6,5,129,9,1,5,[mask]
-6,6,129,9,1,5,[other]
-6,7,112,9,1,5,[other]
-6,8,129,7,1,5,[mask]
-6,9,129,2,1,8,[mask]
-6,10,129,7,1,5,2
-6,11,112,7,1,8,2
-6,12,104,7,[mask],[unknown],2
-6,13,104,7,[mask],[unknown],1
-6,14,104,7,[mask],[unknown],1
-6,15,104,7,[mask],[unknown],1
-6,16,104,9,[mask],1,1
-6,17,104,9,[mask],[unknown],[mask]
-6,18,112,9,1,3,[mask]
-6,19,112,9,1,1,[other]
-6,20,112,9,1,1,[mask]
-6,21,111,9,1,5,2
-6,22,112,9,1,1,[mask]
-6,23,118,7,1,5,2
-6,24,112,7,1,1,[other]
-6,25,112,0,1,1,[mask]
-6,26,118,7,[mask],[unknown],0
-6,27,129,7,1,8,[mask]
-6,28,129,0,[mask],5,[other]
-6,29,112,7,[mask],5,[mask]
-6,30,104,7,[mask],5,[mask]
-6,31,104,7,[mask],8,2
-6,32,129,9,1,8,2
-6,33,104,9,1,5,2
-7,3,129,2,1,5,[mask]
-7,4,129,2,1,5,1
-7,5,112,2,1,5,1
-7,6,107,2,1,5,1
-7,7,107,2,[mask],[unknown],1
-7,8,107,7,[mask],[unknown],1
-7,9,107,7,[mask],[unknown],1
-7,10,107,7,[mask],[unknown],1
-7,11,107,7,[mask],[unknown],1
-7,12,107,7,[mask],[unknown],1
-7,13,104,7,[mask],[unknown],1
-7,14,104,7,[mask],[unknown],1
-7,15,104,9,[mask],[unknown],2
-7,16,104,9,[mask],3,2
-7,17,104,9,[mask],3,2
-7,18,104,9,[mask],3,2
-7,19,104,9,[mask],3,2
-7,20,104,9,[other],1,[mask]
-7,21,112,9,[other],1,[mask]
-7,22,104,9,[other],1,[mask]
-7,23,112,9,[other],1,[mask]
-7,24,104,2,6,1,[mask]
-7,25,110,9,[other],1,[other]
-7,26,110,2,[mask],7,[mask]
-7,27,[other],2,[other],8,[mask]
-7,28,[other],1,[other],8,[other]
-7,29,121,1,[mask],7,[mask]
-7,30,[other],2,1,8,[mask]
-7,31,[other],9,1,5,[other]
-7,32,129,9,1,5,[mask]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv
deleted file mode 100644
index ca8bff47..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,104,2,1,5,[mask]
-8,5,129,9,1,5,1
-8,6,104,2,1,5,[mask]
-8,7,129,7,1,5,1
-8,8,112,2,1,5,1
-8,9,107,7,[mask],[unknown],1
-8,10,107,7,[mask],[unknown],1
-8,11,107,7,[mask],[unknown],1
-8,12,107,7,[mask],[unknown],1
-8,13,123,7,[mask],[unknown],1
-8,14,107,7,[mask],[unknown],1
-8,15,118,7,[mask],[unknown],1
-8,16,104,7,[mask],[unknown],2
-8,17,104,9,[mask],1,2
-8,18,104,9,[mask],1,2
-8,19,104,9,[mask],1,2
-8,20,104,9,[mask],1,[mask]
-8,21,104,9,[other],7,[mask]
-8,22,104,9,[mask],7,[mask]
-8,23,104,9,[other],[unknown],[mask]
-8,24,112,9,[other],1,[mask]
-8,25,109,2,[other],1,[mask]
-8,26,104,9,[other],1,[mask]
-8,27,107,9,[other],1,[mask]
-8,28,107,2,[other],1,[mask]
-8,29,[other],9,[other],8,[other]
-8,30,[other],1,3,8,[other]
-8,31,[other],1,1,8,[other]
-8,32,[other],1,1,8,[other]
-8,33,[other],2,1,5,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv
deleted file mode 100644
index a6cb20c5..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-predictions/sequifier-test-hp-search-custom-eval-run-5-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,104,7,[mask],0,1
-9,4,104,7,[mask],0,1
-9,5,104,7,9,0,1
-9,6,104,7,9,3,0
-9,7,104,0,9,3,1
-9,8,104,[mask],[mask],3,0
-9,9,104,0,9,3,1
-9,10,112,[mask],[mask],3,0
-9,11,[unknown],5,9,3,[other]
-9,12,112,0,[mask],3,0
-9,13,[unknown],7,[mask],[unknown],[other]
-9,14,[unknown],7,[mask],[unknown],0
-9,15,104,7,[mask],[unknown],1
-9,16,104,9,[mask],[unknown],2
-9,17,104,9,[mask],[unknown],0
-9,18,104,9,[other],4,[mask]
-9,19,104,9,[other],4,2
-9,20,104,9,[other],1,[mask]
-9,21,112,9,[other],1,[mask]
-9,22,112,2,[other],1,[mask]
-9,23,[other],2,6,[unknown],[mask]
-9,24,[other],9,[other],1,[other]
-9,25,107,1,[other],1,[mask]
-9,26,118,2,6,[unknown],[other]
-9,27,110,1,1,[unknown],[other]
-9,28,110,2,1,[unknown],[other]
-9,29,129,1,1,8,[other]
-9,30,129,2,1,5,[other]
-9,31,129,2,1,5,[other]
-9,32,129,2,1,5,2
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv
deleted file mode 100644
index 4c3bf37e..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-predictions/sequifier-test-hp-search-custom-eval-run-6-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,113,3,2,[other],[other]
-9,4,119,9,2,7,[mask]
-9,5,129,9,9,[other],[other]
-9,6,119,9,9,7,[mask]
-9,7,119,9,9,7,[other]
-9,8,104,9,9,7,[other]
-9,9,129,9,9,2,[other]
-9,10,112,9,9,2,[other]
-9,11,112,9,9,2,[other]
-9,12,112,9,9,7,[other]
-9,13,112,9,9,7,[other]
-9,14,112,9,9,7,[other]
-9,15,112,9,9,7,[other]
-9,16,112,9,9,7,[other]
-9,17,112,9,9,8,[other]
-9,18,112,9,9,7,[other]
-9,19,112,9,9,7,[other]
-9,20,112,9,9,7,[other]
-9,21,112,9,9,7,[other]
-9,22,112,9,9,7,[other]
-9,23,112,9,9,7,[other]
-9,24,112,9,9,7,[other]
-9,25,112,9,9,8,[other]
-9,26,112,9,9,7,[other]
-9,27,112,9,9,7,[other]
-9,28,112,9,9,7,[other]
-9,29,112,9,9,7,[other]
-9,30,112,9,9,7,[other]
-9,31,112,9,9,7,[other]
-9,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
index 0cf0e0be..5cf025b8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.009402489,0.007112399,0.006294906,0.025800984,0.005330445,0.058306154,0.09866103,0.031214684,0.053123873,0.008108479,0.006941985,0.03378066,0.015496237,0.024613176,0.034951206,0.013683666,0.15753025,0.0058497526,0.010813744,0.012419266,0.00899315,0.0433608,0.024071196,0.0070383553,0.061834656,0.040662296,0.013590249,0.08462712,0.007689061,0.02625725,0.025855735,0.011850585,0.024734205
+0.009402493,0.007112395,0.0062949057,0.02580099,0.005330445,0.05830615,0.09866104,0.031214688,0.053123884,0.008108478,0.0069419877,0.033780657,0.015496235,0.024613172,0.03495121,0.013683665,0.15753023,0.0058497493,0.010813737,0.012419264,0.008993149,0.04336079,0.0240712,0.007038355,0.061834626,0.040662292,0.013590248,0.084627114,0.0076890606,0.026257254,0.025855722,0.011850584,0.024734203
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
index 7a95c833..92b890cf 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008010792,0.007172111,0.0058797314,0.027109819,0.005602905,0.055375658,0.075244375,0.029313534,0.04922832,0.006365298,0.006468321,0.034970127,0.01725487,0.02158429,0.023749217,0.016210178,0.18300672,0.006796729,0.011571762,0.008914565,0.0113315275,0.037273988,0.022148373,0.007266352,0.05995564,0.03540441,0.009834666,0.13455547,0.0071732774,0.021706866,0.02286324,0.011315751,0.01934116
+0.008010796,0.0071721114,0.005879732,0.027109826,0.0056029055,0.055375658,0.075244375,0.029313548,0.04922832,0.0063652983,0.006468322,0.034970116,0.017254872,0.021584291,0.023749208,0.016210178,0.1830067,0.0067967293,0.011571764,0.008914566,0.011331529,0.037273984,0.022148374,0.007266353,0.05995564,0.035404414,0.009834667,0.13455541,0.0071732746,0.021706868,0.022863243,0.011315752,0.019341161
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
index c2b44898..a64649f4 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012989526,0.008451789,0.008343688,0.025282253,0.0077611837,0.049804118,0.12995926,0.073096894,0.043786965,0.012898271,0.010859507,0.029631373,0.019566044,0.019631185,0.046036806,0.028859269,0.060753997,0.012488813,0.016959455,0.0091642905,0.0140438145,0.013560966,0.04981232,0.0114713255,0.055660956,0.039123084,0.016607726,0.044842925,0.010157406,0.035077438,0.033839993,0.018593594,0.030883798
+0.012989529,0.008451787,0.008343695,0.025282247,0.0077611892,0.04980413,0.12995923,0.073096894,0.043786965,0.012898268,0.010859504,0.029631373,0.019566048,0.019631185,0.0460368,0.028859274,0.060754,0.012488811,0.016959468,0.00916429,0.014043812,0.013560963,0.049812313,0.011471328,0.055660956,0.03912309,0.016607722,0.044842925,0.010157409,0.03507745,0.033839986,0.01859359,0.030883793
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
index 45b924b2..9a9a1059 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0084459195,0.005478003,0.005521863,0.022377402,0.0057007815,0.048822228,0.13396627,0.047674347,0.044100903,0.008718914,0.0076107793,0.030250654,0.022718882,0.023471467,0.03372159,0.020928407,0.121317275,0.007938454,0.010315681,0.009893803,0.009422824,0.028209142,0.035702728,0.00701184,0.057384826,0.034776427,0.009679542,0.101385765,0.0068074698,0.034072928,0.021294992,0.011075436,0.02420245
-0.01495256,0.011281415,0.011318162,0.04048125,0.013168076,0.043056514,0.0334818,0.020567056,0.03463795,0.00803167,0.010289714,0.013072755,0.027047135,0.024557816,0.022758493,0.027457278,0.042199492,0.008637161,0.033859514,0.014415746,0.032140847,0.015577597,0.018920667,0.013239309,0.03609407,0.24212311,0.0075512324,0.030155562,0.013689231,0.021117214,0.06719196,0.031909913,0.015017748
+0.0084459195,0.005478003,0.005521865,0.0223774,0.0057007815,0.048822243,0.13396621,0.04767435,0.0441009,0.008718918,0.0076107834,0.030250661,0.022718893,0.023471475,0.03372159,0.020928413,0.12131734,0.007938458,0.010315681,0.009893802,0.0094228275,0.02820915,0.035702728,0.00701184,0.05738481,0.034776427,0.009679547,0.10138574,0.0068074698,0.034072928,0.021294992,0.011075441,0.024202444
+0.014952558,0.011281414,0.01131816,0.040481254,0.013168074,0.043056514,0.0334818,0.020567054,0.034637958,0.008031665,0.010289717,0.013072754,0.027047133,0.024557814,0.02275849,0.027457282,0.042199492,0.00863716,0.033859514,0.014415744,0.03214085,0.015577596,0.018920662,0.013239307,0.03609407,0.24212314,0.0075512314,0.030155566,0.013689236,0.021117216,0.06719196,0.03190991,0.015017746
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
index a80d23ab..6948a5f7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010768626,0.0071453005,0.0061826305,0.020636555,0.007476879,0.053042114,0.121259026,0.055434912,0.06060058,0.011440382,0.008968602,0.034554403,0.018097715,0.020440187,0.040972438,0.030373393,0.08881224,0.010173473,0.015603304,0.009588735,0.01265485,0.026746185,0.037209485,0.0092568,0.06076117,0.03545333,0.017796088,0.065680444,0.0107882945,0.024638936,0.028522931,0.013651196,0.025268821
+0.010768625,0.0071453005,0.0061826278,0.020636555,0.007476882,0.053042125,0.12125908,0.055434898,0.06060058,0.011440382,0.008968601,0.034554392,0.018097715,0.020440182,0.040972427,0.030373385,0.08881222,0.010173463,0.015603303,0.009588735,0.012654849,0.026746185,0.037209474,0.0092568,0.06076114,0.03545333,0.017796088,0.065680444,0.01078829,0.024638936,0.028522924,0.013651196,0.02526882
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
index 9e3adcef..e2c34a6e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008686484,0.005960353,0.008075502,0.039276745,0.0097908955,0.05850304,0.05206207,0.01889337,0.048300233,0.00518526,0.006793675,0.025432974,0.031197455,0.021319006,0.02276724,0.014915394,0.15458107,0.0053153243,0.016606817,0.011906255,0.0143988915,0.038660195,0.019179536,0.007658729,0.052357905,0.0989535,0.005735836,0.11204717,0.008564646,0.016890625,0.025278362,0.020940961,0.013764397
-0.008197727,0.007059469,0.0055742837,0.023094872,0.0057917624,0.051699355,0.11453689,0.0425388,0.043388393,0.00957253,0.0075618536,0.034410797,0.019690048,0.022121822,0.03347103,0.017320948,0.1331481,0.007196628,0.0096377,0.010214877,0.009963169,0.045684543,0.02452887,0.007794731,0.060744867,0.027903598,0.010524433,0.115211345,0.0076923347,0.03072128,0.020051872,0.0100175515,0.022933505
+0.008686484,0.005960354,0.008075498,0.039276745,0.009790896,0.05850304,0.052062083,0.018893363,0.048300244,0.0051852576,0.006793672,0.025432982,0.031197459,0.021319004,0.022767233,0.014915387,0.1545811,0.0053153215,0.01660681,0.011906256,0.014398892,0.03866019,0.019179536,0.0076587293,0.052357905,0.098953515,0.005735833,0.112047225,0.008564644,0.016890615,0.025278356,0.02094096,0.013764391
+0.008197728,0.0070594703,0.005574284,0.02309487,0.005791761,0.051699366,0.11453691,0.0425388,0.04338838,0.009572528,0.0075618555,0.034410793,0.019690046,0.022121828,0.033471037,0.017320944,0.13314813,0.0071966257,0.009637701,0.010214874,0.009963171,0.045684554,0.024528876,0.007794729,0.060744863,0.027903598,0.010524435,0.11521132,0.007692336,0.03072128,0.020051876,0.010017549,0.022933504
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
index f309bc23..b1992ea3 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.011856947,0.007223844,0.0070571597,0.032826528,0.008080109,0.040391415,0.09255855,0.039391916,0.052117735,0.007856196,0.009969194,0.01840535,0.020338742,0.020724263,0.03398534,0.041261356,0.058055624,0.012493181,0.027867151,0.0117967855,0.0220204,0.011738093,0.035083402,0.009734925,0.05840189,0.1281792,0.009151511,0.030461226,0.008224331,0.028107058,0.059158992,0.020592097,0.024889482
+0.011856951,0.0072238427,0.007057155,0.03282652,0.008080111,0.040391415,0.09255859,0.03939192,0.052117724,0.007856198,0.009969188,0.018405346,0.020338742,0.02072426,0.03398535,0.041261356,0.058055628,0.012493185,0.02786714,0.011796777,0.022020405,0.011738096,0.03508339,0.009734918,0.05840191,0.12817924,0.009151513,0.03046122,0.008224325,0.028107053,0.059158992,0.020592093,0.024889484
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
index d6b436bb..9b35f4f2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010474039,0.006521543,0.0074160905,0.023902066,0.0066056135,0.057496417,0.12374015,0.03752914,0.047728445,0.009753353,0.0080655515,0.027190102,0.014648493,0.026415588,0.031456366,0.0145098185,0.13790837,0.006406579,0.010360853,0.01592334,0.009660522,0.059283886,0.031192504,0.007285326,0.06007168,0.037452072,0.011793553,0.05825485,0.008734745,0.027273804,0.028270483,0.011573875,0.02510079
+0.010474039,0.0065215426,0.00741609,0.023902066,0.006605613,0.057496436,0.12374021,0.03752914,0.047728438,0.009753358,0.008065551,0.027190097,0.014648493,0.02641559,0.031456374,0.0145098185,0.13790835,0.0064065787,0.010360852,0.01592333,0.009660522,0.05928386,0.031192506,0.007285326,0.060071684,0.037452076,0.011793552,0.058254838,0.008734744,0.027273806,0.028270485,0.011573874,0.025100792
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
index 4cab3548..b53fb5b7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0018376797,0.0023137012,0.0024238364,0.009155929,0.0033391858,0.037196394,0.11448541,0.03619819,0.020103797,0.0032478096,0.0022368645,0.042428207,0.030106904,0.017886922,0.041678857,0.017872667,0.17781384,0.002194447,0.005525137,0.01080326,0.0041624387,0.026048908,0.038844123,0.0014777568,0.105031826,0.010042691,0.004704572,0.116676375,0.0022287897,0.06710731,0.0103819845,0.011351856,0.023092275
+0.0018302883,0.0023094334,0.0024177611,0.009174009,0.0033216283,0.037184585,0.11439202,0.036135826,0.020121714,0.003243756,0.002227362,0.042347167,0.029958114,0.017858125,0.041660704,0.017803762,0.17849426,0.00218419,0.00554009,0.01079341,0.004177411,0.025839655,0.038697787,0.001474616,0.10490975,0.010089096,0.0046706814,0.1171136,0.0022260093,0.067104094,0.010380509,0.011327853,0.022990765
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
index 4526ff20..e16fcc42 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0048656277,0.007918645,0.0054325806,0.0121447295,0.007898342,0.017814836,0.075257584,0.011748432,0.021347238,0.00770395,0.0054659205,0.023766177,0.042818088,0.014586231,0.021619631,0.062280655,0.065641224,0.0032016544,0.038712315,0.010045757,0.03476402,0.009708087,0.06701612,0.015924746,0.021149337,0.11031023,0.00664596,0.060393848,0.006036425,0.027078018,0.14114997,0.010524234,0.02902941
+0.004845992,0.007889446,0.0054141493,0.012189367,0.007869112,0.017851332,0.07520539,0.011662964,0.021426132,0.0076573635,0.0054324535,0.023717629,0.042729106,0.014602124,0.021532323,0.06217893,0.06529054,0.0031855546,0.038700376,0.0100185275,0.034709904,0.009724645,0.06670463,0.015934361,0.021101989,0.11129304,0.006598829,0.06060536,0.006012587,0.026912445,0.14151993,0.0104822675,0.029001156
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
index a6528ce7..685739bd 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0023688173,0.0032759958,0.0024291654,0.0072109625,0.0038664865,0.03368376,0.037551492,0.020473365,0.020303426,0.00320672,0.002927277,0.019001897,0.0069159083,0.008086997,0.020394612,0.011339872,0.576207,0.0024872613,0.0064319503,0.010710103,0.005667849,0.009287372,0.02400277,0.002055522,0.026548946,0.011598315,0.0042506577,0.04309646,0.002323081,0.030272912,0.01817002,0.013025035,0.010827967
+0.0023620205,0.0032647604,0.002421969,0.0072277756,0.003854628,0.033725373,0.037664812,0.020529818,0.02037918,0.0032004176,0.0029194388,0.019040832,0.0069631515,0.008101235,0.020486154,0.011356625,0.5755077,0.0024824778,0.0064296694,0.010709552,0.0056811264,0.009269812,0.023953447,0.0020496456,0.026671877,0.011577401,0.0042310487,0.043356534,0.0023190514,0.030348403,0.018102089,0.012962391,0.010849504
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
index 29abb8a1..f1b2465f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0066867806,0.009659095,0.006985347,0.018342886,0.008343686,0.02827222,0.13270424,0.009605769,0.023481674,0.0065346137,0.008042574,0.01586629,0.06682908,0.029205475,0.01841571,0.035887375,0.0029870616,0.00378063,0.021729896,0.010461546,0.025363758,0.0364591,0.043761298,0.02097016,0.028998114,0.09809047,0.012051953,0.032568526,0.00945276,0.014552994,0.11844721,0.013847485,0.08161431
-0.0062105553,0.008430051,0.0066421693,0.034530688,0.009992443,0.030213395,0.02225522,0.0048418874,0.034004305,0.006981886,0.0063810414,0.009619785,0.016620198,0.025754781,0.010917337,0.020015128,0.0059674657,0.004531538,0.036805414,0.011499066,0.03323312,0.025179174,0.01618241,0.030543782,0.015805239,0.36270088,0.008696667,0.013364138,0.008154805,0.006193264,0.11358851,0.015613565,0.03853005
+0.0066247783,0.009592915,0.006928904,0.018345714,0.008288779,0.028367922,0.13309619,0.009587395,0.023486814,0.0064903456,0.007973895,0.01584719,0.06669925,0.029178161,0.018443769,0.035900217,0.0030158022,0.003753964,0.021668421,0.010438245,0.025283378,0.036490586,0.043768942,0.020859156,0.029089404,0.09840058,0.011953055,0.03282844,0.009378146,0.014542032,0.118312456,0.013809641,0.08155552
+0.006188416,0.008413583,0.006625596,0.034453783,0.009931466,0.030127268,0.022238078,0.0048153163,0.033921164,0.006972913,0.0063575697,0.009609076,0.016657123,0.025657328,0.010869502,0.020110097,0.005996566,0.0045159073,0.036867622,0.01148467,0.03323851,0.0249987,0.016155457,0.03066806,0.015686167,0.36310777,0.008656854,0.013391952,0.008103484,0.0061835432,0.113910325,0.015572561,0.038513523
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
index b868c8da..84eebabe 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0040303203,0.0029316142,0.0041374513,0.010778411,0.0056618657,0.032110527,0.1331622,0.03412711,0.018952146,0.00459283,0.0029058945,0.023700856,0.08413434,0.023126962,0.058894433,0.033277348,0.027406437,0.0037933602,0.005311759,0.011036976,0.007056581,0.04888226,0.041817285,0.0038434942,0.10182688,0.010030801,0.0061456566,0.10456378,0.0049827173,0.06323838,0.008559678,0.016620507,0.058359172
+0.004018745,0.0029350165,0.0041301805,0.010802765,0.005631038,0.032027874,0.13366385,0.03408293,0.018921215,0.00460021,0.002905895,0.023617709,0.08428678,0.023087565,0.058966562,0.03335568,0.027337305,0.0037721738,0.0053366604,0.011020286,0.0070758117,0.04848865,0.041665643,0.003843431,0.10164474,0.010063016,0.006108716,0.10480515,0.004982868,0.063377164,0.008564493,0.016542716,0.05833719
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
index c5d59b6b..85d8f3a9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0028121886,0.004768915,0.0026038273,0.024106294,0.00428334,0.04001045,0.042205393,0.0055848244,0.051259898,0.0029990096,0.0024799826,0.018063348,0.013305797,0.016817851,0.013413405,0.02079274,0.06953616,0.002443421,0.023023874,0.012457441,0.017435335,0.022667246,0.025499817,0.009044353,0.021868205,0.2366301,0.006028528,0.06351859,0.0038850654,0.008795736,0.16167547,0.015960602,0.03402277
-0.0049432945,0.008161541,0.005641948,0.0334701,0.006687145,0.020943755,0.048687987,0.0067313584,0.026358342,0.0061188038,0.00532264,0.014945954,0.022759516,0.02219019,0.0143439155,0.022385858,0.0037059456,0.0030287185,0.029329462,0.010064615,0.025855048,0.019193592,0.026798103,0.025868736,0.018704478,0.27705717,0.008246222,0.02213319,0.0063955006,0.008136024,0.18003964,0.013626517,0.05212471
+0.0027887959,0.004733092,0.0025837973,0.024125917,0.0042522578,0.04005393,0.042278986,0.0055801696,0.051247988,0.0029797198,0.002457765,0.018063435,0.013389846,0.016801085,0.013436619,0.02084646,0.069999106,0.0024221565,0.02295687,0.012420337,0.017399706,0.022620559,0.02542745,0.008995221,0.021923762,0.23663002,0.0059561864,0.06404682,0.0038476116,0.008797948,0.16114126,0.01581707,0.033978023
+0.004896839,0.008109383,0.005606582,0.033393413,0.0066475375,0.021049256,0.049079787,0.0067314794,0.026396584,0.0060873907,0.00528243,0.014969497,0.022778716,0.0222018,0.014368257,0.022405619,0.003741044,0.0030103682,0.029169861,0.010044832,0.025707247,0.019271096,0.026915964,0.025671674,0.018815916,0.27695572,0.008184225,0.022288233,0.0063604494,0.008164105,0.17993744,0.013626392,0.052130867
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
index 420af8d1..996b347b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008235056,0.006630053,0.008237326,0.0065348125,0.013050176,0.030923633,0.14479923,0.043974973,0.015504926,0.009599815,0.00887134,0.027437832,0.024815973,0.021933725,0.078563236,0.020613033,0.04735088,0.005203651,0.006880082,0.015798736,0.006727233,0.021745805,0.099310055,0.006908331,0.07205278,0.00884878,0.012542185,0.016972749,0.0075840293,0.11026282,0.012869484,0.040418718,0.038798563
+0.008245247,0.0066397716,0.00823672,0.006531572,0.013031284,0.030800205,0.1454552,0.043889422,0.015470334,0.009597873,0.008879567,0.027317327,0.024919743,0.02190608,0.078744985,0.020652225,0.046914577,0.0051893904,0.0069066836,0.015767885,0.0067592845,0.021571003,0.099289015,0.0069529354,0.07191968,0.008895552,0.012519008,0.016933145,0.00758302,0.11038498,0.012901287,0.040280648,0.03891432
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
index 0010dfae..c46f46fb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0014310861,0.002243976,0.0017065166,0.016686594,0.0020667855,0.053174898,0.095879674,0.014162464,0.032807916,0.0018454238,0.0015201237,0.026181472,0.039513864,0.022075271,0.021645933,0.0206909,0.15128985,0.0018012328,0.008061768,0.0071523706,0.007267346,0.053755138,0.028755592,0.00226818,0.050142813,0.034830853,0.0034951526,0.19666047,0.0019271312,0.020879522,0.041771542,0.010048885,0.02625917
+0.0014336656,0.002257306,0.0017123552,0.016647093,0.002066613,0.053203385,0.09665847,0.014232341,0.03271058,0.0018544742,0.001527125,0.025984295,0.03937578,0.022107681,0.021644413,0.02069751,0.152486,0.0017975577,0.008073777,0.0071796924,0.007284085,0.053035576,0.028747264,0.0022616726,0.05012752,0.0346748,0.0034840752,0.19593439,0.0019345195,0.02107136,0.041585583,0.010048055,0.02616101
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index e820f648..f47a7ab9 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.013453301,0.013558816,0.010155084,0.025183218,0.019921565,0.05383606,0.02112372,0.025011655,0.029356034,0.012249369,0.015066982,0.046228647,0.030096248,0.030052193,0.019999536,0.022618225,0.031463686,0.013989219,0.017822728,0.064182125,0.025011655,0.059127342,0.0850275,0.030607535,0.049790215,0.023750037,0.0069251657,0.04524605,0.009766049,0.06673885,0.026443383,0.018245382,0.03795245
-0.016586587,0.009750751,0.00887816,0.044301387,0.015043382,0.030987818,0.10565262,0.038190182,0.021930778,0.0126184095,0.019467577,0.031093914,0.05049505,0.026299018,0.033277556,0.033669822,0.030034423,0.0084716,0.018576091,0.017899383,0.01901661,0.06408156,0.05312548,0.014242788,0.08489428,0.020441733,0.012520213,0.037671603,0.0095995795,0.01046103,0.04588656,0.037451517,0.017382594
-0.007844531,0.012341179,0.0109336795,0.01224514,0.009994192,0.041102406,0.04862002,0.027862249,0.021365335,0.014428301,0.014204611,0.06122185,0.02423375,0.034409486,0.07799866,0.035467107,0.07589478,0.012055296,0.012633842,0.013554134,0.017437879,0.038687624,0.023811487,0.010680402,0.07327296,0.05277653,0.017000694,0.03757064,0.030218352,0.0226325,0.028972154,0.02364927,0.05487891
+0.013400168,0.013558126,0.010154567,0.02523116,0.019998511,0.053623423,0.021163935,0.025010373,0.029211547,0.012248745,0.01518438,0.046046063,0.03008736,0.030080015,0.01995949,0.022617064,0.03151204,0.014043256,0.017821819,0.06417883,0.02503481,0.059124317,0.085355915,0.030591028,0.049787663,0.023748823,0.0069248127,0.04524373,0.009765551,0.06673543,0.026364677,0.018315857,0.037876457
+0.016588878,0.0097520985,0.008879387,0.044221062,0.015104345,0.03090144,0.10566724,0.03819547,0.02193381,0.012620153,0.01943228,0.031098215,0.05050204,0.026276983,0.033282157,0.03360877,0.030089956,0.0084727695,0.018578662,0.017901855,0.018945092,0.06434126,0.05313283,0.014244754,0.08490602,0.020364854,0.012521942,0.037676815,0.009600906,0.010462474,0.045892905,0.037383612,0.017418982
+0.007843096,0.012338922,0.010931679,0.012195169,0.010070735,0.041175235,0.048801392,0.027925251,0.021403193,0.014369421,0.014202012,0.060972024,0.024229323,0.03443681,0.07798441,0.03546063,0.0761779,0.012006101,0.012582285,0.013498821,0.017434688,0.038529754,0.023783902,0.010678447,0.07325957,0.052973416,0.016931316,0.037563775,0.030175973,0.022628365,0.028945653,0.023621868,0.054868884
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 8cfb24ae..9cd7eb50 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.019366598,0.008988731,0.01826444,0.021583911,0.036751345,0.063376725,0.036252305,0.017057614,0.04511672,0.009872189,0.015112207,0.037917957,0.03821535,0.024958435,0.03297588,0.025031663,0.03541252,0.018371774,0.027673723,0.016891846,0.03425602,0.014533268,0.03134313,0.03208641,0.050135113,0.10865299,0.012577496,0.058385372,0.009719135,0.03555112,0.03950552,0.011186656,0.012875763
-0.013347994,0.007877592,0.011918416,0.012393192,0.015243887,0.053833574,0.053833574,0.022816842,0.04135633,0.01283664,0.01146183,0.051569402,0.037071574,0.019804265,0.054895345,0.030598652,0.032779507,0.013452683,0.011825667,0.01938332,0.03758189,0.035721082,0.027969431,0.01409829,0.10959972,0.022375524,0.01506629,0.0386237,0.013825605,0.016676864,0.05299896,0.057082128,0.030080168
-0.015875753,0.01241244,0.019149808,0.021741997,0.01581386,0.039911628,0.030156313,0.041339725,0.060857967,0.017572856,0.011569666,0.094258666,0.045049477,0.01848827,0.052462846,0.019565681,0.055846408,0.022941712,0.021279868,0.01456841,0.028019704,0.031789266,0.013212974,0.019489402,0.0485202,0.059216425,0.014740137,0.050256327,0.02307653,0.019912649,0.020029668,0.019489402,0.021384027
+0.019318607,0.008983985,0.018183628,0.021572521,0.036731947,0.06334328,0.03623317,0.017048607,0.045092907,0.009866976,0.015104229,0.037897944,0.038269855,0.024981832,0.032958474,0.025055127,0.035532355,0.018362077,0.027645616,0.016817106,0.034171134,0.014525594,0.031387832,0.03197566,0.0499133,0.10944736,0.012570854,0.058354557,0.009714004,0.035532355,0.039407626,0.011180749,0.012818792
+0.013353927,0.007942907,0.0119237155,0.012398702,0.015250664,0.05385752,0.05385752,0.022849292,0.04121342,0.012842346,0.011466925,0.051592343,0.0370157,0.01989062,0.055134714,0.030552533,0.0328582,0.0134586645,0.0118772285,0.019391943,0.037598606,0.03587684,0.028024603,0.014104559,0.10879518,0.022363625,0.015072988,0.03864088,0.013885888,0.016684277,0.053022537,0.05710752,0.030093549
+0.015880968,0.012368109,0.019156104,0.021812957,0.015819054,0.039846845,0.030019287,0.041434165,0.06087797,0.017612996,0.0115734665,0.094289646,0.0448886,0.01845826,0.05248009,0.019610377,0.05586477,0.022994122,0.02130766,0.014630232,0.028104289,0.031861883,0.013217314,0.019533923,0.04853615,0.059004948,0.014687494,0.05007685,0.023129247,0.019997157,0.019958138,0.019533923,0.021432877
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 1f2ed7bf..4e2e284d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012260135,0.01367717,0.018951818,0.011836603,0.024954304,0.079014726,0.037033457,0.020253064,0.029310184,0.015080225,0.017873267,0.023197588,0.015742326,0.043296497,0.039191596,0.022287127,0.037763875,0.009736525,0.017561834,0.025771594,0.034925878,0.030522447,0.022330698,0.016692288,0.13546278,0.09237757,0.011836603,0.024109898,0.01633749,0.026002275,0.041637834,0.016823206,0.016147152
-0.011046612,0.008305911,0.0069942786,0.009485649,0.013481825,0.08194348,0.048930686,0.050091043,0.026681205,0.0070491354,0.012323385,0.032058205,0.024869617,0.049314454,0.040963046,0.048172083,0.04605599,0.008050365,0.012420039,0.019124145,0.031980034,0.023615165,0.035277884,0.020437222,0.04724035,0.053113755,0.017480887,0.061613034,0.010176616,0.028699104,0.086549565,0.010256432,0.016198784
-0.015744025,0.013519284,0.00962413,0.008970675,0.017563729,0.06551258,0.0379899,0.038890805,0.030600358,0.007264674,0.013839886,0.02553013,0.017910143,0.035237998,0.062268827,0.09420972,0.030697633,0.007854985,0.018227752,0.01840663,0.032877516,0.020058408,0.021103306,0.026186656,0.08850185,0.0299061,0.016371196,0.045735154,0.0137859285,0.039580476,0.059883345,0.016243795,0.019902313
+0.012296571,0.013664337,0.018971052,0.011825497,0.024949158,0.078940585,0.03707104,0.020234061,0.029225545,0.015066075,0.017891407,0.023130601,0.015666239,0.043425173,0.039078426,0.022331543,0.037581354,0.009765461,0.017545355,0.02581035,0.034893107,0.030523602,0.022266217,0.016709229,0.13533565,0.09301475,0.011825497,0.02400509,0.016386043,0.025968367,0.041598767,0.016840281,0.01616354
+0.011091116,0.008306863,0.0069950796,0.009486735,0.013483369,0.0819529,0.049127843,0.049901493,0.026658228,0.0070499424,0.012324796,0.03199933,0.024921104,0.049320128,0.04104785,0.048177626,0.046151336,0.008051287,0.012421461,0.019089023,0.03210889,0.023594828,0.03528194,0.020479532,0.047245786,0.053119864,0.01741473,0.06137989,0.010177781,0.028621495,0.08655952,0.010257605,0.01620064
+0.015801238,0.013568414,0.009696907,0.008968174,0.017558832,0.065494314,0.037905198,0.038804095,0.030606769,0.00731961,0.013836027,0.02557291,0.017870212,0.035125118,0.06249511,0.09418346,0.030666607,0.0078527955,0.018222671,0.018401498,0.032900464,0.02009202,0.021097422,0.026204934,0.08847717,0.029846711,0.01636663,0.045633186,0.013782086,0.0396468,0.05986665,0.016239265,0.019896762
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index ab6b8364..fffebf7c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.01434968,0.018246295,0.018246295,0.011062379,0.030698866,0.073096424,0.03732033,0.018661758,0.03649588,0.015284481,0.015576431,0.020900974,0.020248648,0.022526698,0.037143078,0.029393356,0.03404698,0.013014595,0.01981463,0.028412612,0.050005887,0.025869979,0.0187521,0.019310784,0.14371982,0.056729846,0.010097493,0.031495996,0.013216318,0.018000763,0.05034006,0.014573876,0.0333467
-0.01807688,0.009905313,0.01062685,0.029172484,0.020458804,0.07633914,0.046557073,0.04852078,0.029202323,0.015531968,0.016718382,0.021956462,0.1231266,0.051788602,0.027899088,0.03077986,0.022778912,0.0100612985,0.016242629,0.027246153,0.033163358,0.017325047,0.022083469,0.016141823,0.05652651,0.031496324,0.012720227,0.037182026,0.011353725,0.025973134,0.047310814,0.022188853,0.013545157
-0.018902939,0.0145501075,0.01169134,0.027147124,0.034455173,0.08530861,0.038063984,0.026669418,0.023067255,0.01891448,0.025368463,0.013871527,0.023945192,0.11464649,0.021188384,0.043950647,0.02112382,0.019579314,0.019284002,0.02452501,0.021680264,0.017422704,0.021336326,0.018038407,0.026665552,0.06965661,0.020803943,0.06600996,0.014377083,0.038264383,0.020744508,0.01648542,0.022261553
-0.016216656,0.01308145,0.014939484,0.025020843,0.017754074,0.057443947,0.035689533,0.043020573,0.033728357,0.014948606,0.010307934,0.024133276,0.07651071,0.026627235,0.044945393,0.018055689,0.031882733,0.012678976,0.014916706,0.023783064,0.031886622,0.03319329,0.015595406,0.023117555,0.07884793,0.06692554,0.014312897,0.06517998,0.01608653,0.026792346,0.0368025,0.01140845,0.024165705
-0.011208817,0.009184176,0.007265281,0.026254857,0.011393673,0.05732051,0.029912245,0.031399224,0.026623994,0.00926186,0.013679789,0.07865209,0.036882624,0.030619256,0.050721206,0.02077786,0.037872672,0.0077945087,0.010975259,0.04301501,0.0168655,0.06145841,0.05736951,0.019180054,0.083752826,0.028208211,0.006933426,0.045228165,0.007977403,0.060601793,0.02146246,0.017494582,0.022652796
-0.017393991,0.019215824,0.0112304045,0.03964368,0.018711211,0.08226599,0.036385193,0.03245042,0.02884637,0.014205229,0.018300092,0.028105145,0.035746835,0.036245,0.021434186,0.035585742,0.0507631,0.011632246,0.033574894,0.023546595,0.024087338,0.01569116,0.026756804,0.050531257,0.07852272,0.03570976,0.005625673,0.044787344,0.014193096,0.04060556,0.028245514,0.018262153,0.021699423
+0.014360843,0.01826049,0.018224861,0.011053937,0.030745259,0.07321582,0.037249196,0.018730072,0.03652873,0.01522744,0.015527775,0.020976687,0.02029163,0.02248273,0.037103977,0.029358827,0.034015287,0.013075697,0.019821575,0.02842257,0.050124273,0.02578917,0.01876669,0.019362409,0.14334963,0.056798242,0.010143663,0.0315205,0.01322983,0.017977389,0.05032045,0.014586994,0.03335737
+0.018110253,0.010020984,0.010625695,0.02915285,0.020481568,0.07639609,0.04651793,0.048370995,0.029231247,0.0155208055,0.016716566,0.021952067,0.123038106,0.0518945,0.02795571,0.030791549,0.022826537,0.01009958,0.016265664,0.027199995,0.033147614,0.01734856,0.022102643,0.016139083,0.056551423,0.031475604,0.012667712,0.037159834,0.011310993,0.02592908,0.047250483,0.02221083,0.013537486
+0.01885988,0.014545349,0.01173326,0.02711961,0.03438508,0.08535359,0.038023666,0.026703235,0.023040008,0.01885988,0.025403539,0.013825159,0.023957819,0.11485581,0.02122545,0.043936264,0.0211427,0.019534715,0.019307127,0.024442276,0.02162296,0.017408503,0.021371044,0.017996224,0.02658371,0.06966357,0.020835249,0.066214286,0.014319844,0.038247116,0.020713525,0.01648204,0.02228757
+0.016209159,0.013075401,0.014932577,0.02497076,0.017732874,0.057466473,0.035681743,0.04312454,0.03374982,0.01487436,0.010303169,0.024190675,0.0761307,0.026587727,0.044930097,0.018082622,0.03186021,0.012623705,0.014932577,0.023722786,0.03187577,0.03325904,0.015527423,0.023105456,0.07916341,0.06692318,0.014304532,0.06537291,0.016083017,0.026770102,0.036814403,0.011404567,0.024214312
+0.011247538,0.009215903,0.007290379,0.026253665,0.011424662,0.05734327,0.029875759,0.031475265,0.026511302,0.009361031,0.013727046,0.07868573,0.03687926,0.03061506,0.05040797,0.020727834,0.037901588,0.007882779,0.010986989,0.042948123,0.016884554,0.061761133,0.057119712,0.01920759,0.08408835,0.028166069,0.007011089,0.045097243,0.007944604,0.060566563,0.021427615,0.017420528,0.022543838
+0.017470723,0.01926294,0.011235971,0.03967968,0.018706782,0.08173608,0.03637658,0.032417256,0.028888872,0.014148223,0.018380824,0.028192118,0.035812616,0.036341075,0.021447431,0.03560339,0.05094972,0.011683561,0.03354442,0.023486488,0.024043452,0.01572195,0.02673089,0.050553225,0.078604825,0.035672996,0.005649801,0.04478768,0.014148223,0.04062066,0.028205885,0.01823778,0.021657905
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index 934e0e29..d521334c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010731453,0.008589347,0.008724609,0.021720512,0.01064794,0.062239602,0.038269885,0.05600958,0.025929496,0.009886278,0.010731453,0.033510234,0.021114009,0.029029025,0.101420246,0.022333529,0.12233617,0.0081322305,0.015133775,0.010731453,0.015312167,0.020029424,0.028453656,0.015312167,0.065481834,0.028537137,0.0071766684,0.026116919,0.010441991,0.10301738,0.022019511,0.016718876,0.01416144
-0.009332455,0.014341904,0.009259829,0.039909847,0.015148069,0.035705153,0.03433731,0.10597313,0.02412395,0.01255819,0.01005145,0.06503362,0.027943308,0.02745641,0.02977456,0.01569009,0.18454129,0.009045325,0.017988779,0.02082663,0.013007542,0.023807997,0.029949531,0.012363493,0.03598519,0.025091063,0.0070445063,0.020107165,0.01634701,0.030421168,0.033937268,0.018200824,0.024696033
-0.0109061785,0.0068784403,0.0077942936,0.033202026,0.011385017,0.01428001,0.043134782,0.029179327,0.029015938,0.011978094,0.015992854,0.041807663,0.070288695,0.025984332,0.04887811,0.056921516,0.06500641,0.008232416,0.014906977,0.010086566,0.017806504,0.044417188,0.06178772,0.016565101,0.09203254,0.017326204,0.007613739,0.040600598,0.01325854,0.019178364,0.023705257,0.06786054,0.021988103
+0.010735215,0.008592358,0.008727668,0.021791875,0.0106101455,0.062261418,0.038208604,0.056248505,0.025929088,0.009889743,0.010735215,0.033587515,0.02101853,0.029095972,0.100666255,0.022330452,0.12237905,0.008135081,0.015139079,0.010693362,0.015317533,0.020095231,0.02846363,0.015257817,0.065761164,0.028561084,0.0071791834,0.026119694,0.010486534,0.10305349,0.02200573,0.016757432,0.014166404
+0.009332249,0.014285676,0.009259625,0.03990897,0.015147736,0.035774168,0.034269553,0.10597081,0.024182387,0.012508955,0.010051228,0.065032184,0.027969994,0.02748263,0.029919641,0.015689746,0.1845372,0.009045126,0.017918251,0.02084652,0.013007255,0.023854017,0.029978134,0.0123632215,0.035984397,0.025164125,0.007044351,0.020106722,0.016314754,0.030361144,0.03393652,0.018129466,0.024623245
+0.010900443,0.006874823,0.0077901953,0.03331444,0.01137903,0.014272501,0.043027967,0.029167539,0.029025465,0.011971795,0.015922125,0.04186736,0.070526674,0.025996035,0.048852395,0.056891568,0.064972214,0.008228086,0.014899139,0.010081262,0.01779714,0.044480614,0.061996922,0.016556391,0.091984116,0.01735095,0.0076097352,0.04057924,0.013251569,0.019205749,0.023646556,0.06756041,0.0220195
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 1593701e..0109cb4c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.011400348,0.008775162,0.008843986,0.014321295,0.012677706,0.08820507,0.044946373,0.040066205,0.019908346,0.008184344,0.014077552,0.03491359,0.020098807,0.075348005,0.033000086,0.041619375,0.031996496,0.008308156,0.014333318,0.024208302,0.029419037,0.049255587,0.036585778,0.027604597,0.05521417,0.034725506,0.015523573,0.052534807,0.012420362,0.048949987,0.04967562,0.016328253,0.016530307
-0.012443908,0.008755383,0.012939616,0.012758654,0.012164551,0.0392054,0.02605592,0.025547082,0.024999686,0.010902902,0.011553777,0.05659787,0.019704089,0.029325426,0.03885284,0.025211142,0.03223767,0.010853769,0.016016373,0.023959927,0.01762386,0.28838542,0.014211388,0.012203965,0.06491157,0.017034778,0.0113316905,0.026981654,0.012987089,0.01789706,0.021515755,0.019766726,0.025063088
-0.01905886,0.015464511,0.014814094,0.01604256,0.025715468,0.038806655,0.051542424,0.019265728,0.03616295,0.019714328,0.03132796,0.026078215,0.05771773,0.025969015,0.039216194,0.0347988,0.042813707,0.016017856,0.028096773,0.035178926,0.029812263,0.0413246,0.026982497,0.011719628,0.06307021,0.0423485,0.016723165,0.029562214,0.02267167,0.01413746,0.04577323,0.034625072,0.027446749
-0.01243457,0.01243457,0.011818892,0.044152383,0.012939381,0.08657686,0.03272077,0.07462879,0.02063925,0.0156059675,0.016217737,0.03191114,0.09086477,0.054746,0.03436231,0.031505693,0.05321489,0.010677479,0.01619301,0.019355675,0.0146282725,0.0225271,0.012729448,0.013341161,0.069316395,0.03227055,0.009992678,0.033357985,0.018552922,0.042900607,0.015252949,0.013815998,0.01831384
-0.017431313,0.024062537,0.015443287,0.035962105,0.04309029,0.05766668,0.06726299,0.017111776,0.023123773,0.01796051,0.017127581,0.0300626,0.02393364,0.038289644,0.024756758,0.03822251,0.036356602,0.01729566,0.031027302,0.015982209,0.018346079,0.019839263,0.018921515,0.012098662,0.166915,0.024002396,0.012779513,0.041348573,0.01895388,0.033669606,0.0148399975,0.013130568,0.01298512
-0.011333878,0.02142407,0.01259456,0.019844584,0.034227207,0.020909905,0.02135618,0.01781424,0.037632555,0.015024115,0.014667127,0.07213027,0.01998162,0.028571296,0.03012072,0.052715294,0.05604668,0.019621471,0.016529996,0.038723007,0.025725903,0.045199998,0.065990776,0.016491702,0.05237854,0.02253048,0.012113549,0.030353254,0.026191704,0.032936145,0.027407028,0.022231326,0.059180837
+0.011403347,0.008811824,0.00895059,0.0143030025,0.0126717575,0.08830369,0.04475013,0.040192187,0.019935472,0.008213523,0.014026358,0.034953695,0.020052625,0.07523569,0.03296447,0.041630324,0.031911284,0.008342868,0.014358983,0.024188083,0.029398052,0.049148407,0.036631156,0.027623659,0.055259038,0.03464783,0.015465233,0.052522976,0.012426664,0.048956797,0.049922384,0.016270857,0.016527088
+0.012426136,0.008742878,0.012921137,0.012720813,0.012138285,0.039182864,0.026063208,0.025609069,0.024954835,0.010880688,0.011537277,0.056346525,0.019663937,0.029260758,0.03887794,0.025175132,0.032230943,0.010838268,0.016017923,0.023899434,0.01762664,0.28952438,0.014246635,0.012138285,0.064601615,0.01705099,0.011358408,0.026929962,0.012971709,0.017834418,0.021491416,0.019721631,0.025015835
+0.019027088,0.015468914,0.014760541,0.015959948,0.02570396,0.038775083,0.05162007,0.01921381,0.03617766,0.019746438,0.031370368,0.02615978,0.057698935,0.026032358,0.0391556,0.03479172,0.042711042,0.016022416,0.028106486,0.035253435,0.029751668,0.04115511,0.026937515,0.01172225,0.063369885,0.042295974,0.016725883,0.029570632,0.022683663,0.014084606,0.045822255,0.034656078,0.027468812
+0.012428584,0.012428584,0.011813202,0.044235602,0.012873298,0.0866091,0.03271299,0.07466172,0.020611685,0.01564995,0.016273372,0.03195519,0.090765566,0.05462368,0.034417097,0.031398386,0.0531503,0.010714109,0.016146732,0.01940074,0.014701767,0.02252724,0.012723318,0.013333924,0.06959237,0.032237284,0.009986647,0.033358194,0.018548507,0.042707466,0.01528742,0.013865086,0.018260945
+0.017474385,0.024095498,0.015481445,0.035995435,0.043080896,0.057520542,0.06724832,0.017136399,0.023127198,0.018029079,0.017169902,0.030075314,0.023954727,0.038316943,0.024727201,0.038316943,0.03634868,0.017338397,0.031151453,0.015941713,0.0183488,0.019820416,0.018894313,0.012151526,0.16644123,0.023978129,0.012784532,0.041349653,0.018931253,0.033814583,0.014830317,0.013138932,0.01298586
+0.011372657,0.021455431,0.012637653,0.01984303,0.034102026,0.020917524,0.021330083,0.017821947,0.037692245,0.014949076,0.014659934,0.07215875,0.019920694,0.028604828,0.030124335,0.052586645,0.055978194,0.019611852,0.016611882,0.03873711,0.02567879,0.045244068,0.06595843,0.016482607,0.052586645,0.02252906,0.012153512,0.030390264,0.0261597,0.032948077,0.027361637,0.022266587,0.059124753
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 99f876c0..8c2ab305 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.016554631,0.020482207,0.017691284,0.025540352,0.036871806,0.06521957,0.034739483,0.021931322,0.029562298,0.022627497,0.019697545,0.034100845,0.027227417,0.046701454,0.022060204,0.042522155,0.05755607,0.019354325,0.0462476,0.027696729,0.03559805,0.025690442,0.032507494,0.01755361,0.04430249,0.027561821,0.015310531,0.037968013,0.016554631,0.025690442,0.045798164,0.022539282,0.018540308
-0.012518779,0.008404711,0.010098492,0.021736387,0.01857397,0.046512906,0.047989387,0.030192723,0.031626258,0.014409027,0.012815653,0.04049012,0.034029573,0.024860121,0.054591935,0.026916321,0.040332265,0.007895496,0.016200505,0.01296672,0.025196189,0.03225003,0.04780229,0.0095984815,0.21009137,0.025030646,0.012916167,0.041450314,0.009903169,0.0090169385,0.029021885,0.019694805,0.014866418
-0.00902418,0.0061539556,0.008477432,0.02116707,0.009238182,0.037993215,0.050727595,0.031682577,0.028151562,0.010265757,0.015410727,0.050727595,0.057034567,0.026796907,0.07497053,0.072663926,0.05152644,0.007539984,0.0119084865,0.010146158,0.022053301,0.039661318,0.028497316,0.019291667,0.09330233,0.027031815,0.007901835,0.03230746,0.011542101,0.026227333,0.028946083,0.05378872,0.017841876
+0.016558,0.020446401,0.017625898,0.025545549,0.03687931,0.06523284,0.03474655,0.021892983,0.029667743,0.022543868,0.019701554,0.034174465,0.027259566,0.046619814,0.022107832,0.04253081,0.057567783,0.019358264,0.046257015,0.027702365,0.035640083,0.025645532,0.032577675,0.017625898,0.04431151,0.02756743,0.015253944,0.03804998,0.016558,0.02567059,0.045628898,0.022543868,0.018507898
+0.012470433,0.00840502,0.010098864,0.021737188,0.01853841,0.046696678,0.04799116,0.030120214,0.03165833,0.014409557,0.0128161255,0.040570777,0.034064077,0.024897482,0.054593947,0.026907457,0.04033375,0.007895785,0.016201101,0.012967198,0.025172522,0.03225122,0.04761769,0.009598835,0.21009907,0.025056025,0.012916644,0.04153289,0.009903534,0.009017271,0.02899463,0.0196571,0.014809004
+0.009016722,0.0061488706,0.008470427,0.021128936,0.0092305485,0.03788775,0.050884057,0.0316564,0.028093988,0.010217286,0.015397993,0.05068568,0.056987446,0.026745368,0.074908584,0.07260389,0.051483866,0.0075337538,0.011898647,0.010137774,0.02201357,0.03970602,0.028418211,0.019275725,0.09395641,0.027032569,0.007895307,0.03224926,0.011532564,0.02622486,0.028936293,0.05374428,0.017896906
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 3381365f..9999fe0a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012163671,0.016177313,0.019475494,0.03201573,0.019628244,0.042621553,0.03087956,0.0340473,0.048865855,0.015436501,0.010860935,0.07137767,0.027397878,0.018224202,0.06918161,0.016821744,0.05757805,0.015020127,0.0148451375,0.02235058,0.02353786,0.062990576,0.016560948,0.019937342,0.05201743,0.050417025,0.013411464,0.056685384,0.024237646,0.030490013,0.01951357,0.020093713,0.015137932
-0.010944551,0.00819709,0.012646405,0.023028629,0.010774871,0.070812054,0.049050074,0.0339097,0.023305774,0.018763326,0.014756296,0.046439677,0.017285585,0.04431305,0.03609669,0.0231781,0.042782236,0.010443363,0.0142187355,0.019548915,0.013357266,0.18802817,0.035606537,0.007944892,0.05282901,0.015404175,0.020090831,0.019663798,0.015524992,0.023804635,0.024000125,0.03457851,0.018671932
-0.011467976,0.012065373,0.009433299,0.024672067,0.0170815,0.060087897,0.15464325,0.041540455,0.018254327,0.013779089,0.019622322,0.028830487,0.023233917,0.035531435,0.024444234,0.05557222,0.022083525,0.010320061,0.013618558,0.018254327,0.019622322,0.040341105,0.03819419,0.012994919,0.046251304,0.023599796,0.027953701,0.026456343,0.015859634,0.012399838,0.065480076,0.036802802,0.019507684
+0.012169482,0.016216686,0.019522885,0.03196851,0.019637613,0.0426419,0.030803923,0.03406355,0.04888918,0.015383666,0.010866124,0.071411744,0.027344117,0.018268555,0.069214635,0.016796945,0.05760553,0.015027304,0.014852231,0.022339424,0.023526108,0.06302065,0.01656886,0.01990794,0.05183937,0.050441086,0.013470389,0.05671244,0.02426106,0.030474791,0.01948479,0.02006408,0.015204442
+0.010947577,0.008199356,0.012649901,0.023091303,0.010820034,0.070831634,0.04906364,0.033919077,0.02334354,0.018695341,0.014760376,0.046452522,0.017256627,0.044152495,0.036036216,0.02321566,0.042794067,0.01044625,0.014222666,0.019535234,0.013360959,0.18808013,0.035616383,0.007947088,0.052843623,0.015408434,0.02007677,0.01968845,0.015529283,0.023857038,0.024028016,0.034453224,0.018677093
+0.011479456,0.012030367,0.009369259,0.02470883,0.017098598,0.060148057,0.15359342,0.041582048,0.018272603,0.013792883,0.019641966,0.028929897,0.0232118,0.035706215,0.024361415,0.055627856,0.022084057,0.010330392,0.013632191,0.018272603,0.019641966,0.040381495,0.03815783,0.013058839,0.046297614,0.02357733,0.027896425,0.026547566,0.01587551,0.012412251,0.06580217,0.036911674,0.019565389
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index b0c6cbe3..76bbd353 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.015029022,0.019297646,0.0163777,0.07805737,0.12882058,0.08037804,0.113240555,0.07114149,0.17607687,0.111921266,0.11592598,0.035355426,0.03837795
-0.0130002005,0.017088441,0.012600226,0.06811618,0.077639334,0.03980949,0.0932858,0.28013662,0.10305648,0.07076048,0.101855844,0.032938834,0.089712076
-0.0252677,0.03453685,0.031941365,0.04289794,0.05886402,0.051947314,0.0674056,0.11280049,0.05979099,0.23786764,0.072545536,0.029425764,0.17470881
+0.015032878,0.01945399,0.0163819,0.07815367,0.12835126,0.08055583,0.1132696,0.07102088,0.17612204,0.111949965,0.1159557,0.03536449,0.038387794
+0.013001445,0.017090078,0.012601433,0.06812271,0.07749526,0.03989114,0.09329473,0.2801634,0.10306636,0.07069819,0.1018656,0.032813556,0.089896075
+0.025228165,0.03434838,0.03189139,0.042914554,0.058944356,0.05196744,0.067275494,0.112624,0.059668303,0.23935817,0.07264455,0.029379725,0.1737554
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index 4f7e8351..2e2a6f5c 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034147155,0.032329876,0.029900245,0.1616392,0.043674875,0.09300306,0.11439571,0.095208555,0.023745617,0.10788543,0.08917852,0.06721694,0.10767492
-0.020598466,0.017345646,0.016042102,0.09035372,0.096557476,0.037299003,0.10258434,0.094136685,0.03356547,0.11222761,0.06780411,0.042265307,0.26922008
-0.034789838,0.054094452,0.03192495,0.06606789,0.05707936,0.08610568,0.10488253,0.034519102,0.022726784,0.13336308,0.059760265,0.033457063,0.281229
+0.034292977,0.03246794,0.029910866,0.1610662,0.043861385,0.09321798,0.11443634,0.09524237,0.02375405,0.107923746,0.08903612,0.067076854,0.10771316
+0.020602653,0.017349172,0.016045362,0.090195745,0.09657709,0.03730658,0.102205165,0.09415581,0.033572294,0.11268975,0.06775169,0.042273894,0.26927477
+0.034788474,0.054092336,0.0319237,0.06609757,0.057021417,0.08601828,0.10508348,0.03451775,0.022725893,0.13335787,0.059699602,0.03345575,0.28121793
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 8aaf1681..ff861119 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.017511223,0.016195234,0.028871126,0.17548077,0.035098363,0.055460665,0.057403073,0.059726194,0.073725,0.2739216,0.06293729,0.095035285,0.048634168
-0.014574757,0.015274214,0.021879302,0.11781726,0.028093578,0.09824792,0.12251057,0.1169004,0.0659042,0.12251057,0.13143465,0.06260283,0.08224971
-0.012698688,0.010946953,0.021601195,0.15775292,0.014389498,0.06268852,0.16339757,0.17667489,0.10145584,0.09401476,0.057232037,0.08443905,0.04270813
+0.017545385,0.01622683,0.028927451,0.17582312,0.03502973,0.05546043,0.05766268,0.06001462,0.07401324,0.27232018,0.063121684,0.09522068,0.048633967
+0.01458249,0.015282317,0.021720547,0.11834111,0.027998893,0.098108225,0.122097656,0.1165064,0.06583057,0.12257554,0.13201904,0.062804475,0.08213275
+0.012693802,0.010942741,0.021592885,0.15769221,0.014383962,0.062756255,0.16333468,0.1766069,0.1014168,0.09416231,0.05715417,0.08457157,0.042691693
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 55d4c245..4293f9e1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.019247921,0.019242048,0.023122676,0.2023879,0.030257503,0.070795946,0.083553046,0.04229355,0.07910523,0.21612567,0.054946132,0.118629664,0.040292792
-0.013859086,0.011220005,0.014865906,0.16772893,0.04046441,0.06960686,0.26081634,0.1302294,0.052909687,0.071038246,0.049805023,0.07961735,0.03783882
-0.03323299,0.028751979,0.053304255,0.20066538,0.029766515,0.13464227,0.11751235,0.17178494,0.043314233,0.059854817,0.047320165,0.053955637,0.025894517
-0.01579723,0.016942782,0.019503407,0.06499989,0.03757122,0.13979305,0.16676494,0.055090666,0.048058998,0.16925676,0.101997375,0.077775694,0.086447895
-0.011209881,0.013517572,0.013519222,0.04491335,0.14133994,0.06666387,0.15254389,0.070202954,0.10669937,0.12986594,0.11656749,0.023598434,0.10935808
-0.01869931,0.029639825,0.041397702,0.19057001,0.0340989,0.11240363,0.078290984,0.15029393,0.0654575,0.08262617,0.078181155,0.09896303,0.01937786
+0.019174946,0.019325335,0.023039225,0.20295402,0.03028449,0.07068914,0.083618164,0.042292938,0.07901361,0.21604344,0.05490511,0.11838204,0.04027746
+0.013898184,0.0113433665,0.015027519,0.16799693,0.04045206,0.06948655,0.2601987,0.13032606,0.05281098,0.07113437,0.049805496,0.07966671,0.03785304
+0.03320181,0.028733771,0.053160068,0.20101622,0.029878395,0.13390566,0.11725172,0.17261124,0.043387983,0.05976947,0.04728154,0.05394451,0.025857594
+0.015875842,0.017032292,0.019604048,0.06506866,0.03764047,0.13930608,0.16607754,0.0550884,0.048237015,0.16935313,0.10211777,0.077763,0.08683565
+0.011254491,0.013575508,0.013575508,0.045037072,0.14090854,0.06673133,0.15235846,0.07019897,0.106779635,0.12981103,0.11681728,0.023640344,0.10931185
+0.018757429,0.02962508,0.041372053,0.19092914,0.034098256,0.11180388,0.07820428,0.15044855,0.06555784,0.08235855,0.07805169,0.099440426,0.019352855
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index f0ffa839..6fdeb3cb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.0173287,0.016026428,0.026839208,0.0828322,0.04565503,0.046283506,0.11124542,0.21780796,0.037408393,0.22123794,0.0596618,0.06350966,0.05416377
-0.025201809,0.025005685,0.034852836,0.05695978,0.0744348,0.05504581,0.049729746,0.18261582,0.07040466,0.15317841,0.16562536,0.048577756,0.058367517
-0.008935493,0.014057503,0.0095863845,0.058039587,0.09719801,0.03352519,0.19865984,0.2736655,0.07825349,0.107588395,0.04577878,0.03985105,0.034860685
+0.0173293,0.016026983,0.026840145,0.08275424,0.045745883,0.046285123,0.11103223,0.21781556,0.03755611,0.22124566,0.059722174,0.06348086,0.054165658
+0.025097355,0.02499951,0.034844227,0.056834597,0.074416414,0.054978497,0.049620453,0.1825707,0.07031857,0.15373994,0.16558443,0.048471,0.05852431
+0.0089370245,0.014059912,0.009588027,0.058219835,0.09721465,0.033596482,0.19869386,0.27371234,0.07796174,0.107606806,0.045674965,0.039935794,0.03479862
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 35299c71..c090598e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.013696977,0.015395226,0.022755446,0.12909949,0.0304831,0.06601206,0.0899556,0.15893084,0.10253051,0.095681295,0.12471578,0.049288597,0.10145512
-0.013873818,0.025609914,0.01354995,0.05140924,0.07923226,0.041689683,0.07039259,0.102127224,0.17122677,0.10499674,0.0796728,0.044621523,0.2015975
-0.018696813,0.024568997,0.015257101,0.05712812,0.06721138,0.033482578,0.07528165,0.19145387,0.17692275,0.123415925,0.09988427,0.09062429,0.026072279
-0.0140934745,0.020499628,0.02925345,0.07079055,0.03404695,0.10040815,0.29820317,0.12686196,0.046179004,0.08807876,0.046114925,0.07786219,0.04760782
-0.026012044,0.044062022,0.06208434,0.15045841,0.038114253,0.13218476,0.09425221,0.09837236,0.040025376,0.08812661,0.07108019,0.09915601,0.05607141
-0.034625676,0.03115023,0.03877182,0.061037596,0.14396825,0.069717765,0.05949643,0.10258124,0.18308024,0.14989388,0.05404332,0.048556603,0.023076938
+0.013744185,0.015574202,0.022660334,0.12888199,0.030492809,0.06577828,0.08997413,0.159148,0.10275368,0.09521741,0.1249167,0.04930175,0.10155656
+0.0139527805,0.025663089,0.013629564,0.051437438,0.07935727,0.041736744,0.07068528,0.102295555,0.17131282,0.1051313,0.07982361,0.04468962,0.20028499
+0.01877227,0.024675684,0.015321485,0.057176925,0.06737088,0.03346519,0.07534136,0.19108021,0.1767203,0.12337066,0.10010392,0.09043637,0.026164724
+0.014150181,0.020588392,0.029262004,0.070919394,0.03421075,0.1003548,0.2978537,0.12710838,0.046125602,0.0878736,0.046035603,0.07792775,0.047589783
+0.02630414,0.04422356,0.06218277,0.1501919,0.0382723,0.13202716,0.094079524,0.09811425,0.039952654,0.08803498,0.0712233,0.09927079,0.056122616
+0.03464103,0.031052018,0.03894796,0.060960393,0.14414504,0.06960196,0.059504647,0.102614194,0.18292959,0.14988713,0.05407388,0.04856628,0.023075903
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 2446af67..116b4471 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034523316,0.053054646,0.040678408,0.0977732,0.05697494,0.09720199,0.06494046,0.109287135,0.109287135,0.045203123,0.11122518,0.13869257,0.041157912
-0.010823801,0.016375888,0.015996542,0.12631503,0.058085933,0.04108866,0.12338894,0.15658872,0.07549984,0.14257565,0.0769138,0.068508826,0.087838314
-0.010226722,0.013234361,0.013869491,0.09971816,0.08315507,0.05762093,0.24392943,0.14737387,0.056230776,0.15997283,0.024628649,0.033927396,0.056112282
+0.034527026,0.052956816,0.040682778,0.09759292,0.057315916,0.09740249,0.06494744,0.10908562,0.10908562,0.045296367,0.11123714,0.1387075,0.041162338
+0.010824752,0.016249876,0.015997946,0.12682055,0.058006,0.041092265,0.12339978,0.15660247,0.07543277,0.14258817,0.076695524,0.06861528,0.08767462
+0.010323669,0.013255853,0.0137839075,0.0998801,0.08345294,0.05774269,0.24242425,0.14761323,0.056443825,0.16023265,0.024668645,0.03398249,0.05619568
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 99292a49..e1e8df04 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.027722225,0.038942188,0.03279261,0.04932378,0.06952372,0.08810105,0.038040094,0.030926377,0.03009219,0.18505847,0.07079118,0.021422075,0.31726402
-0.020294143,0.02752294,0.02110257,0.05809561,0.11678473,0.026572147,0.038662247,0.25112665,0.10716769,0.101861455,0.08781006,0.035202377,0.10779747
-0.023249542,0.02427032,0.028709352,0.10198248,0.054908074,0.052547373,0.06692258,0.23298474,0.07608357,0.08356145,0.1162404,0.028485935,0.110054195
+0.0277286,0.038951144,0.032800153,0.049431577,0.069505766,0.087949365,0.03804884,0.030812891,0.030099113,0.18510105,0.070807464,0.021427004,0.31733704
+0.020127492,0.027511021,0.021093432,0.05807045,0.11719104,0.02656064,0.03849484,0.25101787,0.10712128,0.10181735,0.087772034,0.03504995,0.10817252
+0.023074925,0.024276959,0.028717207,0.1020104,0.054923106,0.05251045,0.06695724,0.23304848,0.07640226,0.0833398,0.11627221,0.028382642,0.11008432
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
index a90a0f9c..c968105d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.011096026,0.012672059,0.016271247,0.02559827,0.100848414,0.047379345,0.043607958,0.02952103,0.112504534,0.14502408,0.3911485,0.017731398,0.046597134
-0.013814418,0.016405025,0.022598883,0.060714453,0.082501985,0.0610787,0.097782366,0.06913524,0.16763797,0.10449618,0.10068946,0.14116533,0.061979994
-0.019121636,0.028040025,0.021837592,0.20881435,0.12739657,0.17860837,0.07736431,0.07641633,0.052181434,0.045159265,0.026341166,0.10706935,0.031649638
+0.011098292,0.012674647,0.01627457,0.025603484,0.100475706,0.047441088,0.04359555,0.02958477,0.11252745,0.14505361,0.39122817,0.01773502,0.04670763
+0.013919137,0.01640075,0.02259299,0.06084698,0.08231952,0.060832124,0.09756611,0.069150954,0.16759425,0.10487779,0.100859985,0.14112851,0.061910894
+0.019119935,0.028037531,0.021835651,0.20879574,0.12713669,0.17929146,0.077329114,0.07641887,0.052074987,0.045331985,0.026338823,0.10664244,0.031646825
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
index c5539a6d..10e0c261 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.028286915,0.033591546,0.04288041,0.0652413,0.20657852,0.08050327,0.12876955,0.036749262,0.08919606,0.09513434,0.02763165,0.06495523,0.10048189
-0.015949441,0.027130906,0.03247144,0.11179749,0.24181513,0.10279171,0.06656204,0.08480224,0.05502044,0.082635805,0.07668217,0.0677755,0.034565672
-0.034620997,0.030552922,0.034756497,0.13375674,0.24031734,0.11263447,0.0917502,0.03892543,0.06997049,0.043424457,0.04176089,0.05420117,0.07332844
+0.028341014,0.033524573,0.04279492,0.06508739,0.20778364,0.08034278,0.12851283,0.036675997,0.08901823,0.095130295,0.027469048,0.06484156,0.10047761
+0.015947836,0.02712818,0.03246818,0.112004794,0.24179083,0.10298231,0.06662037,0.08495948,0.05490756,0.08234555,0.076345675,0.06780178,0.034697466
+0.034501042,0.03056623,0.03477164,0.133815,0.24042203,0.112683535,0.09161106,0.03894239,0.069830276,0.043528304,0.041779082,0.054224778,0.073324576
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
index 263dcbe9..2ee58cc5 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.01560967,0.028822947,0.024944182,0.059488393,0.2579016,0.17794675,0.061197206,0.059604697,0.08186873,0.07943712,0.054059114,0.055341084,0.0437784
-0.013061907,0.019455653,0.01785351,0.16741595,0.14602278,0.07193442,0.058912177,0.032966156,0.17071792,0.03809232,0.052423164,0.1693894,0.041754596
-0.015262611,0.018700138,0.024967961,0.2106938,0.1602874,0.068318814,0.065813944,0.034664582,0.123377606,0.056046806,0.06990481,0.12679777,0.025163786
+0.015611427,0.028826192,0.02494699,0.0594951,0.25793064,0.17866333,0.06114436,0.059611414,0.08167831,0.079112194,0.0540652,0.055131543,0.043783337
+0.013151082,0.019436039,0.017835513,0.16724718,0.14587559,0.07189701,0.05902547,0.032932926,0.17121333,0.038053922,0.05231921,0.16921864,0.041794058
+0.015268664,0.018707559,0.024977867,0.21077739,0.16035098,0.06837931,0.06590439,0.034678336,0.12342657,0.056069046,0.06993255,0.12635355,0.02517377
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
index 682787e3..39f27a25 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.016827505,0.024672894,0.026888637,0.071175836,0.2112891,0.1647487,0.048896004,0.025926603,0.05938984,0.12714048,0.16040213,0.031259973,0.031382322
-0.013997956,0.020205943,0.022106718,0.022882363,0.35228354,0.052252322,0.05392724,0.039436106,0.08225181,0.12857363,0.0942475,0.09183136,0.026003601
-0.05171333,0.03437709,0.04829341,0.060120467,0.07857956,0.1719225,0.10227148,0.05633357,0.054162037,0.15633135,0.024398591,0.116559125,0.044937547
-0.019069683,0.023180008,0.014620337,0.04426263,0.4761877,0.06358374,0.023508424,0.019419778,0.11789036,0.038552396,0.09140214,0.028023642,0.040299125
-0.01085677,0.009961516,0.011466335,0.03731547,0.36767113,0.02872022,0.035538275,0.026009187,0.1229666,0.05771425,0.22166713,0.023857085,0.046255976
-0.031226596,0.02629222,0.023113754,0.112368956,0.13236259,0.08975561,0.1010255,0.043042876,0.09706933,0.12928063,0.06875058,0.09676764,0.048943765
+0.016887669,0.02476413,0.026986418,0.071169205,0.21061277,0.16466737,0.04896162,0.025952585,0.05929014,0.12724507,0.16085285,0.031182745,0.031427313
+0.014183183,0.020237265,0.022139635,0.023021577,0.35177594,0.05240823,0.05402566,0.039391156,0.082258716,0.12840404,0.093942404,0.09212539,0.026086865
+0.05163218,0.03439445,0.04831485,0.060128797,0.07838445,0.17196144,0.10228253,0.056375563,0.054321844,0.15596226,0.024484772,0.116810314,0.044946518
+0.019097615,0.02303612,0.014642649,0.044230174,0.4773803,0.0634187,0.023490468,0.019398358,0.11744495,0.038352862,0.09146622,0.0280048,0.040036753
+0.010892983,0.0099959625,0.0115052825,0.037212107,0.3692803,0.028643182,0.035438757,0.026029024,0.12273083,0.05763524,0.22050735,0.023885578,0.04624343
+0.031303268,0.026257228,0.023171922,0.11228792,0.13282534,0.08952374,0.100654244,0.043037932,0.09660937,0.12974845,0.06882482,0.09698748,0.04876837
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
index b2377fc1..bde6b79b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.016353136,0.013987578,0.015006481,0.11175343,0.4632051,0.05027907,0.040120237,0.0651852,0.05404043,0.042499743,0.01661066,0.0897964,0.021162536
-0.021549432,0.020723887,0.017451258,0.17284249,0.13888285,0.09771629,0.052868646,0.16687156,0.06822529,0.038566373,0.06232491,0.10341074,0.038566373
-0.020322079,0.01738239,0.014523531,0.0958217,0.1388762,0.04725025,0.07889779,0.061897356,0.20206368,0.15193103,0.052148256,0.06576083,0.053124882
+0.016471865,0.014089133,0.014997804,0.11256474,0.45933443,0.05074929,0.040253956,0.0657868,0.054406185,0.04285011,0.01673126,0.09044831,0.021316173
+0.021560203,0.020734245,0.01732411,0.1722547,0.1394961,0.09795627,0.052895073,0.16630408,0.06833443,0.038736675,0.06235607,0.10346244,0.038585655
+0.020330017,0.017389184,0.014529208,0.09567208,0.13893045,0.0472687,0.07885156,0.06191019,0.20214261,0.15199037,0.0521177,0.065722294,0.05314563
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
index 5af54919..982b34a3 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
@@ -1,7 +1,7 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.011690648,0.022708276,0.018247731,0.12031194,0.14736429,0.072557904,0.059560545,0.06837643,0.15909584,0.047002815,0.072019346,0.15372172,0.047342557
-0.016566617,0.022464922,0.019672144,0.07886144,0.21629961,0.08572125,0.0340683,0.04920004,0.16754173,0.04258672,0.09411154,0.057664577,0.115241036
-0.0146416435,0.02312187,0.021978369,0.03468886,0.057679538,0.07068581,0.045911815,0.026580743,0.21460655,0.27920017,0.077151075,0.066860445,0.0668931
-0.019577837,0.020039665,0.0147772245,0.03963531,0.5760377,0.047202054,0.023575775,0.040274374,0.043345284,0.049281806,0.045759395,0.058174577,0.02231905
-0.04962753,0.031602994,0.031358972,0.07433624,0.15786043,0.17700651,0.08936339,0.05545165,0.042340003,0.16477123,0.02115324,0.059756365,0.045371544
-0.028013574,0.027042532,0.030052343,0.06491127,0.083546355,0.118885174,0.099732846,0.06744788,0.054070134,0.14321429,0.2222415,0.029767558,0.03107451
+0.01163391,0.02277809,0.018302709,0.120284714,0.14737591,0.072344445,0.059545252,0.0682523,0.15935136,0.04692053,0.07198328,0.1538465,0.047380984
+0.016511263,0.022392593,0.019761395,0.07892483,0.21580069,0.08592345,0.03414463,0.0493899,0.16741052,0.042493675,0.094276346,0.05768628,0.1152844
+0.014698291,0.023033507,0.02206475,0.03471272,0.057652313,0.07060261,0.045807593,0.026511405,0.21430898,0.2795116,0.07701353,0.06707407,0.0670086
+0.019513203,0.019975943,0.014786988,0.03957203,0.5769814,0.047176875,0.023537416,0.040175576,0.04321801,0.04915209,0.045597192,0.058028545,0.022284778
+0.04979623,0.031714294,0.031344812,0.07435257,0.1585615,0.17619851,0.089293055,0.055389207,0.042344138,0.16487788,0.021043912,0.059743974,0.045339987
+0.02798013,0.027119271,0.029901266,0.06500812,0.083533235,0.11872466,0.09958642,0.06744114,0.05382787,0.14320928,0.222675,0.029901266,0.031092396
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
index 2ebebefb..c2365f59 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034384813,0.038509164,0.031307727,0.06452221,0.055054195,0.106119744,0.09328523,0.07180421,0.12095608,0.13388617,0.041476,0.15470518,0.05398935
-0.014236114,0.022049326,0.0202336,0.099011414,0.2333774,0.08577096,0.054572605,0.04250119,0.1334949,0.14776537,0.06464855,0.047785368,0.034553222
-0.015531263,0.014031327,0.013181212,0.09739672,0.33207306,0.034660105,0.06450115,0.0460519,0.14060858,0.06892985,0.03326726,0.10207088,0.037696745
+0.03437901,0.038352557,0.031180408,0.06441689,0.055206407,0.10610183,0.09326949,0.071722016,0.12140899,0.13438751,0.04130733,0.15407604,0.054191515
+0.014236728,0.02205028,0.020234475,0.0990157,0.23338753,0.08585848,0.054574966,0.04250303,0.13350068,0.14777178,0.06443074,0.047880862,0.034554716
+0.015529931,0.014030124,0.013180082,0.09738835,0.33204457,0.034657136,0.064558625,0.04613797,0.14059651,0.06885666,0.033264406,0.10206211,0.03769351
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
index 295be473..52c7d207 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
@@ -1,4 +1,4 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.037636146,0.032444384,0.023552097,0.16867347,0.20747192,0.13239335,0.05846303,0.050250847,0.09762073,0.026583975,0.028409252,0.04206824,0.09443264
-0.014826758,0.025618346,0.024445197,0.04046106,0.089766786,0.11938711,0.03626906,0.17506963,0.14627615,0.04265201,0.02010807,0.18062693,0.08449295
-0.010266415,0.017463759,0.026733276,0.06694543,0.021734,0.0745373,0.06203545,0.12654482,0.14007233,0.03017467,0.022777036,0.36474198,0.035973556
+0.03764677,0.03245354,0.023466894,0.1687211,0.20753048,0.13243072,0.05847953,0.050265033,0.09745775,0.026591476,0.02841727,0.04208011,0.094459295
+0.0148268975,0.025518704,0.024445422,0.040461432,0.08976761,0.118922755,0.03641135,0.17507124,0.1462775,0.04273579,0.020108255,0.1806286,0.08482443
+0.010265543,0.017462278,0.026731001,0.066939734,0.021817207,0.07453096,0.062030166,0.12653404,0.14006044,0.0301721,0.022775097,0.364711,0.03597049
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
index 8cd9eea8..49604740 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0053320643,0.005789764,0.0048760655,0.03720358,0.0035652393,0.025075478,0.04443915,0.016188094,0.035687704,0.0061186426,0.005490237,0.021655092,0.025955843,0.01783606,0.021378342,0.014357306,0.21706626,0.0067667514,0.016861673,0.011250843,0.018886115,0.02848178,0.022863349,0.006047647,0.06405485,0.14882801,0.010209717,0.07615009,0.0072470545,0.015511747,0.021830576,0.013150852,0.023844091
+0.005332064,0.0057897666,0.004876065,0.037203588,0.0035652376,0.025075484,0.044439144,0.016188093,0.035687704,0.0061186426,0.005490237,0.02165509,0.025955847,0.017836059,0.021378342,0.014357305,0.21706617,0.0067667514,0.016861672,0.011250842,0.018886117,0.028481785,0.022863349,0.00604765,0.06405484,0.14882804,0.0102097215,0.07615008,0.007247054,0.015511746,0.02183058,0.013150859,0.023844091
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
index fafc83a6..b6b22fe5 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005414255,0.0058297557,0.0046986197,0.036962785,0.0031884725,0.025659159,0.04501603,0.017804626,0.042272698,0.0064299456,0.0062360885,0.021977006,0.028345384,0.016593426,0.020529402,0.01365076,0.22053148,0.007261174,0.018225364,0.011267549,0.016855016,0.032483425,0.02040384,0.006038296,0.061356742,0.13856283,0.010517521,0.07868338,0.006724646,0.016816795,0.021524182,0.011967941,0.020171447
+0.0054142578,0.0058297585,0.0046986197,0.036962792,0.0031884734,0.02565916,0.045016028,0.017804626,0.042272698,0.0064299456,0.0062360885,0.021977011,0.02834539,0.016593434,0.020529408,0.01365076,0.2205314,0.007261174,0.018225364,0.011267549,0.016855016,0.032483444,0.02040384,0.006038299,0.06135676,0.13856284,0.010517521,0.07868336,0.006724646,0.016816795,0.021524182,0.011967942,0.02017145
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
index a0bb7db1..be6c1f23 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010200325,0.008255294,0.008725317,0.028328693,0.0071364073,0.031695973,0.050800975,0.014434751,0.036991633,0.010709722,0.009840103,0.023735259,0.027150556,0.015559098,0.017729346,0.021140106,0.07350094,0.010444301,0.03560199,0.015230986,0.03764622,0.023684604,0.0285523,0.010491406,0.051373497,0.22636293,0.02416806,0.02768691,0.010378675,0.017089728,0.040571064,0.013346023,0.031436794
+0.0102003245,0.008255288,0.008725312,0.028328696,0.0071364064,0.03169596,0.050800964,0.014434744,0.036991633,0.010709721,0.009840102,0.023735257,0.027150547,0.015559097,0.017729346,0.021140099,0.07350095,0.01044429,0.035601992,0.015230984,0.03764622,0.023684606,0.028552298,0.010491399,0.05137351,0.22636302,0.024168057,0.027686903,0.010378668,0.017089726,0.040571056,0.013346022,0.031436786
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
index c4c85200..847be8bb 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.006770478,0.0070975553,0.0061097112,0.032388054,0.00469689,0.029014563,0.054458927,0.015219773,0.036924854,0.0077053085,0.0074958084,0.02382878,0.028791584,0.017738653,0.01888046,0.020983137,0.11063638,0.008250365,0.025185334,0.012733429,0.02757118,0.027456699,0.025177583,0.008029096,0.0590341,0.21493421,0.017014707,0.05102823,0.007797203,0.017581401,0.030098991,0.014413131,0.024953334
-0.006460396,0.007043518,0.0061552622,0.043098416,0.0042703887,0.024615532,0.040515337,0.016898174,0.034789506,0.007203162,0.006736328,0.023457414,0.025878392,0.016775291,0.016562976,0.015162665,0.20098755,0.008355655,0.024063585,0.011879982,0.021613166,0.027997697,0.02484585,0.007972056,0.05440866,0.17081757,0.011742848,0.048810504,0.00834056,0.015540331,0.02875625,0.013147283,0.02509769
+0.0067704804,0.0070975586,0.0061097112,0.032388065,0.00469689,0.029014569,0.054458935,0.015219773,0.036924865,0.007705308,0.0074958117,0.023828786,0.02879159,0.017738651,0.018880464,0.020983135,0.1106364,0.008250364,0.025185334,0.012733434,0.027571186,0.027456705,0.025177589,0.008029096,0.059034105,0.21493421,0.017014707,0.051028233,0.0077972026,0.0175814,0.030099008,0.014413131,0.024953345
+0.0064603947,0.007043517,0.0061552613,0.043098435,0.0042703883,0.02461554,0.040515333,0.016898172,0.034789506,0.007203161,0.0067363298,0.023457421,0.025878388,0.01677529,0.01656298,0.01516267,0.20098753,0.008355653,0.024063593,0.011879986,0.021613162,0.027997697,0.024845859,0.007972059,0.05440866,0.17081751,0.011742852,0.048810527,0.008340559,0.015540328,0.028756257,0.0131472815,0.0250977
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
index f67dd196..512da4f8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0055265864,0.0069941594,0.0047999513,0.03426367,0.0041857036,0.022670772,0.03401926,0.020106943,0.023731828,0.0055064764,0.0053447303,0.014821653,0.01768129,0.021764157,0.019465126,0.010148312,0.37970722,0.0066347406,0.0131839905,0.0095445,0.013646816,0.025851715,0.020261735,0.008503725,0.052374292,0.05979142,0.0068904874,0.08166511,0.008144954,0.014931847,0.014926138,0.01588961,0.017021054
+0.005526586,0.0069941585,0.004799951,0.03426365,0.0041857054,0.022670768,0.034019258,0.020106941,0.023731826,0.0055064755,0.0053447275,0.014821659,0.017681288,0.021764155,0.019465124,0.010148311,0.37970722,0.0066347397,0.013183994,0.009544499,0.013646808,0.025851712,0.020261733,0.008503724,0.052374292,0.05979142,0.0068904837,0.08166511,0.008144954,0.0149318455,0.0149261365,0.01588961,0.017021053
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
index 3aa6ada2..5cc07cb2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0079358015,0.008644697,0.008528162,0.050991315,0.006794538,0.027761,0.0358331,0.017671429,0.028613957,0.008022484,0.00793928,0.02051466,0.018362291,0.019954775,0.018463802,0.013146331,0.21864693,0.010478101,0.024880765,0.014835908,0.021379687,0.021088064,0.027753515,0.012069517,0.05148979,0.15231076,0.013349637,0.044053532,0.01088003,0.014509736,0.024076153,0.014572965,0.024447259
-0.005569799,0.005660792,0.005039816,0.03242026,0.0038034345,0.024715062,0.04848758,0.019317342,0.07592577,0.0073058056,0.0071936473,0.028503245,0.039357413,0.016763123,0.02464078,0.017066706,0.107044116,0.006419049,0.021850476,0.009070911,0.016291386,0.042864792,0.028132914,0.005143986,0.0739408,0.13126431,0.012961439,0.079355635,0.007179912,0.021646269,0.03454433,0.011274988,0.029244097
+0.007935801,0.008644695,0.008528165,0.050991338,0.0067945397,0.027761009,0.035833094,0.017671425,0.028613951,0.0080224825,0.007939282,0.020514661,0.01836229,0.019954776,0.018463803,0.013146329,0.21864699,0.010478104,0.024880761,0.014835906,0.021379694,0.021088066,0.027753517,0.012069514,0.051489774,0.15231074,0.013349635,0.04405351,0.010880027,0.014509733,0.024076149,0.014572963,0.02444726
+0.0055697993,0.0056607896,0.0050398135,0.03242026,0.003803431,0.024715057,0.048487585,0.019317342,0.07592577,0.0073058023,0.0071936473,0.02850324,0.039357413,0.016763113,0.024640776,0.017066706,0.107044175,0.0064190463,0.02185047,0.009070911,0.016291386,0.042864796,0.0281329,0.005143986,0.0739408,0.13126434,0.012961427,0.079355635,0.007179909,0.021646269,0.034544323,0.011274983,0.02924409
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
index e7303aa2..fb88a77b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.006583762,0.0068928897,0.0048378515,0.03660268,0.004346066,0.02784286,0.042631052,0.017884584,0.033521336,0.0069473158,0.00634369,0.018111208,0.024766268,0.018511454,0.018494321,0.013050359,0.25470728,0.007486415,0.018900618,0.010490363,0.018111968,0.02067791,0.02412442,0.009464144,0.0647224,0.13972172,0.010750387,0.05095635,0.008137707,0.015822683,0.021603521,0.016543148,0.020411285
+0.006583764,0.0068928925,0.0048378515,0.03660268,0.004346068,0.02784286,0.042631034,0.017884584,0.03352133,0.0069473186,0.0063436897,0.018111207,0.024766268,0.018511454,0.01849432,0.013050358,0.25470734,0.007486411,0.018900616,0.010490363,0.018111968,0.020677898,0.024124417,0.009464148,0.06472242,0.13972169,0.010750386,0.05095634,0.00813771,0.01582268,0.02160351,0.016543156,0.020411283
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
index 27cebee6..b2f4d541 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008046482,0.010288932,0.0069967234,0.03990826,0.007162291,0.03375451,0.03726217,0.021291917,0.027013969,0.007882263,0.0078681735,0.01970519,0.02320241,0.024027871,0.017250976,0.014517644,0.26696384,0.011102694,0.021171335,0.014237669,0.020665448,0.017218802,0.024644138,0.013858891,0.057388205,0.10421708,0.01121666,0.042277344,0.01241735,0.016171634,0.01957094,0.01778471,0.02291343
+0.008046485,0.010288931,0.006996726,0.03990828,0.00716229,0.033754513,0.03726218,0.021291913,0.027013965,0.007882266,0.007868176,0.019705186,0.023202414,0.024027875,0.017250976,0.014517649,0.2669638,0.011102698,0.021171337,0.014237667,0.020665446,0.017218808,0.024644135,0.01385889,0.05738821,0.10421709,0.011216665,0.042277344,0.012417349,0.016171632,0.019570941,0.01778471,0.022913428
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
index 46d86b6f..79a63b85 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0012847829,0.0013862588,0.0013743622,0.0049335216,0.002629261,0.04421479,0.07391029,0.04052809,0.04858283,0.0037130984,0.001400662,0.06348066,0.056958914,0.019132532,0.07674131,0.02009592,0.03749449,0.0011936828,0.0070741074,0.0039621834,0.0048973863,0.08357048,0.02983897,0.0016369526,0.09871048,0.0017693046,0.013229287,0.16916683,0.0018953648,0.019495614,0.012248143,0.0025117958,0.050937597
+0.0012904094,0.0013854031,0.0013756184,0.0049201986,0.0026279963,0.044148285,0.07374064,0.04074432,0.04855693,0.0037117477,0.001404071,0.06342667,0.05690586,0.019039612,0.07667846,0.020172833,0.03751043,0.0011957006,0.0070678433,0.003960929,0.0049003386,0.08352786,0.029856034,0.0016307909,0.09834979,0.00176659,0.013229489,0.1697645,0.0018912563,0.019570507,0.0122591285,0.00250594,0.050883807
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
index 9d87985f..09d5a2d2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.015949078,0.013059553,0.009781349,0.048254672,0.01312685,0.01593082,0.028322889,0.007367935,0.007820241,0.0077832667,0.014817788,0.002786293,0.010103129,0.011100873,0.013910568,0.012986693,0.032847404,0.017707605,0.018416008,0.024353962,0.032842595,0.002778024,0.03274647,0.05513109,0.009160907,0.3112136,0.0045494093,0.0028305363,0.019623127,0.03260526,0.07252648,0.09204435,0.0055211657
+0.015634717,0.01281785,0.009619565,0.048077118,0.012959089,0.015963905,0.028544521,0.007311691,0.007936274,0.007725436,0.014542746,0.0027863763,0.010137666,0.01106823,0.014024269,0.012997272,0.03277003,0.017333027,0.018349424,0.024214767,0.03291807,0.0027870873,0.03269819,0.055223055,0.009176768,0.312866,0.004502388,0.0028817935,0.019294238,0.03272431,0.073402315,0.09117624,0.005535524
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
index 110272db..39f7794f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0006185576,0.0006328632,0.00043876236,0.0064425473,0.0018244528,0.065627284,0.1284394,0.011077401,0.04086932,0.0013103667,0.00057274237,0.05383014,0.0138934385,0.03555001,0.035452865,0.016880834,0.100283615,0.0006685078,0.013122532,0.008209185,0.014611625,0.022872807,0.024365555,0.0038333896,0.035221875,0.08512628,0.003992684,0.14880891,0.00076322147,0.02621189,0.05397747,0.01884453,0.025624936
+0.0006198204,0.00063342374,0.0004384532,0.006420689,0.0018287954,0.06507862,0.1269804,0.011015099,0.04099147,0.001311511,0.0005743433,0.053394005,0.013784635,0.035449795,0.03551256,0.016842777,0.099894196,0.0006676215,0.013255125,0.008228865,0.0147693455,0.022764573,0.02433358,0.0038868464,0.034915894,0.08704853,0.003976894,0.14972544,0.00076197524,0.026150865,0.054506917,0.018916963,0.025319949
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
index d92c0440..84199ce6 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008535063,0.010116155,0.004999788,0.027722772,0.007546305,0.050576966,0.040861025,0.009750304,0.0048526525,0.0060036117,0.007846482,0.01775664,0.0069292323,0.034353506,0.021897852,0.011217101,0.012552527,0.009145636,0.011646283,0.026980927,0.0408808,0.010849433,0.05207903,0.02948556,0.016560948,0.22770509,0.00622852,0.007239846,0.012705731,0.037399787,0.10941857,0.108829156,0.009326774
-0.0008166164,0.0005132967,0.00059320376,0.006916493,0.002581364,0.021923427,0.050397992,0.009789066,0.11758413,0.001491984,0.0006522989,0.013462131,0.023066303,0.012532,0.02107708,0.016029773,0.19077104,0.00079144875,0.01853323,0.005924662,0.009348377,0.015089126,0.008362901,0.0038307437,0.022829419,0.035089612,0.0039825533,0.32077447,0.0008275365,0.014444679,0.024167405,0.0045341994,0.021271454
+0.008406845,0.0100028645,0.004978663,0.028080318,0.0074564624,0.05106336,0.041368242,0.009740752,0.0049058977,0.0059933676,0.007738937,0.017632268,0.006884857,0.034210872,0.021888284,0.011264261,0.012693914,0.009070253,0.011593515,0.026868774,0.041217517,0.010789327,0.05191277,0.029620975,0.016518576,0.22696489,0.006126781,0.0071672136,0.01255229,0.037649535,0.10909438,0.1091829,0.009360133
+0.00081456243,0.00051110506,0.000592047,0.00687621,0.0025706633,0.021814596,0.050301585,0.009817175,0.11709266,0.0014886729,0.00065051304,0.01347012,0.02308426,0.012487394,0.021035971,0.016027665,0.18970953,0.00078846415,0.018496549,0.005921485,0.009317239,0.015121425,0.008347157,0.0038045023,0.022821637,0.034895573,0.003979794,0.3231347,0.0008225577,0.014405082,0.024048084,0.0045162523,0.02123482
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
index 135277b4..9d17464e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0006395509,0.0005395344,0.0006403406,0.0073080775,0.0016307676,0.04628504,0.114420965,0.033811003,0.04748827,0.0018689088,0.0006519,0.033128954,0.04977551,0.015585043,0.048301417,0.019319415,0.11359712,0.0006475441,0.0067579355,0.0037142283,0.0035758512,0.040649626,0.021625642,0.0012714841,0.08745198,0.003365021,0.0064466624,0.21271876,0.0010225932,0.022955855,0.010015839,0.0032300227,0.03955918
+0.000644279,0.00054107426,0.0006426197,0.007318598,0.0016342783,0.046498753,0.114201136,0.03403976,0.04745785,0.0018703459,0.00065563683,0.033165775,0.049821507,0.015586319,0.048468433,0.019384295,0.11359834,0.0006508778,0.006752685,0.0037247261,0.0035874853,0.040763292,0.021660022,0.0012675843,0.087549336,0.003357906,0.006457923,0.2118147,0.0010242914,0.023060877,0.0100330515,0.0032326481,0.03953359
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
index e5581ce7..6de20a20 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0017209036,0.0015223114,0.0009925718,0.017091438,0.0052904864,0.029449824,0.05233395,0.005048244,0.11915878,0.0020151085,0.0016928521,0.011555765,0.01666688,0.016796239,0.025025416,0.015222794,0.24491838,0.002148038,0.034079727,0.010994179,0.035693437,0.0076755397,0.015851151,0.01280274,0.021139963,0.12886791,0.003564564,0.062393457,0.001786263,0.017634643,0.0507273,0.016218634,0.011920487
-0.027149918,0.03692235,0.0297103,0.07504655,0.025415707,0.06079237,0.025890756,0.02272309,0.008084381,0.018789234,0.039250992,0.011783172,0.017395535,0.029108455,0.016487945,0.0140047455,0.021839352,0.052460495,0.012594856,0.029682744,0.036679644,0.009503126,0.053610098,0.036722347,0.033857927,0.029855149,0.0132688945,0.0007574541,0.05655436,0.031545028,0.02689056,0.07513626,0.020486278
+0.00171548,0.001514297,0.0009926202,0.01718312,0.00527013,0.029487936,0.05267179,0.005046223,0.118690774,0.0020176421,0.0016863676,0.011463078,0.016580433,0.016751137,0.024944035,0.0152443405,0.24363711,0.0021403863,0.033990223,0.010999503,0.035781205,0.0076473923,0.015846951,0.0128892865,0.020958621,0.13015136,0.0035353715,0.062455773,0.001783925,0.017721385,0.050996166,0.016302522,0.011903471
+0.027323829,0.037167955,0.029990472,0.07488815,0.025425775,0.060669463,0.025825078,0.022680491,0.008052186,0.018953843,0.03934911,0.011748633,0.017262118,0.02899022,0.01647767,0.0139917005,0.021608276,0.05273841,0.012544975,0.029721929,0.036775358,0.00949796,0.053365026,0.036835194,0.033656303,0.029669568,0.0132582355,0.00075166044,0.05680184,0.031587865,0.026841847,0.075078,0.020470945
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
index fd8cc3b9..f362f47d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.00085969456,0.0007951353,0.00066271354,0.018881613,0.0021610884,0.061503798,0.16510883,0.011513276,0.036499035,0.0013508685,0.0008861966,0.012494069,0.020483125,0.02126216,0.037900373,0.016543375,0.20081232,0.0012051113,0.011987639,0.0088223675,0.012496148,0.008076034,0.025651107,0.005890034,0.034473073,0.09267589,0.00274839,0.04009039,0.0014759087,0.044728093,0.048980862,0.031187622,0.019793583
+0.0008632209,0.00079384743,0.0006646036,0.018937923,0.0021681492,0.06119901,0.1648909,0.011469957,0.03668653,0.0013531677,0.0008869396,0.012405908,0.020405136,0.021181801,0.037859116,0.016523449,0.1998736,0.0012055563,0.012034784,0.008868842,0.012623,0.00803282,0.025626624,0.0059644775,0.03410578,0.09410486,0.0027197315,0.04003143,0.001474787,0.04476594,0.049414117,0.03120969,0.019654365
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
index fc6463da..9cbae14a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0068634474,0.009162191,0.009985762,0.015698336,0.0108448705,0.0904677,0.042359415,0.06670113,0.03318652,0.011744499,0.008607542,0.061853275,0.05889687,0.030315023,0.05424857,0.02326838,0.035711277,0.007862204,0.008442241,0.014744367,0.0083070565,0.06985575,0.029260479,0.0033493678,0.15207788,0.0012586035,0.028393278,0.01334909,0.010879728,0.022950096,0.0053134672,0.0070398482,0.047001738
+0.006902314,0.009175283,0.010000225,0.015625658,0.0108586475,0.09034072,0.04221178,0.06655271,0.033279005,0.011762234,0.008651375,0.061778508,0.059007354,0.030211158,0.05453736,0.023306027,0.035477407,0.007880052,0.008436053,0.014697746,0.008330109,0.07010769,0.029269386,0.0033454658,0.15206087,0.0012518687,0.02848025,0.0133461775,0.010879269,0.022960618,0.005361438,0.006997196,0.046918076
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
index 7cec3722..65ce62b1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
@@ -1,6 +1,6 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03490532,0.031998005,0.028772216,0.027195405,0.027683252,0.029920967,0.03279077,0.02628576,0.034324087,0.030802563,0.027264664,0.027657194,0.027639514,0.029530054,0.03239934,0.030662043,0.029952087,0.030814199,0.028125698,0.0344104,0.033223245,0.032099433,0.032986548,0.029886737,0.03260546,0.030854948,0.027623257,0.031297766,0.026375476,0.030162385,0.030712212,0.027799197,0.031239802
-0.027808135,0.029926706,0.03400922,0.030426748,0.025627963,0.031928588,0.03561634,0.030228743,0.031077368,0.029948497,0.028980114,0.028951673,0.029900176,0.027818188,0.028425556,0.025779901,0.03055118,0.029142106,0.03244458,0.030813629,0.027564976,0.02764529,0.02720103,0.029113112,0.032587785,0.03428341,0.029919686,0.031955376,0.03911367,0.030389592,0.03261028,0.028901618,0.02930875
-0.029915085,0.030543437,0.028834613,0.03042259,0.026555737,0.03297875,0.028495284,0.03012273,0.02954228,0.030395597,0.031444997,0.02910545,0.030163703,0.030961595,0.03490318,0.031798262,0.029193621,0.03289751,0.028760593,0.033082496,0.029938387,0.03467049,0.02820007,0.031503487,0.032639142,0.031294454,0.025620796,0.031352498,0.025782082,0.030491428,0.025737967,0.028993692,0.033657994
-0.025365606,0.025664778,0.026062079,0.032890774,0.026538692,0.03823054,0.028495807,0.029882822,0.033322647,0.031037634,0.025865804,0.029255202,0.02681062,0.03221474,0.03261645,0.034453724,0.027962716,0.028476784,0.023960778,0.027055632,0.026977597,0.03259397,0.03620625,0.032044176,0.032762516,0.037340216,0.030151956,0.033774957,0.029703537,0.0278705,0.028950956,0.030582692,0.03487683
-0.025282321,0.029473947,0.028269365,0.03373335,0.028858237,0.033462204,0.030326918,0.033530205,0.027763087,0.030093702,0.030169412,0.03151387,0.03000025,0.029444912,0.032865997,0.03329371,0.030292938,0.032526247,0.026917703,0.03018704,0.026444286,0.033763483,0.028074782,0.0319315,0.030718518,0.032087997,0.031459063,0.02930742,0.030223848,0.030085208,0.025420504,0.030391077,0.032086886
+0.035109174,0.032228634,0.029122373,0.027463987,0.027945641,0.030163735,0.032845262,0.02594842,0.034922395,0.030951612,0.027564056,0.027116977,0.027191272,0.029049408,0.031170875,0.030246926,0.030047692,0.031019162,0.02844935,0.035145335,0.032860417,0.033282854,0.03376338,0.030274218,0.03223822,0.030229334,0.027426753,0.03153387,0.026364023,0.030059563,0.03067464,0.02712991,0.030460587
+0.027818047,0.029806498,0.034336902,0.030778736,0.026007194,0.031295177,0.036927246,0.030886251,0.03125275,0.02985114,0.028995993,0.02902339,0.02976673,0.027265698,0.028691404,0.025839522,0.03046781,0.029810123,0.03192401,0.030833513,0.027267322,0.027360896,0.026245909,0.028813887,0.031931926,0.032841697,0.03109479,0.031397782,0.03992314,0.030624682,0.032517962,0.029487668,0.028914195
+0.030183403,0.03242052,0.029843858,0.0296451,0.026645405,0.031996164,0.028702747,0.030074446,0.028383184,0.030225128,0.032427616,0.029124603,0.0312291,0.030315014,0.033579975,0.03126286,0.02904425,0.03404973,0.030180257,0.03460662,0.029601347,0.034505613,0.02989275,0.03159202,0.030262558,0.030775197,0.025916839,0.030558,0.026761474,0.02995052,0.025229465,0.028859543,0.032154743
+0.025296634,0.026471082,0.026577652,0.033839412,0.02780684,0.039299954,0.03074986,0.029498447,0.0331007,0.030452378,0.02612987,0.029916089,0.026885556,0.031654015,0.03073531,0.03307961,0.027617231,0.029398216,0.02377331,0.028508658,0.0275827,0.033749986,0.036714476,0.034851376,0.03230926,0.035462867,0.02979035,0.031253695,0.02989457,0.027552374,0.027788708,0.028714927,0.033543948
+0.025478277,0.028292565,0.028266197,0.03619709,0.029252974,0.0347753,0.030366326,0.034015596,0.027419873,0.030975945,0.029481627,0.03149804,0.029024819,0.02893344,0.03221479,0.033274993,0.029894393,0.03219735,0.02582195,0.029421443,0.026916182,0.034794923,0.028513875,0.03288691,0.031072868,0.031177122,0.031510886,0.029790463,0.029789902,0.029741006,0.025122572,0.0299503,0.031930022
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
index 25701dea..f2aa4f3b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
@@ -1,6 +1,6 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.025796872,0.025821052,0.030540183,0.03540621,0.026608227,0.03478382,0.03067776,0.031003399,0.033146046,0.03325188,0.030365815,0.0308043,0.027449563,0.030018117,0.03153884,0.029352657,0.029092524,0.02384649,0.029970763,0.026764145,0.02692843,0.03091666,0.031293344,0.03301509,0.030675245,0.031148765,0.027454989,0.03464577,0.03371407,0.03039152,0.029442204,0.031400308,0.032734923
-0.027645376,0.024716172,0.029397551,0.032479983,0.029057788,0.034062088,0.029561806,0.030762905,0.03275578,0.029999051,0.03151822,0.029582402,0.029124012,0.033837236,0.03152767,0.02815874,0.031342242,0.02682736,0.029414995,0.028265033,0.030185502,0.03224051,0.033708274,0.034261037,0.030121397,0.027603159,0.02855955,0.03343462,0.02645024,0.031482406,0.028913507,0.030383296,0.03262006
-0.024845945,0.025052948,0.0263014,0.032901563,0.025363559,0.03965492,0.028409896,0.029733066,0.03424027,0.03162598,0.025910662,0.028759446,0.026962476,0.032856528,0.03243185,0.032826383,0.029125627,0.02801517,0.024131,0.027621036,0.027519526,0.032559667,0.035635147,0.03229265,0.03338464,0.037081745,0.027807528,0.03353886,0.02957665,0.029155599,0.02823219,0.030683924,0.03576222
-0.02797149,0.029818716,0.028976986,0.028290737,0.03108153,0.03052239,0.02963015,0.03228555,0.028276766,0.028560005,0.03070743,0.032421984,0.03245931,0.03343161,0.031278666,0.030921867,0.03263645,0.03018529,0.032240782,0.0317139,0.03008393,0.02913051,0.034749724,0.031110376,0.02584681,0.027162941,0.032870043,0.028430955,0.02776051,0.031122573,0.029412452,0.030131536,0.028776024
-0.027345048,0.029193792,0.029317684,0.026135022,0.02694352,0.03692952,0.02964658,0.031842634,0.02922959,0.029339928,0.02875359,0.02905978,0.030834356,0.036634963,0.029727926,0.030966414,0.030507686,0.032904536,0.030803429,0.032436226,0.028633079,0.029277464,0.032264367,0.030212997,0.030541789,0.037181064,0.026679141,0.030335572,0.026149832,0.030504115,0.028901305,0.028029755,0.032737307
+0.026077993,0.025926517,0.030068072,0.03636011,0.027121866,0.034626346,0.031620685,0.0315067,0.031596735,0.032528404,0.03069347,0.0314288,0.027027652,0.029328331,0.032343224,0.029408194,0.028625358,0.024174193,0.029095232,0.026811505,0.02688699,0.03139536,0.029974222,0.033895213,0.032119796,0.030733231,0.028192906,0.033536673,0.03282051,0.031133441,0.02963972,0.031351656,0.031950876
+0.028449364,0.024803886,0.0294986,0.03234279,0.029346153,0.03284062,0.029287698,0.031351723,0.03181133,0.02964178,0.031152815,0.030545823,0.028929776,0.032174207,0.032390404,0.029105593,0.031171152,0.027026553,0.02852814,0.027785255,0.029967785,0.031725638,0.03196683,0.033753444,0.030419005,0.026873102,0.030337531,0.033957142,0.027139246,0.032121036,0.03149567,0.03064888,0.031410974
+0.02473332,0.025805464,0.026985694,0.034197494,0.02668728,0.04076238,0.030618208,0.029560212,0.0342524,0.031228403,0.026219497,0.029502872,0.026915431,0.03167128,0.030253854,0.031405352,0.028886192,0.028909547,0.023818407,0.028952114,0.0280342,0.033739235,0.035629865,0.03535153,0.033023417,0.034959845,0.027452167,0.031207168,0.030123003,0.029052667,0.026981011,0.029005291,0.034075227
+0.027612552,0.029588765,0.028897526,0.028167427,0.03146175,0.030992467,0.030953089,0.03307828,0.028132018,0.028136177,0.030538997,0.032958314,0.03215302,0.032910448,0.03135702,0.03136277,0.032575402,0.030259186,0.031658836,0.031609565,0.029420199,0.029490033,0.034439214,0.03181292,0.02608727,0.026992736,0.033126123,0.028185237,0.027118426,0.031353865,0.029262694,0.028808676,0.029499011
+0.028027277,0.029319167,0.029869592,0.028233605,0.0284659,0.03679354,0.030892001,0.03222718,0.030400075,0.029845333,0.028747197,0.028966922,0.030681144,0.0330812,0.02865518,0.030550085,0.030910296,0.03369551,0.029822486,0.032989994,0.029237293,0.031726234,0.03142848,0.03187399,0.030198226,0.033343483,0.026842874,0.03070044,0.026204878,0.029943591,0.027344182,0.027224164,0.031758487
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
index 91c21c03..39b55eb7 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
@@ -1,5 +1,5 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.033804633,0.03133798,0.03240842,0.032249976,0.031567764,0.023804748,0.033451684,0.031181484,0.02904705,0.028814662,0.031253017,0.029255506,0.03155321,0.024031805,0.03100783,0.02836353,0.032650135,0.032079875,0.029468795,0.03124199,0.031478617,0.030717542,0.022784976,0.027685331,0.031394366,0.023952689,0.03493583,0.030857133,0.035084616,0.03195621,0.032159757,0.030788282,0.02763058
-0.025631752,0.029802956,0.030698422,0.029735252,0.029497724,0.031349394,0.033659987,0.030621238,0.02941033,0.029097665,0.030105164,0.031282384,0.032693755,0.030077938,0.02911067,0.026763074,0.03305479,0.026865568,0.036262047,0.029614927,0.028123945,0.028390907,0.031433348,0.029613422,0.029825589,0.03213038,0.0313533,0.028229501,0.03429874,0.030838396,0.030776728,0.03051359,0.02913713
-0.028800884,0.027588563,0.0300563,0.03443287,0.02912721,0.029885918,0.03512253,0.03208439,0.032546118,0.030156115,0.026779838,0.028835671,0.027242702,0.026432548,0.029021429,0.029896194,0.032417633,0.031113513,0.022803584,0.029385664,0.028680721,0.030133814,0.027381856,0.030590741,0.03248337,0.028955875,0.03544596,0.03154471,0.036519587,0.03206369,0.031347316,0.03071588,0.030406838
-0.02862093,0.03000527,0.029652234,0.02862441,0.03345083,0.03031905,0.028309684,0.032872725,0.028253555,0.03009939,0.0307454,0.033102486,0.032555282,0.033184092,0.02956358,0.03107351,0.03461472,0.031375494,0.032400206,0.03102256,0.031462118,0.030181756,0.03170402,0.030066047,0.026040724,0.02641749,0.0313496,0.027319029,0.026150461,0.0320657,0.028823854,0.030786902,0.027786825
+0.034084663,0.031583164,0.032546442,0.032138847,0.029565305,0.02397673,0.032969736,0.031043336,0.027734615,0.02915201,0.031556893,0.02899239,0.030178683,0.024926245,0.032692198,0.02895135,0.031203741,0.03237466,0.029557178,0.030954232,0.029987698,0.030188313,0.023178574,0.026828399,0.03162781,0.025829652,0.03609925,0.031925663,0.034451466,0.03171625,0.032841537,0.031190302,0.027952647
+0.025305754,0.0299227,0.030633897,0.03035362,0.030023782,0.031172622,0.034432754,0.032215968,0.02902592,0.0293114,0.029836485,0.03252568,0.03256719,0.02886391,0.029192211,0.027415669,0.0338695,0.027007982,0.034716416,0.02912123,0.02748679,0.027996587,0.02957499,0.029614456,0.03000929,0.03120561,0.03207879,0.02746605,0.034783393,0.03193119,0.03044039,0.030926643,0.028971093
+0.028770477,0.027449202,0.030062556,0.03365554,0.029413687,0.029721614,0.035221003,0.03161255,0.033253502,0.030102383,0.026639005,0.028710429,0.027786555,0.02702573,0.028572798,0.029887274,0.032881945,0.030976614,0.022857795,0.029277911,0.02882068,0.030132292,0.026810376,0.030622266,0.032237146,0.028753674,0.035257693,0.031009149,0.036623254,0.032417532,0.031938296,0.031099904,0.03039909
+0.028727695,0.029837677,0.029908122,0.02894165,0.033573724,0.030687451,0.028665278,0.033794753,0.027919773,0.029874934,0.030719016,0.03329227,0.03228232,0.032287028,0.029487554,0.031092698,0.03433863,0.031789694,0.032016642,0.030929402,0.031242827,0.030201329,0.03151861,0.030687474,0.026339065,0.026228543,0.031674992,0.02779673,0.025773538,0.031888627,0.028663844,0.029518012,0.028300088
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
index 4b357202..b31d6767 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
@@ -1,9 +1,9 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03365363,0.03131194,0.032062005,0.032612655,0.030762598,0.023996722,0.033894006,0.03161808,0.028858617,0.028537845,0.031369943,0.029386261,0.031269576,0.0240121,0.03156647,0.028297028,0.03270311,0.03169595,0.028860392,0.031636294,0.03105152,0.03050513,0.023036947,0.028302876,0.031403825,0.0235559,0.03512947,0.031095052,0.034982767,0.032558214,0.031808164,0.030668166,0.027796706
-0.032761578,0.029872429,0.029597605,0.029935349,0.026900567,0.032399144,0.02855435,0.026653757,0.03182153,0.03016848,0.032555472,0.028656261,0.029006623,0.03235095,0.03480727,0.028403722,0.028092936,0.029561672,0.03174143,0.03233374,0.033174057,0.034100316,0.03289924,0.032593768,0.032962408,0.029540613,0.02438331,0.032330763,0.023944201,0.029516606,0.027441561,0.028192038,0.0327462
-0.030725045,0.028272143,0.026365971,0.028971849,0.028257614,0.034422457,0.03088777,0.026187155,0.03345275,0.028805315,0.027079374,0.02812067,0.027472235,0.03160347,0.034400526,0.031103302,0.028490582,0.02971379,0.028624378,0.030313667,0.030894523,0.03547529,0.037652876,0.031707373,0.034248933,0.03381967,0.030158943,0.03319742,0.023066722,0.02731636,0.029626993,0.02753793,0.032026935
-0.025067946,0.025263725,0.026296385,0.033531774,0.026559269,0.03958333,0.027857842,0.030703193,0.033559784,0.03183785,0.025875572,0.029518714,0.026663154,0.032539647,0.03160153,0.034056365,0.029149989,0.02885333,0.023305051,0.02679659,0.027072044,0.033060253,0.034554478,0.032332253,0.03327132,0.037098464,0.028822728,0.033504702,0.02874793,0.029003797,0.028169123,0.03096053,0.03478138
-0.024328558,0.03019246,0.02978609,0.030933373,0.02664791,0.03585672,0.029219525,0.03113439,0.030371258,0.031899795,0.027526228,0.031095784,0.02991037,0.029948168,0.030375885,0.03407043,0.028136484,0.03192227,0.028028654,0.029312856,0.025214395,0.031401776,0.030239172,0.030062895,0.029838474,0.03965154,0.029726071,0.030511681,0.035435908,0.027180875,0.027923727,0.030501928,0.03161439
-0.027642481,0.029439252,0.033424422,0.030074678,0.02514382,0.033257872,0.03443825,0.02897289,0.032224122,0.03009342,0.028927157,0.028554885,0.029879149,0.029093452,0.028734183,0.025436884,0.030429183,0.029044895,0.032307327,0.0309847,0.028431837,0.028267644,0.029025367,0.029751183,0.03299192,0.035030864,0.02826856,0.031771418,0.037300482,0.030140122,0.03184548,0.028867587,0.030204555
-0.026242485,0.02856462,0.030316794,0.033157557,0.03299111,0.029088097,0.027124817,0.031284798,0.03198276,0.032629397,0.030249553,0.031380426,0.03177847,0.027848225,0.029764337,0.03523869,0.029829431,0.031178579,0.027072916,0.026859952,0.02733578,0.034253642,0.027440054,0.030315153,0.027151577,0.03023951,0.03223471,0.030502116,0.035234425,0.027642915,0.028225683,0.034010116,0.030831262
-0.031010924,0.027717909,0.028334908,0.029572029,0.031026728,0.028185181,0.03371703,0.03252502,0.028359665,0.028503757,0.029602813,0.028722132,0.03139693,0.029956581,0.033280496,0.027820814,0.037417598,0.029648174,0.03241054,0.03330766,0.034037765,0.030647116,0.027304817,0.027959816,0.031536154,0.023645312,0.031318452,0.029235678,0.025905538,0.035960104,0.030277228,0.029216642,0.030438429
+0.033855878,0.031555194,0.032104813,0.032544944,0.02899993,0.024226334,0.033384256,0.031535964,0.027571268,0.028917968,0.03153802,0.029317284,0.029884605,0.02487106,0.03317497,0.029042145,0.031352926,0.031901088,0.028874978,0.031184962,0.029574865,0.030010147,0.02339816,0.027419098,0.031692605,0.02540574,0.03633683,0.03199998,0.034317236,0.032391068,0.03252312,0.031074857,0.028017707
+0.032645375,0.029211296,0.029622218,0.029929172,0.026831938,0.033350952,0.02760981,0.026967196,0.03171539,0.030520461,0.03234894,0.028386977,0.02834992,0.032421347,0.03404085,0.028682854,0.028301395,0.03028663,0.030858088,0.032663256,0.03316226,0.03464805,0.033819176,0.03297321,0.03245527,0.029456258,0.02407616,0.032931916,0.023553552,0.029962027,0.027696542,0.027935032,0.03258643
+0.03079726,0.02836858,0.026739504,0.029305056,0.029472884,0.034308825,0.030819302,0.026103878,0.033304565,0.029323498,0.027538827,0.028037373,0.027434818,0.030544518,0.032991886,0.030903487,0.029249443,0.029962836,0.028534466,0.030887995,0.030986693,0.037431307,0.03750989,0.032438066,0.033674967,0.032570437,0.029678013,0.032340832,0.022803176,0.027646143,0.029873798,0.027064433,0.03135321
+0.025392696,0.02627636,0.027077502,0.035034917,0.027847782,0.04042039,0.029661613,0.030495398,0.033150274,0.03138564,0.026224535,0.030131755,0.02675575,0.031256,0.029583888,0.032415524,0.02878367,0.03011321,0.023064954,0.028201777,0.027973127,0.03411287,0.03447896,0.035180874,0.033063572,0.03497268,0.028549135,0.031398144,0.029217428,0.028573478,0.026990656,0.029208178,0.03300724
+0.024685439,0.030674905,0.029863117,0.031178279,0.02778255,0.035222333,0.02968134,0.030667255,0.030530095,0.03223433,0.028045597,0.030551089,0.03133845,0.029112432,0.02964785,0.033652943,0.027950848,0.031787325,0.028992387,0.029821344,0.025748067,0.032986067,0.031084312,0.030607037,0.02886095,0.037558336,0.02887098,0.029779462,0.035294015,0.026569769,0.027097931,0.031046059,0.031077052
+0.027818462,0.029526941,0.034052044,0.030185841,0.025486842,0.032498095,0.036262725,0.02962454,0.032617006,0.02988826,0.02899451,0.028541261,0.029611835,0.028164327,0.028638976,0.025260951,0.0305133,0.029842326,0.03165989,0.031313885,0.02806211,0.02798404,0.027728057,0.029662656,0.032396797,0.033439204,0.029236948,0.03131323,0.03833288,0.030617509,0.03189458,0.029205821,0.029624088
+0.02698204,0.029599056,0.030312235,0.033456706,0.03267925,0.028223274,0.027507698,0.03228416,0.028564464,0.032395225,0.030720642,0.031880267,0.03154193,0.028109167,0.030944703,0.035500936,0.02953083,0.032430664,0.027058449,0.027667807,0.027628645,0.033845417,0.027544307,0.029797006,0.02666573,0.029446175,0.033847805,0.029539702,0.034211446,0.027991222,0.028181324,0.034006014,0.029905727
+0.030370781,0.027035689,0.028008454,0.031036004,0.030973144,0.028833287,0.033738982,0.03354538,0.027871111,0.027916575,0.029257156,0.029191185,0.030842572,0.02881783,0.03315752,0.02846031,0.0370996,0.029838745,0.03092998,0.032305267,0.03198498,0.030699644,0.026844112,0.028627403,0.032551426,0.02384607,0.03394906,0.030086126,0.026287593,0.03627954,0.030584106,0.028412709,0.030617632
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
index b58e1cad..d25ff96a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
@@ -1,5 +1,5 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.028346347,0.027353464,0.030344704,0.034818895,0.0283053,0.03005997,0.0355836,0.03275808,0.032522377,0.030344306,0.027134521,0.028696552,0.027392156,0.026236838,0.029147837,0.02928847,0.033233523,0.03095584,0.022680035,0.029963996,0.028415956,0.030157642,0.026274486,0.030956142,0.032518826,0.02823263,0.034416623,0.031546608,0.0369284,0.033256564,0.03058284,0.03078088,0.030765546
-0.02614715,0.0258215,0.027550466,0.034463495,0.028367938,0.032731164,0.030727468,0.032352015,0.028698286,0.028598435,0.027810419,0.029569183,0.029639808,0.030452486,0.033374913,0.030064132,0.03173377,0.029634332,0.027205918,0.028680194,0.030136917,0.029390203,0.031379256,0.029033292,0.032703232,0.029795818,0.034223244,0.030701801,0.03382241,0.031033205,0.03053796,0.030996071,0.03262359
-0.031898044,0.028340638,0.02751846,0.026926287,0.027034767,0.030630607,0.029686758,0.027121056,0.03294052,0.029303096,0.027993035,0.02730981,0.03156264,0.03152772,0.034503076,0.02913811,0.031860307,0.026993573,0.033478737,0.031741023,0.03363095,0.030359568,0.033719897,0.027680459,0.03301379,0.030039832,0.02723801,0.032895096,0.027244497,0.031041017,0.032970347,0.029797535,0.032860763
-0.024685703,0.029492697,0.028752888,0.031134205,0.0272171,0.037294548,0.028083863,0.03050206,0.030993259,0.031962592,0.027071144,0.031027092,0.0293045,0.030799314,0.030283516,0.034436323,0.028116086,0.03160242,0.02737299,0.02817356,0.025712473,0.032479364,0.031470202,0.030321851,0.031176731,0.040771965,0.02915091,0.030929118,0.03230334,0.026936619,0.027898025,0.03060591,0.031937685
+0.028437903,0.027254757,0.030338282,0.034053758,0.028367294,0.029867936,0.035484016,0.032350704,0.033063136,0.030184561,0.027064085,0.028652815,0.027901977,0.026792774,0.028899506,0.029313952,0.033530645,0.030812234,0.022711875,0.029815352,0.028521638,0.029946608,0.02587863,0.030986773,0.032278765,0.027954705,0.034373548,0.031195903,0.037098233,0.033722892,0.031238012,0.031281143,0.030625613
+0.025746314,0.024933858,0.026921023,0.035529517,0.029480062,0.033978727,0.031168824,0.03179294,0.029984899,0.028650545,0.027421633,0.030246083,0.029631928,0.030445006,0.032250527,0.029986437,0.032097112,0.028758565,0.026263552,0.02810667,0.030151488,0.030298227,0.03089507,0.030969998,0.033610586,0.028924711,0.03355842,0.02988841,0.032800015,0.03190985,0.02990864,0.030362934,0.033327445
+0.032267183,0.028918488,0.028129185,0.026897648,0.027366351,0.030486971,0.029913964,0.02686379,0.03348956,0.029299267,0.028501146,0.027437123,0.031400625,0.030622374,0.033201296,0.028855711,0.031837236,0.027683068,0.03356016,0.03306354,0.033076447,0.030550973,0.03371906,0.02831449,0.032340366,0.028730428,0.027243681,0.032467235,0.027846765,0.031897455,0.033190466,0.029134385,0.031693567
+0.025270436,0.02990884,0.02897759,0.03144431,0.028191876,0.03697099,0.02852898,0.02989229,0.031260435,0.03201398,0.027516115,0.03016531,0.030635659,0.029858166,0.029304778,0.033509642,0.027898662,0.031974133,0.02800069,0.029125461,0.02662284,0.034055386,0.032390516,0.03126936,0.030411147,0.038490403,0.028171586,0.030406484,0.03229996,0.02624073,0.02718416,0.030607712,0.031401414
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
index c9e83edb..a425fc01 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
@@ -1,10 +1,10 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02937184,0.030781249,0.033119388,0.028550683,0.025509233,0.03206324,0.03396557,0.028235912,0.032892834,0.029829921,0.028320564,0.02828424,0.030789617,0.02829313,0.028306503,0.025561279,0.031485945,0.029745061,0.0323852,0.031503476,0.02918103,0.028602777,0.028324347,0.028798053,0.033785295,0.0349804,0.028154539,0.03169164,0.035794634,0.030512646,0.032773368,0.02908114,0.029325323
-0.025854453,0.030651193,0.030702325,0.02862513,0.029471286,0.030098798,0.034639407,0.031529788,0.028793877,0.028493036,0.029873291,0.031217096,0.03360634,0.029397383,0.029140493,0.02725684,0.033899527,0.027715212,0.036060736,0.030847853,0.027600395,0.027915468,0.030293047,0.02915647,0.028903313,0.031247552,0.032694653,0.02797646,0.034875035,0.031475507,0.031034105,0.030351939,0.028601935
-0.025826376,0.027251828,0.02673951,0.030150859,0.028219849,0.03097317,0.029552381,0.032052875,0.033098787,0.032143876,0.026451845,0.02858729,0.033598788,0.028336221,0.030879049,0.031548064,0.03830193,0.026910027,0.028226435,0.029459467,0.02940237,0.03086073,0.026327921,0.026920898,0.032483302,0.030522237,0.027463144,0.029610101,0.033719517,0.035605237,0.029845843,0.034718696,0.03421142
-0.02661512,0.030802475,0.032640494,0.031011712,0.03189831,0.030164946,0.03295102,0.030942937,0.031764835,0.032526337,0.032349735,0.03283627,0.03008033,0.029224372,0.02729727,0.029901998,0.029873366,0.025909916,0.03347985,0.027992664,0.02590215,0.030691914,0.030372458,0.033140227,0.026693273,0.03153677,0.029790008,0.029715454,0.033401065,0.02932462,0.029047104,0.03119655,0.028924473
-0.02814665,0.029950805,0.03147324,0.032071218,0.027423255,0.035451543,0.028824573,0.029990712,0.030143596,0.029638186,0.029570555,0.03163136,0.027836546,0.031823345,0.030531272,0.03083544,0.027090862,0.03556963,0.026167938,0.030581702,0.0284657,0.03182817,0.03386525,0.033583216,0.029266147,0.033167303,0.031999893,0.031316996,0.029996168,0.026834186,0.027982738,0.028263362,0.028678378
-0.033281386,0.030130476,0.030221825,0.02840587,0.027716415,0.030989772,0.02878914,0.026732758,0.031153036,0.02907955,0.033042736,0.027861213,0.030571226,0.03249253,0.034299064,0.028136875,0.02837106,0.03143917,0.03291027,0.033115998,0.03358187,0.03396475,0.031448998,0.031539276,0.032543357,0.029416593,0.025101459,0.031998575,0.024063027,0.029163437,0.027996125,0.027905693,0.03253653
-0.024855413,0.026666297,0.02936113,0.031087212,0.02884285,0.0313365,0.029449705,0.033253677,0.030825984,0.029471586,0.028742261,0.02912062,0.034331124,0.030908402,0.028541783,0.028688319,0.03659657,0.030675147,0.027317889,0.029583992,0.029643653,0.028368162,0.027667206,0.028800705,0.030276459,0.029693466,0.030268844,0.028184077,0.036950678,0.034286764,0.029671533,0.033815984,0.032715973
-0.025155025,0.026130434,0.026154922,0.02954241,0.028716099,0.030518187,0.031584375,0.031437334,0.030625204,0.027734924,0.028301684,0.027579084,0.033700705,0.030727208,0.03288186,0.03139454,0.032228857,0.026481455,0.030001791,0.028490013,0.027577322,0.029461626,0.031345062,0.028545208,0.031811137,0.032734197,0.032353926,0.031332225,0.03433076,0.031055674,0.0314804,0.03250627,0.036080144
-0.025650796,0.030116925,0.029914461,0.028343417,0.029756019,0.031495847,0.033737093,0.030940576,0.0293521,0.028435258,0.029297698,0.031161314,0.033550072,0.030386096,0.028744891,0.027141783,0.034294646,0.027434524,0.036070127,0.03000383,0.027885038,0.028399905,0.03164065,0.029249499,0.029878315,0.032681726,0.0320276,0.02796962,0.03277645,0.031302802,0.03109174,0.03045507,0.028814139
+0.029452197,0.0306485,0.03368184,0.029208787,0.02547852,0.031809207,0.035802938,0.028849022,0.033015896,0.029569034,0.028465796,0.027966324,0.029913383,0.027351242,0.028488118,0.025190182,0.031235317,0.030739307,0.03148216,0.032006137,0.028662467,0.028530112,0.027585056,0.028937826,0.03342983,0.03354061,0.029346963,0.03168956,0.03664937,0.030737981,0.032591004,0.029114723,0.028830616
+0.025410624,0.030470952,0.030369842,0.029661402,0.02980678,0.030244453,0.035259236,0.033240262,0.028276877,0.02860871,0.029643638,0.032424532,0.03325316,0.028354872,0.029587941,0.028144129,0.034343008,0.02772666,0.034166075,0.029962657,0.026728328,0.027766544,0.028914964,0.029249499,0.029259473,0.030429421,0.033704724,0.027519733,0.035062674,0.032374986,0.030571992,0.030798515,0.028663335
+0.025337975,0.027083492,0.02744322,0.030657839,0.028513115,0.031966254,0.031017954,0.03185362,0.033549584,0.032333627,0.027234409,0.027745046,0.033427104,0.028093094,0.029163105,0.030709269,0.03713305,0.027277596,0.028337648,0.03032646,0.028630411,0.03181621,0.026347104,0.028802983,0.032475777,0.030587494,0.026387088,0.029571716,0.034264978,0.03554306,0.02840355,0.033852767,0.034113403
+0.025487956,0.028920492,0.031584144,0.032168623,0.030936489,0.03206438,0.032932498,0.032587145,0.030902065,0.0329446,0.030976553,0.03291904,0.028920155,0.029618418,0.028930362,0.031141395,0.029884664,0.026178582,0.030916898,0.027372908,0.025814181,0.03133012,0.030211058,0.03310097,0.02752377,0.032027237,0.030076312,0.030099913,0.03261934,0.029556777,0.02907321,0.031551532,0.029628236
+0.028115489,0.029519554,0.031406675,0.033021085,0.02769064,0.03581125,0.02951925,0.030510264,0.02980014,0.029648412,0.029389903,0.031621452,0.02773173,0.030956693,0.030655386,0.030680252,0.026869236,0.03531161,0.025839085,0.030112183,0.028472967,0.032457538,0.033420917,0.033944823,0.029430335,0.03254275,0.03216424,0.031369843,0.030058293,0.026615452,0.02813199,0.028111547,0.029069062
+0.032825507,0.029050106,0.029823868,0.028476052,0.02723421,0.0323307,0.02776565,0.02706704,0.0309637,0.02922705,0.03263243,0.02747781,0.029745135,0.03299441,0.03421459,0.028740628,0.02831149,0.03199856,0.031706,0.03320294,0.03318537,0.034523427,0.033132866,0.031837754,0.03197639,0.029589491,0.025036542,0.0330899,0.023499442,0.029383754,0.02828902,0.02767611,0.032992084
+0.024933483,0.027506273,0.029821714,0.029868472,0.0295652,0.03086805,0.030575937,0.032271694,0.031803988,0.02926964,0.02913596,0.029194605,0.034982435,0.030616261,0.027791448,0.028632607,0.0363351,0.03158228,0.027264304,0.030072883,0.029574117,0.029052928,0.027467903,0.02984582,0.029604388,0.029606704,0.029627847,0.02741072,0.036741782,0.033972777,0.028591022,0.033564672,0.032846995
+0.024271613,0.02529622,0.025262209,0.031130273,0.029015483,0.032995865,0.03252496,0.032081917,0.030489897,0.027709953,0.02744618,0.028730193,0.033057757,0.030686473,0.03244001,0.032020606,0.032457124,0.025812166,0.02857281,0.02809348,0.026829114,0.030221019,0.031125693,0.03006681,0.032895982,0.031971052,0.032191124,0.030761,0.033361543,0.032130904,0.030909894,0.031800143,0.035640564
+0.025240514,0.02986517,0.029861702,0.02963675,0.030136624,0.031917684,0.03455104,0.032634728,0.028964166,0.02846413,0.02896291,0.03225899,0.033151124,0.02891383,0.028778404,0.02759961,0.034724258,0.027591249,0.034086443,0.029194983,0.02715989,0.028105231,0.029908778,0.029540123,0.030567242,0.03184583,0.03296974,0.02765829,0.03352342,0.032018468,0.03083127,0.03059918,0.028738184
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
index 87c707bb..e15e21d2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
@@ -1,6 +1,6 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.029685352,0.027744511,0.030044284,0.033951394,0.028732833,0.02994491,0.034402113,0.031651333,0.03374663,0.03049073,0.026828067,0.028434549,0.027637096,0.026393222,0.028637927,0.029101957,0.034120213,0.031485118,0.02227293,0.030039264,0.029686691,0.030844377,0.026529493,0.030811347,0.03335207,0.028110413,0.033255078,0.031356674,0.034736305,0.033637084,0.030754102,0.031082017,0.030499963
-0.030537475,0.027899254,0.029378591,0.0297096,0.031716786,0.02693288,0.03433145,0.033088665,0.02740053,0.02792913,0.030472768,0.028694458,0.03220934,0.029565549,0.032662544,0.027634623,0.03633886,0.030167157,0.033250082,0.03298737,0.033357035,0.029501474,0.026456432,0.027505033,0.030585667,0.023674536,0.03291113,0.028946836,0.028558468,0.035006773,0.030973727,0.02936918,0.03024652
-0.033833686,0.030788528,0.032240216,0.032861616,0.030781893,0.024426019,0.03269575,0.030321872,0.029727243,0.029050065,0.03131785,0.029049996,0.0312546,0.024382828,0.031556305,0.027793216,0.03238788,0.031475447,0.02939823,0.031028157,0.032263253,0.03098234,0.023543414,0.02798694,0.03215333,0.024038076,0.03364459,0.031184737,0.03487223,0.03188977,0.031967424,0.030930823,0.028171666
-0.032235544,0.028513635,0.027943423,0.026533665,0.027607339,0.030138258,0.029216288,0.027271325,0.03264726,0.02924624,0.028209869,0.027331444,0.032256123,0.03174167,0.03372429,0.028924001,0.032310896,0.027531,0.034024764,0.031696826,0.034188505,0.029953811,0.03298522,0.027146649,0.032747537,0.029807689,0.027179489,0.032454353,0.027391035,0.031187288,0.03338235,0.03002738,0.032444846
-0.028649943,0.02983568,0.029618803,0.028534783,0.033335116,0.030336117,0.028125117,0.032931272,0.028101504,0.029766463,0.030861683,0.032960948,0.032717913,0.033539113,0.029642688,0.031016078,0.034416024,0.03157327,0.03229032,0.0310232,0.03154139,0.029931167,0.03196622,0.030088337,0.0260188,0.026400523,0.031526826,0.027428826,0.026188847,0.03198211,0.028969765,0.030762099,0.027919022
+0.02990466,0.027794937,0.030200135,0.033223048,0.028840246,0.029801996,0.03408861,0.031201469,0.034209393,0.030435495,0.02688698,0.028210292,0.027972383,0.026755435,0.028162813,0.028899178,0.034516744,0.031520512,0.022439493,0.029979838,0.029862743,0.030803071,0.026144683,0.030880546,0.033226147,0.02804711,0.033052064,0.03126565,0.034700904,0.03395223,0.03138371,0.03145586,0.030181592
+0.02977129,0.027457537,0.028779993,0.030530233,0.0316698,0.027341226,0.034179598,0.033910844,0.027016249,0.027471438,0.030157587,0.029443314,0.031970903,0.02910385,0.032963887,0.028755877,0.03607527,0.030105237,0.031945523,0.031880952,0.031306166,0.029473325,0.026267372,0.027917987,0.031049537,0.024086313,0.035278693,0.02937533,0.028586993,0.035309106,0.03102215,0.028993413,0.030802958
+0.034256108,0.031212777,0.032661483,0.032562193,0.028868396,0.02445675,0.032526903,0.030098625,0.028534148,0.029497765,0.031607904,0.028765576,0.029720742,0.025040075,0.032955986,0.0282345,0.030974323,0.031867743,0.029543443,0.030924428,0.030652856,0.030405918,0.023620382,0.027121766,0.032386478,0.025952253,0.0347015,0.032217287,0.03453304,0.031794257,0.032801624,0.031243311,0.02825948
+0.032598216,0.029093012,0.028690943,0.026613811,0.028034385,0.029812267,0.029320821,0.027035141,0.033263125,0.02949445,0.028751273,0.027457636,0.032119725,0.030603293,0.0324208,0.028617961,0.03232818,0.02807421,0.03443888,0.03276456,0.033602335,0.030092882,0.032640934,0.027553203,0.03204509,0.028555753,0.027167292,0.032216128,0.028183337,0.03194732,0.033609107,0.029577114,0.03127676
+0.028709698,0.029625166,0.029874118,0.028837366,0.033504315,0.030749444,0.028552493,0.033821665,0.027835265,0.029518366,0.030826217,0.033214457,0.0324431,0.032692183,0.02955219,0.031034948,0.034116004,0.03191638,0.03192628,0.030853745,0.03129503,0.029935427,0.03172751,0.030749017,0.026323091,0.026236577,0.03182046,0.027901417,0.02581471,0.031797033,0.028829262,0.029471148,0.028495917
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
index 08bba466..938c691e 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
@@ -1,5 +1,5 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0347061,0.032037724,0.029720293,0.026950317,0.027587326,0.029580832,0.033706103,0.02672045,0.0336072,0.0304948,0.027389366,0.027341463,0.027687615,0.029654475,0.03191975,0.030349722,0.029564764,0.03208654,0.028236479,0.035157055,0.033096552,0.031182028,0.03215033,0.02952733,0.0320103,0.031021614,0.028249703,0.031084804,0.02777836,0.02990182,0.031047527,0.027308928,0.031142361
-0.026352162,0.027308615,0.026728509,0.030381735,0.027763495,0.031036979,0.029111193,0.031803854,0.033297487,0.032028094,0.026604233,0.028412612,0.03335847,0.028347764,0.031116711,0.031521976,0.037724625,0.027063178,0.02756419,0.02944993,0.029611493,0.030874036,0.026384085,0.027145538,0.033017457,0.03050442,0.027128858,0.030084612,0.03357262,0.03557978,0.029888283,0.03474908,0.034483936
-0.031161968,0.030985577,0.028069872,0.029070867,0.02766797,0.031849217,0.028295552,0.029846,0.028211916,0.028096769,0.031157041,0.028553555,0.030611211,0.031491056,0.03540855,0.033342153,0.027244546,0.03506385,0.028400969,0.032600712,0.029665994,0.034185328,0.029610706,0.030778438,0.03282092,0.032744713,0.02873029,0.032103863,0.02512524,0.028251864,0.02723303,0.028246962,0.033373386
-0.025955163,0.027249493,0.029127032,0.030652158,0.03015496,0.03147695,0.028033996,0.034602076,0.030288333,0.029120373,0.028158711,0.029552525,0.03452732,0.030882543,0.02711481,0.029857256,0.03783241,0.03359149,0.025146987,0.029117852,0.029665263,0.028882185,0.026112327,0.028500061,0.030931538,0.02985717,0.031220669,0.028128866,0.03408741,0.03489066,0.02980184,0.03413182,0.03134773
+0.034894887,0.032275356,0.02996399,0.026949542,0.027929796,0.029617915,0.033669554,0.026341518,0.034484576,0.030552784,0.027683964,0.026792545,0.027649531,0.029394316,0.030769546,0.030085139,0.029768215,0.032148227,0.028801225,0.035756513,0.03274599,0.032206167,0.032878507,0.029650897,0.03140312,0.03026901,0.027987143,0.03128506,0.027762262,0.029794872,0.030969823,0.026882874,0.03063518
+0.026114888,0.027567258,0.027771655,0.029752698,0.02820829,0.031714298,0.030308545,0.031291846,0.033901554,0.032065872,0.02763642,0.027419902,0.033871558,0.028658088,0.0289209,0.030442988,0.03671656,0.027796298,0.028483596,0.030759042,0.029196337,0.031400263,0.026616627,0.028743278,0.032370545,0.030774299,0.02558757,0.029802978,0.033976573,0.03539298,0.028634975,0.033863362,0.034237936
+0.03136811,0.032610927,0.028663022,0.02766169,0.027849054,0.030900279,0.029014288,0.029312909,0.027264105,0.027272008,0.032112505,0.028268036,0.032192186,0.031883806,0.034192175,0.032434538,0.027403167,0.03600785,0.030057115,0.03440504,0.029654121,0.03388859,0.032026663,0.030806169,0.030339846,0.032073505,0.028873272,0.03086741,0.025530437,0.02776002,0.026841227,0.027687589,0.032778434
+0.026323346,0.02814542,0.029942133,0.029598957,0.030931026,0.030908601,0.028679792,0.033140037,0.031168977,0.028879615,0.028810753,0.029077675,0.03557963,0.030610638,0.02624658,0.029175146,0.03712324,0.034798477,0.025707724,0.029618949,0.030265389,0.029436663,0.026252486,0.029421646,0.030187503,0.029937852,0.03027628,0.027772948,0.03426053,0.033577643,0.02875072,0.033727318,0.031666294
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index 11f16da0..3c45b01b 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.031412095,0.029059455,0.028553972,0.032031063,0.03308489,0.036242105,0.030798635,0.031423856,0.029171485,0.033814613,0.03109685,0.027981894,0.031494338,0.028169379,0.03019742,0.03245603,0.031989317,0.03025197,0.027243363,0.028524006,0.030921329,0.026887607,0.033563793,0.028870368,0.032369044,0.030635152,0.025346385,0.0325448,0.03363609,0.029415186,0.025767326,0.027363604,0.027682606
+0.031594455,0.029204236,0.028718323,0.032224495,0.03337781,0.035745747,0.030596118,0.03137629,0.029209606,0.03408813,0.031277202,0.027920168,0.03155741,0.028168924,0.030145012,0.03245209,0.031797096,0.030348834,0.027226841,0.02847952,0.03093982,0.026849763,0.033478472,0.028779963,0.031998247,0.030514607,0.025465587,0.0325864,0.033844102,0.029311249,0.025630878,0.027453667,0.027638951
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index 76fb6eb8..84d81025 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02718314,0.030715112,0.031837214,0.035137605,0.030595766,0.03317532,0.031649627,0.029798042,0.03140365,0.031467617,0.03369167,0.027727356,0.028963719,0.027858147,0.025119768,0.030851895,0.03187139,0.03497891,0.02884058,0.031090252,0.028769664,0.028061062,0.032640804,0.026473435,0.030201206,0.02919215,0.031749595,0.031676207,0.03188455,0.02838132,0.029804083,0.0268488,0.030360317
+0.027228992,0.030756999,0.031926684,0.03520301,0.030777134,0.032955393,0.03167837,0.029797489,0.031220395,0.03152831,0.033776756,0.027809406,0.028926183,0.027881108,0.025017528,0.031061377,0.031814765,0.03505755,0.0288942,0.03108295,0.028965265,0.028036237,0.032550003,0.026335774,0.030168375,0.029126244,0.031778075,0.03171285,0.031987205,0.028292153,0.029634615,0.026793996,0.030224688
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index f277d42b..0aaba66d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027903663,0.031013355,0.030396728,0.034959204,0.031085547,0.03232213,0.031048927,0.030432688,0.031227158,0.03177365,0.03364062,0.027822366,0.027917664,0.027073784,0.025336333,0.031431273,0.032738667,0.034493793,0.029069608,0.031952385,0.029592805,0.028263152,0.032590456,0.026255373,0.0297492,0.029384159,0.0307498,0.031711757,0.033399194,0.028632063,0.028912678,0.02644551,0.030674366
+0.027887529,0.031016335,0.030412342,0.03501881,0.03126559,0.032115854,0.03112963,0.0304681,0.031105764,0.031803418,0.0336682,0.02792235,0.027942816,0.02709955,0.025281327,0.031618994,0.032682337,0.034516595,0.029082842,0.03194304,0.029754547,0.02831144,0.032550737,0.026138999,0.029740384,0.029294712,0.030734966,0.031810347,0.033461045,0.028520357,0.028774502,0.026385749,0.03054078
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index 1fd43647..9c566baf 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.032669287,0.02876315,0.030224655,0.027241936,0.029947326,0.031176692,0.029444885,0.029328013,0.028524838,0.03066203,0.027863681,0.03166694,0.034933478,0.032277443,0.03640189,0.029949818,0.028301675,0.026075223,0.029140284,0.027568853,0.0323754,0.03053253,0.029847682,0.033827864,0.032350555,0.0302727,0.027539514,0.029550968,0.028082011,0.03102456,0.030212617,0.03395407,0.028267471
-0.03143606,0.028259205,0.03223914,0.027423099,0.029374318,0.030516151,0.029467668,0.028255668,0.028395286,0.030080462,0.028519074,0.031538118,0.034919344,0.03331209,0.034968793,0.029315263,0.027845407,0.02649866,0.030031862,0.027775807,0.032202866,0.030588228,0.028885216,0.033773314,0.03254831,0.030060904,0.03000054,0.02907594,0.026533103,0.030174984,0.03166629,0.035144202,0.029174669
+0.032903668,0.028901948,0.030422168,0.02729401,0.029996833,0.031081805,0.029144669,0.029184328,0.028582947,0.030838203,0.02802886,0.03143278,0.0348301,0.03218194,0.036367875,0.029662728,0.028250722,0.026152506,0.029222546,0.027613314,0.03228014,0.03035405,0.029751362,0.03398891,0.032090932,0.030233484,0.027714087,0.029350067,0.028187223,0.031162726,0.03029153,0.03414756,0.02835399
+0.031707615,0.028367328,0.03251506,0.02744744,0.029457696,0.030506311,0.029133363,0.028074272,0.028364692,0.03026175,0.028753564,0.031309176,0.034676917,0.033179097,0.034860846,0.029045649,0.02786958,0.026573405,0.030232163,0.027840735,0.032212436,0.03031208,0.028782444,0.03392059,0.0323571,0.030005956,0.030180851,0.028813053,0.02663781,0.030342521,0.03171807,0.035299603,0.029240826
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index 9c79cfda..c967ac22 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027990196,0.031357266,0.032083645,0.035231497,0.030950967,0.03400702,0.031629156,0.029572189,0.031043753,0.03288654,0.03334564,0.027693558,0.030571442,0.027530046,0.027211161,0.030589903,0.030670388,0.03259379,0.02737541,0.03036181,0.0295517,0.027888844,0.034106847,0.02740788,0.030691696,0.027825894,0.02871136,0.031667218,0.032172386,0.02943597,0.029995803,0.02745374,0.028395284
+0.028136749,0.03145643,0.032249775,0.035381794,0.031226981,0.033643033,0.031521108,0.02954274,0.030872017,0.033072893,0.033556398,0.027660204,0.030459512,0.027515223,0.027106846,0.030692818,0.030569497,0.032710854,0.027416352,0.030361334,0.029780567,0.027815046,0.033936046,0.027293991,0.03050649,0.027694564,0.028838307,0.03159304,0.03238606,0.02936506,0.029868718,0.027438898,0.028330646
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index b0bcf281..9a40a897 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0324019,0.026557397,0.03567093,0.028395623,0.03214945,0.030262461,0.029736726,0.02918344,0.02924375,0.031181242,0.030299323,0.031787664,0.031068454,0.03598506,0.03070766,0.028389841,0.028155206,0.027619353,0.033341076,0.028269062,0.029442511,0.028486561,0.028317692,0.032146245,0.031487968,0.02932897,0.03275529,0.028782543,0.025840137,0.02867312,0.032408092,0.031715825,0.030209484
-0.03502903,0.026921587,0.029170362,0.02918356,0.03458919,0.03605911,0.029871225,0.03156951,0.027554981,0.033587333,0.03044571,0.029222019,0.030610982,0.032198284,0.030960076,0.033182647,0.031899124,0.029759252,0.028811408,0.026992684,0.030438298,0.025892453,0.03120597,0.0292835,0.03345653,0.031768657,0.026362149,0.032247383,0.03288019,0.027643709,0.025063304,0.027542502,0.028597245
+0.032555487,0.026670558,0.035933323,0.028395625,0.032177225,0.030180544,0.029458955,0.029028589,0.029231155,0.03132255,0.030461755,0.031708784,0.03087897,0.035742663,0.030636476,0.028290126,0.02821839,0.027709417,0.033597182,0.028345522,0.02946698,0.028262552,0.028240243,0.032148767,0.031331882,0.029351693,0.032904956,0.028570952,0.025903815,0.028767973,0.032443482,0.031798936,0.03026446
+0.035266466,0.027003983,0.029438915,0.029231507,0.03491313,0.035511877,0.029532434,0.031505685,0.027652096,0.033847447,0.0306067,0.029304575,0.030613605,0.03229601,0.030978212,0.0330794,0.03178621,0.02973907,0.02895497,0.02695284,0.03035813,0.025822245,0.03114871,0.029226782,0.033070102,0.03168039,0.026520967,0.032253344,0.032956786,0.027481772,0.024973137,0.027664494,0.02862798
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index cb27b45f..9fc8ce15 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02995065,0.026482854,0.036322676,0.030786272,0.030653637,0.030291405,0.02871537,0.027630482,0.028931543,0.031819902,0.03167783,0.030811392,0.033273786,0.03227268,0.031228758,0.028594691,0.030216627,0.026278855,0.031208765,0.02872123,0.031650018,0.029130049,0.030444073,0.030609695,0.03161625,0.028118087,0.032231912,0.029491147,0.025260435,0.02785982,0.03239316,0.0353287,0.029997202
+0.030240778,0.026598437,0.036707304,0.030852366,0.030847387,0.030217325,0.028302671,0.027445653,0.028811673,0.03210606,0.032096457,0.030536423,0.032865256,0.031971786,0.031098066,0.028391903,0.030326098,0.026407817,0.031497795,0.028856413,0.031845562,0.02874827,0.030271817,0.030664615,0.031373702,0.028012546,0.032444347,0.029172558,0.025456056,0.028003585,0.032420244,0.035377927,0.030031078
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index 006a90b1..d3fd32e1 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.031105757,0.031319205,0.025952792,0.03074406,0.03020676,0.03298403,0.029882036,0.031156622,0.029681066,0.031637415,0.029420717,0.029296836,0.032249913,0.026572065,0.0334537,0.03261439,0.03169759,0.029466227,0.026537174,0.029645033,0.032349445,0.030258182,0.032920852,0.029934868,0.031194655,0.030517844,0.025205344,0.031641852,0.03456159,0.0313981,0.027084364,0.029406648,0.027902825
+0.03120083,0.03140158,0.02597218,0.030870566,0.030339655,0.032766707,0.029818222,0.031150384,0.029739713,0.03175302,0.029487183,0.029177235,0.032368988,0.026637957,0.03345298,0.032503065,0.031507123,0.029519858,0.026424674,0.029624622,0.03226006,0.030312635,0.032872517,0.02997535,0.030994944,0.030409798,0.02527944,0.03171313,0.034694858,0.031350695,0.027044604,0.02949311,0.02788231
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
index a462c215..bc5e64b0 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07371652,0.08747254,0.07478085,0.066930644,0.07320184,0.068280794,0.08230257,0.08293058,0.07530409,0.0836406,0.07084773,0.0773642,0.083227016
+0.074083805,0.087974995,0.075273834,0.06646733,0.07345789,0.06847239,0.0818435,0.08279163,0.07455243,0.08277229,0.07101856,0.07774641,0.08354494
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
index 1ffe1910..b27b3ac2 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.0846475,0.07865937,0.0813897,0.06975984,0.08057894,0.07018437,0.076035455,0.068347126,0.07994533,0.07798379,0.07339249,0.07953821,0.079537906
+0.084869295,0.07880388,0.08157886,0.06951565,0.08075639,0.07072125,0.0760694,0.068276174,0.07958265,0.07747237,0.073025174,0.07977048,0.07955835
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
index b9913b9f..83d9348d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.083180524,0.07846569,0.0812608,0.07024677,0.07927988,0.06894813,0.075323164,0.06711979,0.07884185,0.08134555,0.07181041,0.081516474,0.08266096
+0.083191805,0.0784117,0.08124768,0.07007233,0.07956042,0.06936452,0.07542335,0.067135476,0.07857058,0.08090579,0.071468234,0.081839085,0.08280907
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
index 0ef1760e..6d556ab8 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.06989463,0.07989096,0.07249861,0.08051619,0.07235672,0.081912465,0.08144418,0.08826559,0.074629635,0.07509785,0.079737626,0.071991235,0.07176424
-0.07401593,0.07802295,0.075928785,0.082235724,0.0748795,0.0842309,0.08068106,0.08477475,0.07419375,0.07135396,0.082080215,0.06978505,0.067817405
+0.07036531,0.08049916,0.07311549,0.080371074,0.07224095,0.081713885,0.080874875,0.08775907,0.07437307,0.07501077,0.080234654,0.071844436,0.07159725
+0.07462252,0.07882556,0.076731205,0.08212352,0.0746616,0.08412205,0.08021181,0.08402447,0.07401236,0.071274504,0.08248981,0.069406666,0.067493975
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
index 30d5cd15..7894c70f 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07915403,0.08103098,0.07856794,0.06742558,0.07797891,0.068986446,0.07662418,0.0694121,0.0813124,0.08137604,0.07515498,0.081329875,0.08164642
+0.07968078,0.08148541,0.07910197,0.06712172,0.07808477,0.06944092,0.076500565,0.069036186,0.0806076,0.080700286,0.075026445,0.08142645,0.081786945
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
index 6911d301..f8bde07a 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
@@ -1,3 +1,3 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07836008,0.07883829,0.08306367,0.078179605,0.080974005,0.08502495,0.078466825,0.080645144,0.07336398,0.06484835,0.08606815,0.06423272,0.06793416
-0.0754353,0.091055885,0.07524671,0.07041598,0.07320034,0.07423061,0.087269105,0.0889919,0.069685824,0.073581785,0.071758285,0.069817245,0.07931107
+0.07887929,0.07922841,0.0837173,0.07808471,0.08080305,0.08518249,0.0778711,0.080227315,0.073275805,0.06477759,0.08631688,0.063998096,0.067637995
+0.075657144,0.091537185,0.07574736,0.070048064,0.07345481,0.07445074,0.086469546,0.08901579,0.06910504,0.07258692,0.07215753,0.07019448,0.079575315
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
index aec1ea17..15fe6887 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07717769,0.07774059,0.083538964,0.08017507,0.07791889,0.08045961,0.07922038,0.07712727,0.07463028,0.07106932,0.08220443,0.07093723,0.06780032
+0.07799491,0.078715876,0.08464348,0.07995696,0.07754624,0.08058426,0.07884206,0.075935386,0.074297816,0.07091199,0.08250889,0.070520476,0.067541696
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
index 6558c6eb..59e7177d 100644
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
@@ -1,2 +1,2 @@
 [unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.06911243,0.081835695,0.06873813,0.0740041,0.06911324,0.06964999,0.080704,0.081733346,0.0777465,0.08935496,0.06924195,0.084877074,0.08388858
+0.069228604,0.08208278,0.06885877,0.0737444,0.06933344,0.069567144,0.08054654,0.081645325,0.077265635,0.08891717,0.06935114,0.08529186,0.08416724
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
index baa87136..6be5d40e 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
@@ -224,7 +224,7 @@ def test_distributed_sharding(mock_ws, mock_rank, mock_init, mock_config, datase
 
 
 def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
-    """FSDP exact mode rejects uneven ranks."""
+    """Exact distributed mode rejects uneven ranks."""
     data_dir = tmp_path / "uneven_parquet_data"
     data_dir.mkdir()
 
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
index 992bcf74..6f3c8c98 100644
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
@@ -208,11 +208,24 @@ def test_distributed_sharding(
 
 @patch("sequifier.io.sequifier_dataset_from_folder_pt_lazy.get_worker_info")
 def test_dataloader_worker_sharding_continuous_boundaries(
-    mock_worker_info, mock_config, dataset_path, mock_torch_load
+    mock_worker_info, mock_config, tmp_path
 ):
-    """Worker file boundaries for continuous streaming."""
+    """Worker boundaries can start and stop inside PT files."""
+    data_dir = tmp_path / "data_uneven_worker"
+    data_dir.mkdir()
+    file_specs = [
+        ("file1.pt", 7, 0),
+        ("file2.pt", 11, 100),
+        ("file3.pt", 13, 200),
+        ("file4.pt", 9, 300),
+    ]
+    metadata = _folder_metadata(
+        sum(samples for _, samples, _ in file_specs),
+        [{"path": path, "samples": samples} for path, samples, _ in file_specs],
+    )
+    with open(data_dir / "metadata.json", "w") as f:
+        json.dump(metadata, f)
 
-    # Simulate being DataLoader worker ID 1 out of 2 total workers
     mock_info = MagicMock()
     mock_info.id = 1
     mock_info.num_workers = 2
@@ -220,22 +233,63 @@ def test_dataloader_worker_sharding_continuous_boundaries(
 
     mock_config.training_spec.num_workers = 2
 
-    dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config, shuffle=False)
+    file_offsets = {path: (samples, offset) for path, samples, offset in file_specs}
 
-    # Consume the generator for THIS specific worker
-    batches = list(dataset)
+    def load_uneven_file(path, map_location, weights_only):
+        file_name = path.split("/")[-1]
+        samples, offset = file_offsets[file_name]
+        values = torch.arange(offset, offset + samples * STORED_WIDTH).reshape(
+            samples, STORED_WIDTH
+        )
+        return (
+            {"col1": values, "tgt1": values.clone()},
+            None,
+            None,
+            None,
+            torch.zeros(samples, dtype=torch.int64),
+        )
+
+    with patch("torch.load", side_effect=load_uneven_file) as mock_load:
+        dataset = SequifierDatasetFromFolderPtLazy(
+            str(data_dir), mock_config, shuffle=False
+        )
+        batches = list(dataset)
 
-    # NEW LOGIC: Total samples = 40.
-    # Worker 0 boundary: samples 0 to 20 (overlaps file1.pt, file2.pt)
-    # Worker 1 boundary: samples 20 to 40 (overlaps file3.pt, file4.pt)
-    # Since we are testing Worker 1, it should yield 20 samples -> 4 batches
     assert len(batches) == 4
-    assert mock_torch_load.call_count == 2
+    assert mock_load.call_count == 2
 
-    loaded_files = [call.args[0] for call in mock_torch_load.call_args_list]
+    loaded_files = [call.args[0] for call in mock_load.call_args_list]
     assert any("file3.pt" in f for f in loaded_files)
     assert any("file4.pt" in f for f in loaded_files)
     assert not any("file1.pt" in f for f in loaded_files)
+    assert not any("file2.pt" in f for f in loaded_files)
+
+    sample_ids = []
+    for batch in batches:
+        sample_ids.extend(batch.inputs["col1"][:, 0].tolist())
+
+    assert sample_ids == [
+        200 + 2 * STORED_WIDTH,
+        200 + 3 * STORED_WIDTH,
+        200 + 4 * STORED_WIDTH,
+        200 + 5 * STORED_WIDTH,
+        200 + 6 * STORED_WIDTH,
+        200 + 7 * STORED_WIDTH,
+        200 + 8 * STORED_WIDTH,
+        200 + 9 * STORED_WIDTH,
+        200 + 10 * STORED_WIDTH,
+        200 + 11 * STORED_WIDTH,
+        200 + 12 * STORED_WIDTH,
+        300,
+        300 + STORED_WIDTH,
+        300 + 2 * STORED_WIDTH,
+        300 + 3 * STORED_WIDTH,
+        300 + 4 * STORED_WIDTH,
+        300 + 5 * STORED_WIDTH,
+        300 + 6 * STORED_WIDTH,
+        300 + 7 * STORED_WIDTH,
+        300 + 8 * STORED_WIDTH,
+    ]
 
 
 @patch("torch.distributed.is_initialized", return_value=True)
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
index 2c9d8c78..b3d836ba 100644
--- a/tests/unit/test_infer.py
+++ b/tests/unit/test_infer.py
@@ -74,7 +74,7 @@ def test_sample_with_cumsum(mock_rand):
     sampled_from_log_probs = sample_with_cumsum(log_probs, is_log_probs=True)
     np.testing.assert_array_equal(sampled_from_log_probs, [0, 1])
 
-    # Path 2: Test with logits=False (pre-normalized probabilities)
+    # Path 2: Test with is_log_probs=False (pre-normalized probabilities)
     pure_probs = np.array([[0.1, 0.9], [0.8, 0.2]])
     sampled_from_probs = sample_with_cumsum(pure_probs, is_log_probs=False)
     np.testing.assert_array_equal(sampled_from_probs, [0, 1])
@@ -433,9 +433,10 @@ def test_calculate_item_positions_causal_uses_future_window_tail():
     np.testing.assert_array_equal(positions, [13, 14, 23, 24])
 
 
-def _bert_inference_config(tmp_path, model_type="generative"):
+def _bert_inference_config(tmp_path, model_type="generative", input_columns=None):
     data_path = tmp_path / "data.parquet"
     data_path.touch()
+    input_columns = input_columns or ["target_col"]
 
     return InfererModel(
         project_root=str(tmp_path),
@@ -444,11 +445,11 @@ def _bert_inference_config(tmp_path, model_type="generative"):
         model_type=model_type,
         training_objective="bert",
         data_path=str(data_path),
-        input_columns=["target_col"],
-        categorical_columns=["target_col"],
+        input_columns=input_columns,
+        categorical_columns=input_columns,
         real_columns=[],
         target_columns=["target_col"],
-        column_types={"target_col": "int64"},
+        column_types={col: "int64" for col in input_columns},
         target_column_types={"target_col": "categorical"},
         seed=42,
         device="cpu",
@@ -485,15 +486,15 @@ def _bert_preprocessed_frame():
 def _bert_preprocessed_frame_out_of_order():
     return pl.DataFrame(
         {
-            "sequenceId": [2, 1],
-            "subsequenceId": [0, 0],
-            "startItemPosition": [200, 100],
-            "leftPadLength": [1, 2],
-            "inputCol": ["target_col", "target_col"],
-            "3": [0, 0],
-            "2": [3, 0],
-            "1": [4, 3],
-            "0": [3, 4],
+            "sequenceId": [2, 2, 1, 1],
+            "subsequenceId": [0, 0, 0, 0],
+            "startItemPosition": [200, 200, 100, 100],
+            "leftPadLength": [1, 1, 2, 2],
+            "inputCol": ["aux_col", "target_col", "aux_col", "target_col"],
+            "3": [9, 0, 8, 0],
+            "2": [9, 3, 8, 0],
+            "1": [9, 4, 8, 3],
+            "0": [9, 3, 8, 4],
         }
     )
 
@@ -563,7 +564,7 @@ def capture_write(dataframe, path, write_format):
 
 
 def test_bert_generative_inference_uses_dataframe_order_for_padding_masks(tmp_path):
-    config = _bert_inference_config(tmp_path)
+    config = _bert_inference_config(tmp_path, input_columns=["aux_col", "target_col"])
     data = _bert_preprocessed_frame_out_of_order()
     written = []
 
@@ -580,7 +581,11 @@ def capture_write(dataframe, path, write_format):
         "sequifier.infer.get_probs_preds_from_df", return_value=(None, preds)
     ), patch("sequifier.infer.write_data", side_effect=capture_write):
         infer_generative(
-            config, inferer, "bert-model", [data], {"target_col": torch.int64}
+            config,
+            inferer,
+            "bert-model",
+            [data],
+            {"target_col": torch.int64, "aux_col": torch.int64},
         )
 
     predictions = next(frame for path, frame, _ in written if "predictions" in path)
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
index 837337ff..954076b7 100644
--- a/tests/unit/test_train.py
+++ b/tests/unit/test_train.py
@@ -240,6 +240,28 @@ def _loss_shell_model(
     return model
 
 
+class _IdentityScaler:
+    def scale(self, loss):
+        return loss
+
+    def unscale_(self, optimizer):
+        return None
+
+    def step(self, optimizer):
+        optimizer.step()
+
+    def update(self):
+        return None
+
+
+class _NoopLogger:
+    def info(self, *args, **kwargs):
+        return None
+
+    def warning(self, *args, **kwargs):
+        return None
+
+
 def test_transformer_model_initialization(model, model_config):
     """Expected model layers."""
     # Check if encoder dicts were created
@@ -357,7 +379,6 @@ def test_load_train_config_defaults_missing_metadata_special_token_ids(
                 "column_types": config_values["column_types"],
                 "n_classes": config_values["n_classes"],
                 "id_maps": config_values["id_maps"],
-                "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
             }
         )
     )
@@ -855,25 +876,36 @@ def test_training_loss_enables_fsdp_token_weighting(
     model.hparams = SimpleNamespace(
         training_spec=SimpleNamespace(data_parallelism="FSDP")
     )
-    output = {"real_col": torch.zeros(1, 1, 1)}
+    output = {"real_col": torch.zeros(1, 1, 1, requires_grad=True)}
     targets = {"real_col": torch.ones(1, 1)}
     metadata = {"target_valid_mask": torch.ones(1, 1, dtype=torch.bool)}
+    process_group = object()
+    model._data_parallel_process_group = lambda: process_group
 
     monkeypatch.setattr("sequifier.train.dist.is_available", lambda: True)
     monkeypatch.setattr("sequifier.train.dist.is_initialized", lambda: True)
-    monkeypatch.setattr(
-        "sequifier.train.dist.get_world_size",
-        lambda group=None: 2,
-    )
+
+    def fake_get_world_size(group=None):
+        assert group is process_group
+        return 2
+
+    monkeypatch.setattr("sequifier.train.dist.get_world_size", fake_get_world_size)
+
+    reduce_calls = []
 
     def fake_all_reduce(tensor, op=None, group=None):
-        tensor.fill_(2)
+        reduce_calls.append((tensor.detach().clone(), op, group))
+        tensor.fill_(4)
 
     monkeypatch.setattr("sequifier.train.dist.all_reduce", fake_all_reduce)
 
     loss, _ = TransformerModel._calculate_loss(model, output, targets, metadata)
+    loss.backward()
 
-    assert torch.isclose(loss, torch.tensor(1.0))
+    assert reduce_calls
+    assert reduce_calls[0][2] is process_group
+    assert torch.isclose(loss, torch.tensor(0.5))
+    assert torch.allclose(output["real_col"].grad, torch.tensor([[[-1.0]]]))
 
 
 def test_evaluation_loss_mask_intersects_target_bert_and_sample_masks():
@@ -1322,6 +1354,71 @@ def test_train_epoch_steps_once_for_mixed_empty_and_nonempty_accumulation_window
     assert model.scheduler.get_last_lr()[0] == pytest.approx(0.001)
 
 
+def test_train_epoch_divides_backward_loss_by_accumulation_steps(model, monkeypatch):
+    model.rank = 0
+    model.log_interval = 100
+    model.accumulation_steps = 2
+    model.scheduler_step_on = "batch"
+    model.start_batch = 0
+    model.scaler = _IdentityScaler()
+    model.logger = _NoopLogger()
+    model.save_latest_interval_minutes = None
+    model.save_batch_interval_minutes = None
+    model.save_batch_interval_minutes_val_loss = False
+    model.last_latest_save_time = 0.0
+    model.last_batch_save_time = 0.0
+    tracked_param = next(model.parameters())
+    tracked_value_before = tracked_param.detach().reshape(-1)[0].clone()
+    model.optimizer = torch.optim.SGD([tracked_param], lr=0.1)
+    model.scheduler = torch.optim.lr_scheduler.StepLR(
+        model.optimizer,
+        step_size=1,
+        gamma=0.1,
+    )
+    coefficients = iter([0.2, 0.4])
+
+    def fake_forward(data, metadata=None, return_logits=True):
+        return {}
+
+    def fake_calculate_training_loss(output, targets, metadata):
+        coefficient = next(coefficients)
+        loss = tracked_param.reshape(-1)[0] * coefficient
+        loss_sums = {
+            target_name: loss.detach()
+            for target_name in model._loss_target_names(targets)
+        }
+        count = torch.tensor(1, dtype=torch.int64)
+        return loss, {}, loss_sums, count, count
+
+    monkeypatch.setattr(model, "forward", fake_forward)
+    monkeypatch.setattr(
+        model,
+        "_calculate_training_loss",
+        fake_calculate_training_loss,
+    )
+
+    seq_len = model.window_view.context_length
+    batch = _train_epoch_test_batch(
+        model,
+        torch.ones(1, seq_len, dtype=torch.bool),
+    )
+
+    TransformerModel._train_epoch(
+        model,
+        DataLoader([batch, batch], batch_size=None),
+        DataLoader([], batch_size=None),
+        epoch=1,
+    )
+
+    tracked_value_after = tracked_param.detach().reshape(-1)[0]
+
+    assert torch.allclose(
+        tracked_value_after,
+        tracked_value_before - torch.tensor(0.03),
+        atol=1e-6,
+    )
+
+
 def test_padding_keys_are_masked(bert_model):
     seq_len = bert_model.window_view.context_length
 
diff --git a/tests/unit/test_train_distributed_loss.py b/tests/unit/test_train_distributed_loss.py
index 99c06f0d..0e94cf0f 100644
--- a/tests/unit/test_train_distributed_loss.py
+++ b/tests/unit/test_train_distributed_loss.py
@@ -169,7 +169,7 @@ def _single_process_real_update(seq_lens, selected_counts, target_values, lr):
     return model.param.detach().clone()
 
 
-def _single_process_accumulated_real_update(microbatches, lr):
+def _single_process_accumulated_real_update(microbatches, lr, accumulation_steps):
     model = _TinyRealModel()
     optimizer = torch.optim.SGD(model.parameters(), lr=lr)
     shell = _loss_shell({"real_col": "real"})
@@ -185,12 +185,29 @@ def _single_process_accumulated_real_update(microbatches, lr):
             targets,
             metadata,
         )
-        loss.backward()
+        (loss / accumulation_steps).backward()
     optimizer.step()
     return model.param.detach().clone()
 
 
-def _analytical_accumulated_real_update(microbatches, lr):
+def _analytical_accumulated_real_update(microbatches, lr, accumulation_steps):
+    accumulated_grad = 0.0
+    for microbatch in microbatches:
+        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
+        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
+        token_count = selected_counts.sum().item()
+        if token_count == 0:
+            continue
+
+        weighted_target_mean = (
+            selected_counts * target_values
+        ).sum().item() / token_count
+        accumulated_grad += -2.0 * weighted_target_mean / accumulation_steps
+
+    return torch.tensor(-lr * accumulated_grad)
+
+
+def _analytical_unnormalized_accumulated_real_update(microbatches, lr):
     accumulated_grad = 0.0
     for microbatch in microbatches:
         selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
@@ -367,7 +384,7 @@ def put_result(tensor):
                             metadata,
                         )
                     )
-                    loss.backward()
+                    (loss / case["accumulation_steps"]).backward()
                     accumulated_global_token_count += global_count.detach()
                     optimizer_step_due = (batch_idx + 1) % case[
                         "accumulation_steps"
@@ -415,7 +432,7 @@ def put_result(tensor):
                         targets,
                         metadata,
                     )
-                    loss.backward()
+                    (loss / case["accumulation_steps"]).backward()
                 optimizer.step()
                 if rank == 0:
                     put_result(model.param)
@@ -527,8 +544,8 @@ def _run_ddp_case(case):
 
 
 pytestmark = pytest.mark.skipif(
-    not dist.is_available(),
-    reason="torch.distributed is not available",
+    not dist.is_gloo_available(),
+    reason="torch.is_gloo_available is False",
 )
 
 
@@ -688,7 +705,7 @@ def test_ddp_train_epoch_globally_empty_accumulation_window_leaves_state_unchang
     assert optimizer_state_len == 0
 
 
-def test_ddp_gradient_accumulation_preserves_per_microbatch_weighting():
+def test_ddp_gradient_accumulation_averages_per_microbatch_weighting():
     microbatches = [
         {
             "seq_lens": [1, 1],
@@ -704,14 +721,21 @@ def test_ddp_gradient_accumulation_preserves_per_microbatch_weighting():
     case = {
         "kind": "accumulation_update",
         "microbatches": microbatches,
+        "accumulation_steps": 2,
         "lr": 0.05,
     }
 
     ddp_param = _run_ddp_case(case)
-    reference_param = _analytical_accumulated_real_update(microbatches, case["lr"])
+    reference_param = _analytical_accumulated_real_update(
+        microbatches, case["lr"], case["accumulation_steps"]
+    )
+    unnormalized_param = _analytical_unnormalized_accumulated_real_update(
+        microbatches, case["lr"]
+    )
     combined_batch_param = _analytical_combined_real_update(microbatches, case["lr"])
 
     assert torch.allclose(ddp_param, reference_param, atol=1e-6)
+    assert not torch.allclose(ddp_param, unnormalized_param, atol=1e-3)
     assert not torch.allclose(ddp_param, combined_batch_param, atol=1e-3)
 
 
diff --git a/tests/unit/test_train_fsdp_loss.py b/tests/unit/test_train_fsdp_loss.py
index dbdd09cc..5c092d68 100644
--- a/tests/unit/test_train_fsdp_loss.py
+++ b/tests/unit/test_train_fsdp_loss.py
@@ -174,7 +174,7 @@ def _single_process_real_update(seq_lens, selected_counts, target_values, lr):
     return model.param.detach().clone()
 
 
-def _single_process_accumulated_real_update(microbatches, lr):
+def _single_process_accumulated_real_update(microbatches, lr, accumulation_steps):
     model = _TinyRealModel()
     optimizer = torch.optim.SGD(model.parameters(), lr=lr)
     shell = _loss_shell({"real_col": "real"}, device="cpu")
@@ -190,12 +190,29 @@ def _single_process_accumulated_real_update(microbatches, lr):
             targets,
             metadata,
         )
-        loss.backward()
+        (loss / accumulation_steps).backward()
     optimizer.step()
     return model.param.detach().clone()
 
 
-def _analytical_accumulated_real_update(microbatches, lr):
+def _analytical_accumulated_real_update(microbatches, lr, accumulation_steps):
+    accumulated_grad = 0.0
+    for microbatch in microbatches:
+        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
+        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
+        token_count = selected_counts.sum().item()
+        if token_count == 0:
+            continue
+
+        weighted_target_mean = (
+            selected_counts * target_values
+        ).sum().item() / token_count
+        accumulated_grad += -2.0 * weighted_target_mean / accumulation_steps
+
+    return torch.tensor([-lr * accumulated_grad, 0.0])
+
+
+def _analytical_unnormalized_accumulated_real_update(microbatches, lr):
     accumulated_grad = 0.0
     for microbatch in microbatches:
         selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
@@ -403,7 +420,7 @@ def _fsdp_case_worker(rank, world_size, init_method, case, queue):
                             metadata,
                         )
                     )
-                    loss.backward()
+                    (loss / case["accumulation_steps"]).backward()
                     accumulated_global_token_count += global_count.detach()
                     optimizer_step_due = (batch_idx + 1) % case[
                         "accumulation_steps"
@@ -455,7 +472,7 @@ def _fsdp_case_worker(rank, world_size, init_method, case, queue):
                         targets,
                         metadata,
                     )
-                    loss.backward()
+                    (loss / case["accumulation_steps"]).backward()
             else:
                 seq_len = case["seq_lens"][rank]
                 targets, metadata = _real_batch(
@@ -647,7 +664,7 @@ def test_fsdp_world_size_two_optimizer_update_matches_single_process_reference()
     assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
 
 
-def test_fsdp_gradient_accumulation_preserves_per_microbatch_weighting():
+def test_fsdp_gradient_accumulation_averages_per_microbatch_weighting():
     microbatches = [
         {
             "seq_lens": [1, 1],
@@ -663,14 +680,21 @@ def test_fsdp_gradient_accumulation_preserves_per_microbatch_weighting():
     case = {
         "kind": "accumulation_update",
         "microbatches": microbatches,
+        "accumulation_steps": 2,
         "lr": 0.05,
     }
 
     fsdp_param = _run_fsdp_case(case)
-    reference_param = _analytical_accumulated_real_update(microbatches, case["lr"])
+    reference_param = _analytical_accumulated_real_update(
+        microbatches, case["lr"], case["accumulation_steps"]
+    )
+    unnormalized_param = _analytical_unnormalized_accumulated_real_update(
+        microbatches, case["lr"]
+    )
     combined_batch_param = _analytical_combined_real_update(microbatches, case["lr"])
 
     assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
+    assert not torch.allclose(fsdp_param, unnormalized_param, atol=1e-3)
     assert not torch.allclose(fsdp_param, combined_batch_param, atol=1e-3)
 
 

From d61a1b7409479a570453c14179ac2c53fa64e93b Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Tue, 16 Jun 2026 13:07:23 +0200
Subject: [PATCH 65/81] Make preprocessing resumption more efficient

---
 src/sequifier/preprocess.py | 57 ++++++++++++++++++++++++++++++-------
 1 file changed, 46 insertions(+), 11 deletions(-)

diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index cd5bd024..c2556d03 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -1261,6 +1261,31 @@ def _check_file_has_been_processed(
         return False
 
 
+@beartype
+def _get_processed_prefixes(
+    project_root: str,
+    target_dir: str,
+    write_format: str,
+) -> set[str]:
+    temp_dir = Path(project_root) / "data" / target_dir
+
+    if not temp_dir.is_dir():
+        return set()
+
+    suffix = f".{write_format}"
+    processed = set()
+
+    with os.scandir(temp_dir) as entries:
+        for entry in entries:
+            if not entry.is_file() or not entry.name.endswith(suffix):
+                continue
+
+            if "-split" in entry.name:
+                processed.add(entry.name.rsplit("-split", 1)[0])
+
+    return processed
+
+
 @beartype
 def _process_batches_multiple_files_inner(
     project_root: str,
@@ -1290,6 +1315,12 @@ def _process_batches_multiple_files_inner(
     mask_column: Optional[str],
 ):
     """Process this worker's file shard."""
+
+    processed_prefixes = (
+        _get_processed_prefixes(project_root, target_dir, write_format)
+        if continue_preprocessing and not merge_output
+        else set()
+    )
     n_files = len(file_paths)
     if n_files <= 0:
         raise ValueError("No files found to process.")
@@ -1307,17 +1338,21 @@ def _process_batches_multiple_files_inner(
                 for path in split_paths
             ]
             if continue_preprocessing:
-                file_has_been_processed = _check_file_has_been_processed(
-                    project_root,
-                    data_name_root,
-                    process_id,
-                    split_ratios,
-                    write_format,
-                    target_dir,
-                    merge_output,
-                    file_index_str,
-                )
-
+                file_prefix_str = f"{data_name_root}-{process_id}-{file_index_str}"
+
+                if not merge_output:
+                    file_has_been_processed = file_prefix_str in processed_prefixes
+                else:
+                    file_has_been_processed = _check_file_has_been_processed(
+                        project_root,
+                        data_name_root,
+                        process_id,
+                        split_ratios,
+                        write_format,
+                        target_dir,
+                        merge_output,
+                        file_index_str,
+                    )
                 if file_has_been_processed:
                     logger.info(f"Skipping already processed file: {path}")
                     if max_rows is not None:

From 3fd2784bc04c099017ff6d009c0fc49e3cbe25a2 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Tue, 16 Jun 2026 15:31:23 +0200
Subject: [PATCH 66/81] pyright

---
 pyrightconfig.json     | 3 ++-
 src/sequifier/train.py | 8 ++++----
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/pyrightconfig.json b/pyrightconfig.json
index c31d3b15..2413032c 100644
--- a/pyrightconfig.json
+++ b/pyrightconfig.json
@@ -1,4 +1,5 @@
 {
     "reportMissingImports": false,
-    "reportMissingModuleSource": false
+    "reportMissingModuleSource": false,
+    "useLibraryCodeForTypes": true
 }
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index b0b28bed..3d6123f3 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -21,7 +21,7 @@
 from beartype import beartype  # noqa: E402
 from packaging import version  # noqa: E402
 from torch import Tensor, nn  # noqa: E402
-from torch.amp import GradScaler  # noqa: E402
+from torch.amp.grad_scaler import GradScaler  # noqa: E402
 from torch.distributed.checkpoint.state_dict import (  # noqa: E402
     StateDictOptions,
     get_model_state_dict,
@@ -38,9 +38,9 @@
     )
 else:
     from torch.distributed._composable.fsdp import (  # noqa: E402
-        MixedPrecisionPolicy,
-        OffloadPolicy,
-        fully_shard,
+        MixedPrecisionPolicy,  # type: ignore
+        OffloadPolicy,  # type: ignore
+        fully_shard,  # type: ignore
     )
 
 from torch.distributed.device_mesh import init_device_mesh  # noqa: E402

From d7ee14fba8706428b1dcebee000dbcd48e8cd38d Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Tue, 16 Jun 2026 15:33:01 +0200
Subject: [PATCH 67/81] move tests

---
 .../hyperparameter-search-bayesian.yaml       |   68 -
 tests/configs/hyperparameter-search-bert.yaml |   76 -
 ...arameter-search-custom-eval-inference.yaml |   41 -
 .../hyperparameter-search-custom-eval.yaml    |   92 -
 tests/configs/hyperparameter-search-grid.yaml |   70 -
 .../configs/hyperparameter-search-sample.yaml |   86 -
 ...infer-test-categorical-autoregression.yaml |   25 -
 ...infer-test-categorical-bert-embedding.yaml |   25 -
 .../configs/infer-test-categorical-bert.yaml  |   25 -
 .../infer-test-categorical-embedding.yaml     |   24 -
 .../infer-test-categorical-inf-size-1.yaml    |   24 -
 ...test-categorical-inf-size-3-embedding.yaml |   27 -
 .../infer-test-categorical-inf-size-3.yaml    |   27 -
 .../infer-test-categorical-multitarget.yaml   |   26 -
 tests/configs/infer-test-categorical.yaml     |   24 -
 .../infer-test-distributed-parquet.yaml       |   25 -
 tests/configs/infer-test-distributed.yaml     |   23 -
 tests/configs/infer-test-lazy.yaml            |   21 -
 .../infer-test-real-autoregression.yaml       |   26 -
 tests/configs/infer-test-real.yaml            |   21 -
 .../preprocess-test-categorical-exact-pt.yaml |   15 -
 .../preprocess-test-categorical-exact.yaml    |   15 -
 ...eprocess-test-categorical-interrupted.yaml |   19 -
 ...eprocess-test-categorical-lookahead-0.yaml |   19 -
 ...eprocess-test-categorical-multitarget.yaml |   20 -
 ...ategorical-precomputed-stats-negative.yaml |   15 -
 ...ss-test-categorical-precomputed-stats.yaml |   16 -
 .../configs/preprocess-test-categorical.yaml  |   19 -
 tests/configs/preprocess-test-multi-file.yaml |   21 -
 tests/configs/preprocess-test-real.yaml       |   17 -
 .../configs/train-test-categorical-bert.yaml  |   68 -
 .../train-test-categorical-inf-size-1.yaml    |   61 -
 .../train-test-categorical-inf-size-3.yaml    |   68 -
 ...in-test-categorical-multitarget-eager.yaml |   78 -
 .../train-test-categorical-multitarget.yaml   |   79 -
 tests/configs/train-test-categorical.yaml     |   60 -
 .../train-test-distributed-lazy-parquet.yaml  |   52 -
 tests/configs/train-test-distributed.yaml     |   52 -
 tests/configs/train-test-lazy.yaml            |   48 -
 tests/configs/train-test-real-bert.yaml       |   68 -
 tests/configs/train-test-real.yaml            |   60 -
 tests/configs/train-test-resume-epoch.yaml    |   59 -
 .../configs/train-test-resume-mid-epoch.yaml  |   61 -
 tests/integration-test-log.txt                |   80 -
 tests/integration/conftest.py                 |  926 ----------
 .../integration/test_hyperparameter_search.py |   90 -
 tests/integration/test_identities.py          |   83 -
 tests/integration/test_inference.py           |  536 ------
 tests/integration/test_make.py                |  184 --
 tests/integration/test_onnx_export.py         |  247 ---
 tests/integration/test_preprocessing.py       |  545 ------
 tests/integration/test_training.py            |  148 --
 tests/integration/test_visualize_training.py  |   54 -
 .../source_configs/id_maps/itemId.json        |    1 -
 .../preprocess-manifest.json                  |   35 -
 ...ta-categorical-1-interrupted-split0-0-0.pt |  Bin 4166 -> 0 bytes
 ...ta-categorical-1-interrupted-split0-1-0.pt |  Bin 3910 -> 0 bytes
 ...ta-categorical-1-interrupted-split0-2-0.pt |  Bin 3782 -> 0 bytes
 ...ta-categorical-1-interrupted-split0-3-0.pt |  Bin 3910 -> 0 bytes
 ...ta-categorical-1-interrupted-split1-0-0.pt |  Bin 3206 -> 0 bytes
 ...ta-categorical-1-interrupted-split1-1-0.pt |  Bin 3206 -> 0 bytes
 ...ta-categorical-1-interrupted-split1-2-0.pt |  Bin 3206 -> 0 bytes
 ...ta-categorical-1-interrupted-split1-3-0.pt |  Bin 3334 -> 0 bytes
 ...ta-categorical-1-interrupted-split2-0-0.pt |  Bin 3206 -> 0 bytes
 ...ta-categorical-1-interrupted-split2-1-0.pt |  Bin 3206 -> 0 bytes
 ...ta-categorical-1-interrupted-split2-2-0.pt |  Bin 3206 -> 0 bytes
 ...ta-categorical-1-interrupted-split2-3-0.pt |  Bin 3334 -> 0 bytes
 .../test-data-categorical-1-interrupted.csv   |  201 ---
 .../test-data-categorical-1-lookahead-0.csv   |  201 ---
 .../source_data/test-data-categorical-1.csv   |  201 ---
 .../test-data-categorical-3-equal.csv         |  161 --
 .../test-data-categorical-3-equal.parquet     |  Bin 4744 -> 0 bytes
 .../source_data/test-data-categorical-3.csv   |  201 ---
 .../source_data/test-data-categorical-5.csv   |  201 ---
 .../source_data/test-data-categorical-50.csv  |  201 ---
 .../test-data-categorical-mullti-file-1.csv   |  245 ---
 .../test-data-categorical-mullti-file-2.csv   |  255 ---
 .../test-data-categorical-mullti-file-3.csv   |  255 ---
 .../test-data-categorical-multitarget-5.csv   |  201 ---
 ...categorical-precomputed-stats-negative.csv |  201 ---
 ...est-data-categorical-precomputed-stats.csv |  201 ---
 ...test-data-real-1-split1-autoregression.csv |   20 -
 .../source_data/test-data-real-1.csv          |  201 ---
 .../source_data/test-data-real-3.csv          |  201 ---
 .../source_data/test-data-real-5.csv          |  201 ---
 .../source_data/test-data-real-50.csv         |  201 ---
 .../source_scripts/hp_search_eval_script.py   |   38 -
 ...orical-1-best-embedding-3-0-embeddings.csv |    2 -
 ...orical-1-best-embedding-3-1-embeddings.csv |    2 -
 ...orical-1-best-embedding-3-2-embeddings.csv |    2 -
 ...orical-1-best-embedding-3-3-embeddings.csv |    3 -
 ...orical-1-best-embedding-3-4-embeddings.csv |    2 -
 ...orical-1-best-embedding-3-5-embeddings.csv |    3 -
 ...orical-1-best-embedding-3-6-embeddings.csv |    2 -
 ...orical-1-best-embedding-3-7-embeddings.csv |    2 -
 ...inf-size-best-embedding-3-0-embeddings.csv |    4 -
 ...inf-size-best-embedding-3-1-embeddings.csv |    4 -
 ...inf-size-best-embedding-3-2-embeddings.csv |    4 -
 ...inf-size-best-embedding-3-3-embeddings.csv |    7 -
 ...inf-size-best-embedding-3-4-embeddings.csv |    4 -
 ...inf-size-best-embedding-3-5-embeddings.csv |    7 -
 ...inf-size-best-embedding-3-6-embeddings.csv |    4 -
 ...inf-size-best-embedding-3-7-embeddings.csv |    4 -
 ...cal-bert-best-embedding-1-0-embeddings.csv |    6 -
 ...cal-bert-best-embedding-1-1-embeddings.csv |    6 -
 ...cal-bert-best-embedding-1-2-embeddings.csv |    5 -
 ...cal-bert-best-embedding-1-3-embeddings.csv |    9 -
 ...cal-bert-best-embedding-1-4-embeddings.csv |    5 -
 ...cal-bert-best-embedding-1-5-embeddings.csv |   10 -
 ...cal-bert-best-embedding-1-6-embeddings.csv |    6 -
 ...cal-bert-best-embedding-1-7-embeddings.csv |    5 -
 ...-1-best-3-autoregression-0-predictions.csv |   21 -
 ...-1-best-3-autoregression-1-predictions.csv |   21 -
 ...-1-best-3-autoregression-2-predictions.csv |   21 -
 ...-1-best-3-autoregression-3-predictions.csv |   41 -
 ...-1-best-3-autoregression-4-predictions.csv |   21 -
 ...-1-best-3-autoregression-5-predictions.csv |   41 -
 ...-1-best-3-autoregression-6-predictions.csv |   21 -
 ...-1-best-3-autoregression-7-predictions.csv |   21 -
 ...del-categorical-1-best-3-0-predictions.csv |    2 -
 ...del-categorical-1-best-3-1-predictions.csv |    2 -
 ...del-categorical-1-best-3-2-predictions.csv |    2 -
 ...del-categorical-1-best-3-3-predictions.csv |    3 -
 ...del-categorical-1-best-3-4-predictions.csv |    2 -
 ...del-categorical-1-best-3-5-predictions.csv |    3 -
 ...del-categorical-1-best-3-6-predictions.csv |    2 -
 ...del-categorical-1-best-3-7-predictions.csv |    2 -
 ...orical-1-inf-size-best-3-0-predictions.csv |    4 -
 ...orical-1-inf-size-best-3-1-predictions.csv |    4 -
 ...orical-1-inf-size-best-3-2-predictions.csv |    4 -
 ...orical-1-inf-size-best-3-3-predictions.csv |    7 -
 ...orical-1-inf-size-best-3-4-predictions.csv |    4 -
 ...orical-1-inf-size-best-3-5-predictions.csv |    7 -
 ...orical-1-inf-size-best-3-6-predictions.csv |    4 -
 ...orical-1-inf-size-best-3-7-predictions.csv |    4 -
 ...del-categorical-3-best-3-0-predictions.csv |    2 -
 ...del-categorical-3-best-3-1-predictions.csv |    2 -
 ...del-categorical-3-best-3-2-predictions.csv |    2 -
 ...del-categorical-3-best-3-3-predictions.csv |    3 -
 ...del-categorical-3-best-3-4-predictions.csv |    2 -
 ...del-categorical-3-best-3-5-predictions.csv |    3 -
 ...del-categorical-3-best-3-6-predictions.csv |    2 -
 ...del-categorical-3-best-3-7-predictions.csv |    2 -
 ...orical-3-inf-size-best-3-0-predictions.csv |    4 -
 ...orical-3-inf-size-best-3-1-predictions.csv |    4 -
 ...orical-3-inf-size-best-3-2-predictions.csv |    4 -
 ...orical-3-inf-size-best-3-3-predictions.csv |    7 -
 ...orical-3-inf-size-best-3-4-predictions.csv |    4 -
 ...orical-3-inf-size-best-3-5-predictions.csv |    7 -
 ...orical-3-inf-size-best-3-6-predictions.csv |    4 -
 ...orical-3-inf-size-best-3-7-predictions.csv |    4 -
 ...del-categorical-5-best-3-0-predictions.csv |    2 -
 ...del-categorical-5-best-3-1-predictions.csv |    2 -
 ...del-categorical-5-best-3-2-predictions.csv |    2 -
 ...del-categorical-5-best-3-3-predictions.csv |    3 -
 ...del-categorical-5-best-3-4-predictions.csv |    2 -
 ...del-categorical-5-best-3-5-predictions.csv |    3 -
 ...del-categorical-5-best-3-6-predictions.csv |    2 -
 ...del-categorical-5-best-3-7-predictions.csv |    2 -
 ...el-categorical-50-best-3-0-predictions.csv |    2 -
 ...el-categorical-50-best-3-1-predictions.csv |    2 -
 ...el-categorical-50-best-3-2-predictions.csv |    2 -
 ...el-categorical-50-best-3-3-predictions.csv |    3 -
 ...el-categorical-50-best-3-4-predictions.csv |    2 -
 ...el-categorical-50-best-3-5-predictions.csv |    3 -
 ...el-categorical-50-best-3-6-predictions.csv |    2 -
 ...el-categorical-50-best-3-7-predictions.csv |    2 -
 ...-categorical-bert-best-1-0-predictions.csv |    6 -
 ...-categorical-bert-best-1-1-predictions.csv |    6 -
 ...-categorical-bert-best-1-2-predictions.csv |    5 -
 ...-categorical-bert-best-1-3-predictions.csv |    9 -
 ...-categorical-bert-best-1-4-predictions.csv |    5 -
 ...-categorical-bert-best-1-5-predictions.csv |   10 -
 ...-categorical-bert-best-1-6-predictions.csv |    6 -
 ...-categorical-bert-best-1-7-predictions.csv |    5 -
 ...rical-distributed-best-3-0-predictions.csv |    2 -
 ...rical-distributed-best-3-1-predictions.csv |    2 -
 ...rical-distributed-best-3-2-predictions.csv |    2 -
 ...rical-distributed-best-3-3-predictions.csv |    3 -
 ...rical-distributed-best-3-4-predictions.csv |    2 -
 ...rical-distributed-best-3-5-predictions.csv |    3 -
 ...rical-distributed-best-3-6-predictions.csv |    2 -
 ...rical-distributed-best-3-7-predictions.csv |    2 -
 ...-categorical-lazy-best-3-0-predictions.csv |    2 -
 ...-categorical-lazy-best-3-1-predictions.csv |    2 -
 ...-categorical-lazy-best-3-2-predictions.csv |    2 -
 ...-categorical-lazy-best-3-3-predictions.csv |    3 -
 ...-categorical-lazy-best-3-4-predictions.csv |    2 -
 ...-categorical-lazy-best-3-5-predictions.csv |    3 -
 ...-categorical-lazy-best-3-6-predictions.csv |    2 -
 ...-categorical-lazy-best-3-7-predictions.csv |    2 -
 ...cal-multitarget-5-best-3-0-predictions.csv |    2 -
 ...cal-multitarget-5-best-3-1-predictions.csv |    2 -
 ...cal-multitarget-5-best-3-2-predictions.csv |    2 -
 ...cal-multitarget-5-best-3-3-predictions.csv |    3 -
 ...cal-multitarget-5-best-3-4-predictions.csv |    2 -
 ...cal-multitarget-5-best-3-5-predictions.csv |    3 -
 ...cal-multitarget-5-best-3-6-predictions.csv |    2 -
 ...cal-multitarget-5-best-3-7-predictions.csv |    2 -
 ...cal-multitarget-5-last-3-0-predictions.csv |    2 -
 ...cal-multitarget-5-last-3-1-predictions.csv |    2 -
 ...cal-multitarget-5-last-3-2-predictions.csv |    2 -
 ...cal-multitarget-5-last-3-3-predictions.csv |    3 -
 ...cal-multitarget-5-last-3-4-predictions.csv |    2 -
 ...cal-multitarget-5-last-3-5-predictions.csv |    3 -
 ...cal-multitarget-5-last-3-6-predictions.csv |    2 -
 ...cal-multitarget-5-last-3-7-predictions.csv |    2 -
 ...al-1-best-3-autoregression-predictions.csv |  201 ---
 ...uifier-model-real-1-best-3-predictions.csv |   25 -
 ...uifier-model-real-3-best-3-predictions.csv |   25 -
 ...uifier-model-real-5-best-3-predictions.csv |   25 -
 ...ifier-model-real-50-best-3-predictions.csv |   25 -
 ...custom-eval-run-0-best-3-0-predictions.csv |   31 -
 ...custom-eval-run-0-best-3-1-predictions.csv |   31 -
 ...custom-eval-run-0-best-3-2-predictions.csv |   31 -
 ...custom-eval-run-0-best-3-3-predictions.csv |   61 -
 ...custom-eval-run-0-best-3-4-predictions.csv |   31 -
 ...custom-eval-run-0-best-3-5-predictions.csv |   61 -
 ...custom-eval-run-0-best-3-6-predictions.csv |   31 -
 ...custom-eval-run-0-best-3-7-predictions.csv |   31 -
 ...custom-eval-run-1-best-3-0-predictions.csv |   31 -
 ...custom-eval-run-1-best-3-1-predictions.csv |   31 -
 ...custom-eval-run-1-best-3-2-predictions.csv |   31 -
 ...custom-eval-run-1-best-3-3-predictions.csv |   61 -
 ...custom-eval-run-1-best-3-4-predictions.csv |   31 -
 ...custom-eval-run-1-best-3-5-predictions.csv |   61 -
 ...custom-eval-run-1-best-3-6-predictions.csv |   31 -
 ...custom-eval-run-1-best-3-7-predictions.csv |   31 -
 ...custom-eval-run-2-best-3-0-predictions.csv |   31 -
 ...custom-eval-run-2-best-3-1-predictions.csv |   31 -
 ...custom-eval-run-2-best-3-2-predictions.csv |   31 -
 ...custom-eval-run-2-best-3-3-predictions.csv |   61 -
 ...custom-eval-run-2-best-3-4-predictions.csv |   31 -
 ...custom-eval-run-2-best-3-5-predictions.csv |   61 -
 ...custom-eval-run-2-best-3-6-predictions.csv |   31 -
 ...custom-eval-run-2-best-3-7-predictions.csv |   31 -
 ...custom-eval-run-3-best-3-0-predictions.csv |   31 -
 ...custom-eval-run-3-best-3-1-predictions.csv |   31 -
 ...custom-eval-run-3-best-3-2-predictions.csv |   31 -
 ...custom-eval-run-3-best-3-3-predictions.csv |   61 -
 ...custom-eval-run-3-best-3-4-predictions.csv |   31 -
 ...custom-eval-run-3-best-3-5-predictions.csv |   61 -
 ...custom-eval-run-3-best-3-6-predictions.csv |   31 -
 ...custom-eval-run-3-best-3-7-predictions.csv |   31 -
 ...l-categorical-1-best-3-0-probabilities.csv |    2 -
 ...l-categorical-1-best-3-1-probabilities.csv |    2 -
 ...l-categorical-1-best-3-2-probabilities.csv |    2 -
 ...l-categorical-1-best-3-3-probabilities.csv |    3 -
 ...l-categorical-1-best-3-4-probabilities.csv |    2 -
 ...l-categorical-1-best-3-5-probabilities.csv |    3 -
 ...l-categorical-1-best-3-6-probabilities.csv |    2 -
 ...l-categorical-1-best-3-7-probabilities.csv |    2 -
 ...l-categorical-3-best-3-0-probabilities.csv |    2 -
 ...l-categorical-3-best-3-1-probabilities.csv |    2 -
 ...l-categorical-3-best-3-2-probabilities.csv |    2 -
 ...l-categorical-3-best-3-3-probabilities.csv |    3 -
 ...l-categorical-3-best-3-4-probabilities.csv |    2 -
 ...l-categorical-3-best-3-5-probabilities.csv |    3 -
 ...l-categorical-3-best-3-6-probabilities.csv |    2 -
 ...l-categorical-3-best-3-7-probabilities.csv |    2 -
 ...ical-3-inf-size-best-3-0-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-1-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-2-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-3-probabilities.csv |    7 -
 ...ical-3-inf-size-best-3-4-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-5-probabilities.csv |    7 -
 ...ical-3-inf-size-best-3-6-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-7-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-0-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-1-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-2-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-3-probabilities.csv |    7 -
 ...ical-3-inf-size-best-3-4-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-5-probabilities.csv |    7 -
 ...ical-3-inf-size-best-3-6-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-7-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-0-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-1-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-2-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-3-probabilities.csv |    7 -
 ...ical-3-inf-size-best-3-4-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-5-probabilities.csv |    7 -
 ...ical-3-inf-size-best-3-6-probabilities.csv |    4 -
 ...ical-3-inf-size-best-3-7-probabilities.csv |    4 -
 ...l-categorical-5-best-3-0-probabilities.csv |    2 -
 ...l-categorical-5-best-3-1-probabilities.csv |    2 -
 ...l-categorical-5-best-3-2-probabilities.csv |    2 -
 ...l-categorical-5-best-3-3-probabilities.csv |    3 -
 ...l-categorical-5-best-3-4-probabilities.csv |    2 -
 ...l-categorical-5-best-3-5-probabilities.csv |    3 -
 ...l-categorical-5-best-3-6-probabilities.csv |    2 -
 ...l-categorical-5-best-3-7-probabilities.csv |    2 -
 ...-categorical-50-best-3-0-probabilities.csv |    2 -
 ...-categorical-50-best-3-1-probabilities.csv |    2 -
 ...-categorical-50-best-3-2-probabilities.csv |    2 -
 ...-categorical-50-best-3-3-probabilities.csv |    3 -
 ...-categorical-50-best-3-4-probabilities.csv |    2 -
 ...-categorical-50-best-3-5-probabilities.csv |    3 -
 ...-categorical-50-best-3-6-probabilities.csv |    2 -
 ...-categorical-50-best-3-7-probabilities.csv |    2 -
 ...ategorical-bert-best-1-0-probabilities.csv |    6 -
 ...ategorical-bert-best-1-1-probabilities.csv |    6 -
 ...ategorical-bert-best-1-2-probabilities.csv |    5 -
 ...ategorical-bert-best-1-3-probabilities.csv |    9 -
 ...ategorical-bert-best-1-4-probabilities.csv |    5 -
 ...ategorical-bert-best-1-5-probabilities.csv |   10 -
 ...ategorical-bert-best-1-6-probabilities.csv |    6 -
 ...ategorical-bert-best-1-7-probabilities.csv |    5 -
 ...l-multitarget-5-best-3-0-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-1-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-2-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-3-probabilities.csv |    3 -
 ...l-multitarget-5-best-3-4-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-5-probabilities.csv |    3 -
 ...l-multitarget-5-best-3-6-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-7-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-0-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-1-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-2-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-3-probabilities.csv |    3 -
 ...l-multitarget-5-best-3-4-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-5-probabilities.csv |    3 -
 ...l-multitarget-5-best-3-6-probabilities.csv |    2 -
 ...l-multitarget-5-best-3-7-probabilities.csv |    2 -
 tests/unit/data/empty.parquet                 |    0
 tests/unit/io/test_batch.py                   |   32 -
 ...uifier_dataset_from_folder_parquet_lazy.py |  372 ----
 ...t_sequifier_dataset_from_folder_pt_lazy.py |  421 -----
 tests/unit/test_helpers.py                    |  617 -------
 .../unit/test_hyperparameter_search_config.py |  178 --
 tests/unit/test_infer.py                      |  833 ---------
 tests/unit/test_preprocess.py                 |  787 ---------
 tests/unit/test_probabilities.py              |   51 -
 tests/unit/test_train.py                      | 1529 -----------------
 tests/unit/test_train_distributed_loss.py     |  763 --------
 tests/unit/test_train_fsdp_loss.py            |  754 --------
 336 files changed, 17014 deletions(-)
 delete mode 100644 tests/configs/hyperparameter-search-bayesian.yaml
 delete mode 100644 tests/configs/hyperparameter-search-bert.yaml
 delete mode 100644 tests/configs/hyperparameter-search-custom-eval-inference.yaml
 delete mode 100644 tests/configs/hyperparameter-search-custom-eval.yaml
 delete mode 100644 tests/configs/hyperparameter-search-grid.yaml
 delete mode 100644 tests/configs/hyperparameter-search-sample.yaml
 delete mode 100644 tests/configs/infer-test-categorical-autoregression.yaml
 delete mode 100644 tests/configs/infer-test-categorical-bert-embedding.yaml
 delete mode 100644 tests/configs/infer-test-categorical-bert.yaml
 delete mode 100644 tests/configs/infer-test-categorical-embedding.yaml
 delete mode 100644 tests/configs/infer-test-categorical-inf-size-1.yaml
 delete mode 100644 tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
 delete mode 100644 tests/configs/infer-test-categorical-inf-size-3.yaml
 delete mode 100644 tests/configs/infer-test-categorical-multitarget.yaml
 delete mode 100644 tests/configs/infer-test-categorical.yaml
 delete mode 100644 tests/configs/infer-test-distributed-parquet.yaml
 delete mode 100644 tests/configs/infer-test-distributed.yaml
 delete mode 100644 tests/configs/infer-test-lazy.yaml
 delete mode 100644 tests/configs/infer-test-real-autoregression.yaml
 delete mode 100644 tests/configs/infer-test-real.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-exact-pt.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-exact.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-interrupted.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-lookahead-0.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-multitarget.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical-precomputed-stats.yaml
 delete mode 100644 tests/configs/preprocess-test-categorical.yaml
 delete mode 100644 tests/configs/preprocess-test-multi-file.yaml
 delete mode 100644 tests/configs/preprocess-test-real.yaml
 delete mode 100644 tests/configs/train-test-categorical-bert.yaml
 delete mode 100644 tests/configs/train-test-categorical-inf-size-1.yaml
 delete mode 100644 tests/configs/train-test-categorical-inf-size-3.yaml
 delete mode 100644 tests/configs/train-test-categorical-multitarget-eager.yaml
 delete mode 100644 tests/configs/train-test-categorical-multitarget.yaml
 delete mode 100644 tests/configs/train-test-categorical.yaml
 delete mode 100644 tests/configs/train-test-distributed-lazy-parquet.yaml
 delete mode 100644 tests/configs/train-test-distributed.yaml
 delete mode 100644 tests/configs/train-test-lazy.yaml
 delete mode 100644 tests/configs/train-test-real-bert.yaml
 delete mode 100644 tests/configs/train-test-real.yaml
 delete mode 100644 tests/configs/train-test-resume-epoch.yaml
 delete mode 100644 tests/configs/train-test-resume-mid-epoch.yaml
 delete mode 100644 tests/integration-test-log.txt
 delete mode 100644 tests/integration/conftest.py
 delete mode 100644 tests/integration/test_hyperparameter_search.py
 delete mode 100644 tests/integration/test_identities.py
 delete mode 100644 tests/integration/test_inference.py
 delete mode 100644 tests/integration/test_make.py
 delete mode 100644 tests/integration/test_onnx_export.py
 delete mode 100644 tests/integration/test_preprocessing.py
 delete mode 100644 tests/integration/test_training.py
 delete mode 100644 tests/integration/test_visualize_training.py
 delete mode 100644 tests/resources/source_configs/id_maps/itemId.json
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-0-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-1-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-2-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-3-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-0-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-1-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-2-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-3-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-0-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-1-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-2-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-3-0.pt
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-interrupted.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-1-lookahead-0.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-1.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-3-equal.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-3-equal.parquet
 delete mode 100644 tests/resources/source_data/test-data-categorical-3.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-5.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-50.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-1.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-2.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-3.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-multitarget-5.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-precomputed-stats-negative.csv
 delete mode 100644 tests/resources/source_data/test-data-categorical-precomputed-stats.csv
 delete mode 100644 tests/resources/source_data/test-data-real-1-split1-autoregression.csv
 delete mode 100644 tests/resources/source_data/test-data-real-1.csv
 delete mode 100644 tests/resources/source_data/test-data-real-3.csv
 delete mode 100644 tests/resources/source_data/test-data-real-5.csv
 delete mode 100644 tests/resources/source_data/test-data-real-50.csv
 delete mode 100644 tests/resources/source_scripts/hp_search_eval_script.py
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv
 delete mode 100644 tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
 delete mode 100644 tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
 delete mode 100644 tests/unit/data/empty.parquet
 delete mode 100644 tests/unit/io/test_batch.py
 delete mode 100644 tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
 delete mode 100644 tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
 delete mode 100644 tests/unit/test_helpers.py
 delete mode 100644 tests/unit/test_hyperparameter_search_config.py
 delete mode 100644 tests/unit/test_infer.py
 delete mode 100644 tests/unit/test_preprocess.py
 delete mode 100644 tests/unit/test_probabilities.py
 delete mode 100644 tests/unit/test_train.py
 delete mode 100644 tests/unit/test_train_distributed_loss.py
 delete mode 100644 tests/unit/test_train_fsdp_loss.py

diff --git a/tests/configs/hyperparameter-search-bayesian.yaml b/tests/configs/hyperparameter-search-bayesian.yaml
deleted file mode 100644
index c5f8d733..00000000
--- a/tests/configs/hyperparameter-search-bayesian.yaml
+++ /dev/null
@@ -1,68 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
-hp_search_name: test-hp-search-bayesian
-model_config_write_path: configs
-
-read_format: pt
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-context_length: [8]
-inference_batch_size: 10
-
-# Search Strategy
-search_strategy: bayesian
-n_samples: 4
-
-# Configuration Loading Overrides (set to null to use values from metadata)
-input_columns: null
-# Export Settings
-export_embedding_model: false
-export_generative_model: true
-export_onnx: false
-export_pt: true
-export_with_dropout: false
-
-# Model Hyperparameter Search Space
-model_hyperparameter_sampling:
-  initial_embedding_dim: [40, 80]
-  joint_embedding_dim: [null, null]
-  feature_embedding_dims: null
-  dim_model: [40, 80]
-  n_head: [2, 4]
-  dim_feedforward: [10, 12]
-  num_layers: [2]
-  activation_fn: ["swiglu"]
-  normalization: ["rmsnorm"]
-  positional_encoding: ["rope"]
-  attention_type: ["mqa", "gqa"]
-  norm_first: [ true]
-  n_kv_heads: [1]
-  rope_theta: [10000.0]
-  prediction_length: 1
-
-# Training Hyperparameter Search Space
-training_hyperparameter_sampling:
-  training_objective: ["causal"]
-  device: cpu
-  epochs: [1, 1]
-  save_interval_epochs: 10
-  batch_size: [5, 10]
-  learning_rate: [0.001, 0.01]
-  criterion:
-    itemId: CrossEntropyLoss
-  accumulation_steps: [1]
-  dropout: [0.0]
-  optimizer:
-    - name: Adam
-  scheduler:
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-  log_interval: 5
-  continue_training: false
-
-override_input: true
diff --git a/tests/configs/hyperparameter-search-bert.yaml b/tests/configs/hyperparameter-search-bert.yaml
deleted file mode 100644
index 99a4024e..00000000
--- a/tests/configs/hyperparameter-search-bert.yaml
+++ /dev/null
@@ -1,76 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1-lookahead-0.json
-hp_search_name: test-hp-search-bert
-model_config_write_path: configs
-
-read_format: pt
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-context_length: [8]
-inference_batch_size: 10
-
-search_strategy: sample
-n_samples: 1
-prune_trials: false
-
-input_columns:
-  - [itemId]
-export_embedding_model: false
-export_generative_model: true
-export_onnx: false
-export_pt: true
-export_with_dropout: false
-
-model_hyperparameter_sampling:
-  initial_embedding_dim: [16]
-  joint_embedding_dim: [null]
-  feature_embedding_dims: null
-  dim_model: [16]
-  n_head: [2]
-  dim_feedforward: [8]
-  num_layers: [1]
-  activation_fn: ["relu"]
-  normalization: ["layer_norm"]
-  positional_encoding: ["learned"]
-  attention_type: ["mha"]
-  norm_first: [false]
-  n_kv_heads: [2]
-  rope_theta: [10000.0]
-  prediction_length: 8
-
-training_hyperparameter_sampling:
-  training_objective: ["bert"]
-  bert_spec:
-    masking_probability: [0.5]
-    replacement_distribution:
-      - masked: 0.8
-        random: 0.1
-        identical: 0.1
-    span_masking:
-      - type: GeometricDistribution
-        p: 1.0
-  device: cpu
-  torch_compile: none
-  epochs: [1]
-  save_interval_epochs: 10
-  batch_size: [5]
-  learning_rate: [0.001]
-  criterion:
-    itemId: CrossEntropyLoss
-  loss_weights:
-    itemId: 1.0
-  accumulation_steps: [1]
-  dropout: [0.0]
-  optimizer:
-    - name: Adam
-  scheduler:
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-  log_interval: 5
-  calculate_validation_loss_on_initialization: false
-  continue_training: false
-  layer_autocast: false
-
-override_input: true
diff --git a/tests/configs/hyperparameter-search-custom-eval-inference.yaml b/tests/configs/hyperparameter-search-custom-eval-inference.yaml
deleted file mode 100644
index aa40c11c..00000000
--- a/tests/configs/hyperparameter-search-custom-eval-inference.yaml
+++ /dev/null
@@ -1,41 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: configs/metadata_configs/test-data-categorical-5.json
-
-model_type: generative
-training_objective: causal
-model_path: models/sequifier-test-hp-search-custom-eval-run-0-best-1.pt
-
-data_path: data/test-data-categorical-5-split2
-read_format: pt
-write_format: csv
-
-input_columns:
-- itemId
-- supCat1
-- supCat2
-- supCat3
-- supCat4
-target_columns:
-- itemId
-- supCat1
-- supCat2
-- supCat3
-- supCat4
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supCat2: categorical
-  supCat3: categorical
-  supCat4: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-prediction_length: 1
-inference_batch_size: 10
-enforce_determinism: true
-
-# Autoregression
-autoregression: true
-autoregression_total_steps: 30
diff --git a/tests/configs/hyperparameter-search-custom-eval.yaml b/tests/configs/hyperparameter-search-custom-eval.yaml
deleted file mode 100644
index d93d0792..00000000
--- a/tests/configs/hyperparameter-search-custom-eval.yaml
+++ /dev/null
@@ -1,92 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
-hp_search_name: test-hp-search-custom-eval
-model_config_write_path: configs
-
-read_format: pt
-target_columns:
-- itemId
-- supCat1
-- supCat2
-- supCat3
-- supCat4
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supCat2: categorical
-  supCat3: categorical
-  supCat4: categorical
-context_length: [8]
-inference_batch_size: 10
-
-# Search Strategy
-search_strategy: bayesian
-seed: 101
-n_samples: 4
-
-# Configuration Loading Overrides (set to null to use values from metadata)
-input_columns: null
-# Export Settings
-export_embedding_model: false
-export_generative_model: true
-export_onnx: true
-export_pt: false
-export_with_dropout: false
-
-# Model Hyperparameter Search Space
-model_hyperparameter_sampling:
-  initial_embedding_dim: [40, 80]
-  joint_embedding_dim: [null, null]
-  feature_embedding_dims: null
-  dim_model: [40, 80]
-  n_head: [2, 4]
-  dim_feedforward: [10, 12]
-  num_layers: [2]
-  activation_fn: ["swiglu"]
-  normalization: ["rmsnorm"]
-  positional_encoding: ["rope"]
-  attention_type: ["mqa", "gqa"]
-  norm_first: [ true]
-  n_kv_heads: [1]
-  rope_theta: [10000.0]
-  prediction_length: 1
-
-# Training Hyperparameter Search Space
-training_hyperparameter_sampling:
-  training_objective: ["causal"]
-  device: cpu
-  epochs: [3, 3]
-  save_interval_epochs: 10
-  batch_size: [5, 10]
-  learning_rate: [0.001, 0.0001]
-  criterion:
-    itemId: CrossEntropyLoss
-    supCat1: CrossEntropyLoss
-    supCat2: CrossEntropyLoss
-    supCat3: CrossEntropyLoss
-    supCat4: CrossEntropyLoss
-  accumulation_steps: [1]
-  dropout: [0.5]
-  optimizer:
-    - name: Adam
-  scheduler:
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-  log_interval: 5
-  continue_training: false
-
-
-evaluation_metric_directions:
- - minimize
- - maximize
-evaluation_metrics:
- - max
- - stdev
-evaluation_inference_config: tests/configs/hyperparameter-search-custom-eval-inference.yaml
-evaluation_script: scripts/hp_search_eval_script.py
-
-override_input: true
diff --git a/tests/configs/hyperparameter-search-grid.yaml b/tests/configs/hyperparameter-search-grid.yaml
deleted file mode 100644
index 2922a641..00000000
--- a/tests/configs/hyperparameter-search-grid.yaml
+++ /dev/null
@@ -1,70 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
-hp_search_name: test-hp-search-grid
-model_config_write_path: configs
-
-read_format: pt
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-context_length: [8]
-inference_batch_size: 10
-
-# Search Strategy
-search_strategy: grid
-n_samples: 4
-
-# Configuration Loading Overrides (set to null to use values from metadata)
-input_columns: null
-# Export Settings
-export_embedding_model: false
-export_generative_model: true
-export_onnx: false
-export_pt: true
-export_with_dropout: false
-
-# Model Hyperparameter Search Space
-model_hyperparameter_sampling:
-  initial_embedding_dim: [20]
-  feature_embedding_dims: null
-  joint_embedding_dim: [null]
-  dim_model: [20]
-  n_head: [2]
-  dim_feedforward: [10]
-  num_layers: [2]
-  activation_fn: ["relu", "swiglu"]
-  normalization: ["layer_norm"]
-  positional_encoding: ["learned", "rope"]
-  attention_type: ["mqa"]
-  norm_first: [false]
-  n_kv_heads: [1]
-  rope_theta: [10000.0]
-  prediction_length: 1
-
-# Training Hyperparameter Search Space
-training_hyperparameter_sampling:
-  training_objective: ["causal"]
-  device: cpu
-  epochs: [2]
-  save_interval_epochs: 10
-  batch_size: [5]
-  learning_rate: [0.001]
-  criterion:
-    itemId: CrossEntropyLoss
-  accumulation_steps: [1]
-  dropout: [0.0]
-  optimizer:
-    - name: Adam
-  scheduler:
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-  log_interval: 5
-  continue_training: false
-  layer_type_dtypes:
-    embedding: bfloat16
-    linear: bfloat16
-    norm: bfloat16
-
-
-override_input: true
diff --git a/tests/configs/hyperparameter-search-sample.yaml b/tests/configs/hyperparameter-search-sample.yaml
deleted file mode 100644
index 5d429e8b..00000000
--- a/tests/configs/hyperparameter-search-sample.yaml
+++ /dev/null
@@ -1,86 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
-hp_search_name: test-hp-search-sample
-model_config_write_path: configs
-
-read_format: pt
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-context_length: [8]
-inference_batch_size: 10
-
-# Search Strategy
-search_strategy: sample
-n_samples: 4
-
-# Configuration Loading Overrides
-input_columns: null
-export_embedding_model: false
-export_generative_model: true
-export_onnx: false
-export_pt: true
-export_with_dropout: false
-
-# Model Hyperparameter Search Space using Distributions
-model_hyperparameter_sampling:
-  initial_embedding_dim: [40]
-  joint_embedding_dim: [null]
-  feature_embedding_dims: null
-  dim_model: [40]
-  n_head: [2]
-  # IntDistribution test with step
-  dim_feedforward:
-    low: 10
-    high: 20
-    step: 2
-  # IntDistribution test without step
-  num_layers:
-    low: 1
-    high: 3
-  activation_fn: ["swiglu"]
-  normalization: ["rmsnorm"]
-  positional_encoding: ["rope"]
-  attention_type: ["mqa"]
-  norm_first: [true]
-  n_kv_heads: [1]
-  # FloatDistribution test with log
-  rope_theta:
-    low: 1000.0
-    high: 10000.0
-    log: true
-  prediction_length: 1
-
-# Training Hyperparameter Search Space using Distributions
-training_hyperparameter_sampling:
-  training_objective: ["causal"]
-  device: cpu
-  epochs: [1]
-  save_interval_epochs: 10
-  # IntDistribution test
-  batch_size:
-    low: 5
-    high: 15
-    step: 5
-  learning_rate: [0.001]
-  criterion:
-    itemId: CrossEntropyLoss
-  # IntDistribution test
-  accumulation_steps:
-    low: 1
-    high: 2
-  # FloatDistribution test without log
-  dropout:
-    low: 0.1
-    high: 0.5
-    log: true
-  optimizer:
-    - name: Adam
-  scheduler:
-    - name: StepLR
-      step_size: 1
-      gamma: 0.99
-  log_interval: 5
-  continue_training: false
-
-override_input: true
diff --git a/tests/configs/infer-test-categorical-autoregression.yaml b/tests/configs/infer-test-categorical-autoregression.yaml
deleted file mode 100644
index 7112cd9e..00000000
--- a/tests/configs/infer-test-categorical-autoregression.yaml
+++ /dev/null
@@ -1,25 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-1-best-3-autoregression.onnx
-data_path: tests/project_folder/data/test-data-categorical-1-split2
-read_format: pt
-write_format: csv
-
-input_columns: itemId
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: true
-autoregression_total_steps: 20
diff --git a/tests/configs/infer-test-categorical-bert-embedding.yaml b/tests/configs/infer-test-categorical-bert-embedding.yaml
deleted file mode 100644
index 61724d81..00000000
--- a/tests/configs/infer-test-categorical-bert-embedding.yaml
+++ /dev/null
@@ -1,25 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical-bert.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1-lookahead-0.json
-model_type: embedding
-training_objective: bert
-model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-embedding-1.pt
-data_path: tests/project_folder/data/test-data-categorical-1-lookahead-0-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-prediction_length: null
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-categorical-bert.yaml b/tests/configs/infer-test-categorical-bert.yaml
deleted file mode 100644
index 5a459841..00000000
--- a/tests/configs/infer-test-categorical-bert.yaml
+++ /dev/null
@@ -1,25 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical-bert.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1-lookahead-0.json
-model_type: generative
-training_objective: bert
-model_path: tests/project_folder/models/sequifier-model-categorical-bert-best-1.pt
-data_path: tests/project_folder/data/test-data-categorical-1-lookahead-0-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: true
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-prediction_length: null
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-categorical-embedding.yaml b/tests/configs/infer-test-categorical-embedding.yaml
deleted file mode 100644
index e8ce1543..00000000
--- a/tests/configs/infer-test-categorical-embedding.yaml
+++ /dev/null
@@ -1,24 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
-model_type: embedding
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-1-best-embedding-3.pt
-data_path: tests/project_folder/data/test-data-categorical-1-split2
-read_format: pt
-write_format: csv
-
-input_columns: null
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-categorical-inf-size-1.yaml b/tests/configs/infer-test-categorical-inf-size-1.yaml
deleted file mode 100644
index 8c8c14b9..00000000
--- a/tests/configs/infer-test-categorical-inf-size-1.yaml
+++ /dev/null
@@ -1,24 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-1.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-1-inf-size-best-3.onnx
-data_path: tests/project_folder/data/test-data-categorical-1-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-prediction_length: 3
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml b/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
deleted file mode 100644
index 61732fa0..00000000
--- a/tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
+++ /dev/null
@@ -1,27 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical-inf-size-3.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-3.json
-model_type: embedding
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-3-inf-size-best-embedding-3.pt
-data_path: tests/project_folder/data/test-data-categorical-3-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId, supCat1, supCat2]
-target_columns: [itemId, supCat1, supCat2]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supCat2: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-prediction_length: 3
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-categorical-inf-size-3.yaml b/tests/configs/infer-test-categorical-inf-size-3.yaml
deleted file mode 100644
index 6bdad176..00000000
--- a/tests/configs/infer-test-categorical-inf-size-3.yaml
+++ /dev/null
@@ -1,27 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical-inf-size-3.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-3.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-3-inf-size-best-3.pt
-data_path: tests/project_folder/data/test-data-categorical-3-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId, supCat1, supCat2]
-target_columns: [itemId, supCat1, supCat2]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supCat2: categorical
-
-output_probabilities: true
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-prediction_length: 3
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-categorical-multitarget.yaml b/tests/configs/infer-test-categorical-multitarget.yaml
deleted file mode 100644
index 9202b7ed..00000000
--- a/tests/configs/infer-test-categorical-multitarget.yaml
+++ /dev/null
@@ -1,26 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical-multitarget.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-multitarget-5.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-multitarget-5-best-3.onnx
-data_path: tests/project_folder/data/test-data-categorical-multitarget-5-split2
-read_format: parquet
-write_format: csv
-
-input_columns: null
-target_columns: [itemId, supCat1, supReal3]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supReal3: real
-
-output_probabilities: true
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: true
-
-sample_from_distribution_columns: ['itemId']
-autoregression: false
diff --git a/tests/configs/infer-test-categorical.yaml b/tests/configs/infer-test-categorical.yaml
deleted file mode 100644
index dafe686e..00000000
--- a/tests/configs/infer-test-categorical.yaml
+++ /dev/null
@@ -1,24 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-categorical.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-5.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-5-best-3.pt
-data_path: tests/project_folder/data/test-data-categorical5-split2
-read_format: pt
-write_format: csv
-
-input_columns: null
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: true
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-distributed-parquet.yaml b/tests/configs/infer-test-distributed-parquet.yaml
deleted file mode 100644
index 52ce0a4e..00000000
--- a/tests/configs/infer-test-distributed-parquet.yaml
+++ /dev/null
@@ -1,25 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-categorical-multitarget-5.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-categorical-multitarget-5-last-3.onnx
-data_path: tests/project_folder/data/test-data-categorical-multitarget-5-split2
-read_format: parquet
-write_format: csv
-
-input_columns: null
-target_columns: [itemId, supCat1, supReal3]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supReal3: real
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: true
-
-sample_from_distribution_columns: null
-autoregression: false
diff --git a/tests/configs/infer-test-distributed.yaml b/tests/configs/infer-test-distributed.yaml
deleted file mode 100644
index 61f43431..00000000
--- a/tests/configs/infer-test-distributed.yaml
+++ /dev/null
@@ -1,23 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
-
-# Point to the model generated by the distributed training test
-model_path: models/sequifier-model-categorical-distributed-best-3.onnx
-model_type: generative
-training_objective: causal
-
-# Use the split2 data generated by preprocessing
-data_path: data/test-data-categorical-3-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId, supCat1]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 2
diff --git a/tests/configs/infer-test-lazy.yaml b/tests/configs/infer-test-lazy.yaml
deleted file mode 100644
index 3429ecf5..00000000
--- a/tests/configs/infer-test-lazy.yaml
+++ /dev/null
@@ -1,21 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
-
-model_path: models/sequifier-model-categorical-lazy-best-3.onnx
-model_type: generative
-training_objective: causal
-
-data_path: data/test-data-categorical-3-split2
-read_format: pt
-write_format: csv
-
-input_columns: [itemId, supCat1]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-output_probabilities: false
-map_to_id: true
-device: cpu
-context_length: 8
-inference_batch_size: 2
diff --git a/tests/configs/infer-test-real-autoregression.yaml b/tests/configs/infer-test-real-autoregression.yaml
deleted file mode 100644
index 07d3b21f..00000000
--- a/tests/configs/infer-test-real-autoregression.yaml
+++ /dev/null
@@ -1,26 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-real.yaml
-metadata_config_path: tests/project_folder/configs/metadata_configs/test-data-real-1.json
-model_type: generative
-training_objective: causal
-model_path: tests/project_folder/models/sequifier-model-real-1-best-3-autoregression.pt
-data_path: tests/project_folder/data/test-data-real-1-split1-autoregression.csv
-read_format: csv
-write_format: csv
-
-input_columns: null
-target_columns: [itemValue]
-target_column_types:
- itemValue: real
-
-output_probabilities: false
-map_to_id: false
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: false
-
-sample_from_distribution_columns: [itemValue]
-infer_with_dropout: true
-autoregression: true
-autoregression_total_steps: 20
diff --git a/tests/configs/infer-test-real.yaml b/tests/configs/infer-test-real.yaml
deleted file mode 100644
index 74b40e23..00000000
--- a/tests/configs/infer-test-real.yaml
+++ /dev/null
@@ -1,21 +0,0 @@
-project_root: tests/project_folder
-training_config_path: tests/configs/train-test-real.yaml
-metadata_config_path: configs/metadata_configs/test-data-real-5.json
-model_type: generative
-training_objective: causal
-model_path: models/sequifier-model-real-5-best-3.pt
-data_path: data/test-data-real5-split2.csv
-
-input_columns: null
-target_columns: [itemValue]
-target_column_types:
-  itemValue: real
-
-output_probabilities: false
-map_to_id: false
-device: cpu
-context_length: 8
-inference_batch_size: 10
-enforce_determinism: true
-
-autoregression: false
diff --git a/tests/configs/preprocess-test-categorical-exact-pt.yaml b/tests/configs/preprocess-test-categorical-exact-pt.yaml
deleted file mode 100644
index 2e233b9b..00000000
--- a/tests/configs/preprocess-test-categorical-exact-pt.yaml
+++ /dev/null
@@ -1,15 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-3-equal.parquet
-merge_output: false
-read_format: parquet
-write_format: pt
-selected_columns: null
-
-split_ratios:
-- 1.0
-stored_context_width: 6
-max_target_offset: 1
-stride_by_split:
-- 5
-max_rows: null
-subsequence_start_mode: exact
diff --git a/tests/configs/preprocess-test-categorical-exact.yaml b/tests/configs/preprocess-test-categorical-exact.yaml
deleted file mode 100644
index 3e58bef4..00000000
--- a/tests/configs/preprocess-test-categorical-exact.yaml
+++ /dev/null
@@ -1,15 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-3-equal.csv
-merge_output: true
-read_format: csv
-write_format: parquet
-selected_columns: null
-
-split_ratios:
-- 1.0
-stored_context_width: 6
-max_target_offset: 1
-stride_by_split:
-- 5
-max_rows: null
-subsequence_start_mode: exact
diff --git a/tests/configs/preprocess-test-categorical-interrupted.yaml b/tests/configs/preprocess-test-categorical-interrupted.yaml
deleted file mode 100644
index 7dae9dcf..00000000
--- a/tests/configs/preprocess-test-categorical-interrupted.yaml
+++ /dev/null
@@ -1,19 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-1-interrupted.csv
-merge_output: false
-read_format: csv
-write_format: pt
-selected_columns: null
-continue_preprocessing: true
-split_ratios:
-- 0.6
-- 0.2
-- 0.2
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-- 1
-- 1
-max_rows: null
-subsequence_start_mode: distribute
diff --git a/tests/configs/preprocess-test-categorical-lookahead-0.yaml b/tests/configs/preprocess-test-categorical-lookahead-0.yaml
deleted file mode 100644
index c9653bea..00000000
--- a/tests/configs/preprocess-test-categorical-lookahead-0.yaml
+++ /dev/null
@@ -1,19 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-1.csv
-merge_output: false
-read_format: csv
-write_format: pt
-selected_columns: null
-
-split_ratios:
-- 0.6
-- 0.2
-- 0.2
-stored_context_width: 8
-max_target_offset: 0
-stride_by_split:
-- 1
-- 1
-- 1
-max_rows: null
-subsequence_start_mode: distribute
diff --git a/tests/configs/preprocess-test-categorical-multitarget.yaml b/tests/configs/preprocess-test-categorical-multitarget.yaml
deleted file mode 100644
index 0dca4827..00000000
--- a/tests/configs/preprocess-test-categorical-multitarget.yaml
+++ /dev/null
@@ -1,20 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-multitarget-5.csv
-merge_output: false
-read_format: csv
-write_format: parquet
-selected_columns: null
-
-split_ratios:
-- 0.6
-- 0.2
-- 0.2
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-- 1
-- 1
-max_rows: null
-use_precomputed_maps:
-- "itemId"
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
deleted file mode 100644
index 7372bc0e..00000000
--- a/tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
+++ /dev/null
@@ -1,15 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-precomputed-stats-negative.csv
-read_format: csv
-
-write_format: parquet
-merge_output: true
-selected_columns: null
-
-split_ratios:
-- 1.0
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-max_rows: null
diff --git a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml b/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
deleted file mode 100644
index 1f565c32..00000000
--- a/tests/configs/preprocess-test-categorical-precomputed-stats.yaml
+++ /dev/null
@@ -1,16 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-precomputed-stats.csv
-read_format: csv
-write_format: parquet
-merge_output: true
-selected_columns: null
-
-split_ratios:
-- 1.0
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-max_rows: null
-metadata_config_path: configs/metadata_configs/test-data-categorical-multitarget-5.json
-mask_column: "[mask]"
diff --git a/tests/configs/preprocess-test-categorical.yaml b/tests/configs/preprocess-test-categorical.yaml
deleted file mode 100644
index 4767f7e6..00000000
--- a/tests/configs/preprocess-test-categorical.yaml
+++ /dev/null
@@ -1,19 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-5.csv
-merge_output: false
-read_format: csv
-write_format: pt
-selected_columns: null
-
-split_ratios:
-- 0.6
-- 0.2
-- 0.2
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-- 1
-- 1
-max_rows: null
-subsequence_start_mode: distribute
diff --git a/tests/configs/preprocess-test-multi-file.yaml b/tests/configs/preprocess-test-multi-file.yaml
deleted file mode 100644
index cc76ee4c..00000000
--- a/tests/configs/preprocess-test-multi-file.yaml
+++ /dev/null
@@ -1,21 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-categorical-multi-file
-merge_output: false
-read_format: csv
-process_by_file: false
-write_format: pt
-selected_columns: null
-
-batches_per_file: 1
-
-split_ratios:
-- 0.8
-- 0.1
-- 0.1
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-- 1
-- 1
-max_rows: null
diff --git a/tests/configs/preprocess-test-real.yaml b/tests/configs/preprocess-test-real.yaml
deleted file mode 100644
index 538ba285..00000000
--- a/tests/configs/preprocess-test-real.yaml
+++ /dev/null
@@ -1,17 +0,0 @@
-project_root: tests/project_folder
-data_path: tests/resources/source_data/test-data-real-5.csv
-merge_output: true
-read_format: csv
-write_format: parquet
-selected_columns: null
-
-split_ratios:
-- 0.5
-- 0.5
-stored_context_width: 9
-max_target_offset: 1
-stride_by_split:
-- 1
-- 1
-max_rows: null
-n_cores: 1
diff --git a/tests/configs/train-test-categorical-bert.yaml b/tests/configs/train-test-categorical-bert.yaml
deleted file mode 100644
index c6b4187c..00000000
--- a/tests/configs/train-test-categorical-bert.yaml
+++ /dev/null
@@ -1,68 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-bert
-read_format: pt
-metadata_config_path: configs/metadata_configs/test-data-categorical-1-lookahead-0.json
-
-input_columns: [itemId]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-export_onnx: false
-export_pt: true
-export_with_dropout: false
-
-model_spec:
-  initial_embedding_dim: 16
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 8
-  num_layers: 1
-  prediction_length: 8
-  activation_fn: relu
-  normalization: layer_norm
-  positional_encoding: learned
-  attention_type: mha
-  norm_first: false
-  n_kv_heads: 2
-  rope_theta: 10000.0
-training_spec:
-  training_objective: bert
-  bert_spec:
-    masking_probability: 0.5
-    replacement_distribution:
-      masked: 0.8
-      random: 0.1
-      identical: 0.1
-    span_masking:
-      type: GeometricDistribution
-      p: 1.0
-  device: cpu
-  torch_compile: none
-  epochs: 1
-  save_interval_epochs: 1
-  batch_size: 5
-  log_interval: 1
-  learning_rate: 0.001
-  accumulation_steps: null
-  dropout: 0.0
-  criterion:
-    itemId: CrossEntropyLoss
-  loss_weights:
-    itemId: 1.0
-  optimizer:
-    name: Adam
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  scheduler_step_on: epoch
-  calculate_validation_loss_on_initialization: false
-  continue_training: false
-  enforce_determinism: true
-  layer_autocast: false
diff --git a/tests/configs/train-test-categorical-inf-size-1.yaml b/tests/configs/train-test-categorical-inf-size-1.yaml
deleted file mode 100644
index 5c915af2..00000000
--- a/tests/configs/train-test-categorical-inf-size-1.yaml
+++ /dev/null
@@ -1,61 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-1-inf-size
-read_format: pt
-metadata_config_path: configs/metadata_configs/test-data-categorical-1.json
-
-input_columns: [itemId]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-export_onnx: true
-export_pt: false
-
-model_spec:
-  initial_embedding_dim: 200
-  dim_model: 200
-  n_head: 2
-  dim_feedforward: 12
-  num_layers: 2
-  prediction_length: 3
-  activation_fn: swiglu
-  normalization: rmsnorm
-  positional_encoding: learned
-  attention_type: mha
-  norm_first: false
-  n_kv_heads: 2
-  rope_theta: 10000.0
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  save_interval_epochs: 1
-  batch_size: 5
-  log_interval: 1
-  class_share_log_columns: ["itemId"]
-  learning_rate: 0.003
-  accumulation_steps: 2
-  dropout: 0.3
-  criterion:
-    itemId: CrossEntropyLoss
-  loss_weights:
-    itemId: 1.0
-  optimizer:
-    name: QHAdam
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  scheduler_step_on: epoch
-  continue_training: false
-  enforce_determinism: false
-  layer_type_dtypes:
-    embedding: bfloat16
-    linear: bfloat16
-    norm: bfloat16
-    decoder: bfloat16
diff --git a/tests/configs/train-test-categorical-inf-size-3.yaml b/tests/configs/train-test-categorical-inf-size-3.yaml
deleted file mode 100644
index 93165300..00000000
--- a/tests/configs/train-test-categorical-inf-size-3.yaml
+++ /dev/null
@@ -1,68 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-3-inf-size
-read_format: pt
-metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
-
-input_columns: [itemId, supCat1, supCat2]
-target_columns: [itemId, supCat1, supCat2]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supCat2: categorical
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-export_onnx: false
-export_pt: true
-
-model_spec:
-  initial_embedding_dim: 100
-  feature_embedding_dims:
-    itemId: 50
-    supCat1: 25
-    supCat2: 25
-  joint_embedding_dim: 200
-  dim_model: 200
-  n_head: 2
-  dim_feedforward: 12
-  num_layers: 2
-  prediction_length: 3
-  activation_fn: swiglu
-  normalization: layer_norm
-  positional_encoding: rope
-  attention_type: mqa
-  norm_first: false
-  n_kv_heads: 1
-  rope_theta: 10000.0
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  save_interval_epochs: 1
-  batch_size: 5
-  log_interval: 1
-  class_share_log_columns: ["itemId"]
-  learning_rate: 0.003
-  accumulation_steps: 2
-  dropout: 0.3
-  criterion:
-    itemId: CrossEntropyLoss
-    supCat1: CrossEntropyLoss
-    supCat2: CrossEntropyLoss
-  optimizer:
-    name: QHAdam
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  scheduler_step_on: epoch
-  continue_training: false
-  enforce_determinism: true
-  layer_autocast: false
-  layer_type_dtypes:
-    embedding: bfloat16
-    linear: bfloat16
-    norm: bfloat16
diff --git a/tests/configs/train-test-categorical-multitarget-eager.yaml b/tests/configs/train-test-categorical-multitarget-eager.yaml
deleted file mode 100644
index 2dfbd2c5..00000000
--- a/tests/configs/train-test-categorical-multitarget-eager.yaml
+++ /dev/null
@@ -1,78 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-multitarget-5-eager
-read_format: parquet
-metadata_config_path: configs/metadata_configs/test-data-categorical-multitarget-5.json
-
-input_columns: null
-target_columns: [itemId, supCat1, supReal3]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supReal3: real
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-
-model_spec:
-  initial_embedding_dim: 16
-  feature_embedding_dims:
-    itemId: 6
-    supCat1: 4
-    supReal2: 1
-    supReal3: 1
-    supCat4: 4
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 10
-  num_layers: 2
-  prediction_length: 1
-
-  activation_fn: gelu
-  normalization: rmsnorm
-  positional_encoding: rope
-  attention_type: gqa
-  norm_first: true
-  n_kv_heads: 1
-  rope_theta: 10000.0
-
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  save_interval_epochs: 1
-  batch_size: 10
-  learning_rate: 0.003
-  log_interval: 1
-  accumulation_steps: 2
-  dropout: 0.3
-  criterion:
-    itemId: CrossEntropyLoss
-    supCat1: CrossEntropyLoss
-    supReal3: MSELoss
-  loss_weights:
-    itemId: 1.0
-    supCat1: 0.5
-    supReal3: 0.5
-  optimizer:
-    name: AdamW
-  scheduler:
-    name: OneCycleLR
-    max_lr: 0.001
-    pct_start: 0.03
-    div_factor: 1000
-    final_div_factor: 1000
-    anneal_strategy: cos
-    total_steps: 100
-    three_phase: false
-  scheduler_step_on: 'batch'
-  continue_training: false
-  enforce_determinism: true
-  layer_autocast: true
-  layer_type_dtypes:
-    linear: bfloat16
-    decoder: bfloat16
-    embedding: float32
-    norm: float32
-  load_full_data_to_ram: true
diff --git a/tests/configs/train-test-categorical-multitarget.yaml b/tests/configs/train-test-categorical-multitarget.yaml
deleted file mode 100644
index 4f1a83c1..00000000
--- a/tests/configs/train-test-categorical-multitarget.yaml
+++ /dev/null
@@ -1,79 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-multitarget-5
-read_format: parquet
-metadata_config_path: configs/metadata_configs/test-data-categorical-multitarget-5.json
-
-
-input_columns: null
-target_columns: [itemId, supCat1, supReal3]
-target_column_types:
-  itemId: categorical
-  supCat1: categorical
-  supReal3: real
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-
-model_spec:
-  initial_embedding_dim: 16
-  feature_embedding_dims:
-    itemId: 6
-    supCat1: 4
-    supReal2: 1
-    supReal3: 1
-    supCat4: 4
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 10
-  num_layers: 2
-  prediction_length: 1
-
-  activation_fn: gelu
-  normalization: rmsnorm
-  positional_encoding: rope
-  attention_type: gqa
-  norm_first: true
-  n_kv_heads: 1
-  rope_theta: 10000.0
-
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  save_interval_epochs: 1
-  batch_size: 10
-  learning_rate: 0.003
-  log_interval: 1
-  accumulation_steps: 2
-  dropout: 0.3
-  criterion:
-    itemId: CrossEntropyLoss
-    supCat1: CrossEntropyLoss
-    supReal3: MSELoss
-  loss_weights:
-    itemId: 1.0
-    supCat1: 0.5
-    supReal3: 0.5
-  optimizer:
-    name: AdamW
-  scheduler:
-    name: OneCycleLR
-    max_lr: 0.001
-    pct_start: 0.03
-    div_factor: 1000
-    final_div_factor: 1000
-    anneal_strategy: cos
-    total_steps: 100
-    three_phase: false
-  scheduler_step_on: 'batch'
-  continue_training: false
-  enforce_determinism: true
-  layer_autocast: true
-  layer_type_dtypes:
-    linear: bfloat16
-    decoder: bfloat16
-    embedding: float32
-    norm: float32
-  load_full_data_to_ram: false
diff --git a/tests/configs/train-test-categorical.yaml b/tests/configs/train-test-categorical.yaml
deleted file mode 100644
index e10b673f..00000000
--- a/tests/configs/train-test-categorical.yaml
+++ /dev/null
@@ -1,60 +0,0 @@
-project_root: tests/project_folder
-model_name: default
-read_format: pt
-metadata_config_path: configs/metadata_configs/test-data-categorical-5.json
-
-input_columns: null
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-export_onnx: true
-export_pt: true
-
-model_spec:
-  initial_embedding_dim: 100
-  joint_embedding_dim: 200
-  dim_model: 200
-  n_head: 2
-  dim_feedforward: 12
-  num_layers: 2
-  prediction_length: 1
-  activation_fn: relu
-  normalization: rmsnorm
-  positional_encoding: rope
-  attention_type: mqa
-  norm_first: true
-  n_kv_heads: 1
-  rope_theta: 1000.0
-training_spec:
-  training_objective: causal
-  device: cpu
-  torch_compile: "outer"
-  epochs: 3
-  save_interval_epochs: 1
-  batch_size: 5
-  log_interval: 1
-  class_share_log_columns: ["itemId"]
-  learning_rate: 0.003
-  accumulation_steps: 2
-  dropout: 0.3
-  criterion:
-    itemId: CrossEntropyLoss
-  loss_weights:
-    itemId: 1.0
-  class_weights:
-    itemId: [1.0, 1.0, 1.0, 1.5, 0.5, 0.6, 1.2, 0.66666667, 1.5, 1.2, 0.75, 0.85714286, 0.75, 0.85714286, 0.75, 0.6, 2.,1.5, 1.5, 1.0, 1.0, 0.66666667, 0.75, 0.5, 0.75, 2.0, 1.2, 3.0, 1.0, 0.75, 1.0, 0.85714286, 1.0]
-  optimizer:
-    name: QHAdam
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  scheduler_step_on: epoch
-  continue_training: false
-  enforce_determinism: true
diff --git a/tests/configs/train-test-distributed-lazy-parquet.yaml b/tests/configs/train-test-distributed-lazy-parquet.yaml
deleted file mode 100644
index 1e50b40f..00000000
--- a/tests/configs/train-test-distributed-lazy-parquet.yaml
+++ /dev/null
@@ -1,52 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-distributed-lazy-parquet
-read_format: parquet
-metadata_config_path: configs/metadata_configs/test-data-categorical-multitarget-5.json
-
-input_columns: [itemId, supCat1]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 2
-
-export_generative_model: true
-export_embedding_model: false
-export_onnx: true
-export_pt: true
-
-model_spec:
-  initial_embedding_dim: 16
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 16
-  num_layers: 1
-  prediction_length: 1
-
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  log_interval: 1
-  save_interval_epochs: 1
-  batch_size: 5
-  learning_rate: 0.001
-  criterion:
-    itemId: CrossEntropyLoss
-  optimizer:
-    name: Adam
-  scheduler:
-    name: StepLR
-    step_size: 1
-    gamma: 0.1
-
-  # --- The Magic Combination ---
-  distributed: true             # Triggers mp.spawn + gloo
-  data_parallelism: 'DDP'
-  world_size: 2
-  backend: gloo
-  sampling_strategy: 'oversampling'
-  num_workers: 2                # Triggers thread contention
-  load_full_data_to_ram: false  # Triggers SequifierDatasetFromFolderParquetLazy
-  continue_training: false
diff --git a/tests/configs/train-test-distributed.yaml b/tests/configs/train-test-distributed.yaml
deleted file mode 100644
index ea0efe24..00000000
--- a/tests/configs/train-test-distributed.yaml
+++ /dev/null
@@ -1,52 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
-model_name: model-categorical-distributed
-training_data_path: data/test-data-categorical-3-split0  # Generated by existing preprocess tests
-validation_data_path: data/test-data-categorical-3-split1
-read_format: pt
-
-input_columns: [itemId, supCat1]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 2
-
-export_generative_model: true
-export_embedding_model: false
-export_onnx: true
-export_pt: true
-
-model_spec:
-  initial_embedding_dim: 16
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 16
-  num_layers: 1
-  prediction_length: 1
-
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  log_interval: 1
-  save_interval_epochs: 1
-  batch_size: 5
-  learning_rate: 0.001
-  criterion:
-    itemId: CrossEntropyLoss
-  optimizer:
-    name: Adam
-  scheduler:
-    name: StepLR
-    step_size: 1
-    gamma: 0.1
-
-  # --- Mac Distributed Settings ---
-  distributed: true
-  data_parallelism: 'DDP'
-  world_size: 2
-  backend: gloo
-  load_full_data_to_ram: true
-  continue_training: false
diff --git a/tests/configs/train-test-lazy.yaml b/tests/configs/train-test-lazy.yaml
deleted file mode 100644
index 348c1ada..00000000
--- a/tests/configs/train-test-lazy.yaml
+++ /dev/null
@@ -1,48 +0,0 @@
-project_root: tests/project_folder
-metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
-model_name: model-categorical-lazy
-training_data_path: data/test-data-categorical-3-split0
-validation_data_path: data/test-data-categorical-3-split1
-read_format: pt
-
-input_columns: [itemId, supCat1]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 2
-
-export_generative_model: true
-export_embedding_model: false
-export_onnx: true
-export_pt: true
-
-model_spec:
-  initial_embedding_dim: 16
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 16
-  num_layers: 1
-  prediction_length: 1
-
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  log_interval: 1
-  save_interval_epochs: 1
-  batch_size: 5
-  learning_rate: 0.001
-  criterion:
-    itemId: CrossEntropyLoss
-  optimizer:
-    name: Adam
-  scheduler:
-    name: StepLR
-    step_size: 1
-    gamma: 0.1
-
-  # --- Lazy Loading Setting ---
-  load_full_data_to_ram: false
-  continue_training: false
diff --git a/tests/configs/train-test-real-bert.yaml b/tests/configs/train-test-real-bert.yaml
deleted file mode 100644
index 89ec8ad2..00000000
--- a/tests/configs/train-test-real-bert.yaml
+++ /dev/null
@@ -1,68 +0,0 @@
-project_root: tests/project_folder
-model_name: model-real-bert
-read_format: parquet
-metadata_config_path: configs/metadata_configs/test-data-real-1.json
-
-input_columns: [itemValue]
-target_columns: [itemValue]
-target_column_types:
-  itemValue: real
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: false
-export_onnx: false
-export_pt: true
-export_with_dropout: false
-
-model_spec:
-  initial_embedding_dim: 16
-  dim_model: 16
-  n_head: 2
-  dim_feedforward: 8
-  num_layers: 1
-  prediction_length: 8
-  activation_fn: relu
-  normalization: layer_norm
-  positional_encoding: learned
-  attention_type: mha
-  norm_first: false
-  n_kv_heads: 2
-  rope_theta: 10000.0
-training_spec:
-  training_objective: bert
-  bert_spec:
-    masking_probability: 0.5
-    replacement_distribution:
-      masked: 0.8
-      random: 0.1
-      identical: 0.1
-    span_masking:
-      type: GeometricDistribution
-      p: 1.0
-  device: cpu
-  torch_compile: none
-  epochs: 1
-  save_interval_epochs: 1
-  batch_size: 5
-  log_interval: 1
-  learning_rate: 0.001
-  accumulation_steps: null
-  dropout: 0.0
-  criterion:
-    itemValue: MSELoss
-  loss_weights:
-    itemValue: 1.0
-  optimizer:
-    name: Adam
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  scheduler_step_on: epoch
-  calculate_validation_loss_on_initialization: false
-  continue_training: false
-  enforce_determinism: true
-  layer_autocast: false
diff --git a/tests/configs/train-test-real.yaml b/tests/configs/train-test-real.yaml
deleted file mode 100644
index e44db6a1..00000000
--- a/tests/configs/train-test-real.yaml
+++ /dev/null
@@ -1,60 +0,0 @@
-project_root: tests/project_folder
-model_name: default
-read_format: parquet
-metadata_config_path: configs/metadata_configs/test-data-real-1.json
-
-input_columns: null
-target_columns: [itemValue]
-target_column_types:
-  itemValue: real
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-export_onnx: true
-export_pt: true
-export_with_dropout: true
-
-model_spec:
-  initial_embedding_dim: 100
-  dim_model: 100
-  n_head: 2
-  dim_feedforward: 10
-  num_layers: 2
-  activation_fn: relu
-  normalization: layer_norm
-  positional_encoding: learned
-  attention_type: mha
-  norm_first: false
-  n_kv_heads: null
-  rope_theta: 10000.0
-  prediction_length: 1
-training_spec:
-  training_objective: causal
-  device: cpu
-  torch_compile: "inner"
-  epochs: 3
-  save_interval_epochs: 1
-  save_latest_interval_minutes: 0
-  save_batch_interval_minutes: 0
-  batch_size: 20
-  log_interval: 1
-  learning_rate: 0.003
-  accumulation_steps: null
-  dropout: 0.3
-  criterion:
-    itemValue: MSELoss
-  optimizer:
-    name: AdEMAMix
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  continue_training: false
-  enforce_determinism: true
-  layer_type_dtypes:
-    embedding: bfloat16
-    linear: bfloat16
-    norm: bfloat16
diff --git a/tests/configs/train-test-resume-epoch.yaml b/tests/configs/train-test-resume-epoch.yaml
deleted file mode 100644
index 1ab8bc7f..00000000
--- a/tests/configs/train-test-resume-epoch.yaml
+++ /dev/null
@@ -1,59 +0,0 @@
-project_root: tests/project_folder
-model_name: model-real-1-from-epoch-checkpoint
-read_format: parquet
-metadata_config_path: configs/metadata_configs/test-data-real-1.json
-
-input_columns: null
-target_columns: [itemValue]
-target_column_types:
-  itemValue: real
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: false
-export_embedding_model: true
-export_onnx: true
-export_pt: true
-export_with_dropout: false
-
-model_spec:
-  initial_embedding_dim: 100
-  dim_model: 100
-  n_head: 2
-  dim_feedforward: 10
-  num_layers: 2
-  activation_fn: relu
-  normalization: layer_norm
-  positional_encoding: learned
-  attention_type: mha
-  norm_first: false
-  n_kv_heads: null
-  rope_theta: 10000.0
-  prediction_length: 1
-training_spec:
-  training_objective: causal
-  device: cpu
-  epochs: 3
-  save_interval_epochs: 1
-  save_latest_interval_minutes: null
-  save_batch_interval_minutes: null
-  batch_size: 20
-  log_interval: 1
-  learning_rate: 0.003
-  accumulation_steps: null
-  dropout: 0.3
-  criterion:
-    itemValue: MSELoss
-  optimizer:
-    name: AdEMAMix
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  continue_training: true
-  enforce_determinism: true
-  layer_type_dtypes:
-    embedding: bfloat16
-    linear: bfloat16
-    norm: bfloat16
diff --git a/tests/configs/train-test-resume-mid-epoch.yaml b/tests/configs/train-test-resume-mid-epoch.yaml
deleted file mode 100644
index 7f4e6c2c..00000000
--- a/tests/configs/train-test-resume-mid-epoch.yaml
+++ /dev/null
@@ -1,61 +0,0 @@
-project_root: tests/project_folder
-model_name: model-categorical-3-from-mid-epoch-checkpoint
-read_format: pt
-metadata_config_path: configs/metadata_configs/test-data-categorical-3.json
-
-input_columns: [itemId, supCat1]
-target_columns: [itemId]
-target_column_types:
-  itemId: categorical
-
-context_length: 8
-inference_batch_size: 10
-
-export_generative_model: true
-export_embedding_model: true
-export_onnx: true
-export_pt: true
-
-model_spec:
-  initial_embedding_dim: 100
-  joint_embedding_dim: 200
-  dim_model: 200
-  n_head: 2
-  dim_feedforward: 12
-  num_layers: 2
-  prediction_length: 1
-  activation_fn: relu
-  normalization: rmsnorm
-  positional_encoding: rope
-  attention_type: mqa
-  norm_first: true
-  n_kv_heads: 1
-  rope_theta: 1000.0
-training_spec:
-  training_objective: causal
-  device: cpu
-  torch_compile: "outer"
-  epochs: 3
-  save_interval_epochs: 1
-  batch_size: 5
-  log_interval: 1
-  class_share_log_columns: ["itemId"]
-  learning_rate: 0.003
-  accumulation_steps: 2
-  dropout: 0.3
-  criterion:
-    itemId: CrossEntropyLoss
-  loss_weights:
-    itemId: 1.0
-    supCat1: 1.0
-  class_weights:
-    itemId: [1.0, 1.0, 1.0, 1.5, 0.5, 0.6, 1.2, 0.66666667, 1.5, 1.2, 0.75, 0.85714286, 0.75, 0.85714286, 0.75, 0.6, 2.,1.5, 1.5, 1.0, 1.0, 0.66666667, 0.75, 0.5, 0.75, 2.0, 1.2, 3.0, 1.0, 0.75, 1.0, 0.85714286, 1.0]
-  optimizer:
-    name: QHAdam
-  scheduler:
-    name: StepLR
-    step_size: 1.0
-    gamma: 0.99
-  scheduler_step_on: epoch
-  continue_training: true
-  enforce_determinism: true
diff --git a/tests/integration-test-log.txt b/tests/integration-test-log.txt
deleted file mode 100644
index 5d4ec365..00000000
--- a/tests/integration-test-log.txt
+++ /dev/null
@@ -1,80 +0,0 @@
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical.yaml --data-path tests/resources/source_data/test-data-categorical-1.csv --selected-columns None
-sequifier preprocess --config-path tests/configs/preprocess-test-real.yaml --data-path tests/resources/source_data/test-data-real-1.csv --selected-columns itemValue
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical.yaml --data-path tests/resources/source_data/test-data-categorical-3.csv --selected-columns None
-sequifier preprocess --config-path tests/configs/preprocess-test-real.yaml --data-path tests/resources/source_data/test-data-real-3.csv --selected-columns itemValue supReal1 supReal2
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical.yaml --data-path tests/resources/source_data/test-data-categorical-5.csv --selected-columns None
-sequifier preprocess --config-path tests/configs/preprocess-test-real.yaml --data-path tests/resources/source_data/test-data-real-5.csv --selected-columns itemValue supReal1 supReal2 supReal3 supReal4
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical.yaml --data-path tests/resources/source_data/test-data-categorical-50.csv --selected-columns None
-sequifier preprocess --config-path tests/configs/preprocess-test-real.yaml --data-path tests/resources/source_data/test-data-real-50.csv --selected-columns itemValue supReal1 supReal2 supReal3 supReal4 supReal5 supReal6 supReal7 supReal8 supReal9 supReal10 supReal11 supReal12 supReal13 supReal14 supReal15 supReal16 supReal17 supReal18 supReal19 supReal20 supReal21 supReal22 supReal23 supReal24 supReal25 supReal26 supReal27 supReal28 supReal29 supReal30 supReal31 supReal32 supReal33 supReal34 supReal35 supReal36 supReal37 supReal38 supReal39 supReal40 supReal41 supReal42 supReal43 supReal44 supReal45 supReal46 supReal47 supReal48 supReal49
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-lookahead-0.yaml --data-path tests/resources/source_data/test-data-categorical-1-lookahead-0.csv --selected-columns None
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-multitarget.yaml
-sequifier preprocess --config-path tests/configs/preprocess-test-multi-file.yaml
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-interrupted.yaml
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-exact.yaml
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-exact-pt.yaml
-sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-grid.yaml
-sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-sample.yaml
-sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-bert.yaml
-sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-bayesian.yaml
-sequifier hyperparameter-search --config-path tests/configs/hyperparameter-search-custom-eval.yaml
-sequifier train --config-path tests/configs/train-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-1.json --model-name model-categorical-1 --input-columns itemId
-sequifier train --config-path tests/configs/train-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-1.json --model-name model-real-1 --input-columns None
-sequifier train --config-path tests/configs/train-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-3.json --model-name model-categorical-3 --input-columns itemId supCat1
-sequifier train --config-path tests/configs/train-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-3.json --model-name model-real-3 --input-columns None
-sequifier train --config-path tests/configs/train-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-5.json --model-name model-categorical-5 --input-columns itemId supCat1 supCat2 supCat4
-sequifier train --config-path tests/configs/train-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-5.json --model-name model-real-5 --input-columns None
-sequifier train --config-path tests/configs/train-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-50.json --model-name model-categorical-50 --input-columns itemId supCat1 supCat2 supCat3 supCat4 supCat5 supCat6 supCat7 supCat8 supCat9 supCat10 supCat11 supCat12 supCat13 supCat14 supCat15 supCat16 supCat17 supCat18 supCat19 supCat20 supCat21 supCat22 supCat23 supCat24 supCat25 supCat26 supCat27 supCat28 supCat29 supCat30 supCat31 supCat32 supCat33 supCat34 supCat35 supCat36 supCat37 supCat38 supCat39 supCat40 supCat41 supCat42 supCat43 supCat44 supCat45 supCat46 supCat47 supCat48 supCat49
-sequifier train --config-path tests/configs/train-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-50.json --model-name model-real-50 --input-columns None
-sequifier train --config-path tests/configs/train-test-categorical-inf-size-1.yaml
-sequifier train --config-path tests/configs/train-test-categorical-inf-size-3.yaml
-sequifier train --config-path tests/configs/train-test-categorical-bert.yaml
-sequifier train --config-path tests/configs/train-test-real-bert.yaml
-sequifier train --config-path tests/configs/train-test-categorical-multitarget.yaml
-sequifier train --config-path tests/configs/train-test-categorical-multitarget-eager.yaml
-sequifier train --config-path tests/configs/train-test-distributed.yaml
-sequifier train --config-path tests/configs/train-test-distributed-lazy-parquet.yaml
-sequifier train --config-path tests/configs/train-test-lazy.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-1.json --model-path models/sequifier-model-categorical-1-best-3.onnx --data-path data/test-data-categorical-1-split2 --input-columns itemId
-sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-1.json --model-path models/sequifier-model-real-1-best-3.pt --data-path data/test-data-real-1-split1.parquet --input-columns None
-sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-3.json --model-path models/sequifier-model-categorical-3-best-3.onnx --data-path data/test-data-categorical-3-split2 --input-columns itemId supCat1
-sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-3.json --model-path models/sequifier-model-real-3-best-3.pt --data-path data/test-data-real-3-split1.parquet --input-columns None
-sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-5.json --model-path models/sequifier-model-categorical-5-best-3.onnx --data-path data/test-data-categorical-5-split2 --input-columns itemId supCat1 supCat2 supCat4
-sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-5.json --model-path models/sequifier-model-real-5-best-3.pt --data-path data/test-data-real-5-split1.parquet --input-columns None
-sequifier infer --config-path tests/configs/infer-test-categorical.yaml --metadata-config-path configs/metadata_configs/test-data-categorical-50.json --model-path models/sequifier-model-categorical-50-best-3.onnx --data-path data/test-data-categorical-50-split2 --input-columns itemId supCat1 supCat2 supCat3 supCat4 supCat5 supCat6 supCat7 supCat8 supCat9 supCat10 supCat11 supCat12 supCat13 supCat14 supCat15 supCat16 supCat17 supCat18 supCat19 supCat20 supCat21 supCat22 supCat23 supCat24 supCat25 supCat26 supCat27 supCat28 supCat29 supCat30 supCat31 supCat32 supCat33 supCat34 supCat35 supCat36 supCat37 supCat38 supCat39 supCat40 supCat41 supCat42 supCat43 supCat44 supCat45 supCat46 supCat47 supCat48 supCat49
-sequifier infer --config-path tests/configs/infer-test-real.yaml --metadata-config-path configs/metadata_configs/test-data-real-50.json --model-path models/sequifier-model-real-50-best-3.pt --data-path data/test-data-real-50-split1.parquet --input-columns None
-sequifier infer --config-path tests/configs/infer-test-categorical-multitarget.yaml
-sequifier infer --config-path tests/configs/infer-test-real-autoregression.yaml --input-columns itemValue --randomize
-sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-1.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3.yaml
-sequifier infer --config-path tests/configs/infer-test-distributed.yaml
-sequifier infer --config-path tests/configs/infer-test-distributed-parquet.yaml
-sequifier infer --config-path tests/configs/infer-test-lazy.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-autoregression.yaml  --input-columns itemId
-sequifier infer --config-path tests/configs/infer-test-categorical-embedding.yaml  --input-columns itemId
-sequifier infer --config-path tests/configs/infer-test-categorical-inf-size-3-embedding.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-bert.yaml
-sequifier infer --config-path tests/configs/infer-test-categorical-bert-embedding.yaml
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats.yaml
-sequifier preprocess --config-path tests/configs/preprocess-test-categorical-precomputed-stats-negative.yaml
-sequifier train --config-path tests/configs/train-test-resume-epoch.yaml
-sequifier train --config-path tests/configs/train-test-resume-mid-epoch.yaml
-sequifier visualize-training model-categorical-1 --project-root tests/project_folder
-sequifier visualize-training model-categorical-1-inf-size --project-root tests/project_folder
-sequifier visualize-training model-categorical-3 --project-root tests/project_folder
-sequifier visualize-training model-categorical-3-from-mid-epoch-checkpoint --project-root tests/project_folder
-sequifier visualize-training model-categorical-3-inf-size --project-root tests/project_folder
-sequifier visualize-training model-categorical-5 --project-root tests/project_folder
-sequifier visualize-training model-categorical-50 --project-root tests/project_folder
-sequifier visualize-training model-categorical-bert --project-root tests/project_folder
-sequifier visualize-training model-categorical-distributed --project-root tests/project_folder
-sequifier visualize-training model-categorical-distributed-lazy-parquet --project-root tests/project_folder
-sequifier visualize-training model-categorical-lazy --project-root tests/project_folder
-sequifier visualize-training model-categorical-multitarget-5 --project-root tests/project_folder
-sequifier visualize-training model-categorical-multitarget-5-eager --project-root tests/project_folder
-sequifier visualize-training model-real-1 --project-root tests/project_folder
-sequifier visualize-training model-real-1-from-epoch-checkpoint --project-root tests/project_folder
-sequifier visualize-training model-real-3 --project-root tests/project_folder
-sequifier visualize-training model-real-5 --project-root tests/project_folder
-sequifier visualize-training model-real-50 --project-root tests/project_folder
-sequifier visualize-training model-real-bert --project-root tests/project_folder
-sequifier visualize-training test-hp-search-grid-run-0,test-hp-search-grid-run-1,test-hp-search-grid-run-2,test-hp-search-grid-run-3 --project-root tests/project_folder --log-scale --bucket-training-batches 5
diff --git a/tests/integration/conftest.py b/tests/integration/conftest.py
deleted file mode 100644
index f96c0f61..00000000
--- a/tests/integration/conftest.py
+++ /dev/null
@@ -1,926 +0,0 @@
-import copy
-import os
-import shutil
-import subprocess
-import sys
-import time
-
-import polars as pl
-import pytest
-import torch
-import yaml
-
-SELECTED_COLUMNS = {
-    "categorical": {
-        1: "itemId",
-        3: "itemId supCat1",
-        5: "itemId supCat1 supCat2 supCat4",
-        50: "itemId " + " ".join([f"supCat{i}" for i in range(1, 50)]),
-    },
-    "real": {
-        1: "itemValue",
-        3: "itemValue supReal1 supReal2",
-        5: "itemValue supReal1 supReal2 supReal3 supReal4",
-        50: "itemValue " + " ".join([f"supReal{i}" for i in range(1, 50)]),
-    },
-}
-
-TARGET_VARIABLE_DICT = {"categorical": "itemId", "real": "itemValue"}
-
-
-def run_and_log(command: str) -> None:
-    subprocess.run(command, shell=True, check=True)
-
-    with open(os.path.join("tests", "integration-test-log.txt"), "a+") as f:
-        f.write(f"{command}\n")
-
-
-def pytest_configure(config):
-    os.environ["SEQUIFIER_TESTING"] = "1"
-
-
-@pytest.fixture(scope="session")
-def split_groups():
-    return {"categorical": 3, "real": 2}
-
-
-@pytest.fixture(scope="session")
-def project_root():
-    return os.path.join("tests", "project_folder")
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_cat():
-    return os.path.join("tests", "configs", "preprocess-test-categorical.yaml")
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_cat_multitarget():
-    return os.path.join(
-        "tests", "configs", "preprocess-test-categorical-multitarget.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_cat_lookahead_0():
-    return os.path.join(
-        "tests", "configs", "preprocess-test-categorical-lookahead-0.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_multi_file():
-    return os.path.join("tests", "configs", "preprocess-test-multi-file.yaml")
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_interrupted():
-    return os.path.join(
-        "tests", "configs", "preprocess-test-categorical-interrupted.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_exact():
-    return os.path.join("tests", "configs", "preprocess-test-categorical-exact.yaml")
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_exact_pt():
-    return os.path.join("tests", "configs", "preprocess-test-categorical-exact-pt.yaml")
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_real():
-    return os.path.join("tests", "configs", "preprocess-test-real.yaml")
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_cat_precomputed_stats():
-    return os.path.join(
-        "tests", "configs", "preprocess-test-categorical-precomputed-stats.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def preprocessing_config_path_cat_precomputed_stats_negative():
-    return os.path.join(
-        "tests",
-        "configs",
-        "preprocess-test-categorical-precomputed-stats-negative.yaml",
-    )
-
-
-@pytest.fixture(scope="session")
-def training_config_path_cat():
-    return os.path.join("tests", "configs", "train-test-categorical.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_cat_multitarget():
-    return os.path.join("tests", "configs", "train-test-categorical-multitarget.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_cat_multitarget_eager():
-    return os.path.join(
-        "tests", "configs", "train-test-categorical-multitarget-eager.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def training_config_path_real():
-    return os.path.join("tests", "configs", "train-test-real.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_cat_inf_size_1():
-    return os.path.join("tests", "configs", "train-test-categorical-inf-size-1.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_cat_inf_size_3():
-    return os.path.join("tests", "configs", "train-test-categorical-inf-size-3.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_cat_bert():
-    return os.path.join("tests", "configs", "train-test-categorical-bert.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_real_bert():
-    return os.path.join("tests", "configs", "train-test-real-bert.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_distributed():
-    return os.path.join("tests", "configs", "train-test-distributed.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_distributed_lazy_parquet():
-    return os.path.join("tests", "configs", "train-test-distributed-lazy-parquet.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_lazy():
-    return os.path.join("tests", "configs", "train-test-lazy.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_resume_epoch():
-    return os.path.join("tests", "configs", "train-test-resume-epoch.yaml")
-
-
-@pytest.fixture(scope="session")
-def training_config_path_resume_mid_epoch():
-    return os.path.join("tests", "configs", "train-test-resume-mid-epoch.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat():
-    return os.path.join("tests", "configs", "infer-test-categorical.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat_multitarget():
-    return os.path.join("tests", "configs", "infer-test-categorical-multitarget.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_real():
-    return os.path.join("tests", "configs", "infer-test-real.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat_inf_size_1():
-    return os.path.join("tests", "configs", "infer-test-categorical-inf-size-1.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat_inf_size_3():
-    return os.path.join("tests", "configs", "infer-test-categorical-inf-size-3.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat_bert():
-    return os.path.join("tests", "configs", "infer-test-categorical-bert.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat_bert_embedding():
-    return os.path.join(
-        "tests", "configs", "infer-test-categorical-bert-embedding.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_real_autoregression():
-    return os.path.join("tests", "configs", "infer-test-real-autoregression.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_categorical_autoregression():
-    return os.path.join(
-        "tests", "configs", "infer-test-categorical-autoregression.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_embedding():
-    return os.path.join("tests", "configs", "infer-test-categorical-embedding.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_cat_inf_size_3_embedding():
-    return os.path.join(
-        "tests", "configs", "infer-test-categorical-inf-size-3-embedding.yaml"
-    )
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_distributed():
-    return os.path.join("tests", "configs", "infer-test-distributed.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_distributed_parquet():
-    return os.path.join("tests", "configs", "infer-test-distributed-parquet.yaml")
-
-
-@pytest.fixture(scope="session")
-def inference_config_path_lazy():
-    return os.path.join("tests", "configs", "infer-test-lazy.yaml")
-
-
-@pytest.fixture(scope="session")
-def remove_project_root_contents(project_root):
-    if os.path.exists(project_root):
-        shutil.rmtree(project_root)
-    os.makedirs(project_root)
-
-    log_file_path = os.path.join("tests", "integration-test-log.txt")
-    if os.path.exists(log_file_path):
-        os.remove(log_file_path)
-
-    time.sleep(1)
-
-
-@pytest.fixture(scope="session")
-def hp_search_configs():
-    return {
-        "grid": os.path.join("tests", "configs", "hyperparameter-search-grid.yaml"),
-        "sample": os.path.join("tests", "configs", "hyperparameter-search-sample.yaml"),
-        "bert": os.path.join("tests", "configs", "hyperparameter-search-bert.yaml"),
-        "bayesian": os.path.join(
-            "tests", "configs", "hyperparameter-search-bayesian.yaml"
-        ),
-        "custom-eval": os.path.join(
-            "tests", "configs", "hyperparameter-search-custom-eval.yaml"
-        ),
-    }
-
-
-def reformat_parameter(attr, param, type):
-    if attr.endswith("_path"):
-        if type == "linux->local":
-            return os.path.join(*param.split("/"))
-        elif type == "local->linux":
-            return "/".join(os.path.split(param))
-    else:
-        return param
-
-
-@pytest.fixture(scope="session", autouse=True)
-def format_configs_locally(
-    preprocessing_config_path_cat,
-    preprocessing_config_path_cat_multitarget,
-    preprocessing_config_path_cat_lookahead_0,
-    preprocessing_config_path_real,
-    preprocessing_config_path_multi_file,
-    preprocessing_config_path_interrupted,
-    training_config_path_cat,
-    training_config_path_cat_multitarget,
-    training_config_path_real,
-    training_config_path_cat_inf_size_1,
-    training_config_path_cat_inf_size_3,
-    training_config_path_cat_bert,
-    training_config_path_real_bert,
-    training_config_path_distributed,
-    training_config_path_distributed_lazy_parquet,
-    training_config_path_lazy,
-    training_config_path_resume_epoch,
-    training_config_path_resume_mid_epoch,
-    inference_config_path_cat,
-    inference_config_path_cat_multitarget,
-    training_config_path_cat_multitarget_eager,
-    inference_config_path_real,
-    inference_config_path_real_autoregression,
-    inference_config_path_categorical_autoregression,
-    inference_config_path_cat_inf_size_1,
-    inference_config_path_cat_inf_size_3,
-    inference_config_path_cat_bert,
-    inference_config_path_cat_bert_embedding,
-    inference_config_path_distributed,
-    inference_config_path_distributed_parquet,
-    inference_config_path_lazy,
-    hp_search_configs,
-):
-    config_paths = [
-        preprocessing_config_path_cat,
-        preprocessing_config_path_cat_multitarget,
-        preprocessing_config_path_cat_lookahead_0,
-        preprocessing_config_path_real,
-        preprocessing_config_path_multi_file,
-        preprocessing_config_path_interrupted,
-        training_config_path_cat,
-        training_config_path_cat_multitarget,
-        training_config_path_real,
-        training_config_path_cat_inf_size_1,
-        training_config_path_cat_inf_size_3,
-        training_config_path_cat_bert,
-        training_config_path_real_bert,
-        training_config_path_distributed,
-        training_config_path_distributed_lazy_parquet,
-        training_config_path_lazy,
-        training_config_path_resume_epoch,
-        training_config_path_resume_mid_epoch,
-        inference_config_path_cat,
-        inference_config_path_cat_multitarget,
-        training_config_path_cat_multitarget_eager,
-        inference_config_path_real,
-        inference_config_path_real_autoregression,
-        inference_config_path_categorical_autoregression,
-        inference_config_path_cat_inf_size_1,
-        inference_config_path_cat_inf_size_3,
-        inference_config_path_cat_bert,
-        inference_config_path_cat_bert_embedding,
-        inference_config_path_distributed,
-        inference_config_path_distributed_parquet,
-        inference_config_path_lazy,
-        hp_search_configs["grid"],
-        hp_search_configs["sample"],
-        hp_search_configs["bert"],
-        hp_search_configs["bayesian"],
-        hp_search_configs["custom-eval"],
-    ]
-    original_configs = {}
-
-    cuda_available = torch.cuda.is_available()
-
-    for config_path in config_paths:
-        if sys.platform.startswith("win") or cuda_available:
-            with open(config_path, "r") as f:
-                config = yaml.safe_load(f)
-
-            original_config = copy.deepcopy(config)
-            original_configs[config_path] = original_config
-
-            assert config is not None, config_path
-
-            if sys.platform.startswith("win"):
-                config = {
-                    attr: reformat_parameter(attr, param, "linux->local")
-                    for attr, param in config.items()
-                }
-
-            if cuda_available:
-                if "training_spec" in config:
-                    config["training_spec"]["device"] = "cuda"  # type: ignore
-                if "device" in config:
-                    config["device"] = "cuda"
-
-            with open(config_path, "w") as f:
-                yaml.dump(config, f, default_flow_style=False, sort_keys=False)
-
-    yield
-
-    if sys.platform.startswith("win") or cuda_available:
-        for config_path in config_paths:
-            with open(config_path, "w") as f:
-                yaml.dump(
-                    original_configs[config_path],
-                    f,
-                    default_flow_style=False,
-                    sort_keys=False,
-                )
-
-
-@pytest.fixture(scope="session")
-def copy_interrupted_data(project_root, remove_project_root_contents):
-    os.makedirs(os.path.join(project_root, "data"), exist_ok=True)
-
-    source_path = os.path.join(
-        "tests", "resources", "source_data", "test-data-categorical-1-interrupted-temp"
-    )
-    target_path = os.path.join(
-        project_root, "data", "test-data-categorical-1-interrupted-temp"
-    )
-
-    shutil.copytree(source_path, target_path, dirs_exist_ok=True)
-
-
-@pytest.fixture(scope="session")
-def run_preprocessing(
-    project_root,
-    preprocessing_config_path_cat,
-    preprocessing_config_path_cat_multitarget,
-    preprocessing_config_path_cat_lookahead_0,
-    preprocessing_config_path_real,
-    preprocessing_config_path_multi_file,
-    preprocessing_config_path_interrupted,
-    preprocessing_config_path_exact,
-    preprocessing_config_path_exact_pt,
-    format_configs_locally,
-    remove_project_root_contents,
-    copy_interrupted_data,
-):
-    for data_number in [1, 3, 5, 50]:
-        data_path_cat = os.path.join(
-            "tests",
-            "resources",
-            "source_data",
-            f"test-data-categorical-{data_number}.csv",
-        )
-        run_and_log(
-            f"sequifier preprocess --config-path {preprocessing_config_path_cat} --data-path {data_path_cat} --selected-columns None"
-        )
-
-        data_path_real = os.path.join(
-            "tests", "resources", "source_data", f"test-data-real-{data_number}.csv"
-        )
-        run_and_log(
-            f"sequifier preprocess --config-path {preprocessing_config_path_real} --data-path {data_path_real} --selected-columns {SELECTED_COLUMNS['real'][data_number]}"
-        )
-
-    data_path_cat_lookahead_0 = os.path.join(
-        "tests",
-        "resources",
-        "source_data",
-        "test-data-categorical-1-lookahead-0.csv",
-    )
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_cat_lookahead_0} "
-        f"--data-path {data_path_cat_lookahead_0} --selected-columns None "
-    )
-
-    source_path = os.path.join("tests", "resources", "source_configs", "id_maps")
-    target_path = os.path.join(project_root, "configs", "id_maps")
-    shutil.copytree(source_path, target_path, dirs_exist_ok=True)
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_cat_multitarget}"
-    )
-    shutil.rmtree(target_path)
-
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_multi_file}"
-    )
-
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_interrupted}"
-    )
-
-    run_and_log(f"sequifier preprocess --config-path {preprocessing_config_path_exact}")
-
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_exact_pt}"
-    )
-
-    source_path = os.path.join(
-        "tests",
-        "resources",
-        "source_data",
-        "test-data-real-1-split1-autoregression.csv",
-    )
-
-    target_path = os.path.join(
-        "tests", "project_folder", "data", "test-data-real-1-split1-autoregression.csv"
-    )
-
-    shutil.copyfile(source_path, target_path)
-
-    os.makedirs(os.path.join(project_root, "scripts"))
-    source_path = os.path.join(
-        "tests", "resources", "source_scripts", "hp_search_eval_script.py"
-    )
-    target_path = os.path.join(project_root, "scripts", "hp_search_eval_script.py")
-    shutil.copyfile(source_path, target_path)
-
-    source_path = os.path.join(
-        "tests", "configs", "hyperparameter-search-custom-eval-inference.yaml"
-    )
-    target_path = os.path.join(
-        project_root, "configs", "hyperparameter-search-custom-eval-inference.yaml"
-    )
-    shutil.copyfile(source_path, target_path)
-
-
-@pytest.fixture(scope="session")
-def run_training(
-    run_preprocessing,
-    project_root,
-    training_config_path_cat,
-    training_config_path_real,
-    training_config_path_cat_inf_size_1,
-    training_config_path_cat_inf_size_3,
-    training_config_path_cat_bert,
-    training_config_path_real_bert,
-    training_config_path_distributed,
-    training_config_path_distributed_lazy_parquet,
-    training_config_path_lazy,
-    training_config_path_cat_multitarget,
-    training_config_path_cat_multitarget_eager,
-):
-    for model_number in [1, 3, 5, 50]:
-        metadata_config_path_cat = os.path.join(
-            "configs", "metadata_configs", f"test-data-categorical-{model_number}.json"
-        )
-        model_name_cat = f"model-categorical-{model_number}"
-        run_and_log(
-            f"sequifier train --config-path {training_config_path_cat} --metadata-config-path {metadata_config_path_cat} --model-name {model_name_cat} --input-columns {SELECTED_COLUMNS['categorical'][model_number]}"
-        )
-
-        metadata_config_path_real = os.path.join(
-            "configs", "metadata_configs", f"test-data-real-{model_number}.json"
-        )
-        model_name_real = f"model-real-{model_number}"
-        run_and_log(
-            f"sequifier train --config-path {training_config_path_real} --metadata-config-path {metadata_config_path_real} --model-name {model_name_real} --input-columns None"
-        )
-
-    run_and_log(f"sequifier train --config-path {training_config_path_cat_inf_size_1}")
-
-    run_and_log(f"sequifier train --config-path {training_config_path_cat_inf_size_3}")
-
-    run_and_log(f"sequifier train --config-path {training_config_path_cat_bert}")
-
-    run_and_log(f"sequifier train --config-path {training_config_path_real_bert}")
-
-    run_and_log(f"sequifier train --config-path {training_config_path_cat_multitarget}")
-
-    run_and_log(
-        f"sequifier train --config-path {training_config_path_cat_multitarget_eager}"
-    )
-
-    run_and_log(f"sequifier train --config-path {training_config_path_distributed}")
-
-    run_and_log(
-        f"sequifier train --config-path {training_config_path_distributed_lazy_parquet}"
-    )
-
-    run_and_log(f"sequifier train --config-path {training_config_path_lazy}")
-
-    source_path = os.path.join(
-        project_root, "models", "sequifier-model-real-1-best-3.pt"
-    )
-    target_path = os.path.join(
-        project_root, "models", "sequifier-model-real-1-best-3-autoregression.pt"
-    )
-
-    shutil.copy(source_path, target_path)
-
-
-@pytest.fixture(scope="session")
-def copy_checkpoints(run_training, project_root):
-    src_ckpt = os.path.join(project_root, "checkpoints", "model-real-1-epoch-1.pt")
-    dst_ckpt = os.path.join(
-        project_root, "checkpoints", "model-real-1-from-epoch-checkpoint-epoch-1.pt"
-    )
-    shutil.copy(src_ckpt, dst_ckpt)
-
-    src_ckpt = os.path.join(
-        project_root, "checkpoints", "model-categorical-3-epoch-1.pt"
-    )
-    dst_ckpt = os.path.join(
-        project_root,
-        "checkpoints",
-        "model-categorical-3-from-mid-epoch-checkpoint-epoch-1.pt",
-    )
-    shutil.copy(src_ckpt, dst_ckpt)
-
-
-@pytest.fixture(scope="session")
-def run_training_from_checkpoint(
-    copy_checkpoints,
-    training_config_path_resume_epoch,
-    training_config_path_resume_mid_epoch,
-):
-    run_and_log(f"sequifier train --config-path {training_config_path_resume_epoch}")
-
-    run_and_log(
-        f"sequifier train --config-path {training_config_path_resume_mid_epoch}"
-    )
-
-
-@pytest.fixture(scope="session")
-def run_hp_search(
-    project_root, hp_search_configs, format_configs_locally, run_preprocessing
-):
-    run_and_log(
-        f"sequifier hyperparameter-search --config-path {hp_search_configs['grid']}"
-    )
-
-    run_and_log(
-        f"sequifier hyperparameter-search --config-path {hp_search_configs['sample']}"
-    )
-
-    run_and_log(
-        f"sequifier hyperparameter-search --config-path {hp_search_configs['bert']}"
-    )
-
-    run_and_log(
-        f"sequifier hyperparameter-search --config-path {hp_search_configs['bayesian']}"
-    )
-
-    run_and_log(
-        f"sequifier hyperparameter-search --config-path {hp_search_configs['custom-eval']}"
-    )
-
-
-@pytest.fixture(scope="session")
-def copy_autoregression_model(project_root, run_training):
-    model_path = os.path.join(
-        project_root, "models", "sequifier-model-categorical-1-best-3.onnx"
-    )
-    target_path = os.path.join(
-        project_root,
-        "models",
-        "sequifier-model-categorical-1-best-3-autoregression.onnx",
-    )
-    shutil.copyfile(model_path, target_path)
-
-
-@pytest.fixture(scope="session")
-def run_inference(
-    project_root,
-    run_training,
-    copy_autoregression_model,
-    inference_config_path_cat,
-    inference_config_path_cat_multitarget,
-    inference_config_path_real,
-    inference_config_path_real_autoregression,
-    inference_config_path_categorical_autoregression,
-    inference_config_path_embedding,
-    inference_config_path_cat_inf_size_1,
-    inference_config_path_cat_inf_size_3,
-    inference_config_path_cat_bert,
-    inference_config_path_cat_bert_embedding,
-    inference_config_path_distributed,
-    inference_config_path_distributed_parquet,
-    inference_config_path_lazy,
-    inference_config_path_cat_inf_size_3_embedding,
-):
-    for model_number in [1, 3, 5, 50]:
-        model_path_cat = os.path.join(
-            "models", f"sequifier-model-categorical-{model_number}-best-3.onnx"
-        )
-        data_path_cat = os.path.join(
-            "data", f"test-data-categorical-{model_number}-split2"
-        )
-        metadata_config_path_cat = os.path.join(
-            "configs", "metadata_configs", f"test-data-categorical-{model_number}.json"
-        )
-        run_and_log(
-            f"sequifier infer --config-path {inference_config_path_cat} --metadata-config-path {metadata_config_path_cat} --model-path {model_path_cat} --data-path {data_path_cat} --input-columns {SELECTED_COLUMNS['categorical'][model_number]}"
-        )
-
-        model_path_real = os.path.join(
-            "models", f"sequifier-model-real-{model_number}-best-3.pt"
-        )
-        data_path_real = os.path.join(
-            "data", f"test-data-real-{model_number}-split1.parquet"
-        )
-        metadata_config_path_real = os.path.join(
-            "configs", "metadata_configs", f"test-data-real-{model_number}.json"
-        )
-        run_and_log(
-            f"sequifier infer --config-path {inference_config_path_real} --metadata-config-path {metadata_config_path_real} --model-path {model_path_real} --data-path {data_path_real} --input-columns None"
-        )
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_cat_multitarget}"
-    )
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_real_autoregression} --input-columns {SELECTED_COLUMNS['real'][1]} --randomize"
-    )
-
-    run_and_log(f"sequifier infer --config-path {inference_config_path_cat_inf_size_1}")
-
-    run_and_log(f"sequifier infer --config-path {inference_config_path_cat_inf_size_3}")
-
-    run_and_log(f"sequifier infer --config-path {inference_config_path_distributed}")
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_distributed_parquet}"
-    )
-
-    run_and_log(f"sequifier infer --config-path {inference_config_path_lazy}")
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_categorical_autoregression}  --input-columns itemId"
-    )
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_embedding}  --input-columns itemId"
-    )
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_cat_inf_size_3_embedding}"
-    )
-
-    run_and_log(f"sequifier infer --config-path {inference_config_path_cat_bert}")
-
-    run_and_log(
-        f"sequifier infer --config-path {inference_config_path_cat_bert_embedding}"
-    )
-
-
-@pytest.fixture()
-def model_names_preds():
-    model_names_preds = [
-        f"model-{variant}-{model_number}-best-3"
-        for variant in ["categorical", "real"]
-        for model_number in [1, 3, 5, 50]
-    ]
-    model_names_preds += [
-        "model-categorical-multitarget-5-best-3",
-        "model-categorical-multitarget-5-last-3",
-        "model-real-1-best-3-autoregression",
-        "model-categorical-1-best-3-autoregression",
-        "model-categorical-1-inf-size-best-3",
-        "model-categorical-3-inf-size-best-3",
-        "model-categorical-distributed-best-3",
-        "model-categorical-lazy-best-3",
-    ]
-
-    return model_names_preds
-
-
-@pytest.fixture()
-def model_names_probs():
-    model_names_probs = [
-        f"model-categorical-{model_number}-best-3-itemId"
-        for model_number in [1, 3, 5, 50]
-    ]
-    model_names_probs += [
-        f"model-categorical-multitarget-5-best-3-{col}" for col in ["itemId", "supCat1"]
-    ]
-    model_names_probs += [
-        f"model-categorical-3-inf-size-best-3-{col}"
-        for col in ["itemId", "supCat1", "supCat2"]
-    ]
-    return model_names_probs
-
-
-@pytest.fixture()
-def model_names_embeddings():
-    model_names_embeddings = [
-        "model-categorical-1-best-embedding-3",
-        "model-categorical-3-inf-size-best-embedding-3",
-    ]
-    return model_names_embeddings
-
-
-@pytest.fixture()
-def targets(model_names_preds, model_names_probs, model_names_embeddings):
-    target_dict = {"preds": {}, "probs": {}, "embeds": {}}
-    for model_name in model_names_preds:
-        target_type = "categorical" if "categorical" in model_name else "real"
-        file_name = f"sequifier-{model_name}-predictions"
-        target_path = os.path.join(
-            "tests", "resources", "target_outputs", "predictions", file_name
-        )
-        target_dict["preds"][model_name] = read_multi_file_preds(
-            target_path, target_type
-        )
-
-    for model_name in model_names_probs:
-        target_type = "categorical" if "categorical" in model_name else "real"
-        file_name = f"sequifier-{model_name}-probabilities"
-        target_path = os.path.join(
-            "tests", "resources", "target_outputs", "probabilities", file_name
-        )
-        target_dict["probs"][model_name] = read_multi_file_preds(
-            target_path, target_type
-        )
-
-    for model_name in model_names_embeddings:
-        target_type = "categorical" if "categorical" in model_name else "real"
-        file_name = f"sequifier-{model_name}-embeddings"
-        target_path = os.path.join(
-            "tests", "resources", "target_outputs", "embeddings", file_name
-        )
-        target_dict["embeds"][model_name] = read_multi_file_preds(
-            target_path, target_type
-        )
-
-    return target_dict
-
-
-def read_multi_file_preds(path, target_type, file_suffix=None):
-    dtype = (
-        {
-            **{TARGET_VARIABLE_DICT[target_type]: str},
-            **{f"supCat{i+1}": str for i in range(50)},
-        }
-        if target_type == "categorical"
-        else None
-    )
-    if target_type == "categorical":
-        contents = []
-        for root, dirs, files in os.walk(path):
-            for file in sorted(list(files)):
-                if file_suffix is None or file.endswith(file_suffix):
-                    contents.append(
-                        pl.read_csv(os.path.join(root, file), schema_overrides=dtype)
-                    )
-        assert len(contents) > 0, f"no files found for {path}"
-        return pl.concat(contents, how="vertical")
-    else:
-        return pl.read_csv(f"{path}.csv", separator=",", schema_overrides=dtype)
-
-
-@pytest.fixture()
-def predictions(run_inference, model_names_preds, project_root):
-    preds = {}
-    for model_name in model_names_preds:
-        target_type = "categorical" if "categorical" in model_name else "real"
-
-        prediction_path = os.path.join(
-            project_root,
-            "outputs",
-            "predictions",
-            f"sequifier-{model_name}-predictions",
-        )
-
-        preds[model_name] = read_multi_file_preds(prediction_path, target_type)
-
-    return preds
-
-
-@pytest.fixture()
-def probabilities(run_inference, model_names_probs, project_root):
-    probs = {}
-    for model_name in model_names_probs:
-        probabilities_path = os.path.join(
-            project_root,
-            "outputs",
-            "probabilities",
-            f"sequifier-{model_name}-probabilities",
-        )
-        probs[model_name] = read_multi_file_preds(
-            probabilities_path, "categorical", "csv"
-        )
-
-    return probs
-
-
-@pytest.fixture()
-def embeddings(run_inference, model_names_embeddings, project_root):
-    embeds = {}
-    for model_name in model_names_embeddings:
-        embeddings_path = os.path.join(
-            project_root,
-            "outputs",
-            "embeddings",
-            f"sequifier-{model_name}-embeddings",
-        )
-        embeds[model_name] = read_multi_file_preds(
-            embeddings_path, "categorical", "csv"
-        )
-    return embeds
-
-
-@pytest.fixture()
-def bert_predictions(run_inference, project_root):
-    prediction_path = os.path.join(
-        project_root,
-        "outputs",
-        "predictions",
-        "sequifier-model-categorical-bert-best-1-predictions",
-    )
-    return read_multi_file_preds(prediction_path, "categorical")
-
-
-@pytest.fixture()
-def bert_probabilities(run_inference, project_root):
-    probabilities_path = os.path.join(
-        project_root,
-        "outputs",
-        "probabilities",
-        "sequifier-model-categorical-bert-best-1-itemId-probabilities",
-    )
-    return read_multi_file_preds(probabilities_path, "categorical", "csv")
-
-
-@pytest.fixture()
-def bert_embeddings(run_inference, project_root):
-    embeddings_path = os.path.join(
-        project_root,
-        "outputs",
-        "embeddings",
-        "sequifier-model-categorical-bert-best-embedding-1-embeddings",
-    )
-    return read_multi_file_preds(embeddings_path, "categorical", "csv")
diff --git a/tests/integration/test_hyperparameter_search.py b/tests/integration/test_hyperparameter_search.py
deleted file mode 100644
index 17c83460..00000000
--- a/tests/integration/test_hyperparameter_search.py
+++ /dev/null
@@ -1,90 +0,0 @@
-import glob
-import json
-import os
-
-import yaml
-
-
-def test_hp_search_grid_outputs(run_hp_search, project_root):
-    hp_name = "test-hp-search-grid"
-    config_dir = os.path.join(project_root, "configs")
-
-    generated_configs = glob.glob(os.path.join(config_dir, f"{hp_name}-run-*.yaml"))
-    assert (
-        len(generated_configs) == 4
-    ), f"Expected 4 grid configs, found {len(generated_configs)}"
-
-
-def test_hp_search_sample_outputs(run_hp_search, project_root):
-    hp_name = "test-hp-search-sample"
-    config_dir = os.path.join(project_root, "configs")
-
-    generated_configs = glob.glob(os.path.join(config_dir, f"{hp_name}-run-*.yaml"))
-    assert (
-        len(generated_configs) == 4
-    ), f"Expected 4 sample configs, found {len(generated_configs)}"
-
-
-def test_hp_search_bert_outputs(run_hp_search, project_root):
-    hp_name = "test-hp-search-bert"
-    config_dir = os.path.join(project_root, "configs")
-
-    generated_configs = glob.glob(os.path.join(config_dir, f"{hp_name}-run-*.yaml"))
-    assert (
-        len(generated_configs) == 1
-    ), f"Expected 1 BERT sample config, found {len(generated_configs)}"
-
-    with open(generated_configs[0], "r") as f:
-        generated_config = yaml.safe_load(f)
-
-    assert generated_config["metadata_config_path"].endswith(
-        "test-data-categorical-1-lookahead-0.json"
-    )
-    assert "storage_layout" not in generated_config
-    assert "window_view" not in generated_config
-    assert generated_config["target_offset"] == 0
-    assert generated_config["training_spec"]["training_objective"] == "bert"
-
-
-def test_hp_search_bayesian_outputs(run_hp_search, project_root):
-    hp_name = "test-hp-search-bayesian"
-    config_dir = os.path.join(project_root, "configs")
-
-    generated_configs = glob.glob(os.path.join(config_dir, f"{hp_name}-run-*.yaml"))
-    assert (
-        len(generated_configs) == 4
-    ), f"Expected 4 bayesian configs, found {len(generated_configs)}"
-
-
-def test_hp_search_state(run_hp_search, project_root):
-    state_dir = os.path.join(project_root, "state", "optuna")
-
-    assert os.path.exists(os.path.join(state_dir, "test-hp-search-sample.db"))
-    assert os.path.exists(os.path.join(state_dir, "test-hp-search-grid.db"))
-    assert os.path.exists(os.path.join(state_dir, "test-hp-search-bert.db"))
-    assert os.path.exists(os.path.join(state_dir, "test-hp-search-bayesian.db"))
-    assert os.path.exists(os.path.join(state_dir, "test-hp-search-custom-eval.db"))
-
-
-def test_hp_search_inference_feedback_loop(run_hp_search, project_root):
-    # Verify that the evaluations directory was populated
-    eval_dir = os.path.join(project_root, "outputs", "evaluations")
-    assert os.path.exists(eval_dir), f"Evaluation directory {eval_dir} was not created."
-
-    eval_files = [
-        f
-        for f in os.listdir(eval_dir)
-        if f.startswith("test-hp-search-custom-eval-run-") and f.endswith(".json")
-    ]
-
-    assert len(eval_files) == 4, f"Expected 4 evaluation JSONs, found {len(eval_files)}"
-
-    for f in eval_files:
-        with open(os.path.join(eval_dir, f), "r") as fp:
-            metrics = json.load(fp)
-            assert "max" in metrics, f"'max' missing in {f}"
-            assert "stdev" in metrics, f"'stdev' missing in {f}"
-
-            # Sanity check that metrics were actually calculated
-            assert isinstance(metrics["max"], int)
-            assert isinstance(metrics["stdev"], float)
diff --git a/tests/integration/test_identities.py b/tests/integration/test_identities.py
deleted file mode 100644
index f2ecee46..00000000
--- a/tests/integration/test_identities.py
+++ /dev/null
@@ -1,83 +0,0 @@
-import numpy as np
-import pytest
-
-
-def _frame_mismatches(actual, expected, *, rtol=1e-5, atol=1e-6):
-    mismatches = []
-    for column in actual.columns:
-        actual_values = actual[column].to_numpy()
-        expected_values = expected[column].to_numpy()
-        if np.issubdtype(actual_values.dtype, np.floating):
-            try:
-                np.testing.assert_allclose(
-                    actual_values,
-                    expected_values,
-                    rtol=rtol,
-                    atol=atol,
-                )
-            except AssertionError as exc:
-                mismatches.append((column, str(exc)))
-        elif not np.array_equal(actual_values, expected_values):
-            mismatches.append(
-                (
-                    column,
-                    f"{actual_values = } != {expected_values = }",
-                )
-            )
-    return mismatches
-
-
-@pytest.mark.optional
-def test_identities(targets, predictions, probabilities, embeddings):
-    failed_models = []
-    for model_name, preds in predictions.items():
-        mismatches = _frame_mismatches(preds, targets["preds"][model_name])
-        if model_name != "model-real-1-best-3-autoregression":
-            if mismatches:
-                failed_models.append(
-                    (
-                        model_name,
-                        mismatches,
-                        f"{model_name} preds differ from target: {mismatches = }",
-                    )
-                )
-        else:
-            equal = preds.to_numpy() == targets["preds"][model_name].to_numpy()
-            mean_equal = np.mean(equal.astype(int))
-            if mean_equal == 1.0:
-                failed_models.append(
-                    (
-                        model_name,
-                        equal,
-                        f"{model_name} preds are not randomized, {preds.to_numpy() = } == {targets['preds'][model_name].to_numpy() = }: {equal = }, {mean_equal = }",
-                    )
-                )
-
-    for model_name, probs in probabilities.items():
-        mismatches = _frame_mismatches(probs, targets["probs"][model_name])
-        if mismatches:
-            failed_models.append(
-                (
-                    model_name,
-                    mismatches,
-                    f"{model_name} probs differ from target: {mismatches = }",
-                )
-            )
-
-    for model_name, embeds in embeddings.items():
-        mismatches = _frame_mismatches(embeds, targets["embeds"][model_name])
-        if mismatches:
-            failed_models.append(
-                (
-                    model_name,
-                    mismatches,
-                    f"{model_name} embeddings differ from target: {mismatches = }",
-                )
-            )
-
-    if len(failed_models):
-        print(failed_models)
-        failed_models_subset = [(model, message) for model, _, message in failed_models]
-        assert (
-            len(failed_models) == 0
-        ), f"{len(failed_models) = } - {failed_models_subset}"
diff --git a/tests/integration/test_inference.py b/tests/integration/test_inference.py
deleted file mode 100644
index 1f9d9887..00000000
--- a/tests/integration/test_inference.py
+++ /dev/null
@@ -1,536 +0,0 @@
-import json
-import os
-import re
-from collections import Counter
-
-import numpy as np
-import polars as pl
-import torch
-
-from sequifier.helpers import (
-    ModelWindowView,
-    StoredWindowLayout,
-    resolve_window_view,
-    stored_window_layout_from_metadata,
-)
-
-TARGET_VARIABLE_DICT = {"categorical": "itemId", "real": "itemValue"}
-BERT_SEQ_LENGTH = 8
-BERT_EMBEDDING_DIM = 16
-BERT_DATA_NAME = "test-data-categorical-1-lookahead-0"
-CAUSAL_CONTEXT_LENGTH = 8
-
-
-def _project_path(project_root, path):
-    if os.path.isabs(path) or path.startswith(project_root):
-        return path
-    return os.path.join(project_root, path)
-
-
-def _causal_storage_layout(data_path):
-    metadata_path = (
-        os.path.join(data_path, "metadata.json") if os.path.isdir(data_path) else None
-    )
-    if metadata_path is not None and os.path.exists(metadata_path):
-        with open(metadata_path, "r") as f:
-            return stored_window_layout_from_metadata(json.load(f))
-
-    return StoredWindowLayout(
-        stored_context_width=CAUSAL_CONTEXT_LENGTH + 1,
-        max_target_offset=1,
-        version=2,
-    )
-
-
-def _target_valid_mask(left_pad_lengths, storage_layout, prediction_length):
-    resolved_view = resolve_window_view(
-        storage_layout,
-        ModelWindowView(
-            context_length=CAUSAL_CONTEXT_LENGTH,
-            objective="causal",
-            target_offset=1,
-        ),
-    )
-    masks = resolved_view.build_masks(torch.tensor(left_pad_lengths, dtype=torch.int64))
-    return masks["target_valid_mask"][:, -prediction_length:].reshape(-1).numpy()
-
-
-def _read_pt_window_metadata(data_path):
-    contents = []
-    for root, _, files in os.walk(data_path):
-        for file in sorted(files):
-            if not file.endswith(".pt"):
-                continue
-
-            loaded = torch.load(os.path.join(root, file), weights_only=False)
-            if len(loaded) == 5:
-                _, sequence_ids, _, start_positions, left_pad_lengths = loaded
-            else:
-                _, sequence_ids, _, start_positions = loaded
-                left_pad_lengths = torch.zeros_like(sequence_ids)
-
-            contents.append(
-                pl.DataFrame(
-                    {
-                        "sequenceId": sequence_ids.detach().cpu().numpy(),
-                        "startItemPosition": start_positions.detach().cpu().numpy(),
-                        "leftPadLength": left_pad_lengths.detach().cpu().numpy(),
-                    }
-                )
-            )
-
-    assert len(contents) > 0, f"no files found for {data_path}"
-    return pl.concat(contents, how="vertical")
-
-
-def _read_long_window_metadata(data_path):
-    paths = []
-    if os.path.isdir(data_path):
-        for root, _, files in os.walk(data_path):
-            paths.extend(
-                os.path.join(root, file)
-                for file in sorted(files)
-                if file.endswith((".csv", ".parquet"))
-            )
-    else:
-        paths = [data_path]
-
-    contents = []
-    for path in paths:
-        data = pl.read_csv(path) if path.endswith(".csv") else pl.read_parquet(path)
-        if "leftPadLength" not in data.columns:
-            data = data.with_columns(pl.lit(0).alias("leftPadLength"))
-        contents.append(
-            data.group_by(["sequenceId", "subsequenceId"], maintain_order=True).agg(
-                pl.col("startItemPosition").first(),
-                pl.col("leftPadLength").first(),
-            )
-        )
-
-    assert len(contents) > 0, f"no files found for {data_path}"
-    return pl.concat(contents, how="vertical")
-
-
-def _prediction_source_spec(project_root, model_name):
-    if model_name == "model-categorical-multitarget-5-best-3":
-        return {
-            "path": _project_path(
-                project_root, "data/test-data-categorical-multitarget-5-split2"
-            ),
-            "format": "parquet",
-            "prediction_length": 1,
-            "autoregression_total_steps": None,
-        }
-    if model_name == "model-categorical-multitarget-5-last-3":
-        return {
-            "path": _project_path(
-                project_root, "data/test-data-categorical-multitarget-5-split2"
-            ),
-            "format": "parquet",
-            "prediction_length": 1,
-            "autoregression_total_steps": None,
-        }
-    if model_name == "model-categorical-distributed-best-3":
-        return {
-            "path": _project_path(project_root, "data/test-data-categorical-3-split2"),
-            "format": "pt",
-            "prediction_length": 1,
-            "autoregression_total_steps": None,
-        }
-    if model_name == "model-categorical-lazy-best-3":
-        return {
-            "path": _project_path(project_root, "data/test-data-categorical-3-split2"),
-            "format": "pt",
-            "prediction_length": 1,
-            "autoregression_total_steps": None,
-        }
-    if model_name == "model-categorical-1-best-3-autoregression":
-        return {
-            "path": _project_path(project_root, "data/test-data-categorical-1-split2"),
-            "format": "pt",
-            "prediction_length": 1,
-            "autoregression_total_steps": 20,
-        }
-    if model_name == "model-real-1-best-3-autoregression":
-        return {
-            "path": _project_path(
-                project_root, "data/test-data-real-1-split1-autoregression.csv"
-            ),
-            "format": "csv",
-            "prediction_length": 1,
-            "autoregression_total_steps": 20,
-            "csv_autoregression": True,
-        }
-
-    match = re.fullmatch(r"model-categorical-(\d+)-inf-size-best-3", model_name)
-    if match is not None:
-        data_number = int(match.group(1))
-        return {
-            "path": _project_path(
-                project_root, f"data/test-data-categorical-{data_number}-split2"
-            ),
-            "format": "pt",
-            "prediction_length": 3,
-            "autoregression_total_steps": None,
-        }
-
-    match = re.fullmatch(r"model-categorical-(\d+)-best-3", model_name)
-    if match is not None:
-        data_number = int(match.group(1))
-        return {
-            "path": _project_path(
-                project_root, f"data/test-data-categorical-{data_number}-split2"
-            ),
-            "format": "pt",
-            "prediction_length": 1,
-            "autoregression_total_steps": None,
-        }
-
-    match = re.fullmatch(r"model-real-(\d+)-best-3", model_name)
-    if match is not None:
-        data_number = int(match.group(1))
-        return {
-            "path": _project_path(
-                project_root, f"data/test-data-real-{data_number}-split1.parquet"
-            ),
-            "format": "parquet",
-            "prediction_length": 1,
-            "autoregression_total_steps": None,
-        }
-
-    raise AssertionError(f"No source metadata mapping for {model_name}")
-
-
-def _expected_prediction_positions(project_root, model_name):
-    spec = _prediction_source_spec(project_root, model_name)
-    metadata = (
-        _read_pt_window_metadata(spec["path"])
-        if spec["format"] == "pt"
-        else _read_long_window_metadata(spec["path"])
-    )
-    storage_layout = _causal_storage_layout(spec["path"])
-    prediction_length = spec["prediction_length"]
-
-    if spec["autoregression_total_steps"] is not None:
-        total_steps = spec["autoregression_total_steps"]
-        if spec.get("csv_autoregression", False):
-            metadata = metadata.group_by("sequenceId", maintain_order=True).head(1)
-
-        sequence_ids = np.repeat(metadata["sequenceId"].to_numpy(), total_steps)
-        item_positions = np.concatenate(
-            [
-                np.arange(start, start + total_steps)
-                for start in (
-                    metadata["startItemPosition"].to_numpy() + CAUSAL_CONTEXT_LENGTH
-                )
-            ],
-            axis=0,
-        )
-        valid_mask = np.repeat(
-            _target_valid_mask(
-                metadata["leftPadLength"].to_numpy(), storage_layout, prediction_length
-            ),
-            total_steps,
-        )
-    else:
-        starts = metadata["startItemPosition"].to_numpy()
-        offsets = np.arange(-prediction_length + 1, 1)
-        sequence_ids = np.repeat(metadata["sequenceId"].to_numpy(), prediction_length)
-        item_positions = np.repeat(
-            starts + CAUSAL_CONTEXT_LENGTH, prediction_length
-        ) + np.tile(offsets, len(starts))
-        valid_mask = _target_valid_mask(
-            metadata["leftPadLength"].to_numpy(), storage_layout, prediction_length
-        )
-
-    return pl.DataFrame(
-        {
-            "sequenceId": sequence_ids[valid_mask],
-            "itemPosition": item_positions[valid_mask],
-        }
-    )
-
-
-def _categorical_metadata(project_root):
-    metadata_path = os.path.join(
-        project_root, "configs", "metadata_configs", f"{BERT_DATA_NAME}.json"
-    )
-    with open(metadata_path, "r") as f:
-        return json.load(f)
-
-
-def _bert_inference_metadata(project_root):
-    data_path = os.path.join(project_root, "data", f"{BERT_DATA_NAME}-split2")
-    contents = []
-    for root, _, files in os.walk(data_path):
-        for file in sorted(files):
-            if not file.endswith(".pt"):
-                continue
-
-            loaded = torch.load(os.path.join(root, file), weights_only=False)
-            if len(loaded) == 5:
-                _, sequence_ids, subsequence_ids, _, left_pad_lengths = loaded
-            else:
-                _, sequence_ids, subsequence_ids, _ = loaded
-                left_pad_lengths = None
-
-            if left_pad_lengths is None:
-                left_pad_lengths = torch.zeros_like(sequence_ids)
-
-            valid_counts = torch.clamp(
-                BERT_SEQ_LENGTH - left_pad_lengths, min=0, max=BERT_SEQ_LENGTH
-            )
-            contents.append(
-                pl.DataFrame(
-                    {
-                        "sequenceId": sequence_ids.detach().cpu().numpy(),
-                        "subsequenceId": subsequence_ids.detach().cpu().numpy(),
-                        "valid_count": valid_counts.detach().cpu().numpy(),
-                    }
-                )
-            )
-
-    assert len(contents) > 0, f"no files found for {data_path}"
-    return pl.concat(contents, how="vertical")
-
-
-def _expected_bert_valid_counts(project_root, group_columns):
-    metadata = _bert_inference_metadata(project_root)
-
-    return (
-        metadata.group_by(group_columns)
-        .agg(pl.col("valid_count").sum().alias("expected_len"))
-        .filter(pl.col("expected_len") > 0)
-    )
-
-
-def _assert_counts_match_expected(actual, expected, group_columns):
-    actual_keys = set(actual.select(group_columns).iter_rows())
-    expected_keys = set(expected.select(group_columns).iter_rows())
-
-    assert actual_keys == expected_keys
-
-    comparison = actual.join(expected, on=group_columns)
-    assert comparison.select(
-        (pl.col("len") == pl.col("expected_len")).all()
-    ).item(), comparison
-
-
-def test_predictions_real(predictions):
-    for model_name, model_predictions in predictions.items():
-        if "categorical" not in model_name or "multitarget" in model_name:
-            if "multitarget" not in model_name:
-                assert np.all(
-                    [
-                        v > -10.0 and v < 10.0
-                        for v in model_predictions[
-                            TARGET_VARIABLE_DICT["real"]
-                        ].to_numpy()
-                    ]
-                )
-            else:
-                assert np.all(
-                    [
-                        v > -10.0 and v < 10.0
-                        for v in model_predictions["supReal3"].to_numpy()
-                    ]
-                ), model_predictions
-
-
-def test_predictions_cat(predictions):
-    valid_values = [str(x) for x in np.arange(100, 130)] + [
-        "[unknown]",
-        "[other]",
-        "[mask]",
-    ]
-    for model_name, model_predictions in predictions.items():
-        if "categorical" in model_name or "multitarget" in model_name:
-            assert np.all(
-                [
-                    v in valid_values
-                    for v in model_predictions[
-                        TARGET_VARIABLE_DICT["categorical"]
-                    ].to_numpy()
-                ]
-            ), model_predictions
-
-            if "multitarget" in model_name:
-                admssible_vals = [str(x) for x in np.arange(0, 10)] + [
-                    "[unknown]",
-                    "[other]",
-                    "[mask]",
-                ]
-                assert np.all(
-                    [
-                        v in admssible_vals
-                        for v in model_predictions["supCat1"].to_numpy()
-                    ]
-                ), model_predictions
-
-            if "inf" in model_name:
-                prediction_length = 3
-                n_test_rows = model_predictions.height
-                baseline_preds = predictions["model-categorical-1-best-3"]
-                n_baseline_rows = baseline_preds.height
-
-                assert n_test_rows == n_baseline_rows * prediction_length, (
-                    f"Expected {n_baseline_rows * prediction_length} rows for prediction_length={prediction_length}, "
-                    f"but found {n_test_rows} rows."
-                )
-
-                baseline_rows_per_seq = (
-                    baseline_preds.group_by("sequenceId").len().height
-                )
-                test_rows_per_seq_groups = model_predictions.group_by(
-                    "sequenceId"
-                ).len()
-
-                assert baseline_rows_per_seq == test_rows_per_seq_groups.height
-                assert (
-                    test_rows_per_seq_groups["len"] == prediction_length
-                ).all(), (
-                    f"Test should have {prediction_length} predictions per sequence"
-                )
-
-
-def test_probabilities(probabilities):
-    for model_name, model_probabilities in probabilities.items():
-        if "itemId" in model_name:
-            assert model_probabilities.shape[1] == 33
-        elif "supCat1" in model_name:
-            assert model_probabilities.shape[1] == 13
-
-        np.testing.assert_almost_equal(
-            model_probabilities.sum_horizontal(),
-            np.ones(model_probabilities.shape[0]),
-            decimal=5,
-        )
-
-
-def test_bert_generative_predictions_default_to_context_length(
-    bert_predictions, project_root
-):
-    metadata = _categorical_metadata(project_root)
-    valid_values = {str(v) for v in metadata["id_maps"]["itemId"].keys()}.union(
-        {"[unknown]", "[other]"}
-    )
-
-    assert bert_predictions.height > 0
-    assert set(bert_predictions["itemId"].to_list()).issubset(valid_values)
-
-    rows_per_sequence = bert_predictions.group_by("sequenceId").len()
-    expected_rows_per_sequence = _expected_bert_valid_counts(
-        project_root, ["sequenceId"]
-    )
-    _assert_counts_match_expected(
-        rows_per_sequence, expected_rows_per_sequence, ["sequenceId"]
-    )
-
-
-def test_bert_probabilities(bert_predictions, bert_probabilities, project_root):
-    metadata = _categorical_metadata(project_root)
-    expected_class_count = metadata["n_classes"]["itemId"]
-
-    assert bert_probabilities.height == bert_predictions.height
-    assert bert_probabilities.shape[1] == expected_class_count
-    np.testing.assert_almost_equal(
-        bert_probabilities.sum_horizontal(),
-        np.ones(bert_probabilities.shape[0]),
-        decimal=5,
-    )
-
-
-def test_multi_pred(predictions):
-    multitarget_models = [name for name in predictions.keys() if "multitarget" in name]
-
-    for model_name in multitarget_models:
-        preds = predictions[model_name]
-
-        assert preds.shape[0] > 0, f"{model_name} has no predictions"
-        assert preds.shape[1] == 5, f"{model_name} should have 5 columns"
-
-        admssible_vals = [str(x) for x in np.arange(0, 10)] + [
-            "[unknown]",
-            "[other]",
-            "[mask]",
-        ]
-
-        assert np.all(
-            [v in admssible_vals for v in preds["supCat1"]]
-        ), f"Invalid supCat1 values in {model_name}"
-        assert np.all(preds["supReal3"].to_numpy() > -4.0) and np.all(
-            preds["supReal3"].to_numpy() < 4.0
-        ), f"supReal3 out of bounds in {model_name}"
-
-
-def test_embeddings(embeddings):
-    for model_name, model_embeddings in embeddings.items():
-        if "categorical-1" in model_name:
-            assert model_embeddings.shape[0] == 10
-            assert model_embeddings.shape[1] == 203
-            assert np.abs(model_embeddings[:, 1:].to_numpy().mean()) < 0.3
-        if "categorical-3" in model_name:
-            assert model_embeddings.shape[0] == 30
-            assert model_embeddings.shape[1] == 203
-            assert np.abs(model_embeddings[:, 1:].to_numpy().mean()) < 0.3
-
-
-def test_bert_embeddings(bert_predictions, bert_embeddings, project_root):
-    expected_embedding_cols = [str(i) for i in range(BERT_EMBEDDING_DIM)]
-    expected_columns = ["sequenceId", "subsequenceId", "itemPosition"]
-
-    assert bert_embeddings.height == bert_predictions.height
-    assert bert_embeddings.shape[1] == len(expected_columns) + BERT_EMBEDDING_DIM
-    assert all(col in bert_embeddings.columns for col in expected_columns)
-    assert all(col in bert_embeddings.columns for col in expected_embedding_cols)
-
-    embedding_values = bert_embeddings.select(expected_embedding_cols).to_numpy()
-    assert np.isfinite(embedding_values).all()
-
-    rows_per_subsequence = bert_embeddings.group_by(
-        ["sequenceId", "subsequenceId"]
-    ).len()
-    expected_rows_per_subsequence = _expected_bert_valid_counts(
-        project_root, ["sequenceId", "subsequenceId"]
-    )
-    _assert_counts_match_expected(
-        rows_per_subsequence,
-        expected_rows_per_subsequence,
-        ["sequenceId", "subsequenceId"],
-    )
-
-
-def test_predictions_item_position(predictions, project_root):
-    """Check itemPosition values against source preprocessing metadata."""
-    for model_name, preds_df in predictions.items():
-        expected_positions = _expected_prediction_positions(project_root, model_name)
-        actual_pairs = list(preds_df.select(["sequenceId", "itemPosition"]).iter_rows())
-        expected_pairs = list(expected_positions.iter_rows())
-
-        duplicate_pairs = [
-            pair for pair, count in Counter(actual_pairs).items() if count > 1
-        ]
-        assert duplicate_pairs == [], (
-            f"Model '{model_name}': Found duplicate sequence-position pairs:\n"
-            f"{duplicate_pairs[:10]}"
-        )
-
-        assert Counter(actual_pairs) == Counter(
-            expected_pairs
-        ), f"Model '{model_name}': itemPosition values do not match source metadata."
-
-
-def test_embeddings_subsequence_id(embeddings):
-    """Check subsequenceId increments within each sequence."""
-    for model_name, embeds_df in embeddings.items():
-        for sequence_id, sequence_df in embeds_df.group_by(
-            "sequenceId", maintain_order=True
-        ):
-            subsequence_ids = sorted(
-                sequence_df.get_column("subsequenceId").unique().to_list()
-            )
-            expected_ids = list(range(len(subsequence_ids)))
-            assert subsequence_ids == expected_ids, (
-                f"Model '{model_name}', sequenceId {sequence_id}: expected "
-                f"subsequenceIds {expected_ids}, found {subsequence_ids}."
-            )
diff --git a/tests/integration/test_make.py b/tests/integration/test_make.py
deleted file mode 100644
index 88d1f366..00000000
--- a/tests/integration/test_make.py
+++ /dev/null
@@ -1,184 +0,0 @@
-import os
-import shutil
-import sys
-import time
-
-import pytest
-import yaml
-from conftest import reformat_parameter
-
-test_project_name = os.path.join("tests", "sequifier-make-test-project")
-
-
-@pytest.fixture
-def setup_for_test_make():
-    if os.path.exists(test_project_name):
-        shutil.rmtree(test_project_name)
-        time.sleep(1)
-    os.system(f"sequifier make {test_project_name}")
-
-
-@pytest.fixture
-def config_strings(setup_for_test_make):
-    def load_config_string(path):
-        with open(path, "r") as f:
-            config_string = f.read()
-
-        return config_string
-
-    config_strings = {
-        config_name: load_config_string(
-            os.path.join(test_project_name, "configs", f"{config_name}.yaml")
-        )
-        for config_name in ["preprocess", "train", "infer"]
-    }
-    return config_strings
-
-
-@pytest.fixture
-def adapt_configs(config_strings):
-    preprocess_config_path = os.path.join(
-        test_project_name, "configs", "preprocess.yaml"
-    )
-    with open(preprocess_config_path, "r") as f:
-        preprocess_config_string = f.read()
-
-        preprocess_config_string = (
-            preprocess_config_string.replace(
-                "project_root: .", f"project_root: {test_project_name}"
-            )
-            .replace(
-                "data_path: PLEASE FILL",
-                "data_path: tests/resources/source_data/test-data-categorical-1.csv",
-            )
-            .replace(
-                "selected_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "selected_columns: "
-            )
-            .replace("input_columns: [EXAMPLE_INPUT_COLUMN_NAME]", "input_columns: ")
-            .replace("stored_context_width: 49", "stored_context_width: 11")
-            .replace("max_rows: null", "max_rows: null\nn_cores: 1")
-        )
-
-    with open(preprocess_config_path, "w") as f:
-        f.write(preprocess_config_string)
-
-    train_config_path = os.path.join(test_project_name, "configs", "train.yaml")
-    with open(train_config_path, "r") as f:
-        train_config_string = f.read()
-
-        train_config_string = (
-            train_config_string.replace(
-                "project_root: .", f"project_root: {test_project_name}"
-            )
-            .replace("model_name: PLEASE FILL", "model_name: default")
-            .replace(
-                "metadata_config_path: PLEASE FILL",
-                f"metadata_config_path: {test_project_name}/configs/metadata_configs/test-data-categorical-1.json",
-            )
-            .replace(
-                "export_generative_model: PLEASE FILL", "export_generative_model: true"
-            )
-            .replace(
-                "export_embedding_model: PLEASE FILL", "export_embedding_model: false"
-            )
-            .replace("[EXAMPLE_INPUT_COLUMN_NAME]", "[itemId]")
-            .replace("[EXAMPLE_TARGET_COLUMN_NAME]", "[itemId]")
-            .replace("EXAMPLE_TARGET_COLUMN_NAME: real", "itemId: categorical")
-            .replace("EXAMPLE_INPUT_COLUMN_NAME:", "itemId: 128")
-            .replace("EXAMPLE_TARGET_COLUMN_NAME: MSELoss", "itemId: CrossEntropyLoss")
-            .replace("epochs: 10", "epochs: 3")
-            .replace("device: cuda", "device: cpu")
-            .replace("context_length: 48", "context_length: 10")
-            .replace("total_steps: PLEASE FILL", "total_steps: 10000")
-        )
-
-    with open(train_config_path, "w") as f:
-        f.write(train_config_string)
-
-    infer_config_path = os.path.join(test_project_name, "configs", "infer.yaml")
-    with open(infer_config_path, "r") as f:
-        infer_config_string = f.read()
-
-        infer_config_string = (
-            infer_config_string.replace(
-                "project_root: .", f"project_root: {test_project_name}"
-            )
-            .replace("model_type: PLEASE FILL", "model_type: generative")
-            .replace(
-                "metadata_config_path: PLEASE FILL",
-                f"metadata_config_path: {test_project_name}/configs/metadata_configs/test-data-categorical-1.json",
-            )
-            .replace(
-                "model_path: PLEASE FILL",
-                f"model_path: {test_project_name}/models/sequifier-default-best-3.onnx",
-            )
-            .replace(
-                "data_path: PLEASE FILL",
-                f"data_path: {test_project_name}/data/test-data-categorical-1-split2.parquet",
-            )
-            .replace("[EXAMPLE_INPUT_COLUMN_NAME]", "[itemId]")
-            .replace("[EXAMPLE_TARGET_COLUMN_NAME]", "[itemId]")
-            .replace("context_length: 48", "context_length: 10")
-            .replace("autoregression: true", "autoregression: false")
-            .replace("EXAMPLE_TARGET_COLUMN_NAME: real", "itemId: categorical")
-        )
-
-    with open(infer_config_path, "w") as f:
-        f.write(infer_config_string)
-
-    if sys.platform.startswith("win"):
-        for config_path in [
-            preprocess_config_path,
-            train_config_path,
-            infer_config_path,
-        ]:
-            with open(config_path, "r") as f:
-                config = yaml.safe_load(f)
-
-            config_formatted = {
-                attr: reformat_parameter(attr, param, "linux->local")
-                for attr, param in config.items()
-            }
-
-            with open(config_path, "w") as f:
-                yaml.dump(
-                    config_formatted, f, default_flow_style=False, sort_keys=False
-                )
-
-
-def test_make(adapt_configs):
-    return_code = os.system(
-        f"sequifier preprocess --config-path {test_project_name}/configs/preprocess.yaml"
-    )
-    if return_code == 0:
-        return_code = os.system(
-            f"sequifier train --config-path {test_project_name}/configs/train.yaml"
-        )
-        if return_code == 0:
-            return_code = os.system(
-                f"sequifier infer --config-path {test_project_name}/configs/infer.yaml"
-            )
-            assert (
-                return_code == 0
-            ), f"Inference for 'sequifier infer --config-path {test_project_name}/configs/infer.yaml' was unsuccessful"
-        else:
-            assert False, f"Training for 'sequifier train --config-path {test_project_name}/configs/train.yaml' was unsuccessful"
-    else:
-        assert False, f"Preprocessing for 'sequifier preprocess --config-path {test_project_name}/configs/preprocess.yaml' was unsuccessful"
-
-    # clean up, only if tests didn't fail
-    shutil.rmtree(test_project_name)
-
-
-def test_config_folder(config_strings):
-    from sequifier.make import (
-        infer_config_string,
-        preprocess_config_string,
-        train_config_string,
-    )
-
-    assert config_strings["preprocess"].strip() == preprocess_config_string.strip()
-
-    assert config_strings["train"].strip() == train_config_string.strip()
-
-    assert config_strings["infer"].strip() == infer_config_string.strip()
diff --git a/tests/integration/test_onnx_export.py b/tests/integration/test_onnx_export.py
deleted file mode 100644
index c7f2bd72..00000000
--- a/tests/integration/test_onnx_export.py
+++ /dev/null
@@ -1,247 +0,0 @@
-import os
-
-import numpy as np
-import onnxruntime
-import torch
-
-from sequifier.config.train_config import ModelSpecModel, TrainingSpecModel, TrainModel
-from sequifier.helpers import ModelWindowView, StoredWindowLayout
-from sequifier.train import TransformerModel
-
-
-def test_bert_onnx_export_accepts_attention_valid_mask(tmp_path):
-    project_root = str(tmp_path)
-    (tmp_path / "logs").mkdir()
-
-    context_length = 4
-    inference_batch_size = 2
-    config = TrainModel(
-        project_root=project_root,
-        model_name="bert-onnx-mask",
-        metadata_config_path="metadata.json",
-        training_data_path="data/train.pt",
-        validation_data_path="data/val.pt",
-        input_columns=["cat_col", "real_col"],
-        target_columns=["cat_col", "real_col"],
-        target_column_types={"cat_col": "categorical", "real_col": "real"},
-        column_types={"cat_col": "int64", "real_col": "float64"},
-        categorical_columns=["cat_col"],
-        real_columns=["real_col"],
-        id_maps={"cat_col": {"a": 3, "b": 4, "c": 5}},
-        n_classes={"cat_col": 6},
-        storage_layout=StoredWindowLayout(
-            stored_context_width=context_length + 1,
-            max_target_offset=1,
-            version=2,
-        ),
-        window_view=ModelWindowView(
-            context_length=context_length,
-            objective="bert",
-            target_offset=0,
-        ),
-        inference_batch_size=inference_batch_size,
-        seed=42,
-        export_generative_model=True,
-        export_embedding_model=False,
-        export_onnx=True,
-        export_pt=False,
-        model_spec=ModelSpecModel(
-            initial_embedding_dim=8,
-            dim_model=8,
-            n_head=2,
-            dim_feedforward=8,
-            num_layers=1,
-            prediction_length=context_length,
-            feature_embedding_dims={"cat_col": 7, "real_col": 1},
-            activation_fn="relu",
-            normalization="layer_norm",
-            positional_encoding="learned",
-            attention_type="mha",
-            norm_first=False,
-            n_kv_heads=2,
-        ),
-        training_spec=TrainingSpecModel(
-            training_objective="bert",
-            bert_spec={
-                "masking_probability": 0.5,
-                "replacement_distribution": {
-                    "masked": 1.0,
-                    "random": 0.0,
-                    "identical": 0.0,
-                },
-                "span_masking": {"type": "GeometricDistribution", "p": 1.0},
-            },
-            device="cpu",
-            epochs=1,
-            save_interval_epochs=1,
-            batch_size=2,
-            learning_rate=0.001,
-            criterion={"cat_col": "CrossEntropyLoss", "real_col": "MSELoss"},
-            optimizer={"name": "Adam"},
-            scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
-            loss_weights={"cat_col": 1.0, "real_col": 1.0},
-            torch_compile="none",
-            layer_autocast=False,
-        ),
-    )
-    model = TransformerModel(config)
-    model.eval()
-
-    model._export_model(model, "best", 1)
-    export_path = os.path.join(
-        project_root, "models", "sequifier-bert-onnx-mask-best-1.onnx"
-    )
-
-    session = onnxruntime.InferenceSession(
-        export_path, providers=["CPUExecutionProvider"]
-    )
-    input_names = [session_input.name for session_input in session.get_inputs()]
-
-    assert "attention_valid_mask" in input_names
-
-    ort_inputs = {
-        "cat_col_in": np.array([[3, 4, 5, 3], [0, 0, 3, 4]], dtype=np.int64),
-        "real_col_in": np.array(
-            [[0.1, 0.2, 0.3, 0.4], [0.0, 0.0, 0.5, 0.6]], dtype=np.float32
-        ),
-        "attention_valid_mask": np.array(
-            [[True, True, True, True], [False, False, True, True]], dtype=np.bool_
-        ),
-    }
-    outputs = session.run(None, ort_inputs)
-
-    assert len(outputs) == 2
-    assert outputs[0].shape[1] == inference_batch_size
-
-
-def test_onnx_export_preserves_feature_name_order(tmp_path):
-    project_root = str(tmp_path)
-    (tmp_path / "logs").mkdir()
-
-    context_length = 4
-    inference_batch_size = 2
-    config = TrainModel(
-        project_root=project_root,
-        model_name="onnx-feature-order",
-        metadata_config_path="metadata.json",
-        training_data_path="data/train.pt",
-        validation_data_path="data/val.pt",
-        input_columns=["cat2", "cat10"],
-        target_columns=["cat2"],
-        target_column_types={"cat2": "categorical"},
-        column_types={"cat2": "int64", "cat10": "int64"},
-        categorical_columns=["cat2", "cat10"],
-        real_columns=[],
-        id_maps={
-            "cat2": {"a": 3, "b": 4, "c": 5, "d": 6},
-            "cat10": {"x": 3, "y": 4, "z": 5},
-        },
-        n_classes={"cat2": 7, "cat10": 6},
-        storage_layout=StoredWindowLayout(
-            stored_context_width=context_length + 1,
-            max_target_offset=1,
-            version=2,
-        ),
-        window_view=ModelWindowView(
-            context_length=context_length,
-            objective="causal",
-            target_offset=1,
-        ),
-        inference_batch_size=inference_batch_size,
-        seed=42,
-        export_generative_model=True,
-        export_embedding_model=False,
-        export_onnx=True,
-        export_pt=False,
-        model_spec=ModelSpecModel(
-            initial_embedding_dim=8,
-            dim_model=8,
-            n_head=2,
-            dim_feedforward=8,
-            num_layers=1,
-            prediction_length=1,
-            feature_embedding_dims={"cat2": 4, "cat10": 4},
-            activation_fn="relu",
-            normalization="layer_norm",
-            positional_encoding="learned",
-            attention_type="mha",
-            norm_first=False,
-            n_kv_heads=2,
-        ),
-        training_spec=TrainingSpecModel(
-            training_objective="causal",
-            device="cpu",
-            epochs=1,
-            save_interval_epochs=1,
-            batch_size=2,
-            learning_rate=0.001,
-            criterion={"cat2": "CrossEntropyLoss"},
-            optimizer={"name": "Adam"},
-            scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
-            loss_weights={"cat2": 1.0},
-            torch_compile="none",
-            layer_autocast=False,
-        ),
-    )
-    model = TransformerModel(config)
-
-    model._export_model(model, "best", 1)
-    export_path = os.path.join(
-        project_root, "models", "sequifier-onnx-feature-order-best-1.onnx"
-    )
-
-    session = onnxruntime.InferenceSession(
-        export_path, providers=["CPUExecutionProvider"]
-    )
-    input_names = [session_input.name for session_input in session.get_inputs()]
-
-    assert input_names == ["cat2_in", "cat10_in", "attention_valid_mask"]
-
-    ort_inputs = {
-        "cat2_in": np.array([[3, 4, 5, 6], [6, 5, 4, 3]], dtype=np.int64),
-        "cat10_in": np.array([[3, 4, 5, 3], [5, 4, 3, 5]], dtype=np.int64),
-        "attention_valid_mask": np.ones(
-            (inference_batch_size, context_length), dtype=np.bool_
-        ),
-    }
-    outputs = session.run(None, ort_inputs)
-
-    assert outputs[0].shape == (1, inference_batch_size, config.n_classes["cat2"])
-
-    def assert_onnx_matches_torch(onnx_inputs):
-        torch_inputs_for_case = {
-            "cat2": torch.tensor(onnx_inputs["cat2_in"], dtype=torch.long),
-            "cat10": torch.tensor(onnx_inputs["cat10_in"], dtype=torch.long),
-        }
-        torch_metadata_for_case = {
-            "attention_valid_mask": torch.tensor(
-                onnx_inputs["attention_valid_mask"], dtype=torch.bool
-            )
-        }
-        with torch.no_grad():
-            torch_outputs = model(torch_inputs_for_case, torch_metadata_for_case)
-
-        onnx_outputs = session.run(None, onnx_inputs)
-        np.testing.assert_allclose(
-            onnx_outputs[0],
-            torch_outputs["cat2"].detach().numpy(),
-            rtol=1e-4,
-            atol=1e-5,
-        )
-        return onnx_outputs[0]
-
-    base_outputs = assert_onnx_matches_torch(ort_inputs)
-
-    cat2_changed_inputs = {
-        **ort_inputs,
-        "cat2_in": np.array([[6, 6, 5, 4], [3, 3, 4, 5]], dtype=np.int64),
-    }
-    cat2_changed_outputs = assert_onnx_matches_torch(cat2_changed_inputs)
-    assert not np.allclose(base_outputs, cat2_changed_outputs, rtol=1e-5, atol=1e-6)
-
-    cat10_changed_inputs = {
-        **ort_inputs,
-        "cat10_in": np.array([[5, 5, 4, 3], [3, 3, 5, 4]], dtype=np.int64),
-    }
-    cat10_changed_outputs = assert_onnx_matches_torch(cat10_changed_inputs)
-    assert not np.allclose(base_outputs, cat10_changed_outputs, rtol=1e-5, atol=1e-6)
diff --git a/tests/integration/test_preprocessing.py b/tests/integration/test_preprocessing.py
deleted file mode 100644
index a214e7ef..00000000
--- a/tests/integration/test_preprocessing.py
+++ /dev/null
@@ -1,545 +0,0 @@
-import json
-import os
-
-import numpy as np
-import polars as pl
-import pytest
-import torch
-from conftest import run_and_log
-
-
-@pytest.fixture()
-def metadata_configs(run_preprocessing, project_root):
-    metadata_configs = {}
-    for data_number in [1, 3, 5, 50]:
-        for variant in ["categorical", "real"]:
-            file_name = f"test-data-{variant}-{data_number}.json"
-            with open(
-                os.path.join(project_root, "configs", "metadata_configs", file_name),
-                "r",
-            ) as f:
-                dd_conf = json.loads(f.read())
-            metadata_configs[file_name] = dd_conf
-    return metadata_configs
-
-
-def test_metadata_config(metadata_configs):
-    expected_metadata_keys = [
-        "n_classes",
-        "id_maps",
-        "special_token_ids",
-        "split_paths",
-        "column_types",
-        "selected_columns_statistics",
-        "stored_context_width",
-        "max_target_offset",
-        "stored_window_layout_version",
-    ]
-
-    for file_name, metadata_config in metadata_configs.items():
-        print(f"Verifying metadata_config for: {file_name}")
-        assert list(metadata_config.keys()) == expected_metadata_keys
-        assert metadata_config["stored_window_layout_version"] == 2
-        assert metadata_config["special_token_ids"] == {
-            "[unknown]": 0,
-            "[other]": 1,
-            "[mask]": 2,
-        }
-
-        assert metadata_config["split_paths"][0].endswith(
-            "split0.parquet"
-        ) or metadata_config["split_paths"][0].endswith("split0")
-
-        if "itemId" in metadata_config["n_classes"]:
-            assert len(metadata_config["id_maps"]["itemId"]) == 30
-            assert metadata_config["n_classes"]["itemId"] == 33
-
-            id_map_keys = np.array(
-                sorted(list(metadata_config["id_maps"]["itemId"].keys()))
-            )
-            # assert False, np.array([str(x) for x in range(100, 130)])
-            assert np.all(id_map_keys == np.array([str(x) for x in range(100, 130)]))
-
-        for col in metadata_config["id_maps"].keys():
-            id_map_values = np.array(
-                sorted(list(metadata_config["id_maps"][col].values()))
-            )
-            # assert False, id_map_values
-            assert np.all(
-                id_map_values == np.arange(3, len(id_map_values) + 3)
-            ), id_map_values
-
-        if "itemValue" in metadata_config["selected_columns_statistics"]:
-            assert "std" in metadata_config["selected_columns_statistics"]["itemValue"]
-            assert "mean" in metadata_config["selected_columns_statistics"]["itemValue"]
-
-
-def load_parquet_folder_outputs(path):
-    """Sorted frame from Parquet chunks."""
-    # Polars natively supports reading all matching files via glob patterns
-    data = pl.read_parquet(os.path.join(path, "*.parquet"))
-
-    other_cols = [
-        col
-        for col in data.columns
-        if col
-        not in [
-            "sequenceId",
-            "subsequenceId",
-            "startItemPosition",
-            "leftPadLength",
-            "inputCol",
-        ]
-    ]
-
-    return data[
-        [
-            "sequenceId",
-            "subsequenceId",
-            "startItemPosition",
-            "leftPadLength",
-            "inputCol",
-        ]
-        + other_cols
-    ].sort(["sequenceId", "subsequenceId", "startItemPosition", "inputCol"])
-
-
-def load_pt_outputs(path):
-    contents = []
-    for root, _, files in os.walk(path):
-        for file in sorted(list(files)):
-            if file.endswith("pt"):
-                (
-                    sequences,
-                    sequence_id,
-                    subsequence_id,
-                    start_item_position,
-                    *maybe_left_pad_length,
-                ) = torch.load(os.path.join(root, file))
-                left_pad_length = (
-                    maybe_left_pad_length[0]
-                    if maybe_left_pad_length
-                    else torch.zeros_like(sequence_id)
-                )
-                sequences2 = {}
-                for col, vals in sequences.items():
-                    vals2 = vals.numpy()
-
-                    for offset in range(vals2.shape[1] - 1, -1, -1):
-                        sequences2[str(offset)] = np.concatenate(
-                            [sequences2.get(str(offset), []), vals2[:, -(offset + 1)]],
-                            axis=0,
-                        )
-
-                    sequences2["sequenceId"] = np.concatenate(
-                        [
-                            sequences2.get("sequenceId", []),
-                            sequence_id.numpy().astype(int),
-                        ],
-                        axis=0,
-                    )
-                    sequences2["inputCol"] = np.concatenate(
-                        [
-                            sequences2.get("inputCol", []),
-                            np.repeat(col, vals2.shape[0]),
-                        ],
-                        axis=0,
-                    )
-                    sequences2["subsequenceId"] = np.concatenate(
-                        [sequences2.get("subsequenceId", []), subsequence_id],
-                        axis=0,
-                    )
-                    sequences2["startItemPosition"] = np.concatenate(
-                        [
-                            sequences2.get("startItemPosition", []),
-                            start_item_position,
-                        ]
-                    )
-                    sequences2["leftPadLength"] = np.concatenate(
-                        [
-                            sequences2.get("leftPadLength", []),
-                            left_pad_length,
-                        ]
-                    )
-
-                content = pl.DataFrame(sequences2)
-                contents.append(content)
-
-    assert len(contents) > 0, f"no files found for {path}"
-    data = pl.concat(contents, how="vertical")
-    other_cols = [
-        col
-        for col in data.columns
-        if col
-        not in [
-            "sequenceId",
-            "subsequenceId",
-            "startItemPosition",
-            "leftPadLength",
-            "inputCol",
-        ]
-    ]
-    return data[
-        [
-            "sequenceId",
-            "subsequenceId",
-            "startItemPosition",
-            "leftPadLength",
-            "inputCol",
-        ]
-        + other_cols
-    ].sort(["sequenceId", "subsequenceId", "startItemPosition", "inputCol"])
-
-
-def read_preprocessing_outputs(path, variant):
-    if variant == "real":
-        return pl.read_parquet(f"{path}.parquet")
-    elif variant == "categorical":
-        if os.path.isdir(path) and any(
-            f.endswith(".parquet") for f in os.listdir(path)
-        ):
-            return load_parquet_folder_outputs(path)
-        return load_pt_outputs(path)
-    else:
-        raise ValueError(f"Invalid variant: {variant}")
-
-
-@pytest.fixture()
-def data_splits(project_root, split_groups):
-    data_split_values = {
-        f"{j}-{variant}": [
-            read_preprocessing_outputs(
-                os.path.join(project_root, "data", f"test-data-{variant}-{j}-split{i}"),
-                variant,
-            )
-            for i in range(split_groups[variant])
-        ]
-        for variant in ["categorical", "real"]
-        for j in [1, 3, 5, 50]
-    }
-
-    return data_split_values
-
-
-def test_preprocessed_data_real(data_splits):
-    for j in [1, 3, 5, 50]:
-        name = f"{j}-real"
-        assert len(data_splits[name]) == 2
-
-        for i, data in enumerate(data_splits[name]):
-            number_expected_columns = 14
-            assert data.shape[1] == (
-                number_expected_columns
-            ), f"{name = } - {i = }: {data.shape = } - {data.columns = }"
-            for sequenceId, group in data.group_by("sequenceId"):
-                # offset by j in either direction as that is the number of columns in the input
-                # data, thus an offset by 1 'observation' requires an offset by j values
-                assert np.all((group["1"].to_numpy()[:-j] == group["2"].to_numpy()[j:]))
-                assert np.all((group["5"].to_numpy()[:-j] == group["6"].to_numpy()[j:]))
-
-
-def test_preprocessed_data_categorical(data_splits):
-    for j in [1, 3, 5, 50]:
-        name = f"{j}-categorical"
-        assert len(data_splits[name]) == 3
-
-        for i, data in enumerate(data_splits[name]):
-            number_expected_columns = 14
-            assert data.shape[1] == (
-                number_expected_columns
-            ), f"{name = } - {i = }: {data.shape = } - {data.columns = }"
-
-            for sequenceId, group in data.group_by("sequenceId"):
-                # offset by j in either direction as that is the number of columns in the input
-                # data, thus an offset by 1 'observation' requires an offset by j values
-                assert np.all(
-                    np.abs(group["1"].to_numpy()[:-j] - group["2"].to_numpy()[j:])
-                    < 0.0001
-                ), f'{list(group["1"].to_numpy()[:-j]) = } != {list(group["2"].to_numpy()[j:]) = }'
-                assert np.all(
-                    np.abs(group["5"].to_numpy()[:-j] - group["6"].to_numpy()[j:])
-                    < 0.0001
-                ), f'{list(group["5"].to_numpy()[:-j]) = } != {list(group["6"].to_numpy()[j:]) = }'
-
-
-def test_preprocessed_data_categorical_lookahead_0(run_preprocessing, project_root):
-    metadata_path = os.path.join(
-        project_root,
-        "configs",
-        "metadata_configs",
-        "test-data-categorical-1-lookahead-0.json",
-    )
-    with open(metadata_path, "r") as f:
-        metadata_config = json.load(f)
-
-    assert metadata_config["max_target_offset"] == 0
-    assert metadata_config["stored_context_width"] == 8
-
-    for split in range(3):
-        data = read_preprocessing_outputs(
-            os.path.join(
-                project_root,
-                "data",
-                f"test-data-categorical-1-lookahead-0-split{split}",
-            ),
-            "categorical",
-        )
-
-        assert data is not None
-        assert data.shape[1] == 13
-        assert "8" not in data.columns
-        assert set(str(i) for i in range(8)).issubset(set(data.columns))
-
-        for sequence_id, group in data.group_by("sequenceId"):
-            assert np.all(
-                np.abs(group["1"].to_numpy()[:-1] - group["2"].to_numpy()[1:]) < 0.0001
-            ), (
-                f"{sequence_id = }: {list(group['1'].to_numpy()[:-1]) = } "
-                f"!= {list(group['2'].to_numpy()[1:]) = }"
-            )
-            assert np.all(
-                np.abs(group["5"].to_numpy()[:-1] - group["6"].to_numpy()[1:]) < 0.0001
-            ), (
-                f"{sequence_id = }: {list(group['5'].to_numpy()[:-1]) = } "
-                f"!= {list(group['6'].to_numpy()[1:]) = }"
-            )
-
-
-def unnest(list_var):
-    return [x for y in list_var for x in y]
-
-
-def test_preprocessed_data_multi_file(run_preprocessing):
-    for split in range(3):
-        file_list = []
-        for root, _, files in os.walk(
-            os.path.join(
-                "tests",
-                "project_folder",
-                "data",
-                f"test-data-categorical-multi-file-split{split}",
-            )
-        ):
-            for file in files:
-                file_list.append(file)
-
-        file_list = sorted(file_list)
-
-        expected_file_list = sorted(
-            ["metadata.json"]
-            + unnest(
-                [
-                    [
-                        f"test-data-categorical-multi-file-{source_file}-0-split{split}-0-{str(seq_id).zfill(2)}.pt"
-                    ]
-                    for source_file in range(3)
-                    for seq_id in range(13)
-                ]
-            )
-        )
-        assert len(file_list) == len(
-            expected_file_list
-        ), f"{file_list = }, {expected_file_list = }"
-        assert np.all(
-            np.array(file_list) == np.array(expected_file_list)
-        ), f"for split: {split}:\n{set(file_list).difference(set(expected_file_list))} not found\n{set(expected_file_list).difference(set(file_list))} extra"
-
-
-def test_preprocessed_data_exact(run_preprocessing):
-    parquet_out_path = os.path.join(
-        "tests",
-        "project_folder",
-        "data",
-        "test-data-categorical-3-equal-split0.parquet",
-    )
-    pt_out_path = os.path.join(
-        "tests", "project_folder", "data", "test-data-categorical-3-equal-split0"
-    )
-
-    parquet_output = pl.read_parquet(parquet_out_path)
-    pt_output = load_pt_outputs(pt_out_path)
-
-    assert np.all(
-        parquet_output.to_numpy()[:, [0, 1, 2, 3, 5, 6, 7, 8, 9]]
-        == pt_output.to_numpy()[:, [0, 1, 2, 3, 5, 6, 7, 8, 9]].astype(int)
-    ), f"{np.sum(parquet_output.to_numpy()[:,[0,1,2,3,5,6,7,8,9]] == pt_output.to_numpy()[:,[0,1,2,3,5,6,7,8,9]].astype(int)) = }"
-
-    assert np.all(
-        parquet_output["sequenceId"].to_numpy() == np.repeat(np.arange(10), 9)
-    )
-
-    assert np.all(
-        parquet_output["subsequenceId"].to_numpy()
-        == np.tile(np.repeat(np.arange(3), 3), 10)
-    )
-
-
-def test_preprocessing_interrupted(run_preprocessing, metadata_configs):
-    with open(
-        os.path.join(
-            "tests",
-            "project_folder",
-            "configs",
-            "metadata_configs",
-            "test-data-categorical-1-interrupted.json",
-        ),
-        "r",
-    ) as f:
-        interrupted_metadata_config = json.loads(f.read())
-    baseline_metadata_config = metadata_configs["test-data-categorical-1.json"]
-
-    interrupted_metadata_config_adapted = interrupted_metadata_config
-    interrupted_metadata_config_adapted["split_paths"] = [
-        path.replace("categorical-1-interrupted", "categorical-1")
-        for path in interrupted_metadata_config_adapted["split_paths"]
-    ]
-
-    assert str(interrupted_metadata_config_adapted) == str(
-        baseline_metadata_config
-    ), f"{interrupted_metadata_config_adapted = } != {baseline_metadata_config = }"
-
-    baseline_output = {
-        split: load_pt_outputs(
-            os.path.join(
-                "tests",
-                "project_folder",
-                "data",
-                f"test-data-categorical-1-split{split}",
-            )
-        )
-        for split in range(3)
-    }
-    interrupted_output = {
-        split: load_pt_outputs(
-            os.path.join(
-                "tests",
-                "project_folder",
-                "data",
-                f"test-data-categorical-1-interrupted-split{split}",
-            )
-        )
-        for split in range(3)
-    }
-
-    for split in range(3):
-        assert np.all(
-            interrupted_output[split].to_numpy() == baseline_output[split].to_numpy()
-        ), f"interrupted output != baseline output for split {split}"
-
-
-def test_preprocessing_from_precomputed_stats(
-    run_preprocessing,
-    project_root,
-    preprocessing_config_path_cat_precomputed_stats,
-    preprocessing_config_path_cat_precomputed_stats_negative,
-):
-    """Compare precomputed-stat and standard preprocessing outputs."""
-
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_cat_precomputed_stats}"
-    )
-
-    run_and_log(
-        f"sequifier preprocess --config-path {preprocessing_config_path_cat_precomputed_stats_negative}"
-    )
-
-    preprocessing_output_path = os.path.join(
-        "tests",
-        "project_folder",
-        "data",
-        "test-data-categorical-precomputed-stats-split0.parquet",
-    )
-
-    preprocessing_output = pl.read_parquet(preprocessing_output_path)
-
-    preprocessing_output_negative_path = os.path.join(
-        "tests",
-        "project_folder",
-        "data",
-        "test-data-categorical-precomputed-stats-negative-split0.parquet",
-    )
-
-    preprocessing_output_negative = pl.read_parquet(preprocessing_output_negative_path)
-
-    assert np.all(
-        preprocessing_output[
-            ["sequenceId", "subsequenceId", "startItemPosition"]
-        ].to_numpy()
-        == preprocessing_output_negative[
-            ["sequenceId", "subsequenceId", "startItemPosition"]
-        ].to_numpy()
-    )
-
-    metadata_cols = [
-        "sequenceId",
-        "subsequenceId",
-        "startItemPosition",
-        "leftPadLength",
-        "inputCol",
-    ]
-
-    assert (
-        np.mean(
-            np.isclose(
-                preprocessing_output.filter(pl.col("inputCol") == "supReal2")
-                .drop(metadata_cols)
-                .to_numpy()
-                * 2,
-                preprocessing_output_negative.filter(pl.col("inputCol") == "supReal2")
-                .drop(metadata_cols)
-                .to_numpy(),
-                rtol=1e-01,
-            )
-        )
-        > 0.5
-    )
-
-    assert (
-        np.mean(
-            np.isclose(
-                preprocessing_output.filter(pl.col("inputCol") == "supReal3")
-                .drop(metadata_cols)
-                .to_numpy()
-                / 2,
-                preprocessing_output_negative.filter(pl.col("inputCol") == "supReal3")
-                .drop(metadata_cols)
-                .to_numpy(),
-                rtol=1e-01,
-            )
-        )
-        > 0.5
-    )
-
-    assert (
-        np.mean(
-            preprocessing_output.filter(pl.col("inputCol") == "itemId")
-            .drop(metadata_cols)
-            .to_numpy()
-            != preprocessing_output_negative.filter(pl.col("inputCol") == "itemId")
-            .drop(metadata_cols)
-            .to_numpy()
-        )
-        > 0.1
-    )
-    assert (
-        np.mean(
-            preprocessing_output.filter(pl.col("inputCol") == "supCat1")
-            .drop(metadata_cols)
-            .to_numpy()
-            != preprocessing_output_negative.filter(pl.col("inputCol") == "supCat1")
-            .drop(metadata_cols)
-            .to_numpy()
-        )
-        > 0.1
-    )
-    assert (
-        np.mean(
-            preprocessing_output.filter(pl.col("inputCol") == "supCat4")
-            .drop(metadata_cols)
-            .to_numpy()
-            != preprocessing_output_negative.filter(pl.col("inputCol") == "supCat4")
-            .drop(metadata_cols)
-            .to_numpy()
-        )
-        > 0.1
-    )
diff --git a/tests/integration/test_training.py b/tests/integration/test_training.py
deleted file mode 100644
index ca8a4ab6..00000000
--- a/tests/integration/test_training.py
+++ /dev/null
@@ -1,148 +0,0 @@
-import os
-
-
-def test_checkpoint_files_exists(
-    run_training, run_training_from_checkpoint, run_hp_search, project_root
-):
-    found_items = sorted(os.listdir(os.path.join(project_root, "checkpoints")))
-    expected_items = sorted(
-        [
-            f"model-{model_type}-{j}-epoch-{i}.pt"
-            for model_type in ["categorical", "real"]
-            for j in [1, 3, 5, 50]
-            for i in range(1, 4)
-        ]
-        + [
-            f"model-real-{j}-epoch-{i}-batch-1.pt"
-            for j in [1, 3, 5, 50]
-            for i in range(1, 4)
-        ]
-        + [f"model-real-{j}-latest.pt" for j in [1, 3, 5, 50]]
-        + [f"model-real-1-from-epoch-checkpoint-epoch-{i}.pt" for i in range(1, 4)]
-        + [
-            f"model-categorical-3-from-mid-epoch-checkpoint-epoch-{i}.pt"
-            for i in range(1, 4)
-        ]
-        + [
-            f"model-categorical-{j}-inf-size-epoch-{i}.pt"
-            for j in [1, 3]
-            for i in range(1, 4)
-        ]
-        + ["model-categorical-bert-epoch-1.pt"]
-        + ["model-real-bert-epoch-1.pt"]
-        + [f"model-categorical-multitarget-5-epoch-{i}.pt" for i in range(1, 4)]
-        + [f"model-categorical-multitarget-5-eager-epoch-{i}.pt" for i in range(1, 4)]
-        + [f"model-categorical-distributed-epoch-{i}.pt" for i in range(1, 4)]
-        + [
-            f"model-categorical-distributed-lazy-parquet-epoch-{i}.pt"
-            for i in range(1, 4)
-        ]
-        + [f"model-categorical-lazy-epoch-{i}.pt" for i in range(1, 4)]
-    )
-
-    assert set(found_items) == set(expected_items), {
-        "missing": sorted(set(expected_items) - set(found_items)),
-        "unexpected": sorted(set(found_items) - set(expected_items)),
-    }
-
-
-def test_model_files_exists(
-    run_training, run_hp_search, run_training_from_checkpoint, project_root
-):
-    model_type_formats = {"categorical": ["onnx", "pt"], "real": ["onnx", "pt"]}
-    found_items = sorted(
-        [
-            f
-            for f in os.listdir(os.path.join(project_root, "models"))
-            if not f.endswith(".onnx.data")
-        ]
-    )
-
-    expected_items = sorted(
-        [
-            f"sequifier-model-{model_type2}-{j}-{kind}{model_type}-3.{model_type_format}"
-            for model_type2 in ["categorical", "real"]
-            for model_type in ["", "-embedding"]
-            for model_type_format in model_type_formats[model_type2]
-            for j in [1, 3, 5, 50]
-            for kind in ["best", "last"]
-        ]
-        + [
-            f"sequifier-model-categorical-3-from-mid-epoch-checkpoint-{kind}{model_type}-3.{model_type_format}"
-            for model_type in ["", "-embedding"]
-            for kind in ["best", "last"]
-            for model_type_format in ["pt", "onnx"]
-        ]
-        + [
-            f"sequifier-model-real-1-from-epoch-checkpoint-{kind}-embedding-3.{model_type_format}"
-            for kind in ["best", "last"]
-            for model_type_format in ["pt", "onnx"]
-        ]
-        + [
-            "sequifier-model-categorical-multitarget-5-best-3.onnx",
-            "sequifier-model-categorical-multitarget-5-last-3.onnx",
-            "sequifier-model-categorical-multitarget-5-best-embedding-3.onnx",
-            "sequifier-model-categorical-multitarget-5-last-embedding-3.onnx",
-            "sequifier-model-categorical-multitarget-5-eager-best-3.onnx",
-            "sequifier-model-categorical-multitarget-5-eager-last-3.onnx",
-            "sequifier-model-categorical-multitarget-5-eager-best-embedding-3.onnx",
-            "sequifier-model-categorical-multitarget-5-eager-last-embedding-3.onnx",
-            "sequifier-model-real-1-best-3-autoregression.pt",
-            "sequifier-model-categorical-1-best-3-autoregression.onnx",
-            "sequifier-model-categorical-1-inf-size-best-3.onnx",
-            "sequifier-model-categorical-1-inf-size-last-3.onnx",
-            "sequifier-model-categorical-1-inf-size-best-embedding-3.onnx",
-            "sequifier-model-categorical-1-inf-size-last-embedding-3.onnx",
-            "sequifier-model-categorical-3-inf-size-best-3.pt",
-            "sequifier-model-categorical-3-inf-size-last-3.pt",
-            "sequifier-model-categorical-3-inf-size-best-embedding-3.pt",
-            "sequifier-model-categorical-3-inf-size-last-embedding-3.pt",
-            "sequifier-model-categorical-bert-best-1.pt",
-            "sequifier-model-categorical-bert-last-1.pt",
-            "sequifier-model-categorical-bert-best-embedding-1.pt",
-            "sequifier-model-categorical-bert-last-embedding-1.pt",
-            "sequifier-model-real-bert-best-1.pt",
-            "sequifier-model-real-bert-last-1.pt",
-            "sequifier-model-categorical-distributed-best-3.pt",
-            "sequifier-model-categorical-distributed-best-3.onnx",
-            "sequifier-model-categorical-distributed-last-3.pt",
-            "sequifier-model-categorical-distributed-last-3.onnx",
-            "sequifier-model-categorical-distributed-lazy-parquet-best-3.pt",
-            "sequifier-model-categorical-distributed-lazy-parquet-best-3.onnx",
-            "sequifier-model-categorical-distributed-lazy-parquet-last-3.pt",
-            "sequifier-model-categorical-distributed-lazy-parquet-last-3.onnx",
-            "sequifier-model-categorical-lazy-best-3.pt",
-            "sequifier-model-categorical-lazy-best-3.onnx",
-            "sequifier-model-categorical-lazy-last-3.pt",
-            "sequifier-model-categorical-lazy-last-3.onnx",
-        ]
-        + [
-            f"sequifier-test-hp-search-sample-run-{i}-{suffix}-1.pt"
-            for i in range(4)
-            for suffix in ["best", "last"]
-        ]
-        + [
-            f"sequifier-test-hp-search-bert-run-0-{suffix}-1.pt"
-            for suffix in ["best", "last"]
-        ]
-        + [
-            f"sequifier-test-hp-search-grid-run-{i}-{suffix}-2.pt"
-            for i in range(4)
-            for suffix in ["best", "last"]
-        ]
-        + [
-            f"sequifier-test-hp-search-bayesian-run-{i}-{suffix}-1.pt"
-            for i in range(4)
-            for suffix in ["best", "last"]
-        ]
-        + [
-            f"sequifier-test-hp-search-custom-eval-run-{i}-{suffix}-3.onnx"
-            for i in range(4)
-            for suffix in ["best", "last"]
-        ]
-    )
-
-    assert set(found_items) == set(expected_items), {
-        "missing": sorted(set(expected_items) - set(found_items)),
-        "unexpected": sorted(set(found_items) - set(expected_items)),
-    }
diff --git a/tests/integration/test_visualize_training.py b/tests/integration/test_visualize_training.py
deleted file mode 100644
index a619748a..00000000
--- a/tests/integration/test_visualize_training.py
+++ /dev/null
@@ -1,54 +0,0 @@
-import glob
-import os
-
-from conftest import run_and_log
-
-
-def test_visualize_training(
-    run_training, run_training_from_checkpoint, run_hp_search, project_root
-):
-    # Dynamically find all models that were trained and logged
-    log_dir = os.path.join(project_root, "logs")
-    log_files = glob.glob(os.path.join(log_dir, "sequifier-*-rank0-*.txt"))
-
-    models = set()
-    for lf in log_files:
-        filename = os.path.basename(lf)
-        # Extract model name by stripping known prefix and suffix
-        name = filename[len("sequifier-") :]
-        name = name.rsplit("-rank0-", 1)[0]
-        models.add(name)
-
-    single_models = sorted([m for m in models if "categorical" in m or "real" in m])
-    assert len(single_models) > 0, "No single models found in logs directory"
-
-    hp_models_grid = sorted([m for m in models if "hp-search-grid-run" in m])
-    assert len(hp_models_grid) > 0, "No hp grid models found in logs directory"
-
-    for model in single_models:
-        # visualize_training.py expects to find "logs/" relative to the working directory
-        command = f"sequifier visualize-training {model} --project-root {project_root}"
-        run_and_log(command)
-
-        output_path = os.path.join(
-            project_root,
-            "outputs",
-            "visualization",
-            f"{model}-training-visualization.html",
-        )
-        assert os.path.exists(
-            output_path
-        ), f"Visualization output not found for {model}"
-
-    models_str = ",".join(hp_models_grid)
-    command_joint = f"sequifier visualize-training {models_str} --project-root {project_root} --log-scale --bucket-training-batches 5"
-    run_and_log(command_joint)
-
-    output_path_joint = os.path.join(
-        project_root,
-        "outputs",
-        "visualization",
-        "multi-model-training-visualization.html",
-    )
-
-    assert os.path.exists(output_path_joint), "Joint visualization output not found"
diff --git a/tests/resources/source_configs/id_maps/itemId.json b/tests/resources/source_configs/id_maps/itemId.json
deleted file mode 100644
index 77d5f018..00000000
--- a/tests/resources/source_configs/id_maps/itemId.json
+++ /dev/null
@@ -1 +0,0 @@
-{"100": 3, "101": 4, "102": 5, "103": 6, "104": 7, "105": 8, "106": 9, "107": 10, "108": 11, "109": 12, "110": 13, "111": 14, "112": 15, "113": 16, "114": 17, "115": 18, "116": 19, "117": 20, "118": 21, "119": 22, "120": 23, "121": 24, "122": 25, "123": 26, "124": 27, "125": 28, "126": 29, "127": 30, "128": 31, "129": 32}
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
deleted file mode 100644
index 78cac8f9..00000000
--- a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/preprocess-manifest.json
+++ /dev/null
@@ -1,35 +0,0 @@
-{
-    "manifest_version": 1,
-    "preprocessing_config": {
-        "stored_context_width": 9,
-        "max_target_offset": 1,
-        "stored_window_layout_version": 2,
-        "read_format": "csv",
-        "write_format": "pt",
-        "merge_output": false,
-        "selected_columns": null,
-        "data_columns": [
-            "itemId"
-        ],
-        "split_ratios": [
-            0.6,
-            0.2,
-            0.2
-        ],
-        "stride_by_split": [
-            1,
-            1,
-            1
-        ],
-        "max_rows": null,
-        "process_by_file": true,
-        "subsequence_start_mode": "distribute",
-        "mask_column": null,
-        "metadata_config_path": null,
-        "use_precomputed_maps": null
-    },
-    "effective_metadata_digest": {
-        "algorithm": "sha256",
-        "value": "b245092f219b7b0bed5e4e24ec7aa7be6dead1885e61adc00f6395d6164c460c"
-    }
-}
diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-0-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-0-0.pt
deleted file mode 100644
index f796d2f4aa9749df94e9ff7acbcf40d3ebb0154e..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 4166
zcmeHKzi-n(7<C%jq)vd+7TWToP$-a6k|xd%n*mTG8c`XE5I~ApqSP^og`4E;3lszr
z5-NfX78WFskQiZM=*q}Q%Zd;ZEQkT|A8=o6x3;O9+EOJ3zUq3Bb?@H0^Skdp=Xw7&
ziV6m)zk;PEDJ-fuS}EglR48My+E65+EJqVjsg6ZOX*ID}iK<Om!tp5in}U(C<_$TQ
zJx*QQT-@)9+(T5chtNw{oS&^ApMV>RP>YpXSdvw`q=<7ZNv@QzSXUcL=_W&?u1KU9
zAPAL=ELTP3*C6zTM!kAjdq$&eMIRXwN8Lp~>4jzs_3#`<flEBa2l(Ja6y$@v?;+}a
zEC>x*76mLd>MDIvsfdbLIVTC2M)X7hg)j;i(0276+EMe>2zAY0)8<Z~owfFyeXDcs
z(&p@cTm$F;MuP>kTeYAVw1=Pq7!9>ii)fhlEus+(u@|F#1+?D+NTLH8U{nVj1i&E;
zFox0L0y<&=q|mqqII07V0pPd>h+ssNoxo@^mgfV(cTX0U?~;vzmUr~R8EV+A7h=~N
z%6u7@Zi<RZHa3?XT-Oo_5V7>@^8<KU(40j~Q+=*Q=ng5BWvN=97wb5e4O?_%QXt^$
zE%c*{kVA78z`5Urx!YK5fTNt&PtFdOMbNN#`r8N`18B|~($tXKO@LE570+b{EDP6<
z=XSyd(XDmVsNV0<ds^?q4$r;f_UZGv<uh=G?#=47;|$%M*5`yHNwXbJ`_c(FsN?PW
zH;Xg)!8u=+et&&W&JMJVoO8gFa}v4i$a>Dv3}o!1*Z1fZ_y7HC1~UBX);Ex4AizIY
z-aKE~I{z>m`bSgY{y*rRef{ARN1)I)^8HayzE9?|W6r)esvAQ&TwM@wgFryd0b!g|
z<D8iHeRAUMt*z6a`bYF<sByRZzgjaK>i7bZG&o`TMfGau`+J58xm6NtU!JO{NTh<e
zQ-%_;Bvo?RfFUfwuxuik=2(tl<0&SQ;@EgR&2q`iDK?qNq&SXEGc3b`%G#@!KO@8w
z(DJ;mZ55;PHrHV+7{?MG>u@`Xxz0LcAqW#OE=g<!o}J5FQcYX9o_gT=tB$kNnX4x_
zQ)|&rj-tz@tyXPz7V}M;UZWttp1Ehy>M}cpIT>3iK9U^bav^OBQes=ZX3m>diazok
z=!Jr@Q<zhmmEwa(isa@g9`$-#1^E=PZj?E#WGt`yJh^vWZj_$X>>TE#Fk|6ZanG@S
zy2|rCJty_4NBz-Nl92w(=w<z~TF@U59DHR{-*;*Q;f~Am-R&zZL~<#N*T((<PW%YY

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-1-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-1-0.pt
deleted file mode 100644
index 04c99a991c8c7573702cbdd2a036590e35440b53..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3910
zcmd^COK;Oa5O$iTNu2<F(Fe5UO#>yV?IewRpwNg$LPih*h=PD9bxdO6##wt)3ZfDc
zYNehK2RL%%FTmfRXKv-f1^xm%vEABf-PDmPA+geGBWu6?zMYw!ndAh2Cq;!q)UJ`H
zCMhiGI94j+VoWSzsjO<USX5$cOs--{)9N)Wm125LkugI)GE=ZKUVEWrGZWO~gN?Tx
z(N#n(bP;+POG`IP$R}b|6D#pT9m|SN7c^<HE-R%1ma4j{6_z;~bwr~JL4wf9&SF_Y
zegi_^Qmf^A#y1*uF8Ih61_c)UWEE?5)FnhQ3f>hcAs~d-QAh|0zID{SA&RP^NFtWi
zs!re0N|Gj(Zpb305q&9-!Wc#J=!m|Gj#hjXLS6A!jI&vEtkO8MXZy@v<IKK|3PAfY
z8pxyLx&y_bL4pckG}J(?p<%(dhDHp;35-tW(WnEEKw}1A+ytBgz-a?;2BWiibj|@t
zqVopef(f_?fC&RYOhA-f#%MC06M`Z1B>3nJF%-<4Yb?xC!(L+{{#@0Tiny>WX*w}%
zHZ!niNU$K{>z^O*z*#WeMa)q>o<*1uQYb2Nxw<4(aW)fi=$KA|fNL+&uMR>E(_H}X
z-3p7Zu%tpnx!s?<9jr*8VeRVH2s{Ir?iy0mkXIwXtDIu8nSRH`&F6WPu)*jxmIgcg
zw7z$n-}}uK-VzLJZ5KCa9OIeK!UatHL3K9cA|}4o{oHV26AzxNKit~9PGT44ex4g^
z%X4ftGqRWG3<v)2_IHcRR(`YM7yMWLt&Sd^|G0hq2PqFms}*$OXCLoJ$zGWI#aXJY
zI7?(Rtb3ffRddh|!_p5K6ozGA(p)qK=pViPUO7Dd$$v!uG?nyf|Lq#dcy>U72Frf`
zTz=5He$7*1ugc<$;-O2LOe*N>BD6>axs=TWEuk!zPP2&=pXRwVljPVWpJtduf@f!V
zW`<+=6qn=@WCfK^2fw6u{O@aoiq(1A`;fF5GdoM0eP~W_lGyu<*`3qLuLu$2k))}?
zY38!G6mt%)4{dPm>SLPe?A-(Jpz9+~(c{&&yOw4ad!<V`SU!1YG5V5b3VSkkQshX@
z_jr*81!*yzVPMarPKqA#uc93au9?D~>YNnc+N4MvoMI)^-YrNg>@<`;txP-I&6B>#
zqoK?s*UVv03iA$*P466gr>mUM)pf-@dfXq|Cb4=rYgNo*wL%+P2oC;bQa{Zm6KY&e
P2tb2QGHIBC{l?zE4T}d2

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-2-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-2-0.pt
deleted file mode 100644
index 97a48607d7c3264932d1c7e277f44b6e72a0ec5a..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3782
zcmd5<O>fgk7@jn36Wju&?Y0zX>GB!2B()tk&PtRO5d<NxL<k^NT%zF2CKhhev8P>u
z7HK6^BqWgFvd6vejT1=x3vR5$jT0Akj~toT_Ea9zl-dG`kwzmY^UU+i>-WnwBRz;v
zHj6$RI$A&!yENZw(q`UlQrzx2mf5uPO5W;H>^SRd6u0v3nr%@zujb`N*jZS+V^=Gu
z(XHK$KL>L630fIu^cKadS6U=x(w<{>3XOGY*{;-Z@a=WWZZ#<Gx;>|{p-N;Rms`m&
zgv)j|+nA(%i1cf(+rIAqCXvCF6x$+`p_MdS&9!whY|K-Vxn>|^(8xX{StDzt9+Ewe
zOtWX(*rZmk>q=Lh7Itv!qGeKvNN4I~gp$!Z+3VgX`#Px(qwb_T{@DuI-|@~I+dA`r
zf9Ck34nQX;nXHo^Tmhw$DTW%PWZFY*l7mKSlg#*tLzEn@lOqD4NM?P&TmU!<fO#MA
zBPGY`<hTIP$O#{CG60+cz-b?lql78@iIRmv&B$cOUX^G6V1|NO>+1{W(ZRUBQ24Fq
ztTt(513NA=Y_&4EtxG5%;^lvDAHd6k*<Zv3G!|P#;318sZMD0rxJ#>*Q9(yZ0|CGM
zPTmhNa+v)El;d0B_Fam5;8FeV&tiw#7&L4yzng(DfZ1Qe5}J-{1c=H-xmuYJE*^yE
zUd9Gv#$V7>u=3}*;F%=}Z%hX3M6mLGad9I7J}C}j8V2Y8l1|RYahw+{|M=hOP0#ms
zjDaOIo5=D?wKB7v<^31H^>aP^dAIjTisN>O%g1q?cegsDxIO;sW1gPL7H)ia`|a$n
z?r49Aco_Xw(D%;^vOLfBz}zV^mlOF%v0C}LzkhtGiZV&wfYb?t--Fv5^h@LD**yCX
z=GWJ+eJA=$=v-X;Z`G!9;ulbqz{0QpZr=pU!|ZS2OK2pn!h)B)UF=w_g57IEi_)-K
z)k=m7R#aVAic4j^tm?9+Dq2~Wm2y!n=_OfHbhW5ynp`T%pz<<)^CZU<!mJr7&sBKm
z7w!Xe_P;n1_JL87aGgIdFpM%WF-al~MwBaD(k}>H&k}Hb)<;C?!qvMh(2cWDjCr-K
zt|7`2u2f3`%U|(X{JtPc5l+S;#n@2ni+B{Q#S?ukoKHoHr>yyjX)BLHjZ%bDo=73H
z4~HZa#obf<pOAvJqN1U~X{98*?&s`-E#@tPWEbTKCxuG_N1fH^m>k=utC}%9d^R|G
qE}h>Z;XN5&?V!p07&HeA2VZ%pwa<q@jjI_$&>*u+I)Y%ozxNa8ferEi

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-3-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split0-3-0.pt
deleted file mode 100644
index 1099e78dc545785794a1956f1ba6107a976bf323..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3910
zcmdT{OK;Oa5Ox~!s9PZPK`D>&rh$@R=i!6{uLvR`BN2j-0zyQoV-ka#WbG{!L?tBD
zN<DJvkpl<*0fH+BegJUg210P+#uawF-e9Lqse>xPN~^Ww{pOpUnVp#wWdA0P3x~NC
zBg0K`ST%91TEUf=Qo(AiX=qACkBKp@fmOqpZDF+<Gh4cbg;+8soPeG2)?GcHo8)e+
zZ#?dc&LgfAB=j0qXD(KePr*$?smIH+Skq0uY^c+-nqDnqwP7}m^1TF)`l8WNh#*X|
zvr<!$pF;Rc%|`7i{f$SPN<OkhK!K8<tV(MZ1?4D4p(`>c2juV~3d>>Hw}|?e6s4)_
zs)Ds<!{jd;Rn<_d=QRcMh(B6H0~ie!(PncVZK?a}gu3ppQ`sWgTDN3I+A_CMnL|r;
zfDU6cQbePs3zb0I2`Yfmn1xzEJ7nJi8mEYz80{*e-7Y{9?V*4P8?YAu`zT;PMhA-M
zpbL;fhbZ8%4LAaTNeUn)AetS+XewTiLt%6?bo~)A6wE@`SUAn?@EQy8+f8Gpg3I?*
z!z6~y=SEfy2@#Zd_5I@$a2Cw&N}S^&o|Uj8q+HRpT4P3S;Cyb-rH7OP1<u?@-}(qS
z%<c+gy<1`W0alw3QQgi@?he*f&|%@!r*XIjFuUuJ=El4l0dD09A)gy|9o)X2V}uRH
z1YIl(z~hkpK5DPK?Un7{Y`^!4WAeb_FxJb*a1jdoYupxF4wDPkhY_F)2WbYzFkLuy
zZe|DY+{@?lcg~X7g}IjJ_VnaAF`pY>&2!X&|GWMF6kp6Qo&JWL4ugI7>NmC@{C8;E
zS99b1msrz(>>S-zQk<;;dcIix@H0x@V6GKs>7L>&na_#c<E&dXr`plf4bSX8v-7a;
zp#PV%Z@+Gw{;7XNe~C+ZwST)t3Z5O1<iWD<p4G1Xxqe;X2E1w(w~B|U8XBpf4=d0j
zm9=U<7h;NviA+XJrn8xBA|s>{Vk(;v1nHzGrNm5H$Y#@0N=itQ09w8|eCx+=|NE>^
zVV$S556Q5gX|Fx@p#|AZ;_Nd{w}r#QAW_C6Nk@az$>nS*&bhdr_Q18GkLjdyb`Mz>
z-4MBo9*@@US~^*rl`idKS@zCC`;txyXEJtE6i7|>co2(%w3zNNaOP1rMTC_5UMLcs
z6wXxVrg+^WMRNTVbK%}jL0VzAp`2+&a=AO(@0r3*a-AH`q;SE-@!mVf>glQ=2ZP6L
p(G&hyn`FR#VykW!+c}<oAULc2`)oIf^zTW)Y8?hhE`|Mc?=P7163PGo

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-0-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-0-0.pt
deleted file mode 100644
index 82d2dd60705a7eb7a007dd9ba4f5a51ad54813eb..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3206
zcmcIm%Wl&^6t$Z+Nu2=cy9HXH6dEW_Z6~hN1*JqZ5;77YRi%hc<T^uQ;U*b-QUs!a
zRJBr<Y*1JH1}ymn)<t5$iWMKghMCxj>^#~yDkEL_HD}Jb*E44(Bl`v^Dj1~NnnW#8
z2$+a1C`e&d1%YD4&{Rcdc~&bUU>LP30tMEr>Kfu%j^&m~&sg=Po=z=N&wFbh2BLeA
z%K0(9hQQ9_0;E+`G1O8lUqhO1GI;}R*EGG5N1$w0jQp;^z=3Eq7r+P;_f(1i`YZ_Z
zq*5+!TJIP*n4@tQ4i4pf_^MWG&@TrO3_O!5nU;fxFenep^dTHRQq_vC0~KkNvdKI(
z3cvt`N1BQl$lS@o5Q5<>ykPFZizT{*sY||+H8u|~mFi<gnq!VwV~!q`2y_g=@hqG$
zT_^!gVib+wR2_8yr)By8&RB@c2wus;t1dts&RT#u8*q&Pu3LZ`2+n8WO&1^m7c9Un
z8*rNd7A-&&L9Fa9f=jWC90+c{KiPVV3q>N+buO$?(_ZI7>_x@cQBZyt7$z=kIyK&L
zO7KL)mp?!D$<HFuUBp8w;#q_}L-LBQ70WxIjMAyFOUFur2zc-c{vN>OB)SVodauIv
zYXmA}Ms>>{uN|ZVqT%5FuNkroNOadAQd3@wKvv~4mrjkj61SHpCOfrF+Fu(6F>60J
zu0*-|ZsYGdeox{w<+FV$=hLZ~PRgwWFlEPfmhKT7Mf4ASTVvPpGl}!pFZ86J=+J)7
z15#%ar|FpJtB&z>YNWl6n5y3XKY!w!_*bZLulYAUDAHDdI72f3{<*ky_WIwTLSC6;
z^~GlbL&FaAMj=~FUMr+ifrb!XkR(1XCZ(hxaS4G>Bqfd$#iS^TaZwWDtN7(ACx}F)
zu%A1L;=e^ABhy51bLV#YxI{j<TA%xPD~Z#m*IWc)EXE^A%LdWP<+S97F0PM#aJB6N
zt#nTJO}gkt@$tc9YE9Q~WpO&C=wkWgoyGF?Rtl$?xhXREpyx4>ItAXM-E+<9Q*MgL
zP=6Fc?-a*<Qp9_wn8sUhKZU|a33s8KwzA^--N)W3Y!hqcaGJt~i(?cYkv*o-*{(9O
w-+#{@dd|l-NkaBr<ErBcM%q_$HpdO)>URra>)ATq4Uq#5cG3_<`dfYf0a7jDMgRZ+

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-1-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-1-0.pt
deleted file mode 100644
index 85a4f29e9a6e7ec61d9b24fbb8e79501b5f0200c..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3206
zcmcImO>YuG7+y*XbU~$GwpCketro3NcKKK?)|OZjvrR;rCg?#zmk~B@f!Q5uG&Y);
zHumVn^iuEs0{t7FHR-{VCy#n`UY3#_D5Oj3n@naG=9%Z2ci(wt(z3Ik<2)X&EhKJ%
zBgi1ZLZ0M<N}ix%MOT%)78HVNnIK)SRS7BtjjE;+f&Lg?gpr}@b1jve=boG{-u4A{
zFqd^vdX=E92L<d<NJUplp<Ioqn!)FEv{_TNLXM!aQPFeTB9Hq5fvlS%3_3DjMA&IU
z__a#8xNiQ&<NmCJ4hiFdtdsUiwT4}?hhX<3nUfu|XAgViLD{i~2lo}FqG?DWYNc%O
z5A_1lQQ^L-5FYclGT2M7FN4n+JNSIbQKHl(XUUvfz!ysOIsMH!hs-&L_e%gBA$T-{
z#|#D~;)@jJAb7lv+Qk#HV;4`Fh)V=t&fqHyAcCh%z;pv}6#z3P;2OcR8GM}qMDd&n
zxX}RI1i-ur2oOw_-6nV;l$PC|^*7%)UeiQDOLwe=C2qoQErgy`^sPL}Z6n>FiA^O(
zPb>)mM11-E;~jh~Xq`o@aDLk&8WobuYihB)g~}wA^f5Z(Q4nzN1^(4X$)R-?kg%VH
z&6fmKprSghpY9!^A<(dU_va*R18AK!#JF+0Nq}9s7)~Wen1wgCCnY;oz#f7aGdqrN
z^q~A{EI6}F-=TF=KGl<QA(fmwNx8WIO4&6!>4q;20s0R!YlqA98QR(F7rW|*l7|2t
ze~#XN1|x>nO~+VId5olzL+yD4ZV2F4k7m*T<By&Zf1De&yZ)vJ`P&i@;bG<9J{32h
zBA|87{3_?Q%N(k2J_G40b)Z*yIAU^YA(eC=3lT&~5+bpLln|wGR1~5KDJ+DgL?RlK
z;_+xACPbyU7?D8bwEXvRfGUKRmL2sHK0djvJ}%M!+34{cZzZw%^b*71qhf55v>Xtv
zTvkh7VYoi@z}0pRw9;AKH^I;iQwy=Vw5IE~vRIum#;|;}&tm#|D}~j}Sc){g=-FII
zoq`_GY^_;+ily)m^hP1JQdq5sr8wx3B650)33>$gk|=tWU=wAvl{oXc5A0JkOsti|
zY6`0i$1uGj+g!%Uc9oW0t~-sXr=7tjiMR1|+-tPJAil!goCd|!pXPk#yLr7EfC~<F
LQZEPN&9VOg5wPLY

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-2-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-2-0.pt
deleted file mode 100644
index f57e27bb9375269831dd4d90df4354345270ad48..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3206
zcmcImO-~a+7~WE7OQ~9*2&f?92UXnC?KcMz8fh?VVl5$QyfkGywi}mjXLpLxNHj5!
zaP(rh=wI-tf5RJzCLW9jPyPkx?Y7O%mNsmIZ!+0_%rnn3@4oZAlNFtV9Ov<H9g*Y~
zIf8V;mx`pwOGSdpswPWCg%^0aN|2`2YXp^ey{5=S;3Ir^31)_B&y`GifqQzocsmf>
z!(84)>1Bd;9+t2}BB~};LWMe!72R9V&~{x`N(F+dx~dg+BVIfZ4CdVwq0^bgGQv&+
z;(er6%bUh;UObq0&?#X&ly}mnRI6i`=poqsMC3$==sCb1aaeR5;Ne3_QWXVBL{_W1
z_pw$&8Y(@ICE~^2+c`W!urG(t>3jHm#ZjTu6=%g*Tfi48jWzx4HAjs##||q19Va-D
z!xK7#ir|YB<sf*nf!fDYqGKOV8;DB;U(Vqx3?PbU48Uv?a1{V^2H+aO^ErH-0mSeP
z18}nmxCMX(0}v#bD!W7QVkj%RJ)3X7Z@s38f|2c73oG1|-C77eQ?;EUDeNLmr-{v^
z11FY*03yEp{qYW77L4v9);Pay5zPuI6cxE#-9c57N&6Tbi5LjD{{sIRpyV*R3rN||
z!uCsoRH&$K`=@7zC<rv{-}^NU#{fom4RLPLZW7>BE`>Acac1Mq<4MU*6_6PSI7<*C
z#)B$#dp^1Y$E)-^j9$uT`cf`r($gm?H#R^idqy|iU>^{m|MTyS^;P;A#@XwS^wi(d
z2RZ`B?gcPn7`=2%^p(eGCOz7jM@rQQ;Fo{SqyNVrJ0t!C7qGkkwg>q;5)k#m&cA&s
zZ^4^@(LM9)wCZhk#8Bh#=}42Q1HCFjiz&#ZOxoQNB1Do&AsSC5Q;}pi77=2pWLQWf
z;_+xKyc}Cj2;pcn9!n)aWnyk45TpuWWJO1#bXuL;?BhxLKN~hxxB9r1#O%{641<q~
zu}NZS5LPa;C9g4D@B85D=mS<dv-_qPx-n{bHn-MxeJhLEDdP;w2m36BuUje1X2w!v
z=|#`xMj8~fMYFYL_9>RaKhz&Z#7bedB9`K)Pm1X2DW+%(?k7?7D#0eoY%2-obsyQM
zXquRn!)yxc496JlaW=PcvR!3Gm+NkG=~*Y=CTZOPT2F-*7{u3RbEUY(ryD-wcjJ0D
P1Q#6Yq$3>6H|G8WXH(tC

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-3-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split1-3-0.pt
deleted file mode 100644
index 39785a599fcaa5222a628869d56b160b8c164bf9..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3334
zcmcJS&u`LT7{?22g8@_advk8ipF=lTTVQ1eXGR=}sfmbLLQIsfQlYT}-u7iivn3>E
zOLkKa9*qb81QU&3yn6KR&C?PS<H5V{Lp#=%ft1WQc`0q*@B4Y4_s8?RP4TWCit>1<
zrovJ46c$w+%I9z{B;>GIC@YeXlS52MDq&Gksue8eLuy5qFcXS|=tXE5t~{2Li8<<i
zXQk#0?jS1DNAMCBH*e*UL%?N4D2B6DEXk@jtB4y_NzP}nSW?SMc8m2QXE2y?6NO4z
z<_aQmX%g@4a;b1nd*($w83$=Xquz{*T!l&%_3;6W+;@43ck-S+<l#NMV-NN33qo0z
zMFC6YlIp#q<V8iy-;@OGMc%7v<ip6HMg!^&I#qNO3A*SiYJD^4bg|xNpwZ``*5}ZE
z5v0Qyjik}2YErRijHsL#jn`GXXo7d_qDf702BWDoI%^U{(6lC)(FNx~Fslj9V{{>n
zE}8^UbV(Cj)&*BUFsBKE7!k{^VKg63@ovxf!Pke|WT2p?TCRl^YQpYX2tO(-n>n1_
z5*3vUY%(!&?2=%>h!4NMKZnVJ+S-U!Dq!0PeT8ImvQ#K-iY1&(_)T^!MZtg@Ptbvr
zz@fG_AZ~vYHlAX!3@fUY|77oASp*w)um6~YZ2+~k4KZrm?jXRfT%?nUVUy$f_9U>5
z%BbF|U%yUuq-t(7rav0dui*Ln^@V(m+(T{W`E*yFGs(o{ah_`fK#*-qE58TZs+wO~
z{XQH&@c-lXpIVJbpq{+{Y+L^!HbNmupkBb92fvqB1LO{BJ9`GYnvX~_(SI}_8ujQG
z`0M=VpZ|{^Jt6)Q<+tm9BNloZ0}%0o$KT!**8d%U)+nD{Vz|EfRZ)>h1bmi*6ql9q
z$%OmR0EXo_CK8Ktah9W_EEA1$G#!gaW0BZ$jANq_I?mEjjs`0~ePL2TGF4DhyaNnw
zoE(gKgo78R(&ahA3Nhx16%)WujIjx^Bq%G(m=ab^te0J|n(~kpXN>-F6K;rXQJbzc
zvW}HvjI=Qm<+Xi^mRGC@V=yx#UXU1L(~&xY91+ZG%^067h~DlHtQBDlL1x4($<N(*
z5$PQ9wo3##%9sbr7*>`{(>+U$&o;f#gO`<I3<_%|#w+`b<HJ>o@9SI8yPk1{8W5lU
k@9-+?Ni42^%@7@YB~b7DT0`x4*$W3O5=ng&wAb4H1s#gu6951J

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-0-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-0-0.pt
deleted file mode 100644
index 28153cd611d254af22afd12fa3516870c5e3af52..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3206
zcmcImT~E_c815KsW1Bi*2&kYU$`m)+eRW)ju*ecq6XyuYVoa9O9k<4^?X+hQi9{0v
z3700uaOL07zu~neUU=n|f5G#%Woh>T>w<4`(hkmZp67jg&hwtU;2vfepO5JXo>^oF
z(nz!{5-}=?1Xb#)EQv~#i^?^ERJGY4s2tTAicI2BHp(u+%vj^KlFKeIFZzr3Ly>*V
z6g`w)CTROf8M`D>SEXvK)FiT^`AaI=YRXEvL{LqutEHW|9}h($MK48Ybf#EA*lj@k
zPwTbHhVeH)9xl4*6c&#Z-SjCnn%E=w2=+b~7{Mj@4zW)d6<mjS^hlEGih?8}*K3;p
znOa6FDnFJb;>Z4b1sotaSil#weSER%s#5BzyK1b>;Y(G0%}{I2%f^~xM^%816C5t!
z3C)Iz<0};9B6w0q9pEX!b%3W0#8rZ?74UT%Ac1ELz^n<l0f3tZ;1<Di1$^5ENaA?|
zaK{AP1;By<h!9Mb-6wc4mKVIfjdwpb-_k<C$oJd}E6kMBy%2j<SGPq{+Ci#D3!BS^
zyKV^%M11}G^8<V=7`;WTF(Jny%pFn^6}eK|Mm3Vl25mahNf7X87ylWe<S=>*$T+XU
z)*FKAu%mkAPtOif5NJ4f_<I_T0gT=nQp}{&BEYF!VsqJXTjJ*Nq-3WC95V=N5_*Kq
zC#Zw~<y)UB`BnN3<22<n11aZn+39Y|jRYuV*EmZzY&8hbzcHljuF`iH=dVBBmwwP8
z|C&Q!#4t|NF)>gb6S?f=&N@;mJ^nv`@|^h7OxS7uEe{HH6d>V;%)fuBY@U7o*O`D*
z=9qr?G^EPZf%ZgbF(tX2%X&|QaB-gJ5~&QIiSukS&LuNE%O;ZPWu9MVS&sh7q|(cY
z6sYuVzTAt@uLUD7xIl2Lb6b6!hkw{&z&_qiV)f}28$pnYaY)kEAlkXCmb_-;`Zxer
zM;~aXv$}7_MmI*U4-Qjnxqdr~)hSapme0;v3}0`ju$q~jB2PWsVIn#OZPE6<X7wpM
zMQCI&ig^DN#{*I%`lpzp&25lE(W`{LP*z(>+dlWPa|+YM+BvMIux{fRqdm}J8r|(G
uFL*o;%%x}D(H2R-d^>q6<|W~q*?cKZf13{)6OHTLh<<elP%nk~#@v6E<KD3V

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-1-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-1-0.pt
deleted file mode 100644
index 10e60e18fc2c03add4e64ab9173d41081b418dba..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3206
zcmcImO-~a+7~WE7OQ~AE1Qh&0lq$BgyY03fL};W5S);Xtr0St5+p*oaw9M{+(MU8Q
z5IK4=oV@!B^e1@rtO*Ab6HgvJ>b%{h*;!g(8+?<=e$6w_^Ul8W%w#2ZC&&4G+<{1P
zQyf7$373kb7?z6!m1~+J7u9eqtW*iow6#@&N@0CfRR~XCMyH`?Wc8VvNl$W*j~8z`
zLK~RNdnmm^(8@vyyJS++<VvKlMif=|7c{iIrl_R?K~=q`71nt_?g)kQUW(9Z&te&2
zw*m3rt5wU3#ydam%)4lpDDKL;=~Z4`!yd^;u=k<FNiNB^jeSzL<l4sFJF;9;RU{Lo
zR@MCvv=Y)#>8>IZKlWeG;U0nmIebdrz^5y&3Z<^NE5_I{e5O(#GuRlj*BG;JrvlJ^
zf(LSVP-jp)K1)$9f`{s;Ej%o_w(y97I7jgL93Eu=0v<B}7firK09-NvmkA!v;VTRv
zjwcMjRTFRx0FwqFL@-r$gW#!1R`U85Uw>PAMJEbcwq-5MaKlb(A@a1QtrSUN9cen9
z*i3q$X-ULD#HZiiH{oYNYb|1q3py5IR!E_!D&^`5s*+4Pz~~U;AmGk({HueKLu)M{
z<$M;FUl3G-ifWZVT{}cYpkeFwj}ce~&{}Iqa6?Wf0aoR7G?VUU5;vD8B|9_%=GRg4
z3hE$&#)T^N`ks7w6qu#Y&<;~R)|T>ECOy(jxsd>+Y#K-D?lMuJe`+&*f0jN&J9+*5
zq57fZAwb8Uy|*91h@l;(W3a7x6f)`FgYyX75WuhOjidj^A3q^}ksENDf5U@<2PQ!9
zL*`#Ul$VZP|MOgrQ|3s0@##oYr~|z$!WL6dN}05GUr3BkCSyV(l}z!;Xq=D5Q^{zQ
zj|zeqjf-MT5QW4okxz)Aa&O`1-w^#*(6W*X1UGhWtB)t)0~2lT<8~6OPtPy}0V>8J
ziEV?hb6G8Uj^TRO2G@aoz)oj%-xNdFM~@E<Q){@soyF>u35Mmpa~8wb?G#osV=1!q
zpyx1=ItAUL*;=#u6iX4Lw%QH_Z>O+Y5lgY#CWUZ(ieb9RwUY%sO0W}UwH1;1-Mh{y
zOcS$nSWRJ`;pn4BWQS=qx2vq=@!T|rzTgfwNP5h>{i{j~4B|7K&1q0v{cghCjE&=6
PSN#alL%kIG8-4!)sKDgU

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-2-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-2-0.pt
deleted file mode 100644
index b333a65fdf3a51a5ecdf7fae56cc51b1bb62eed2..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3206
zcmcIm&rcIU6y8#3%Ys^<NKrvWlq$CDZhvwRp^+wLO{^s(jhCiu$9BWgc6O&2jYJaz
zk)s#G$$vvn`Zstp(Zqv^i6{R86X$i?W~ZeM+u%!HwhQyk_s!ck-%MI|4sx8w!*xWQ
zTjU5bNU)G6`Jj>~s94ceC9eg=pjswK*K1XR3PGc)sYD7&L176-hN|0ID!ITtK3TjP
z2<%}l>!S23K|A*g*rAY$u9QN#8c{Wa&*^BZrfP*8L1m+&=XNC?4+H{PH$@n9WWI>7
z(}eI3D&^va`8OU9W*u~hfQPb9`c$el?2<hMyC2D%?2tVN*dq_ijsrY=s3;XpLkdwV
zWrKgH7m$t$_f&=On7^69BLt6T@EKzdpDj5`l)B_BnRAQyT&X^%w>9UOIp_Fc37``M
z`!aaaU{Df1Pf-qnr|PJE?3W$;c-ll<AoyYiUt$1ZJYxc88-U9IxMBjX5<Hi|*BC$q
z&zpej4ZsZmESP`*!Bp8Tf)_(++3neQ`EBzBEfi$BYcH&De!IO8dQ#DM@+7y5bb}T)
zmGm9k5+aEB@aOw$_*jtLMXYgN+aekrlFMsqvAl!IB$XUxbi^Vc;O<lWdw`Ndb{CMa
zUxlq_1XZA;x~-q?9ik!7uz%<0G;9ON?i!-pl-(l0u3QpQ$q8oRjqORvP86_*AV$mw
z$2Xp!eEaX4U#s*TvX}CizLblp<n(dM%>_`(p3zOWw}Arv{@UTERr(Hj`ue4w`k~_>
zK!@+EdjX6XvX_p@zUmlGCC57J2;30Bul|@r|BpX%O8ha-XLtQA5At>tAk4$czq~JQ
z!k2*TUio#}{Wed;P<`_mNLQ%?J<r1tlT!<+q`N6Zl;UwQ98JU%Qe22gVk8k4#8@IG
zh?1}zTaHJCs1S=p>o)mWSe^<{g^+34QE#2*$!+!VIQ>5xHdQyz@pclcPp>cxqg0Gd
zlC}e)oy%&;YYf-hKDav0fp$8p`z9E=acUtpm)3Iqb{4BsMj4iO_E}6{Z>O-D8B39-
z7d@K`sZ-G2V|&f&Q!Isds6Ps+ox*BGEX7fu6ycLo_~{YcPod~lf-RKQR$|QOKC(~I
zFtK(Ht0}BA9OLwgY;zgM+f`b2xo$V6o^=LWB+a`(^QqAWgZKz{a~c#^|202qzMI#(
PA-LdBCmrEnygBwC@fYCy

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-3-0.pt b/tests/resources/source_data/test-data-categorical-1-interrupted-temp/test-data-categorical-1-interrupted-split2-3-0.pt
deleted file mode 100644
index f7edbe8a70928d848d9044808b78f01afb7bb039..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3334
zcmcInO=}ZD7~UlPNR!mYephX+pN-Z{v%6_GIaq110a?KqDFi`7x>=LOZPMADRun4*
zX`wgu;88?8d-dSWn@8`02M<zE#DCzt*=^j}HZdFO3&SMsJoCKId_3<YBl$WQ#_wnL
zg~&`ZSTQhLEZ_nw7qC*Q>Z)AOSdLXISkd*hRjd?QV^vcz&kAg821Z6#?`!GQ6mz?^
z*z`s=5R>a7^eR?Xt`(6-##LP|NAqh~)r>%1SC-dQt(eD3#i;7}bv}T+kw~taAPh3H
zP*RZ3gaocvE2UfJ-vX#3=OII4s59pyy}Y`Hx}-2h-8Ush@=E?~<d^)CXB+kG$Z}QF
z6d9}4iV?V>7ZqJ8UR7lrK!FQc6vQZ$MZLxbI#TwO33b_5Hs|J0U%57CxIX7mbI$&q
zGC&6~8qA_0gF^9Wn4r8Ejnq(EXjJlSp)nJ245Q;&bb<m1Xxs!$Sb&oNm^1;WFgl$@
zXDC1%oizdHEWmjHOqqZPMkKR~7)?hrQn!C(_tV|SWTBvC8lHuDX4LIjh~BH}D+Qci
zS9F6cY&tb~;E~`UiTA(0JcZ4I);Ng;ChR&1>xASBnp&!?C>5Meg{VB{;*h|V2WZzz
z$e}e(AmM%$mLFoJ3MZ<OesXuPra%r`m%oj{HGtMQhgoLC?IOUfoQb7V1C(&<dJ?jh
z0`3sRu-P#KR^MtjWDO){Q;UA92jx$`zx%pLzC&xKe7r5?Tsk#&kaBYY2xZe~q#Jgr
zm1v~9&jN#fbvYW^B;TPOzJ9)`{ylwr`vUpg`JG${lU$%R%cr-k{SeZrp1=D+D5wDb
zCF#SD|HmIcB>p)j<W_&(3;p#45CWk1^Q+R*zx`*C3A$yB)-J!H=qho*M+GQxd9|2M
zb?+s>@uJ8Hvk5W5i?KM*#S>zT6BBWf7h)Wj<b`Bh6q7;{vJC!w^D#np3tC3<K*IHx
zgT0Q3@DD25d`CD*>~&(EA_$RWT#`5nl#|O|5*8?~=WTH9t3ysYyZa|7x_)v+T}rE0
z9Vd(3X=f>xm+o22y5gj;n;A{<jChPoAvFr}M4-=_-Jcv3o$XQZP71pP(G-)!`P)$;
zv`+E5O$zdqp%=<-D|6I#Pm<@eOBI%RIXUd6ut;&daL;kDU1g-Mu1nU`6F#<160|<;
i^_o@064tL7f`eZP%-fJT(0pEY)=GDfcqxoG$Nm9e>fkj1

diff --git a/tests/resources/source_data/test-data-categorical-1-interrupted.csv b/tests/resources/source_data/test-data-categorical-1-interrupted.csv
deleted file mode 100644
index 9c89f83e..00000000
--- a/tests/resources/source_data/test-data-categorical-1-interrupted.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId
-0,0,112
-0,1,104
-0,2,129
-0,3,121
-0,4,109
-0,5,111
-0,6,121
-0,7,109
-0,8,126
-0,9,112
-0,10,104
-0,11,103
-0,12,119
-0,13,103
-0,14,105
-0,15,121
-0,16,107
-0,17,128
-0,18,128
-0,19,116
-0,20,106
-0,21,103
-0,22,119
-0,23,105
-0,24,100
-1,25,113
-1,26,116
-1,27,105
-1,28,113
-1,29,108
-1,30,118
-1,31,102
-1,32,102
-1,33,121
-1,34,104
-1,35,111
-1,36,110
-1,37,102
-1,38,102
-1,39,120
-1,40,109
-1,41,101
-1,42,111
-1,43,127
-1,44,105
-1,45,115
-1,46,104
-2,47,112
-2,48,124
-2,49,121
-2,50,124
-2,51,129
-2,52,103
-2,53,119
-2,54,118
-2,55,110
-2,56,102
-2,57,121
-2,58,108
-2,59,101
-2,60,102
-2,61,110
-2,62,106
-2,63,102
-2,64,110
-2,65,109
-2,66,115
-3,67,122
-3,68,129
-3,69,101
-3,70,117
-3,71,120
-3,72,127
-3,73,122
-3,74,112
-3,75,127
-3,76,115
-3,77,117
-3,78,102
-3,79,120
-3,80,108
-3,81,126
-3,82,116
-3,83,102
-3,84,121
-3,85,118
-3,86,105
-4,87,113
-4,88,127
-4,89,105
-4,90,104
-4,91,106
-4,92,118
-4,93,112
-4,94,122
-4,95,105
-4,96,121
-4,97,119
-4,98,126
-4,99,120
-4,100,103
-4,101,117
-4,102,114
-5,103,120
-5,104,123
-5,105,127
-5,106,105
-5,107,112
-5,108,108
-5,109,123
-5,110,108
-5,111,129
-5,112,122
-5,113,117
-5,114,122
-5,115,109
-5,116,112
-5,117,126
-5,118,120
-6,119,127
-6,120,119
-6,121,127
-6,122,102
-6,123,113
-6,124,119
-6,125,103
-6,126,126
-6,127,128
-6,128,113
-6,129,111
-6,130,117
-6,131,124
-6,132,100
-6,133,119
-6,134,117
-6,135,103
-6,136,110
-6,137,108
-6,138,125
-6,139,107
-7,140,111
-7,141,112
-7,142,113
-7,143,113
-7,144,113
-7,145,121
-7,146,100
-7,147,129
-7,148,100
-7,149,125
-7,150,118
-7,151,121
-7,152,127
-7,153,121
-7,154,129
-7,155,113
-7,156,110
-8,157,101
-8,158,119
-8,159,103
-8,160,113
-8,161,105
-8,162,111
-8,163,124
-8,164,108
-8,165,118
-8,166,124
-8,167,109
-8,168,121
-8,169,110
-8,170,120
-8,171,114
-8,172,107
-8,173,100
-8,174,111
-8,175,107
-8,176,109
-8,177,114
-8,178,102
-8,179,126
-8,180,115
-9,181,128
-9,182,120
-9,183,103
-9,184,116
-9,185,122
-9,186,100
-9,187,128
-9,188,102
-9,189,122
-9,190,110
-9,191,122
-9,192,107
-9,193,123
-9,194,103
-9,195,128
-9,196,106
-9,197,108
-9,198,119
-9,199,129
diff --git a/tests/resources/source_data/test-data-categorical-1-lookahead-0.csv b/tests/resources/source_data/test-data-categorical-1-lookahead-0.csv
deleted file mode 100644
index 9c89f83e..00000000
--- a/tests/resources/source_data/test-data-categorical-1-lookahead-0.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId
-0,0,112
-0,1,104
-0,2,129
-0,3,121
-0,4,109
-0,5,111
-0,6,121
-0,7,109
-0,8,126
-0,9,112
-0,10,104
-0,11,103
-0,12,119
-0,13,103
-0,14,105
-0,15,121
-0,16,107
-0,17,128
-0,18,128
-0,19,116
-0,20,106
-0,21,103
-0,22,119
-0,23,105
-0,24,100
-1,25,113
-1,26,116
-1,27,105
-1,28,113
-1,29,108
-1,30,118
-1,31,102
-1,32,102
-1,33,121
-1,34,104
-1,35,111
-1,36,110
-1,37,102
-1,38,102
-1,39,120
-1,40,109
-1,41,101
-1,42,111
-1,43,127
-1,44,105
-1,45,115
-1,46,104
-2,47,112
-2,48,124
-2,49,121
-2,50,124
-2,51,129
-2,52,103
-2,53,119
-2,54,118
-2,55,110
-2,56,102
-2,57,121
-2,58,108
-2,59,101
-2,60,102
-2,61,110
-2,62,106
-2,63,102
-2,64,110
-2,65,109
-2,66,115
-3,67,122
-3,68,129
-3,69,101
-3,70,117
-3,71,120
-3,72,127
-3,73,122
-3,74,112
-3,75,127
-3,76,115
-3,77,117
-3,78,102
-3,79,120
-3,80,108
-3,81,126
-3,82,116
-3,83,102
-3,84,121
-3,85,118
-3,86,105
-4,87,113
-4,88,127
-4,89,105
-4,90,104
-4,91,106
-4,92,118
-4,93,112
-4,94,122
-4,95,105
-4,96,121
-4,97,119
-4,98,126
-4,99,120
-4,100,103
-4,101,117
-4,102,114
-5,103,120
-5,104,123
-5,105,127
-5,106,105
-5,107,112
-5,108,108
-5,109,123
-5,110,108
-5,111,129
-5,112,122
-5,113,117
-5,114,122
-5,115,109
-5,116,112
-5,117,126
-5,118,120
-6,119,127
-6,120,119
-6,121,127
-6,122,102
-6,123,113
-6,124,119
-6,125,103
-6,126,126
-6,127,128
-6,128,113
-6,129,111
-6,130,117
-6,131,124
-6,132,100
-6,133,119
-6,134,117
-6,135,103
-6,136,110
-6,137,108
-6,138,125
-6,139,107
-7,140,111
-7,141,112
-7,142,113
-7,143,113
-7,144,113
-7,145,121
-7,146,100
-7,147,129
-7,148,100
-7,149,125
-7,150,118
-7,151,121
-7,152,127
-7,153,121
-7,154,129
-7,155,113
-7,156,110
-8,157,101
-8,158,119
-8,159,103
-8,160,113
-8,161,105
-8,162,111
-8,163,124
-8,164,108
-8,165,118
-8,166,124
-8,167,109
-8,168,121
-8,169,110
-8,170,120
-8,171,114
-8,172,107
-8,173,100
-8,174,111
-8,175,107
-8,176,109
-8,177,114
-8,178,102
-8,179,126
-8,180,115
-9,181,128
-9,182,120
-9,183,103
-9,184,116
-9,185,122
-9,186,100
-9,187,128
-9,188,102
-9,189,122
-9,190,110
-9,191,122
-9,192,107
-9,193,123
-9,194,103
-9,195,128
-9,196,106
-9,197,108
-9,198,119
-9,199,129
diff --git a/tests/resources/source_data/test-data-categorical-1.csv b/tests/resources/source_data/test-data-categorical-1.csv
deleted file mode 100644
index 9c89f83e..00000000
--- a/tests/resources/source_data/test-data-categorical-1.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId
-0,0,112
-0,1,104
-0,2,129
-0,3,121
-0,4,109
-0,5,111
-0,6,121
-0,7,109
-0,8,126
-0,9,112
-0,10,104
-0,11,103
-0,12,119
-0,13,103
-0,14,105
-0,15,121
-0,16,107
-0,17,128
-0,18,128
-0,19,116
-0,20,106
-0,21,103
-0,22,119
-0,23,105
-0,24,100
-1,25,113
-1,26,116
-1,27,105
-1,28,113
-1,29,108
-1,30,118
-1,31,102
-1,32,102
-1,33,121
-1,34,104
-1,35,111
-1,36,110
-1,37,102
-1,38,102
-1,39,120
-1,40,109
-1,41,101
-1,42,111
-1,43,127
-1,44,105
-1,45,115
-1,46,104
-2,47,112
-2,48,124
-2,49,121
-2,50,124
-2,51,129
-2,52,103
-2,53,119
-2,54,118
-2,55,110
-2,56,102
-2,57,121
-2,58,108
-2,59,101
-2,60,102
-2,61,110
-2,62,106
-2,63,102
-2,64,110
-2,65,109
-2,66,115
-3,67,122
-3,68,129
-3,69,101
-3,70,117
-3,71,120
-3,72,127
-3,73,122
-3,74,112
-3,75,127
-3,76,115
-3,77,117
-3,78,102
-3,79,120
-3,80,108
-3,81,126
-3,82,116
-3,83,102
-3,84,121
-3,85,118
-3,86,105
-4,87,113
-4,88,127
-4,89,105
-4,90,104
-4,91,106
-4,92,118
-4,93,112
-4,94,122
-4,95,105
-4,96,121
-4,97,119
-4,98,126
-4,99,120
-4,100,103
-4,101,117
-4,102,114
-5,103,120
-5,104,123
-5,105,127
-5,106,105
-5,107,112
-5,108,108
-5,109,123
-5,110,108
-5,111,129
-5,112,122
-5,113,117
-5,114,122
-5,115,109
-5,116,112
-5,117,126
-5,118,120
-6,119,127
-6,120,119
-6,121,127
-6,122,102
-6,123,113
-6,124,119
-6,125,103
-6,126,126
-6,127,128
-6,128,113
-6,129,111
-6,130,117
-6,131,124
-6,132,100
-6,133,119
-6,134,117
-6,135,103
-6,136,110
-6,137,108
-6,138,125
-6,139,107
-7,140,111
-7,141,112
-7,142,113
-7,143,113
-7,144,113
-7,145,121
-7,146,100
-7,147,129
-7,148,100
-7,149,125
-7,150,118
-7,151,121
-7,152,127
-7,153,121
-7,154,129
-7,155,113
-7,156,110
-8,157,101
-8,158,119
-8,159,103
-8,160,113
-8,161,105
-8,162,111
-8,163,124
-8,164,108
-8,165,118
-8,166,124
-8,167,109
-8,168,121
-8,169,110
-8,170,120
-8,171,114
-8,172,107
-8,173,100
-8,174,111
-8,175,107
-8,176,109
-8,177,114
-8,178,102
-8,179,126
-8,180,115
-9,181,128
-9,182,120
-9,183,103
-9,184,116
-9,185,122
-9,186,100
-9,187,128
-9,188,102
-9,189,122
-9,190,110
-9,191,122
-9,192,107
-9,193,123
-9,194,103
-9,195,128
-9,196,106
-9,197,108
-9,198,119
-9,199,129
diff --git a/tests/resources/source_data/test-data-categorical-3-equal.csv b/tests/resources/source_data/test-data-categorical-3-equal.csv
deleted file mode 100644
index bc370970..00000000
--- a/tests/resources/source_data/test-data-categorical-3-equal.csv
+++ /dev/null
@@ -1,161 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-0,0,112,3,3
-0,1,104,9,8
-0,2,129,7,2
-0,3,121,9,2
-0,4,109,2,1
-0,5,111,6,1
-0,6,121,9,1
-0,7,109,3,7
-0,8,126,4,1
-0,9,112,3,5
-0,10,104,4,4
-0,11,103,9,8
-0,12,119,2,1
-0,13,103,0,8
-0,14,105,7,6
-0,15,121,6,5
-1,0,113,9,8
-1,1,116,1,2
-1,2,105,5,1
-1,3,113,6,1
-1,4,108,1,7
-1,5,118,5,5
-1,6,102,3,6
-1,7,102,0,2
-1,8,121,6,1
-1,9,104,3,1
-1,10,111,1,8
-1,11,110,6,4
-1,12,102,4,2
-1,13,102,6,9
-1,14,120,0,0
-1,15,109,6,1
-2,0,112,3,6
-2,1,124,5,3
-2,2,121,0,6
-2,3,124,8,1
-2,4,129,6,1
-2,5,103,3,0
-2,6,119,6,6
-2,7,118,4,7
-2,8,110,0,2
-2,9,102,3,0
-2,10,121,8,6
-2,11,108,5,0
-2,12,101,6,2
-2,13,102,4,4
-2,14,110,8,9
-2,15,106,9,4
-3,0,122,9,1
-3,1,129,5,6
-3,2,101,0,6
-3,3,117,2,7
-3,4,120,0,3
-3,5,127,8,1
-3,6,122,4,3
-3,7,112,4,8
-3,8,127,7,0
-3,9,115,9,9
-3,10,117,0,2
-3,11,102,5,9
-3,12,120,6,0
-3,13,108,7,3
-3,14,126,8,3
-3,15,116,6,1
-4,0,113,4,9
-4,1,127,3,6
-4,2,105,7,4
-4,3,104,9,2
-4,4,106,9,4
-4,5,118,0,4
-4,6,112,0,8
-4,7,122,0,3
-4,8,105,0,6
-4,9,121,6,2
-4,10,119,9,5
-4,11,126,1,4
-4,12,120,9,9
-4,13,103,6,1
-4,14,117,1,3
-4,15,114,6,0
-5,0,120,4,2
-5,1,123,4,5
-5,2,127,0,0
-5,3,105,4,5
-5,4,112,6,0
-5,5,108,4,7
-5,6,123,8,4
-5,7,108,6,5
-5,8,129,9,1
-5,9,122,0,2
-5,10,117,7,6
-5,11,122,2,2
-5,12,109,3,8
-5,13,112,1,8
-5,14,126,9,0
-5,15,120,9,6
-6,0,127,0,7
-6,1,119,8,4
-6,2,127,4,2
-6,3,102,3,8
-6,4,113,6,0
-6,5,119,5,3
-6,6,103,2,8
-6,7,126,1,5
-6,8,128,4,0
-6,9,113,5,5
-6,10,111,7,2
-6,11,117,9,8
-6,12,124,8,8
-6,13,100,2,8
-6,14,119,4,3
-6,15,117,4,4
-7,0,111,9,9
-7,1,112,3,2
-7,2,113,7,8
-7,3,113,1,0
-7,4,113,8,4
-7,5,121,1,4
-7,6,100,4,6
-7,7,129,6,7
-7,8,100,4,3
-7,9,125,6,0
-7,10,118,9,6
-7,11,121,7,3
-7,12,127,8,3
-7,13,121,1,2
-7,14,129,1,3
-7,15,113,8,6
-8,0,101,6,7
-8,1,119,4,2
-8,2,103,5,3
-8,3,113,7,9
-8,4,105,6,1
-8,5,111,9,0
-8,6,124,7,5
-8,7,108,9,1
-8,8,118,4,9
-8,9,124,2,0
-8,10,109,1,6
-8,11,121,8,0
-8,12,110,4,6
-8,13,120,4,3
-8,14,114,4,2
-8,15,107,8,4
-9,0,128,0,1
-9,1,120,0,6
-9,2,103,5,7
-9,3,116,5,7
-9,4,122,6,9
-9,5,100,6,5
-9,6,128,1,4
-9,7,102,7,8
-9,8,122,6,5
-9,9,110,4,8
-9,10,122,7,5
-9,11,107,5,3
-9,12,123,2,2
-9,13,103,8,2
-9,14,128,9,2
-9,15,106,7,9
diff --git a/tests/resources/source_data/test-data-categorical-3-equal.parquet b/tests/resources/source_data/test-data-categorical-3-equal.parquet
deleted file mode 100644
index d86b09844edfaf67e0b72caaced91b9f5328b65b..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 4744
zcmcgwTWs6b8Kz|2?8T|0go2!T0Hel=3)rbHmJ=5lQ0i)Gwrt4~by<g$6e)``MadF%
zbC%8Vvi0)NX3gSsD-Z+&wgJPQ`qH5fLol>MQ4Co3+I2u5h7H)mkQGB8`cUjYhmtK>
zYO?`B34IjL|DW?;zVn}-f;Q-cn>3L3b)>LLo*@Y<L2MAjq>j){S_nO03}65l0VaTX
z(sIQ@XxDi6n<U7~<du2yfu5WqO{8hYZt1D)y}5*qm>VGl3p4T<+#l$528y6`l%5)+
z43v>FQD(|QjXQs@pCJv?KEgCLZJ(%%I}K6Ei;O_(Z4dOePxQ98_0E@dPRsw}b9x?9
zCP+O2w<-7{2pI3_$phUgNnsB5bdMr|umBn3fC<3Y04D&GK+Z`phps2IFXZD*9qAfi
zr>7faXIKg*vd{L|d1}g&Y`*^XN1uOgAx>GGpO}t#IU%WHd81O7RH>q9%zR+5{Y~$D
z$LM^;0MrbdM%ytgB}P*7P)B|W)Lg;Tv>&#r5LXd!1JD7K0Tsaa09$|-KmzOlRDcX<
z0tx^H&<1=LAOcE&T|fiyeLx*h16%{_0R+Hxz%~G={7^@Jtn1@)iI_8>Fzg025w7yX
zUO5?`qF$umr%X405Ib9%efMYeCx7yfne)^ub0<rGnsom4jpTb13&u}b^G~zy|I+pK
z`w}&_{;umUhD>3fjqcsBWq<X7J9p>orC%H8e(?v}8*gzhY*@c>Cn_ww_HXK6XYYmH
zYX0ZX_g=l{+@G8w_1*e3p?&Dp=9>oF?+nflC!FtCW{!n9lI0CD);of8Nwn$c44DC_
zKurJxF|M13!X%6Ubgs10`t5N(&^o&TZs^KB%p4VvHJ~$+dH>GJ3(p%Oq43^;A!&Lk
zIela4AZ7j0dB$kv3gI0(l46q#=c%4LFy5^H(7K$`)jFZjxz63>+!?FyZt70q#o&eH
z{_S(l&!&!YH`g200H?aO^835q_D{X@vy;wSCvoz9F^x<f8c?y9{f1!-G9@hIW&>_P
zs1-rm!7T{2B4|TbkiGtEG!>&8&HyD2Z(!4QI(^QYyx`^bm(7=MM_#+j7wZw%)Jyy0
zuC}wFyC7b^mB=#elNU~P1nYP_Yd!V++>Z;9TV|?ye($#L^}A-i9?aNCqhsHBpFBF2
zT}nI}CGU;d4#u3fPCH+=B1%TZHj0{Q*+<w-^?Xq*bL0ecs#$%1AJhZ`4FEbD%<0zR
ztbO%2r>CH;Ytr5>&7K7@g~iBVFVfi=5I>8=&|Rlzrl0$2+V5A^%pTKeeO!hAxU0qX
zb1Uqd(_;JRi8BQBa=WR@DFRow-!xxyNQxk~ujVUqqpZ{&&rsjK=GX=qJUMEdvL!m^
zDTl(9MGz<rS%y(v<!UMnUHG|Dg-7?2HW2aP`gcI1S-gtPi(SKmb_$Xx3s(naT3}bz
zP8B(HG!!Xc5wsW@<!a|pQQ=EsUPZB$MU@jcm4oTm9PlHcB1p;>TANCS7cM)l_XRT;
zh2!W~1F|y^%`x4vs3b*QTEx5z#B?a`3kgwneI6p~rs?Q^B@eqju6@jo8DX?bO@7Q-
z1M+b%4X1e||FS&0ZGYHbpJIeVuV=JFP5x+yc=@rd#4jMM7~yc#MxqQg`EiH1uxlcx
zR%*E3WQnhFwGJw`YKN=UDlHU#Q>@kD7Z;MceHUF9Jq{=VN7s3u(sR*!(e1cS*iX=j
z#CrM}t()8bIQcZX+H{zv={Nvg-=gV7+P{TlTXbLx%l!TUeG3Ughx*l4P#eP?TfusU
zrdu?ugKuOckb)r^gCq{~{g;@s+{i`P#f&G`<X2dgSJtF;$<K?OXpKwzwwY38i&3IQ
zzAVU$)Y@Xoz9OGap*am9-L9_&XesYa6!S_PWOS=LZsr5b5)%x0)&j6biMeta8tK~w
z)>UT$(Gu@zyZN-QfPCaKMLFMr`C%2}8MdD*%Q7EH6!>yj&27TGh|kSOTF6c-3+uC)
zSV35kTUZx}y7{sU_7lB$-F(LH%XkyMd<4Z!OS~t#lS{{3T0U-Bh=jpLqKI;Zyk))|
zYvv-r3yLWaW!X?xT@OT;kHx-==BK@N=`gN9bT^j?yK`w@DVLJBHT$`0F5TYFczSsO
zUmmtCq}fK+!$xxf=8N*E@=Dwj)AYc#yCt&AE-~Ym@t(lG3Bl~h9C(GT^DDsf%0O&Y
zA;PMu$nq|iNmO~yqJ;Kph}%GP<h*<&40RIT?dK8p&J`=gu^sTK=2ztYedv#0%>lo#
z-jzMvzxMKK0M*|SFA$f^e+h2`dc_Yl33cc0<3wSR&vs10oDOp~sQ;g0%6r*X+!J=K
zZDN~cF3r|+;pIj)?bh}|iMjKkSWyUuFb|`664Gd|);ituGgOa*`oVhqg1y`oGX7vT
z6Yts_QDeQhH7-V;lSk~IdqAINUD<S^C`3YsXTNtI@?{p}%bo7|sdMRQC!2}-aGffC
zA3S$>Zyxg_l-W++BR6>ooh_yv+gxm|mKdq9iTcFho_cbb=q{IDZeSZFrm@ypY%_`r
zbuQEwX?mqQ7xgpL@78gD%Oiab-GaCYLVu*`jSc9ps2#=`EQ}!GVXKApNVn44M~1F8
zVLo={ufWsOp!L&5w1&>28}6Uz0al^gVU5Itn`k~XTbhn0;ps9wgaSB-01~2p-Mf$}
zKR#=lX&e^H3zo*U{IHWCgY2P+UH>S_e*bG|jL!w@Zvw9_#2J#&tGiwQ8a-$cyiCA;
z7UhfAK`@%*Bm3KEX<vTu9)kS(`xT5K6<EK8?BG<<96E*X-z4l*m8-%3GwMRVT3x^|
dNviu2q_o%L^m8-te^1|s`1l0;;ROFd{TB@Fj^F?Q

diff --git a/tests/resources/source_data/test-data-categorical-3.csv b/tests/resources/source_data/test-data-categorical-3.csv
deleted file mode 100644
index 5bf5157c..00000000
--- a/tests/resources/source_data/test-data-categorical-3.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-0,0,112,3,3
-0,1,104,9,8
-0,2,129,7,2
-0,3,121,9,2
-0,4,109,2,1
-0,5,111,6,1
-0,6,121,9,1
-0,7,109,3,7
-0,8,126,4,1
-0,9,112,3,5
-0,10,104,4,4
-0,11,103,9,8
-0,12,119,2,1
-0,13,103,0,8
-0,14,105,7,6
-0,15,121,6,5
-0,16,107,1,8
-0,17,128,4,4
-0,18,128,6,2
-0,19,116,9,2
-0,20,106,3,1
-0,21,103,9,6
-0,22,119,6,0
-0,23,105,8,2
-0,24,100,4,1
-1,25,113,9,8
-1,26,116,1,2
-1,27,105,5,1
-1,28,113,6,1
-1,29,108,1,7
-1,30,118,5,5
-1,31,102,3,6
-1,32,102,0,2
-1,33,121,6,1
-1,34,104,3,1
-1,35,111,1,8
-1,36,110,6,4
-1,37,102,4,2
-1,38,102,6,9
-1,39,120,0,0
-1,40,109,6,1
-1,41,101,8,2
-1,42,111,1,6
-1,43,127,2,8
-1,44,105,0,5
-1,45,115,7,9
-1,46,104,5,5
-2,47,112,3,6
-2,48,124,5,3
-2,49,121,0,6
-2,50,124,8,1
-2,51,129,6,1
-2,52,103,3,0
-2,53,119,6,6
-2,54,118,4,7
-2,55,110,0,2
-2,56,102,3,0
-2,57,121,8,6
-2,58,108,5,0
-2,59,101,6,2
-2,60,102,4,4
-2,61,110,8,9
-2,62,106,9,4
-2,63,102,5,7
-2,64,110,4,8
-2,65,109,4,6
-2,66,115,2,5
-3,67,122,9,1
-3,68,129,5,6
-3,69,101,0,6
-3,70,117,2,7
-3,71,120,0,3
-3,72,127,8,1
-3,73,122,4,3
-3,74,112,4,8
-3,75,127,7,0
-3,76,115,9,9
-3,77,117,0,2
-3,78,102,5,9
-3,79,120,6,0
-3,80,108,7,3
-3,81,126,8,3
-3,82,116,6,1
-3,83,102,5,6
-3,84,121,0,4
-3,85,118,8,4
-3,86,105,5,5
-4,87,113,4,9
-4,88,127,3,6
-4,89,105,7,4
-4,90,104,9,2
-4,91,106,9,4
-4,92,118,0,4
-4,93,112,0,8
-4,94,122,0,3
-4,95,105,0,6
-4,96,121,6,2
-4,97,119,9,5
-4,98,126,1,4
-4,99,120,9,9
-4,100,103,6,1
-4,101,117,1,3
-4,102,114,6,0
-5,103,120,4,2
-5,104,123,4,5
-5,105,127,0,0
-5,106,105,4,5
-5,107,112,6,0
-5,108,108,4,7
-5,109,123,8,4
-5,110,108,6,5
-5,111,129,9,1
-5,112,122,0,2
-5,113,117,7,6
-5,114,122,2,2
-5,115,109,3,8
-5,116,112,1,8
-5,117,126,9,0
-5,118,120,9,6
-6,119,127,0,7
-6,120,119,8,4
-6,121,127,4,2
-6,122,102,3,8
-6,123,113,6,0
-6,124,119,5,3
-6,125,103,2,8
-6,126,126,1,5
-6,127,128,4,0
-6,128,113,5,5
-6,129,111,7,2
-6,130,117,9,8
-6,131,124,8,8
-6,132,100,2,8
-6,133,119,4,3
-6,134,117,4,4
-6,135,103,7,8
-6,136,110,4,7
-6,137,108,7,6
-6,138,125,5,0
-6,139,107,8,3
-7,140,111,9,9
-7,141,112,3,2
-7,142,113,7,8
-7,143,113,1,0
-7,144,113,8,4
-7,145,121,1,4
-7,146,100,4,6
-7,147,129,6,7
-7,148,100,4,3
-7,149,125,6,0
-7,150,118,9,6
-7,151,121,7,3
-7,152,127,8,3
-7,153,121,1,2
-7,154,129,1,3
-7,155,113,8,6
-7,156,110,3,3
-8,157,101,6,7
-8,158,119,4,2
-8,159,103,5,3
-8,160,113,7,9
-8,161,105,6,1
-8,162,111,9,0
-8,163,124,7,5
-8,164,108,9,1
-8,165,118,4,9
-8,166,124,2,0
-8,167,109,1,6
-8,168,121,8,0
-8,169,110,4,6
-8,170,120,4,3
-8,171,114,4,2
-8,172,107,8,4
-8,173,100,2,2
-8,174,111,6,3
-8,175,107,9,7
-8,176,109,4,6
-8,177,114,1,3
-8,178,102,6,0
-8,179,126,2,2
-8,180,115,2,4
-9,181,128,0,1
-9,182,120,0,6
-9,183,103,5,7
-9,184,116,5,7
-9,185,122,6,9
-9,186,100,6,5
-9,187,128,1,4
-9,188,102,7,8
-9,189,122,6,5
-9,190,110,4,8
-9,191,122,7,5
-9,192,107,5,3
-9,193,123,2,2
-9,194,103,8,2
-9,195,128,9,2
-9,196,106,7,9
-9,197,108,3,4
-9,198,119,4,4
-9,199,129,4,3
diff --git a/tests/resources/source_data/test-data-categorical-5.csv b/tests/resources/source_data/test-data-categorical-5.csv
deleted file mode 100644
index 4a342018..00000000
--- a/tests/resources/source_data/test-data-categorical-5.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,0,112,3,9,7,2
-0,1,104,9,1,8,1
-0,2,129,7,4,0,1
-0,3,121,9,9,4,0
-0,4,109,2,9,5,2
-0,5,111,6,3,9,0
-0,6,121,9,0,2,0
-0,7,109,3,5,8,1
-0,8,126,4,9,6,0
-0,9,112,3,3,3,2
-0,10,104,4,5,5,1
-0,11,103,9,6,0,0
-0,12,119,2,4,7,2
-0,13,103,0,8,3,1
-0,14,105,7,9,3,2
-0,15,121,6,4,8,0
-0,16,107,1,7,7,2
-0,17,128,4,5,1,1
-0,18,128,6,1,0,2
-0,19,116,9,9,8,1
-0,20,106,3,3,7,2
-0,21,103,9,0,2,1
-0,22,119,6,9,5,0
-0,23,105,8,2,6,0
-0,24,100,4,5,5,1
-1,25,113,9,6,4,1
-1,26,116,1,4,3,0
-1,27,105,5,9,6,1
-1,28,113,6,3,6,1
-1,29,108,1,3,2,2
-1,30,118,5,2,2,0
-1,31,102,3,5,7,1
-1,32,102,0,3,8,1
-1,33,121,6,3,6,1
-1,34,104,3,8,7,2
-1,35,111,1,9,5,1
-1,36,110,6,7,8,0
-1,37,102,4,9,4,2
-1,38,102,6,3,7,0
-1,39,120,0,9,8,2
-1,40,109,6,2,5,1
-1,41,101,8,8,8,0
-1,42,111,1,9,9,0
-1,43,127,2,2,4,2
-1,44,105,0,9,1,1
-1,45,115,7,8,8,2
-1,46,104,5,0,0,0
-2,47,112,3,0,0,0
-2,48,124,5,0,3,0
-2,49,121,0,4,0,2
-2,50,124,8,3,7,0
-2,51,129,6,7,0,1
-2,52,103,3,1,6,0
-2,53,119,6,5,6,1
-2,54,118,4,6,4,1
-2,55,110,0,1,8,1
-2,56,102,3,1,4,0
-2,57,121,8,9,3,0
-2,58,108,5,5,6,1
-2,59,101,6,4,2,2
-2,60,102,4,0,4,1
-2,61,110,8,9,8,2
-2,62,106,9,3,8,1
-2,63,102,5,1,3,0
-2,64,110,4,1,6,0
-2,65,109,4,5,2,1
-2,66,115,2,2,7,0
-3,67,122,9,9,0,0
-3,68,129,5,6,4,1
-3,69,101,0,2,8,0
-3,70,117,2,4,9,0
-3,71,120,0,0,5,0
-3,72,127,8,8,9,0
-3,73,122,4,7,4,1
-3,74,112,4,9,5,2
-3,75,127,7,3,7,0
-3,76,115,9,1,6,1
-3,77,117,0,4,7,0
-3,78,102,5,6,2,2
-3,79,120,6,1,7,2
-3,80,108,7,0,4,1
-3,81,126,8,1,7,2
-3,82,116,6,9,5,2
-3,83,102,5,9,2,0
-3,84,121,0,3,3,2
-3,85,118,8,2,5,0
-3,86,105,5,8,4,0
-4,87,113,4,3,4,0
-4,88,127,3,9,3,0
-4,89,105,7,8,0,0
-4,90,104,9,5,8,0
-4,91,106,9,1,5,1
-4,92,118,0,9,1,2
-4,93,112,0,3,8,1
-4,94,122,0,0,2,2
-4,95,105,0,8,3,2
-4,96,121,6,1,7,2
-4,97,119,9,2,8,1
-4,98,126,1,5,0,1
-4,99,120,9,7,3,1
-4,100,103,6,8,4,0
-4,101,117,1,2,4,0
-4,102,114,6,7,0,1
-5,103,120,4,3,6,0
-5,104,123,4,4,4,0
-5,105,127,0,2,8,0
-5,106,105,4,7,7,1
-5,107,112,6,0,7,0
-5,108,108,4,9,2,1
-5,109,123,8,3,9,0
-5,110,108,6,9,4,1
-5,111,129,9,0,7,2
-5,112,122,0,3,0,0
-5,113,117,7,2,5,1
-5,114,122,2,6,4,1
-5,115,109,3,7,5,1
-5,116,112,1,7,2,2
-5,117,126,9,8,9,1
-5,118,120,9,5,6,2
-6,119,127,0,8,2,0
-6,120,119,8,2,0,2
-6,121,127,4,5,7,1
-6,122,102,3,2,5,1
-6,123,113,6,3,1,0
-6,124,119,5,5,0,2
-6,125,103,2,1,0,2
-6,126,126,1,2,0,0
-6,127,128,4,2,0,2
-6,128,113,5,7,2,1
-6,129,111,7,3,1,0
-6,130,117,9,8,5,2
-6,131,124,8,3,3,2
-6,132,100,2,4,5,1
-6,133,119,4,6,1,1
-6,134,117,4,7,9,0
-6,135,103,7,4,7,1
-6,136,110,4,4,2,2
-6,137,108,7,4,1,0
-6,138,125,5,6,4,0
-6,139,107,8,6,4,0
-7,140,111,9,5,1,2
-7,141,112,3,4,2,1
-7,142,113,7,8,0,0
-7,143,113,1,7,4,2
-7,144,113,8,9,8,1
-7,145,121,1,0,1,1
-7,146,100,4,4,1,0
-7,147,129,6,8,5,0
-7,148,100,4,8,2,1
-7,149,125,6,9,8,1
-7,150,118,9,8,6,1
-7,151,121,7,3,3,2
-7,152,127,8,5,5,1
-7,153,121,1,8,4,0
-7,154,129,1,1,4,0
-7,155,113,8,1,8,0
-7,156,110,3,9,5,1
-8,157,101,6,1,4,2
-8,158,119,4,8,7,0
-8,159,103,5,7,7,1
-8,160,113,7,2,9,2
-8,161,105,6,0,0,2
-8,162,111,9,7,3,0
-8,163,124,7,9,6,1
-8,164,108,9,7,7,1
-8,165,118,4,5,4,1
-8,166,124,2,2,8,0
-8,167,109,1,2,8,0
-8,168,121,8,0,1,0
-8,169,110,4,8,4,0
-8,170,120,4,6,8,0
-8,171,114,4,2,9,2
-8,172,107,8,2,3,2
-8,173,100,2,2,6,0
-8,174,111,6,9,4,1
-8,175,107,9,0,8,2
-8,176,109,4,5,6,0
-8,177,114,1,1,3,0
-8,178,102,6,8,0,2
-8,179,126,2,2,4,1
-8,180,115,2,0,6,2
-9,181,128,0,9,2,0
-9,182,120,0,2,7,1
-9,183,103,5,5,2,2
-9,184,116,5,2,8,1
-9,185,122,6,0,6,0
-9,186,100,6,2,5,0
-9,187,128,1,1,0,1
-9,188,102,7,8,3,1
-9,189,122,6,1,2,0
-9,190,110,4,3,9,1
-9,191,122,7,8,8,0
-9,192,107,5,2,0,2
-9,193,123,2,2,6,2
-9,194,103,8,4,7,0
-9,195,128,9,6,6,1
-9,196,106,7,9,4,1
-9,197,108,3,3,8,1
-9,198,119,4,4,5,1
-9,199,129,4,6,4,1
diff --git a/tests/resources/source_data/test-data-categorical-50.csv b/tests/resources/source_data/test-data-categorical-50.csv
deleted file mode 100644
index badf6dc6..00000000
--- a/tests/resources/source_data/test-data-categorical-50.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4,supCat5,supCat6,supCat7,supCat8,supCat9,supCat10,supCat11,supCat12,supCat13,supCat14,supCat15,supCat16,supCat17,supCat18,supCat19,supCat20,supCat21,supCat22,supCat23,supCat24,supCat25,supCat26,supCat27,supCat28,supCat29,supCat30,supCat31,supCat32,supCat33,supCat34,supCat35,supCat36,supCat37,supCat38,supCat39,supCat40,supCat41,supCat42,supCat43,supCat44,supCat45,supCat46,supCat47,supCat48,supCat49
-0,0,112,3,9,7,2,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9,7,2,3,9
-0,1,104,9,1,8,1,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1,8,1,9,1
-0,2,129,7,4,0,1,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4,0,1,7,4
-0,3,121,9,9,4,0,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9,4,0,9,9
-0,4,109,2,9,5,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9,5,2,2,9
-0,5,111,6,3,9,0,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3,9,0,6,3
-0,6,121,9,0,2,0,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0,2,0,9,0
-0,7,109,3,5,8,1,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5,8,1,3,5
-0,8,126,4,9,6,0,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9,6,0,4,9
-0,9,112,3,3,3,2,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3,3,2,3,3
-0,10,104,4,5,5,1,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5
-0,11,103,9,6,0,0,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6,0,0,9,6
-0,12,119,2,4,7,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4,7,2,2,4
-0,13,103,0,8,3,1,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8,3,1,0,8
-0,14,105,7,9,3,2,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9,3,2,7,9
-0,15,121,6,4,8,0,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4,8,0,6,4
-0,16,107,1,7,7,2,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7,7,2,1,7
-0,17,128,4,5,1,1,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5,1,1,4,5
-0,18,128,6,1,0,2,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1,0,2,6,1
-0,19,116,9,9,8,1,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9,8,1,9,9
-0,20,106,3,3,7,2,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3,7,2,3,3
-0,21,103,9,0,2,1,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0,2,1,9,0
-0,22,119,6,9,5,0,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9,5,0,6,9
-0,23,105,8,2,6,0,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2,6,0,8,2
-0,24,100,4,5,5,1,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5,5,1,4,5
-1,25,113,9,6,4,1,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6,4,1,9,6
-1,26,116,1,4,3,0,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4,3,0,1,4
-1,27,105,5,9,6,1,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9,6,1,5,9
-1,28,113,6,3,6,1,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3
-1,29,108,1,3,2,2,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3,2,2,1,3
-1,30,118,5,2,2,0,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2,2,0,5,2
-1,31,102,3,5,7,1,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5,7,1,3,5
-1,32,102,0,3,8,1,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3
-1,33,121,6,3,6,1,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3,6,1,6,3
-1,34,104,3,8,7,2,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8,7,2,3,8
-1,35,111,1,9,5,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9,5,1,1,9
-1,36,110,6,7,8,0,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7,8,0,6,7
-1,37,102,4,9,4,2,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9,4,2,4,9
-1,38,102,6,3,7,0,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3,7,0,6,3
-1,39,120,0,9,8,2,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9,8,2,0,9
-1,40,109,6,2,5,1,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2,5,1,6,2
-1,41,101,8,8,8,0,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8,8,0,8,8
-1,42,111,1,9,9,0,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9,9,0,1,9
-1,43,127,2,2,4,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2,4,2,2,2
-1,44,105,0,9,1,1,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9,1,1,0,9
-1,45,115,7,8,8,2,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8,8,2,7,8
-1,46,104,5,0,0,0,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0
-2,47,112,3,0,0,0,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0
-2,48,124,5,0,3,0,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0,3,0,5,0
-2,49,121,0,4,0,2,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4,0,2,0,4
-2,50,124,8,3,7,0,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3,7,0,8,3
-2,51,129,6,7,0,1,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7
-2,52,103,3,1,6,0,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1,6,0,3,1
-2,53,119,6,5,6,1,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5,6,1,6,5
-2,54,118,4,6,4,1,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6
-2,55,110,0,1,8,1,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1,8,1,0,1
-2,56,102,3,1,4,0,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1,4,0,3,1
-2,57,121,8,9,3,0,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9,3,0,8,9
-2,58,108,5,5,6,1,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5,6,1,5,5
-2,59,101,6,4,2,2,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4,2,2,6,4
-2,60,102,4,0,4,1,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0,4,1,4,0
-2,61,110,8,9,8,2,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9,8,2,8,9
-2,62,106,9,3,8,1,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3,8,1,9,3
-2,63,102,5,1,3,0,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1,3,0,5,1
-2,64,110,4,1,6,0,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1,6,0,4,1
-2,65,109,4,5,2,1,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5,2,1,4,5
-2,66,115,2,2,7,0,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2,7,0,2,2
-3,67,122,9,9,0,0,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9,0,0,9,9
-3,68,129,5,6,4,1,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6,4,1,5,6
-3,69,101,0,2,8,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2
-3,70,117,2,4,9,0,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4,9,0,2,4
-3,71,120,0,0,5,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0,5,0,0,0
-3,72,127,8,8,9,0,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8,9,0,8,8
-3,73,122,4,7,4,1,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7,4,1,4,7
-3,74,112,4,9,5,2,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9,5,2,4,9
-3,75,127,7,3,7,0,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3,7,0,7,3
-3,76,115,9,1,6,1,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1,6,1,9,1
-3,77,117,0,4,7,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4,7,0,0,4
-3,78,102,5,6,2,2,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6,2,2,5,6
-3,79,120,6,1,7,2,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1
-3,80,108,7,0,4,1,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0,4,1,7,0
-3,81,126,8,1,7,2,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1,7,2,8,1
-3,82,116,6,9,5,2,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9,5,2,6,9
-3,83,102,5,9,2,0,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9,2,0,5,9
-3,84,121,0,3,3,2,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3,3,2,0,3
-3,85,118,8,2,5,0,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2,5,0,8,2
-3,86,105,5,8,4,0,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8,4,0,5,8
-4,87,113,4,3,4,0,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3,4,0,4,3
-4,88,127,3,9,3,0,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9,3,0,3,9
-4,89,105,7,8,0,0,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8
-4,90,104,9,5,8,0,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5,8,0,9,5
-4,91,106,9,1,5,1,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1,5,1,9,1
-4,92,118,0,9,1,2,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9,1,2,0,9
-4,93,112,0,3,8,1,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3,8,1,0,3
-4,94,122,0,0,2,2,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0,2,2,0,0
-4,95,105,0,8,3,2,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8,3,2,0,8
-4,96,121,6,1,7,2,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1,7,2,6,1
-4,97,119,9,2,8,1,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2,8,1,9,2
-4,98,126,1,5,0,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5,0,1,1,5
-4,99,120,9,7,3,1,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7,3,1,9,7
-4,100,103,6,8,4,0,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8,4,0,6,8
-4,101,117,1,2,4,0,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2,4,0,1,2
-4,102,114,6,7,0,1,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7,0,1,6,7
-5,103,120,4,3,6,0,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3,6,0,4,3
-5,104,123,4,4,4,0,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4,4,0,4,4
-5,105,127,0,2,8,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2,8,0,0,2
-5,106,105,4,7,7,1,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7,7,1,4,7
-5,107,112,6,0,7,0,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0,7,0,6,0
-5,108,108,4,9,2,1,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9,2,1,4,9
-5,109,123,8,3,9,0,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3,9,0,8,3
-5,110,108,6,9,4,1,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9
-5,111,129,9,0,7,2,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0,7,2,9,0
-5,112,122,0,3,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3,0,0,0,3
-5,113,117,7,2,5,1,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2,5,1,7,2
-5,114,122,2,6,4,1,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6,4,1,2,6
-5,115,109,3,7,5,1,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7,5,1,3,7
-5,116,112,1,7,2,2,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7,2,2,1,7
-5,117,126,9,8,9,1,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8,9,1,9,8
-5,118,120,9,5,6,2,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5,6,2,9,5
-6,119,127,0,8,2,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8,2,0,0,8
-6,120,119,8,2,0,2,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2,0,2,8,2
-6,121,127,4,5,7,1,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5,7,1,4,5
-6,122,102,3,2,5,1,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2,5,1,3,2
-6,123,113,6,3,1,0,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3,1,0,6,3
-6,124,119,5,5,0,2,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5,0,2,5,5
-6,125,103,2,1,0,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1,0,2,2,1
-6,126,126,1,2,0,0,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2,0,0,1,2
-6,127,128,4,2,0,2,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2,0,2,4,2
-6,128,113,5,7,2,1,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7,2,1,5,7
-6,129,111,7,3,1,0,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3,1,0,7,3
-6,130,117,9,8,5,2,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8,5,2,9,8
-6,131,124,8,3,3,2,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3,3,2,8,3
-6,132,100,2,4,5,1,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4,5,1,2,4
-6,133,119,4,6,1,1,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6,1,1,4,6
-6,134,117,4,7,9,0,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7,9,0,4,7
-6,135,103,7,4,7,1,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4,7,1,7,4
-6,136,110,4,4,2,2,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4,2,2,4,4
-6,137,108,7,4,1,0,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4,1,0,7,4
-6,138,125,5,6,4,0,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6,4,0,5,6
-6,139,107,8,6,4,0,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6,4,0,8,6
-7,140,111,9,5,1,2,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5,1,2,9,5
-7,141,112,3,4,2,1,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4,2,1,3,4
-7,142,113,7,8,0,0,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8,0,0,7,8
-7,143,113,1,7,4,2,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7,4,2,1,7
-7,144,113,8,9,8,1,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9,8,1,8,9
-7,145,121,1,0,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0
-7,146,100,4,4,1,0,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4,1,0,4,4
-7,147,129,6,8,5,0,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8,5,0,6,8
-7,148,100,4,8,2,1,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8,2,1,4,8
-7,149,125,6,9,8,1,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9,8,1,6,9
-7,150,118,9,8,6,1,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8,6,1,9,8
-7,151,121,7,3,3,2,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3,3,2,7,3
-7,152,127,8,5,5,1,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5,5,1,8,5
-7,153,121,1,8,4,0,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8,4,0,1,8
-7,154,129,1,1,4,0,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1,4,0,1,1
-7,155,113,8,1,8,0,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1,8,0,8,1
-7,156,110,3,9,5,1,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9,5,1,3,9
-8,157,101,6,1,4,2,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1,4,2,6,1
-8,158,119,4,8,7,0,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8,7,0,4,8
-8,159,103,5,7,7,1,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7,7,1,5,7
-8,160,113,7,2,9,2,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2,9,2,7,2
-8,161,105,6,0,0,2,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0,0,2,6,0
-8,162,111,9,7,3,0,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7,3,0,9,7
-8,163,124,7,9,6,1,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9,6,1,7,9
-8,164,108,9,7,7,1,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7,7,1,9,7
-8,165,118,4,5,4,1,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5,4,1,4,5
-8,166,124,2,2,8,0,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2,8,0,2,2
-8,167,109,1,2,8,0,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2,8,0,1,2
-8,168,121,8,0,1,0,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0,1,0,8,0
-8,169,110,4,8,4,0,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8,4,0,4,8
-8,170,120,4,6,8,0,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6,8,0,4,6
-8,171,114,4,2,9,2,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2,9,2,4,2
-8,172,107,8,2,3,2,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2,3,2,8,2
-8,173,100,2,2,6,0,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2,6,0,2,2
-8,174,111,6,9,4,1,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9,4,1,6,9
-8,175,107,9,0,8,2,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0,8,2,9,0
-8,176,109,4,5,6,0,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5,6,0,4,5
-8,177,114,1,1,3,0,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1,3,0,1,1
-8,178,102,6,8,0,2,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8,0,2,6,8
-8,179,126,2,2,4,1,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2,4,1,2,2
-8,180,115,2,0,6,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0,6,2,2,0
-9,181,128,0,9,2,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9,2,0,0,9
-9,182,120,0,2,7,1,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2,7,1,0,2
-9,183,103,5,5,2,2,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5,2,2,5,5
-9,184,116,5,2,8,1,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2,8,1,5,2
-9,185,122,6,0,6,0,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0,6,0
-9,186,100,6,2,5,0,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2,5,0,6,2
-9,187,128,1,1,0,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1,0,1,1,1
-9,188,102,7,8,3,1,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8,3,1,7,8
-9,189,122,6,1,2,0,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1,2,0,6,1
-9,190,110,4,3,9,1,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3,9,1,4,3
-9,191,122,7,8,8,0,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8,8,0,7,8
-9,192,107,5,2,0,2,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2,0,2,5,2
-9,193,123,2,2,6,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2,6,2,2,2
-9,194,103,8,4,7,0,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4,7,0,8,4
-9,195,128,9,6,6,1,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6,6,1,9,6
-9,196,106,7,9,4,1,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9,4,1,7,9
-9,197,108,3,3,8,1,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3,8,1,3,3
-9,198,119,4,4,5,1,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4,5,1,4,4
-9,199,129,4,6,4,1,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6,4,1,4,6
diff --git a/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-1.csv b/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-1.csv
deleted file mode 100644
index 69d1879e..00000000
--- a/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-1.csv
+++ /dev/null
@@ -1,245 +0,0 @@
-sequenceId,itemPosition,Unnamed: 0,itemId,supCat1,supReal2,supReal3,supCat4
-0,0,0,112,3,0.47512151913436,-0.5820086906467112,2
-0,1,1,104,9,-1.1261734468621212,-0.9089247216662516,1
-0,2,2,129,7,-1.2162003117631153,-0.5154371712252781,1
-0,3,3,121,9,-1.4933477031742597,-0.8796536665660435,0
-0,4,4,109,2,-0.9670816286563634,-0.0946835025191322,2
-0,5,5,111,6,0.2951585961176835,0.573453806301526,0
-0,6,6,121,9,-0.8256811879776176,-0.6726355586632142,0
-0,7,7,109,3,-0.2626654773388363,0.7338377851923579,1
-0,8,8,126,4,1.3128198055960525,0.4123843865529258,0
-0,9,9,112,3,-0.1724514169048517,1.0002758074691138,2
-0,10,10,104,4,0.9984558771504448,-1.6793033775681865,1
-0,11,11,103,9,0.8241769523505718,0.5192150697838572,0
-0,12,12,119,2,1.1524820376873743,-0.0684948380162701,2
-0,13,13,103,0,-0.4532287722802309,-1.012670179240466,1
-0,14,14,105,7,0.7406957728589633,0.4598005719286187,2
-0,15,15,121,6,-0.3046679002763786,2.228549215957172,0
-0,16,16,107,1,-0.0981028713050338,1.580761752579522,2
-0,17,17,128,4,-0.338687599861625,1.0274608136462915,1
-0,18,18,128,6,-0.5926329276128396,0.7105776900317172,2
-0,19,19,116,9,-0.8683614880285571,0.7000235215395574,1
-0,20,20,106,3,-0.2223944052810371,-0.1421408408449271,2
-0,21,21,103,9,0.2238516678750416,0.7671249818780963,1
-0,22,22,119,6,-0.4794418411187094,0.284269664190007,0
-0,23,23,105,8,-0.208702368387007,0.5367982644094035,0
-0,24,24,100,4,0.1273379928547726,0.9143038302645866,1
-1,0,25,113,9,-0.2656666807034851,0.4806224256538509,1
-1,1,26,116,1,-0.9754261015074585,0.2940960032574599,0
-1,2,27,105,5,-2.0639707193059498,1.2133396475903873,1
-1,3,28,113,6,0.7319848912899831,-0.4229377963857184,1
-1,4,29,108,1,-0.0908278640955465,0.5666495515437896,2
-1,5,30,118,5,0.2854174163780527,0.6009858839082124,0
-1,6,31,102,3,0.2014682827257396,1.7092664244525773,1
-1,7,32,102,0,0.1453056346528572,0.3399258904044571,1
-1,8,33,121,6,0.8095167753996217,-0.1079289200633795,1
-1,9,34,104,3,1.5900425772732012,0.6294206328381015,2
-1,10,35,111,1,-0.8178820250244837,1.4633194614817098,1
-1,11,36,110,6,-1.1811248427188683,0.5503790726147177,0
-1,12,37,102,4,-0.5521109827498613,0.2288111251968629,2
-1,13,38,102,6,0.3949748622436907,1.5447878604004366,0
-1,14,39,120,0,-0.6463976133595792,0.1630326559403528,2
-1,15,40,109,6,-1.2805957817399582,-1.1826482612552474,1
-1,16,41,101,8,0.5508900386572504,-0.1671730495935667,0
-1,17,42,111,1,0.0853086889333529,-0.0132633758837144,0
-1,18,43,127,2,-0.218769711385981,-0.7019854368140658,2
-1,19,44,105,0,-0.6197723706627636,2.138611526896666,1
-1,20,45,115,7,-1.1247441521636312,-0.4641464411878229,2
-1,21,46,104,5,-0.072555647708791,-0.3929353813928705,0
-2,0,47,112,3,1.3399026572550203,1.878090363680033,0
-2,1,48,124,5,-0.024381870075387,0.5316335161706456,0
-2,2,49,121,0,-0.9970857305430262,-1.383512008752633,2
-2,3,50,124,8,-0.8334960910074487,-0.9116180632797354,0
-2,4,51,129,6,0.3845601295654262,0.1807163239900383,1
-2,5,52,103,3,-0.2294227620946844,0.3477709783565773,0
-2,6,53,119,6,1.0169773111385545,-0.5910886720642256,1
-2,7,54,118,4,0.3363413463024886,-1.0001172593271828,1
-2,8,55,110,0,0.0097818169970301,-0.6059838835052326,1
-2,9,56,102,3,0.5901905216267234,0.3696348407961316,0
-2,10,57,121,8,-0.3329927633569184,-2.1592296837868172,0
-2,11,58,108,5,0.5469937351204301,0.1158729118533183,1
-2,12,59,101,6,1.4387947715516716,0.510184749224577,2
-2,13,60,102,4,-1.5928420817380986,-1.0181941589104413,1
-2,14,61,110,8,0.4144898850289152,-2.1012260841117647,2
-2,15,62,106,9,0.9626552542957164,-0.2641403390810031,1
-2,16,63,102,5,0.5938953228849851,1.126424989282014,0
-2,17,64,110,4,-0.6852888501385631,-0.2788158895740412,0
-2,18,65,109,4,-1.802304268569908,-0.3884284583050837,1
-2,19,66,115,2,-0.0129419618714691,-0.9695615693196264,0
-3,0,67,122,9,0.728096157555939,1.57860530312412,0
-3,1,68,129,5,0.7156279104852876,-0.4852580169423038,1
-3,2,69,101,0,1.3546839988588728,1.3115417566478216,0
-3,3,70,117,2,-1.232361381356479,-0.9375197390131628,0
-3,4,71,120,0,-0.0435872778479013,-1.4933926504379784,0
-3,5,72,127,8,0.5484525334475734,1.66698134913082,0
-3,6,73,122,4,1.307847634805456,0.4215958062784253,1
-3,7,74,112,4,-2.618532214839832,1.6374672768528042,2
-3,8,75,127,7,-0.8496513139782857,0.2645026021626481,0
-3,9,76,115,9,-0.7521255704289022,0.7526070369449078,1
-3,10,77,117,0,-2.126705535677708,0.2182067272217519,0
-3,11,78,102,5,0.8752645615044675,0.2730385089817146,2
-3,12,79,120,6,0.3561666568718871,-1.1795205381405394,2
-3,13,80,108,7,-0.5208614133489836,-0.300497076461078,1
-3,14,81,126,8,0.1299212462358277,-0.8405052153381654,2
-3,15,82,116,6,-0.3047310126274782,0.7356818372550097,2
-3,16,83,102,5,1.9115397604052724,-0.2388956656409064,0
-3,17,84,121,0,-0.8567878621140438,1.643155432693797,2
-3,18,85,118,8,0.9366327938393536,0.5319856670267262,0
-3,19,86,105,5,0.1183670068806765,-2.1326672344812105,0
-4,0,87,113,4,-0.1473154301210917,-1.6042550026469249,0
-4,1,88,127,3,-1.3056540391037692,-0.5997440870462475,0
-4,2,89,105,7,3.0274943343874083,-0.0520796489093672,0
-4,3,90,104,9,0.3617294328556023,-0.0883800576540255,0
-4,4,91,106,9,-0.4453650036042871,-0.8334261447948773,1
-4,5,92,118,0,0.6282853705023487,1.350321469218172,2
-4,6,93,112,0,0.9642562105511686,-0.9992263103431246,1
-4,7,94,122,0,-0.8330901074711051,-0.7635405661499053,2
-4,8,95,105,0,0.7981435937792382,-0.6666971915003546,2
-4,9,96,121,6,0.7995168560801881,-3.730120038128068,2
-4,10,97,119,9,-0.4778765668524142,-1.8018134368021577,1
-4,11,98,126,1,1.7731643504472827,-2.966586951844483,1
-4,12,99,120,9,1.2401001274351815,0.9763019660890628,1
-4,13,100,103,6,-0.6172799146920306,-1.9341599869280055,0
-4,14,101,117,1,0.7526864964032826,0.0115401685101198,0
-4,15,102,114,6,-0.3922040087053289,-1.7973367751171303,1
-5,0,103,120,4,1.3164444471305916,-0.5782800824093337,0
-5,1,104,123,4,0.420020710446368,-0.1850434000530602,0
-5,2,105,127,0,0.8921979069031415,-0.2332258635179755,0
-5,3,106,105,4,0.4279062963690861,-1.5395002171602334,1
-5,4,107,112,6,-0.1809155329521617,-1.106476206464435,0
-5,5,108,108,4,0.5335886571652085,-2.129243520644403,1
-5,6,109,123,8,-0.7825217022412374,0.4768143516638856,0
-5,7,110,108,6,-1.3297057074440568,-0.1723342300598751,1
-5,8,111,129,9,-0.9899126599978328,-0.6345577264308937,2
-5,9,112,122,0,-1.2241592816111664,-1.3588134360661253,0
-5,10,113,117,7,-1.6081151211440765,-1.115309203106232,1
-5,11,114,122,2,0.6893365374951431,1.660083512711184,1
-5,12,115,109,3,1.3184511797851892,0.5981510743069248,1
-5,13,116,112,1,0.2530927744153326,0.1762152669186038,2
-5,14,117,126,9,-0.7218134212123581,0.0995596156598381,1
-5,15,118,120,9,-0.0069204705176007,-0.134973127015259,2
-6,0,119,127,0,-0.4138548174567836,-0.7491928547518463,0
-6,1,120,119,8,-0.9217336181544192,2.0705334246918015,2
-6,2,121,127,4,-1.086483742151834,-0.714833705558571,1
-6,3,122,102,3,1.474288547233567,0.1631413469278299,1
-6,4,123,113,6,1.5875581988032164,1.027268717867479,0
-6,5,124,119,5,-1.8988657509221156,1.8808868244932129,2
-6,6,125,103,2,-0.4050814416560676,0.4437277986745326,2
-6,7,126,126,1,0.8726496980525889,0.1240530371932496,0
-6,8,127,128,4,1.8224737121284704,0.4209771365993368,2
-6,9,128,113,5,0.3293755829411718,0.6439655653176443,1
-6,10,129,111,7,1.1353736264714218,-0.0632902816234641,0
-6,11,130,117,9,-2.088924140703356,0.7873992206662046,2
-6,12,131,124,8,1.1776401371612333,0.1655764555283259,2
-6,13,132,100,2,0.3101310578176122,-0.543886007962999,1
-6,14,133,119,4,0.0489690373962899,-1.3584102327948733,1
-6,15,134,117,4,-1.122565585229101,-0.1804465985609838,0
-6,16,135,103,7,-0.1379191118897307,1.0044578374950142,1
-6,17,136,110,4,0.477337569358077,0.7537311601289707,2
-6,18,137,108,7,-0.0207534690671556,0.5841218654437593,0
-6,19,138,125,5,0.9348832049035216,-0.7226482893432934,0
-6,20,139,107,8,-0.804227609128954,-0.5571601176468748,0
-7,0,140,111,9,0.1431939714725609,0.0422524056965653,2
-7,1,141,112,3,-0.4605569829841085,-0.4325683214460127,1
-7,2,142,113,7,-1.0553504705140826,0.8465631724299651,0
-7,3,143,113,1,0.7206026881536415,0.2871208611901518,2
-7,4,144,113,8,-1.718237086897971,-0.7868091893675095,1
-7,5,145,121,1,0.527556676509205,-1.9318448023128916,1
-7,6,146,100,4,1.4729892021615898,0.8391818770701904,0
-7,7,147,129,6,-1.0059562338044414,0.0019652114728092,0
-7,8,148,100,4,1.6305569491478604,1.0951871930560133,1
-7,9,149,125,6,2.547370551387279,-1.0208678213187818,1
-7,10,150,118,9,2.033324551178638,-0.1154719415024776,1
-7,11,151,121,7,0.4043282940651028,-0.8707042594597949,2
-7,12,152,127,8,0.4191280421631976,0.0957855998835066,1
-7,13,153,121,1,-1.5587069063508987,0.1791648322349623,0
-7,14,154,129,1,-0.7830858337970409,0.0648510199038132,0
-7,15,155,113,8,0.1371199788701356,0.2161753878561426,0
-7,16,156,110,3,1.5499797081722422,1.0344576013849434,1
-8,0,157,101,6,1.818575785843776,-1.9949939475490075,2
-8,1,158,119,4,0.1183152450709592,-1.0344467157761543,0
-8,2,159,103,5,0.7652662329923993,-1.0019028093921996,1
-8,3,160,113,7,1.5686505707558396,-0.5766936717058257,2
-8,4,161,105,6,0.1255815698914305,-0.9386728917772926,2
-8,5,162,111,9,0.4338170324237214,-1.0166286003580864,0
-8,6,163,124,7,1.449813536152996,1.2624300972050062,1
-8,7,164,108,9,0.8621272110483021,-0.5284653415605093,1
-8,8,165,118,4,0.3234054613524826,1.1616916884353274,1
-8,9,166,124,2,0.5064088364815167,-0.4651367065068575,0
-8,10,167,109,1,1.4913472073701488,0.2991262309237505,0
-8,11,168,121,8,1.023767695317657,-0.055448324522217,0
-8,12,169,110,4,-1.4437814578576325,0.599379484168683,0
-8,13,170,120,4,-1.2314791793635456,0.3901346999802171,0
-8,14,171,114,4,0.7673842918563573,0.7395058195373676,2
-8,15,172,107,8,1.6847926118412546,-0.2618624769340897,2
-8,16,173,100,2,0.4514898135283557,-0.4452332175533571,0
-8,17,174,111,6,1.5702390064548917,0.7613097400139596,1
-8,18,175,107,9,-1.4233773553881364,-2.2941794906187303,2
-8,19,176,109,4,-0.8348156343282267,0.3184559910398074,0
-8,20,177,114,1,1.64301604024024,-0.8105101252279568,0
-8,21,178,102,6,0.5931810417739123,-0.4433977354954991,2
-8,22,179,126,2,0.2933912934374598,-0.2155182381754514,1
-8,23,180,115,2,1.0447321534051053,1.6744425413368984,2
-9,0,181,128,0,0.6181051753418312,-1.3708388861688323,0
-9,1,182,128,0,0.6181051753418312,-1.3708388861688323,0
-9,2,183,128,0,0.6181051753418312,-1.3708388861688323,0
-9,3,184,128,0,0.6181051753418312,-1.3708388861688323,0
-9,4,185,128,0,0.6181051753418312,-1.3708388861688323,0
-9,5,186,128,0,0.6181051753418312,-1.3708388861688323,0
-9,6,187,128,0,0.6181051753418312,-1.3708388861688323,0
-9,7,188,128,0,0.6181051753418312,-1.3708388861688323,0
-9,8,189,128,0,0.6181051753418312,-1.3708388861688323,0
-10,0,190,120,0,-0.4304087997159057,0.3153721331558907,1
-10,1,191,103,5,0.4244180231779655,-0.9094229514381224,2
-10,2,192,116,5,-0.0579477086770905,-0.7201079456603995,1
-10,3,193,122,6,0.3949805394801389,-0.3897487397244845,0
-10,4,194,100,6,-1.6105230975007343,-0.7172381117410999,0
-10,5,195,128,1,-2.243399918867632,0.2448124704232933,1
-10,6,196,102,7,0.9299617676237236,-0.099403340668543,1
-10,7,197,122,6,-1.4146582334712663,-1.521788254894823,0
-10,8,198,110,4,-0.7481546299358196,0.4304399046873204,1
-10,9,199,122,7,0.9978354826902832,0.8417705179306674,0
-10,10,200,107,5,-1.3269464311869232,-0.1530678319918289,2
-10,11,201,123,2,-1.5774782605931386,-0.6719432184447797,2
-10,12,202,103,8,-0.0946765671355123,1.828172446219632,0
-10,13,203,128,9,-1.0427121479775012,-1.2067768937681949,1
-10,14,204,106,7,-0.8627459284058191,-1.1178231135428731,1
-10,15,205,108,3,0.0995309724676549,0.6450065530334209,1
-10,16,206,119,4,-0.571201566340334,2.8820698132437697,1
-10,17,207,129,4,-1.0436603387986798,-0.5667820105949781,1
-11,0,208,120,0,-0.4304087997159057,0.3153721331558907,1
-11,1,209,103,5,0.4244180231779655,-0.9094229514381224,2
-11,2,210,116,5,-0.0579477086770905,-0.7201079456603995,1
-11,3,211,122,6,0.3949805394801389,-0.3897487397244845,0
-11,4,212,100,6,-1.6105230975007343,-0.7172381117410999,0
-11,5,213,128,1,-2.243399918867632,0.2448124704232933,1
-11,6,214,102,7,0.9299617676237236,-0.099403340668543,1
-11,7,215,122,6,-1.4146582334712663,-1.521788254894823,0
-11,8,216,110,4,-0.7481546299358196,0.4304399046873204,1
-11,9,217,122,7,0.9978354826902832,0.8417705179306674,0
-11,10,218,107,5,-1.3269464311869232,-0.1530678319918289,2
-11,11,219,123,2,-1.5774782605931386,-0.6719432184447797,2
-11,12,220,103,8,-0.0946765671355123,1.828172446219632,0
-11,13,221,128,9,-1.0427121479775012,-1.2067768937681949,1
-11,14,222,106,7,-0.8627459284058191,-1.1178231135428731,1
-11,15,223,108,3,0.0995309724676549,0.6450065530334209,1
-11,16,224,119,4,-0.571201566340334,2.8820698132437697,1
-11,17,225,129,4,-1.0436603387986798,-0.5667820105949781,1
-12,0,226,120,0,-0.4304087997159057,0.3153721331558907,1
-12,1,227,103,5,0.4244180231779655,-0.9094229514381224,2
-12,2,228,116,5,-0.0579477086770905,-0.7201079456603995,1
-12,3,229,122,6,0.3949805394801389,-0.3897487397244845,0
-12,4,230,100,6,-1.6105230975007343,-0.7172381117410999,0
-12,5,231,128,1,-2.243399918867632,0.2448124704232933,1
-12,6,232,102,7,0.9299617676237236,-0.099403340668543,1
-12,7,233,122,6,-1.4146582334712663,-1.521788254894823,0
-12,8,234,110,4,-0.7481546299358196,0.4304399046873204,1
-12,9,235,122,7,0.9978354826902832,0.8417705179306674,0
-12,10,236,107,5,-1.3269464311869232,-0.1530678319918289,2
-12,11,237,123,2,-1.5774782605931386,-0.6719432184447797,2
-12,12,238,103,8,-0.0946765671355123,1.828172446219632,0
-12,13,239,128,9,-1.0427121479775012,-1.2067768937681949,1
-12,14,240,106,7,-0.8627459284058191,-1.1178231135428731,1
-12,15,241,108,3,0.0995309724676549,0.6450065530334209,1
-12,16,242,119,4,-0.571201566340334,2.8820698132437697,1
-12,17,243,129,4,-1.0436603387986798,-0.5667820105949781,1
diff --git a/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-2.csv b/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-2.csv
deleted file mode 100644
index fd2991a4..00000000
--- a/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-2.csv
+++ /dev/null
@@ -1,255 +0,0 @@
-sequenceId,itemPosition,Unnamed: 0,itemId,supCat1,supReal2,supReal3,supCat4
-13,0,0,112,3,0.47512151913436,-0.5820086906467112,2
-13,1,1,104,9,-1.1261734468621212,-0.9089247216662516,1
-13,2,2,129,7,-1.2162003117631153,-0.5154371712252781,1
-13,3,3,121,9,-1.4933477031742597,-0.8796536665660435,0
-13,4,4,109,2,-0.9670816286563634,-0.0946835025191322,2
-13,5,5,111,6,0.2951585961176835,0.573453806301526,0
-13,6,6,121,9,-0.8256811879776176,-0.6726355586632142,0
-13,7,7,109,3,-0.2626654773388363,0.7338377851923579,1
-13,8,8,126,4,1.3128198055960525,0.4123843865529258,0
-13,9,9,112,3,-0.1724514169048517,1.0002758074691138,2
-13,10,10,104,4,0.9984558771504448,-1.6793033775681865,1
-13,11,11,103,9,0.8241769523505718,0.5192150697838572,0
-13,12,12,119,2,1.1524820376873743,-0.0684948380162701,2
-13,13,13,103,0,-0.4532287722802309,-1.012670179240466,1
-13,14,14,105,7,0.7406957728589633,0.4598005719286187,2
-13,15,15,121,6,-0.3046679002763786,2.228549215957172,0
-13,16,16,107,1,-0.0981028713050338,1.580761752579522,2
-13,17,17,128,4,-0.338687599861625,1.0274608136462915,1
-13,18,18,128,6,-0.5926329276128396,0.7105776900317172,2
-13,19,19,116,9,-0.8683614880285571,0.7000235215395574,1
-13,20,20,106,3,-0.2223944052810371,-0.1421408408449271,2
-13,21,21,103,9,0.2238516678750416,0.7671249818780963,1
-13,22,22,119,6,-0.4794418411187094,0.284269664190007,0
-13,23,23,105,8,-0.208702368387007,0.5367982644094035,0
-13,24,24,100,4,0.1273379928547726,0.9143038302645866,1
-14,0,25,113,9,-0.2656666807034851,0.4806224256538509,1
-14,1,26,116,1,-0.9754261015074585,0.2940960032574599,0
-14,2,27,105,5,-2.0639707193059498,1.2133396475903873,1
-14,3,28,113,6,0.7319848912899831,-0.4229377963857184,1
-14,4,29,108,1,-0.0908278640955465,0.5666495515437896,2
-14,5,30,118,5,0.2854174163780527,0.6009858839082124,0
-14,6,31,102,3,0.2014682827257396,1.7092664244525773,1
-14,7,32,102,0,0.1453056346528572,0.3399258904044571,1
-14,8,33,121,6,0.8095167753996217,-0.1079289200633795,1
-14,9,34,104,3,1.5900425772732012,0.6294206328381015,2
-14,10,35,111,1,-0.8178820250244837,1.4633194614817098,1
-14,11,36,110,6,-1.1811248427188683,0.5503790726147177,0
-14,12,37,102,4,-0.5521109827498613,0.2288111251968629,2
-14,13,38,102,6,0.3949748622436907,1.5447878604004366,0
-14,14,39,120,0,-0.6463976133595792,0.1630326559403528,2
-14,15,40,109,6,-1.2805957817399582,-1.1826482612552474,1
-14,16,41,101,8,0.5508900386572504,-0.1671730495935667,0
-14,17,42,111,1,0.0853086889333529,-0.0132633758837144,0
-14,18,43,127,2,-0.218769711385981,-0.7019854368140658,2
-14,19,44,105,0,-0.6197723706627636,2.138611526896666,1
-14,20,45,115,7,-1.1247441521636312,-0.4641464411878229,2
-14,21,46,104,5,-0.072555647708791,-0.3929353813928705,0
-15,0,47,112,3,1.3399026572550203,1.878090363680033,0
-15,1,48,124,5,-0.024381870075387,0.5316335161706456,0
-15,2,49,121,0,-0.9970857305430262,-1.383512008752633,2
-15,3,50,124,8,-0.8334960910074487,-0.9116180632797354,0
-15,4,51,129,6,0.3845601295654262,0.1807163239900383,1
-15,5,52,103,3,-0.2294227620946844,0.3477709783565773,0
-15,6,53,119,6,1.0169773111385545,-0.5910886720642256,1
-15,7,54,118,4,0.3363413463024886,-1.0001172593271828,1
-15,8,55,110,0,0.0097818169970301,-0.6059838835052326,1
-15,9,56,102,3,0.5901905216267234,0.3696348407961316,0
-15,10,57,121,8,-0.3329927633569184,-2.1592296837868172,0
-15,11,58,108,5,0.5469937351204301,0.1158729118533183,1
-15,12,59,101,6,1.4387947715516716,0.510184749224577,2
-15,13,60,102,4,-1.5928420817380986,-1.0181941589104413,1
-15,14,61,110,8,0.4144898850289152,-2.1012260841117647,2
-15,15,62,106,9,0.9626552542957164,-0.2641403390810031,1
-15,16,63,102,5,0.5938953228849851,1.126424989282014,0
-15,17,64,110,4,-0.6852888501385631,-0.2788158895740412,0
-15,18,65,109,4,-1.802304268569908,-0.3884284583050837,1
-15,19,66,115,2,-0.0129419618714691,-0.9695615693196264,0
-16,0,67,122,9,0.728096157555939,1.57860530312412,0
-16,1,68,129,5,0.7156279104852876,-0.4852580169423038,1
-16,2,69,101,0,1.3546839988588728,1.3115417566478216,0
-16,3,70,117,2,-1.232361381356479,-0.9375197390131628,0
-16,4,71,120,0,-0.0435872778479013,-1.4933926504379784,0
-16,5,72,127,8,0.5484525334475734,1.66698134913082,0
-16,6,73,122,4,1.307847634805456,0.4215958062784253,1
-16,7,74,112,4,-2.618532214839832,1.6374672768528042,2
-16,8,75,127,7,-0.8496513139782857,0.2645026021626481,0
-16,9,76,115,9,-0.7521255704289022,0.7526070369449078,1
-16,10,77,117,0,-2.126705535677708,0.2182067272217519,0
-16,11,78,102,5,0.8752645615044675,0.2730385089817146,2
-16,12,79,120,6,0.3561666568718871,-1.1795205381405394,2
-16,13,80,108,7,-0.5208614133489836,-0.300497076461078,1
-16,14,81,126,8,0.1299212462358277,-0.8405052153381654,2
-16,15,82,116,6,-0.3047310126274782,0.7356818372550097,2
-16,16,83,102,5,1.9115397604052724,-0.2388956656409064,0
-16,17,84,121,0,-0.8567878621140438,1.643155432693797,2
-16,18,85,118,8,0.9366327938393536,0.5319856670267262,0
-16,19,86,105,5,0.1183670068806765,-2.1326672344812105,0
-17,0,87,113,4,-0.1473154301210917,-1.6042550026469249,0
-17,1,88,127,3,-1.3056540391037692,-0.5997440870462475,0
-17,2,89,105,7,3.0274943343874083,-0.0520796489093672,0
-17,3,90,104,9,0.3617294328556023,-0.0883800576540255,0
-17,4,91,106,9,-0.4453650036042871,-0.8334261447948773,1
-17,5,92,118,0,0.6282853705023487,1.350321469218172,2
-17,6,93,112,0,0.9642562105511686,-0.9992263103431246,1
-17,7,94,122,0,-0.8330901074711051,-0.7635405661499053,2
-17,8,95,105,0,0.7981435937792382,-0.6666971915003546,2
-17,9,96,121,6,0.7995168560801881,-3.730120038128068,2
-17,10,97,119,9,-0.4778765668524142,-1.8018134368021577,1
-17,11,98,126,1,1.7731643504472827,-2.966586951844483,1
-17,12,99,120,9,1.2401001274351815,0.9763019660890628,1
-17,13,100,103,6,-0.6172799146920306,-1.9341599869280055,0
-17,14,101,117,1,0.7526864964032826,0.0115401685101198,0
-17,15,102,114,6,-0.3922040087053289,-1.7973367751171303,1
-18,0,103,120,4,1.3164444471305916,-0.5782800824093337,0
-18,1,104,123,4,0.420020710446368,-0.1850434000530602,0
-18,2,105,127,0,0.8921979069031415,-0.2332258635179755,0
-18,3,106,105,4,0.4279062963690861,-1.5395002171602334,1
-18,4,107,112,6,-0.1809155329521617,-1.106476206464435,0
-18,5,108,108,4,0.5335886571652085,-2.129243520644403,1
-18,6,109,123,8,-0.7825217022412374,0.4768143516638856,0
-18,7,110,108,6,-1.3297057074440568,-0.1723342300598751,1
-18,8,111,129,9,-0.9899126599978328,-0.6345577264308937,2
-18,9,112,122,0,-1.2241592816111664,-1.3588134360661253,0
-18,10,113,117,7,-1.6081151211440765,-1.115309203106232,1
-18,11,114,122,2,0.6893365374951431,1.660083512711184,1
-18,12,115,109,3,1.3184511797851892,0.5981510743069248,1
-18,13,116,112,1,0.2530927744153326,0.1762152669186038,2
-18,14,117,126,9,-0.7218134212123581,0.0995596156598381,1
-18,15,118,120,9,-0.0069204705176007,-0.134973127015259,2
-19,0,119,127,0,-0.4138548174567836,-0.7491928547518463,0
-19,1,120,119,8,-0.9217336181544192,2.0705334246918015,2
-19,2,121,127,4,-1.086483742151834,-0.714833705558571,1
-19,3,122,102,3,1.474288547233567,0.1631413469278299,1
-19,4,123,113,6,1.5875581988032164,1.027268717867479,0
-19,5,124,119,5,-1.8988657509221156,1.8808868244932129,2
-19,6,125,103,2,-0.4050814416560676,0.4437277986745326,2
-19,7,126,126,1,0.8726496980525889,0.1240530371932496,0
-19,8,127,128,4,1.8224737121284704,0.4209771365993368,2
-19,9,128,113,5,0.3293755829411718,0.6439655653176443,1
-19,10,129,111,7,1.1353736264714218,-0.0632902816234641,0
-19,11,130,117,9,-2.088924140703356,0.7873992206662046,2
-19,12,131,124,8,1.1776401371612333,0.1655764555283259,2
-19,13,132,100,2,0.3101310578176122,-0.543886007962999,1
-19,14,133,119,4,0.0489690373962899,-1.3584102327948733,1
-19,15,134,117,4,-1.122565585229101,-0.1804465985609838,0
-19,16,135,103,7,-0.1379191118897307,1.0044578374950142,1
-19,17,136,110,4,0.477337569358077,0.7537311601289707,2
-19,18,137,108,7,-0.0207534690671556,0.5841218654437593,0
-19,19,138,125,5,0.9348832049035216,-0.7226482893432934,0
-19,20,139,107,8,-0.804227609128954,-0.5571601176468748,0
-20,0,140,111,9,0.1431939714725609,0.0422524056965653,2
-20,1,141,112,3,-0.4605569829841085,-0.4325683214460127,1
-20,2,142,113,7,-1.0553504705140826,0.8465631724299651,0
-20,3,143,113,1,0.7206026881536415,0.2871208611901518,2
-20,4,144,113,8,-1.718237086897971,-0.7868091893675095,1
-20,5,145,121,1,0.527556676509205,-1.9318448023128916,1
-20,6,146,100,4,1.4729892021615898,0.8391818770701904,0
-20,7,147,129,6,-1.0059562338044414,0.0019652114728092,0
-20,8,148,100,4,1.6305569491478604,1.0951871930560133,1
-20,9,149,125,6,2.547370551387279,-1.0208678213187818,1
-20,10,150,118,9,2.033324551178638,-0.1154719415024776,1
-20,11,151,121,7,0.4043282940651028,-0.8707042594597949,2
-20,12,152,127,8,0.4191280421631976,0.0957855998835066,1
-20,13,153,121,1,-1.5587069063508987,0.1791648322349623,0
-20,14,154,129,1,-0.7830858337970409,0.0648510199038132,0
-20,15,155,113,8,0.1371199788701356,0.2161753878561426,0
-20,16,156,110,3,1.5499797081722422,1.0344576013849434,1
-21,0,157,101,6,1.818575785843776,-1.9949939475490075,2
-21,1,158,119,4,0.1183152450709592,-1.0344467157761543,0
-21,2,159,103,5,0.7652662329923993,-1.0019028093921996,1
-21,3,160,113,7,1.5686505707558396,-0.5766936717058257,2
-21,4,161,105,6,0.1255815698914305,-0.9386728917772926,2
-21,5,162,111,9,0.4338170324237214,-1.0166286003580864,0
-21,6,163,124,7,1.449813536152996,1.2624300972050062,1
-21,7,164,108,9,0.8621272110483021,-0.5284653415605093,1
-21,8,165,118,4,0.3234054613524826,1.1616916884353274,1
-21,9,166,124,2,0.5064088364815167,-0.4651367065068575,0
-21,10,167,109,1,1.4913472073701488,0.2991262309237505,0
-21,11,168,121,8,1.023767695317657,-0.055448324522217,0
-21,12,169,110,4,-1.4437814578576325,0.599379484168683,0
-21,13,170,120,4,-1.2314791793635456,0.3901346999802171,0
-21,14,171,114,4,0.7673842918563573,0.7395058195373676,2
-21,15,172,107,8,1.6847926118412546,-0.2618624769340897,2
-21,16,173,100,2,0.4514898135283557,-0.4452332175533571,0
-21,17,174,111,6,1.5702390064548917,0.7613097400139596,1
-21,18,175,107,9,-1.4233773553881364,-2.2941794906187303,2
-21,19,176,109,4,-0.8348156343282267,0.3184559910398074,0
-21,20,177,114,1,1.64301604024024,-0.8105101252279568,0
-21,21,178,102,6,0.5931810417739123,-0.4433977354954991,2
-21,22,179,126,2,0.2933912934374598,-0.2155182381754514,1
-21,23,180,115,2,1.0447321534051053,1.6744425413368984,2
-22,0,181,128,0,0.6181051753418312,-1.3708388861688323,0
-22,1,182,120,0,-0.4304087997159057,0.3153721331558907,1
-22,2,183,103,5,0.4244180231779655,-0.9094229514381224,2
-22,3,184,116,5,-0.0579477086770905,-0.7201079456603995,1
-22,4,185,122,6,0.3949805394801389,-0.3897487397244845,0
-22,5,186,100,6,-1.6105230975007343,-0.7172381117410999,0
-22,6,187,128,1,-2.243399918867632,0.2448124704232933,1
-22,7,188,102,7,0.9299617676237236,-0.099403340668543,1
-22,8,189,122,6,-1.4146582334712663,-1.521788254894823,0
-22,9,190,110,4,-0.7481546299358196,0.4304399046873204,1
-22,10,191,122,7,0.9978354826902832,0.8417705179306674,0
-22,11,192,107,5,-1.3269464311869232,-0.1530678319918289,2
-22,12,193,123,2,-1.5774782605931386,-0.6719432184447797,2
-22,13,194,103,8,-0.0946765671355123,1.828172446219632,0
-22,14,195,128,9,-1.0427121479775012,-1.2067768937681949,1
-22,15,196,106,7,-0.8627459284058191,-1.1178231135428731,1
-22,16,197,108,3,0.0995309724676549,0.6450065530334209,1
-22,17,198,119,4,-0.571201566340334,2.8820698132437697,1
-22,18,199,129,4,-1.0436603387986798,-0.5667820105949781,1
-23,0,200,120,0,-0.4304087997159057,0.3153721331558907,1
-23,1,201,103,5,0.4244180231779655,-0.9094229514381224,2
-23,2,202,116,5,-0.0579477086770905,-0.7201079456603995,1
-23,3,203,122,6,0.3949805394801389,-0.3897487397244845,0
-23,4,204,100,6,-1.6105230975007343,-0.7172381117410999,0
-23,5,205,128,1,-2.243399918867632,0.2448124704232933,1
-23,6,206,102,7,0.9299617676237236,-0.099403340668543,1
-23,7,207,122,6,-1.4146582334712663,-1.521788254894823,0
-23,8,208,110,4,-0.7481546299358196,0.4304399046873204,1
-23,9,209,122,7,0.9978354826902832,0.8417705179306674,0
-23,10,210,107,5,-1.3269464311869232,-0.1530678319918289,2
-23,11,211,123,2,-1.5774782605931386,-0.6719432184447797,2
-23,12,212,103,8,-0.0946765671355123,1.828172446219632,0
-23,13,213,128,9,-1.0427121479775012,-1.2067768937681949,1
-23,14,214,106,7,-0.8627459284058191,-1.1178231135428731,1
-23,15,215,108,3,0.0995309724676549,0.6450065530334209,1
-23,16,216,119,4,-0.571201566340334,2.8820698132437697,1
-23,17,217,129,4,-1.0436603387986798,-0.5667820105949781,1
-24,0,218,120,0,-0.4304087997159057,0.3153721331558907,1
-24,1,219,103,5,0.4244180231779655,-0.9094229514381224,2
-24,2,220,116,5,-0.0579477086770905,-0.7201079456603995,1
-24,3,221,122,6,0.3949805394801389,-0.3897487397244845,0
-24,4,222,100,6,-1.6105230975007343,-0.7172381117410999,0
-24,5,223,128,1,-2.243399918867632,0.2448124704232933,1
-24,6,224,102,7,0.9299617676237236,-0.099403340668543,1
-24,7,225,122,6,-1.4146582334712663,-1.521788254894823,0
-24,8,226,110,4,-0.7481546299358196,0.4304399046873204,1
-24,9,227,122,7,0.9978354826902832,0.8417705179306674,0
-24,10,228,107,5,-1.3269464311869232,-0.1530678319918289,2
-24,11,229,123,2,-1.5774782605931386,-0.6719432184447797,2
-24,12,230,103,8,-0.0946765671355123,1.828172446219632,0
-24,13,231,128,9,-1.0427121479775012,-1.2067768937681949,1
-24,14,232,106,7,-0.8627459284058191,-1.1178231135428731,1
-24,15,233,108,3,0.0995309724676549,0.6450065530334209,1
-24,16,234,119,4,-0.571201566340334,2.8820698132437697,1
-24,17,235,129,4,-1.0436603387986798,-0.5667820105949781,1
-25,0,236,120,0,-0.4304087997159057,0.3153721331558907,1
-25,1,237,103,5,0.4244180231779655,-0.9094229514381224,2
-25,2,238,116,5,-0.0579477086770905,-0.7201079456603995,1
-25,3,239,122,6,0.3949805394801389,-0.3897487397244845,0
-25,4,240,100,6,-1.6105230975007343,-0.7172381117410999,0
-25,5,241,128,1,-2.243399918867632,0.2448124704232933,1
-25,6,242,102,7,0.9299617676237236,-0.099403340668543,1
-25,7,243,122,6,-1.4146582334712663,-1.521788254894823,0
-25,8,244,110,4,-0.7481546299358196,0.4304399046873204,1
-25,9,245,122,7,0.9978354826902832,0.8417705179306674,0
-25,10,246,107,5,-1.3269464311869232,-0.1530678319918289,2
-25,11,247,123,2,-1.5774782605931386,-0.6719432184447797,2
-25,12,248,103,8,-0.0946765671355123,1.828172446219632,0
-25,13,249,128,9,-1.0427121479775012,-1.2067768937681949,1
-25,14,250,106,7,-0.8627459284058191,-1.1178231135428731,1
-25,15,251,108,3,0.0995309724676549,0.6450065530334209,1
-25,16,252,119,4,-0.571201566340334,2.8820698132437697,1
-25,17,253,129,4,-1.0436603387986798,-0.5667820105949781,1
diff --git a/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-3.csv b/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-3.csv
deleted file mode 100644
index cb7a3a2a..00000000
--- a/tests/resources/source_data/test-data-categorical-multi-file/test-data-categorical-mullti-file-3.csv
+++ /dev/null
@@ -1,255 +0,0 @@
-sequenceId,itemPosition,Unnamed: 0,itemId,supCat1,supReal2,supReal3,supCat4
-26,0,0,112,3,0.47512151913436,-0.5820086906467112,2
-26,1,1,104,9,-1.1261734468621212,-0.9089247216662516,1
-26,2,2,129,7,-1.2162003117631153,-0.5154371712252781,1
-26,3,3,121,9,-1.4933477031742597,-0.8796536665660435,0
-26,4,4,109,2,-0.9670816286563634,-0.0946835025191322,2
-26,5,5,111,6,0.2951585961176835,0.573453806301526,0
-26,6,6,121,9,-0.8256811879776176,-0.6726355586632142,0
-26,7,7,109,3,-0.2626654773388363,0.7338377851923579,1
-26,8,8,126,4,1.3128198055960525,0.4123843865529258,0
-26,9,9,112,3,-0.1724514169048517,1.0002758074691138,2
-26,10,10,104,4,0.9984558771504448,-1.6793033775681865,1
-26,11,11,103,9,0.8241769523505718,0.5192150697838572,0
-26,12,12,119,2,1.1524820376873743,-0.0684948380162701,2
-26,13,13,103,0,-0.4532287722802309,-1.012670179240466,1
-26,14,14,105,7,0.7406957728589633,0.4598005719286187,2
-26,15,15,121,6,-0.3046679002763786,2.228549215957172,0
-26,16,16,107,1,-0.0981028713050338,1.580761752579522,2
-26,17,17,128,4,-0.338687599861625,1.0274608136462915,1
-26,18,18,128,6,-0.5926329276128396,0.7105776900317172,2
-26,19,19,116,9,-0.8683614880285571,0.7000235215395574,1
-26,20,20,106,3,-0.2223944052810371,-0.1421408408449271,2
-26,21,21,103,9,0.2238516678750416,0.7671249818780963,1
-26,22,22,119,6,-0.4794418411187094,0.284269664190007,0
-26,23,23,105,8,-0.208702368387007,0.5367982644094035,0
-26,24,24,100,4,0.1273379928547726,0.9143038302645866,1
-27,0,25,113,9,-0.2656666807034851,0.4806224256538509,1
-27,1,26,116,1,-0.9754261015074585,0.2940960032574599,0
-27,2,27,105,5,-2.0639707193059498,1.2133396475903873,1
-27,3,28,113,6,0.7319848912899831,-0.4229377963857184,1
-27,4,29,108,1,-0.0908278640955465,0.5666495515437896,2
-27,5,30,118,5,0.2854174163780527,0.6009858839082124,0
-27,6,31,102,3,0.2014682827257396,1.7092664244525773,1
-27,7,32,102,0,0.1453056346528572,0.3399258904044571,1
-27,8,33,121,6,0.8095167753996217,-0.1079289200633795,1
-27,9,34,104,3,1.5900425772732012,0.6294206328381015,2
-27,10,35,111,1,-0.8178820250244837,1.4633194614817098,1
-27,11,36,110,6,-1.1811248427188683,0.5503790726147177,0
-27,12,37,102,4,-0.5521109827498613,0.2288111251968629,2
-27,13,38,102,6,0.3949748622436907,1.5447878604004366,0
-27,14,39,120,0,-0.6463976133595792,0.1630326559403528,2
-27,15,40,109,6,-1.2805957817399582,-1.1826482612552474,1
-27,16,41,101,8,0.5508900386572504,-0.1671730495935667,0
-27,17,42,111,1,0.0853086889333529,-0.0132633758837144,0
-27,18,43,127,2,-0.218769711385981,-0.7019854368140658,2
-27,19,44,105,0,-0.6197723706627636,2.138611526896666,1
-27,20,45,115,7,-1.1247441521636312,-0.4641464411878229,2
-27,21,46,104,5,-0.072555647708791,-0.3929353813928705,0
-28,0,47,112,3,1.3399026572550203,1.878090363680033,0
-28,1,48,124,5,-0.024381870075387,0.5316335161706456,0
-28,2,49,121,0,-0.9970857305430262,-1.383512008752633,2
-28,3,50,124,8,-0.8334960910074487,-0.9116180632797354,0
-28,4,51,129,6,0.3845601295654262,0.1807163239900383,1
-28,5,52,103,3,-0.2294227620946844,0.3477709783565773,0
-28,6,53,119,6,1.0169773111385545,-0.5910886720642256,1
-28,7,54,118,4,0.3363413463024886,-1.0001172593271828,1
-28,8,55,110,0,0.0097818169970301,-0.6059838835052326,1
-28,9,56,102,3,0.5901905216267234,0.3696348407961316,0
-28,10,57,121,8,-0.3329927633569184,-2.1592296837868172,0
-28,11,58,108,5,0.5469937351204301,0.1158729118533183,1
-28,12,59,101,6,1.4387947715516716,0.510184749224577,2
-28,13,60,102,4,-1.5928420817380986,-1.0181941589104413,1
-28,14,61,110,8,0.4144898850289152,-2.1012260841117647,2
-28,15,62,106,9,0.9626552542957164,-0.2641403390810031,1
-28,16,63,102,5,0.5938953228849851,1.126424989282014,0
-28,17,64,110,4,-0.6852888501385631,-0.2788158895740412,0
-28,18,65,109,4,-1.802304268569908,-0.3884284583050837,1
-28,19,66,115,2,-0.0129419618714691,-0.9695615693196264,0
-29,0,67,122,9,0.728096157555939,1.57860530312412,0
-29,1,68,129,5,0.7156279104852876,-0.4852580169423038,1
-29,2,69,101,0,1.3546839988588728,1.3115417566478216,0
-29,3,70,117,2,-1.232361381356479,-0.9375197390131628,0
-29,4,71,120,0,-0.0435872778479013,-1.4933926504379784,0
-29,5,72,127,8,0.5484525334475734,1.66698134913082,0
-29,6,73,122,4,1.307847634805456,0.4215958062784253,1
-29,7,74,112,4,-2.618532214839832,1.6374672768528042,2
-29,8,75,127,7,-0.8496513139782857,0.2645026021626481,0
-29,9,76,115,9,-0.7521255704289022,0.7526070369449078,1
-29,10,77,117,0,-2.126705535677708,0.2182067272217519,0
-29,11,78,102,5,0.8752645615044675,0.2730385089817146,2
-29,12,79,120,6,0.3561666568718871,-1.1795205381405394,2
-29,13,80,108,7,-0.5208614133489836,-0.300497076461078,1
-29,14,81,126,8,0.1299212462358277,-0.8405052153381654,2
-29,15,82,116,6,-0.3047310126274782,0.7356818372550097,2
-29,16,83,102,5,1.9115397604052724,-0.2388956656409064,0
-29,17,84,121,0,-0.8567878621140438,1.643155432693797,2
-29,18,85,118,8,0.9366327938393536,0.5319856670267262,0
-29,19,86,105,5,0.1183670068806765,-2.1326672344812105,0
-30,0,87,113,4,-0.1473154301210917,-1.6042550026469249,0
-30,1,88,127,3,-1.3056540391037692,-0.5997440870462475,0
-30,2,89,105,7,3.0274943343874083,-0.0520796489093672,0
-30,3,90,104,9,0.3617294328556023,-0.0883800576540255,0
-30,4,91,106,9,-0.4453650036042871,-0.8334261447948773,1
-30,5,92,118,0,0.6282853705023487,1.350321469218172,2
-30,6,93,112,0,0.9642562105511686,-0.9992263103431246,1
-30,7,94,122,0,-0.8330901074711051,-0.7635405661499053,2
-30,8,95,105,0,0.7981435937792382,-0.6666971915003546,2
-30,9,96,121,6,0.7995168560801881,-3.730120038128068,2
-30,10,97,119,9,-0.4778765668524142,-1.8018134368021577,1
-30,11,98,126,1,1.7731643504472827,-2.966586951844483,1
-30,12,99,120,9,1.2401001274351815,0.9763019660890628,1
-30,13,100,103,6,-0.6172799146920306,-1.9341599869280055,0
-30,14,101,117,1,0.7526864964032826,0.0115401685101198,0
-30,15,102,114,6,-0.3922040087053289,-1.7973367751171303,1
-31,0,103,120,4,1.3164444471305916,-0.5782800824093337,0
-31,1,104,123,4,0.420020710446368,-0.1850434000530602,0
-31,2,105,127,0,0.8921979069031415,-0.2332258635179755,0
-31,3,106,105,4,0.4279062963690861,-1.5395002171602334,1
-31,4,107,112,6,-0.1809155329521617,-1.106476206464435,0
-31,5,108,108,4,0.5335886571652085,-2.129243520644403,1
-31,6,109,123,8,-0.7825217022412374,0.4768143516638856,0
-31,7,110,108,6,-1.3297057074440568,-0.1723342300598751,1
-31,8,111,129,9,-0.9899126599978328,-0.6345577264308937,2
-31,9,112,122,0,-1.2241592816111664,-1.3588134360661253,0
-31,10,113,117,7,-1.6081151211440765,-1.115309203106232,1
-31,11,114,122,2,0.6893365374951431,1.660083512711184,1
-31,12,115,109,3,1.3184511797851892,0.5981510743069248,1
-31,13,116,112,1,0.2530927744153326,0.1762152669186038,2
-31,14,117,126,9,-0.7218134212123581,0.0995596156598381,1
-31,15,118,120,9,-0.0069204705176007,-0.134973127015259,2
-32,0,119,127,0,-0.4138548174567836,-0.7491928547518463,0
-32,1,120,119,8,-0.9217336181544192,2.0705334246918015,2
-32,2,121,127,4,-1.086483742151834,-0.714833705558571,1
-32,3,122,102,3,1.474288547233567,0.1631413469278299,1
-32,4,123,113,6,1.5875581988032164,1.027268717867479,0
-32,5,124,119,5,-1.8988657509221156,1.8808868244932129,2
-32,6,125,103,2,-0.4050814416560676,0.4437277986745326,2
-32,7,126,126,1,0.8726496980525889,0.1240530371932496,0
-32,8,127,128,4,1.8224737121284704,0.4209771365993368,2
-32,9,128,113,5,0.3293755829411718,0.6439655653176443,1
-32,10,129,111,7,1.1353736264714218,-0.0632902816234641,0
-32,11,130,117,9,-2.088924140703356,0.7873992206662046,2
-32,12,131,124,8,1.1776401371612333,0.1655764555283259,2
-32,13,132,100,2,0.3101310578176122,-0.543886007962999,1
-32,14,133,119,4,0.0489690373962899,-1.3584102327948733,1
-32,15,134,117,4,-1.122565585229101,-0.1804465985609838,0
-32,16,135,103,7,-0.1379191118897307,1.0044578374950142,1
-32,17,136,110,4,0.477337569358077,0.7537311601289707,2
-32,18,137,108,7,-0.0207534690671556,0.5841218654437593,0
-32,19,138,125,5,0.9348832049035216,-0.7226482893432934,0
-32,20,139,107,8,-0.804227609128954,-0.5571601176468748,0
-33,0,140,111,9,0.1431939714725609,0.0422524056965653,2
-33,1,141,112,3,-0.4605569829841085,-0.4325683214460127,1
-33,2,142,113,7,-1.0553504705140826,0.8465631724299651,0
-33,3,143,113,1,0.7206026881536415,0.2871208611901518,2
-33,4,144,113,8,-1.718237086897971,-0.7868091893675095,1
-33,5,145,121,1,0.527556676509205,-1.9318448023128916,1
-33,6,146,100,4,1.4729892021615898,0.8391818770701904,0
-33,7,147,129,6,-1.0059562338044414,0.0019652114728092,0
-33,8,148,100,4,1.6305569491478604,1.0951871930560133,1
-33,9,149,125,6,2.547370551387279,-1.0208678213187818,1
-33,10,150,118,9,2.033324551178638,-0.1154719415024776,1
-33,11,151,121,7,0.4043282940651028,-0.8707042594597949,2
-33,12,152,127,8,0.4191280421631976,0.0957855998835066,1
-33,13,153,121,1,-1.5587069063508987,0.1791648322349623,0
-33,14,154,129,1,-0.7830858337970409,0.0648510199038132,0
-33,15,155,113,8,0.1371199788701356,0.2161753878561426,0
-33,16,156,110,3,1.5499797081722422,1.0344576013849434,1
-34,0,157,101,6,1.818575785843776,-1.9949939475490075,2
-34,1,158,119,4,0.1183152450709592,-1.0344467157761543,0
-34,2,159,103,5,0.7652662329923993,-1.0019028093921996,1
-34,3,160,113,7,1.5686505707558396,-0.5766936717058257,2
-34,4,161,105,6,0.1255815698914305,-0.9386728917772926,2
-34,5,162,111,9,0.4338170324237214,-1.0166286003580864,0
-34,6,163,124,7,1.449813536152996,1.2624300972050062,1
-34,7,164,108,9,0.8621272110483021,-0.5284653415605093,1
-34,8,165,118,4,0.3234054613524826,1.1616916884353274,1
-34,9,166,124,2,0.5064088364815167,-0.4651367065068575,0
-34,10,167,109,1,1.4913472073701488,0.2991262309237505,0
-34,11,168,121,8,1.023767695317657,-0.055448324522217,0
-34,12,169,110,4,-1.4437814578576325,0.599379484168683,0
-34,13,170,120,4,-1.2314791793635456,0.3901346999802171,0
-34,14,171,114,4,0.7673842918563573,0.7395058195373676,2
-34,15,172,107,8,1.6847926118412546,-0.2618624769340897,2
-34,16,173,100,2,0.4514898135283557,-0.4452332175533571,0
-34,17,174,111,6,1.5702390064548917,0.7613097400139596,1
-34,18,175,107,9,-1.4233773553881364,-2.2941794906187303,2
-34,19,176,109,4,-0.8348156343282267,0.3184559910398074,0
-34,20,177,114,1,1.64301604024024,-0.8105101252279568,0
-34,21,178,102,6,0.5931810417739123,-0.4433977354954991,2
-34,22,179,126,2,0.2933912934374598,-0.2155182381754514,1
-34,23,180,115,2,1.0447321534051053,1.6744425413368984,2
-35,0,181,128,0,0.6181051753418312,-1.3708388861688323,0
-35,1,182,120,0,-0.4304087997159057,0.3153721331558907,1
-35,2,183,103,5,0.4244180231779655,-0.9094229514381224,2
-35,3,184,116,5,-0.0579477086770905,-0.7201079456603995,1
-35,4,185,122,6,0.3949805394801389,-0.3897487397244845,0
-35,5,186,100,6,-1.6105230975007343,-0.7172381117410999,0
-35,6,187,128,1,-2.243399918867632,0.2448124704232933,1
-35,7,188,102,7,0.9299617676237236,-0.099403340668543,1
-35,8,189,122,6,-1.4146582334712663,-1.521788254894823,0
-35,9,190,110,4,-0.7481546299358196,0.4304399046873204,1
-35,10,191,122,7,0.9978354826902832,0.8417705179306674,0
-35,11,192,107,5,-1.3269464311869232,-0.1530678319918289,2
-35,12,193,123,2,-1.5774782605931386,-0.6719432184447797,2
-35,13,194,103,8,-0.0946765671355123,1.828172446219632,0
-35,14,195,128,9,-1.0427121479775012,-1.2067768937681949,1
-35,15,196,106,7,-0.8627459284058191,-1.1178231135428731,1
-35,16,197,108,3,0.0995309724676549,0.6450065530334209,1
-35,17,198,119,4,-0.571201566340334,2.8820698132437697,1
-35,18,199,129,4,-1.0436603387986798,-0.5667820105949781,1
-36,0,200,120,0,-0.4304087997159057,0.3153721331558907,1
-36,1,201,103,5,0.4244180231779655,-0.9094229514381224,2
-36,2,202,116,5,-0.0579477086770905,-0.7201079456603995,1
-36,3,203,122,6,0.3949805394801389,-0.3897487397244845,0
-36,4,204,100,6,-1.6105230975007343,-0.7172381117410999,0
-36,5,205,128,1,-2.243399918867632,0.2448124704232933,1
-36,6,206,102,7,0.9299617676237236,-0.099403340668543,1
-36,7,207,122,6,-1.4146582334712663,-1.521788254894823,0
-36,8,208,110,4,-0.7481546299358196,0.4304399046873204,1
-36,9,209,122,7,0.9978354826902832,0.8417705179306674,0
-36,10,210,107,5,-1.3269464311869232,-0.1530678319918289,2
-36,11,211,123,2,-1.5774782605931386,-0.6719432184447797,2
-36,12,212,103,8,-0.0946765671355123,1.828172446219632,0
-36,13,213,128,9,-1.0427121479775012,-1.2067768937681949,1
-36,14,214,106,7,-0.8627459284058191,-1.1178231135428731,1
-36,15,215,108,3,0.0995309724676549,0.6450065530334209,1
-36,16,216,119,4,-0.571201566340334,2.8820698132437697,1
-36,17,217,129,4,-1.0436603387986798,-0.5667820105949781,1
-37,0,218,120,0,-0.4304087997159057,0.3153721331558907,1
-37,1,219,103,5,0.4244180231779655,-0.9094229514381224,2
-37,2,220,116,5,-0.0579477086770905,-0.7201079456603995,1
-37,3,221,122,6,0.3949805394801389,-0.3897487397244845,0
-37,4,222,100,6,-1.6105230975007343,-0.7172381117410999,0
-37,5,223,128,1,-2.243399918867632,0.2448124704232933,1
-37,6,224,102,7,0.9299617676237236,-0.099403340668543,1
-37,7,225,122,6,-1.4146582334712663,-1.521788254894823,0
-37,8,226,110,4,-0.7481546299358196,0.4304399046873204,1
-37,9,227,122,7,0.9978354826902832,0.8417705179306674,0
-37,10,228,107,5,-1.3269464311869232,-0.1530678319918289,2
-37,11,229,123,2,-1.5774782605931386,-0.6719432184447797,2
-37,12,230,103,8,-0.0946765671355123,1.828172446219632,0
-37,13,231,128,9,-1.0427121479775012,-1.2067768937681949,1
-37,14,232,106,7,-0.8627459284058191,-1.1178231135428731,1
-37,15,233,108,3,0.0995309724676549,0.6450065530334209,1
-37,16,234,119,4,-0.571201566340334,2.8820698132437697,1
-37,17,235,129,4,-1.0436603387986798,-0.5667820105949781,1
-38,0,236,120,0,-0.4304087997159057,0.3153721331558907,1
-38,1,237,103,5,0.4244180231779655,-0.9094229514381224,2
-38,2,238,116,5,-0.0579477086770905,-0.7201079456603995,1
-38,3,239,122,6,0.3949805394801389,-0.3897487397244845,0
-38,4,240,100,6,-1.6105230975007343,-0.7172381117410999,0
-38,5,241,128,1,-2.243399918867632,0.2448124704232933,1
-38,6,242,102,7,0.9299617676237236,-0.099403340668543,1
-38,7,243,122,6,-1.4146582334712663,-1.521788254894823,0
-38,8,244,110,4,-0.7481546299358196,0.4304399046873204,1
-38,9,245,122,7,0.9978354826902832,0.8417705179306674,0
-38,10,246,107,5,-1.3269464311869232,-0.1530678319918289,2
-38,11,247,123,2,-1.5774782605931386,-0.6719432184447797,2
-38,12,248,103,8,-0.0946765671355123,1.828172446219632,0
-38,13,249,128,9,-1.0427121479775012,-1.2067768937681949,1
-38,14,250,106,7,-0.8627459284058191,-1.1178231135428731,1
-38,15,251,108,3,0.0995309724676549,0.6450065530334209,1
-38,16,252,119,4,-0.571201566340334,2.8820698132437697,1
-38,17,253,129,4,-1.0436603387986798,-0.5667820105949781,1
diff --git a/tests/resources/source_data/test-data-categorical-multitarget-5.csv b/tests/resources/source_data/test-data-categorical-multitarget-5.csv
deleted file mode 100644
index bb36376c..00000000
--- a/tests/resources/source_data/test-data-categorical-multitarget-5.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal2,supReal3,supCat4
-0,0,112,3,0.47512151913436,-0.5820086906467112,2
-0,1,104,9,-1.1261734468621212,-0.9089247216662516,1
-0,2,129,7,-1.2162003117631153,-0.5154371712252781,1
-0,3,121,9,-1.4933477031742597,-0.8796536665660435,0
-0,4,109,2,-0.9670816286563634,-0.0946835025191322,2
-0,5,111,6,0.2951585961176835,0.573453806301526,0
-0,6,121,9,-0.8256811879776176,-0.6726355586632142,0
-0,7,109,3,-0.2626654773388363,0.7338377851923579,1
-0,8,126,4,1.3128198055960525,0.4123843865529258,0
-0,9,112,3,-0.1724514169048517,1.0002758074691138,2
-0,10,104,4,0.9984558771504448,-1.6793033775681865,1
-0,11,103,9,0.8241769523505718,0.5192150697838572,0
-0,12,119,2,1.1524820376873743,-0.0684948380162701,2
-0,13,103,0,-0.4532287722802309,-1.012670179240466,1
-0,14,105,7,0.7406957728589633,0.4598005719286187,2
-0,15,121,6,-0.3046679002763786,2.228549215957172,0
-0,16,107,1,-0.0981028713050338,1.580761752579522,2
-0,17,128,4,-0.338687599861625,1.0274608136462915,1
-0,18,128,6,-0.5926329276128396,0.7105776900317172,2
-0,19,116,9,-0.8683614880285571,0.7000235215395574,1
-0,20,106,3,-0.2223944052810371,-0.1421408408449271,2
-0,21,103,9,0.2238516678750416,0.7671249818780963,1
-0,22,119,6,-0.4794418411187094,0.284269664190007,0
-0,23,105,8,-0.208702368387007,0.5367982644094035,0
-0,24,100,4,0.1273379928547726,0.9143038302645866,1
-1,25,113,9,-0.2656666807034851,0.4806224256538509,1
-1,26,116,1,-0.9754261015074585,0.2940960032574599,0
-1,27,105,5,-2.0639707193059498,1.2133396475903873,1
-1,28,113,6,0.7319848912899831,-0.4229377963857184,1
-1,29,108,1,-0.0908278640955465,0.5666495515437896,2
-1,30,118,5,0.2854174163780527,0.6009858839082124,0
-1,31,102,3,0.2014682827257396,1.7092664244525773,1
-1,32,102,0,0.1453056346528572,0.3399258904044571,1
-1,33,121,6,0.8095167753996217,-0.1079289200633795,1
-1,34,104,3,1.5900425772732012,0.6294206328381015,2
-1,35,111,1,-0.8178820250244837,1.4633194614817098,1
-1,36,110,6,-1.1811248427188683,0.5503790726147177,0
-1,37,102,4,-0.5521109827498613,0.2288111251968629,2
-1,38,102,6,0.3949748622436907,1.5447878604004366,0
-1,39,120,0,-0.6463976133595792,0.1630326559403528,2
-1,40,109,6,-1.2805957817399582,-1.1826482612552474,1
-1,41,101,8,0.5508900386572504,-0.1671730495935667,0
-1,42,111,1,0.0853086889333529,-0.0132633758837144,0
-1,43,127,2,-0.218769711385981,-0.7019854368140658,2
-1,44,105,0,-0.6197723706627636,2.138611526896666,1
-1,45,115,7,-1.1247441521636312,-0.4641464411878229,2
-1,46,104,5,-0.072555647708791,-0.3929353813928705,0
-2,47,112,3,1.3399026572550203,1.878090363680033,0
-2,48,124,5,-0.024381870075387,0.5316335161706456,0
-2,49,121,0,-0.9970857305430262,-1.383512008752633,2
-2,50,124,8,-0.8334960910074487,-0.9116180632797354,0
-2,51,129,6,0.3845601295654262,0.1807163239900383,1
-2,52,103,3,-0.2294227620946844,0.3477709783565773,0
-2,53,119,6,1.0169773111385545,-0.5910886720642256,1
-2,54,118,4,0.3363413463024886,-1.0001172593271828,1
-2,55,110,0,0.0097818169970301,-0.6059838835052326,1
-2,56,102,3,0.5901905216267234,0.3696348407961316,0
-2,57,121,8,-0.3329927633569184,-2.1592296837868172,0
-2,58,108,5,0.5469937351204301,0.1158729118533183,1
-2,59,101,6,1.4387947715516716,0.510184749224577,2
-2,60,102,4,-1.5928420817380986,-1.0181941589104413,1
-2,61,110,8,0.4144898850289152,-2.1012260841117647,2
-2,62,106,9,0.9626552542957164,-0.2641403390810031,1
-2,63,102,5,0.5938953228849851,1.126424989282014,0
-2,64,110,4,-0.6852888501385631,-0.2788158895740412,0
-2,65,109,4,-1.802304268569908,-0.3884284583050837,1
-2,66,115,2,-0.0129419618714691,-0.9695615693196264,0
-3,67,122,9,0.728096157555939,1.57860530312412,0
-3,68,129,5,0.7156279104852876,-0.4852580169423038,1
-3,69,101,0,1.3546839988588728,1.3115417566478216,0
-3,70,117,2,-1.232361381356479,-0.9375197390131628,0
-3,71,120,0,-0.0435872778479013,-1.4933926504379784,0
-3,72,127,8,0.5484525334475734,1.66698134913082,0
-3,73,122,4,1.307847634805456,0.4215958062784253,1
-3,74,112,4,-2.618532214839832,1.6374672768528042,2
-3,75,127,7,-0.8496513139782857,0.2645026021626481,0
-3,76,115,9,-0.7521255704289022,0.7526070369449078,1
-3,77,117,0,-2.126705535677708,0.2182067272217519,0
-3,78,102,5,0.8752645615044675,0.2730385089817146,2
-3,79,120,6,0.3561666568718871,-1.1795205381405394,2
-3,80,108,7,-0.5208614133489836,-0.300497076461078,1
-3,81,126,8,0.1299212462358277,-0.8405052153381654,2
-3,82,116,6,-0.3047310126274782,0.7356818372550097,2
-3,83,102,5,1.9115397604052724,-0.2388956656409064,0
-3,84,121,0,-0.8567878621140438,1.643155432693797,2
-3,85,118,8,0.9366327938393536,0.5319856670267262,0
-3,86,105,5,0.1183670068806765,-2.1326672344812105,0
-4,87,113,4,-0.1473154301210917,-1.6042550026469249,0
-4,88,127,3,-1.3056540391037692,-0.5997440870462475,0
-4,89,105,7,3.0274943343874083,-0.0520796489093672,0
-4,90,104,9,0.3617294328556023,-0.0883800576540255,0
-4,91,106,9,-0.4453650036042871,-0.8334261447948773,1
-4,92,118,0,0.6282853705023487,1.350321469218172,2
-4,93,112,0,0.9642562105511686,-0.9992263103431246,1
-4,94,122,0,-0.8330901074711051,-0.7635405661499053,2
-4,95,105,0,0.7981435937792382,-0.6666971915003546,2
-4,96,121,6,0.7995168560801881,-3.730120038128068,2
-4,97,119,9,-0.4778765668524142,-1.8018134368021577,1
-4,98,126,1,1.7731643504472827,-2.966586951844483,1
-4,99,120,9,1.2401001274351815,0.9763019660890628,1
-4,100,103,6,-0.6172799146920306,-1.9341599869280055,0
-4,101,117,1,0.7526864964032826,0.0115401685101198,0
-4,102,114,6,-0.3922040087053289,-1.7973367751171303,1
-5,103,120,4,1.3164444471305916,-0.5782800824093337,0
-5,104,123,4,0.420020710446368,-0.1850434000530602,0
-5,105,127,0,0.8921979069031415,-0.2332258635179755,0
-5,106,105,4,0.4279062963690861,-1.5395002171602334,1
-5,107,112,6,-0.1809155329521617,-1.106476206464435,0
-5,108,108,4,0.5335886571652085,-2.129243520644403,1
-5,109,123,8,-0.7825217022412374,0.4768143516638856,0
-5,110,108,6,-1.3297057074440568,-0.1723342300598751,1
-5,111,129,9,-0.9899126599978328,-0.6345577264308937,2
-5,112,122,0,-1.2241592816111664,-1.3588134360661253,0
-5,113,117,7,-1.6081151211440765,-1.115309203106232,1
-5,114,122,2,0.6893365374951431,1.660083512711184,1
-5,115,109,3,1.3184511797851892,0.5981510743069248,1
-5,116,112,1,0.2530927744153326,0.1762152669186038,2
-5,117,126,9,-0.7218134212123581,0.0995596156598381,1
-5,118,120,9,-0.0069204705176007,-0.134973127015259,2
-6,119,127,0,-0.4138548174567836,-0.7491928547518463,0
-6,120,119,8,-0.9217336181544192,2.0705334246918015,2
-6,121,127,4,-1.086483742151834,-0.714833705558571,1
-6,122,102,3,1.474288547233567,0.1631413469278299,1
-6,123,113,6,1.5875581988032164,1.027268717867479,0
-6,124,119,5,-1.8988657509221156,1.8808868244932129,2
-6,125,103,2,-0.4050814416560676,0.4437277986745326,2
-6,126,126,1,0.8726496980525889,0.1240530371932496,0
-6,127,128,4,1.8224737121284704,0.4209771365993368,2
-6,128,113,5,0.3293755829411718,0.6439655653176443,1
-6,129,111,7,1.1353736264714218,-0.0632902816234641,0
-6,130,117,9,-2.088924140703356,0.7873992206662046,2
-6,131,124,8,1.1776401371612333,0.1655764555283259,2
-6,132,100,2,0.3101310578176122,-0.543886007962999,1
-6,133,119,4,0.0489690373962899,-1.3584102327948733,1
-6,134,117,4,-1.122565585229101,-0.1804465985609838,0
-6,135,103,7,-0.1379191118897307,1.0044578374950142,1
-6,136,110,4,0.477337569358077,0.7537311601289707,2
-6,137,108,7,-0.0207534690671556,0.5841218654437593,0
-6,138,125,5,0.9348832049035216,-0.7226482893432934,0
-6,139,107,8,-0.804227609128954,-0.5571601176468748,0
-7,140,111,9,0.1431939714725609,0.0422524056965653,2
-7,141,112,3,-0.4605569829841085,-0.4325683214460127,1
-7,142,113,7,-1.0553504705140826,0.8465631724299651,0
-7,143,113,1,0.7206026881536415,0.2871208611901518,2
-7,144,113,8,-1.718237086897971,-0.7868091893675095,1
-7,145,121,1,0.527556676509205,-1.9318448023128916,1
-7,146,100,4,1.4729892021615898,0.8391818770701904,0
-7,147,129,6,-1.0059562338044414,0.0019652114728092,0
-7,148,100,4,1.6305569491478604,1.0951871930560133,1
-7,149,125,6,2.547370551387279,-1.0208678213187818,1
-7,150,118,9,2.033324551178638,-0.1154719415024776,1
-7,151,121,7,0.4043282940651028,-0.8707042594597949,2
-7,152,127,8,0.4191280421631976,0.0957855998835066,1
-7,153,121,1,-1.5587069063508987,0.1791648322349623,0
-7,154,129,1,-0.7830858337970409,0.0648510199038132,0
-7,155,113,8,0.1371199788701356,0.2161753878561426,0
-7,156,110,3,1.5499797081722422,1.0344576013849434,1
-8,157,101,6,1.818575785843776,-1.9949939475490075,2
-8,158,119,4,0.1183152450709592,-1.0344467157761543,0
-8,159,103,5,0.7652662329923993,-1.0019028093921996,1
-8,160,113,7,1.5686505707558396,-0.5766936717058257,2
-8,161,105,6,0.1255815698914305,-0.9386728917772926,2
-8,162,111,9,0.4338170324237214,-1.0166286003580864,0
-8,163,124,7,1.449813536152996,1.2624300972050062,1
-8,164,108,9,0.8621272110483021,-0.5284653415605093,1
-8,165,118,4,0.3234054613524826,1.1616916884353274,1
-8,166,124,2,0.5064088364815167,-0.4651367065068575,0
-8,167,109,1,1.4913472073701488,0.2991262309237505,0
-8,168,121,8,1.023767695317657,-0.055448324522217,0
-8,169,110,4,-1.4437814578576325,0.599379484168683,0
-8,170,120,4,-1.2314791793635456,0.3901346999802171,0
-8,171,114,4,0.7673842918563573,0.7395058195373676,2
-8,172,107,8,1.6847926118412546,-0.2618624769340897,2
-8,173,100,2,0.4514898135283557,-0.4452332175533571,0
-8,174,111,6,1.5702390064548917,0.7613097400139596,1
-8,175,107,9,-1.4233773553881364,-2.2941794906187303,2
-8,176,109,4,-0.8348156343282267,0.3184559910398074,0
-8,177,114,1,1.64301604024024,-0.8105101252279568,0
-8,178,102,6,0.5931810417739123,-0.4433977354954991,2
-8,179,126,2,0.2933912934374598,-0.2155182381754514,1
-8,180,115,2,1.0447321534051053,1.6744425413368984,2
-9,181,128,0,0.6181051753418312,-1.3708388861688323,0
-9,182,120,0,-0.4304087997159057,0.3153721331558907,1
-9,183,103,5,0.4244180231779655,-0.9094229514381224,2
-9,184,116,5,-0.0579477086770905,-0.7201079456603995,1
-9,185,122,6,0.3949805394801389,-0.3897487397244845,0
-9,186,100,6,-1.6105230975007343,-0.7172381117410999,0
-9,187,128,1,-2.243399918867632,0.2448124704232933,1
-9,188,102,7,0.9299617676237236,-0.099403340668543,1
-9,189,122,6,-1.4146582334712663,-1.521788254894823,0
-9,190,110,4,-0.7481546299358196,0.4304399046873204,1
-9,191,122,7,0.9978354826902832,0.8417705179306674,0
-9,192,107,5,-1.3269464311869232,-0.1530678319918289,2
-9,193,123,2,-1.5774782605931386,-0.6719432184447797,2
-9,194,103,8,-0.0946765671355123,1.828172446219632,0
-9,195,128,9,-1.0427121479775012,-1.2067768937681949,1
-9,196,106,7,-0.8627459284058191,-1.1178231135428731,1
-9,197,108,3,0.0995309724676549,0.6450065530334209,1
-9,198,119,4,-0.571201566340334,2.8820698132437697,1
-9,199,129,4,-1.0436603387986798,-0.5667820105949781,1
diff --git a/tests/resources/source_data/test-data-categorical-precomputed-stats-negative.csv b/tests/resources/source_data/test-data-categorical-precomputed-stats-negative.csv
deleted file mode 100644
index 02a630c2..00000000
--- a/tests/resources/source_data/test-data-categorical-precomputed-stats-negative.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal2,supReal3,supCat4
-0,0,113,3,0.23756075956718,-1.1640173812934225,3
-0,1,105,9,-0.5630867234310606,-1.8178494433325032,1
-0,2,129,7,-0.6081001558815576,-1.0308743424505562,1
-0,3,121,9,-0.7466738515871298,-1.759307333132087,1
-0,4,109,3,-0.4835408143281817,-0.1893670050382644,3
-0,5,111,7,0.14757929805884176,1.146907612603052,1
-0,6,121,9,-0.4128405939888088,-1.3452711173264285,1
-0,7,109,3,-0.13133273866941814,1.4676755703847157,1
-0,8,127,5,0.6564099027980262,0.8247687731058516,1
-0,9,113,3,-0.08622570845242584,2.0005516149382276,3
-0,10,105,5,0.4992279385752224,-3.358606755136373,1
-0,11,103,9,0.4120884761752859,1.0384301395677145,1
-0,12,119,3,0.5762410188436872,-0.1369896760325402,3
-0,13,103,1,-0.22661438614011545,-2.025340358480932,1
-0,14,105,7,0.37034788642948163,0.9196011438572373,3
-0,15,121,7,-0.1523339501381893,4.457098431914344,1
-0,16,107,1,-0.0490514356525169,3.161523505159044,3
-0,17,129,5,-0.1693437999308125,2.054921627292583,1
-0,18,129,7,-0.2963164638064198,1.4211553800634344,3
-0,19,117,9,-0.43418074401427853,1.4000470430791148,1
-0,20,107,3,-0.11119720264051855,-0.2842816816898542,3
-0,21,103,9,0.1119258339375208,1.5342499637561926,1
-0,22,119,7,-0.2397209205593547,0.568539328380014,1
-0,23,105,9,-0.1043511841935035,1.073596528818807,1
-0,24,101,5,0.0636689964273863,1.8286076605291732,1
-1,25,113,9,-0.13283334035174255,0.9612448513077018,1
-1,26,117,1,-0.4877130507537292,0.5881920065149198,1
-1,27,105,5,-1.0319853596529749,2.4266792951807745,1
-1,28,113,7,0.36599244564499156,-0.8458755927714368,1
-1,29,109,1,-0.04541393204777325,1.1332991030875792,3
-1,30,119,5,0.14270870818902634,1.2019717678164248,1
-1,31,103,3,0.1007341413628698,3.4185328489051545,1
-1,32,103,1,0.0726528173264286,0.6798517808089142,1
-1,33,121,7,0.4047583876998109,-0.215857840126759,1
-1,34,105,3,0.7950212886366006,1.258841265676203,3
-1,35,111,1,-0.40894101251224185,2.9266389229634195,1
-1,36,111,7,-0.5905624213594342,1.1007581452294355,1
-1,37,103,5,-0.27605549137493063,0.4576222503937258,3
-1,38,103,7,0.19748743112184536,3.089575720800873,1
-1,39,121,1,-0.3231988066797896,0.3260653118807056,3
-1,40,109,7,-0.6402978908699791,-2.3652965225104947,1
-1,41,101,9,0.2754450193286252,-0.3343460991871334,1
-1,42,111,1,0.04265434446667645,-0.0265267517674288,1
-1,43,127,3,-0.1093848556929905,-1.4039708736281316,3
-1,44,105,1,-0.3098861853313818,4.277223053793332,1
-1,45,115,7,-0.5623720760818156,-0.9282928823756458,3
-1,46,105,5,-0.0362778238543955,-0.785870762785741,1
-2,47,113,3,0.6699513286275102,3.756180727360066,1
-2,48,125,5,-0.0121909350376935,1.0632670323412912,1
-2,49,121,1,-0.4985428652715131,-2.767024017505266,3
-2,50,125,9,-0.41674804550372435,-1.8232361265594708,1
-2,51,129,7,0.1922800647827131,0.3614326479800766,1
-2,52,103,3,-0.1147113810473422,0.6955419567131546,1
-2,53,119,7,0.5084886555692772,-1.182177344128451,1
-2,54,119,5,0.1681706731512443,-2.0002345186543655,1
-2,55,111,1,0.00489090849851505,-1.2119677670104652,1
-2,56,103,3,0.2950952608133617,0.7392696815922633,1
-2,57,121,9,-0.1664963816784592,-4.3184593675736345,1
-2,58,109,5,0.2734968675602151,0.2317458237066366,1
-2,59,101,7,0.7193973857758358,1.020369498449154,3
-2,60,103,5,-0.7964210408690493,-2.0363883178208826,1
-2,61,111,9,0.2072449425144576,-4.202452168223529,3
-2,62,107,9,0.4813276271478582,-0.5282806781620062,1
-2,63,103,5,0.29694766144249257,2.252849978564028,1
-2,64,111,5,-0.34264442506928156,-0.5576317791480824,1
-2,65,109,5,-0.901152134284954,-0.7768569166101674,1
-2,66,115,3,-0.00647098093573455,-1.939123138639253,1
-3,67,123,9,0.3640480787779695,3.15721060624824,1
-3,68,129,5,0.3578139552426438,-0.9705160338846076,1
-3,69,101,1,0.6773419994294364,2.623083513295643,1
-3,70,117,3,-0.6161806906782396,-1.8750394780263255,1
-3,71,121,1,-0.02179363892395065,-2.986785300875957,1
-3,72,127,9,0.2742262667237867,3.33396269826164,1
-3,73,123,5,0.653923817402728,0.8431916125568506,1
-3,74,113,5,-1.309266107419916,3.2749345537056085,3
-3,75,127,7,-0.4248256569891429,0.5290052043252962,1
-3,76,115,9,-0.3760627852144511,1.5052140738898157,1
-3,77,117,1,-1.063352767838854,0.4364134544435038,1
-3,78,103,5,0.43763228075223376,0.5460770179634292,3
-3,79,121,7,0.17808332843594354,-2.3590410762810787,3
-3,80,109,7,-0.2604307066744918,-0.600994152922156,1
-3,81,127,9,0.06496062311791385,-1.6810104306763307,3
-3,82,117,7,-0.1523655063137391,1.4713636745100194,3
-3,83,103,5,0.9557698802026362,-0.4777913312818128,1
-3,84,121,1,-0.4283939310570219,3.286310865387594,3
-3,85,119,9,0.4683163969196768,1.0639713340534525,1
-3,86,105,5,0.05918350344033825,-4.265334468962421,1
-4,87,113,5,-0.07365771506054584,-3.2085100052938498,1
-4,88,127,3,-0.6528270195518846,-1.199488174092495,1
-4,89,105,7,1.5137471671937042,-0.1041592978187344,1
-4,90,105,9,0.18086471642780114,-0.176760115308051,1
-4,91,107,9,-0.22268250180214355,-1.6668522895897546,1
-4,92,119,1,0.31414268525117434,2.700642938436344,3
-4,93,113,1,0.4821281052755843,-1.9984526206862492,1
-4,94,123,1,-0.41654505373555256,-1.5270811322998106,3
-4,95,105,1,0.3990717968896191,-1.3333943830007091,3
-4,96,121,7,0.39975842804009404,-7.460240076256136,3
-4,97,119,9,-0.2389382834262071,-3.6036268736043153,1
-4,98,127,1,0.8865821752236414,-5.933173903688966,1
-4,99,121,9,0.6200500637175907,1.9526039321781257,1
-4,100,103,7,-0.3086399573460153,-3.868319973856011,1
-4,101,117,1,0.3763432482016413,0.0230803370202396,1
-4,102,115,7,-0.19610200435266445,-3.5946735502342606,1
-5,103,121,5,0.6582222235652958,-1.1565601648186674,1
-5,104,123,5,0.210010355223184,-0.3700868001061204,1
-5,105,127,1,0.44609895345157075,-0.466451727035951,1
-5,106,105,5,0.21395314818454306,-3.079000434320467,1
-5,107,113,7,-0.09045776647608085,-2.21295241292887,1
-5,108,109,5,0.26679432858260427,-4.258487041288806,1
-5,109,123,9,-0.3912608511206187,0.9536287033277712,1
-5,110,109,7,-0.6648528537220284,-0.3446684601197502,1
-5,111,129,9,-0.4949563299989164,-1.2691154528617874,3
-5,112,123,1,-0.6120796408055832,-2.7176268721322505,1
-5,113,117,7,-0.8040575605720383,-2.230618406212464,1
-5,114,123,3,0.34466826874757156,3.320167025422368,1
-5,115,109,3,0.6592255898925946,1.1963021486138496,1
-5,116,113,1,0.1265463872076663,0.3524305338372076,3
-5,117,127,9,-0.36090671060617907,0.1991192313196762,1
-5,118,121,9,-0.00346023525880035,-0.269946254030518,3
-6,119,127,1,-0.2069274087283918,-1.4983857095036925,1
-6,120,119,9,-0.4608668090772096,4.141066849383603,3
-6,121,127,5,-0.543241871075917,-1.429667411117142,1
-6,122,103,3,0.7371442736167835,0.3262826938556598,1
-6,123,113,7,0.7937790994016082,2.054537435734958,1
-6,124,119,5,-0.9494328754610578,3.7617736489864257,3
-6,125,103,3,-0.2025407208280338,0.8874555973490652,3
-6,126,127,1,0.4363248490262944,0.2481060743864992,1
-6,127,129,5,0.9112368560642352,0.8419542731986736,3
-6,128,113,5,0.1646877914705859,1.2879311306352885,1
-6,129,111,7,0.5676868132357109,-0.1265805632469282,1
-6,130,117,9,-1.044462070351678,1.5747984413324092,3
-6,131,125,9,0.5888200685806166,0.3311529110566518,3
-6,132,101,3,0.1550655289088061,-1.087772015925998,1
-6,133,119,5,0.02448451869814495,-2.7168204655897465,1
-6,134,117,5,-0.5612827926145505,-0.3608931971219676,1
-6,135,103,7,-0.06895955594486534,2.0089156749900283,1
-6,136,111,5,0.2386687846790385,1.5074623202579414,3
-6,137,109,7,-0.0103767345335778,1.1682437308875186,1
-6,138,125,5,0.4674416024517608,-1.4452965786865868,1
-6,139,107,9,-0.402113804564477,-1.1143202352937496,1
-7,140,111,9,0.07159698573628046,0.0845048113931306,3
-7,141,113,3,-0.23027849149205426,-0.8651366428920254,1
-7,142,113,7,-0.5276752352570413,1.6931263448599303,1
-7,143,113,1,0.36030134407682074,0.5742417223803036,3
-7,144,113,9,-0.8591185434489855,-1.573618378735019,1
-7,145,121,1,0.2637783382546025,-3.863689604625783,1
-7,146,101,5,0.7364946010807949,1.6783637541403809,1
-7,147,129,7,-0.5029781169022207,0.0039304229456184,1
-7,148,101,5,0.8152784745739302,2.1903743861120266,1
-7,149,125,7,1.2736852756936394,-2.0417356426375637,1
-7,150,119,9,1.016662275589319,-0.2309438830049552,1
-7,151,121,7,0.2021641470325514,-1.7414085189195898,3
-7,152,127,9,0.2095640210815988,0.1915711997670132,1
-7,153,121,1,-0.7793534531754494,0.3583296644699246,1
-7,154,129,1,-0.39154291689852044,0.1297020398076264,1
-7,155,113,9,0.0685599894350678,0.4323507757122852,1
-7,156,111,3,0.7749898540861211,2.0689152027698867,1
-8,157,101,7,0.909287892921888,-3.989987895098015,3
-8,158,119,5,0.0591576225354796,-2.0688934315523086,1
-8,159,103,5,0.38263311649619963,-2.003805618784399,1
-8,160,113,7,0.7843252853779198,-1.1533873434116515,3
-8,161,105,7,0.06279078494571524,-1.8773457835545853,3
-8,162,111,9,0.2169085162118607,-2.033257200716173,1
-8,163,125,7,0.724906768076498,2.5248601944100124,1
-8,164,109,9,0.43106360552415107,-1.0569306831210186,1
-8,165,119,5,0.1617027306762413,2.323383376870655,1
-8,166,125,3,0.25320441824075834,-0.930273413013715,1
-8,167,109,1,0.7456736036850744,0.598252461847501,1
-8,168,121,9,0.5118838476588286,-0.110896649044434,1
-8,169,111,5,-0.7218907289288162,1.198758968337366,1
-8,170,121,5,-0.6157395896817728,0.7802693999604342,1
-8,171,115,5,0.38369214592817863,1.4790116390747352,3
-8,172,107,9,0.8423963059206273,-0.5237249538681794,3
-8,173,101,3,0.22574490676417786,-0.8904664351067142,1
-8,174,111,7,0.7851195032274458,1.5226194800279191,1
-8,175,107,9,-0.7116886776940682,-4.5883589812374606,3
-8,176,109,5,-0.41740781716411335,0.6369119820796147,1
-8,177,115,1,0.82150802012012,-1.6210202504559137,1
-8,178,103,7,0.29659052088695614,-0.8867954709909982,3
-8,179,127,3,0.1466956467187299,-0.4310364763509028,1
-8,180,115,3,0.5223660767025526,3.348885082673797,3
-9,181,129,1,0.3090525876709156,-2.7416777723376646,1
-9,182,121,1,-0.21520439985795284,0.6307442663117814,1
-9,183,103,5,0.21220901158898275,-1.8188459028762447,3
-9,184,117,5,-0.02897385433854525,-1.440215891320799,1
-9,185,123,7,0.19749026974006945,-0.779497479448969,1
-9,186,101,7,-0.8052615487503672,-1.4344762234821997,1
-9,187,129,1,-1.121699959433816,0.4896249408465866,1
-9,188,103,7,0.4649808838118618,-0.198806681337086,1
-9,189,123,7,-0.7073291167356331,-3.043576509789646,1
-9,190,111,5,-0.3740773149679098,0.8608798093746408,1
-9,191,123,7,0.4989177413451416,1.6835410358613347,1
-9,192,107,5,-0.6634732155934616,-0.3061356639836578,3
-9,193,123,3,-0.7887391302965693,-1.3438864368895593,3
-9,194,103,9,-0.04733828356775615,3.656344892439264,1
-9,195,129,9,-0.5213560739887506,-2.4135537875363897,1
-9,196,107,7,-0.43137296420290955,-2.2356462270857462,1
-9,197,109,3,0.04976548623382745,1.2900131060668418,1
-9,198,119,5,-0.285600783170167,5.764139626487539,1
-9,199,129,5,-0.5218301693993399,-1.1335640211899563,1
diff --git a/tests/resources/source_data/test-data-categorical-precomputed-stats.csv b/tests/resources/source_data/test-data-categorical-precomputed-stats.csv
deleted file mode 100644
index 734bd449..00000000
--- a/tests/resources/source_data/test-data-categorical-precomputed-stats.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal2,supReal3,supCat4,[mask]
-0,0,113,3,0.23756075956718,-1.1640173812934225,3,0
-0,1,105,9,-0.5630867234310606,-1.8178494433325032,1,0
-0,2,129,7,-0.6081001558815576,-1.0308743424505562,1,0
-0,3,121,9,-0.7466738515871298,-1.759307333132087,1,0
-0,4,109,3,-0.4835408143281817,-0.1893670050382644,3,0
-0,5,111,7,0.1475792980588417,1.146907612603052,1,0
-0,6,121,9,-0.4128405939888088,-1.3452711173264285,1,0
-0,7,109,3,-0.1313327386694181,1.4676755703847155,1,0
-0,8,127,5,0.6564099027980262,0.8247687731058516,1,0
-0,9,113,3,-0.0862257084524258,2.0005516149382276,3,0
-0,10,105,5,0.4992279385752224,-3.358606755136373,1,0
-0,11,103,9,0.4120884761752859,1.0384301395677145,1,0
-0,12,119,3,0.5762410188436872,-0.1369896760325402,3,0
-0,13,103,1,-0.2266143861401154,-2.025340358480932,1,0
-0,14,105,7,0.3703478864294816,0.9196011438572372,3,0
-0,15,121,7,-0.1523339501381893,4.457098431914344,1,0
-0,16,107,1,-0.0490514356525169,3.161523505159044,3,0
-0,17,129,5,-0.1693437999308125,2.054921627292583,1,0
-0,18,129,7,-0.2963164638064198,1.4211553800634344,3,0
-0,19,117,9,-0.4341807440142785,1.4000470430791148,1,0
-0,20,107,3,-0.1111972026405185,-0.2842816816898542,3,0
-0,21,103,9,0.1119258339375208,1.5342499637561926,1,0
-0,22,119,7,-0.2397209205593547,0.568539328380014,1,0
-0,23,105,9,-0.1043511841935035,1.073596528818807,1,0
-0,24,101,5,0.0636689964273863,1.8286076605291732,1,0
-1,25,113,9,-0.1328333403517425,0.9612448513077018,1,0
-1,26,117,1,-0.4877130507537292,0.5881920065149198,1,0
-1,27,105,5,-1.0319853596529749,2.4266792951807745,1,0
-1,28,113,7,0.3659924456449915,-0.8458755927714368,1,0
-1,29,109,1,-0.0454139320477732,1.1332991030875792,3,0
-1,30,119,5,0.1427087081890263,1.2019717678164248,1,0
-1,31,103,3,0.1007341413628698,3.4185328489051545,1,0
-1,32,103,1,0.0726528173264286,0.6798517808089142,1,0
-1,33,121,7,0.4047583876998109,-0.215857840126759,1,0
-1,34,105,3,0.7950212886366006,1.258841265676203,3,0
-1,35,111,1,-0.4089410125122418,2.9266389229634195,1,0
-1,36,111,7,-0.5905624213594342,1.1007581452294355,1,0
-1,37,103,5,-0.2760554913749306,0.4576222503937258,3,0
-1,38,103,7,0.1974874311218453,3.089575720800873,1,0
-1,39,121,1,-0.3231988066797896,0.3260653118807056,3,0
-1,40,109,7,-0.6402978908699791,-2.3652965225104947,1,0
-1,41,101,9,0.2754450193286252,-0.3343460991871334,1,0
-1,42,111,1,0.0426543444666764,-0.0265267517674288,1,0
-1,43,127,3,-0.1093848556929905,-1.4039708736281316,3,0
-1,44,105,1,-0.3098861853313818,4.277223053793332,1,0
-1,45,115,7,-0.5623720760818156,-0.9282928823756458,3,0
-1,46,105,5,-0.0362778238543955,-0.785870762785741,1,0
-2,47,113,3,0.6699513286275102,3.756180727360066,1,0
-2,48,125,5,-0.0121909350376935,1.0632670323412912,1,0
-2,49,121,1,-0.4985428652715131,-2.767024017505266,3,0
-2,50,125,9,-0.4167480455037243,-1.8232361265594708,1,0
-2,51,129,7,0.1922800647827131,0.3614326479800766,1,0
-2,52,103,3,-0.1147113810473422,0.6955419567131546,1,0
-2,53,119,7,0.5084886555692772,-1.182177344128451,1,0
-2,54,119,5,0.1681706731512443,-2.0002345186543655,1,0
-2,55,111,1,0.004890908498515,-1.2119677670104652,1,0
-2,56,103,3,0.2950952608133617,0.7392696815922633,1,0
-2,57,121,9,-0.1664963816784592,-4.3184593675736345,1,0
-2,58,109,5,0.2734968675602151,0.2317458237066366,1,0
-2,59,101,7,0.7193973857758358,1.020369498449154,3,0
-2,60,103,5,-0.7964210408690493,-2.0363883178208826,1,0
-2,61,111,9,0.2072449425144576,-4.202452168223529,3,0
-2,62,107,9,0.4813276271478582,-0.5282806781620062,1,0
-2,63,103,5,0.2969476614424925,2.252849978564028,1,0
-2,64,111,5,-0.3426444250692815,-0.5576317791480824,1,0
-2,65,109,5,-0.901152134284954,-0.7768569166101674,1,0
-2,66,115,3,-0.0064709809357345,-1.939123138639253,1,0
-3,67,123,9,0.3640480787779695,3.15721060624824,1,0
-3,68,129,5,0.3578139552426438,-0.9705160338846076,1,0
-3,69,101,1,0.6773419994294364,2.623083513295643,1,0
-3,70,117,3,-0.6161806906782396,-1.8750394780263255,1,0
-3,71,121,1,-0.0217936389239506,-2.986785300875957,1,0
-3,72,127,9,0.2742262667237867,3.33396269826164,1,0
-3,73,123,5,0.653923817402728,0.8431916125568506,1,0
-3,74,113,5,-1.309266107419916,3.2749345537056085,3,0
-3,75,127,7,-0.4248256569891429,0.5290052043252962,1,0
-3,76,115,9,-0.3760627852144511,1.5052140738898157,1,0
-3,77,117,1,-1.063352767838854,0.4364134544435038,1,0
-3,78,103,5,0.4376322807522337,0.5460770179634292,3,0
-3,79,121,7,0.1780833284359435,-2.3590410762810787,3,0
-3,80,109,7,-0.2604307066744918,-0.600994152922156,1,0
-3,81,127,9,0.0649606231179138,-1.6810104306763307,3,0
-3,82,117,7,-0.1523655063137391,1.4713636745100194,3,0
-3,83,103,5,0.9557698802026362,-0.4777913312818128,1,0
-3,84,121,1,-0.4283939310570219,3.286310865387594,3,0
-3,85,119,9,0.4683163969196768,1.0639713340534525,1,0
-3,86,105,5,0.0591835034403382,-4.265334468962421,1,0
-4,87,113,5,-0.0736577150605458,-3.2085100052938498,1,0
-4,88,127,3,-0.6528270195518846,-1.199488174092495,1,0
-4,89,105,7,1.5137471671937042,-0.1041592978187344,1,0
-4,90,105,9,0.1808647164278011,-0.176760115308051,1,0
-4,91,107,9,-0.2226825018021435,-1.6668522895897546,1,0
-4,92,119,1,0.3141426852511743,2.700642938436344,3,0
-4,93,113,1,0.4821281052755843,-1.9984526206862492,1,0
-4,94,123,1,-0.4165450537355525,-1.5270811322998106,3,0
-4,95,105,1,0.3990717968896191,-1.3333943830007091,3,0
-4,96,121,7,0.399758428040094,-7.460240076256136,3,0
-4,97,119,9,-0.2389382834262071,-3.6036268736043153,1,0
-4,98,127,1,0.8865821752236414,-5.933173903688966,1,0
-4,99,121,9,0.6200500637175907,1.9526039321781257,1,0
-4,100,103,7,-0.3086399573460153,-3.868319973856011,1,0
-4,101,117,1,0.3763432482016413,0.0230803370202396,1,0
-4,102,115,7,-0.1961020043526644,-3.5946735502342606,1,0
-5,103,121,5,0.6582222235652958,-1.1565601648186674,1,0
-5,104,123,5,0.210010355223184,-0.3700868001061204,1,0
-5,105,127,1,0.4460989534515707,-0.466451727035951,1,0
-5,106,105,5,0.213953148184543,-3.079000434320467,1,0
-5,107,113,7,-0.0904577664760808,-2.21295241292887,1,0
-5,108,109,5,0.2667943285826042,-4.258487041288806,1,0
-5,109,123,9,-0.3912608511206187,0.9536287033277712,1,0
-5,110,109,7,-0.6648528537220284,-0.3446684601197502,1,0
-5,111,129,9,-0.4949563299989164,-1.2691154528617874,3,0
-5,112,123,1,-0.6120796408055832,-2.7176268721322505,1,0
-5,113,117,7,-0.8040575605720383,-2.230618406212464,1,0
-5,114,123,3,0.3446682687475715,3.320167025422368,1,0
-5,115,109,3,0.6592255898925946,1.1963021486138496,1,0
-5,116,113,1,0.1265463872076663,0.3524305338372076,3,0
-5,117,127,9,-0.360906710606179,0.1991192313196762,1,0
-5,118,121,9,-0.0034602352588003,-0.269946254030518,3,0
-6,119,127,1,-0.2069274087283918,-1.4983857095036923,1,0
-6,120,119,9,-0.4608668090772096,4.141066849383603,3,0
-6,121,127,5,-0.543241871075917,-1.429667411117142,1,0
-6,122,103,3,0.7371442736167835,0.3262826938556598,1,0
-6,123,113,7,0.7937790994016082,2.054537435734958,1,0
-6,124,119,5,-0.9494328754610578,3.7617736489864257,3,0
-6,125,103,3,-0.2025407208280338,0.8874555973490652,3,0
-6,126,127,1,0.4363248490262944,0.2481060743864992,1,0
-6,127,129,5,0.9112368560642352,0.8419542731986736,3,0
-6,128,113,5,0.1646877914705859,1.2879311306352883,1,0
-6,129,111,7,0.5676868132357109,-0.1265805632469282,1,0
-6,130,117,9,-1.044462070351678,1.5747984413324092,3,0
-6,131,125,9,0.5888200685806166,0.3311529110566518,3,0
-6,132,101,3,0.1550655289088061,-1.087772015925998,1,0
-6,133,119,5,0.0244845186981449,-2.7168204655897465,1,0
-6,134,117,5,-0.5612827926145505,-0.3608931971219676,1,0
-6,135,103,7,-0.0689595559448653,2.0089156749900283,1,0
-6,136,111,5,0.2386687846790385,1.5074623202579414,3,0
-6,137,109,7,-0.0103767345335778,1.1682437308875186,1,0
-6,138,125,5,0.4674416024517608,-1.4452965786865868,1,0
-6,139,107,9,-0.402113804564477,-1.1143202352937496,1,0
-7,140,111,9,0.0715969857362804,0.0845048113931306,3,0
-7,141,113,3,-0.2302784914920542,-0.8651366428920254,1,0
-7,142,113,7,-0.5276752352570413,1.6931263448599303,1,0
-7,143,113,1,0.3603013440768207,0.5742417223803036,3,0
-7,144,113,9,-0.8591185434489855,-1.573618378735019,1,0
-7,145,121,1,0.2637783382546025,-3.863689604625783,1,0
-7,146,101,5,0.7364946010807949,1.6783637541403809,1,0
-7,147,129,7,-0.5029781169022207,0.0039304229456184,1,0
-7,148,101,5,0.8152784745739302,2.1903743861120266,1,0
-7,149,125,7,1.2736852756936394,-2.0417356426375637,1,0
-7,150,119,9,1.016662275589319,-0.2309438830049552,1,0
-7,151,121,7,0.2021641470325514,-1.7414085189195898,3,0
-7,152,127,9,0.2095640210815988,0.1915711997670132,1,0
-7,153,121,1,-0.7793534531754494,0.3583296644699246,1,0
-7,154,129,1,-0.3915429168985204,0.1297020398076264,1,0
-7,155,113,9,0.0685599894350678,0.4323507757122852,1,0
-7,156,111,3,0.7749898540861211,2.0689152027698867,1,0
-8,157,101,7,0.909287892921888,-3.989987895098015,3,0
-8,158,119,5,0.0591576225354796,-2.0688934315523086,1,0
-8,159,103,5,0.3826331164961996,-2.003805618784399,1,0
-8,160,113,7,0.7843252853779198,-1.1533873434116515,3,0
-8,161,105,7,0.0627907849457152,-1.8773457835545853,3,0
-8,162,111,9,0.2169085162118607,-2.033257200716173,1,0
-8,163,125,7,0.724906768076498,2.5248601944100124,1,0
-8,164,109,9,0.431063605524151,-1.0569306831210186,1,0
-8,165,119,5,0.1617027306762413,2.323383376870655,1,0
-8,166,125,3,0.2532044182407583,-0.930273413013715,1,0
-8,167,109,1,0.7456736036850744,0.598252461847501,1,0
-8,168,121,9,0.5118838476588286,-0.110896649044434,1,0
-8,169,111,5,-0.7218907289288162,1.198758968337366,1,0
-8,170,121,5,-0.6157395896817728,0.7802693999604342,1,0
-8,171,115,5,0.3836921459281786,1.4790116390747352,3,0
-8,172,107,9,0.8423963059206273,-0.5237249538681794,3,0
-8,173,101,3,0.2257449067641778,-0.8904664351067142,1,0
-8,174,111,7,0.7851195032274458,1.5226194800279191,1,0
-8,175,107,9,-0.7116886776940682,-4.5883589812374606,3,0
-8,176,109,5,-0.4174078171641133,0.6369119820796147,1,0
-8,177,115,1,0.82150802012012,-1.6210202504559137,1,0
-8,178,103,7,0.2965905208869561,-0.8867954709909982,3,0
-8,179,127,3,0.1466956467187299,-0.4310364763509028,1,0
-8,180,115,3,0.5223660767025526,3.348885082673797,3,0
-9,181,129,1,0.3090525876709156,-2.7416777723376646,1,0
-9,182,121,1,-0.2152043998579528,0.6307442663117814,1,0
-9,183,103,5,0.2122090115889827,-1.8188459028762447,3,0
-9,184,117,5,-0.0289738543385452,-1.440215891320799,1,0
-9,185,123,7,0.1974902697400694,-0.779497479448969,1,0
-9,186,101,7,-0.8052615487503672,-1.4344762234821995,1,0
-9,187,129,1,-1.121699959433816,0.4896249408465866,1,0
-9,188,103,7,0.4649808838118618,-0.198806681337086,1,0
-9,189,123,7,-0.7073291167356331,-3.043576509789646,1,0
-9,190,111,5,-0.3740773149679098,0.8608798093746408,1,0
-9,191,123,7,0.4989177413451416,1.6835410358613347,1,0
-9,192,107,5,-0.6634732155934616,-0.3061356639836578,3,0
-9,193,123,3,-0.7887391302965693,-1.343886436889559,3,0
-9,194,103,9,-0.0473382835677561,3.656344892439264,1,0
-9,195,129,9,-0.5213560739887506,-2.4135537875363897,1,0
-9,196,107,7,-0.4313729642029095,-2.2356462270857462,1,0
-9,197,109,3,0.0497654862338274,1.2900131060668418,1,0
-9,198,119,5,-0.285600783170167,5.764139626487539,1,0
-9,199,129,5,-0.5218301693993399,-1.1335640211899565,1,0
diff --git a/tests/resources/source_data/test-data-real-1-split1-autoregression.csv b/tests/resources/source_data/test-data-real-1-split1-autoregression.csv
deleted file mode 100644
index 5174b6ba..00000000
--- a/tests/resources/source_data/test-data-real-1-split1-autoregression.csv
+++ /dev/null
@@ -1,20 +0,0 @@
-sequenceId,subsequenceId,startItemPosition,leftPadLength,inputCol,8,7,6,5,4,3,2,1,0
-0,13,13,0,itemValue,-1.2113837237467866,1.1957454087349089,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-0.5880064200454953
-0,14,14,0,itemValue,1.1957454087349089,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-inf,-1.6146336831548016
-0,15,15,0,itemValue,-1.3864879826232943,-1.4489901585552898,0.7671059681713814,2.217195960015043,0.8077394764090938,-1.1685193385087969,-inf,-inf,-0.5557687212758077
-1,11,11,0,itemValue,-0.3979099218270587,0.1472006972746456,1.0744329431771849,2.098645946368935,-1.3233038228636609,0.3701880711691477,1.6509216329279448,0.695362468181269,-0.3227447762317759
-1,12,12,0,itemValue,0.1472006972746456,1.0744329431771849,2.098645946368935,-1.3233038228636609,0.3701880711691477,1.6509216329279448,0.695362468181269,-inf,-2.737064338735829
-2,9,9,0,itemValue,-1.9364972938429672,0.381774018107753,0.7895053840838986,0.071840823669383,-0.3274088457533277,0.8707377962958142,1.4490373682705944,1.0572550888530108,0.0765690379670138
-2,10,10,0,itemValue,0.381774018107753,0.7895053840838986,0.071840823669383,-0.3274088457533277,0.8707377962958142,1.4490373682705944,1.0572550888530108,-inf,0.476280644045447
-3,9,9,0,itemValue,-0.1291893885530777,-0.8243404591242441,-1.6194781231152326,0.1530304968195626,-0.4118502341991996,-0.7016026835022849,0.2909377683444284,1.9914944653082405,0.2070996307773291
-3,10,10,0,itemValue,-0.8243404591242441,-1.6194781231152326,0.1530304968195626,-0.4118502341991996,-0.7016026835022849,0.2909377683444284,1.9914944653082405,-inf,-0.0923048918846115
-4,6,6,0,itemValue,-0.3473441905826143,-0.9526372634835768,0.3152110308127366,1.0910573998928226,0.4164594405061328,0.142507586519504,1.0387354362777137,0.4461925335963277,1.6935115804711565
-5,6,6,0,itemValue,0.607763764789741,-2.458568025269152,-0.8214502614252478,-1.5396443182660882,-0.3113729142601391,-1.7463565527326972,0.441061415952748,-0.7445626882829559,-0.4361008211094133
-6,10,10,0,itemValue,1.5641135703291782,0.8372376866758972,-0.927489932456818,1.1789898362189837,-1.7300169078644538,-0.0513293403405074,0.013139305117334,0.6322625662411738,-1.0028799371032169
-6,11,11,0,itemValue,0.8372376866758972,-0.927489932456818,1.1789898362189837,-1.7300169078644538,-0.0513293403405074,0.013139305117334,0.6322625662411738,-inf,0.1685537328210896
-7,7,7,0,itemValue,-1.7680402671702622,-1.1479572969796248,0.366441270630024,0.9519780800395158,0.3286902962731986,1.3789080143759465,0.3820587792655491,1.5380644602548283,0.4910444675915603
-8,12,12,0,itemValue,-0.7095652049712206,0.116749400524584,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,1.2543512347210088
-8,13,13,0,itemValue,0.116749400524584,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,-inf,-0.6162672551654905
-8,14,14,0,itemValue,-0.2458275472980247,1.105017102819967,0.4151322337650286,0.6678888872763786,-0.390515971944902,0.6056203083985299,-inf,-inf,1.0886264665030447
-9,8,8,0,itemValue,0.2102757881460243,-0.756823800221343,1.0186924371356785,0.618472230767961,0.2340549739792952,-0.2583133906283459,-1.781087365783252,0.5179010995396939,-0.2841126785745299
-9,9,9,0,itemValue,-0.756823800221343,1.0186924371356785,0.618472230767961,0.2340549739792952,-0.2583133906283459,-1.781087365783252,0.5179010995396939,-inf,1.2530292363758264
diff --git a/tests/resources/source_data/test-data-real-1.csv b/tests/resources/source_data/test-data-real-1.csv
deleted file mode 100644
index 66d09a2c..00000000
--- a/tests/resources/source_data/test-data-real-1.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemValue
-0,0,0.0944992664850621
-0,1,1.3739480999676186
-0,2,-1.107338794524096
-0,3,0.2881636605223107
-0,4,-0.5561339361410481
-0,5,2.2131767781475062
-0,6,0.0676768816653241
-0,7,0.9543574448876856
-0,8,-0.7449642120093283
-0,9,0.3111266693612036
-0,10,-0.2953998645085581
-0,11,-0.2009927247615343
-0,12,0.6930672112064217
-0,13,-1.2113837237467866
-0,14,1.1957454087349089
-0,15,-1.3864879826232943
-0,16,-1.4489901585552898
-0,17,0.7671059681713814
-0,18,2.217195960015043
-0,19,0.8077394764090938
-0,20,-1.1685193385087969
-0,21,0.1835258824568895
-0,22,1.0433230518948158
-0,23,-2.7886385588069658
-0,24,1.8634021111854504
-1,25,-1.877830762650686
-1,26,-0.1469148118828995
-1,27,-0.8823233933435725
-1,28,-0.8089917990141724
-1,29,-0.3943085765676132
-1,30,-0.0337570976576666
-1,31,0.0213140323834337
-1,32,1.1282923510450278
-1,33,-0.8221263385452469
-1,34,1.8168113502009495
-1,35,1.660137633752651
-1,36,-0.3979099218270587
-1,37,0.1472006972746456
-1,38,1.0744329431771849
-1,39,2.098645946368935
-1,40,-1.3233038228636609
-1,41,0.3701880711691477
-1,42,1.6509216329279448
-1,43,0.695362468181269
-1,44,-0.6121856652552128
-1,45,0.3051518245496207
-1,46,0.0659683229061858
-2,47,0.8083981254449394
-2,48,-0.2711079427384384
-2,49,-0.4294965087158796
-2,50,0.3670981012520712
-2,51,0.3410588078178669
-2,52,-0.6929642964935003
-2,53,0.0199389347673413
-2,54,0.2456847767142994
-2,55,-1.0992550106377
-2,56,-1.9364972938429672
-2,57,0.381774018107753
-2,58,0.7895053840838986
-2,59,0.071840823669383
-2,60,-0.3274088457533277
-2,61,0.8707377962958142
-2,62,1.4490373682705944
-2,63,1.0572550888530108
-2,64,-0.9835633674955572
-2,65,1.3265001144866806
-2,66,1.1741329477581235
-3,67,-0.674407462500132
-3,68,-1.0081812552074476
-3,69,-1.360915051647735
-3,70,0.6786803068042123
-3,71,1.577569579504721
-3,72,-1.3526971290727938
-3,73,-0.0583685307039785
-3,74,-0.934751987891592
-3,75,-0.9347505632972628
-3,76,-0.1291893885530777
-3,77,-0.8243404591242441
-3,78,-1.6194781231152326
-3,79,0.1530304968195626
-3,80,-0.4118502341991996
-3,81,-0.7016026835022849
-3,82,0.2909377683444284
-3,83,1.9914944653082405
-3,84,0.4918751257304392
-3,85,0.9085888103495288
-3,86,0.5038597933688415
-4,87,0.1086217083079374
-4,88,0.205672331266788
-4,89,0.1226449026453264
-4,90,-1.036603186664228
-4,91,-0.7458776587447739
-4,92,-0.5935092863545317
-4,93,-0.3473441905826143
-4,94,-0.9526372634835768
-4,95,0.3152110308127366
-4,96,1.0910573998928226
-4,97,0.4164594405061328
-4,98,0.142507586519504
-4,99,1.0387354362777137
-4,100,0.4461925335963277
-4,101,-1.991929158454544
-4,102,-0.0072582312611647
-5,103,-0.9376599918993224
-5,104,-0.0781662582621091
-5,105,-0.0496679031411692
-5,106,0.8757687954793659
-5,107,1.714495826544065
-5,108,0.166882277370868
-5,109,0.607763764789741
-5,110,-2.4585680252691517
-5,111,-0.8214502614252478
-5,112,-1.5396443182660882
-5,113,-0.3113729142601391
-5,114,-1.7463565527326972
-5,115,0.441061415952748
-5,116,-0.7445626882829559
-5,117,2.414825155527423
-5,118,-0.0929611777149686
-6,119,0.9924512124248496
-6,120,-1.471083324581432
-6,121,0.6301997266816972
-6,122,0.4611731584188029
-6,123,0.7521129894916484
-6,124,-0.6007669114616375
-6,125,-0.5125436647631566
-6,126,-0.1866480466015074
-6,127,-0.3415916219347562
-6,128,-0.2405580845441048
-6,129,1.5641135703291782
-6,130,0.8372376866758972
-6,131,-0.927489932456818
-6,132,1.1789898362189837
-6,133,-1.7300169078644538
-6,134,-0.0513293403405074
-6,135,0.013139305117334
-6,136,0.6322625662411738
-6,137,-0.5465139969051596
-6,138,0.6671200729278953
-6,139,-1.6313917054076197
-7,140,-0.054998716485888
-7,141,0.135173804066526
-7,142,1.3068582712356975
-7,143,0.4006903121515655
-7,144,-0.5169207673195412
-7,145,0.2551214405007075
-7,146,2.0827074501312857
-7,147,-1.7680402671702622
-7,148,-1.1479572969796248
-7,149,0.366441270630024
-7,150,0.9519780800395158
-7,151,0.3286902962731986
-7,152,1.3789080143759465
-7,153,0.3820587792655491
-7,154,1.5380644602548283
-7,155,-0.8696259068344739
-7,156,-0.8222787595288253
-8,157,-0.9631386888242454
-8,158,-0.8540642049866167
-8,159,-1.7854123996764923
-8,160,0.1809010985046613
-8,161,1.0597440854253286
-8,162,1.5142826396791564
-8,163,-0.0294168686901563
-8,164,-0.5760408952983338
-8,165,0.643841661526263
-8,166,0.8585268109580487
-8,167,-0.4275970481511105
-8,168,-2.3153389439023173
-8,169,-0.7095652049712206
-8,170,0.116749400524584
-8,171,-0.2458275472980247
-8,172,1.105017102819967
-8,173,0.4151322337650286
-8,174,0.6678888872763786
-8,175,-0.390515971944902
-8,176,0.6056203083985299
-8,177,-2.4432428802949824
-8,178,0.0713429957179392
-8,179,0.3474149331272206
-8,180,1.0900435015932757
-9,181,-2.3620753256776106
-9,182,-0.7864276689543984
-9,183,0.8562240736290871
-9,184,0.0176927129678913
-9,185,-1.604293154787326
-9,186,0.7908622857384023
-9,187,-0.3924475723433511
-9,188,0.7208186380322986
-9,189,0.2102757881460243
-9,190,-0.756823800221343
-9,191,1.0186924371356785
-9,192,0.618472230767961
-9,193,0.2340549739792952
-9,194,-0.2583133906283459
-9,195,-1.781087365783252
-9,196,0.5179010995396939
-9,197,-0.1764631825530115
-9,198,-0.5558421321699659
-9,199,0.0972090807826861
diff --git a/tests/resources/source_data/test-data-real-3.csv b/tests/resources/source_data/test-data-real-3.csv
deleted file mode 100644
index 56516148..00000000
--- a/tests/resources/source_data/test-data-real-3.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemValue,supReal1,supReal2
-0,0,1.143956134511096,0.3156304921985627,0.47512151913436
-0,1,1.3956844750568178,-0.1822429283186853,-1.1261734468621212
-0,2,-0.1715630160735158,0.4013693011536511,-1.2162003117631153
-0,3,-1.5550080594262743,1.132244341738739,-1.4933477031742597
-0,4,-0.5094348768207044,-0.0429108458575636,-0.9670816286563634
-0,5,1.1182473810170177,0.9886909669935166,0.2951585961176835
-0,6,-0.243231053806412,0.6038345786438268,-0.8256811879776176
-0,7,-0.8727324468448932,0.4894333898098869,-0.2626654773388363
-0,8,-1.0162859523607557,0.4068867065004759,1.3128198055960525
-0,9,-2.14929390417939,-0.1378925185193623,-0.1724514169048517
-0,10,0.3589567696659215,0.0559662205535733,0.9984558771504448
-0,11,2.309501640646432,0.3821088546568114,0.8241769523505718
-0,12,-0.4907019447464217,-0.3845457511930897,1.1524820376873743
-0,13,-0.5708572484953972,-0.4576193768022207,-0.4532287722802309
-0,14,-1.5153605543961426,-0.2529049831249026,0.7406957728589633
-0,15,-2.260734238316745,0.4879281081569628,-0.3046679002763786
-0,16,-0.6956194490834757,0.4519709748946399,-0.0981028713050338
-0,17,0.0055059262205625,0.0919939041535107,-0.338687599861625
-0,18,-0.6552582609410857,-0.022186056317866,-0.5926329276128396
-0,19,-0.3138800389908938,0.6891927961331837,-0.8683614880285571
-0,20,-1.4552229530328598,-0.6102957802065466,-0.2223944052810371
-0,21,-0.917132691276364,-0.1581526752807625,0.2238516678750416
-0,22,-0.018834301332115,-1.731035691880018,-0.4794418411187094
-0,23,0.7626086190249719,-1.7639258336612718,-0.208702368387007
-0,24,-0.6655988005097199,0.0060505157900411,0.1273379928547726
-1,25,-0.1809480099771767,-1.1712166898091485,-0.2656666807034851
-1,26,1.5281650764902397,-1.38055496916859,-0.9754261015074585
-1,27,0.5058577252234937,-2.297511461684582,-2.0639707193059498
-1,28,-0.2541536631753138,0.6851795700949505,0.7319848912899831
-1,29,0.619739160843701,-3.008671660328942,-0.0908278640955465
-1,30,0.146312811906593,1.707571601719115,0.2854174163780527
-1,31,-2.184392118921295,1.5360119663914973,0.2014682827257396
-1,32,1.1174230750757488,-0.0489276784398041,0.1453056346528572
-1,33,0.3367012529248535,-1.1192225290167672,0.8095167753996217
-1,34,0.6670158490102784,-0.3066059502009673,1.5900425772732012
-1,35,0.755078941326885,-0.6835362360694909,-0.8178820250244837
-1,36,-0.3676490200669038,-1.0611474620151,-1.1811248427188683
-1,37,-0.2710739102647702,-0.0916301757399661,-0.5521109827498613
-1,38,-0.2661160554693632,0.8005779139015006,0.3949748622436907
-1,39,-0.7351109737492553,0.1853694198501086,-0.6463976133595792
-1,40,0.4429417228949345,-1.4889159327953163,-1.2805957817399582
-1,41,-0.4069743745970505,1.3650539499845882,0.5508900386572504
-1,42,-0.0082860827227986,0.2518340709808602,0.0853086889333529
-1,43,0.0036562589715113,-0.8363773494419445,-0.218769711385981
-1,44,0.6757516080507786,-0.0137686912539288,-0.6197723706627636
-1,45,-1.707314817367056,0.802155664664332,-1.1247441521636312
-1,46,0.2136335797454943,0.3104002443287074,-0.072555647708791
-2,47,-0.433268572989774,0.0244865632771758,1.3399026572550203
-2,48,-0.3054040190508337,-1.2193671668665338,-0.024381870075387
-2,49,-0.1002680245383688,0.1492885707520549,-0.9970857305430262
-2,50,-1.467295698454029,-0.2767660813092324,-0.8334960910074487
-2,51,0.5548011155624822,0.236707130088914,0.3845601295654262
-2,52,-0.8783684024032121,2.336972941172497,-0.2294227620946844
-2,53,1.2017845416788098,-0.2258905493628782,1.0169773111385545
-2,54,0.7535596895960655,-0.4913060708854921,0.3363413463024886
-2,55,0.0897222058021331,0.4172472822212967,0.0097818169970301
-2,56,-1.8197908527299689,0.9623048127332562,0.5901905216267234
-2,57,0.7335101691112972,-1.6732811417083997,-0.3329927633569184
-2,58,-0.823192331827531,-1.1068971276083273,0.5469937351204301
-2,59,-1.029747086224018,-0.6167477123403153,1.4387947715516716
-2,60,1.426655949860882,1.8489599120038709,-1.5928420817380986
-2,61,0.0104728571593871,0.0596136644264906,0.4144898850289152
-2,62,-0.9525513995666236,-0.9180104246891572,0.9626552542957164
-2,63,-0.1861821362644952,-1.293207193364908,0.5938953228849851
-2,64,1.5198235795189494,0.8977741478374721,-0.6852888501385631
-2,65,1.0646831051806884,1.9978983207730696,-1.802304268569908
-2,66,-0.6824862408874952,-0.4106770921604671,-0.0129419618714691
-3,67,0.0263588534786415,0.0512576888774666,0.728096157555939
-3,68,0.9157116179597412,1.10900057064673,0.7156279104852876
-3,69,-2.6428526570265496,0.4872815209095086,1.3546839988588728
-3,70,-1.022737078099231,0.3109816751133513,-1.232361381356479
-3,71,-1.9048298880429355,1.981791566013018,-0.0435872778479013
-3,72,0.2061905652506893,1.7982660199482663,0.5484525334475734
-3,73,0.4677446998552286,-1.3021452513832306,1.307847634805456
-3,74,-0.3242057413171001,0.2039687473138783,-2.618532214839832
-3,75,-0.954384415885333,0.0319028974699685,-0.8496513139782857
-3,76,0.1542312008748267,0.0463832263973931,-0.7521255704289022
-3,77,-0.9423760670055648,1.160576963665477,-2.126705535677708
-3,78,1.5710522049038294,0.766752760239803,0.8752645615044675
-3,79,0.0651964031994713,1.2587228109353903,0.3561666568718871
-3,80,-1.75122379690588,-0.3008387767317566,-0.5208614133489836
-3,81,1.383460264628571,0.9379248586904108,0.1299212462358277
-3,82,-0.0608910740225044,1.3323944583229077,-0.3047310126274782
-3,83,1.6230751657815008,-0.0630906778233005,1.9115397604052724
-3,84,2.2408288480418097,0.8103089375173486,-0.8567878621140438
-3,85,-0.9727773430855462,0.363533033619507,0.9366327938393536
-3,86,0.7902367502999565,-0.3627126190829625,0.1183670068806765
-4,87,-2.0447369464898366,0.3023432233196432,-0.1473154301210917
-4,88,1.869964721158462,-0.864875745417614,-1.3056540391037692
-4,89,-0.3804968586445018,-1.2991572101520728,3.0274943343874083
-4,90,-0.6849867971231273,-0.6971021337130133,0.3617294328556023
-4,91,-0.1893749607437728,0.1572174669954087,-0.4453650036042871
-4,92,-0.0473841748369682,-0.2885714864228671,0.6282853705023487
-4,93,1.355145254330449,0.8030287320554321,0.9642562105511686
-4,94,-0.812919327386319,-0.278953756610509,-0.8330901074711051
-4,95,1.761852889925791,1.0654125004131518,0.7981435937792382
-4,96,1.5513843109017444,-0.2196020449778792,0.7995168560801881
-4,97,-0.0351860754981054,-0.9365730657119276,-0.4778765668524142
-4,98,0.6231377615829627,-0.5729404959546416,1.7731643504472827
-4,99,-0.3997316463290774,-0.8806733580293892,1.2401001274351815
-4,100,0.2372394789402002,-0.1054132415697752,-0.6172799146920306
-4,101,-0.1212227670247407,0.2660023798346163,0.7526864964032826
-4,102,2.0205193300657247,-2.364509701006651,-0.3922040087053289
-5,103,-1.38640492450024,0.1640158418645363,1.3164444471305916
-5,104,-0.0921642607233633,0.9692400537679028,0.420020710446368
-5,105,-0.3149420755060182,1.5473868473347665,0.8921979069031415
-5,106,-1.2635106392550968,0.7619417833495526,0.4279062963690861
-5,107,-0.1778576632950099,-0.3282050844487043,-0.1809155329521617
-5,108,-0.4341462186773626,-0.478029836620016,0.5335886571652085
-5,109,1.4092433547182477,-0.6852162193360752,-0.7825217022412374
-5,110,0.3859950895821014,2.0669881917476105,-1.3297057074440568
-5,111,-0.7182644036676693,-0.146096739029817,-0.9899126599978328
-5,112,-1.273385819360683,-0.9772625270992248,-1.2241592816111664
-5,113,-0.1359745388983841,0.9827239663467208,-1.6081151211440765
-5,114,0.2113980657466534,-0.1136468939335117,0.6893365374951431
-5,115,0.7724693369876277,0.8486283795297606,1.3184511797851892
-5,116,-1.189419775035365,2.175160590481025,0.2530927744153326
-5,117,-0.2944814094803828,0.3844036744981713,-0.7218134212123581
-5,118,-1.1942449437258529,1.489763278407007,-0.0069204705176007
-6,119,0.0493501681912529,-2.510681111908385,-0.4138548174567836
-6,120,0.2096106989069184,-0.2183747775499781,-0.9217336181544192
-6,121,-1.5429639704955027,-1.098282706675557,-1.086483742151834
-6,122,-0.6450828357377884,0.2354221990787906,1.474288547233567
-6,123,-0.446605922679624,1.2013091752960456,1.5875581988032164
-6,124,0.4686929978268352,-0.0961431446440496,-1.8988657509221156
-6,125,-0.1580299316500071,-0.1578710620547832,-0.4050814416560676
-6,126,-1.759532925155835,-0.0693937481215322,0.8726496980525889
-6,127,-0.3015247329700108,-1.3923361315518956,1.8224737121284704
-6,128,-0.4450013498626253,-0.4113434043871031,0.3293755829411718
-6,129,0.7903245001231152,-0.4011588652459638,1.1353736264714218
-6,130,-0.6295261828313713,0.292502475077484,-2.088924140703356
-6,131,0.8531138131638296,1.6468836783647591,1.1776401371612333
-6,132,0.5187188498926204,-0.4647811355098913,0.3101310578176122
-6,133,-1.3489685142027716,-0.3756787727697905,0.0489690373962899
-6,134,-0.0946907184934595,0.3892264192035026,-1.122565585229101
-6,135,-0.7474712355198262,1.2901653302909308,-0.1379191118897307
-6,136,-0.4068537594387174,-0.6838974781340557,0.477337569358077
-6,137,0.0266763706342167,-1.007283482879746,-0.0207534690671556
-6,138,-0.0032743175101055,1.2232128286131143,0.9348832049035216
-6,139,-1.4245234141757734,-1.5302479419210997,-0.804227609128954
-7,140,1.3385463652897756,0.0713020705966725,0.1431939714725609
-7,141,0.8725755275379272,1.678570135587483,-0.4605569829841085
-7,142,0.7739773642814373,2.152287076973953,-1.0553504705140826
-7,143,0.4365841243143215,1.1854010966339783,0.7206026881536415
-7,144,1.7451551239261762,0.8043762766414709,-1.718237086897971
-7,145,-1.6815565231304883,0.7973424761299034,0.527556676509205
-7,146,-1.6595644566768657,-0.2695223816290026,1.4729892021615898
-7,147,-0.1021694125843338,-1.383641623013081,-1.0059562338044414
-7,148,0.9372663261906188,-0.4184403128098172,1.6305569491478604
-7,149,1.073148960167395,-2.438952753391153,2.547370551387279
-7,150,-0.0717409847677429,1.323885747839544,2.033324551178638
-7,151,-0.5685414310705261,-0.6367891774048193,0.4043282940651028
-7,152,-0.1316454577882007,0.1369545874797834,0.4191280421631976
-7,153,1.2953199335014074,0.6667546621883552,-1.5587069063508987
-7,154,0.9833972715686036,-0.5470404748851541,-0.7830858337970409
-7,155,1.1579688470041969,-0.9611805800659046,0.1371199788701356
-7,156,0.1827239769253915,-1.5230847499914129,1.5499797081722422
-8,157,-1.0562245233729604,-0.2219192589824737,1.818575785843776
-8,158,-0.9383243525460702,0.9546211427025948,0.1183152450709592
-8,159,0.0513695722870604,-0.6758643545674766,0.7652662329923993
-8,160,0.3874642120923474,0.5043116930108147,1.5686505707558396
-8,161,-0.243642091045886,-0.2025208427125454,0.1255815698914305
-8,162,0.7717151527853235,-1.506391286970976,0.4338170324237214
-8,163,1.387099745160974,0.6487698593069315,1.449813536152996
-8,164,-1.602068215416117,-0.4577977113344454,0.8621272110483021
-8,165,1.236572267392626,0.044167628496001,0.3234054613524826
-8,166,1.5622214470091351,1.1754844396047714,0.5064088364815167
-8,167,-0.0485551549391928,-0.0990325767481106,1.4913472073701488
-8,168,0.4285506018746526,-0.0475450617002715,1.023767695317657
-8,169,0.497472174299677,0.5034484743308774,-1.4437814578576325
-8,170,1.0604287367676706,0.6921642133956442,-1.2314791793635456
-8,171,1.412582247970505,0.5559150837807887,0.7673842918563573
-8,172,-0.535552885754127,0.5499993252896678,1.6847926118412546
-8,173,0.2408773901470533,-0.4530529438523386,0.4514898135283557
-8,174,0.471447663060989,2.0916787764915914,1.5702390064548917
-8,175,0.3794929622741819,-1.0422876011718514,-1.4233773553881364
-8,176,1.740091064639636,-1.0655283950718082,-0.8348156343282267
-8,177,0.9174869800162324,0.4217188487657152,1.64301604024024
-8,178,-0.795087240847617,0.8445655979794515,0.5931810417739123
-8,179,-0.0110151195914141,0.9582441198728032,0.2933912934374598
-8,180,-0.1134702145020138,0.5930174775979217,1.0447321534051053
-9,181,0.5320720682670665,1.0962183807331836,0.6181051753418312
-9,182,-0.0834163446834311,0.6793317541568811,-0.4304087997159057
-9,183,-0.457604347411888,0.05847542907819,0.4244180231779655
-9,184,0.6138956500945167,0.3941151592170865,-0.0579477086770905
-9,185,-1.282571875410227,0.5588548106709973,0.3949805394801389
-9,186,1.125490928479345,-0.6045051491381532,-1.6105230975007343
-9,187,0.3386017615816551,-2.576017924887878,-2.243399918867632
-9,188,0.5914720853242976,0.3974530789128242,0.9299617676237236
-9,189,1.738580470075832,-0.8958775205971541,-1.4146582334712663
-9,190,-0.0415027573287599,-0.1068012186985196,-0.7481546299358196
-9,191,-0.7906458359030527,0.3913039526310852,0.9978354826902832
-9,192,0.7004059336701238,0.7501424988056595,-1.3269464311869232
-9,193,0.6334042909883103,0.2327825827474678,-1.5774782605931386
-9,194,1.3817463103025862,-0.5904344844700138,-0.0946765671355123
-9,195,-1.5609804115046182,-1.879944960324204,-1.0427121479775012
-9,196,-1.2189431636039487,1.8513283087111012,-0.8627459284058191
-9,197,1.6991040334285772,0.9436017873408368,0.0995309724676549
-9,198,0.9410236320395357,1.5543229265388008,-0.571201566340334
-9,199,2.083194838563287,0.7115442580156478,-1.0436603387986798
diff --git a/tests/resources/source_data/test-data-real-5.csv b/tests/resources/source_data/test-data-real-5.csv
deleted file mode 100644
index 78e3348b..00000000
--- a/tests/resources/source_data/test-data-real-5.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemValue,supReal1,supReal2,supReal3,supReal4
-0,0,1.2157364754725126,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886
-0,1,-0.4357209731834651,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225
-0,2,-0.6339475697946867,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471
-0,3,0.6368626136290272,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664
-0,4,0.6110081460367197,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679
-0,5,0.9905452893218942,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873
-0,6,-0.3345643150202115,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203
-0,7,0.1256628904239142,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406
-0,8,0.2480230016822292,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459
-0,9,-0.1922427914035724,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191
-0,10,0.5924561539591094,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862
-0,11,0.8745787252568672,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572
-0,12,0.5909207435392255,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035
-0,13,-1.238310596736537,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925
-0,14,-0.0313947023454899,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497
-0,15,2.2431544596409108,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702
-0,16,0.008841402510734,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912
-0,17,1.1787037174986643,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137
-0,18,-1.2232374764781524,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254
-0,19,-1.1887591553907793,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754
-0,20,1.2833234626994587,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509
-0,21,-0.832095086163681,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411
-0,22,0.6549478021989834,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851
-0,23,1.4144054511713913,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458
-0,24,1.434700880302887,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296
-1,25,-0.1484614438126814,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405
-1,26,-0.907149052022494,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361
-1,27,-1.315525084311021,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477
-1,28,0.321371903856307,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585
-1,29,0.0588998328655788,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805
-1,30,-1.6705347852472077,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652
-1,31,1.6517289984560022,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368
-1,32,-0.4256480977552965,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937
-1,33,0.514013235792249,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382
-1,34,0.0557968554012388,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087
-1,35,-1.1716537478584226,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054
-1,36,0.448353692105597,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053
-1,37,0.7305124247576416,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647
-1,38,-0.6380221609658,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902
-1,39,1.4755589311816426,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116
-1,40,-0.260160568050281,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335
-1,41,-0.3439026752059935,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823
-1,42,-0.5938983834903411,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052
-1,43,0.7148213037098026,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834
-1,44,-0.9990236390866448,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783
-1,45,-2.035389167985359,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322
-1,46,-0.1241090076273592,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941
-2,47,0.6859234003054556,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419
-2,48,-0.8499007946906604,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136
-2,49,0.9408956310929066,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277
-2,50,1.6646131411816578,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542
-2,51,-1.4938369847202195,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181
-2,52,0.8496529584712066,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197
-2,53,-0.0459189534252687,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343
-2,54,1.5365245662265232,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612
-2,55,0.6256680895244052,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304
-2,56,-2.260359396826676,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609
-2,57,0.1914661609037843,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176
-2,58,0.5158076460748913,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599
-2,59,-1.0947810018348507,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016
-2,60,-0.9865581243255936,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634
-2,61,-1.4338664192702113,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373
-2,62,-0.2172906414350655,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621
-2,63,0.0199528357410374,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672
-2,64,0.0168235208889687,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168
-2,65,2.1259103285431804,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221
-2,66,0.0094460765129033,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992
-3,67,-1.473158658721011,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244
-3,68,0.8552751411665621,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708
-3,69,-0.6933356133030083,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643
-3,70,1.4678284201004026,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273
-3,71,0.24217591143051,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959
-3,72,-2.10339281517975,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244
-3,73,1.0711650587517474,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615
-3,74,1.7463744130531542,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745
-3,75,-0.3363388568076421,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721
-3,76,-0.3242937117544179,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233
-3,77,-0.9411647896878936,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143
-3,78,1.069879949761595,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957
-3,79,-0.8667154099610572,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826
-3,80,0.1198519014290499,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804
-3,81,1.044964805654135,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499
-3,82,-1.8702057892269273,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173
-3,83,0.886557763359472,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713
-3,84,-0.4732550989352198,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431
-3,85,0.0634333684969395,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982
-3,86,0.2665413517198428,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551
-4,87,-0.9188292311366768,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927
-4,88,-0.2034731183127446,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394
-4,89,-0.215148925629062,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372
-4,90,0.6064317194064263,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425
-4,91,0.5244272005602573,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239
-4,92,-0.4160522494759603,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421
-4,93,-0.2625459626333661,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964
-4,94,-1.0851148931024432,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218
-4,95,-0.2912304897933477,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471
-4,96,-0.8187074748916447,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201
-4,97,0.0492718080501761,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814
-4,98,-0.8062622557548957,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774
-4,99,-0.4153917948141301,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907
-4,100,-0.2224607283709459,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114
-4,101,-1.5140509754678362,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313
-4,102,1.0334281623735169,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983
-5,103,-0.6864230377948968,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032
-5,104,0.4121725852133129,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658
-5,105,-0.910130468154296,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267
-5,106,-1.70552576193866,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745
-5,107,-0.5989313128779504,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032
-5,108,-1.768663887404953,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915
-5,109,0.0821914280814239,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433
-5,110,1.390366451549932,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324
-5,111,0.3326169890263175,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795
-5,112,0.6734144260342821,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436
-5,113,0.0305206223002543,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962
-5,114,-0.0784091712011275,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194
-5,115,2.064172173892823,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067
-5,116,-1.9779839409772504,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663
-5,117,0.2051671337582474,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233
-5,118,1.6220394426605342,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675
-6,119,0.0473166159950642,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145
-6,120,-0.4115115284299588,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176
-6,121,-1.402681056942516,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413
-6,122,-1.3275034743892011,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453
-6,123,0.7014665551150057,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988
-6,124,-0.2035315039990889,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934
-6,125,0.0367104465103882,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717
-6,126,-0.5108362781439201,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115
-6,127,0.3179021236908106,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816
-6,128,0.6725802644112402,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604
-6,129,0.8596257938486648,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782
-6,130,0.4715510300846484,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951
-6,131,0.5315089928295402,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643
-6,132,0.1817244592342696,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258
-6,133,-0.2256707905927475,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091
-6,134,-2.16768470520022,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833
-6,135,0.3087019855449177,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803
-6,136,-1.2897394787144127,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744
-6,137,0.4621743751557871,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001
-6,138,-0.3134765343841647,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725
-6,139,1.0522402191888502,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555
-7,140,0.2927037400921888,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131
-7,141,-0.8730668573777406,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113
-7,142,-0.2922851815210742,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609
-7,143,-0.0649021232364764,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969
-7,144,1.2214127004424862,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388
-7,145,-0.5043967829478707,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822
-7,146,-0.0708070086155714,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244
-7,147,-0.0692762501832949,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663
-7,148,1.3045814471832968,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197
-7,149,-0.4043913713107616,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621
-7,150,-0.0446396090232183,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756
-7,151,0.512499015351023,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713
-7,152,0.5357132143740251,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026
-7,153,-2.197173079151018,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117
-7,154,-0.0886891865646971,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711
-7,155,-0.2605697307740011,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251
-7,156,0.6174561809867044,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264
-8,157,-1.0038074507140595,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726
-8,158,-1.059730211003972,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695
-8,159,-0.3692993280894203,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448
-8,160,-0.4171826559718362,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761
-8,161,0.2393065429868705,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186
-8,162,-0.676895959096811,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605
-8,163,1.1797212483448458,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671
-8,164,0.3968374847149752,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678
-8,165,0.2299432745676143,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952
-8,166,-1.0984706199955088,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802
-8,167,1.5304566581109758,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236
-8,168,0.7606981599458013,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769
-8,169,-0.1082555524626026,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754
-8,170,-1.7075689205534992,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697
-8,171,1.5659864989309202,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477
-8,172,-0.8565031425715999,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823
-8,173,-0.7512773910842393,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512
-8,174,-1.3291106873309495,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194
-8,175,0.0402955848732538,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751
-8,176,-0.2798439837471386,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605
-8,177,-0.6708471763560321,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511
-8,178,1.3065606154170957,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498
-8,179,-0.4208095265891202,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123
-8,180,-1.290508372602977,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558
-9,181,-0.0236678016825857,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157
-9,182,-1.3482004671679793,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377
-9,183,-1.6032504940224688,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836
-9,184,0.8862769243448122,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203
-9,185,-0.6916024517502692,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431
-9,186,0.4105014455674474,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504
-9,187,-0.479142102490158,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529
-9,188,0.4483153370081094,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308
-9,189,0.0831798773951058,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378
-9,190,0.8891055731718962,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787
-9,191,0.1744572947226649,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506
-9,192,-1.04626647271007,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671
-9,193,0.9239324161922192,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253
-9,194,-0.2719797371819717,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154
-9,195,-0.5394421420914198,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914
-9,196,0.8822731035358525,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066
-9,197,-0.2855852303824563,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809
-9,198,0.1982091219091618,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367
-9,199,0.5878676284104647,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885
diff --git a/tests/resources/source_data/test-data-real-50.csv b/tests/resources/source_data/test-data-real-50.csv
deleted file mode 100644
index 931141e3..00000000
--- a/tests/resources/source_data/test-data-real-50.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemValue,supReal1,supReal2,supReal3,supReal4,supReal5,supReal6,supReal7,supReal8,supReal9,supReal10,supReal11,supReal12,supReal13,supReal14,supReal15,supReal16,supReal17,supReal18,supReal19,supReal20,supReal21,supReal22,supReal23,supReal24,supReal25,supReal26,supReal27,supReal28,supReal29,supReal30,supReal31,supReal32,supReal33,supReal34,supReal35,supReal36,supReal37,supReal38,supReal39,supReal40,supReal41,supReal42,supReal43,supReal44,supReal45,supReal46,supReal47,supReal48,supReal49
-0,0,1.2157364754725126,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436,-0.5820086906467112,-0.0769931567360886,-1.1517403862404294,0.47512151913436
-0,1,-0.4357209731834651,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212,-0.9089247216662516,0.3791596715157225,-0.0206890519736249,-1.1261734468621212
-0,2,-0.6339475697946867,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153,-0.5154371712252781,-0.6449188963332471,0.5834646194670815,-1.2162003117631153
-0,3,0.6368626136290272,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597,-0.8796536665660435,-1.8579306192075664,0.1427987669192363,-1.4933477031742597
-0,4,0.6110081460367197,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634,-0.0946835025191322,1.576673513358679,-0.3530020167338319,-0.9670816286563634
-0,5,0.9905452893218942,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835,0.573453806301526,-0.1647553645757873,1.4233611396452903,0.2951585961176835
-0,6,-0.3345643150202115,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176,-0.6726355586632142,-1.461972740712203,-0.9799644518478924,-0.8256811879776176
-0,7,0.1256628904239142,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363,0.7338377851923579,2.748534947455406,2.229642972063479,-0.2626654773388363
-0,8,0.2480230016822292,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525,0.4123843865529258,-0.4185007661778459,-0.5732593838301187,1.3128198055960525
-0,9,-0.1922427914035724,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517,1.0002758074691138,1.5772388807424191,-0.5873894980380264,-0.1724514169048517
-0,10,0.5924561539591094,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448,-1.6793033775681865,-0.2018030185838862,-0.2477508115702648,0.9984558771504448
-0,11,0.8745787252568672,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718,0.5192150697838572,0.5151521190062572,0.375295245422501,0.8241769523505718
-0,12,0.5909207435392255,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743,-0.0684948380162701,1.2255888487339035,-0.028236201979088,1.1524820376873743
-0,13,-1.238310596736537,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309,-1.012670179240466,1.054142287849925,-0.7885639649483744,-0.4532287722802309
-0,14,-0.0313947023454899,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633,0.4598005719286187,0.1955793181435497,-0.9023833307495563,0.7406957728589633
-0,15,2.2431544596409108,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786,2.228549215957172,0.4303494463769702,0.044437463783281,-0.3046679002763786
-0,16,0.008841402510734,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338,1.580761752579522,-0.7216115292517912,-0.1687177623299483,-0.0981028713050338
-0,17,1.1787037174986643,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625,1.0274608136462915,0.5810592781295137,0.9883773158686636,-0.338687599861625
-0,18,-1.2232374764781524,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396,0.7105776900317172,1.4366479972728254,-0.7690976383064227,-0.5926329276128396
-0,19,-1.1887591553907793,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571,0.7000235215395574,-0.335667291930754,-0.6433843829871714,-0.8683614880285571
-0,20,1.2833234626994587,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371,-0.1421408408449271,1.4195302188448509,-0.0519121264937029,-0.2223944052810371
-0,21,-0.832095086163681,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416,0.7671249818780963,0.5754851039040411,-0.4591761627606048,0.2238516678750416
-0,22,0.6549478021989834,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094,0.284269664190007,-0.3111877446448851,-0.3898668641222089,-0.4794418411187094
-0,23,1.4144054511713913,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007,0.5367982644094035,-1.0458755769373458,-1.2837459164682858,-0.208702368387007
-0,24,1.434700880302887,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726,0.9143038302645866,-0.0429160593265296,0.1902826239053188,0.1273379928547726
-1,25,-0.1484614438126814,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851,0.4806224256538509,-1.664559241162405,-1.4806482485447472,-0.2656666807034851
-1,26,-0.907149052022494,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585,0.2940960032574599,0.1593963161317361,0.7237431360170201,-0.9754261015074585
-1,27,-1.315525084311021,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498,1.2133396475903873,-0.8143597963095477,0.1929718727825684,-2.0639707193059498
-1,28,0.321371903856307,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831,-0.4229377963857184,-0.0408137348154585,1.7796982141489022,0.7319848912899831
-1,29,0.0588998328655788,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465,0.5666495515437896,2.5570118821919805,-0.8850362307026824,-0.0908278640955465
-1,30,-1.6705347852472077,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527,0.6009858839082124,0.0611023753041652,1.2978731439161524,0.2854174163780527
-1,31,1.6517289984560022,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396,1.7092664244525773,-0.5819529585304368,-0.2591452925227064,0.2014682827257396
-1,32,-0.4256480977552965,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572,0.3399258904044571,-0.6904391262963937,1.4982259796054014,0.1453056346528572
-1,33,0.514013235792249,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217,-0.1079289200633795,0.591072574117382,-0.0768490761377993,0.8095167753996217
-1,34,0.0557968554012388,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012,0.6294206328381015,0.1651597212565087,-1.9367263299369712,1.5900425772732012
-1,35,-1.1716537478584226,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837,1.4633194614817098,-0.638500712064054,0.0506897125076762,-0.8178820250244837
-1,36,0.448353692105597,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683,0.5503790726147177,-0.4576388791956053,-0.2116134506350607,-1.1811248427188683
-1,37,0.7305124247576416,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613,0.2288111251968629,-1.3597763965249647,0.6897281340221474,-0.5521109827498613
-1,38,-0.6380221609658,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907,1.5447878604004366,0.2631408387300902,0.4499808493098527,0.3949748622436907
-1,39,1.4755589311816426,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792,0.1630326559403528,1.352357018704116,0.8871824140419462,-0.6463976133595792
-1,40,-0.260160568050281,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582,-1.1826482612552474,-0.2882233178544335,0.419432629431521,-1.2805957817399582
-1,41,-0.3439026752059935,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504,-0.1671730495935667,1.9502633738336823,-1.119232635198605,0.5508900386572504
-1,42,-0.5938983834903411,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529,-0.0132633758837144,0.2930435082013052,2.2684789495928994,0.0853086889333529
-1,43,0.7148213037098026,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981,-0.7019854368140658,1.408893632658834,0.5078629724212651,-0.218769711385981
-1,44,-0.9990236390866448,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636,2.138611526896666,1.6224252204407783,1.3656424777113103,-0.6197723706627636
-1,45,-2.035389167985359,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312,-0.4641464411878229,-1.3845837181487322,-1.3562245646187732,-1.1247441521636312
-1,46,-0.1241090076273592,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791,-0.3929353813928705,-0.1806118894466941,-1.2589744121175277,-0.072555647708791
-2,47,0.6859234003054556,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203,1.878090363680033,-0.451915812567419,-1.551896108778869,1.3399026572550203
-2,48,-0.8499007946906604,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387,0.5316335161706456,-2.076928596557136,0.5089022493346776,-0.024381870075387
-2,49,0.9408956310929066,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262,-1.383512008752633,-0.7200539514595277,-0.4616503845204617,-0.9970857305430262
-2,50,1.6646131411816578,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487,-0.9116180632797354,0.676131854304542,0.4064830425417453,-0.8334960910074487
-2,51,-1.4938369847202195,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262,0.1807163239900383,-0.5520111418892181,0.9541145888616588,0.3845601295654262
-2,52,0.8496529584712066,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844,0.3477709783565773,-0.0700644269350197,-0.1491201158577285,-0.2294227620946844
-2,53,-0.0459189534252687,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545,-0.5910886720642256,-0.1758591223402343,0.4926583207467876,1.0169773111385545
-2,54,1.5365245662265232,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886,-1.0001172593271828,-1.3009627710612612,-1.9417151819704204,0.3363413463024886
-2,55,0.6256680895244052,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301,-0.6059838835052326,-1.9462420775220304,-0.4657423912926528,0.0097818169970301
-2,56,-2.260359396826676,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234,0.3696348407961316,0.8349123959613609,-0.9894874210300914,0.5901905216267234
-2,57,0.1914661609037843,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184,-2.159229683786817,1.8919225969657176,0.7699124251573882,-0.3329927633569184
-2,58,0.5158076460748913,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301,0.1158729118533183,0.0480150229911599,0.1342074200813592,0.5469937351204301
-2,59,-1.0947810018348507,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716,0.510184749224577,0.0521734016312016,-0.0620290424804897,1.4387947715516716
-2,60,-0.9865581243255936,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986,-1.0181941589104413,-0.122488689422634,-0.0623725199141001,-1.5928420817380986
-2,61,-1.4338664192702113,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152,-2.1012260841117647,-0.5029723690001373,-0.8841317706278249,0.4144898850289152
-2,62,-0.2172906414350655,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164,-0.2641403390810031,0.784556160076621,-1.021775165064795,0.9626552542957164
-2,63,0.0199528357410374,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851,1.126424989282014,0.4504849416729672,-0.6089787081800524,0.5938953228849851
-2,64,0.0168235208889687,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631,-0.2788158895740412,0.9155610696581168,-0.8498994741129242,-0.6852888501385631
-2,65,2.1259103285431804,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908,-0.3884284583050837,-0.2164343072108221,1.0272836436902952,-1.802304268569908
-2,66,0.0094460765129033,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691,-0.9695615693196264,0.9838979133549992,0.0853085749390771,-0.0129419618714691
-3,67,-1.473158658721011,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939,1.57860530312412,1.781026952291244,0.3785703689466918,0.728096157555939
-3,68,0.8552751411665621,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876,-0.4852580169423038,1.2210276536011708,-0.0731007657759648,0.7156279104852876
-3,69,-0.6933356133030083,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728,1.3115417566478216,0.1115062283716643,1.360751332964618,1.3546839988588728
-3,70,1.4678284201004026,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479,-0.9375197390131628,2.2180940643695273,-0.834846466609614,-1.232361381356479
-3,71,0.24217591143051,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013,-1.4933926504379784,2.063977986283959,-1.809855630735837,-0.0435872778479013
-3,72,-2.10339281517975,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734,1.66698134913082,1.358559600078244,-2.1594379807383564,0.5484525334475734
-3,73,1.0711650587517474,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456,0.4215958062784253,-0.7316269607111615,-1.5313215972642025,1.307847634805456
-3,74,1.7463744130531542,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832,1.6374672768528042,-0.8610589522771745,0.687052936489978,-2.618532214839832
-3,75,-0.3363388568076421,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857,0.2645026021626481,-0.3201115163187721,-0.3698804048569127,-0.8496513139782857
-3,76,-0.3242937117544179,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022,0.7526070369449078,0.0701835488158233,1.7261527406887782,-0.7521255704289022
-3,77,-0.9411647896878936,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708,0.2182067272217519,-0.4656191153685143,-0.9060432443210892,-2.126705535677708
-3,78,1.069879949761595,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675,0.2730385089817146,1.010133405147957,0.3555883594964702,0.8752645615044675
-3,79,-0.8667154099610572,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871,-1.1795205381405394,-1.1624184233132826,-0.181205600784726,0.3561666568718871
-3,80,0.1198519014290499,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836,-0.300497076461078,-0.3509048690778804,-0.0490761453581302,-0.5208614133489836
-3,81,1.044964805654135,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277,-0.8405052153381654,0.3234805183583499,-0.507367498913363,0.1299212462358277
-3,82,-1.8702057892269273,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782,0.7356818372550097,1.0955279766605173,-0.7170222640656372,-0.3047310126274782
-3,83,0.886557763359472,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724,-0.2388956656409064,2.141356993565713,-0.4907544149793494,1.9115397604052724
-3,84,-0.4732550989352198,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438,1.643155432693797,-0.5315980822923431,-0.1847655376603063,-0.8567878621140438
-3,85,0.0634333684969395,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536,0.5319856670267262,-1.6832107216853982,1.9391554888589777,0.9366327938393536
-3,86,0.2665413517198428,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765,-2.1326672344812105,1.6293148762619551,-0.1407819188950014,0.1183670068806765
-4,87,-0.9188292311366768,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917,-1.6042550026469249,0.934842571948927,0.5486353617814611,-0.1473154301210917
-4,88,-0.2034731183127446,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692,-0.5997440870462475,-1.6268707626302394,0.827624617684303,-1.3056540391037692
-4,89,-0.215148925629062,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083,-0.0520796489093672,1.0706727669407372,0.2834781571881385,3.0274943343874083
-4,90,0.6064317194064263,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023,-0.0883800576540255,1.8298688366760425,-0.8498887240727157,0.3617294328556023
-4,91,0.5244272005602573,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871,-0.8334261447948773,0.8460104326165239,-0.8306787084232922,-0.4453650036042871
-4,92,-0.4160522494759603,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487,1.350321469218172,0.6572555679296421,-0.9931133230551088,0.6282853705023487
-4,93,-0.2625459626333661,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686,-0.9992263103431246,0.6359063166710964,-0.2163318860335712,0.9642562105511686
-4,94,-1.0851148931024432,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051,-0.7635405661499053,-0.8288880276599218,-0.1516555757419315,-0.8330901074711051
-4,95,-0.2912304897933477,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382,-0.6666971915003546,0.7571819436119471,-0.8026449667542606,0.7981435937792382
-4,96,-0.8187074748916447,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881,-3.730120038128068,0.6010894307864201,1.486736945275556,0.7995168560801881
-4,97,0.0492718080501761,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142,-1.8018134368021572,-0.5882845110059814,0.1996588544375249,-0.4778765668524142
-4,98,-0.8062622557548957,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827,-2.966586951844483,0.3460687892555774,0.7122258787189174,1.7731643504472827
-4,99,-0.4153917948141301,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815,0.9763019660890628,-0.7464387784593907,-0.6432674437141003,1.2401001274351815
-4,100,-0.2224607283709459,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306,-1.9341599869280055,-0.5571429776627114,-1.0509036442167718,-0.6172799146920306
-4,101,-1.5140509754678362,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826,0.0115401685101198,-0.4361955547709313,1.176673643162231,0.7526864964032826
-4,102,1.0334281623735169,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289,-1.7973367751171303,0.6067623680262983,-1.2244068035018794,-0.3922040087053289
-5,103,-0.6864230377948968,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916,-0.5782800824093337,0.0443200602767032,1.3462077769017655,1.3164444471305916
-5,104,0.4121725852133129,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368,-0.1850434000530602,0.2991100780439658,-0.1375589867919389,0.420020710446368
-5,105,-0.910130468154296,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415,-0.2332258635179755,0.3259625438150267,-0.9377838244466902,0.8921979069031415
-5,106,-1.70552576193866,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861,-1.5395002171602334,-2.427626324562745,-1.0107549325024228,0.4279062963690861
-5,107,-0.5989313128779504,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617,-1.106476206464435,-0.0836893357947032,0.2910156134691695,-0.1809155329521617
-5,108,-1.768663887404953,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085,-2.129243520644403,-0.3347671453032915,-0.1324631105096152,0.5335886571652085
-5,109,0.0821914280814239,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374,0.4768143516638856,2.1555402812119433,1.7507054168018197,-0.7825217022412374
-5,110,1.390366451549932,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568,-0.1723342300598751,-1.833112610191324,-0.3134080474285327,-1.3297057074440568
-5,111,0.3326169890263175,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328,-0.6345577264308937,0.7987667198270795,-1.3845614226225136,-0.9899126599978328
-5,112,0.6734144260342821,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664,-1.3588134360661253,0.590422932866436,1.1726184657626748,-1.2241592816111664
-5,113,0.0305206223002543,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765,-1.115309203106232,0.5874688773032962,-0.2400575224647247,-1.6081151211440765
-5,114,-0.0784091712011275,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431,1.660083512711184,1.484213803841194,-0.0316580690339819,0.6893365374951431
-5,115,2.064172173892823,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892,0.5981510743069248,0.3784041225930067,0.205692450226748,1.3184511797851892
-5,116,-1.9779839409772504,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326,0.1762152669186038,-1.1598019361532663,1.232834205749278,0.2530927744153326
-5,117,0.2051671337582474,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581,0.0995596156598381,1.1771905557314233,-1.3464989957288118,-0.7218134212123581
-5,118,1.6220394426605342,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007,-0.134973127015259,-0.0541151698997675,1.253570686211627,-0.0069204705176007
-6,119,0.0473166159950642,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836,-0.7491928547518463,-1.417592127415145,0.0364353437915598,-0.4138548174567836
-6,120,-0.4115115284299588,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192,2.070533424691801,0.6783386557605176,-1.2066020838043194,-0.9217336181544192
-6,121,-1.402681056942516,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834,-0.714833705558571,2.1854436787350413,-0.6967053433394235,-1.086483742151834
-6,122,-1.3275034743892011,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567,0.1631413469278299,-2.3538868063575453,-0.7112384571825194,1.474288547233567
-6,123,0.7014665551150057,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164,1.027268717867479,0.9478340426501988,0.7918858755313075,1.5875581988032164
-6,124,-0.2035315039990889,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116,1.8808868244932129,-1.028398215435934,-0.3360562566812991,-1.898865750922116
-6,125,0.0367104465103882,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676,0.4437277986745326,0.4009052069266717,-1.0404799003110452,-0.4050814416560676
-6,126,-0.5108362781439201,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889,0.1240530371932496,-0.8704155160564115,-1.771859996402141,0.8726496980525889
-6,127,0.3179021236908106,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704,0.4209771365993368,-0.3367093052485816,-0.4402677398140227,1.8224737121284704
-6,128,0.6725802644112402,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718,0.6439655653176443,1.2432492588266604,-0.4657105598503153,0.3293755829411718
-6,129,0.8596257938486648,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218,-0.0632902816234641,0.4016836942875782,0.401291498036223,1.1353736264714218
-6,130,0.4715510300846484,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356,0.7873992206662046,-0.1137723552581951,-0.0933585093103445,-2.088924140703356
-6,131,0.5315089928295402,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333,0.1655764555283259,0.1093916180874643,-0.7592687790730879,1.1776401371612333
-6,132,0.1817244592342696,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122,-0.543886007962999,0.9732071392260258,-0.7251882238055091,0.3101310578176122
-6,133,-0.2256707905927475,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899,-1.3584102327948733,-2.439816596914091,0.9479551036955036,0.0489690373962899
-6,134,-2.16768470520022,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101,-0.1804465985609838,-0.3162304747063833,0.1754295291808443,-1.122565585229101
-6,135,0.3087019855449177,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307,1.0044578374950142,-0.4863328940227803,-0.2306481659713534,-0.1379191118897307
-6,136,-1.2897394787144127,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077,0.7537311601289707,0.4938374222149744,1.4379717883808354,0.477337569358077
-6,137,0.4621743751557871,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556,0.5841218654437593,0.4106303082042001,1.61525984431583,-0.0207534690671556
-6,138,-0.3134765343841647,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216,-0.7226482893432934,-1.759339909789725,0.0102548402674546,0.9348832049035216
-6,139,1.0522402191888502,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954,-0.5571601176468748,-0.4346723973463555,1.361121866942182,-0.804227609128954
-7,140,0.2927037400921888,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609,0.0422524056965653,-0.0240101159271131,1.4489292794691402,0.1431939714725609
-7,141,-0.8730668573777406,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085,-0.4325683214460127,1.706962509491113,0.2734455827127003,-0.4605569829841085
-7,142,-0.2922851815210742,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826,0.8465631724299651,-1.1265125524432609,-0.444622959301189,-1.0553504705140826
-7,143,-0.0649021232364764,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415,0.2871208611901518,-1.09204951025969,0.3400713838645086,0.7206026881536415
-7,144,1.2214127004424862,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971,-0.7868091893675095,0.959537720230388,0.2976658248329601,-1.718237086897971
-7,145,-0.5043967829478707,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205,-1.931844802312892,-0.228977733264822,-0.8053415549445475,0.527556676509205
-7,146,-0.0708070086155714,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898,0.8391818770701904,-0.1782134719224244,0.6898227086652969,1.4729892021615898
-7,147,-0.0692762501832949,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414,0.0019652114728092,-0.3639456962549663,-1.8001008946429178,-1.0059562338044414
-7,148,1.3045814471832968,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604,1.0951871930560133,1.6354854564047197,0.9341867423433384,1.6305569491478604
-7,149,-0.4043913713107616,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279,-1.0208678213187818,0.483662199983621,-0.8412597918918427,2.547370551387279
-7,150,-0.0446396090232183,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638,-0.1154719415024776,1.398128876416756,1.7083320785047869,2.033324551178638
-7,151,0.512499015351023,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028,-0.8707042594597949,-1.109461649094713,0.7551263629789535,0.4043282940651028
-7,152,0.5357132143740251,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976,0.0957855998835066,-0.9450227018489026,1.5294480503629078,0.4191280421631976
-7,153,-2.197173079151018,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987,0.1791648322349623,0.8008029898905117,0.4742598910276621,-1.5587069063508987
-7,154,-0.0886891865646971,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409,0.0648510199038132,1.734280560242711,-0.2552280989334006,-0.7830858337970409
-7,155,-0.2605697307740011,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356,0.2161753878561426,-1.933525731251251,0.9073813550655982,0.1371199788701356
-7,156,0.6174561809867044,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422,1.0344576013849434,0.6160296162846264,-0.5684186138849224,1.5499797081722422
-8,157,-1.0038074507140595,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776,-1.994993947549008,1.145952516208726,1.5216917853720442,1.818575785843776
-8,158,-1.059730211003972,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592,-1.0344467157761543,-0.2535340665870695,-1.4120672507051637,0.1183152450709592
-8,159,-0.3692993280894203,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993,-1.0019028093921996,-1.1000707452790448,-0.0416008039388945,0.7652662329923993
-8,160,-0.4171826559718362,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396,-0.5766936717058257,-1.096630862526761,0.4920064434795458,1.5686505707558396
-8,161,0.2393065429868705,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305,-0.9386728917772926,-0.2402721355505186,0.5365509439884095,0.1255815698914305
-8,162,-0.676895959096811,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214,-1.0166286003580864,0.467624899919605,0.3571814292703247,0.4338170324237214
-8,163,1.1797212483448458,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996,1.2624300972050062,-1.103864704777671,-0.8680772050526456,1.449813536152996
-8,164,0.3968374847149752,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021,-0.5284653415605093,-1.46937511468678,-0.2382931058528679,0.8621272110483021
-8,165,0.2299432745676143,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826,1.1616916884353274,-0.9305269292427952,0.2415737241591204,0.3234054613524826
-8,166,-1.0984706199955088,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167,-0.4651367065068575,-1.1849221842266802,-1.697582153440078,0.5064088364815167
-8,167,1.5304566581109758,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488,0.2991262309237505,-0.6698539026023236,-0.5550596231066798,1.4913472073701488
-8,168,0.7606981599458013,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657,-0.055448324522217,-0.7030253871947769,1.6289405324846409,1.023767695317657
-8,169,-0.1082555524626026,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325,0.599379484168683,0.1495451075683754,-0.35701450358857,-1.4437814578576325
-8,170,-1.7075689205534992,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456,0.3901346999802171,-2.0219509709027697,-0.3146916574309019,-1.2314791793635456
-8,171,1.5659864989309202,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573,0.7395058195373676,2.3514902202349477,-1.177203933584577,0.7673842918563573
-8,172,-0.8565031425715999,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546,-0.2618624769340897,-0.7346368194505823,-0.7864615726217858,1.6847926118412546
-8,173,-0.7512773910842393,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557,-0.4452332175533571,-0.4942753974182512,-0.8744163016610587,0.4514898135283557
-8,174,-1.3291106873309495,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917,0.7613097400139596,-0.1173079903540194,-0.2329089288259748,1.5702390064548917
-8,175,0.0402955848732538,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364,-2.2941794906187303,0.7973437933746751,-1.426007541608154,-1.4233773553881364
-8,176,-0.2798439837471386,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267,0.3184559910398074,0.2107444848003605,0.5596656048644648,-0.8348156343282267
-8,177,-0.6708471763560321,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024,-0.8105101252279568,0.7090347648978511,0.5943452906649828,1.64301604024024
-8,178,1.3065606154170957,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123,-0.4433977354954991,-0.7527960826082498,-0.7841599801769913,0.5931810417739123
-8,179,-0.4208095265891202,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598,-0.2155182381754514,0.8143933428375123,-1.1853946796917243,0.2933912934374598
-8,180,-1.290508372602977,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053,1.6744425413368984,0.996548099996558,-1.0978350788620952,1.0447321534051053
-9,181,-0.0236678016825857,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312,-1.3708388861688323,-0.5493807575389157,1.787101883196328,0.6181051753418312
-9,182,-1.3482004671679793,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057,0.3153721331558907,1.6524931520319377,0.5966684370848218,-0.4304087997159057
-9,183,-1.6032504940224688,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655,-0.9094229514381224,0.1951099471970836,-0.9985225164246512,0.4244180231779655
-9,184,0.8862769243448122,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905,-0.7201079456603995,1.9966008914240203,0.5450880972968815,-0.0579477086770905
-9,185,-0.6916024517502692,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389,-0.3897487397244845,0.0512459191431431,-0.3199173778759451,0.3949805394801389
-9,186,0.4105014455674474,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343,-0.7172381117410999,-1.9727159516935504,-2.048659318810247,-1.6105230975007343
-9,187,-0.479142102490158,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632,0.2448124704232933,-1.162959243638529,0.480689116023008,-2.243399918867632
-9,188,0.4483153370081094,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236,-0.099403340668543,-1.4228761207951308,0.5990631250227616,0.9299617676237236
-9,189,0.0831798773951058,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663,-1.521788254894823,1.1017620346543378,-1.8720492067245145,-1.4146582334712663
-9,190,0.8891055731718962,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196,0.4304399046873204,0.7137451767766787,1.1352709446954192,-0.7481546299358196
-9,191,0.1744572947226649,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832,0.8417705179306674,0.4910955558600506,-1.031973816978013,0.9978354826902832
-9,192,-1.04626647271007,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232,-0.1530678319918289,1.2299202343474671,1.4428401709642211,-1.3269464311869232
-9,193,0.9239324161922192,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386,-0.6719432184447797,-0.6076462824810253,1.3917639604269478,-1.5774782605931386
-9,194,-0.2719797371819717,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123,1.828172446219632,0.4572182235815154,-2.0796199749608832,-0.0946765671355123
-9,195,-0.5394421420914198,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012,-1.2067768937681949,-0.1934015136944914,-0.0429551689588962,-1.0427121479775012
-9,196,0.8822731035358525,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191,-1.1178231135428731,-0.2116301848583066,-0.2733847034116038,-0.8627459284058191
-9,197,-0.2855852303824563,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549,0.6450065530334209,0.2315283210129809,-0.5750141548928486,0.0995309724676549
-9,198,0.1982091219091618,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334,2.8820698132437697,-1.896381798526367,-0.1053054527556563,-0.571201566340334
-9,199,0.5878676284104647,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798,-0.5667820105949781,-2.0175121351827885,0.5531593772082467,-1.0436603387986798
diff --git a/tests/resources/source_scripts/hp_search_eval_script.py b/tests/resources/source_scripts/hp_search_eval_script.py
deleted file mode 100644
index 899783bd..00000000
--- a/tests/resources/source_scripts/hp_search_eval_script.py
+++ /dev/null
@@ -1,38 +0,0 @@
-import json
-import os
-import sys
-
-import polars as pl
-
-
-def main():
-    if len(sys.argv) < 2:
-        print("Error: Missing run_name argument.")
-        sys.exit(1)
-
-    run_name = sys.argv[1]
-
-    preds_path = f"outputs/predictions/sequifier-{run_name}-predictions"
-
-    dfs = []
-    for root, dir, files in os.walk(preds_path):
-        for file in sorted(list(files)):
-            df = pl.read_csv(os.path.join(preds_path, file), infer_schema_length=0)
-
-            df = df.with_columns(pl.all().cast(pl.Int64, strict=False).fill_null(-1))
-            dfs.append(df)
-    df = pl.concat(dfs)
-
-    max_freqs = df["itemId"].value_counts()["count"].max()
-    stdev_freqs = df["itemId"].value_counts()["count"].std()
-
-    eval_dir = "outputs/evaluations"
-    os.makedirs(eval_dir, exist_ok=True)
-    eval_json_path = os.path.join(eval_dir, f"{run_name}.json")
-
-    with open(eval_json_path, "w") as f:
-        f.write(json.dumps({"max": max_freqs, "stdev": stdev_freqs}, indent=2))
-
-
-if __name__ == "__main__":
-    main()
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
deleted file mode 100644
index ef886f9d..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-0-embeddings.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,3,0.12990583,-0.57160014,0.1838459,0.48382023,-0.95800894,-0.06879183,-1.5230852,0.32024524,-0.032266203,-0.23246907,1.2026421,1.264635,-1.6631967,0.8923387,1.008154,2.1465178,1.2605029,-0.024123702,-0.4261095,0.7181141,0.6665921,-0.8360398,0.7274644,0.08159177,-0.46056688,-0.7607759,-0.59745675,0.5428708,0.041339252,0.6946458,0.54585075,-1.3706634,0.99218065,-1.1147714,-1.1846339,-0.65351975,1.6928166,-0.61697114,1.1777916,-0.33404547,-1.2148314,0.16786565,1.4557565,-0.7030481,0.40415096,-2.0602548,-0.44774878,-0.9443907,-0.3784696,-1.4340262,-1.0351793,-0.54817986,-1.341067,1.0574965,0.66676235,1.8221258,-1.4848418,1.5296896,0.25698647,-2.118917,0.2704801,-0.8023632,-0.34508163,-0.7845013,0.29995131,-1.5465924,0.3353845,-0.8597248,0.96339536,0.10576437,0.7733713,1.4520824,-2.3267972,0.81377566,-0.99766785,0.25014007,-0.27935317,0.8346441,-0.33092317,-1.087323,0.4937019,-0.47276843,-0.7608273,-0.80224836,1.3239352,-2.2216246,0.5520812,-1.1424915,1.1881708,-0.45950648,-0.09527827,0.9202227,0.5705651,-2.0961478,1.6434134,-1.0467834,-0.116270155,-0.37290683,-0.6135416,-0.88942945,-1.1429784,-0.78733057,0.57443386,1.4698808,0.43613678,-0.075808465,-0.14003229,-2.95055,-0.016151935,0.6177054,0.6845007,0.56070817,0.87257427,-1.3993288,0.35658634,-0.0063095656,-0.25653526,-0.3163344,-3.5138404,0.887773,1.6519105,0.40932044,-0.027439797,-0.7984332,-1.1049061,0.07724614,0.5687538,-1.0056508,0.48330805,-0.15724464,-1.6178644,0.19340196,-1.9760467,0.058967575,0.5110821,0.08960242,-0.9684786,-1.094302,-0.72294104,1.4727569,0.47599727,-0.628843,-2.055541,-1.0116285,0.992143,0.55828834,0.50531685,-0.09756176,0.7318141,0.67830867,0.34729233,-1.7636684,0.27717403,0.28878212,0.52713984,1.5502089,0.16203246,-0.50224864,1.5970831,-0.6694091,-0.53928715,-1.9245727,-0.013257689,-0.10813395,-2.0357018,-1.6700076,0.21033196,1.2394766,-1.1461272,0.036453612,0.4096094,-0.8327397,-0.008255711,-0.30468205,0.5700773,0.61941594,-1.6036438,0.21153934,0.6522951,-0.39073592,0.21115246,-0.46797752,-0.018942907,0.09517552,-0.61526686,0.6709447,-0.45691368,-0.42524138,-0.8716745,0.07637033,-0.72497517,-1.246546,-0.44291162,1.5433346,-0.82140267,1.8114822,-0.9000441,-0.6661404,0.8775227,0.9152606
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
deleted file mode 100644
index 91bd4f31..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-1-embeddings.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,3,0.6310361,0.33467388,0.16728823,0.36186728,-0.94780934,-0.39266327,-1.3003972,-0.044595696,-0.20526414,-0.026398215,0.1938047,0.81182855,-2.039287,1.2787219,1.2542659,1.4919837,1.0886672,-0.010551577,-0.8448256,0.43219796,0.5695074,-0.80267984,0.7473274,-0.42725766,-1.0895563,-0.21274823,-0.7626765,0.026918093,-0.5983658,0.28117266,0.73443246,-0.9515915,0.8178587,-1.6458836,-0.71132463,0.23738661,1.6034894,0.019240078,1.3241775,-0.20347083,-0.7043624,0.38013235,1.2755847,-1.4004868,0.46275395,-1.9880222,-0.91980124,-1.0473715,-0.65675545,-0.84835905,-1.3224108,-0.70712143,-0.7341194,1.1134468,0.6568125,1.3994766,-0.95416814,1.3808515,0.25189096,-1.9545913,0.23087955,-0.26487717,-0.64339334,-0.2622316,0.6039643,-1.415046,0.6588125,0.08482099,1.1321739,-0.09947382,0.01097981,1.8312315,-1.8722234,0.88451487,-0.7610804,0.08002127,-0.06698804,1.0715898,-0.93518686,-1.0435194,0.026436012,-0.41959038,-0.5251393,-0.54125756,2.0294697,-2.7480886,0.5300638,-1.0863824,0.3388017,-0.14623287,0.6506474,1.0939751,-0.4888243,-1.8103579,1.3499825,-0.8964366,-0.008765843,-0.060274,-0.4804681,-1.2652204,-1.4560695,-0.9467429,-0.2770497,1.9105395,0.20772837,-0.30490705,-0.34610507,-3.309444,-0.46867862,0.6501004,0.43324322,0.559345,0.87965095,-0.87755597,0.5619585,0.29790485,0.5441109,-0.49651536,-2.4436905,1.2963796,1.9646293,0.41595292,0.07784585,-0.8488378,-1.2093579,-0.25940973,0.12893897,-0.51058507,0.64491534,-0.017900245,-2.127124,0.27657425,-1.7817887,-0.66130847,0.2365224,-0.07435237,-1.0712522,-0.8090829,-0.7708737,1.1399105,0.71278065,-0.9480938,-2.0319617,-0.7065952,1.1998773,0.56364936,0.62384117,-0.25050887,-0.123555444,0.011005579,0.41714785,-2.2401168,1.005311,0.10580762,0.57452756,1.6763165,0.39520282,-0.37343678,2.0440457,-0.50678766,-0.59976757,-2.5602221,0.28748658,-0.06885617,-1.7209247,-1.531891,0.72574735,0.8677975,-0.944349,0.32327467,0.09680159,-1.0634657,0.23432283,-0.73151994,0.61498916,0.50479555,-1.6433319,0.36874565,1.5886225,-0.4850403,-0.5036866,-0.24184072,-0.20931797,-0.046914864,-0.6832405,0.32949412,0.5522636,-0.77527416,-1.4231143,0.22487274,-0.16668202,-1.7301934,0.025722574,1.6323113,-1.0889417,1.2082905,-0.36733398,0.19397217,0.32195246,1.1962752
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
deleted file mode 100644
index 6fefa7ce..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-2-embeddings.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,2,0.14896777,-0.7527335,0.09861636,1.5360669,-0.8321218,-0.8661579,0.8799402,0.7707242,0.89909035,0.3597984,1.5049816,0.99816555,-2.204535,0.4618939,1.3522851,0.811366,-0.37875888,-0.4156246,-1.6569219,0.918114,0.88935566,-0.6614972,0.3558165,1.0705256,0.6181391,-1.6933414,-1.6523575,0.47211194,-0.64157414,-0.8052875,-1.1905366,-1.3483986,0.75206506,-1.4446423,0.032763835,-0.77299273,0.8698887,-0.90122116,-0.105313264,-0.032954592,-1.7241205,0.4064759,1.3450825,0.00073461805,0.2343334,-0.9771892,-0.10764814,1.2662542,-0.7311291,-0.6664417,0.0063853106,-0.31623444,0.10896421,0.43619284,0.28484625,0.34632155,0.1304682,1.7500312,-0.3157372,-2.1822789,-0.07376069,-0.63938814,-2.4171436,-1.382503,1.4659556,0.061222862,-2.076287,-1.1751353,-0.08180853,-1.6416469,0.61133516,1.4318693,-1.5806879,1.4651626,-0.42543525,0.27195483,-1.9691069,0.51244324,0.77850366,-1.2194266,-1.2140155,-0.2127739,-0.64433223,-0.3459279,1.4803112,-1.6828287,0.5201632,-0.4554113,-0.17800674,-0.29835144,0.06779454,0.17683099,-0.5690358,-0.47959915,0.24317142,0.70385575,0.47528574,0.38569346,-0.92215097,0.12652385,-0.52221704,0.79115444,0.106221735,1.2142566,0.5760093,0.36402392,1.4705237,-1.8675978,-0.151888,0.84478027,0.55229247,0.25622958,0.5075333,-0.8731629,-1.790598,-0.88261557,-0.80747306,1.2090179,-2.0752997,-0.18883318,0.35716256,0.12954298,-0.58638185,-0.34076798,-0.8663316,0.82491094,1.212744,-1.0448644,-0.53810245,-0.63174623,-0.5035392,0.4781915,-0.9165807,-0.21077545,0.49530777,0.6614037,-0.06105711,-0.011890445,-1.4830858,-0.5170972,1.2860987,-0.48179644,-1.2840594,-0.123924315,-0.561227,0.5283733,1.6894126,-0.015183194,1.619416,0.61651087,0.84680295,-1.2590885,-0.19867837,0.058812138,0.8715905,1.7141719,-2.1852307,-0.17253566,1.748117,0.72871757,-0.93332094,-0.5228718,-1.1484399,-1.0014701,-1.5860217,-0.9670559,0.5662045,-0.63588244,0.62854993,1.4611295,-1.5053114,-0.7354125,-1.1782357,0.4759481,0.56338674,0.29015055,-0.5925469,0.44083118,-0.23582503,-1.1164836,1.0048939,-0.39400432,-0.705857,0.9913283,-1.3044467,2.0163794,-1.8661263,0.013690455,-0.9851868,-0.28402337,-1.4385033,-0.7005213,-1.5944669,1.7985251,0.98409235,1.8579992,-0.7487591,1.5471251,0.7375818,0.14302124
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
deleted file mode 100644
index baf0de12..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-3-embeddings.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,2,0.38537654,-0.7623611,0.5445376,1.1489445,-0.93219095,-1.0841125,-0.27961814,0.19042616,0.06699963,0.6020508,0.7307117,0.975402,-2.3184752,0.97368747,1.6034342,0.94340354,1.2762038,-0.30378082,-0.97008324,0.62717956,1.3900623,-0.9464736,0.48373678,-0.030571124,-0.42904106,-1.1276956,-0.7923078,0.5639612,-0.40978295,-0.23345235,-0.27136517,-0.8576837,0.69629353,-1.600303,0.11132369,-0.5791752,1.9399357,-0.92576593,0.5349059,-0.49527636,-1.3456329,-0.17603783,1.3494829,-1.1282393,0.20037884,-0.955993,-0.8822727,-0.0751942,-0.7717005,-0.93502855,-0.5532879,-0.9401976,-0.10448796,1.1707094,0.6575927,1.3083345,-0.611235,1.6713382,0.16979468,-2.0898654,0.0098398905,-0.44183353,-1.1921822,-1.2657422,1.6432862,-1.1305616,-0.46437377,-0.8430222,0.7453205,-0.6572058,0.3585215,1.6237743,-2.0383017,1.1427608,-0.8917951,0.20761041,-1.3489532,0.62074584,-0.447139,-1.8621218,-0.52442676,-0.45581636,-0.39755735,-0.43458155,1.8005574,-2.0100255,0.35888514,-1.2393867,0.95991373,-0.65878403,0.74196094,0.43640032,-0.37811738,-1.3537853,1.012442,0.12569553,0.45669976,-0.0621398,-0.68999076,-0.7770921,-1.2683296,-0.42139223,-0.41232187,1.7224118,0.3818442,-0.5022627,0.1969878,-2.7483199,-0.18802676,1.315932,0.45021772,-0.2638872,0.5703608,-0.85277075,-0.4612713,0.23827244,-0.42989627,0.41151485,-2.6187692,0.564632,1.1139395,-0.03292864,0.102194116,-0.6601094,-1.3054988,0.09395442,0.43201578,-1.4673973,0.08933548,-0.71541005,-1.6057762,0.40757605,-1.6277497,0.057484966,-0.18920068,0.10566316,-0.89184695,-0.17913231,-1.4436156,0.07747503,0.6109399,-0.83478105,-1.3056535,-0.87130284,0.28275958,0.87232304,0.8705628,-0.1486141,0.91572547,0.6586048,0.6749444,-1.7701942,0.16297537,0.47285134,1.0638492,1.7032028,-0.65427315,-0.017882247,2.1696198,-0.047487043,-0.773447,-2.2596717,-1.181566,-0.58230144,-1.6384531,-1.8893386,0.9176424,0.32815474,0.4039305,0.17879185,-0.32735547,-1.1835169,-1.0626794,-0.124136135,-0.13082623,0.5998316,-1.793989,0.5362147,1.2261003,-1.4384094,0.4796257,-0.45061588,-0.100991115,0.5150622,-1.253895,1.3633755,-0.5889575,-0.21253206,-1.333326,0.5079265,-0.97876984,-1.5163475,-0.6666502,1.8272516,-1.481488,2.3199954,-0.6350159,0.1867249,0.83364856,0.9136824
-4,0,2,1.8327769,0.9652212,-0.25314742,0.2162136,1.9031242,0.70369756,-0.93062437,-0.8883426,0.5527405,0.5570374,1.0132976,-0.9646425,-0.5195063,-0.05579895,-0.37479118,0.44072798,0.16671617,0.5790385,0.41375798,0.80011284,-1.535962,-0.22593139,1.7854862,-0.7154213,-1.0921632,-0.041371867,-0.21795613,-0.33691156,-1.2671199,-0.74733275,0.18677123,-1.8774186,2.0318227,0.27333388,-0.9697611,-0.5860972,-0.36415997,0.22731379,0.47210854,-1.3540338,-0.25599036,0.6586218,1.5556432,-0.8154355,1.1687236,-2.214086,0.42116502,0.47876915,-0.64788955,0.44233435,0.20754997,-0.55014664,-0.5956206,-0.6086165,-1.3033571,-0.9219002,1.9484233,-0.028180521,0.9555862,-0.6107659,1.3185904,0.42626792,-1.1561952,-0.51619065,-0.83520895,0.2209806,-0.3262739,-1.1956955,1.7369201,-1.1766452,0.07778294,-1.4228518,-2.164931,0.61946815,-0.44674355,-0.0108849695,-0.4363428,1.737323,0.6609657,-0.20005423,-0.5650333,0.29500026,0.11585065,-0.054047324,1.7766253,-1.624335,0.48191375,1.0377663,-1.2305183,0.8004891,-1.9219418,1.6361127,0.13642037,-1.2401029,0.56233984,-0.158208,1.5966983,-0.25904053,-0.55149657,0.43704,0.40011016,-0.12492106,1.6866363,2.0976322,-0.49593005,0.85817003,0.21792285,-0.6346643,-1.7050064,-1.6924843,-0.083759315,-0.38175458,-0.32693613,-0.96400416,-0.9026496,-0.594135,0.19496153,-0.8319494,-0.55289817,-0.802287,1.6503073,1.2991755,-1.083528,0.3950478,0.29286572,-0.13108596,-0.38219288,-0.59767187,-1.3788453,-0.26507127,0.3421463,1.0462469,1.5181253,-0.59406966,0.58952194,2.7725646,-1.0286808,1.1897799,0.33696112,-1.0014181,-0.029820807,-0.5898688,-0.99967515,0.010425864,1.4441029,-0.87508553,0.5764213,2.5007524,-1.539079,-1.6458069,-1.7197638,-0.74723464,1.1642295,-0.313455,-0.005579108,1.5905834,-0.9592405,-0.4174034,-0.6339656,0.85305965,1.299676,-0.114394456,0.39143074,0.46125907,-3.0907335,-0.27514967,0.20107278,-1.4204426,-1.7638285,-0.48746756,-1.254607,-0.23415706,-0.101100445,-0.24766745,1.738794,-0.75891614,0.7503048,0.58364636,0.10834089,0.79890466,-0.48462954,-0.62584233,0.50277036,1.1866002,-0.28132573,-0.8015658,0.44897038,1.1347669,0.072793104,-0.6131346,0.88922364,0.71706647,0.6583793,0.3322992,1.1745046,-0.7435528,0.5451967,0.36237976,1.4138212,0.3774721
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
deleted file mode 100644
index a053e062..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-4-embeddings.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,2,0.8678485,-1.1492082,-0.29577842,0.9786945,-1.9542456,-0.0766423,-0.039730456,-0.097185545,-0.46550742,0.0883074,0.014938131,1.1039449,-1.9457139,0.79943234,1.8773037,1.5537525,0.39788535,-0.49531272,-1.4370352,1.0003623,0.8410879,-0.9913289,0.15745485,0.44232905,-1.5118521,-1.9748081,-1.3705083,0.11752842,0.047184132,0.72272474,0.1562442,-1.3321223,0.82741535,-0.50209045,-0.85796386,-0.8208208,1.3705729,-0.32510337,0.92136246,-0.45679024,-2.0167284,0.77819073,1.3592293,-1.6879604,0.35116142,-1.6433353,-0.8796013,0.4710182,-0.49356297,-1.540319,-0.56203073,-0.74792975,0.11485019,0.59294385,1.0806789,0.85203356,-0.2844916,2.3426995,-0.7176778,-2.077216,0.44693428,-0.5111819,-1.8134439,-0.31994978,0.52750957,-0.79248077,-0.9060044,-0.5945146,0.62720716,-0.7667804,1.4406855,0.8350828,-1.3691337,0.436756,-0.9342305,-0.58701885,-0.5398224,1.3742958,-0.36840814,-1.3804276,-1.3538328,-1.2029362,-0.6361641,-0.20616648,1.1179501,-0.8696952,0.463748,-0.9316529,0.36905104,0.18406911,0.57265526,-0.09701394,-0.30845135,-1.0851074,1.229493,-0.4344818,0.47142524,0.29762283,-0.6675888,-0.24110633,-0.8684386,-0.1543391,0.32171702,1.7343627,0.6421661,-0.39611164,1.2057976,-3.07743,0.80410504,1.0713863,0.3196743,0.8850115,0.54137725,-1.8904045,-0.6433146,-0.7131403,-0.89828914,0.4916221,-1.7892698,1.2660217,1.2143362,0.16972245,-0.42226088,-0.9209111,-1.2874365,-0.30553946,0.78019506,-1.3130425,-0.54638034,-0.06269859,-1.2766078,0.3067184,-1.197986,-0.50415367,0.24134831,0.99854076,-0.20058848,-0.19003294,-0.91562223,0.7388306,0.6062894,-0.39632004,-1.4018264,0.2661708,1.2859259,1.7502286,1.1291435,-0.25935975,1.0141768,-0.0024153837,-0.012609187,-1.6622946,0.84452695,0.40896052,0.24173,0.8491717,-1.2459071,0.44094783,2.3070037,-0.20285298,-0.9055851,-1.99059,-0.629158,-0.7049506,-1.5236279,-1.641153,0.2873126,0.23528822,-0.8580843,0.8726923,-0.8224612,-1.3814898,-0.8996295,0.007593808,1.4490062,0.6224926,-1.3583354,-0.39259246,0.5550518,-1.1764413,1.2858899,-0.5120128,-0.36201936,0.028119579,-1.0583311,0.62619597,-0.93444824,-0.6852235,-0.5574866,0.12772661,-1.117098,-1.7487797,-0.62091154,1.1936653,-0.5524274,1.6852114,-1.5539129,0.9307747,0.54523605,-0.32935366
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
deleted file mode 100644
index 22cbffa0..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-5-embeddings.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,3,0.47420266,1.2638454,-0.29428756,0.60634655,0.32506496,-0.011651864,-1.3833889,-1.2344038,-0.45170534,1.037789,1.8840408,-0.33119506,-1.0696626,0.9726885,0.8081458,0.93543977,0.71671206,0.42845142,-0.015223262,0.8131241,-1.2162951,-1.1050558,0.6366258,-1.6387162,-0.981911,-0.4953308,0.12158294,-0.6746963,-1.7174507,-0.21087167,1.0931985,-1.6311551,1.6655928,-1.1295464,-0.82908833,-0.2783438,0.35710457,0.7679168,1.2862791,-1.565548,-0.47575852,0.38371316,0.8257988,-0.6541346,-0.06021297,-1.6886668,0.3157164,-1.34399,-1.3430762,0.0015743595,-0.87187505,-1.2474334,-0.82622164,0.5390428,0.43257558,0.7165556,-0.0131348735,1.0078024,1.2916025,-1.6230769,0.18352437,1.4117194,-0.9933953,-0.17660153,-0.85399514,0.17099203,1.0725635,-0.23263934,1.7940859,0.49845862,-0.5827904,0.9100559,-2.1336787,-0.08051071,-0.15371832,0.5364136,0.036808893,0.2722593,0.4312951,-0.5169293,0.86170137,0.37777555,-0.33405718,-0.50667346,2.3349266,-3.2664793,0.23907636,0.13636166,-0.13565257,-0.2975336,-0.56251824,1.5495615,-0.25081053,-2.0128424,1.4139546,-0.1662151,0.78444827,0.5935687,-0.8553964,-0.56905305,-0.5415987,-0.8971025,1.8330035,2.7651489,1.0394961,-0.234266,-0.5788035,-1.1803628,-1.0954615,-0.30837297,0.20563039,-0.45784846,0.7992813,-0.58580524,0.8425376,0.11244689,0.720191,-1.2291318,-1.6669607,-0.4446297,2.2034054,1.0664696,0.81338793,-0.45597473,-1.262895,-0.22252297,-1.3892789,-0.7460112,0.09792392,-0.9173114,-1.1128796,0.4023646,-0.7147693,-0.7736084,0.39257377,2.3405874,-0.27565384,0.40534794,-0.8130043,-0.097918786,0.61959296,-1.450402,-1.5624176,-1.1680084,1.2857002,-1.2578357,0.41585746,1.0423982,-0.53722286,-0.38153228,-0.519493,-1.5859611,1.3256052,-0.09761098,0.5610493,2.6576302,1.2429906,-0.7566362,0.9100397,0.73991334,-0.10965712,-2.0155513,-0.009600936,-0.68587065,-1.7821826,-1.0589895,-0.13436031,0.42209822,-1.1379088,0.4846043,-0.61625975,-0.3349506,0.5386397,0.18685767,0.2667846,-0.20175363,-0.72045606,0.21726064,1.1120164,0.79319555,-0.36998132,-0.67134815,0.72676235,-0.18856223,0.4676755,-1.068567,0.32332298,-0.06708162,-1.2248513,-0.78454924,0.5460869,-0.6168076,1.5106833,1.7673894,-0.6841837,0.44412613,0.79369456,-0.8888351,0.039817635,1.6578221
-7,0,2,-0.087797664,-0.72064584,0.39195046,0.3027774,-1.0405561,-0.877984,-0.79764,0.07123449,-0.29967046,0.27734974,-0.10566526,0.7710878,-1.7671878,0.82492906,1.5373951,1.5010266,1.4846263,-0.78963494,-0.8870087,0.6919234,0.6787521,-0.43306896,0.68626046,-0.29046977,-0.96195453,-0.36133754,-0.9279964,0.1886242,-0.13149588,0.33348688,0.3524693,-0.8627565,0.7647197,-1.5118191,-0.24293655,-0.6279771,1.9434764,-0.15230925,1.2640053,-0.5305171,-1.5717331,-0.23882796,1.2569784,-1.515566,0.31609678,-1.4623228,-1.5276749,-0.6301773,-0.15530233,-1.2590178,-1.11366,-0.8283696,-0.3316043,1.6386755,0.71456283,1.4172398,-0.9240341,1.1554213,-0.20543054,-2.2690706,0.03277201,-0.60741836,-0.84084237,-0.43909258,1.1674479,-1.5343083,0.39456686,-0.63195604,0.69497895,-0.17052427,0.66388005,1.6770406,-1.6856107,0.8501753,-1.0100727,-0.069561884,-0.49753237,0.4581591,-0.8491859,-1.5388991,-0.21994227,-0.6159067,-0.4368516,-0.54839456,1.8440411,-1.9056602,0.23943676,-0.8901972,1.0178717,-0.41829002,0.47147444,0.28248173,-0.4255799,-1.3693498,1.3788788,-0.832486,0.2925802,0.038576376,-0.9931526,-0.8897898,-1.3541447,-0.38844064,-0.5666752,2.0802116,0.20410384,-0.56880385,-0.25536695,-2.847082,0.22910748,1.3398811,0.4140531,0.37236091,1.0213392,-0.6121137,0.20765178,0.22666255,-0.052260183,0.2003284,-2.3570495,1.430299,1.474341,-0.28836238,0.44256905,-0.7141607,-0.87645155,-0.20459776,0.5313733,-1.1576004,0.28822646,-0.3185678,-2.0415008,0.5557355,-1.3157727,0.27283114,0.18704557,-0.345221,-0.70967424,-0.43224388,-1.6015024,1.0732054,0.91898453,-0.6834103,-1.5872614,-0.59699357,0.9072513,1.1939424,0.62792987,-0.49085766,0.9603636,0.668255,0.6667969,-1.698892,0.26019427,0.57732755,0.53784436,1.2200284,0.3280351,-0.15348202,2.1410975,-0.0141839655,-1.0378197,-2.6077065,-0.30667314,-0.067144245,-1.3892245,-1.7354017,0.5355618,0.9041617,-0.32855982,0.12410726,-0.40722403,-1.3432459,-0.7320008,-0.25400352,0.17272069,1.1034757,-2.285054,0.6256184,1.7550694,-1.3452939,-0.017986381,-0.5405325,0.05288163,0.28705782,-0.98898715,0.76313794,0.15638609,-0.6830085,-1.3387135,0.62448525,-0.7181843,-1.9862365,0.070949845,1.8409837,-1.5815736,1.6396283,-0.85405725,0.19265085,0.4421183,0.6777789
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
deleted file mode 100644
index 3df20b08..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-6-embeddings.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,3,1.6896036,0.2908516,-0.23292227,1.333588,0.4545924,0.34372577,-0.41658887,-0.4220771,0.60733056,1.2122439,0.8359348,0.58897126,-1.7851542,1.4177668,0.74911904,0.5095494,0.26941735,0.21018945,-0.8802367,0.56112623,0.49656504,-0.8069377,1.1553066,-0.012634999,-0.1019033,-1.7098432,-0.7785347,0.08457122,-1.0872964,-0.46575698,0.17051879,-1.0981392,1.2801946,-0.20283177,-0.377797,-1.0450277,0.8350198,-0.44490838,0.37728155,-1.326767,-1.3647465,0.49804258,1.0723178,-1.5584799,0.62113315,-1.7105298,0.9363782,0.962023,-0.5543043,-0.22153258,0.51130754,-0.517027,0.119101755,-0.16186844,0.33424366,-0.80313826,1.655926,0.4785711,0.5240289,-0.8545002,1.1488025,-0.12382955,-2.314366,-1.7480171,0.34907153,0.3368917,-0.9671285,-1.1980494,1.7140709,-1.4368649,0.6655941,-0.27778602,-1.7404888,1.058552,-0.34576625,-0.026606066,-1.9635391,2.325346,0.718786,-0.63990754,-2.1113021,-0.7639529,-0.37198016,-0.00446413,1.4715579,-1.4121689,0.046070803,0.04943272,-0.67208874,0.77220094,-0.7848665,1.6651015,-0.50486845,-1.0169153,0.19552442,0.60602796,1.06237,0.15507467,-0.32467127,0.27232736,-0.9548156,-0.060560405,1.6082474,1.8419516,-0.6165264,0.5530548,0.531371,-1.6408411,-0.08376483,-0.6223387,0.053751606,0.17250703,-0.58378434,-1.9890922,-2.1878114,-0.63507706,-1.3977063,-0.09614923,-0.996112,-0.87689984,0.8914697,1.3250511,-1.3865025,0.07022147,0.14449127,-0.4635789,0.22937027,-1.0663304,-1.1240877,0.1710589,0.38474682,1.002527,0.06262817,-1.4390726,-0.07946953,2.1103234,-1.7293843,1.1202103,-0.19353478,-1.0497701,-0.18288228,-0.8191537,-0.91999054,-0.001152092,1.4552096,0.02856364,1.5781829,1.566296,-0.37651303,-0.75987285,-0.9691587,-1.9078715,1.1331105,-0.11489664,0.5372741,2.1304917,-2.0829363,0.2657544,0.73497146,-0.30037382,0.7455473,-1.2468303,-0.9414173,-0.96966416,-2.6400287,-1.1223292,0.87630415,-0.9678274,-0.9243957,1.250168,-1.4089605,-0.8066182,-0.8789456,0.509323,0.8661667,-1.0657727,-0.44514495,0.1678456,-0.8246042,-0.62025964,0.22096853,-0.5744515,-0.051140007,0.9549694,-0.7317687,0.21573786,-0.73088586,1.3564854,-0.5641902,-0.35502443,-1.0255394,-0.57220846,-0.90364045,0.36711854,-0.095699474,0.6653445,-0.5859865,1.2525156,1.7926835,0.031228162
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
deleted file mode 100644
index 80e7457d..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-1-best-embedding-3-embeddings/sequifier-model-categorical-1-best-embedding-3-7-embeddings.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,2,-0.1672056,-0.9560093,0.28606495,0.60610676,-0.3845802,-0.20699823,-1.1377773,0.2095035,-0.13692239,-0.3980985,1.1865553,1.430365,-1.054823,0.47550726,0.9458505,2.1207294,2.1566586,-0.24435459,-0.64201486,1.7221576,0.9315173,-0.73526466,0.3490929,0.724757,-0.53412163,-0.38585144,-0.78077614,0.5533652,0.0011221742,1.0769403,0.17883608,-1.3198036,1.3214654,-1.3146971,-0.7111261,-0.4876436,1.6924083,-0.51583886,1.3672888,-0.55709535,-0.89117086,-0.7621175,1.6342846,-1.2171476,0.2161075,-2.0004432,-0.90466124,-0.6608028,0.09917505,-1.515609,-1.4084437,-1.4207675,-1.6178051,1.1580858,0.932027,1.5196207,-1.0533146,1.2685808,0.001325828,-1.9838752,-0.020375686,-0.64693415,-1.3806916,-1.0592812,0.5494306,-1.396179,-0.15649273,-0.5929404,0.96867883,-0.40543112,1.0557821,0.8631483,-2.2666907,0.67865187,-0.9861259,-0.46339324,-0.5806145,1.3590288,-0.9761064,-1.6902065,0.015556082,-1.0601697,-0.6836848,-0.48810717,1.0633674,-2.1536198,0.27614254,-0.9428517,0.88297933,-0.13327223,-0.69135034,0.05794356,0.2954672,-1.8476416,1.6050957,-0.7725617,-0.34552756,-0.399923,-0.91350776,-1.4197913,-1.434273,0.05444653,0.38957354,1.6343312,-0.01868396,0.20149925,-0.51843625,-2.798834,-0.2381288,0.39474154,0.77599514,0.26147756,0.769546,-0.8783735,0.24983686,0.2294478,-0.5027371,0.1792123,-2.9645298,0.849348,0.8776484,0.17855346,-0.31652552,-0.49537584,-1.4034135,-0.5197447,0.6516527,-1.1936587,0.43764108,-0.31979695,-1.1554341,-0.09437473,-1.4360908,-0.666431,0.19690526,0.22096193,-0.7755801,-0.64809597,-1.0999116,1.1237248,0.32265925,-0.71968305,-2.2545598,-1.4951686,0.55628073,1.1043607,1.1391095,0.12772316,0.3919614,0.6803616,-0.48576495,-1.285276,0.8191475,0.3090739,-0.07296532,1.5340542,0.30991074,-0.0071372786,1.7126004,-0.8783499,-0.07996719,-2.1059916,-0.3130492,-0.31512108,-1.8248317,-1.577193,0.5107035,0.71821797,-0.25824943,-0.4562349,-0.26445264,-0.95276433,-0.7283461,0.17178622,0.211934,0.949343,-1.3720604,0.15322232,0.7012931,-0.5517222,-0.15245192,0.062418915,-0.6320953,0.5529452,-0.42877352,1.0122427,-0.0774351,-0.28508034,-1.1758733,0.21220773,-1.3312134,-1.4357238,0.21157022,0.591768,-1.174219,1.9612663,-1.2883047,0.06013686,1.073086,0.5143656
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
deleted file mode 100644
index 8e069e93..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-0-embeddings.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-0,0,1,0.24902344,-0.004058838,-1.1171875,-0.29101562,1.21875,0.76953125,-0.56640625,1.03125,0.9609375,-0.14453125,0.66796875,2.828125,0.4453125,0.28320312,-0.91015625,-0.04663086,-0.67578125,0.5625,3.265625,1.6875,1.015625,-0.8984375,0.63671875,0.25585938,0.028808594,0.7890625,-0.56640625,-0.4453125,-0.91796875,-0.875,0.095703125,0.056152344,0.16601562,1.1328125,-1.125,0.00289917,0.2890625,0.2265625,0.19140625,0.016479492,-0.37890625,-0.42578125,-0.25195312,-1.078125,0.11669922,0.76171875,0.14355469,-1.484375,0.13183594,0.13085938,0.49804688,-1.546875,0.0703125,0.119140625,-1.3515625,0.05053711,-0.33789062,-0.44726562,1.234375,-1.2578125,-0.06982422,-0.359375,2.21875,-0.40820312,-0.6953125,-1.15625,0.02746582,-1.4140625,-0.25976562,0.8125,0.044433594,1.40625,-0.53125,-0.77734375,1.0859375,2.875,-0.91015625,-0.13769531,-2.328125,0.02319336,0.4921875,-0.06640625,0.546875,0.3125,0.045654297,-0.296875,-1.15625,-1.34375,0.38671875,1.65625,1.421875,0.671875,-1.625,1.640625,-0.013549805,-0.024658203,-0.32617188,0.546875,-0.76953125,-0.25195312,-0.1640625,0.24316406,1.0078125,-1.2421875,-1.4765625,0.26953125,-0.63671875,0.69140625,-0.45507812,-0.14648438,1.203125,0.49804688,0.13867188,-0.24902344,1.4296875,-1.53125,-1.0,-0.4140625,0.81640625,0.18652344,-0.21289062,0.39257812,0.4765625,-0.12597656,-1.4765625,2.421875,-0.49804688,-1.265625,2.21875,1.4296875,0.57421875,-2.328125,-1.6484375,-0.40625,-1.6875,0.546875,-1.3671875,-0.609375,-0.08544922,-0.78515625,1.4140625,-1.640625,-0.87890625,2.03125,-0.671875,-0.37695312,-0.14160156,1.0,-0.76953125,0.39257812,1.09375,-0.9375,0.54296875,-0.875,-1.0390625,-0.08886719,1.046875,0.515625,0.6015625,-1.15625,0.11279297,1.375,0.73046875,2.28125,0.064941406,0.080078125,-1.28125,-2.359375,1.7734375,0.03564453,0.37890625,-0.7890625,-0.3359375,0.7421875,-0.95703125,-0.19726562,-0.90234375,-1.109375,-0.69140625,0.6484375,-0.76953125,-0.053222656,0.41210938,1.109375,-1.4453125,0.9453125,-0.9609375,-1.2890625,1.8046875,-0.93359375,-1.4296875,-0.53515625,0.43359375,-0.578125,-0.20507812,1.375,-0.421875,0.31054688,0.33398438,1.21875
-0,0,2,0.29882812,0.50390625,0.66015625,-0.47851562,0.26171875,-0.026367188,0.6796875,-0.6171875,0.58984375,-0.66796875,0.68359375,0.095703125,1.6015625,-0.73046875,0.00793457,-0.31054688,-0.19433594,-0.004638672,2.625,0.16601562,-0.15136719,-1.3984375,0.6640625,0.05517578,-0.703125,-1.0,0.29492188,-0.3125,-0.37890625,1.609375,0.64453125,-1.0625,1.515625,-1.4140625,0.19726562,0.20996094,0.07080078,-1.046875,-0.84765625,0.82421875,-2.140625,0.29101562,-0.82421875,-0.64453125,-0.27734375,-0.51953125,0.31445312,-0.984375,1.3515625,0.671875,-0.78515625,0.33984375,-0.93359375,1.5390625,0.20019531,-0.30273438,0.140625,-0.421875,-1.1484375,-0.828125,1.3359375,0.50390625,-0.051757812,1.25,-0.1953125,0.44921875,-0.05859375,-1.796875,-1.265625,1.5390625,-0.6640625,0.578125,0.640625,0.06738281,0.31640625,1.578125,0.94921875,0.98046875,-1.125,-0.012817383,-0.80078125,-0.66796875,0.5625,-0.1796875,-0.27734375,-0.609375,-1.2421875,-1.953125,1.484375,-0.50390625,1.1953125,0.953125,-0.9609375,0.33007812,-1.109375,-0.74609375,-1.8984375,-0.75,-1.296875,0.04248047,-1.375,1.125,1.796875,-0.018920898,0.44335938,2.609375,-1.03125,0.5859375,0.28515625,-0.625,0.69921875,1.15625,0.640625,-0.07714844,0.46484375,0.30859375,-0.21582031,0.24707031,-0.06347656,1.296875,0.5390625,-0.6171875,1.1875,-0.9140625,-1.1171875,1.1171875,-1.8203125,-1.40625,-0.18652344,-0.10205078,1.109375,-0.609375,-0.54296875,-1.3515625,0.38476562,0.40234375,0.40039062,-0.40625,0.21582031,-1.8828125,2.484375,0.57421875,-0.671875,2.34375,-0.25,-1.5234375,2.21875,-1.0703125,-1.3359375,0.82421875,-0.9609375,0.20019531,1.0390625,-1.4140625,-0.32226562,-0.03173828,1.2578125,-0.82421875,1.0703125,0.32421875,-0.38867188,-0.53515625,2.390625,1.7734375,-0.94921875,-0.7734375,-1.15625,1.015625,1.1015625,-0.39453125,-0.30664062,-0.50390625,-1.734375,-0.9140625,-0.88671875,0.265625,0.98046875,-1.4375,0.51171875,0.2578125,-1.9296875,-1.015625,1.171875,0.26953125,-1.5390625,-0.41796875,0.43164062,-0.20898438,-0.65625,-0.05859375,-1.7109375,1.7890625,1.1875,-0.53125,-1.0859375,0.8984375,1.796875,-0.34765625,1.359375,1.0703125
-0,0,3,0.6640625,1.21875,-1.203125,-1.3125,-0.515625,-1.8984375,0.31445312,1.1875,-0.53125,-0.68359375,-0.05419922,-1.1796875,-1.828125,-0.33203125,-0.9375,1.0703125,-0.27929688,0.91796875,0.1640625,0.022338867,0.90234375,-0.23535156,-0.03564453,1.328125,-1.7265625,-0.96484375,-2.390625,0.6640625,1.984375,0.9765625,0.76953125,1.3984375,-1.515625,-0.3828125,0.73828125,-0.01928711,0.9453125,1.359375,-0.044433594,0.62109375,-2.15625,-0.27734375,0.072265625,1.7421875,0.71484375,2.03125,0.43359375,0.58984375,0.1875,1.25,0.28320312,-0.23144531,-1.3515625,-0.80078125,-0.1328125,-0.43359375,-0.9765625,0.013122559,-0.06591797,-0.60546875,-0.1640625,0.53515625,0.34765625,-0.38085938,1.328125,0.23535156,0.16503906,0.0625,-1.0625,0.58984375,0.049560547,-0.16796875,0.390625,-0.828125,0.67578125,0.08935547,2.09375,-0.044677734,-0.42578125,0.03930664,1.2734375,0.609375,1.578125,0.9140625,-0.08691406,0.9765625,-1.328125,0.51953125,-1.75,0.35351562,0.9609375,2.203125,1.7890625,-0.40820312,-0.06640625,-1.421875,-0.33203125,0.014221191,0.97265625,1.03125,0.87890625,-0.18457031,1.0234375,-0.17382812,-0.94921875,-0.34570312,-1.515625,-1.4375,-0.8046875,-0.0061035156,-0.37890625,-0.22851562,2.546875,-0.8828125,-0.47851562,-0.88671875,-0.35546875,-1.03125,0.9921875,0.6875,2.0625,-1.8828125,-0.86328125,-2.953125,-0.21777344,0.19042969,-0.98046875,1.2421875,0.17285156,-1.9765625,0.90234375,-0.37890625,0.67578125,-1.078125,-1.875,0.58984375,0.53515625,-0.546875,0.11035156,-2.1875,0.32226562,0.890625,1.71875,1.21875,-0.62109375,-0.49414062,-1.625,1.078125,-0.37695312,0.12988281,-0.041259766,-1.359375,-0.7890625,1.4296875,0.19433594,-0.71484375,-0.6015625,1.0546875,-0.37695312,1.90625,-2.0625,0.96875,-0.33007812,0.083496094,-1.296875,-0.14648438,-0.26757812,0.29492188,0.47070312,-0.5859375,-0.32421875,0.47265625,-1.109375,0.9921875,0.953125,0.24804688,0.9921875,-1.546875,0.04248047,-0.103027344,1.0,0.52734375,0.62109375,-0.88671875,-0.50390625,-0.32421875,-0.24609375,0.0625,-0.54296875,-1.046875,-1.2421875,0.31445312,-0.036132812,-0.86328125,1.296875,0.119140625,1.921875,-0.23046875,0.09277344,0.42382812
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
deleted file mode 100644
index 4c736c3e..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-1-embeddings.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-1,0,1,1.0546875,0.8984375,-0.5546875,0.82421875,-0.65625,-0.375,0.77734375,1.6484375,0.5625,-1.6171875,-0.53515625,0.5390625,0.76171875,-0.09814453,-1.4453125,0.82421875,-0.056884766,0.5859375,0.33984375,-0.37304688,0.8359375,-0.9375,-0.03515625,-0.91796875,0.16015625,-0.9453125,-1.515625,-0.14648438,-1.1484375,-0.8828125,1.328125,-1.328125,1.71875,0.7578125,0.22460938,-1.578125,-0.12792969,-1.7265625,-0.7265625,-1.125,-1.2421875,-1.3046875,-1.28125,0.018188477,-1.875,0.9296875,0.48046875,1.015625,1.125,1.546875,-1.3125,-0.12792969,-0.734375,1.7578125,-0.671875,0.5078125,1.15625,0.66796875,-0.15332031,-0.22460938,0.05493164,0.515625,0.084472656,-0.12011719,0.7265625,-0.5859375,-0.90234375,-0.40625,-0.703125,0.25,1.59375,-1.8125,0.5703125,-1.046875,2.296875,2.578125,1.5390625,0.82421875,-0.34375,-1.125,-0.38671875,-1.25,-0.42578125,-1.265625,0.9375,-1.1875,1.28125,0.59375,-0.30859375,1.4921875,2.234375,-1.34375,0.16894531,0.734375,-0.4609375,0.56640625,0.18457031,-0.67578125,-0.21484375,0.58984375,-1.578125,1.15625,1.03125,-0.028076172,2.0,-0.55859375,0.40234375,0.57421875,0.8828125,1.5859375,-0.31835938,0.015380859,0.296875,-0.61328125,1.0,-0.29101562,-1.6875,0.17382812,-0.15039062,-0.28710938,-0.6640625,-0.40820312,0.22265625,-0.02709961,-0.578125,0.58984375,0.57421875,-0.58984375,-0.49804688,-0.515625,-0.37890625,0.2109375,-0.2578125,-1.28125,-1.75,-0.46875,-0.60546875,0.37890625,-0.49414062,-2.328125,2.171875,-0.6640625,-0.51953125,-1.9375,-0.30664062,-0.19335938,-0.7265625,-1.0859375,0.61328125,3.640625,-0.44921875,1.0546875,-1.8046875,0.828125,0.19824219,-0.984375,0.14550781,1.2109375,0.296875,-0.859375,0.2578125,0.36914062,-0.02758789,0.12451172,-0.8359375,0.53125,0.14941406,0.83203125,1.96875,-2.34375,-0.33398438,0.296875,-0.4140625,-0.25390625,0.27734375,1.15625,1.0078125,1.203125,-0.4296875,0.38085938,-0.828125,1.4375,0.87109375,0.13378906,0.32226562,-1.203125,1.4375,0.91796875,-1.3203125,-0.19238281,-0.40429688,-0.828125,0.81640625,0.29101562,-1.6875,1.109375,0.05078125,-0.03540039,-0.35546875,-0.38085938
-1,0,2,-0.5546875,1.265625,-1.359375,-1.5078125,0.30859375,-1.9609375,0.34375,1.6328125,1.6015625,-0.58203125,0.21777344,1.03125,-0.953125,0.51953125,-1.2109375,0.53515625,0.26953125,0.22363281,-0.37109375,-0.071777344,0.33789062,-0.39257812,1.328125,-0.037109375,-0.73828125,-0.9296875,-2.421875,1.1015625,1.578125,-0.4765625,1.375,0.20800781,-1.125,0.1484375,1.5703125,-1.09375,0.6640625,0.8671875,-0.7734375,1.0546875,-1.5546875,0.43164062,-0.21679688,-0.265625,-0.3203125,2.609375,-0.059814453,1.8125,1.15625,1.609375,-1.0234375,1.0859375,-0.640625,-0.08886719,-0.828125,-0.088378906,-0.51953125,0.484375,-1.109375,1.3984375,0.39453125,0.9453125,-0.15527344,0.609375,0.546875,-1.1953125,-0.12451172,0.71875,-1.6484375,1.2578125,0.53125,0.78125,2.0625,-0.5390625,0.5625,2.078125,0.5078125,0.703125,-0.83203125,-0.40429688,0.48242188,-0.36328125,0.765625,-0.13183594,-0.119628906,0.9375,-1.421875,-0.26953125,-1.15625,-0.68359375,0.58203125,1.5390625,0.99609375,-0.26171875,0.78515625,-1.875,-0.5390625,-0.25976562,0.41015625,-0.5625,0.14746094,0.49609375,0.6953125,0.15820312,-0.76171875,0.06542969,-1.0546875,-0.40039062,0.044677734,0.5703125,-1.6484375,-0.06689453,1.6640625,-2.328125,0.64453125,-0.1484375,-1.578125,-0.31054688,0.921875,0.083496094,1.1796875,-2.359375,0.028686523,-1.3671875,0.37890625,0.97265625,-0.25976562,-1.359375,1.03125,-0.24902344,-1.3203125,-0.91015625,-0.5390625,0.5859375,-1.90625,1.5390625,1.1875,-0.19335938,0.91015625,-2.03125,1.3046875,0.12597656,0.6484375,0.828125,-1.234375,-1.3359375,-0.81640625,-1.3046875,-1.453125,-0.6953125,0.60546875,-0.61328125,-0.48632812,0.87109375,-0.73828125,1.8125,0.5625,0.39648438,0.45898438,0.29492188,-0.26953125,1.1171875,0.7265625,0.36523438,0.02722168,-0.7265625,-0.38671875,0.7578125,-0.49414062,-2.0,-1.03125,-0.059326172,-1.578125,-0.6953125,0.8515625,-0.125,1.0703125,0.83984375,-0.31445312,0.022705078,0.08300781,-1.1171875,0.30664062,0.16894531,0.33203125,0.06591797,-0.83203125,-0.84765625,-1.7734375,-0.99609375,0.29101562,0.8125,-0.7265625,0.6796875,-0.81640625,2.625,2.03125,0.10107422,-0.8984375,-0.48632812
-1,0,3,1.2578125,0.56640625,-0.9921875,0.10546875,-1.359375,-0.11669922,0.61328125,1.0078125,-1.15625,-1.4609375,-0.21484375,-0.38085938,-0.43164062,0.45507812,-1.75,-0.8203125,-1.0625,0.2890625,-0.953125,-0.53125,0.59765625,-0.22167969,-0.57421875,-0.46289062,-0.10205078,-2.1875,-2.390625,-0.43554688,-0.012329102,0.19042969,-0.3828125,-1.453125,0.28710938,0.87890625,-0.32617188,-0.546875,-0.09863281,-2.375,0.453125,-0.08154297,0.41210938,-0.21386719,-0.859375,1.2421875,0.94140625,0.34960938,-1.0234375,0.60546875,-0.019897461,1.171875,0.16308594,-0.19824219,-1.65625,0.703125,-0.90625,-0.671875,-0.640625,1.265625,-1.0625,-1.0234375,-0.28125,0.765625,1.2734375,0.7734375,1.921875,-0.34960938,-0.28125,-0.5078125,-1.9375,0.49804688,1.2109375,-0.36328125,1.671875,-0.69140625,0.890625,0.828125,2.515625,1.2265625,0.60546875,0.10253906,-0.4140625,-0.38867188,0.375,-1.3359375,-0.28125,1.0,0.32421875,-0.60546875,0.24023438,0.984375,0.5390625,1.34375,0.22167969,0.578125,-0.9609375,-1.28125,-0.7421875,-0.13378906,1.1484375,1.078125,0.38671875,1.2421875,-0.30078125,-1.2421875,1.2734375,-1.078125,-0.71484375,0.6875,0.41210938,-0.3359375,-0.9609375,1.0078125,1.578125,0.27148438,0.119628906,0.46875,-0.30664062,0.515625,0.12597656,1.921875,0.30664062,-2.09375,-1.1484375,-0.6015625,-0.055664062,-1.2578125,0.171875,-1.046875,0.002670288,-0.17089844,0.43359375,-0.5703125,-0.19433594,-0.022949219,-1.046875,2.453125,-0.91015625,-1.703125,1.3984375,-2.3125,0.6484375,-0.671875,-0.28320312,-0.65234375,0.875,0.31054688,0.37695312,0.19921875,2.1875,3.578125,0.421875,0.265625,-1.6171875,1.0234375,-0.008728027,0.29492188,0.66015625,1.140625,-1.3515625,0.625,0.9765625,1.734375,0.80859375,-0.62109375,-0.5078125,-0.546875,-0.34765625,1.6796875,0.51953125,-1.7578125,-0.61328125,-0.93359375,0.23632812,-0.55078125,0.703125,1.390625,-0.31445312,1.1171875,1.328125,-0.14257812,0.06738281,0.123535156,1.359375,-0.640625,0.48828125,-0.29101562,0.640625,1.4140625,-0.0029296875,-1.5234375,0.21484375,-1.6015625,0.54296875,-1.7265625,-0.80859375,1.2109375,0.30664062,-1.0859375,-1.2421875,-0.35742188
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
deleted file mode 100644
index 4202cade..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-2-embeddings.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-2,0,0,1.2109375,-0.096191406,-1.3359375,-0.6015625,0.9609375,-1.484375,-0.8203125,3.140625,-0.5234375,-1.6640625,0.0234375,1.5859375,-0.53515625,-0.75,-0.5703125,1.75,-0.27148438,-1.40625,0.059570312,0.48632812,-1.375,-0.6796875,-0.44921875,1.2421875,-0.095703125,0.33007812,-2.484375,-0.15527344,-0.890625,-0.06933594,-0.020019531,0.7890625,0.69140625,0.7734375,-0.07324219,-1.0546875,1.25,-0.171875,0.515625,0.6953125,-0.06201172,-0.008239746,0.11279297,-0.78125,-0.115234375,0.65625,-0.46484375,0.96484375,-0.43945312,2.21875,-0.609375,-2.234375,-1.3359375,-0.048095703,-0.15039062,-0.33398438,0.921875,0.29882812,0.40234375,-1.046875,-0.3671875,1.4765625,-0.12988281,0.53515625,-0.26953125,-0.029663086,-0.000705719,1.4765625,0.16796875,-0.11328125,-0.4140625,-0.2890625,0.9140625,-0.67578125,1.6328125,2.140625,0.62890625,0.26953125,-0.24414062,-0.75390625,1.671875,0.7578125,0.53515625,-0.6328125,-0.44140625,-1.1484375,1.328125,1.84375,0.5078125,1.65625,1.734375,-0.7734375,-1.046875,1.375,-0.28320312,0.033691406,-1.7265625,-0.73828125,-1.0859375,-0.21875,-1.0703125,-0.24804688,1.4296875,0.94140625,-0.46875,0.25390625,-1.640625,-0.62109375,-0.28515625,0.016723633,-1.140625,-1.9453125,1.421875,-2.265625,0.84765625,0.27539062,-2.0625,1.2734375,-0.30273438,-1.21875,1.2578125,-0.47265625,-0.92578125,-0.36523438,0.4140625,2.71875,-0.43945312,-0.41210938,0.4453125,-1.8671875,-1.484375,-1.640625,0.3671875,-1.1015625,-0.8984375,-0.012634277,1.3203125,0.3515625,0.43554688,-1.3515625,1.453125,-1.8203125,-0.6328125,-0.15917969,0.16210938,0.47070312,-0.32226562,0.83984375,-0.73828125,0.828125,-0.6171875,-0.019165039,0.80859375,-0.453125,0.046142578,-0.76171875,0.31835938,0.014831543,1.6640625,1.015625,-1.1484375,1.859375,1.2890625,0.8203125,-0.57421875,0.77734375,0.17285156,1.3203125,1.2578125,0.7734375,-1.2890625,-0.16796875,0.06640625,1.09375,-1.28125,0.4296875,0.35351562,-0.6953125,-0.057128906,-1.171875,0.46679688,0.38476562,0.3046875,-1.0859375,0.22363281,-0.24023438,-1.40625,0.01373291,-0.84375,-1.34375,0.21386719,1.4375,-0.11621094,-0.42773438,0.123046875,1.1953125,0.49804688,0.6015625,-1.0703125,0.45703125
-2,0,1,0.31445312,-0.33398438,-0.72265625,-0.44140625,0.40625,-0.546875,-0.0046691895,1.578125,-0.13769531,-0.8359375,-0.5390625,-0.37695312,1.2265625,0.50390625,-1.6484375,0.44140625,-0.5234375,-1.984375,0.42578125,-0.6953125,0.5625,-0.328125,-0.27734375,-0.06689453,-1.328125,-1.8125,-1.625,-0.088378906,-2.0625,0.48828125,-0.6875,-1.1171875,0.55859375,0.076171875,0.546875,1.671875,0.625,-1.453125,-1.8671875,1.1875,-1.4765625,0.34960938,-0.546875,-0.040771484,0.23144531,1.0625,0.5546875,-0.55078125,0.2421875,1.734375,-2.109375,-1.3203125,-1.4140625,0.7265625,-1.96875,0.6171875,0.12695312,-0.671875,-0.19042969,-2.171875,1.1875,-0.16015625,0.42773438,-0.07324219,0.17675781,0.055419922,-1.0859375,0.19042969,-0.66796875,2.0,0.90625,-1.609375,-0.45898438,-0.5,0.04248047,1.921875,1.109375,1.1015625,-0.61328125,-0.8828125,1.5546875,0.32226562,0.27539062,-0.51171875,-0.671875,-0.028320312,0.28515625,0.9609375,-0.6640625,0.6796875,2.109375,-0.17285156,0.15039062,0.41796875,-1.3984375,-0.19824219,-0.30664062,-0.53515625,1.421875,-0.33203125,-2.25,0.13964844,2.21875,1.15625,-0.47070312,-0.23046875,1.484375,-0.9296875,0.40820312,0.7109375,0.27539062,0.35742188,0.859375,-1.4375,0.3046875,-0.47070312,0.18554688,-0.26953125,-0.65625,-0.65625,1.3359375,-0.55078125,0.4140625,-1.171875,-0.24511719,1.2734375,-0.4453125,-2.3125,-0.00579834,-0.04345703,0.6796875,-0.061035156,0.19140625,-1.3046875,-1.328125,-0.20019531,-0.35351562,-0.46875,-0.018432617,-1.8046875,3.09375,-0.7734375,0.38671875,-0.052246094,-0.07080078,1.8359375,-1.21875,-0.017944336,-1.9296875,1.921875,0.23535156,1.140625,1.0234375,-0.6015625,-0.083496094,-0.89453125,0.43359375,1.625,1.4453125,0.66796875,-0.71875,1.34375,1.3671875,0.49609375,-1.3359375,0.53515625,-1.1171875,-0.33789062,2.375,-0.25390625,0.22949219,-0.14746094,-0.66796875,-0.31445312,0.11816406,0.453125,-0.9375,-0.91796875,0.9296875,-0.5859375,-0.17871094,-0.107421875,1.265625,0.48242188,0.6640625,-0.30078125,1.328125,0.8203125,-0.53515625,-1.046875,-0.23632812,0.53515625,2.15625,0.026489258,0.828125,-0.15429688,1.4140625,-0.2109375,1.2265625,0.26757812
-2,0,2,0.7734375,0.110839844,-0.37109375,-0.22460938,0.20996094,-0.76171875,-0.26953125,1.4140625,0.5546875,-1.09375,0.5859375,1.421875,0.44140625,0.55078125,-1.59375,0.796875,0.18847656,-1.203125,-0.11621094,0.11767578,0.9921875,-0.5390625,-0.39648438,0.578125,-0.29492188,-0.72265625,-0.51171875,0.060791016,-1.0078125,-0.061035156,-0.68359375,-0.51171875,1.796875,-0.453125,1.28125,1.59375,0.828125,-0.15917969,-2.21875,1.7265625,-1.8359375,0.375,-0.59375,-0.052001953,0.62890625,0.81640625,-0.3359375,-0.1328125,-0.4453125,1.90625,-1.3671875,-0.10986328,-0.69140625,-0.08984375,-0.796875,0.71484375,0.6171875,0.609375,-0.08105469,-1.25,1.609375,-0.21484375,1.8671875,-0.34765625,-1.7421875,0.8515625,-1.171875,0.53515625,-0.52734375,1.265625,-0.103027344,-1.171875,-0.09863281,0.609375,-0.15820312,1.9375,1.390625,0.38867188,-1.0,-1.46875,1.3828125,-0.20019531,0.59765625,-1.2890625,-1.375,-0.3359375,0.19042969,1.40625,-0.8125,0.359375,1.96875,-0.035888672,-0.8203125,0.6171875,-1.5390625,0.9453125,-0.17675781,0.8828125,0.7421875,0.9921875,-2.46875,0.012023926,2.0,1.03125,-1.3984375,0.35351562,0.90625,-1.6640625,-1.0703125,0.18164062,0.578125,-0.36523438,1.546875,-1.5078125,-0.35742188,-1.21875,0.953125,-0.044677734,-0.3046875,-1.2890625,2.140625,0.44140625,-0.59765625,-1.2578125,0.60546875,0.41992188,-0.73046875,-2.171875,0.26171875,-0.36523438,0.18066406,-1.171875,-0.54296875,-0.78515625,-1.8515625,1.6640625,0.31054688,-0.625,-0.5390625,-1.3515625,1.8125,-1.6953125,-0.68359375,-0.3828125,-0.17675781,1.3125,-0.5859375,0.51953125,-2.6875,-0.1171875,0.2578125,1.1328125,0.69921875,-1.03125,-0.23144531,-0.77734375,0.89453125,0.4765625,1.1171875,0.546875,-0.41015625,0.36523438,-0.265625,0.91796875,-1.5625,1.2734375,-1.1015625,-0.578125,1.3984375,0.08886719,-1.6796875,-0.7890625,-0.640625,-0.34960938,0.1640625,0.36523438,0.80078125,-0.46875,0.30859375,-0.83203125,-0.49804688,-0.03173828,0.640625,-0.49023438,1.234375,-1.5625,0.47070312,0.61328125,0.24804688,-0.86328125,1.09375,1.546875,0.859375,-0.7421875,-0.7109375,-0.66796875,2.21875,1.6484375,1.2109375,-0.2421875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
deleted file mode 100644
index e1904552..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-3-embeddings.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-3,0,0,1.09375,-0.07519531,-1.4921875,0.24316406,0.8515625,-1.703125,-0.57421875,2.953125,-0.059814453,-0.76171875,0.22753906,2.6875,0.265625,-0.36328125,-0.703125,1.734375,0.27929688,-1.328125,0.35742188,0.921875,-0.24414062,-0.734375,-0.65234375,0.55078125,0.94140625,1.328125,-2.703125,0.38085938,-0.60546875,-1.0234375,0.34570312,-0.042236328,1.6953125,1.3515625,0.55078125,-0.62890625,0.99609375,-0.015625,-0.40429688,0.107910156,0.044433594,0.30859375,-0.18066406,-1.65625,-0.13964844,0.87890625,-1.5703125,0.44726562,0.072265625,0.7890625,-0.8515625,-1.421875,-1.1796875,-1.046875,-0.18945312,-0.20703125,0.74609375,0.29882812,-0.24804688,-1.0703125,0.21289062,0.71484375,1.3125,0.80859375,-1.265625,-0.32421875,0.3125,0.28320312,-0.57421875,-0.40234375,0.4140625,-0.42578125,0.50390625,-0.46679688,2.265625,3.09375,1.0390625,0.024291992,-0.8125,-0.84375,1.4453125,0.68359375,0.82421875,-1.609375,0.03930664,-0.39648438,0.47265625,1.3125,0.578125,2.203125,1.8671875,-0.765625,-1.0078125,1.359375,-0.6171875,-0.091308594,-1.390625,0.00680542,-0.14648438,0.6796875,-0.51171875,0.16601562,0.055664062,1.265625,-1.0390625,-0.18261719,-1.109375,0.03173828,-1.3828125,-0.359375,-1.1328125,-1.53125,1.765625,-1.8828125,0.61328125,-0.036865234,-1.4609375,0.6953125,-0.0054016113,-0.39257812,1.2734375,0.11328125,-0.8828125,-0.109375,-0.045166016,2.421875,-0.7578125,-0.62890625,1.40625,-1.546875,-1.1640625,-2.8125,0.10205078,-0.89453125,-1.203125,0.33984375,1.1640625,-0.27929688,0.18359375,-1.1875,0.92578125,-1.2734375,-1.3984375,-0.55078125,-0.20117188,0.53515625,-0.8203125,0.765625,-0.34960938,0.02746582,0.41796875,-0.54296875,0.21386719,-1.203125,0.28125,-0.12988281,0.96875,0.044433594,1.328125,0.58984375,-0.18164062,2.46875,0.42578125,0.94921875,-0.064453125,0.3984375,0.18652344,0.48046875,1.2734375,1.2578125,-1.03125,-0.640625,-0.88671875,0.9140625,-1.2890625,0.29492188,0.14257812,-0.29882812,0.23535156,-0.3046875,-0.546875,-0.32421875,0.49023438,-1.203125,0.546875,-0.5234375,-1.1015625,-0.90625,-0.51953125,-0.9921875,0.5703125,0.96875,-0.4296875,-0.80859375,-0.16503906,1.9296875,-0.1875,0.93359375,-1.4609375,-0.16992188
-3,0,1,-0.31054688,0.45898438,-1.703125,-1.1796875,-0.609375,0.15917969,-0.45703125,2.109375,0.5625,-0.81640625,1.7421875,1.1171875,1.453125,-0.3671875,0.08105469,0.95703125,-0.072753906,-0.93359375,0.76171875,-0.52734375,0.625,-0.012329102,0.8671875,0.96484375,-0.055419922,0.13378906,0.22558594,-0.30664062,-1.2890625,-1.7265625,-0.3125,1.6796875,1.2578125,2.65625,0.9453125,-0.14648438,0.30078125,0.24414062,0.52734375,0.74609375,-0.203125,-0.453125,0.16503906,-1.5078125,0.2734375,0.83984375,0.23828125,0.5546875,0.39648438,-0.30273438,-2.21875,0.796875,-0.9140625,0.35546875,-0.8671875,-1.015625,-0.24023438,-0.6171875,0.28320312,-0.047607422,-0.12011719,0.87109375,0.6875,1.5390625,1.4140625,0.08886719,-1.28125,0.54296875,-0.765625,0.54296875,-0.40625,0.119628906,0.48632812,0.099609375,-0.62890625,2.421875,-0.609375,0.018310547,-0.14453125,-0.032226562,-0.88671875,0.092285156,1.0,-0.8359375,0.8828125,0.76171875,-0.61328125,-0.41210938,1.0859375,-0.34570312,1.390625,-0.24707031,-0.875,0.8359375,-0.114746094,-1.1171875,0.66796875,-0.8359375,-1.3984375,-1.21875,-1.4140625,0.6796875,3.140625,-0.8125,-1.2421875,-1.1796875,-0.6796875,0.3046875,-0.15136719,0.9140625,-0.546875,-0.47070312,0.87890625,-1.25,1.765625,-0.3046875,-0.5546875,-0.359375,0.015197754,-0.64453125,-0.0015182495,-0.82421875,0.45898438,-0.43945312,-1.0234375,1.28125,-0.88671875,-1.65625,1.59375,0.52734375,-0.9296875,-0.78125,0.37304688,-0.984375,-1.734375,1.4921875,0.69140625,0.33398438,-0.43359375,-1.96875,2.5,-1.375,-1.0078125,-0.064453125,-0.1796875,-0.26367188,0.328125,-1.59375,-1.1953125,0.2421875,-0.06982422,1.59375,1.4140625,-2.109375,0.10546875,-0.18457031,0.41210938,-1.03125,1.6875,-0.578125,0.42382812,-0.27539062,0.90234375,-0.71484375,-0.01361084,2.265625,-0.9453125,-0.110839844,1.59375,-0.25585938,-1.625,-1.1015625,-2.078125,-1.5546875,-0.734375,0.484375,-0.03149414,0.0038757324,-1.328125,-1.265625,-0.13671875,-0.48632812,0.7421875,-0.029785156,0.73828125,-1.0546875,-0.45703125,0.15429688,-1.1015625,1.0859375,1.734375,-0.21386719,0.69140625,-0.46484375,-0.94921875,0.75,1.90625,1.6015625,0.65625,1.4296875
-3,0,2,0.57421875,0.41992188,-0.9765625,-0.28125,-1.546875,-1.1796875,0.51171875,1.3671875,-0.12890625,-0.9140625,1.6796875,1.40625,0.7578125,0.19824219,0.4296875,0.94921875,-0.8359375,-0.28125,1.0625,-0.7265625,0.796875,-0.2890625,2.296875,1.9921875,-0.28125,0.5546875,-0.17382812,-1.1640625,-0.578125,-0.32226562,0.35546875,0.64453125,1.0703125,2.125,-0.0005455017,-0.4765625,-0.20605469,0.5,0.60546875,-1.2734375,-1.0703125,-1.84375,1.0,-0.4765625,0.66796875,0.20019531,0.3515625,-0.19335938,-2.375,-0.12988281,-0.34960938,0.14355469,-1.65625,0.95703125,1.6328125,-0.11035156,-0.53515625,-0.022827148,-0.53125,0.029052734,0.24609375,0.26171875,-1.40625,-1.1796875,1.140625,1.53125,-1.0703125,-0.23925781,1.0234375,0.08984375,-1.4375,0.89453125,-0.890625,0.5,0.9375,0.65625,1.0703125,0.609375,-0.671875,0.39257812,-0.77734375,0.54296875,1.7578125,0.029663086,1.0390625,0.059326172,-0.39257812,1.2578125,0.06640625,0.82421875,1.2265625,-0.90625,-1.3046875,0.52734375,-0.42578125,-0.7578125,1.5625,-0.31640625,-0.51171875,-0.51171875,-0.91796875,-1.015625,2.421875,-0.26171875,-1.953125,-1.6875,0.71484375,-1.515625,0.859375,0.640625,-0.26171875,-1.609375,0.92578125,-1.2421875,0.2734375,-0.53515625,0.028930664,0.42382812,0.3515625,-1.359375,1.359375,0.47265625,-0.3125,-2.140625,0.45117188,2.03125,-0.73828125,-0.119628906,1.796875,-1.59375,0.44921875,1.25,1.203125,-2.609375,-2.28125,-0.28515625,-1.0078125,0.5625,-0.028442383,-1.4765625,0.89453125,-0.1015625,-0.29101562,0.29296875,-0.40234375,0.54296875,-0.9921875,-0.30078125,-0.80078125,0.64453125,-0.43945312,0.29882812,-0.6953125,-0.15136719,-0.2890625,-2.109375,-0.7734375,-0.55859375,0.53515625,1.46875,-0.80859375,-1.375,0.075683594,-0.84765625,0.8984375,-0.041503906,-2.0,-1.5234375,2.046875,-0.60546875,-0.31054688,0.54296875,-1.5234375,-0.47265625,-0.22558594,-0.09326172,1.546875,0.22070312,-0.3515625,0.11376953,0.47851562,1.765625,0.48046875,-0.25585938,0.63671875,-0.70703125,-0.37695312,0.31445312,0.52734375,1.5390625,0.828125,0.6015625,-0.21679688,-0.5078125,0.4140625,-1.546875,0.765625,0.13378906,1.8046875,0.20996094
-4,0,0,2.046875,0.01586914,-1.2265625,1.3046875,1.1015625,-1.015625,-0.7109375,1.5546875,-0.13476562,-0.9296875,1.03125,1.0625,1.0546875,0.23828125,-1.7578125,0.84765625,-0.37304688,-0.08251953,0.37695312,0.010253906,0.88671875,0.49609375,0.6328125,0.36523438,0.40820312,0.08544922,-1.7734375,0.33007812,-1.6640625,0.45898438,0.16015625,0.79296875,1.34375,1.7578125,0.7109375,0.41015625,0.24414062,-1.2421875,0.73046875,0.9140625,-0.80078125,0.20703125,-1.0546875,-0.73046875,0.40429688,-0.106933594,1.296875,-0.20605469,-0.81640625,-0.78125,-0.4453125,-1.1953125,-2.203125,-1.546875,-1.6796875,-0.8203125,0.28125,0.29492188,0.38085938,-0.6953125,0.099121094,0.23046875,0.046142578,0.75390625,-0.21777344,-1.2578125,-0.05053711,0.20996094,0.13378906,0.921875,1.1953125,-2.828125,0.03955078,-0.037841797,0.6171875,2.03125,0.40039062,-0.65234375,0.14941406,-1.9375,-0.48046875,0.29882812,0.114746094,-1.6328125,1.125,0.64453125,0.25195312,-0.36914062,1.2734375,-0.07128906,1.953125,0.9296875,0.00340271,0.4765625,0.50390625,-1.546875,-0.18945312,-0.6328125,-0.67578125,-0.06640625,-0.8984375,1.3203125,0.9453125,-0.29296875,-0.62890625,-1.859375,-1.25,-0.1796875,-0.119140625,1.390625,-0.37695312,0.51171875,0.38867188,-0.3203125,1.0,1.296875,-0.77734375,-0.8671875,0.07861328,-0.26757812,-0.12597656,-0.72265625,-1.0859375,0.83984375,-1.3671875,1.515625,-0.90625,-1.09375,0.14941406,0.29101562,0.609375,-0.7890625,-0.40039062,-0.6171875,-0.83984375,0.703125,0.13183594,-1.171875,0.484375,-2.28125,1.9765625,-1.984375,-1.1015625,0.4453125,0.76953125,0.30859375,-1.328125,0.66796875,1.421875,2.46875,-0.028198242,0.84765625,0.45898438,0.29101562,1.5234375,1.0859375,-0.953125,0.26953125,0.78515625,-0.97265625,-0.18945312,1.5,1.0078125,1.15625,-0.67578125,0.88671875,-0.26367188,-0.15332031,1.1171875,-1.125,-1.2109375,-1.3828125,0.100097656,-1.828125,-1.6328125,-0.828125,0.35351562,-1.8515625,0.5390625,-0.09033203,-0.4921875,-0.8828125,0.52734375,-2.0,1.5,0.39648438,0.58984375,-0.50390625,-0.052246094,-2.09375,1.125,-0.057861328,1.15625,-0.47070312,-0.16796875,2.65625,0.609375,0.014770508,0.296875,1.0234375
-4,0,1,0.38671875,1.015625,-0.84375,-1.046875,1.4921875,0.36328125,-1.2421875,1.2734375,0.828125,-0.13476562,0.16992188,0.70703125,-0.64453125,0.14550781,-0.390625,-0.62890625,-1.2109375,0.21484375,1.8671875,1.171875,0.6640625,-0.39257812,0.94921875,0.6796875,-0.84765625,-0.78515625,-2.0625,-1.0625,-2.046875,-0.030639648,-0.390625,0.075683594,0.22460938,1.046875,-0.5703125,0.096191406,0.3828125,-0.23925781,0.54296875,0.8671875,-1.6640625,0.47460938,-0.041503906,0.08251953,0.26953125,0.38085938,0.78125,-0.87109375,0.70703125,0.65234375,0.98046875,-1.28125,0.10498047,0.28320312,-2.0,-0.6484375,-0.34179688,-1.0078125,0.625,-1.390625,0.2890625,0.76171875,2.0625,-0.72265625,0.47460938,-1.0625,-0.048583984,-1.265625,-0.171875,1.484375,0.50390625,1.390625,0.048339844,-1.0390625,-0.04321289,0.8359375,-0.37890625,0.30273438,-0.71875,0.064941406,0.515625,-0.921875,0.46289062,-0.32617188,-0.37890625,-0.26367188,-0.5078125,-1.34375,1.5078125,0.38671875,1.3125,1.7421875,-0.99609375,1.4453125,-0.53515625,-0.7421875,-0.41601562,-0.023925781,-0.9765625,-0.09423828,-0.62890625,1.2265625,1.6640625,-1.6015625,-1.1015625,0.5859375,-0.9296875,0.111328125,0.31445312,0.01928711,0.3515625,0.3984375,0.10888672,0.38671875,0.99609375,-0.52734375,-0.5703125,-0.31835938,0.671875,0.07519531,0.25195312,-0.22265625,-1.0,0.22558594,-1.1640625,2.09375,-0.18554688,-2.1875,1.359375,2.84375,0.6171875,-1.6171875,-2.03125,0.49023438,-1.6875,-0.6953125,-1.1640625,-1.46875,0.40625,-1.6875,2.140625,-2.296875,-0.9921875,1.8671875,0.796875,-0.7578125,0.78125,0.22460938,-0.36914062,1.75,0.375,-0.22363281,1.171875,-0.38085938,0.052978516,0.8359375,1.6875,0.35351562,0.7578125,-2.03125,0.875,1.1875,0.97265625,2.203125,-1.03125,1.0390625,-1.6171875,-1.4140625,2.03125,-0.81640625,0.390625,-1.515625,0.11816406,-0.18847656,-0.71875,-1.046875,-0.49023438,-1.375,0.23046875,0.004486084,-0.82421875,-0.609375,0.8359375,0.60546875,-0.35546875,1.21875,-0.064453125,-0.56640625,1.0703125,-0.9765625,-1.109375,0.036865234,0.46289062,-0.5859375,-1.03125,1.9453125,0.7890625,0.66796875,-0.421875,0.66015625
-4,0,2,1.2578125,-0.80078125,-0.62890625,0.20019531,-0.296875,-0.51953125,-0.765625,0.88671875,-0.68359375,-0.62890625,0.32421875,1.1953125,1.0546875,0.52734375,-0.6640625,1.5078125,-0.3125,-0.45117188,2.265625,-0.46289062,1.6953125,0.14453125,-0.20410156,0.04711914,2.1875,-0.53515625,-0.984375,-1.3125,-0.546875,0.49023438,-0.3046875,0.5390625,1.0390625,-0.012756348,-1.984375,-0.13378906,0.5390625,-1.640625,-2.015625,-0.6953125,-1.5234375,-0.859375,-1.0625,-0.24023438,-0.10058594,0.91796875,0.19628906,-0.68359375,-1.2578125,-0.10205078,-0.44335938,-1.0078125,-1.5859375,-0.59375,-0.50390625,1.9609375,0.49023438,0.018676758,0.23535156,-0.71875,1.3203125,-0.578125,0.52734375,1.53125,0.58203125,0.18652344,-0.48828125,-0.45898438,2.265625,-0.023071289,0.18457031,0.032470703,-0.38085938,0.28320312,1.0859375,2.375,-0.25390625,0.083984375,-1.6484375,-0.34375,0.88671875,-0.5234375,-0.45703125,0.111816406,0.99609375,-0.15722656,0.31640625,1.5859375,0.5625,1.6015625,1.5859375,-0.40429688,0.28515625,0.97265625,-1.7265625,1.5546875,-0.91015625,-0.028198242,0.375,-0.21972656,-0.83984375,0.19140625,-0.22167969,0.28710938,-0.91796875,-0.29296875,-0.4921875,-2.6875,-0.64453125,0.32617188,0.484375,-0.953125,0.6484375,-0.67578125,-0.61328125,-0.099609375,-0.79296875,0.96484375,-0.24609375,0.21582031,0.77734375,1.171875,-0.25390625,-1.703125,-1.2109375,0.8984375,-1.421875,-0.9453125,0.025268555,0.23046875,0.9765625,-0.30078125,-0.29296875,-1.296875,-0.61328125,-0.20507812,1.8671875,-0.62109375,-1.6875,-2.640625,2.171875,0.640625,-1.0703125,-0.12597656,-0.83984375,-0.20703125,0.6796875,-0.87109375,0.29882812,1.9140625,-0.36328125,-0.111328125,1.3203125,-2.09375,-0.30664062,-1.5859375,1.625,1.2734375,1.3984375,-0.83984375,-0.48242188,-0.80859375,0.36914062,-0.28320312,-1.1328125,0.1875,0.099121094,-1.0859375,1.828125,-0.390625,0.30859375,-0.28125,-1.2890625,0.45703125,-0.515625,1.6015625,0.2109375,-0.123046875,0.96875,0.09033203,-1.3359375,0.22558594,-0.020263672,0.5,0.018676758,-0.5546875,1.0,-0.44335938,1.5546875,-0.89453125,-0.16308594,-0.5390625,1.25,-0.671875,-0.8203125,0.48632812,0.7109375,2.78125,2.171875,0.546875
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
deleted file mode 100644
index 25531275..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-4-embeddings.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-5,0,0,0.58984375,0.99609375,-0.34179688,-1.34375,-0.48828125,-0.9140625,-0.88671875,0.46484375,0.94921875,-1.2421875,-0.19433594,-0.5859375,-1.09375,1.5390625,-1.671875,0.66796875,-0.33789062,0.453125,-0.8125,0.87890625,1.6640625,0.703125,0.063964844,0.34765625,0.40625,-0.58984375,-1.8828125,-0.15234375,-1.0,1.2265625,-0.9296875,-0.2265625,0.84375,0.2109375,-0.22949219,0.90625,0.3984375,-0.060791016,-0.58984375,0.62109375,-1.7890625,0.38085938,-0.12158203,0.41210938,-0.0056152344,1.3359375,0.69921875,0.13671875,-0.40234375,1.2578125,-0.41796875,-0.6796875,-1.5234375,-0.018310547,-0.041503906,1.8125,0.014404297,-0.23632812,0.40429688,-1.546875,0.29296875,0.39648438,1.2890625,0.24316406,0.96875,1.1796875,-1.578125,1.46875,0.012939453,-0.22949219,-0.80078125,-1.203125,-0.76171875,-0.07324219,-0.5859375,0.015197754,0.49609375,0.5546875,0.30664062,0.6484375,1.078125,-1.046875,1.0078125,0.42773438,-0.53125,0.49609375,0.3515625,0.640625,-0.58984375,-0.50390625,3.59375,0.41992188,0.796875,1.2734375,-0.51953125,1.0390625,-0.98828125,0.31835938,-0.58203125,0.34570312,-0.828125,0.17089844,1.3828125,-1.140625,0.045410156,0.17773438,-0.390625,-1.8671875,-0.4453125,0.83203125,-0.15625,-0.40625,0.4296875,-0.80859375,-0.625,-0.609375,-0.34960938,0.6953125,0.4296875,-0.12792969,0.734375,-1.0703125,-0.62890625,-1.8515625,-0.828125,1.65625,-0.35742188,-1.625,-1.1015625,-0.96875,1.3203125,0.2578125,-2.375,-0.5859375,-1.5703125,2.015625,0.14648438,-0.453125,-1.34375,-2.5625,1.1015625,-2.21875,0.5859375,-0.092285156,0.6484375,0.35742188,-0.20410156,-0.76953125,-0.087890625,1.5625,-1.5859375,0.59765625,0.60546875,-0.09814453,1.296875,-0.06542969,0.41015625,0.9375,0.7890625,0.41601562,-0.87890625,-0.66015625,-0.37109375,0.09082031,-2.6875,2.390625,0.22851562,-0.014282227,1.4140625,-0.9140625,-1.8984375,-0.86328125,-1.1171875,-0.9296875,-0.45117188,1.5078125,1.1875,-1.0,-0.18261719,-0.91015625,0.44921875,1.3125,0.50390625,-0.041015625,1.0078125,-1.109375,-0.13085938,1.6015625,0.578125,-1.3125,-0.23730469,1.359375,1.21875,0.38085938,-0.546875,0.43554688,3.265625,-0.09814453,-0.18554688,0.45117188
-5,0,1,-0.44140625,-0.69921875,-0.38476562,-1.4375,0.17578125,-1.203125,-0.9609375,1.640625,0.057373047,0.8203125,-0.8671875,-0.53125,1.3984375,0.875,-2.15625,0.93359375,-0.51953125,-0.85546875,0.1015625,0.5078125,2.46875,-0.18164062,-0.8515625,0.2421875,0.48046875,-0.38671875,-2.078125,-1.125,-0.42578125,1.75,-0.26757812,0.50390625,-0.98046875,0.5546875,-1.25,0.49804688,-0.37695312,-0.49414062,-0.27734375,-0.6015625,-0.9453125,0.33203125,-0.6953125,-0.28710938,0.53515625,2.484375,-0.106933594,-0.3125,1.390625,1.0703125,-1.0078125,-2.421875,-2.34375,0.43554688,-0.609375,1.6640625,-0.28125,-0.80859375,-0.26757812,-1.5546875,-0.87890625,-0.80859375,1.578125,0.96484375,1.4375,2.015625,-1.671875,1.6171875,-0.953125,0.6171875,1.296875,-1.2890625,-0.6953125,-0.265625,-0.234375,0.25195312,0.265625,-0.33984375,0.7265625,0.734375,1.578125,-0.42382812,0.27929688,1.734375,-0.84375,0.64453125,-0.8203125,0.049316406,0.13476562,-0.111816406,1.4765625,-0.36523438,1.78125,-0.09765625,-0.15332031,0.025878906,-1.84375,0.78125,-0.5390625,0.37695312,-0.2890625,-0.27148438,0.36328125,0.17675781,0.36523438,1.3125,-1.671875,-2.609375,-1.0703125,-0.61328125,-0.19921875,0.78125,-1.0546875,-0.53125,-0.099121094,-2.359375,-0.09667969,0.81640625,0.33007812,1.5859375,0.11376953,-0.3828125,-0.15917969,-1.6484375,-1.2578125,0.6328125,-1.359375,-0.77734375,-0.78515625,0.48046875,2.078125,-0.7578125,-1.71875,-0.546875,-0.28515625,0.8359375,-0.13867188,0.078125,-0.09765625,-2.046875,-0.32226562,1.609375,0.58984375,0.95703125,0.52734375,0.16015625,-0.51953125,0.20507812,-0.59375,1.0234375,0.034179688,-0.90625,0.54296875,-0.52734375,0.43554688,-0.17382812,1.4140625,1.0078125,0.55078125,0.35546875,-0.19433594,0.875,0.41601562,0.5,-1.7890625,1.734375,0.73828125,0.20800781,0.46289062,0.03955078,-0.49804688,0.63671875,-0.036865234,0.18261719,-1.1171875,1.984375,0.18164062,-1.671875,-0.57421875,-0.09863281,-0.6640625,0.875,0.91796875,0.3828125,0.28320312,-0.28710938,0.28710938,0.39453125,0.3203125,-0.79296875,-1.453125,0.7578125,1.5078125,0.14746094,-0.546875,0.2734375,2.375,-0.3984375,1.0078125,1.125
-5,0,2,0.91015625,0.5625,0.828125,0.203125,-1.140625,-0.080078125,-1.1953125,-0.03112793,2.609375,-1.46875,1.8359375,-0.040283203,0.35351562,1.3671875,-0.62890625,-0.42578125,-0.46289062,0.18359375,2.265625,1.1484375,0.2421875,-0.5078125,-0.15820312,0.58203125,-0.484375,-0.53515625,0.6484375,-0.52734375,0.90625,0.5078125,-0.40625,-0.008361816,0.78125,-0.66796875,0.37695312,0.90625,0.94921875,-0.19042969,-0.20898438,1.3671875,-2.828125,-0.94140625,-1.578125,0.8203125,-0.24804688,-0.23339844,0.5078125,-0.29296875,0.87890625,0.68359375,-0.15917969,-0.26367188,-0.74609375,0.95703125,0.9296875,0.40820312,-0.65234375,-0.49804688,-0.578125,0.77734375,1.1875,-0.3828125,1.2578125,0.09765625,0.64453125,0.515625,-0.20703125,-1.2578125,-1.453125,0.41210938,-0.110839844,1.3828125,0.78125,-0.63671875,-0.65234375,0.47851562,-0.52734375,0.48046875,-1.09375,1.2578125,-1.8984375,-0.71484375,0.92578125,-1.1640625,-0.703125,0.51171875,-0.36523438,-1.5078125,0.47851562,0.28710938,1.5390625,1.1953125,-0.47070312,0.875,-0.453125,0.1640625,-1.78125,0.73046875,-0.640625,-0.36523438,-1.6953125,0.609375,0.9140625,-1.296875,-0.34570312,2.171875,-0.63671875,-0.7890625,-0.375,0.78125,-0.47460938,0.26171875,1.203125,-0.578125,0.265625,0.34960938,-0.625,0.453125,-0.41015625,0.5546875,0.6875,-0.0032958984,0.4453125,-1.6796875,-1.90625,0.5625,-0.16699219,-0.1953125,-0.34570312,1.140625,1.078125,0.61328125,-1.5078125,-0.265625,-1.1015625,1.578125,0.16601562,-0.73046875,-0.41601562,-1.75,1.671875,-0.26953125,-0.62890625,1.5,-0.28320312,-0.075683594,1.4609375,-0.73046875,-1.1796875,0.22070312,-1.078125,-0.16601562,2.140625,-0.578125,-0.60546875,0.19335938,1.40625,-0.39648438,0.65625,0.390625,-0.75,-0.30273438,1.390625,1.2421875,-1.171875,-0.14550781,-0.6640625,-0.625,0.6484375,-1.703125,-2.109375,-0.25585938,-3.0,-0.40234375,-0.578125,2.5,0.80078125,-1.046875,-0.47460938,0.36328125,-0.33398438,-0.09277344,-0.068847656,-0.6640625,-1.1640625,0.20605469,1.125,0.58984375,-0.38867188,-0.8203125,-1.5234375,0.296875,0.546875,-0.86328125,-2.796875,1.2578125,3.171875,0.20019531,0.34570312,0.859375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
deleted file mode 100644
index dc74307e..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-5-embeddings.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-6,0,1,0.13867188,0.10498047,0.2734375,-0.88671875,1.3984375,0.6171875,0.49023438,1.0234375,-0.4921875,-1.0390625,0.12792969,-0.26757812,0.58203125,0.35351562,-1.25,0.48828125,-0.546875,-1.90625,1.1328125,-0.5234375,-0.0859375,-0.6640625,1.28125,0.53125,-1.234375,-1.5078125,-0.95703125,-0.86328125,-1.7890625,0.44335938,-0.63671875,-0.26367188,0.011413574,-0.46484375,-0.52734375,0.953125,0.9140625,-0.7421875,-1.4609375,1.578125,-1.7890625,0.029663086,-0.075683594,0.09765625,0.38476562,0.52734375,0.63671875,-0.703125,0.37304688,1.8671875,-1.1953125,-1.6015625,-0.96875,1.109375,-2.421875,0.09277344,-0.21289062,-0.8046875,1.328125,-1.71875,1.75,0.18945312,0.29101562,0.29882812,-0.51171875,-0.064453125,-1.078125,0.16601562,0.546875,1.8359375,0.060791016,-0.31640625,0.57421875,-0.13769531,-0.671875,1.8671875,0.765625,1.1796875,-1.78125,-0.5234375,1.2890625,0.890625,-0.12011719,0.12109375,-0.63671875,-0.20214844,-0.46679688,1.4609375,-0.35742188,0.9921875,1.3828125,0.21289062,-0.921875,0.36914062,-1.421875,0.22460938,-0.59375,-0.640625,0.62109375,-0.890625,-2.375,-0.17675781,2.515625,0.77734375,-0.79296875,0.0059814453,0.06689453,-1.0703125,0.48632812,0.984375,0.703125,0.03515625,0.5703125,-0.796875,0.49023438,-0.31835938,-0.47460938,-0.35351562,-0.48632812,-1.1875,0.7109375,-0.7890625,0.6015625,-1.5703125,-0.22363281,1.2578125,-1.3203125,-2.421875,0.016235352,0.17382812,-0.18554688,0.044921875,-0.022460938,-1.3671875,-1.03125,-0.13769531,-0.65625,0.25390625,0.20605469,-1.734375,3.375,-0.40625,1.3125,1.1015625,-1.2265625,1.0703125,-0.6015625,-0.6640625,-2.515625,1.1015625,-0.31054688,-0.032958984,1.0625,-0.48828125,-1.09375,-1.1953125,0.38085938,1.3515625,1.015625,1.421875,-0.26171875,-0.012084961,1.828125,1.2421875,-1.78125,1.65625,-1.671875,-0.24511719,2.109375,-0.20117188,0.3671875,-0.40234375,0.09765625,0.6171875,0.6875,0.0390625,-0.8671875,-0.84375,0.86328125,-0.609375,-0.3046875,-0.88671875,0.13964844,0.85546875,0.14746094,0.5390625,0.59765625,0.040527344,0.86328125,-1.828125,0.095703125,1.375,1.2578125,0.28710938,0.4375,-0.18261719,1.6796875,0.15722656,1.4140625,0.51953125
-6,0,2,0.82421875,0.82421875,-0.3984375,0.84375,1.390625,0.62890625,-0.35546875,0.39257812,-0.10058594,-1.4375,1.109375,-0.19238281,0.53515625,-0.54296875,-1.0,-0.37109375,-0.81640625,-0.40039062,1.109375,1.2421875,0.20507812,-1.8984375,0.36132812,0.359375,-0.90234375,-1.609375,-1.3828125,-0.35742188,0.22753906,0.05883789,-0.23632812,-0.32617188,1.71875,0.20800781,-0.328125,-0.16894531,0.31640625,-0.6015625,-0.26367188,0.78125,-0.60546875,0.71484375,-0.26953125,1.2734375,-0.27539062,0.25195312,-0.734375,-1.0234375,0.43945312,1.71875,1.140625,-1.765625,-0.4921875,-0.26367188,-1.0859375,-2.125,0.38476562,1.3046875,-0.109375,-1.015625,1.6171875,1.59375,1.734375,-0.34960938,-1.0390625,-0.7109375,0.05029297,-2.078125,-0.30273438,0.85546875,0.3828125,0.16992188,1.34375,-1.4921875,0.4609375,0.515625,0.18457031,-0.20605469,-1.7109375,0.39648438,-0.27539062,0.71875,-0.018310547,0.31835938,-0.76171875,0.7421875,-2.828125,0.453125,1.5859375,1.203125,1.421875,1.1953125,1.125,0.22753906,-0.6328125,-0.9375,-0.875,1.2421875,-0.22949219,0.123535156,-0.54296875,-0.39453125,0.89453125,-0.11621094,-1.0390625,0.77734375,-0.9921875,-0.18164062,-0.34960938,-0.33203125,0.18261719,1.96875,0.13378906,-0.6328125,-0.57421875,0.671875,-0.41601562,0.33203125,-0.46484375,0.83984375,0.05102539,-1.1015625,-1.15625,-0.59375,-1.328125,1.78125,-0.70703125,-0.59765625,-0.2421875,0.76171875,-0.04321289,-2.515625,-0.9375,0.7421875,-1.3203125,2.546875,-0.31835938,0.01574707,1.8125,-1.3125,1.421875,-0.75390625,0.119628906,1.171875,0.6640625,-0.20214844,-0.22460938,-0.27929688,-0.56640625,0.20019531,0.43164062,-0.32421875,-1.34375,1.015625,-1.1328125,1.234375,0.6953125,-0.36523438,-0.38867188,1.046875,0.78515625,1.15625,1.0078125,3.046875,-2.671875,1.609375,-2.234375,1.1875,0.16699219,-0.29492188,-1.7734375,-1.828125,0.734375,-0.65234375,0.65625,0.27539062,-0.66796875,-1.984375,1.40625,-0.35742188,0.10107422,0.15136719,0.84375,-0.115722656,-1.0234375,1.578125,-0.48828125,-0.26953125,0.5078125,-0.36523438,0.66015625,1.546875,-0.31640625,-1.0390625,-0.82421875,-0.47070312,-0.46875,-0.20019531,0.19726562,0.51953125
-6,0,3,-0.91015625,0.041748047,-0.34375,1.078125,1.359375,-0.030395508,-0.43945312,1.0546875,0.59765625,-0.6640625,0.27539062,1.84375,-0.29296875,-0.6640625,-0.97265625,0.23046875,0.0031280518,-1.046875,1.9765625,-0.4765625,-0.12402344,-2.0625,-0.8203125,0.54296875,-0.12060547,-0.62890625,0.7109375,0.73828125,0.69140625,0.41015625,0.4375,0.2734375,1.3203125,-0.52734375,2.140625,0.88671875,1.1796875,-0.828125,-1.015625,0.6640625,-0.40429688,-0.80859375,-1.859375,-1.875,-0.82421875,0.8984375,-0.40039062,-0.94140625,0.8515625,1.7578125,1.125,-0.5703125,-1.15625,0.31445312,-0.123046875,-2.03125,0.35351562,1.515625,0.43359375,0.15722656,0.78125,0.828125,-0.3515625,1.28125,-1.3125,0.55078125,-1.3359375,-1.90625,0.47460938,-0.0546875,0.8203125,1.4453125,1.7734375,-1.7734375,0.81640625,0.921875,-0.08251953,-0.69140625,-0.2890625,0.9296875,-1.0390625,0.578125,-0.29101562,-0.9140625,0.041992188,-1.3671875,-0.43554688,-0.19140625,1.5703125,0.859375,1.4453125,-0.13769531,-0.20117188,1.0859375,-0.31835938,-0.27539062,-0.82421875,-0.9609375,-1.03125,-0.21972656,0.083984375,0.40625,1.3984375,1.1796875,-1.0703125,1.359375,0.6171875,0.19921875,-0.049804688,0.51953125,0.75390625,0.83984375,2.5,-1.59375,0.76953125,-0.35351562,-1.53125,0.671875,-0.84765625,-0.7890625,1.3125,0.46289062,-0.671875,-0.6484375,-1.296875,2.03125,-0.06982422,0.76953125,0.8125,-1.2734375,1.4765625,-2.3125,0.15332031,0.19628906,-0.10644531,0.8984375,0.7265625,-0.12988281,0.41015625,-0.99609375,1.5625,0.265625,-1.453125,1.3828125,-0.36523438,-1.0234375,0.49414062,-0.31640625,-0.18554688,-1.2578125,0.78515625,0.22363281,2.671875,-0.92578125,-0.8203125,0.040527344,0.13964844,-1.578125,1.1875,0.119140625,0.63671875,0.59765625,1.125,0.48046875,0.052734375,0.7265625,0.024047852,0.984375,-0.33203125,-0.004211426,-1.1875,-1.3984375,-2.359375,-1.28125,-0.4140625,0.9921875,-0.5,-1.5234375,-1.6328125,-0.953125,0.703125,0.15820312,0.25585938,-1.234375,-0.828125,-1.21875,-0.25,0.18945312,-1.2578125,1.078125,-1.21875,-0.030395508,-0.87109375,-1.1953125,-1.6171875,0.578125,-0.1875,0.21777344,0.20019531,1.609375
-7,0,0,0.86328125,1.2109375,-0.83203125,-0.984375,-1.2890625,0.32617188,-0.49804688,1.515625,-0.7890625,-0.70703125,1.625,0.0032348633,-0.060791016,-0.28125,-1.15625,1.6328125,-0.2890625,-1.5234375,0.15039062,-0.6640625,1.34375,0.84765625,0.47851562,1.125,0.8984375,-0.859375,-0.93359375,-1.9375,-1.6171875,-0.63671875,-1.2578125,2.453125,1.0078125,2.515625,-0.16796875,0.14746094,-0.40234375,-0.59765625,0.8671875,0.5625,-0.625,0.28125,0.63671875,0.20019531,1.3046875,0.36328125,0.578125,0.26953125,0.65625,-1.1796875,-0.84765625,-0.51171875,-1.890625,-1.609375,-1.03125,0.59765625,0.08544922,-0.33984375,0.3125,-0.74609375,-0.19042969,-0.515625,1.8515625,1.0,1.1796875,0.70703125,-0.88671875,-0.13574219,0.22558594,0.25195312,-0.036376953,-0.36132812,-0.671875,-0.18847656,-0.4765625,-0.15820312,-0.234375,-0.13183594,-0.014892578,0.45507812,-0.48242188,-0.546875,1.1328125,0.42382812,1.109375,1.2890625,-0.97265625,0.52734375,1.375,-1.0234375,1.3359375,0.41015625,0.546875,1.2109375,0.21972656,-1.3203125,0.765625,0.99609375,-1.703125,-0.296875,-0.6484375,0.33007812,2.21875,-1.7734375,-1.2265625,-1.4765625,-2.015625,-1.7109375,-0.47851562,0.93359375,0.20996094,-1.0078125,0.088378906,0.5703125,0.53125,0.20117188,0.0043640137,-0.98046875,-0.27148438,0.22851562,-0.3671875,-0.31054688,-1.4375,-1.0,-1.4609375,1.0625,-1.75,-0.55078125,1.0390625,0.984375,0.9140625,-0.484375,-1.1328125,-0.6484375,-1.7734375,0.82421875,1.1171875,-0.18359375,-1.6640625,-2.4375,1.484375,-0.5546875,-0.640625,0.59375,0.4765625,0.18652344,-0.24804688,-0.5234375,0.72265625,1.5390625,-0.75390625,0.2734375,1.2734375,-1.21875,0.6796875,0.3125,0.59375,-0.859375,0.76953125,-1.140625,0.80078125,-0.40234375,0.64453125,-0.64453125,-1.140625,3.453125,-0.76171875,-1.1796875,0.328125,-1.015625,-1.0,-0.47070312,-1.2890625,-1.1171875,-0.38476562,1.1796875,0.7890625,-0.72265625,1.046875,-0.49804688,0.39453125,0.3515625,0.73046875,-1.2890625,1.1484375,-0.053710938,0.09082031,0.28515625,0.24609375,0.37890625,2.03125,-0.14257812,-0.033203125,-0.7265625,-0.009887695,0.8359375,2.0625,1.796875,0.8515625,0.66796875
-7,0,1,0.5859375,1.3984375,-0.61328125,-0.51953125,-1.9140625,-0.7578125,1.0234375,0.9765625,0.18945312,0.010009766,0.115234375,0.43945312,-0.021118164,-0.828125,-0.51171875,1.796875,-1.2578125,-0.82421875,0.34960938,0.28125,0.45703125,-0.6015625,1.046875,0.3671875,1.2890625,-0.515625,-1.984375,-0.72265625,0.5,-0.62890625,-0.73828125,-0.009216309,2.21875,-0.43554688,-0.625,-0.076660156,0.26171875,-1.28125,-0.51953125,-0.68359375,-0.4296875,0.18847656,0.29882812,-0.103027344,-0.73046875,0.15332031,0.37890625,0.65625,-0.28125,-0.20117188,0.89453125,0.546875,-3.75,-0.15527344,1.15625,1.71875,-0.14941406,0.34375,0.47070312,-0.095703125,-1.0546875,-1.1171875,1.0390625,0.33007812,1.59375,-0.375,-0.234375,-0.34765625,0.040771484,0.35742188,-0.32226562,1.125,-0.045410156,-0.984375,2.453125,0.83203125,0.87890625,-0.13183594,-1.4375,0.34179688,0.59765625,-0.28125,1.53125,0.125,0.37304688,0.0048828125,-1.203125,1.1328125,1.6015625,-0.24316406,0.8828125,-0.3515625,1.0,0.52734375,-1.265625,0.57421875,-0.83984375,1.828125,-1.3984375,0.35742188,0.33789062,0.97265625,-0.5625,0.71875,-0.6171875,-0.20898438,-1.6796875,-1.75,-0.29101562,0.33398438,-1.46875,-2.296875,0.30859375,-0.20117188,-1.4140625,1.15625,-0.66796875,1.03125,-1.140625,-0.546875,-0.12988281,0.41210938,-1.203125,-1.3359375,0.49023438,0.76171875,-1.625,-0.10107422,-0.18554688,-1.125,1.140625,-0.23339844,-0.11035156,-2.171875,-0.765625,-0.6640625,1.6484375,0.048339844,-1.4296875,-2.796875,1.4453125,0.14941406,-1.390625,-1.21875,-0.33007812,-0.44921875,0.40039062,0.77734375,2.25,1.9765625,-0.5390625,-1.0703125,0.17675781,-0.31835938,-0.23925781,-1.2734375,1.234375,0.084472656,0.9921875,0.096191406,-0.2734375,-0.07128906,0.22753906,-0.40429688,-1.1484375,0.625,-0.515625,-1.375,0.578125,0.46679688,0.84765625,0.609375,-0.8046875,0.8046875,-1.6015625,1.1328125,2.75,-0.35351562,1.6796875,1.8984375,-0.25195312,0.65234375,0.62109375,-1.0234375,0.043945312,-1.046875,0.4921875,-0.087402344,0.83984375,-0.043701172,0.18652344,0.96875,-0.011230469,0.31445312,-1.09375,-0.20605469,0.013671875,2.0,1.0703125,0.027832031
-7,0,2,0.9609375,0.67578125,-1.0078125,-0.953125,0.055419922,-0.69921875,-0.1484375,1.9921875,0.8203125,0.62109375,-1.328125,1.8359375,-0.36328125,1.15625,0.020263672,0.7734375,0.62109375,-0.35742188,0.875,1.421875,0.234375,-0.05126953,-0.75390625,-0.328125,-0.6171875,1.375,-0.64453125,0.026977539,0.8984375,-2.203125,0.5625,1.7890625,1.1796875,0.15917969,-0.59765625,-0.41601562,0.22753906,1.4609375,0.08642578,-0.104003906,-0.984375,-0.07714844,-0.52734375,0.49414062,0.265625,0.05053711,-0.23925781,-0.33203125,0.06738281,0.17675781,0.09814453,-1.390625,-0.359375,-0.7421875,0.06982422,0.50390625,-1.6484375,0.15429688,0.28320312,-0.63671875,-1.1484375,-0.69140625,1.6171875,-0.34179688,0.453125,0.55859375,0.25390625,-0.17871094,-0.296875,1.0390625,0.22070312,2.09375,-0.86328125,-0.7421875,0.8671875,1.5859375,0.171875,-0.8125,-2.859375,0.7109375,1.6171875,0.578125,1.84375,2.0625,-0.66796875,0.8046875,-0.6796875,-0.65234375,-0.5625,1.375,1.1484375,1.0,-0.16992188,1.2890625,0.021362305,0.11230469,-1.6328125,1.03125,-0.13183594,0.453125,0.66796875,-0.50390625,-0.48828125,-0.34375,-0.59765625,0.94140625,-0.34375,-0.90625,-1.7265625,-0.87109375,-0.17871094,-0.47070312,1.5859375,-0.83203125,-0.15136719,-1.0859375,0.21484375,0.19140625,0.8984375,0.80859375,1.2265625,0.06738281,-1.078125,-0.78125,-0.46875,0.73046875,-0.18164062,-0.23535156,1.21875,-0.54296875,0.8984375,-1.859375,0.29296875,-1.1640625,-1.359375,0.7734375,-1.5703125,0.16601562,0.22167969,-2.0625,0.0703125,-0.13085938,-1.359375,1.2421875,-0.24121094,0.06738281,-0.53515625,1.71875,0.21972656,-1.8828125,1.5703125,-2.09375,0.34765625,0.20410156,0.14941406,0.46679688,1.4140625,0.68359375,-0.54296875,0.23535156,-1.578125,2.25,-0.83203125,0.796875,1.15625,-0.78515625,-0.484375,-2.859375,0.23925781,1.25,-0.55859375,0.7421875,-1.1171875,1.3203125,0.092285156,1.703125,0.22558594,-1.4921875,-0.765625,0.82421875,-0.609375,0.51953125,-0.41992188,0.47070312,-1.7109375,0.29882812,-1.6953125,-2.046875,1.484375,-0.97265625,-1.5703125,-0.92578125,-1.5390625,0.484375,0.072265625,0.06689453,0.23925781,-0.38085938,-0.85546875,1.265625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
deleted file mode 100644
index 24fc7499..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-6-embeddings.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-8,0,1,1.0703125,0.24707031,-0.1640625,-0.29101562,0.11035156,-0.42382812,-1.28125,0.7890625,0.23632812,0.46875,0.13476562,0.91796875,1.2421875,-0.13476562,-0.56640625,-0.13183594,-1.078125,-2.234375,0.93359375,0.21582031,1.78125,-1.03125,-1.2890625,-0.20996094,1.40625,-0.30273438,-0.71875,-0.009643555,0.2578125,-0.14453125,-0.84375,-0.6171875,1.9453125,-0.953125,-0.9296875,-0.41796875,0.09423828,-1.140625,-2.578125,-0.46484375,0.44726562,-0.87890625,-0.70703125,-0.29882812,-1.2265625,0.5234375,-0.31445312,0.50390625,1.7578125,1.4375,-0.27148438,0.359375,-1.65625,1.2578125,-0.103027344,0.3828125,-0.5,1.8828125,0.7109375,-0.82421875,-0.6953125,-1.9453125,0.76953125,0.8046875,1.3359375,1.4765625,-1.4609375,-1.8671875,0.41992188,0.9375,1.421875,1.21875,-0.41992188,-0.94921875,2.90625,1.78125,-0.35546875,-0.11279297,-0.49414062,0.2734375,-0.34375,0.26757812,0.0005950928,2.359375,-0.48046875,-0.58984375,-0.41601562,2.5,1.0625,0.92578125,0.8125,0.5390625,0.81640625,0.06640625,-0.84375,-0.13378906,0.55078125,1.0390625,0.51171875,-0.94140625,0.1953125,1.0390625,-0.47070312,0.36914062,-0.16601562,0.7734375,-1.171875,-1.140625,0.37890625,0.33007812,-1.0234375,-0.15527344,-0.5625,-1.9453125,-1.78125,0.10498047,-0.6328125,1.21875,-1.265625,0.4453125,0.61328125,0.41796875,-2.21875,-1.3671875,-0.8046875,0.18652344,-0.5625,0.44140625,0.6640625,-0.6328125,1.7734375,-0.25,0.14453125,-1.5,-0.65625,0.56640625,0.55078125,0.19335938,-0.74609375,-2.5,1.078125,2.578125,-0.71484375,-1.21875,-1.265625,1.2265625,0.32421875,-0.15625,0.32421875,0.60546875,0.5625,0.98046875,0.6328125,-0.62109375,-0.78125,-1.453125,0.29101562,-0.05419922,1.171875,-0.65625,0.35742188,0.57421875,0.078125,-1.0625,0.5859375,-0.28320312,0.578125,-0.74609375,-1.28125,1.265625,0.14160156,0.65234375,-0.8984375,-0.53125,-1.703125,1.84375,-0.17773438,-0.578125,0.07519531,-0.48242188,1.25,0.40625,0.81640625,-0.68359375,-1.0703125,-1.5625,0.18847656,0.36328125,0.69140625,0.62890625,-0.30078125,0.36132812,0.796875,-0.9375,-0.640625,-1.84375,-0.6953125,1.2578125,1.0234375,0.7265625
-8,0,2,0.875,0.9921875,-0.1796875,0.12597656,0.080566406,-0.609375,0.765625,1.0703125,1.0703125,-0.80078125,1.328125,1.546875,0.05419922,-1.03125,-0.14648438,0.296875,-0.6796875,-1.0078125,0.83984375,0.33398438,0.026733398,-0.36523438,-0.24511719,1.625,-0.18164062,0.14550781,-2.78125,0.21484375,0.3125,0.7890625,-0.5078125,-0.8984375,1.3671875,-0.77734375,0.578125,0.59375,0.53515625,-0.9375,-0.22070312,1.078125,-1.203125,1.1171875,-0.33984375,-2.40625,-0.8671875,-0.78125,0.2578125,-0.23046875,1.296875,1.609375,0.5625,-0.47460938,-0.92578125,0.67578125,0.47265625,0.265625,0.19921875,-0.515625,-1.0625,-0.035888672,1.4765625,0.2109375,0.6328125,0.86328125,0.50390625,-0.4296875,-0.140625,0.5625,-1.359375,0.67578125,-0.10058594,0.54296875,0.84765625,-0.42773438,1.8359375,0.8046875,1.3828125,1.0546875,-0.35351562,0.37109375,-0.08251953,-1.1328125,0.8359375,-1.4453125,-0.453125,0.36523438,-0.03149414,-0.032470703,1.7421875,0.19042969,1.3515625,-0.5390625,-1.03125,1.2421875,-1.1015625,-0.59375,-2.734375,0.49414062,-0.6015625,-0.37109375,-1.4609375,1.1640625,0.38085938,0.19433594,-0.59765625,1.8359375,-0.75390625,0.16699219,-1.15625,0.30664062,-0.56640625,-0.42773438,0.91796875,-0.26171875,0.36328125,1.765625,-0.5078125,0.5546875,-0.15917969,-0.44140625,0.57421875,-0.625,-0.49023438,-0.9375,-1.328125,2.109375,-1.265625,-1.046875,0.71875,-1.2890625,1.3203125,-1.21875,-0.6875,-0.76953125,-1.6015625,0.65234375,1.09375,-0.5234375,0.05810547,-2.46875,1.703125,-0.94921875,-1.765625,1.375,0.71484375,-0.04248047,0.8671875,-0.89453125,-1.1640625,0.9140625,-1.2890625,-0.703125,0.68359375,-1.015625,1.1171875,0.35546875,1.6796875,-0.8671875,1.1640625,-0.20800781,-0.0703125,1.6484375,2.453125,1.453125,-1.5234375,0.09863281,-0.5390625,0.084472656,0.859375,-0.15722656,-1.015625,-0.29492188,-1.640625,-0.1484375,-1.5390625,1.0390625,0.9140625,-1.1640625,0.29492188,0.050048828,-1.2734375,-1.40625,0.54296875,-1.0,-0.45507812,0.24804688,0.021484375,0.56640625,-1.6171875,-1.6328125,-0.23339844,2.46875,-0.3671875,-0.97265625,-1.5625,1.03125,1.4921875,1.09375,-0.41992188,-0.59375
-8,0,3,1.2578125,1.5546875,0.1015625,-0.25585938,-1.5859375,-0.018310547,-1.328125,1.4375,1.4140625,-2.65625,1.7578125,-0.6875,0.03100586,1.1171875,-0.4921875,0.46484375,0.2890625,0.40625,1.859375,-0.014160156,-0.8046875,-0.036132812,-0.390625,-0.25,-0.86328125,-1.0625,-0.71484375,-0.22070312,0.100097656,-0.20605469,-0.25,0.018432617,1.0546875,0.71484375,0.09716797,0.29101562,1.40625,0.1484375,-0.88671875,1.6953125,-2.3125,-0.22363281,0.114746094,2.140625,0.25976562,0.91796875,-0.87109375,0.7109375,-0.061767578,0.87890625,-0.64453125,-0.09033203,-0.3515625,-0.25,-0.41210938,-0.45507812,0.19433594,0.80859375,-1.1328125,0.16601562,0.546875,0.65625,1.65625,-0.40820312,1.3671875,0.43164062,-0.36328125,-1.265625,-0.98046875,0.07763672,-0.27148438,-0.23730469,0.6171875,0.037841797,-0.91015625,0.734375,0.14160156,0.34179688,-1.53125,1.3203125,-0.81640625,-0.66015625,0.953125,-1.2265625,-0.1796875,0.33398438,0.24902344,-1.3046875,0.1015625,0.51171875,1.9453125,1.5859375,-0.35742188,1.0546875,-0.89453125,-0.14648438,-0.68359375,0.004760742,0.19335938,-0.703125,-2.390625,0.87890625,1.7265625,-1.7421875,0.03515625,1.546875,-1.109375,-0.46289062,0.0546875,0.20410156,-0.091308594,-0.3125,1.3359375,-0.73046875,0.063964844,0.080566406,-0.38085938,0.24902344,-0.25390625,0.859375,1.6484375,-0.56640625,-0.15429688,-1.796875,-0.515625,0.20117188,0.39257812,-1.5546875,-0.11816406,1.390625,-0.890625,1.109375,-1.3984375,0.6484375,-2.078125,1.515625,0.90625,-0.6015625,-0.7109375,-2.328125,1.5859375,-1.171875,-0.76171875,0.033691406,-0.19140625,0.203125,0.48046875,-0.59375,-0.7578125,1.234375,-1.3984375,0.29492188,0.58203125,-0.119628906,-0.31835938,0.09667969,1.1953125,0.34570312,0.52734375,0.037353516,-0.67578125,0.16894531,0.4375,0.12695312,-1.265625,0.83984375,-0.6015625,0.6796875,0.671875,-2.015625,-2.109375,-1.3359375,-2.171875,-0.49023438,0.34765625,2.296875,0.3671875,-0.5234375,0.037597656,-0.4453125,0.64453125,-0.27929688,-0.2578125,0.20898438,-0.38476562,0.14257812,0.91796875,1.4375,-0.4765625,0.045410156,0.080566406,0.25976562,0.78125,-1.4140625,-3.125,1.2578125,3.359375,0.6484375,-0.44921875,0.49609375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
deleted file mode 100644
index c18d2d33..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-embeddings/sequifier-model-categorical-3-inf-size-best-embedding-3-7-embeddings.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199
-9,0,0,0.7890625,1.0,-0.060302734,0.65234375,0.36132812,-0.49804688,0.3359375,1.0859375,-0.15917969,-1.0625,0.72265625,-0.25,-1.125,-0.859375,-1.0546875,0.1484375,-1.0859375,1.375,-0.43554688,0.29882812,-0.024902344,0.79296875,-0.33398438,-0.13867188,-0.609375,-1.4765625,-2.765625,0.3125,-0.63671875,0.51171875,-1.046875,-1.1171875,-0.38476562,0.21972656,-0.63671875,-0.8984375,0.515625,-1.2578125,0.63671875,-0.103515625,-0.7578125,0.44140625,0.40039062,1.59375,0.29882812,0.69921875,0.05444336,0.296875,-2.15625,0.8125,1.28125,-1.2890625,-2.296875,-0.14355469,-0.92578125,-0.44726562,0.07861328,1.0,-0.2734375,-0.8125,-0.07714844,1.265625,-0.072753906,-0.17089844,1.3125,-0.5859375,0.33789062,-0.0016403198,0.22070312,0.94140625,0.18261719,-2.015625,0.90234375,-1.1171875,-0.1328125,1.265625,1.765625,-0.13085938,0.23046875,-0.3671875,0.90234375,0.47265625,-0.17480469,-1.09375,-0.46289062,1.0078125,0.072265625,-1.3203125,0.14453125,0.609375,0.62109375,1.6796875,0.88671875,0.59765625,-0.27734375,-0.20898438,-1.09375,0.80078125,0.12109375,0.60546875,1.6796875,0.67578125,-0.90625,-0.2109375,0.50390625,-0.35351562,-0.94921875,-0.52734375,0.42382812,-0.1640625,-1.59375,2.015625,0.07763672,0.5,-0.36328125,0.16796875,-0.4765625,1.5703125,-0.30078125,1.4453125,1.4609375,-1.296875,-0.77734375,-1.03125,0.02331543,0.20605469,0.33789062,-1.0546875,-2.203125,0.18164062,0.22167969,-0.76953125,-0.85546875,0.21289062,0.2421875,1.0,-0.44335938,-1.1015625,1.2421875,-1.90625,1.3515625,-0.5390625,-0.44140625,-0.118652344,1.1796875,0.33789062,-0.97265625,-0.119140625,2.203125,3.46875,-0.27734375,-0.75,-2.375,2.21875,-0.48828125,0.2421875,0.203125,1.2578125,-1.4921875,0.44335938,-0.95703125,1.0625,0.66015625,1.75,-1.9296875,0.23535156,-1.265625,0.828125,1.2109375,-1.1484375,-0.06640625,-1.0703125,1.8359375,-0.80078125,-0.47851562,0.0023651123,0.03930664,-0.7265625,1.21875,0.17675781,-0.5546875,0.31054688,1.546875,0.30664062,-0.068359375,0.44335938,1.296875,0.50390625,0.94921875,-2.96875,-0.125,-0.1015625,0.0075683594,-1.2421875,-1.3984375,1.6640625,1.234375,-0.5390625,-0.88671875,-0.5078125
-9,0,1,-0.16308594,1.359375,-0.18359375,-0.94140625,0.55859375,0.32421875,0.57421875,-0.81640625,1.578125,-1.4609375,2.109375,-1.296875,-0.46289062,-0.18066406,-0.43164062,-0.48242188,-1.0546875,0.03125,0.9296875,0.640625,0.30664062,-0.984375,1.2578125,0.07373047,-1.609375,-2.046875,-0.65234375,-0.01586914,-0.09667969,1.328125,0.62890625,-0.27539062,0.703125,0.83984375,0.115722656,-0.17578125,0.59375,0.64453125,-0.3046875,-0.001373291,-0.45117188,0.5703125,1.0546875,0.90625,-0.7265625,1.0,0.07128906,-1.6484375,-0.70703125,2.25,0.578125,-1.1171875,-0.609375,0.84765625,0.34765625,-0.51171875,-0.008117676,1.09375,0.546875,-0.171875,0.96875,1.6875,0.20800781,0.5,1.0,0.24511719,-0.8515625,-0.8984375,0.24511719,0.9765625,-1.1015625,0.5,0.390625,-0.5390625,-0.53515625,-0.34179688,-0.22949219,-0.55078125,-0.6328125,1.296875,1.109375,0.091796875,-0.037109375,2.1875,0.28320312,0.020263672,-2.21875,-0.55078125,0.671875,-0.13183594,2.203125,0.043945312,1.3203125,-0.72265625,-0.11621094,-0.04711914,-1.2109375,0.21582031,-0.81640625,-1.5390625,-0.41210938,-0.625,1.4921875,-0.45117188,-0.54296875,1.1015625,-2.03125,0.33203125,-0.21191406,-0.34179688,-0.4921875,2.109375,-0.43359375,-2.59375,-0.58203125,0.65625,-0.98046875,0.85546875,-0.24902344,0.1328125,0.55859375,-1.1796875,0.024902344,-2.21875,-1.2734375,2.28125,-1.46875,-0.70703125,-0.41796875,-0.37890625,1.0859375,0.07519531,-1.0546875,-0.040039062,-1.078125,2.34375,-0.5390625,0.62890625,0.72265625,-0.89453125,1.4296875,0.22265625,0.17089844,2.203125,0.0625,-1.078125,-0.052490234,-0.75390625,-1.265625,0.6484375,-0.5,-0.6484375,-0.68359375,0.05444336,-0.87109375,-0.05078125,0.20410156,-0.27148438,-1.234375,2.296875,-0.69921875,-0.64453125,1.5,1.140625,-2.4375,1.921875,-0.78515625,2.015625,-0.028442383,-0.14746094,-1.2421875,-0.026123047,-0.48242188,-1.1171875,0.6015625,0.33203125,-0.11328125,-2.6875,-0.49414062,-0.94140625,1.2890625,0.57421875,0.5078125,0.8046875,-1.0078125,1.7734375,-1.2734375,-0.25585938,-0.72265625,0.125,0.48828125,0.66796875,0.36523438,0.41796875,-0.12695312,-1.71875,0.2109375,-0.03466797,1.0078125,0.7265625
-9,0,2,-0.31445312,0.34179688,0.5546875,-2.0625,-0.025634766,0.671875,1.8046875,-0.69921875,1.0390625,-0.98046875,1.125,-1.3671875,1.3671875,-0.6484375,-0.54296875,0.69140625,-0.043701172,-0.62890625,2.109375,-0.734375,-0.78125,-1.4140625,1.25,-0.15429688,-2.46875,-1.6171875,0.18066406,-0.79296875,-0.515625,1.65625,0.53125,-0.546875,-0.12207031,0.099121094,1.109375,0.7734375,-0.20898438,-0.14160156,-1.8046875,0.9609375,-1.5390625,0.51953125,0.95703125,0.96875,0.59765625,1.3671875,-0.98828125,-1.1484375,-0.2265625,1.609375,-1.78125,0.25,-1.0,1.890625,-0.25195312,0.62109375,0.20996094,0.28125,-0.047851562,-0.34375,0.60546875,-0.045654297,-0.703125,0.52734375,0.6328125,1.421875,-0.78125,-1.1015625,-0.98828125,2.109375,-1.109375,-0.57421875,0.080078125,0.859375,-0.4140625,0.76953125,1.5703125,0.1953125,-0.21386719,0.22753906,1.125,0.45898438,-0.16113281,1.2578125,0.16894531,-0.82421875,-0.65625,-1.28125,-0.57421875,-0.32421875,0.8671875,0.037109375,-0.83203125,-1.0078125,-1.5390625,0.484375,-0.5703125,-0.60546875,-0.25195312,-0.625,-1.375,0.06225586,2.3125,1.1796875,0.17871094,1.78125,-0.9296875,-0.10107422,0.53125,-0.79296875,0.40039062,1.015625,0.90234375,-2.59375,0.53515625,-0.34375,0.28515625,0.19140625,-0.76953125,0.53515625,1.75,-0.3828125,1.234375,-2.0625,0.83203125,0.15625,-2.140625,-1.6171875,-0.19238281,0.24902344,0.71875,0.68359375,0.32617188,-1.515625,0.578125,0.40234375,-0.25585938,0.6015625,-0.32617188,-0.62890625,2.546875,1.671875,-0.25585938,1.7109375,-0.87109375,-0.35546875,0.1875,-0.15332031,-2.421875,0.6640625,-0.54296875,-0.13867188,-0.41210938,-1.2578125,-1.234375,-0.93359375,0.3359375,-0.07519531,-0.05419922,2.09375,-2.0625,-0.671875,1.171875,-0.014465332,-1.6875,-0.10888672,-0.17285156,1.34375,0.024536133,0.46679688,0.3515625,0.49804688,-1.375,-0.29492188,0.03564453,-0.15332031,0.765625,-1.2890625,-0.25390625,0.109375,-0.26953125,-1.140625,1.0390625,0.79296875,-0.63671875,-0.17382812,0.3671875,0.02331543,-1.1484375,-0.17089844,-0.5390625,1.203125,1.2578125,0.13671875,-0.03173828,-0.45703125,1.34375,-0.69140625,2.0625,0.5859375
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
deleted file mode 100644
index 6b50fe9d..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-0-embeddings.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-0,0,0,-0.6006612,0.27331486,-0.34914157,-0.41479835,-0.3665798,-0.5533585,-0.14582494,-0.91373837,0.9844944,-0.95687056,-0.4577148,0.91134745,1.9406763,-0.3906231,2.3219848,-1.3473451
-0,0,1,0.82811767,-0.5268697,2.0814955,-0.004021469,0.5512644,-1.179125,-0.043747347,0.63015777,-1.3691996,1.6033845,-0.75416285,-0.42233095,0.45330113,0.27154657,-0.41820994,-1.7461473
-0,0,2,-0.84707606,-2.0609052,-0.1081782,-1.045911,-0.42114437,0.60394675,0.59907866,0.25174326,1.3924737,0.7319598,-0.9297085,1.5825528,-1.3019253,0.7942725,0.87738246,-0.1327626
-0,0,3,1.0684123,0.57142717,0.075573,0.11559761,-0.86974573,0.1249011,-0.2048351,0.22256361,0.48389286,-1.8543849,-0.93811846,-1.1520655,0.44447225,2.7120762,-0.38763985,-0.48254395
-0,0,4,-0.17874089,-1.1804047,0.6036014,-1.2056729,-0.21152717,-0.11572556,0.7904921,0.64713424,1.58864,0.2770208,-0.6086349,-0.9361127,-1.0290424,2.055118,-1.3742863,0.8467823
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
deleted file mode 100644
index 19f8ea9b..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-1-embeddings.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-1,0,0,0.1883566,-0.51069915,1.0757822,0.8057133,-1.7141013,-1.0322051,0.8189624,2.455926,-0.086683646,-0.548176,0.2890321,-0.5181182,-0.046274558,0.71348274,-1.3307245,-0.6185085
-1,0,1,1.5297848,-1.2117101,-0.13439633,1.253659,-0.67018974,-0.5551331,0.7888967,1.1364132,-0.023180284,-1.6521875,1.607315,0.5899527,-0.32935065,-0.41081616,-1.3123547,-0.64372915
-1,0,2,0.8035976,-0.1564827,0.7187234,0.23841827,-1.041676,0.36406025,-0.5592999,0.6205595,0.22290729,-1.892453,-0.571009,-0.9368423,0.4768633,2.6517987,-0.04318322,-0.9589635
-1,0,3,1.6627725,-0.8503898,0.30536455,0.9283896,0.027027875,-0.6029172,-1.3817337,-1.0517935,0.88064855,-0.30291262,0.46670607,1.2897946,-1.2118512,-0.20403868,-1.3723617,1.4220868
-1,0,4,0.3392744,-0.80427676,2.273018,-0.6253882,0.75351644,0.86384004,0.31538412,-0.83312535,1.189355,-1.0529702,-1.1474402,0.37553895,-1.7151668,-0.13089456,0.64540154,-0.4849727
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
deleted file mode 100644
index 0eb7cf80..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-2-embeddings.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-2,0,0,-0.45247898,-0.6700307,-1.3980862,-0.8169967,1.1174649,-1.5679284,1.3004793,0.55547404,0.17673689,2.2888486,0.61724645,-0.30405724,0.7114118,-0.48643318,-0.40007603,-0.69985884
-2,0,1,0.96580744,-0.7271615,1.9331442,0.35922375,-1.1587405,-0.7792273,-0.84466994,-0.14708672,-1.1819658,1.8822826,0.33108777,-0.21132755,-0.08135607,0.38287959,-1.4923096,0.7527195
-2,0,2,0.3168712,0.017159618,-0.608047,0.8562339,1.3965137,-1.2637002,-0.35587037,0.14391245,0.03870829,-0.47761393,0.3009319,-2.3473392,1.9502603,0.8832216,0.0150891915,-0.9161799
-2,0,3,0.9275735,-1.1297508,1.0120924,-0.20396979,0.64359254,0.12881663,-1.0047234,-0.9111546,0.6870843,-0.58317107,0.87117267,1.0960609,-1.4761081,-0.9214843,-1.0393838,1.9111625
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
deleted file mode 100644
index 05af7ab7..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-3-embeddings.csv
+++ /dev/null
@@ -1,9 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-3,0,0,-0.5061456,-0.7763785,-1.4411104,-0.52552813,1.0513479,-1.8644568,1.2233515,0.5684989,0.3443002,2.1877117,0.73235464,-0.2621535,0.5448747,-0.17720893,-0.4192689,-0.7051907
-3,0,1,-0.66679925,-1.3665861,-0.22060487,-0.44057253,-1.2946736,0.76740754,0.73320085,0.6129007,0.31359667,-0.85040903,0.46061102,2.6040866,-0.12284056,0.15048675,0.7167322,-1.4336061
-3,0,2,1.1156372,-0.22328217,-0.16104864,-1.1401838,-1.9943426,-0.2494216,1.7525011,-0.9386309,0.7549316,-1.3333093,0.040007867,0.77046984,1.3616405,0.5641874,0.22289011,-0.625249
-3,0,3,0.56514716,0.021596653,0.6965276,-0.2659085,-0.47595215,0.41413322,0.15071203,0.4543448,0.3236465,-2.2226622,-0.6223092,-1.3502904,0.022095501,2.6971529,-0.082981095,-0.391887
-4,0,0,1.2965182,-0.5597377,0.87963766,-1.3444073,-0.53000945,0.48690176,-0.06573936,0.76831144,0.30042756,0.42209244,-2.325122,-1.1396146,-0.7086506,1.4786966,0.15237255,0.84046805
-4,0,1,0.81161726,-0.8216032,2.040686,0.17894322,0.19341934,-0.7916967,-0.28501594,0.5449751,-1.6567922,1.2058699,-0.6466604,0.02430841,0.7775131,0.4694338,-0.07763458,-2.0113945
-4,0,2,0.89558095,-0.9144592,-1.5348337,-0.88432336,0.38815454,0.113910615,0.2328737,1.2174358,0.79367036,0.35744292,-0.73602664,-1.4213881,-0.011158632,0.26761565,-1.0480565,2.2604616
-4,0,3,0.1480164,-0.8613011,0.67656535,-0.3105859,0.05321045,-0.9875798,-0.100072786,-0.68129635,1.649064,1.3399781,2.217513,0.4515288,-0.27188015,-1.0483067,-1.1040651,-1.1851294
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
deleted file mode 100644
index 28a36481..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-4-embeddings.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-5,0,0,0.14931662,-0.53661007,-0.4178915,1.0650523,1.4260073,-1.4046125,-0.4454908,0.43748403,0.17220889,-0.3263942,0.46434885,-2.1899922,1.6828609,0.9274285,0.095608875,-1.1405207
-5,0,1,0.9416289,0.41291103,-0.4414328,0.27544495,-0.43491212,0.074826315,-1.4218235,0.46812102,-0.17113322,0.63309056,1.3964156,-0.91853875,0.35744792,1.9670777,-1.7801918,-1.3875116
-5,0,2,1.1062784,0.10729322,-0.4857815,-0.008219642,-1.3734647,0.07514935,-0.89442974,0.041828547,0.040950842,0.41346258,1.0223055,1.4269317,0.4183222,-1.0225612,1.4259869,-2.336927
-5,0,3,1.1709383,-0.3431763,0.86983496,-1.6041002,-0.6716347,0.9073796,0.5058333,0.52458864,0.07157109,-0.31755292,-2.013162,-1.166504,-0.6992124,1.7159256,0.378617,0.6107121
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
deleted file mode 100644
index 964edf93..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-5-embeddings.csv
+++ /dev/null
@@ -1,10 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-6,0,0,0.75977767,-0.8941403,1.8381127,-0.2947546,0.2300058,-0.83663684,0.25620785,0.12637803,-1.7378011,1.182791,-0.50745416,0.036903758,0.838191,0.390844,0.6659436,-2.1043644
-6,0,1,1.2344869,-0.87115544,1.6973119,0.45785195,-0.79289585,-1.1952786,-0.7739234,-0.41190162,-0.6169048,2.0888212,0.17353193,-0.35224319,-0.42837173,0.29140207,-1.3883605,0.87579113
-6,0,2,-0.04716062,-1.1540091,0.7190903,0.21688245,-1.0973245,0.6634729,-1.0191848,1.0636467,0.86636645,0.95576775,0.40887043,-2.1949317,0.34050336,-0.5335777,1.6200428,-0.8434403
-6,0,3,0.5315266,-0.75266963,1.6877258,0.7681073,-0.9721382,-1.0216631,0.68959737,1.0582479,0.28320122,-0.731182,-0.9680603,-0.9397336,0.115028106,-0.47067833,-1.2997086,1.9737223
-6,0,4,1.4608288,-0.97320133,0.2406221,-0.8215309,0.9401043,-0.24250522,0.4038392,0.020989973,-0.51842433,-1.293551,-0.90764123,0.74466497,-0.6153507,2.456806,-1.1631211,0.23413561
-7,0,0,-0.16067488,-1.4088781,-0.52943593,-0.53733695,-0.881949,1.020826,0.9473788,0.22936635,0.48022568,-0.3948946,0.26959687,2.6491475,-0.31377748,-0.2948756,0.564145,-1.6736668
-7,0,1,0.7797228,-1.711448,-0.27516288,1.2757864,0.5518668,1.0981587,-2.2466977,0.15571526,-0.89171106,0.9792887,0.612105,-1.1947432,0.38149735,0.55929846,0.13911384,-0.20466185
-7,0,2,1.4645267,0.6052429,-0.6147724,1.2570165,-1.6225443,-0.0040931324,-0.46623838,0.37951505,0.81095076,1.1644993,0.37510827,-2.0000474,-0.28104275,0.7649079,-0.69417703,-1.187673
-7,0,3,1.3395354,-0.7207316,2.0857847,0.21226466,-1.0327022,-0.8066197,-0.42594796,-0.64940375,-1.1074684,1.6851654,0.36267513,-0.4739317,-0.4367329,0.49596614,-1.2793999,0.7298483
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
deleted file mode 100644
index a62eae9c..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-6-embeddings.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-8,0,0,-0.13603266,-0.7307862,-0.52509326,0.74821085,1.3972374,-1.0351514,-0.087998375,0.14008258,-0.16702926,-0.7757859,0.767171,-2.0736523,1.9220414,0.7747719,0.8359873,-1.0989997
-8,0,1,0.2691684,-0.58000475,0.22857264,0.0726761,0.36003938,-0.88419944,-0.48701763,-0.6307923,1.4697176,1.9778647,1.8036274,0.28785348,-0.23820582,-1.2125394,-1.566934,-0.87618756
-8,0,2,-0.5361712,-0.7612994,-1.4328666,-0.8012423,0.8185195,-1.415421,1.1227995,0.81984663,-0.20334002,2.2054589,0.822599,-0.029563703,1.0102043,-0.28044125,-0.27897397,-1.0894494
-8,0,3,1.0546969,0.12370178,-0.39167616,-0.19105366,-1.0928397,0.34877908,-0.921601,0.05444164,-0.1952575,0.56712765,1.1011527,1.3574673,0.31019345,-1.4148287,1.472835,-2.2225912
-8,0,4,1.0152535,-1.0647243,0.87573284,-0.06597459,0.7320688,0.2022493,-1.0524291,-0.9378761,0.6166235,-0.6034941,0.89612216,1.1415883,-1.5616627,-0.86831933,-1.1108083,1.7946959
diff --git a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv b/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
deleted file mode 100644
index 64e66679..00000000
--- a/tests/resources/target_outputs/embeddings/sequifier-model-categorical-bert-best-embedding-1-embeddings/sequifier-model-categorical-bert-best-embedding-1-7-embeddings.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-sequenceId,subsequenceId,itemPosition,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15
-9,0,0,-0.4572789,0.4083919,-0.31473643,-0.2729684,0.24014777,-0.4612613,-0.54261976,-1.0134654,1.0257438,-0.6418706,-0.78112227,0.8925522,1.8929882,-0.6186886,2.1505845,-1.5657275
-9,0,1,-0.1670796,-1.1557673,0.45290652,0.3810859,-0.8798305,0.950252,-1.0811738,0.9893472,0.62552804,0.80969876,0.35180464,-1.9200974,0.22418688,-0.624056,2.0414052,-1.0304397
-9,0,2,-0.20206067,-0.93332684,-1.3126967,-0.8317151,0.009184291,0.7905838,1.1029824,-0.8864819,1.4854951,0.84356177,-1.2966992,1.5094204,-1.4029254,0.9184137,0.5197256,-0.3381626
-9,0,3,0.54572934,-1.7311612,-0.3354268,0.58204854,1.634565,1.6117706,-1.2522428,-0.40521324,-1.0268775,0.31409565,0.66306615,-1.6557356,-0.32661834,0.54005146,0.6954313,0.15141636
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv
deleted file mode 100644
index b31efdb2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-0-predictions.csv
+++ /dev/null
@@ -1,21 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,113
-0,5,113
-0,6,113
-0,7,113
-0,8,113
-0,9,113
-0,10,113
-0,11,113
-0,12,113
-0,13,113
-0,14,113
-0,15,113
-0,16,113
-0,17,113
-0,18,113
-0,19,113
-0,20,113
-0,21,113
-0,22,113
-0,23,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
deleted file mode 100644
index 72488290..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-1-predictions.csv
+++ /dev/null
@@ -1,21 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,113
-1,5,113
-1,6,113
-1,7,113
-1,8,113
-1,9,113
-1,10,113
-1,11,113
-1,12,113
-1,13,113
-1,14,113
-1,15,113
-1,16,113
-1,17,113
-1,18,113
-1,19,113
-1,20,113
-1,21,113
-1,22,113
-1,23,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
deleted file mode 100644
index 9966bc14..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-2-predictions.csv
+++ /dev/null
@@ -1,21 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,103
-2,4,122
-2,5,122
-2,6,122
-2,7,122
-2,8,122
-2,9,122
-2,10,122
-2,11,122
-2,12,122
-2,13,122
-2,14,122
-2,15,122
-2,16,122
-2,17,122
-2,18,122
-2,19,122
-2,20,122
-2,21,122
-2,22,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
deleted file mode 100644
index bcce6828..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-3-predictions.csv
+++ /dev/null
@@ -1,41 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,103
-3,4,124
-3,5,124
-3,6,124
-3,7,124
-3,8,124
-3,9,124
-3,10,124
-3,11,124
-3,12,124
-3,13,124
-3,14,124
-3,15,124
-3,16,124
-3,17,124
-3,18,124
-3,19,124
-3,20,124
-3,21,124
-3,22,124
-4,3,122
-4,4,122
-4,5,122
-4,6,122
-4,7,122
-4,8,122
-4,9,122
-4,10,122
-4,11,122
-4,12,122
-4,13,122
-4,14,122
-4,15,122
-4,16,122
-4,17,122
-4,18,122
-4,19,122
-4,20,122
-4,21,122
-4,22,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
deleted file mode 100644
index 90a13a9d..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-4-predictions.csv
+++ /dev/null
@@ -1,21 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,103
-5,4,122
-5,5,122
-5,6,122
-5,7,122
-5,8,122
-5,9,122
-5,10,122
-5,11,122
-5,12,122
-5,13,122
-5,14,122
-5,15,122
-5,16,122
-5,17,122
-5,18,122
-5,19,122
-5,20,122
-5,21,122
-5,22,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
deleted file mode 100644
index f87b5d08..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-5-predictions.csv
+++ /dev/null
@@ -1,41 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,113
-6,5,113
-6,6,113
-6,7,113
-6,8,113
-6,9,113
-6,10,113
-6,11,113
-6,12,113
-6,13,113
-6,14,113
-6,15,113
-6,16,113
-6,17,113
-6,18,113
-6,19,113
-6,20,113
-6,21,113
-6,22,113
-6,23,113
-7,3,113
-7,4,113
-7,5,113
-7,6,113
-7,7,113
-7,8,113
-7,9,113
-7,10,113
-7,11,113
-7,12,113
-7,13,113
-7,14,113
-7,15,113
-7,16,113
-7,17,113
-7,18,113
-7,19,113
-7,20,113
-7,21,113
-7,22,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
deleted file mode 100644
index fcb6691b..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-6-predictions.csv
+++ /dev/null
@@ -1,21 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,122
-8,5,122
-8,6,122
-8,7,122
-8,8,122
-8,9,122
-8,10,113
-8,11,122
-8,12,122
-8,13,122
-8,14,122
-8,15,122
-8,16,122
-8,17,113
-8,18,113
-8,19,113
-8,20,113
-8,21,113
-8,22,113
-8,23,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
deleted file mode 100644
index 2b8cfd75..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-autoregression-predictions/sequifier-model-categorical-1-best-3-autoregression-7-predictions.csv
+++ /dev/null
@@ -1,21 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,113
-9,4,113
-9,5,113
-9,6,113
-9,7,113
-9,8,113
-9,9,113
-9,10,113
-9,11,113
-9,12,113
-9,13,113
-9,14,113
-9,15,113
-9,16,113
-9,17,113
-9,18,113
-9,19,113
-9,20,113
-9,21,113
-9,22,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv
deleted file mode 100644
index 783e9350..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
deleted file mode 100644
index 9eb44d36..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
deleted file mode 100644
index ac8d014f..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
deleted file mode 100644
index 19f45567..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,103
-4,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
deleted file mode 100644
index e7767b59..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
deleted file mode 100644
index e09b589d..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,113
-7,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
deleted file mode 100644
index d9b241c6..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
deleted file mode 100644
index d5752b57..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-best-3-predictions/sequifier-model-categorical-1-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv
deleted file mode 100644
index 461de2e0..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-0-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId
-0,2,119
-0,3,103
-0,4,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
deleted file mode 100644
index d32d6aab..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-1-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId
-1,2,122
-1,3,124
-1,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
deleted file mode 100644
index 118f3041..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-2-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId
-2,1,121
-2,2,102
-2,3,126
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
deleted file mode 100644
index d7dc85f6..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-3-predictions.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-sequenceId,itemPosition,itemId
-3,1,121
-3,2,109
-3,3,102
-4,1,127
-4,2,119
-4,3,120
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv
deleted file mode 100644
index a133efa3..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-4-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId
-5,1,126
-5,2,104
-5,3,112
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
deleted file mode 100644
index 0a48eeb8..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-5-predictions.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-sequenceId,itemPosition,itemId
-6,2,102
-6,3,118
-6,4,113
-7,1,109
-7,2,103
-7,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
deleted file mode 100644
index 12b5fd01..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-6-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId
-8,2,104
-8,3,121
-8,4,112
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
deleted file mode 100644
index de154d9e..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-1-inf-size-best-3-predictions/sequifier-model-categorical-1-inf-size-best-3-7-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId
-9,1,118
-9,2,118
-9,3,118
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv
deleted file mode 100644
index 783e9350..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
deleted file mode 100644
index 4ac41fe6..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,127
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv
deleted file mode 100644
index 5b9010cb..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
deleted file mode 100644
index 19f45567..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,103
-4,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
deleted file mode 100644
index e7767b59..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
deleted file mode 100644
index 6ed65b7f..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,122
-7,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
deleted file mode 100644
index c2bb5aec..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,103
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
deleted file mode 100644
index 40d6064e..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-best-3-predictions/sequifier-model-categorical-3-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
deleted file mode 100644
index 4f60920f..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-0-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-0,2,119,5,7
-0,3,103,4,5
-0,4,111,6,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
deleted file mode 100644
index 79b88c4f..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-1-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-1,2,122,0,1
-1,3,121,9,1
-1,4,108,9,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
deleted file mode 100644
index 98346781..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-2-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-2,1,121,6,1
-2,2,127,7,5
-2,3,112,4,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
deleted file mode 100644
index 248bef03..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-3-predictions.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-3,1,121,6,1
-3,2,109,3,1
-3,3,110,0,2
-4,1,121,6,1
-4,2,121,3,1
-4,3,102,0,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
deleted file mode 100644
index 2dbc2594..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-4-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-5,1,113,6,1
-5,2,113,4,0
-5,3,121,4,5
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
deleted file mode 100644
index 36d5bae9..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-5-predictions.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-6,2,102,4,5
-6,3,118,9,1
-6,4,121,4,6
-7,1,109,3,1
-7,2,121,0,2
-7,3,108,5,7
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
deleted file mode 100644
index 3e0792ba..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-6-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-8,2,102,8,8
-8,3,121,4,1
-8,4,121,3,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
deleted file mode 100644
index 20f73580..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-3-inf-size-best-3-predictions/sequifier-model-categorical-3-inf-size-best-3-7-predictions.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2
-9,1,108,9,1
-9,2,118,4,8
-9,3,103,4,8
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv
deleted file mode 100644
index 783e9350..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv
deleted file mode 100644
index 9eb44d36..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
deleted file mode 100644
index 53c89ed7..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
deleted file mode 100644
index 6b9ee33a..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,122
-4,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv
deleted file mode 100644
index 2b757dfc..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
deleted file mode 100644
index 579df4c3..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,113
-7,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv
deleted file mode 100644
index 3f1e127e..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv
deleted file mode 100644
index d5752b57..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-5-best-3-predictions/sequifier-model-categorical-5-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
deleted file mode 100644
index d5e918cd..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
deleted file mode 100644
index e87652fa..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
deleted file mode 100644
index 8ace2dd0..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
deleted file mode 100644
index 47d128ca..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,122
-4,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
deleted file mode 100644
index 081fa079..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,124
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
deleted file mode 100644
index 12c49755..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,113
-7,3,128
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
deleted file mode 100644
index 3f1e127e..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
deleted file mode 100644
index d302645a..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-50-best-3-predictions/sequifier-model-categorical-50-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
deleted file mode 100644
index c0c4c994..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-0-predictions.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-sequenceId,itemPosition,itemId
-0,0,116
-0,1,125
-0,2,116
-0,3,102
-0,4,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
deleted file mode 100644
index 6f05ec83..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-1-predictions.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-sequenceId,itemPosition,itemId
-1,0,100
-1,1,124
-1,2,102
-1,3,119
-1,4,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
deleted file mode 100644
index d7879885..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-2-predictions.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-sequenceId,itemPosition,itemId
-2,0,123
-2,1,125
-2,2,125
-2,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
deleted file mode 100644
index 2b1e4f45..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-3-predictions.csv
+++ /dev/null
@@ -1,9 +0,0 @@
-sequenceId,itemPosition,itemId
-3,0,123
-3,1,118
-3,2,119
-3,3,102
-4,0,122
-4,1,125
-4,2,112
-4,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
deleted file mode 100644
index df93b1d1..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-4-predictions.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-sequenceId,itemPosition,itemId
-5,0,125
-5,1,100
-5,2,119
-5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
deleted file mode 100644
index 0ba233fe..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-5-predictions.csv
+++ /dev/null
@@ -1,10 +0,0 @@
-sequenceId,itemPosition,itemId
-6,0,125
-6,1,103
-6,2,113
-6,3,120
-6,4,102
-7,0,118
-7,1,125
-7,2,129
-7,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
deleted file mode 100644
index 04a51b82..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-6-predictions.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-sequenceId,itemPosition,itemId
-8,0,125
-8,1,113
-8,2,123
-8,3,115
-8,4,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
deleted file mode 100644
index 7b4ec0b2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-bert-best-1-predictions/sequifier-model-categorical-bert-best-1-7-predictions.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-sequenceId,itemPosition,itemId
-9,0,116
-9,1,113
-9,2,114
-9,3,113
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
deleted file mode 100644
index 0502ed89..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
deleted file mode 100644
index 79a8ecae..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
deleted file mode 100644
index 20ab5369..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
deleted file mode 100644
index e4054674..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,115
-4,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
deleted file mode 100644
index f37176b4..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
deleted file mode 100644
index 25b28cdd..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,115
-7,3,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
deleted file mode 100644
index 8eb4c4a0..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
deleted file mode 100644
index d302645a..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-distributed-best-3-predictions/sequifier-model-categorical-distributed-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
deleted file mode 100644
index 0502ed89..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-0,4,100
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv
deleted file mode 100644
index 79a8ecae..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-1,4,115
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
deleted file mode 100644
index 53c89ed7..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-2,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
deleted file mode 100644
index e4054674..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-3,3,115
-4,3,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
deleted file mode 100644
index f37176b4..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-5,3,122
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
deleted file mode 100644
index 25b28cdd..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId
-6,4,115
-7,3,111
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
deleted file mode 100644
index 59349bc1..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-8,4,102
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
deleted file mode 100644
index d302645a..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-lazy-best-3-predictions/sequifier-model-categorical-lazy-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId
-9,3,121
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
deleted file mode 100644
index 10a61d90..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-0,4,109,[other],-0.007879399
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
deleted file mode 100644
index 6989e535..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-1,4,102,[unknown],-0.09629029
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
deleted file mode 100644
index f70f3d04..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-2,3,[mask],[unknown],-0.1264407
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
deleted file mode 100644
index 0b7fcd80..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-3,3,103,4,0.035191707
-4,3,110,2,0.025370859
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
deleted file mode 100644
index 4300c9d3..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-5,3,110,9,-0.07751563
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
deleted file mode 100644
index d82f9c83..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-6,4,122,7,-0.05601125
-7,3,113,[other],0.044996656
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
deleted file mode 100644
index 7ed5dffb..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-8,4,110,[mask],-0.10512776
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
deleted file mode 100644
index 418d29d1..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-best-3-predictions/sequifier-model-categorical-multitarget-5-best-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-9,3,114,6,-0.004389014
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
deleted file mode 100644
index c21215f2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-0-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-0,4,102,[other],0.006249614
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
deleted file mode 100644
index c249c329..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-1-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-1,4,100,[unknown],-0.09367853
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
deleted file mode 100644
index 57aca839..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-2-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-2,3,100,[unknown],-0.12620808
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
deleted file mode 100644
index fdadb1a4..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-3-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-3,3,111,4,0.047561854
-4,3,128,4,0.038341478
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
deleted file mode 100644
index 3ec68b25..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-4-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-5,3,100,6,-0.069281995
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
deleted file mode 100644
index c01739fe..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-5-predictions.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-6,4,110,7,-0.04092831
-7,3,102,4,0.06709013
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
deleted file mode 100644
index e00e6b7c..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-6-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-8,4,128,[mask],-0.09031482
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
deleted file mode 100644
index 7b52a58c..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-categorical-multitarget-5-last-3-predictions/sequifier-model-categorical-multitarget-5-last-3-7-predictions.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supReal3
-9,3,125,6,-0.0026362315
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
deleted file mode 100644
index 73c62198..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-autoregression-predictions.csv
+++ /dev/null
@@ -1,201 +0,0 @@
-sequenceId,itemPosition,itemValue
-0,21,-0.21923348
-0,22,0.26796454
-0,23,0.35781664
-0,24,0.32586923
-0,25,0.18709765
-0,26,0.16812888
-0,27,0.12320283
-0,28,-0.028172985
-0,29,0.014381966
-0,30,0.02785978
-0,31,-0.052008748
-0,32,-0.047266554
-0,33,-0.11041261
-0,34,0.052818693
-0,35,0.25398755
-0,36,-0.13537154
-0,37,-0.14335838
-0,38,0.0024016856
-0,39,-0.069979176
-0,40,-0.10791672
-1,19,0.35182652
-1,20,0.20307136
-1,21,0.3678002
-1,22,0.27595142
-1,23,0.0662965
-1,24,-0.17031401
-1,25,0.13717982
-1,26,0.14117326
-1,27,0.04607979
-1,28,-0.21623841
-1,29,0.12370201
-1,30,-0.033539154
-1,31,-0.09843233
-1,32,-0.044271484
-1,33,-0.10042905
-1,34,0.04008965
-1,35,0.18410258
-1,36,-0.11740111
-1,37,0.06030637
-1,38,-0.0854537
-2,17,0.1561486
-2,18,0.087262
-2,19,0.052818693
-2,20,0.011574087
-2,21,0.030106083
-2,22,0.29392183
-2,23,0.07827678
-2,24,0.030979645
-2,25,0.15115681
-2,26,0.08526528
-2,27,-0.1313781
-2,28,0.02761019
-2,29,-0.030668877
-2,30,0.12270366
-2,31,-0.05175916
-2,32,0.049574036
-2,33,-0.09593645
-2,34,-0.089447126
-2,35,-0.21124664
-2,36,-0.041026827
-3,17,0.087262
-3,18,0.32986265
-3,19,0.13418475
-3,20,-0.038530935
-3,21,0.1401749
-3,22,0.23002699
-3,23,-0.011762499
-3,24,0.08975788
-3,25,0.0343491
-3,26,0.03709458
-3,27,-0.11390686
-3,28,-0.023680381
-3,29,-0.0050235917
-3,30,0.15215518
-3,31,-0.044271484
-3,32,-0.16931565
-3,33,-0.006021948
-3,34,0.08127186
-3,35,-0.088947944
-3,36,-0.15933208
-4,14,0.14316997
-4,15,0.00483518
-4,16,-0.09443891
-4,17,0.008454223
-4,18,-0.041526005
-4,19,-0.032540794
-4,20,0.07727843
-4,21,0.086263634
-4,22,-0.11191015
-4,23,0.05057239
-4,24,0.04258554
-4,25,0.06579733
-4,26,-0.015506336
-4,27,0.10872666
-4,28,0.1721223
-4,29,-0.021933256
-4,30,0.2230385
-4,31,-0.07097753
-4,32,0.039091293
-4,33,-0.1383666
-5,14,-0.14635345
-5,15,0.21105821
-5,16,0.21804671
-5,17,0.31388897
-5,18,0.16912724
-5,19,0.16413546
-5,20,0.12470037
-5,21,0.124201186
-5,22,-0.1084159
-5,23,0.03360033
-5,24,0.033849917
-5,25,-0.018314214
-5,26,-0.029545726
-5,27,-0.0031516731
-5,28,0.13618147
-5,29,0.038841702
-5,30,0.043583896
-5,31,0.16014202
-5,32,-0.050261624
-5,33,-0.022182846
-6,18,0.18310423
-6,19,0.15015846
-6,20,0.19208944
-6,21,0.072785825
-6,22,0.23102535
-6,23,0.04008965
-6,24,-0.07297424
-6,25,-0.048264913
-6,26,0.008267031
-6,27,-0.15134524
-6,28,-0.12339125
-6,29,-0.12039618
-6,30,-0.036284633
-6,31,-0.08794959
-6,32,0.030230876
-6,33,-0.07596931
-6,34,-0.10941426
-6,35,-0.009017018
-6,36,-0.18528937
-6,37,-0.08844877
-7,15,0.17511737
-7,16,0.2090615
-7,17,-0.112908505
-7,18,0.0022612917
-7,19,0.1721223
-7,20,-0.036534224
-7,21,0.13917653
-7,22,-0.15833373
-7,23,0.08776117
-7,24,-0.03453751
-7,25,-0.004649208
-7,26,0.0055839475
-7,27,0.096746385
-7,28,-0.053755872
-7,29,0.057560887
-7,30,-0.062491495
-7,31,0.11321927
-7,32,0.01725224
-7,33,0.18809602
-7,34,-0.042274773
-8,20,0.26397112
-8,21,0.18410258
-8,22,0.14316997
-8,23,0.104234055
-8,24,0.19109108
-8,25,0.2230385
-8,26,-0.110911794
-8,27,-0.09992987
-8,28,0.19708122
-8,29,-0.10941426
-8,30,0.015255528
-8,31,-0.022182846
-8,32,-0.20325978
-8,33,-0.18928279
-8,34,0.08276939
-8,35,-0.032540794
-8,36,-0.11340769
-8,37,-0.028547369
-8,38,0.03634581
-8,39,0.18410258
-9,16,-0.004961194
-9,17,0.0662965
-9,18,0.17711408
-9,19,0.31588566
-9,20,0.35781664
-9,21,0.34383965
-9,22,0.18410258
-9,23,-0.024054764
-9,24,0.14416832
-9,25,-0.063989036
-9,26,-0.0071450993
-9,27,0.057560887
-9,28,0.036096223
-9,29,-0.17031401
-9,30,-0.019062981
-9,31,0.012322854
-9,32,0.058309656
-9,33,-0.09044548
-9,34,0.014756349
-9,35,-0.03378874
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
deleted file mode 100644
index 88320a94..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-1-best-3-predictions.csv
+++ /dev/null
@@ -1,25 +0,0 @@
-sequenceId,itemPosition,itemValue
-0,8,0.2899284
-0,9,-0.1753058
-0,10,0.37778378
-0,11,0.26596785
-0,12,-0.30309543
-1,8,0.28194156
-1,9,-0.00092096993
-1,10,0.33784953
-2,8,-0.15034688
-2,9,0.118211046
-3,8,0.3678002
-3,9,0.29791525
-4,7,-0.13537154
-5,7,0.1401749
-6,8,0.3358528
-6,9,0.26996127
-6,10,0.31388897
-7,8,-0.15533866
-8,8,0.16812888
-8,9,-0.25717103
-8,10,0.41771805
-8,11,0.40773448
-9,8,0.31388897
-9,9,0.257981
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
deleted file mode 100644
index fbc8a7dc..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-3-best-3-predictions.csv
+++ /dev/null
@@ -1,25 +0,0 @@
-sequenceId,itemPosition,itemValue
-0,8,-0.19301088
-0,9,-0.08141631
-0,10,0.019747147
-0,11,0.13387236
-0,12,0.30965853
-1,8,0.08696313
-1,9,-0.027594142
-1,10,-0.23646365
-2,8,0.047213733
-2,9,-0.18807307
-3,8,0.08844448
-3,9,0.17436244
-4,7,0.3787879
-5,7,-0.17622232
-6,8,0.2997829
-6,9,0.33731028
-6,10,0.009555192
-7,8,0.6651811
-8,8,0.1990515
-8,9,0.19016345
-8,10,-0.03203817
-8,11,-0.012780699
-9,8,0.0013537919
-9,9,-0.2532522
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
deleted file mode 100644
index 699aaf44..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-5-best-3-predictions.csv
+++ /dev/null
@@ -1,25 +0,0 @@
-sequenceId,itemPosition,itemValue
-0,8,-0.07336837
-0,9,-0.34461385
-0,10,-0.06304177
-0,11,-0.2298738
-0,12,-0.23170963
-1,8,-0.22712004
-1,9,-0.109626226
-1,10,-0.01875211
-2,8,-0.3611364
-2,9,-0.30055368
-3,8,-0.13670488
-3,9,0.23918352
-4,7,0.103331305
-5,7,-0.3464497
-6,8,0.099659614
-6,9,-0.11192103
-6,10,0.15381691
-7,8,0.12719724
-8,8,-0.1766344
-8,9,-0.10939674
-8,10,-0.29688197
-8,11,-0.29137447
-9,8,-0.14358929
-9,9,-0.018981587
diff --git a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
deleted file mode 100644
index 7436f650..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-model-real-50-best-3-predictions.csv
+++ /dev/null
@@ -1,25 +0,0 @@
-sequenceId,itemPosition,itemValue
-0,8,-0.07899063
-0,9,-0.21977666
-0,10,-0.058853757
-0,11,-0.06705767
-0,12,-0.0972343
-1,8,-0.05463706
-1,9,-0.029308192
-1,10,-0.12614879
-2,8,-0.25741142
-2,9,-0.24088883
-3,8,0.014293022
-3,9,0.37136406
-4,7,0.360349
-5,7,-0.1284436
-6,8,0.3419906
-6,9,0.27773616
-6,10,0.31812465
-7,8,0.35667732
-8,8,-0.038143177
-8,9,-0.012097187
-8,10,-0.054866537
-8,11,-0.12614879
-9,8,0.0005242191
-9,9,0.19145167
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv
deleted file mode 100644
index 32cbdbc6..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,121,1,1,2,1
-0,5,113,0,[other],8,2
-0,6,113,1,1,8,0
-0,7,105,3,[other],8,1
-0,8,113,1,1,8,0
-0,9,105,3,1,1,1
-0,10,128,1,8,8,0
-0,11,113,1,0,7,1
-0,12,105,8,9,1,1
-0,13,113,1,8,0,1
-0,14,105,1,9,7,1
-0,15,105,1,7,1,1
-0,16,113,1,9,1,1
-0,17,113,1,[other],1,2
-0,18,113,1,[other],7,1
-0,19,105,8,[other],2,1
-0,20,113,1,[other],8,1
-0,21,113,1,[other],7,1
-0,22,105,8,9,2,1
-0,23,113,1,8,0,1
-0,24,113,1,9,1,1
-0,25,113,1,[other],1,1
-0,26,113,1,[other],7,1
-0,27,105,8,[other],2,1
-0,28,113,1,[other],2,1
-0,29,113,1,[other],7,1
-0,30,105,8,[other],2,1
-0,31,113,1,8,2,1
-0,32,113,3,9,7,1
-0,33,105,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv
deleted file mode 100644
index 9d999e02..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,121,5,[other],9,1
-1,5,121,5,[other],7,1
-1,6,121,5,[other],7,1
-1,7,121,5,[other],7,1
-1,8,121,5,[other],0,1
-1,9,122,1,0,0,1
-1,10,113,1,7,8,1
-1,11,113,1,7,0,1
-1,12,105,1,7,2,1
-1,13,113,1,7,2,1
-1,14,113,[mask],7,1,1
-1,15,113,1,1,2,1
-1,16,113,[mask],[other],1,1
-1,17,113,1,1,2,1
-1,18,113,3,1,7,1
-1,19,105,1,[other],2,1
-1,20,113,1,1,7,1
-1,21,105,8,9,1,1
-1,22,113,1,8,0,1
-1,23,113,1,9,7,1
-1,24,105,1,[other],1,1
-1,25,113,1,[other],7,2
-1,26,105,8,[other],2,1
-1,27,113,1,[other],2,1
-1,28,113,1,[other],7,1
-1,29,105,8,[other],2,1
-1,30,113,1,[other],2,1
-1,31,113,1,[other],7,1
-1,32,105,8,9,2,1
-1,33,113,1,1,2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv
deleted file mode 100644
index c3a6b4fb..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,128,1,8,[unknown],2
-2,4,105,0,[other],7,2
-2,5,105,8,9,0,1
-2,6,128,1,8,0,2
-2,7,105,1,[other],6,1
-2,8,113,1,9,2,1
-2,9,113,1,[other],0,2
-2,10,113,1,[other],2,1
-2,11,113,1,[other],8,1
-2,12,113,1,[other],8,1
-2,13,113,1,[other],7,1
-2,14,105,8,[other],2,1
-2,15,113,1,8,2,1
-2,16,113,3,[other],7,1
-2,17,105,8,[other],2,1
-2,18,113,1,1,2,1
-2,19,113,3,[other],7,1
-2,20,105,8,[other],2,1
-2,21,113,1,[other],2,1
-2,22,113,1,[other],7,1
-2,23,105,8,[other],2,1
-2,24,113,1,[other],2,1
-2,25,113,1,[other],7,1
-2,26,105,8,[other],2,1
-2,27,113,1,[other],2,1
-2,28,113,1,9,7,1
-2,29,105,1,[other],7,1
-2,30,105,8,[other],2,1
-2,31,113,1,[other],2,1
-2,32,113,[mask],9,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv
deleted file mode 100644
index 5cb291b4..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,102,5,1,1,0
-3,4,128,0,1,5,[unknown]
-3,5,105,0,8,5,0
-3,6,127,[other],8,[mask],2
-3,7,109,0,8,[unknown],1
-3,8,127,0,8,[mask],[unknown]
-3,9,102,4,[other],[mask],1
-3,10,124,1,7,8,[unknown]
-3,11,113,1,7,0,2
-3,12,113,1,[other],6,1
-3,13,113,1,[other],8,1
-3,14,113,1,[other],2,1
-3,15,113,1,[other],2,1
-3,16,113,8,[other],1,1
-3,17,113,1,[other],2,1
-3,18,113,3,1,1,1
-3,19,103,1,[other],1,1
-3,20,103,1,9,7,1
-3,21,105,5,8,0,1
-3,22,105,1,8,2,1
-3,23,113,1,9,7,1
-3,24,105,1,9,1,1
-3,25,113,1,9,1,2
-3,26,113,1,[other],1,2
-3,27,113,1,[other],1,1
-3,28,113,1,[other],7,1
-3,29,105,8,[other],2,1
-3,30,113,1,8,2,1
-3,31,113,3,[other],1,1
-3,32,113,1,9,1,1
-4,3,127,0,[other],[unknown],1
-4,4,127,2,1,2,1
-4,5,127,3,[other],8,1
-4,6,127,1,1,0,1
-4,7,127,0,[other],0,1
-4,8,113,1,1,0,1
-4,9,113,3,[other],8,1
-4,10,113,1,[other],0,1
-4,11,113,1,[other],0,1
-4,12,113,1,9,0,1
-4,13,113,1,[other],1,1
-4,14,113,1,[other],1,1
-4,15,113,1,[other],1,1
-4,16,113,1,[other],1,1
-4,17,113,1,9,1,1
-4,18,105,0,[other],1,1
-4,19,113,0,8,0,1
-4,20,105,0,8,2,1
-4,21,113,0,8,0,1
-4,22,113,0,8,2,1
-4,23,113,3,[other],8,1
-4,24,113,1,[other],8,1
-4,25,113,1,[other],8,1
-4,26,113,1,[other],8,1
-4,27,113,1,[other],8,1
-4,28,113,1,[other],7,1
-4,29,105,1,[other],7,1
-4,30,105,1,9,0,1
-4,31,113,1,5,0,1
-4,32,113,1,5,1,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv
deleted file mode 100644
index 79f882a9..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,128,1,[other],4,1
-5,4,105,1,[other],7,1
-5,5,105,1,5,0,1
-5,6,113,1,[other],0,1
-5,7,113,1,[other],2,1
-5,8,113,1,5,2,1
-5,9,113,1,[other],1,1
-5,10,113,1,[other],2,1
-5,11,113,8,9,1,1
-5,12,113,1,[other],7,1
-5,13,105,8,9,1,1
-5,14,103,1,8,0,1
-5,15,113,1,8,7,1
-5,16,105,5,9,1,1
-5,17,113,1,8,0,1
-5,18,113,1,[other],7,1
-5,19,105,8,[other],2,1
-5,20,113,1,[other],8,1
-5,21,113,1,[other],7,1
-5,22,105,8,[other],2,1
-5,23,113,1,[other],2,1
-5,24,113,1,9,7,1
-5,25,105,1,[other],7,1
-5,26,105,8,5,2,1
-5,27,113,1,[other],2,1
-5,28,113,[mask],9,7,1
-5,29,113,1,[other],2,1
-5,30,113,1,[other],7,1
-5,31,105,8,[other],7,1
-5,32,113,1,[other],2,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv
deleted file mode 100644
index 359be249..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,122,4,7,8,1
-6,5,113,1,0,8,1
-6,6,113,1,7,8,1
-6,7,113,1,7,8,1
-6,8,113,1,1,1,1
-6,9,120,0,9,1,1
-6,10,102,0,8,7,1
-6,11,105,0,7,8,1
-6,12,113,1,1,0,1
-6,13,113,1,[other],7,1
-6,14,105,8,5,7,1
-6,15,105,1,[other],8,1
-6,16,113,1,7,0,1
-6,17,113,1,[other],7,1
-6,18,113,8,[other],2,1
-6,19,113,1,[other],8,1
-6,20,113,1,[other],0,1
-6,21,113,1,[other],2,1
-6,22,113,1,[other],7,1
-6,23,105,8,5,2,1
-6,24,113,1,6,8,1
-6,25,113,1,[other],1,1
-6,26,113,1,1,7,1
-6,27,105,8,9,1,1
-6,28,113,1,8,0,1
-6,29,113,1,9,7,1
-6,30,105,1,[other],1,1
-6,31,113,1,9,7,2
-6,32,105,[mask],9,1,1
-6,33,113,1,[other],0,1
-7,3,105,0,9,8,0
-7,4,105,0,1,1,2
-7,5,105,0,9,6,[unknown]
-7,6,113,0,9,1,1
-7,7,113,0,9,1,2
-7,8,113,0,9,6,2
-7,9,113,9,8,1,1
-7,10,113,1,[other],8,1
-7,11,113,1,[other],8,1
-7,12,113,1,[other],8,1
-7,13,113,1,[other],7,1
-7,14,108,1,9,0,1
-7,15,113,1,0,7,1
-7,16,108,1,[other],7,1
-7,17,105,8,[other],7,1
-7,18,105,1,5,2,1
-7,19,113,1,[other],8,1
-7,20,113,1,[other],0,1
-7,21,113,8,[other],2,1
-7,22,113,1,[other],8,1
-7,23,113,1,[other],2,1
-7,24,113,8,[other],7,1
-7,25,105,1,[other],2,1
-7,26,113,1,9,7,1
-7,27,105,1,5,1,1
-7,28,113,1,9,8,1
-7,29,113,1,1,7,1
-7,30,105,8,9,1,1
-7,31,113,1,8,8,1
-7,32,113,1,[other],7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv
deleted file mode 100644
index 67187a34..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,124,1,[other],7,2
-8,5,121,8,[other],6,1
-8,6,114,1,[other],8,2
-8,7,113,1,[other],0,1
-8,8,113,1,[other],0,1
-8,9,113,8,[other],2,1
-8,10,113,1,[other],2,1
-8,11,113,1,[other],0,1
-8,12,113,1,7,2,1
-8,13,113,1,9,1,1
-8,14,113,1,[other],1,1
-8,15,113,1,[other],1,1
-8,16,113,1,9,1,1
-8,17,103,1,9,1,1
-8,18,103,0,8,1,1
-8,19,113,0,8,7,1
-8,20,105,0,9,2,1
-8,21,113,0,8,0,1
-8,22,113,0,[other],8,1
-8,23,113,0,[other],8,1
-8,24,113,1,[other],8,1
-8,25,113,1,[other],8,1
-8,26,113,1,[other],8,1
-8,27,113,1,[other],8,1
-8,28,113,1,[other],7,1
-8,29,108,8,[other],7,1
-8,30,108,1,[other],7,1
-8,31,108,8,5,7,1
-8,32,105,1,[other],0,1
-8,33,113,1,7,7,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv
deleted file mode 100644
index 6793444f..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-predictions/sequifier-test-hp-search-custom-eval-run-0-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,108,1,[other],8,1
-9,4,113,1,[other],7,1
-9,5,105,8,[other],7,1
-9,6,105,5,[other],2,1
-9,7,113,1,7,7,1
-9,8,113,8,[other],1,1
-9,9,113,1,[other],2,1
-9,10,113,1,[other],2,1
-9,11,113,8,5,2,1
-9,12,113,1,[other],8,1
-9,13,113,1,1,7,1
-9,14,105,8,9,7,1
-9,15,105,1,8,0,1
-9,16,113,1,8,7,1
-9,17,105,8,9,7,1
-9,18,105,1,8,2,1
-9,19,113,1,[other],7,2
-9,20,113,8,[other],7,1
-9,21,113,1,[other],2,1
-9,22,113,1,[other],8,2
-9,23,113,1,[other],7,1
-9,24,105,8,[other],2,1
-9,25,113,1,[other],2,1
-9,26,113,1,[other],2,1
-9,27,113,1,[other],0,1
-9,28,113,1,[other],2,1
-9,29,113,3,1,1,1
-9,30,105,1,[other],[mask],1
-9,31,113,1,9,7,1
-9,32,105,1,9,1,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv
deleted file mode 100644
index f02aeba2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,119,9,2,7,[other]
-0,5,104,9,2,7,[other]
-0,6,111,9,9,7,[other]
-0,7,111,9,9,7,[other]
-0,8,111,9,9,7,[other]
-0,9,104,9,9,7,[other]
-0,10,116,1,9,7,[other]
-0,11,104,6,9,7,[other]
-0,12,104,1,9,7,[other]
-0,13,102,1,9,7,[other]
-0,14,104,6,9,7,[other]
-0,15,104,6,9,7,[other]
-0,16,104,6,9,7,[other]
-0,17,104,6,9,7,[other]
-0,18,104,6,9,7,[other]
-0,19,104,6,9,7,[other]
-0,20,104,6,9,7,[other]
-0,21,104,6,9,7,[other]
-0,22,104,6,9,7,[other]
-0,23,104,6,9,7,[other]
-0,24,104,6,9,7,[other]
-0,25,104,6,9,7,[other]
-0,26,104,6,9,7,[other]
-0,27,104,6,9,7,[other]
-0,28,104,6,9,7,[other]
-0,29,104,6,9,7,[other]
-0,30,104,6,9,7,[other]
-0,31,104,6,9,7,[other]
-0,32,104,6,9,7,[other]
-0,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv
deleted file mode 100644
index 756d6e7c..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,111,9,2,7,[other]
-1,5,111,1,9,7,[other]
-1,6,111,1,9,7,[other]
-1,7,111,1,9,7,[other]
-1,8,111,1,9,7,[other]
-1,9,111,1,9,7,[other]
-1,10,102,1,9,7,[other]
-1,11,129,1,9,7,[other]
-1,12,102,1,9,[unknown],[other]
-1,13,104,6,9,[unknown],[other]
-1,14,104,6,9,[unknown],[other]
-1,15,104,6,9,7,[other]
-1,16,104,6,9,7,[other]
-1,17,104,6,9,7,[other]
-1,18,104,6,9,7,[other]
-1,19,104,6,9,7,[other]
-1,20,104,6,9,7,[other]
-1,21,104,6,9,7,[other]
-1,22,104,6,9,7,[other]
-1,23,104,6,9,7,[other]
-1,24,104,6,9,7,[other]
-1,25,104,6,9,7,[other]
-1,26,104,6,9,7,[other]
-1,27,104,6,9,7,[other]
-1,28,104,6,9,7,[other]
-1,29,104,6,9,7,[other]
-1,30,104,6,9,7,[other]
-1,31,104,6,9,7,[other]
-1,32,104,6,9,7,[other]
-1,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv
deleted file mode 100644
index 50574151..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,115,9,2,2,2
-2,4,115,9,2,2,[mask]
-2,5,118,9,2,7,[mask]
-2,6,112,9,9,7,0
-2,7,112,9,9,2,[unknown]
-2,8,115,9,9,7,[other]
-2,9,112,9,9,7,[other]
-2,10,112,9,9,7,1
-2,11,112,9,9,7,[other]
-2,12,112,9,9,7,[other]
-2,13,112,9,9,7,[other]
-2,14,112,9,9,7,[other]
-2,15,112,9,9,7,[other]
-2,16,112,9,9,7,0
-2,17,112,9,9,8,[other]
-2,18,112,9,9,7,[other]
-2,19,112,9,9,7,[other]
-2,20,112,9,9,7,[other]
-2,21,112,9,9,7,[other]
-2,22,112,9,9,7,[other]
-2,23,112,9,9,7,[other]
-2,24,112,9,9,7,[other]
-2,25,112,9,9,7,[other]
-2,26,112,9,9,7,0
-2,27,112,9,9,7,[other]
-2,28,112,9,9,7,0
-2,29,112,9,9,7,[other]
-2,30,112,9,9,7,0
-2,31,112,9,9,7,[other]
-2,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv
deleted file mode 100644
index 08e5981d..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,112,9,2,2,2
-3,4,115,9,2,9,[mask]
-3,5,118,9,2,2,[mask]
-3,6,112,9,2,2,0
-3,7,112,9,9,2,[other]
-3,8,112,9,2,2,[other]
-3,9,112,9,2,7,[mask]
-3,10,112,9,9,7,0
-3,11,112,9,9,7,[other]
-3,12,112,9,9,7,0
-3,13,112,9,9,7,[other]
-3,14,112,9,9,7,[other]
-3,15,112,9,9,7,[other]
-3,16,112,9,9,7,0
-3,17,112,9,9,7,[other]
-3,18,112,9,9,7,[other]
-3,19,112,9,9,7,0
-3,20,112,9,9,7,[other]
-3,21,112,9,9,7,0
-3,22,112,9,9,7,[other]
-3,23,112,9,9,7,[other]
-3,24,112,9,9,7,[other]
-3,25,112,9,9,7,0
-3,26,112,9,9,7,[other]
-3,27,112,9,9,7,[other]
-3,28,112,9,9,7,0
-3,29,112,9,9,7,[other]
-3,30,112,9,9,7,0
-3,31,112,9,9,7,[other]
-3,32,112,9,9,7,[other]
-4,3,119,9,9,7,[other]
-4,4,112,9,9,7,[other]
-4,5,112,9,9,7,[other]
-4,6,112,9,9,7,[other]
-4,7,112,9,9,7,[other]
-4,8,112,9,9,7,[other]
-4,9,112,9,9,7,0
-4,10,112,9,9,7,[other]
-4,11,112,9,9,7,0
-4,12,112,9,9,7,[other]
-4,13,112,9,9,7,0
-4,14,112,9,9,7,[other]
-4,15,112,9,9,7,0
-4,16,112,9,9,7,[other]
-4,17,112,9,9,7,[other]
-4,18,112,9,9,7,0
-4,19,112,9,9,7,[other]
-4,20,112,9,9,7,[other]
-4,21,112,9,9,7,[other]
-4,22,112,9,9,7,0
-4,23,112,9,9,7,[other]
-4,24,112,9,9,7,0
-4,25,112,9,9,7,[other]
-4,26,112,9,9,7,[other]
-4,27,112,9,9,7,0
-4,28,112,9,9,7,[other]
-4,29,112,9,9,7,[other]
-4,30,112,9,9,7,[other]
-4,31,112,9,9,7,0
-4,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv
deleted file mode 100644
index 5ade6550..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,111,3,2,2,[other]
-5,4,111,3,2,8,[other]
-5,5,111,1,2,7,[other]
-5,6,111,1,9,7,[other]
-5,7,111,3,9,7,[other]
-5,8,111,1,9,7,[other]
-5,9,102,1,9,7,[other]
-5,10,102,1,9,7,[other]
-5,11,102,4,9,[unknown],[other]
-5,12,102,6,9,[unknown],[other]
-5,13,104,6,9,5,[other]
-5,14,104,6,9,5,[other]
-5,15,104,6,9,5,[other]
-5,16,104,9,9,5,[other]
-5,17,104,9,9,7,[other]
-5,18,104,9,9,7,[other]
-5,19,104,6,9,7,[other]
-5,20,104,6,9,7,[other]
-5,21,104,6,9,7,[other]
-5,22,104,6,9,7,[other]
-5,23,104,6,9,7,[other]
-5,24,104,6,9,7,[other]
-5,25,104,6,9,7,[other]
-5,26,104,6,9,7,[other]
-5,27,104,6,9,7,[other]
-5,28,104,6,9,7,[other]
-5,29,104,6,9,7,[other]
-5,30,104,6,9,7,[other]
-5,31,104,6,9,7,[other]
-5,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv
deleted file mode 100644
index 7090f23b..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,111,9,9,7,1
-6,5,111,9,2,7,[other]
-6,6,111,1,2,7,[other]
-6,7,102,1,2,7,[other]
-6,8,104,9,9,7,[other]
-6,9,111,1,9,7,[other]
-6,10,102,1,9,7,[other]
-6,11,104,6,9,7,[other]
-6,12,104,1,9,7,[other]
-6,13,104,6,9,7,[other]
-6,14,104,6,9,7,[other]
-6,15,104,6,9,7,[other]
-6,16,104,6,9,7,[other]
-6,17,104,6,9,7,[other]
-6,18,104,6,9,7,[other]
-6,19,104,6,9,7,[other]
-6,20,104,6,9,7,[other]
-6,21,104,6,9,7,[other]
-6,22,104,6,9,7,[other]
-6,23,104,6,9,7,[other]
-6,24,104,6,9,7,[other]
-6,25,104,6,9,7,[other]
-6,26,104,6,9,7,[other]
-6,27,104,6,9,7,[other]
-6,28,104,6,9,7,[other]
-6,29,104,6,9,7,[other]
-6,30,104,6,9,7,[other]
-6,31,104,6,9,7,[other]
-6,32,104,6,9,7,[other]
-6,33,104,6,9,7,[other]
-7,3,111,2,9,7,[other]
-7,4,111,5,9,7,[other]
-7,5,111,1,9,7,[other]
-7,6,117,1,9,7,[other]
-7,7,129,6,9,7,[other]
-7,8,104,6,9,7,[other]
-7,9,104,6,9,7,[other]
-7,10,104,6,9,7,[other]
-7,11,104,6,9,7,[other]
-7,12,104,6,9,7,[other]
-7,13,104,6,9,7,[other]
-7,14,104,6,9,7,[other]
-7,15,104,6,9,7,[other]
-7,16,104,6,9,7,[other]
-7,17,104,6,9,7,[other]
-7,18,104,6,9,7,[other]
-7,19,104,6,9,7,[other]
-7,20,104,6,9,7,[other]
-7,21,104,6,9,7,[other]
-7,22,104,6,9,7,[other]
-7,23,104,6,9,7,[other]
-7,24,104,6,9,7,[other]
-7,25,104,6,9,7,[other]
-7,26,104,6,9,7,[other]
-7,27,104,6,9,7,[other]
-7,28,104,6,9,7,[other]
-7,29,104,6,9,7,[other]
-7,30,104,6,9,7,[other]
-7,31,104,6,9,7,[other]
-7,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv
deleted file mode 100644
index 200bb342..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,111,9,9,7,[other]
-8,5,111,9,9,7,[other]
-8,6,111,9,9,7,[other]
-8,7,111,1,9,7,[other]
-8,8,111,1,9,7,[other]
-8,9,111,1,9,7,[other]
-8,10,102,1,9,7,[other]
-8,11,102,1,9,7,[other]
-8,12,102,1,9,[unknown],[other]
-8,13,104,6,9,[unknown],[other]
-8,14,104,6,9,7,[other]
-8,15,104,6,9,7,[other]
-8,16,104,6,9,7,[other]
-8,17,104,6,9,7,[other]
-8,18,104,6,9,7,[other]
-8,19,104,6,9,7,[other]
-8,20,104,6,9,7,[other]
-8,21,104,6,9,7,[other]
-8,22,104,6,9,7,[other]
-8,23,104,6,9,7,[other]
-8,24,104,6,9,7,[other]
-8,25,104,6,9,7,[other]
-8,26,104,6,9,7,[other]
-8,27,104,6,9,7,[other]
-8,28,104,6,9,7,[other]
-8,29,104,6,9,7,[other]
-8,30,104,6,9,7,[other]
-8,31,104,6,9,7,[other]
-8,32,104,6,9,7,[other]
-8,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv
deleted file mode 100644
index 8b618b25..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-predictions/sequifier-test-hp-search-custom-eval-run-1-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,113,3,2,6,[other]
-9,4,111,4,2,7,[other]
-9,5,117,4,2,7,[mask]
-9,6,117,4,9,7,[mask]
-9,7,104,4,9,7,[other]
-9,8,104,4,9,7,[other]
-9,9,117,4,9,7,[mask]
-9,10,104,9,9,[other],1
-9,11,121,3,9,7,1
-9,12,104,9,9,7,1
-9,13,112,9,9,2,1
-9,14,104,9,9,2,[other]
-9,15,112,9,9,7,[other]
-9,16,119,9,9,8,1
-9,17,112,9,9,7,[other]
-9,18,119,9,9,8,[other]
-9,19,104,9,9,7,[other]
-9,20,129,1,9,7,[other]
-9,21,104,6,9,7,[other]
-9,22,104,6,9,7,[other]
-9,23,104,6,9,7,[other]
-9,24,104,6,9,7,[other]
-9,25,104,6,9,7,[other]
-9,26,104,6,9,7,[other]
-9,27,104,6,9,7,[other]
-9,28,104,6,9,7,[other]
-9,29,104,6,9,7,[other]
-9,30,104,6,9,7,[other]
-9,31,104,6,9,7,[other]
-9,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv
deleted file mode 100644
index f02aeba2..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,119,9,2,7,[other]
-0,5,104,9,2,7,[other]
-0,6,111,9,9,7,[other]
-0,7,111,9,9,7,[other]
-0,8,111,9,9,7,[other]
-0,9,104,9,9,7,[other]
-0,10,116,1,9,7,[other]
-0,11,104,6,9,7,[other]
-0,12,104,1,9,7,[other]
-0,13,102,1,9,7,[other]
-0,14,104,6,9,7,[other]
-0,15,104,6,9,7,[other]
-0,16,104,6,9,7,[other]
-0,17,104,6,9,7,[other]
-0,18,104,6,9,7,[other]
-0,19,104,6,9,7,[other]
-0,20,104,6,9,7,[other]
-0,21,104,6,9,7,[other]
-0,22,104,6,9,7,[other]
-0,23,104,6,9,7,[other]
-0,24,104,6,9,7,[other]
-0,25,104,6,9,7,[other]
-0,26,104,6,9,7,[other]
-0,27,104,6,9,7,[other]
-0,28,104,6,9,7,[other]
-0,29,104,6,9,7,[other]
-0,30,104,6,9,7,[other]
-0,31,104,6,9,7,[other]
-0,32,104,6,9,7,[other]
-0,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv
deleted file mode 100644
index 756d6e7c..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,111,9,2,7,[other]
-1,5,111,1,9,7,[other]
-1,6,111,1,9,7,[other]
-1,7,111,1,9,7,[other]
-1,8,111,1,9,7,[other]
-1,9,111,1,9,7,[other]
-1,10,102,1,9,7,[other]
-1,11,129,1,9,7,[other]
-1,12,102,1,9,[unknown],[other]
-1,13,104,6,9,[unknown],[other]
-1,14,104,6,9,[unknown],[other]
-1,15,104,6,9,7,[other]
-1,16,104,6,9,7,[other]
-1,17,104,6,9,7,[other]
-1,18,104,6,9,7,[other]
-1,19,104,6,9,7,[other]
-1,20,104,6,9,7,[other]
-1,21,104,6,9,7,[other]
-1,22,104,6,9,7,[other]
-1,23,104,6,9,7,[other]
-1,24,104,6,9,7,[other]
-1,25,104,6,9,7,[other]
-1,26,104,6,9,7,[other]
-1,27,104,6,9,7,[other]
-1,28,104,6,9,7,[other]
-1,29,104,6,9,7,[other]
-1,30,104,6,9,7,[other]
-1,31,104,6,9,7,[other]
-1,32,104,6,9,7,[other]
-1,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv
deleted file mode 100644
index 50574151..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,115,9,2,2,2
-2,4,115,9,2,2,[mask]
-2,5,118,9,2,7,[mask]
-2,6,112,9,9,7,0
-2,7,112,9,9,2,[unknown]
-2,8,115,9,9,7,[other]
-2,9,112,9,9,7,[other]
-2,10,112,9,9,7,1
-2,11,112,9,9,7,[other]
-2,12,112,9,9,7,[other]
-2,13,112,9,9,7,[other]
-2,14,112,9,9,7,[other]
-2,15,112,9,9,7,[other]
-2,16,112,9,9,7,0
-2,17,112,9,9,8,[other]
-2,18,112,9,9,7,[other]
-2,19,112,9,9,7,[other]
-2,20,112,9,9,7,[other]
-2,21,112,9,9,7,[other]
-2,22,112,9,9,7,[other]
-2,23,112,9,9,7,[other]
-2,24,112,9,9,7,[other]
-2,25,112,9,9,7,[other]
-2,26,112,9,9,7,0
-2,27,112,9,9,7,[other]
-2,28,112,9,9,7,0
-2,29,112,9,9,7,[other]
-2,30,112,9,9,7,0
-2,31,112,9,9,7,[other]
-2,32,112,9,9,7,0
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv
deleted file mode 100644
index 08e5981d..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,112,9,2,2,2
-3,4,115,9,2,9,[mask]
-3,5,118,9,2,2,[mask]
-3,6,112,9,2,2,0
-3,7,112,9,9,2,[other]
-3,8,112,9,2,2,[other]
-3,9,112,9,2,7,[mask]
-3,10,112,9,9,7,0
-3,11,112,9,9,7,[other]
-3,12,112,9,9,7,0
-3,13,112,9,9,7,[other]
-3,14,112,9,9,7,[other]
-3,15,112,9,9,7,[other]
-3,16,112,9,9,7,0
-3,17,112,9,9,7,[other]
-3,18,112,9,9,7,[other]
-3,19,112,9,9,7,0
-3,20,112,9,9,7,[other]
-3,21,112,9,9,7,0
-3,22,112,9,9,7,[other]
-3,23,112,9,9,7,[other]
-3,24,112,9,9,7,[other]
-3,25,112,9,9,7,0
-3,26,112,9,9,7,[other]
-3,27,112,9,9,7,[other]
-3,28,112,9,9,7,0
-3,29,112,9,9,7,[other]
-3,30,112,9,9,7,0
-3,31,112,9,9,7,[other]
-3,32,112,9,9,7,[other]
-4,3,119,9,9,7,[other]
-4,4,112,9,9,7,[other]
-4,5,112,9,9,7,[other]
-4,6,112,9,9,7,[other]
-4,7,112,9,9,7,[other]
-4,8,112,9,9,7,[other]
-4,9,112,9,9,7,0
-4,10,112,9,9,7,[other]
-4,11,112,9,9,7,0
-4,12,112,9,9,7,[other]
-4,13,112,9,9,7,0
-4,14,112,9,9,7,[other]
-4,15,112,9,9,7,0
-4,16,112,9,9,7,[other]
-4,17,112,9,9,7,[other]
-4,18,112,9,9,7,0
-4,19,112,9,9,7,[other]
-4,20,112,9,9,7,[other]
-4,21,112,9,9,7,[other]
-4,22,112,9,9,7,0
-4,23,112,9,9,7,[other]
-4,24,112,9,9,7,0
-4,25,112,9,9,7,[other]
-4,26,112,9,9,7,[other]
-4,27,112,9,9,7,0
-4,28,112,9,9,7,[other]
-4,29,112,9,9,7,[other]
-4,30,112,9,9,7,[other]
-4,31,112,9,9,7,0
-4,32,112,9,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv
deleted file mode 100644
index 5ade6550..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,111,3,2,2,[other]
-5,4,111,3,2,8,[other]
-5,5,111,1,2,7,[other]
-5,6,111,1,9,7,[other]
-5,7,111,3,9,7,[other]
-5,8,111,1,9,7,[other]
-5,9,102,1,9,7,[other]
-5,10,102,1,9,7,[other]
-5,11,102,4,9,[unknown],[other]
-5,12,102,6,9,[unknown],[other]
-5,13,104,6,9,5,[other]
-5,14,104,6,9,5,[other]
-5,15,104,6,9,5,[other]
-5,16,104,9,9,5,[other]
-5,17,104,9,9,7,[other]
-5,18,104,9,9,7,[other]
-5,19,104,6,9,7,[other]
-5,20,104,6,9,7,[other]
-5,21,104,6,9,7,[other]
-5,22,104,6,9,7,[other]
-5,23,104,6,9,7,[other]
-5,24,104,6,9,7,[other]
-5,25,104,6,9,7,[other]
-5,26,104,6,9,7,[other]
-5,27,104,6,9,7,[other]
-5,28,104,6,9,7,[other]
-5,29,104,6,9,7,[other]
-5,30,104,6,9,7,[other]
-5,31,104,6,9,7,[other]
-5,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv
deleted file mode 100644
index 7090f23b..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,111,9,9,7,1
-6,5,111,9,2,7,[other]
-6,6,111,1,2,7,[other]
-6,7,102,1,2,7,[other]
-6,8,104,9,9,7,[other]
-6,9,111,1,9,7,[other]
-6,10,102,1,9,7,[other]
-6,11,104,6,9,7,[other]
-6,12,104,1,9,7,[other]
-6,13,104,6,9,7,[other]
-6,14,104,6,9,7,[other]
-6,15,104,6,9,7,[other]
-6,16,104,6,9,7,[other]
-6,17,104,6,9,7,[other]
-6,18,104,6,9,7,[other]
-6,19,104,6,9,7,[other]
-6,20,104,6,9,7,[other]
-6,21,104,6,9,7,[other]
-6,22,104,6,9,7,[other]
-6,23,104,6,9,7,[other]
-6,24,104,6,9,7,[other]
-6,25,104,6,9,7,[other]
-6,26,104,6,9,7,[other]
-6,27,104,6,9,7,[other]
-6,28,104,6,9,7,[other]
-6,29,104,6,9,7,[other]
-6,30,104,6,9,7,[other]
-6,31,104,6,9,7,[other]
-6,32,104,6,9,7,[other]
-6,33,104,6,9,7,[other]
-7,3,111,2,9,7,[other]
-7,4,111,5,9,7,[other]
-7,5,111,1,9,7,[other]
-7,6,117,1,9,7,[other]
-7,7,129,6,9,7,[other]
-7,8,104,6,9,7,[other]
-7,9,104,6,9,7,[other]
-7,10,104,6,9,7,[other]
-7,11,104,6,9,7,[other]
-7,12,104,6,9,7,[other]
-7,13,104,6,9,7,[other]
-7,14,104,6,9,7,[other]
-7,15,104,6,9,7,[other]
-7,16,104,6,9,7,[other]
-7,17,104,6,9,7,[other]
-7,18,104,6,9,7,[other]
-7,19,104,6,9,7,[other]
-7,20,104,6,9,7,[other]
-7,21,104,6,9,7,[other]
-7,22,104,6,9,7,[other]
-7,23,104,6,9,7,[other]
-7,24,104,6,9,7,[other]
-7,25,104,6,9,7,[other]
-7,26,104,6,9,7,[other]
-7,27,104,6,9,7,[other]
-7,28,104,6,9,7,[other]
-7,29,104,6,9,7,[other]
-7,30,104,6,9,7,[other]
-7,31,104,6,9,7,[other]
-7,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv
deleted file mode 100644
index 200bb342..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,111,9,9,7,[other]
-8,5,111,9,9,7,[other]
-8,6,111,9,9,7,[other]
-8,7,111,1,9,7,[other]
-8,8,111,1,9,7,[other]
-8,9,111,1,9,7,[other]
-8,10,102,1,9,7,[other]
-8,11,102,1,9,7,[other]
-8,12,102,1,9,[unknown],[other]
-8,13,104,6,9,[unknown],[other]
-8,14,104,6,9,7,[other]
-8,15,104,6,9,7,[other]
-8,16,104,6,9,7,[other]
-8,17,104,6,9,7,[other]
-8,18,104,6,9,7,[other]
-8,19,104,6,9,7,[other]
-8,20,104,6,9,7,[other]
-8,21,104,6,9,7,[other]
-8,22,104,6,9,7,[other]
-8,23,104,6,9,7,[other]
-8,24,104,6,9,7,[other]
-8,25,104,6,9,7,[other]
-8,26,104,6,9,7,[other]
-8,27,104,6,9,7,[other]
-8,28,104,6,9,7,[other]
-8,29,104,6,9,7,[other]
-8,30,104,6,9,7,[other]
-8,31,104,6,9,7,[other]
-8,32,104,6,9,7,[other]
-8,33,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv
deleted file mode 100644
index 8b618b25..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-predictions/sequifier-test-hp-search-custom-eval-run-2-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,113,3,2,6,[other]
-9,4,111,4,2,7,[other]
-9,5,117,4,2,7,[mask]
-9,6,117,4,9,7,[mask]
-9,7,104,4,9,7,[other]
-9,8,104,4,9,7,[other]
-9,9,117,4,9,7,[mask]
-9,10,104,9,9,[other],1
-9,11,121,3,9,7,1
-9,12,104,9,9,7,1
-9,13,112,9,9,2,1
-9,14,104,9,9,2,[other]
-9,15,112,9,9,7,[other]
-9,16,119,9,9,8,1
-9,17,112,9,9,7,[other]
-9,18,119,9,9,8,[other]
-9,19,104,9,9,7,[other]
-9,20,129,1,9,7,[other]
-9,21,104,6,9,7,[other]
-9,22,104,6,9,7,[other]
-9,23,104,6,9,7,[other]
-9,24,104,6,9,7,[other]
-9,25,104,6,9,7,[other]
-9,26,104,6,9,7,[other]
-9,27,104,6,9,7,[other]
-9,28,104,6,9,7,[other]
-9,29,104,6,9,7,[other]
-9,30,104,6,9,7,[other]
-9,31,104,6,9,7,[other]
-9,32,104,6,9,7,[other]
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv
deleted file mode 100644
index 7269a244..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-0-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-0,4,102,6,9,5,1
-0,5,104,6,5,5,1
-0,6,104,6,9,5,1
-0,7,104,6,9,8,1
-0,8,104,6,9,5,1
-0,9,104,6,9,8,1
-0,10,104,4,9,5,1
-0,11,104,6,9,8,0
-0,12,103,4,9,5,1
-0,13,102,6,9,5,0
-0,14,103,6,9,5,1
-0,15,102,6,9,5,1
-0,16,102,6,9,5,1
-0,17,102,6,9,5,1
-0,18,102,6,9,8,1
-0,19,104,6,9,5,1
-0,20,102,6,9,8,1
-0,21,104,4,9,5,1
-0,22,102,6,9,8,1
-0,23,102,6,9,5,1
-0,24,102,6,9,5,1
-0,25,102,6,9,5,1
-0,26,102,6,9,5,1
-0,27,102,6,9,5,1
-0,28,102,6,9,5,1
-0,29,102,6,9,5,1
-0,30,102,6,9,5,1
-0,31,102,6,9,5,1
-0,32,102,6,9,5,1
-0,33,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv
deleted file mode 100644
index f57b3e81..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-1-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-1,4,104,6,9,5,1
-1,5,104,6,9,8,1
-1,6,104,6,9,5,1
-1,7,104,6,9,8,1
-1,8,104,6,9,5,1
-1,9,102,6,9,8,1
-1,10,104,6,9,5,1
-1,11,112,6,9,5,1
-1,12,102,6,9,5,1
-1,13,102,6,9,5,1
-1,14,102,6,9,5,1
-1,15,102,6,9,5,1
-1,16,102,6,9,5,1
-1,17,102,6,9,5,1
-1,18,102,6,9,5,1
-1,19,102,6,9,5,1
-1,20,102,6,9,5,1
-1,21,102,6,9,5,1
-1,22,102,6,9,5,1
-1,23,102,6,9,5,1
-1,24,102,6,9,5,1
-1,25,102,6,9,5,1
-1,26,102,6,9,5,1
-1,27,102,6,9,5,1
-1,28,102,6,9,5,1
-1,29,102,6,9,5,1
-1,30,102,6,9,5,1
-1,31,102,6,9,5,1
-1,32,102,6,9,5,1
-1,33,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv
deleted file mode 100644
index 5d574330..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-2-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-2,3,112,6,5,5,1
-2,4,102,6,5,5,1
-2,5,102,6,1,5,1
-2,6,102,6,5,5,1
-2,7,102,6,1,5,1
-2,8,102,6,1,0,1
-2,9,102,6,1,5,1
-2,10,102,6,9,0,1
-2,11,102,6,9,5,1
-2,12,102,6,9,5,1
-2,13,102,6,9,5,1
-2,14,102,6,9,5,1
-2,15,102,6,9,5,1
-2,16,102,6,9,5,1
-2,17,102,6,9,5,1
-2,18,102,6,9,5,1
-2,19,102,6,9,5,1
-2,20,102,6,9,5,1
-2,21,102,6,9,5,1
-2,22,102,6,9,5,1
-2,23,102,6,9,5,1
-2,24,102,6,9,5,1
-2,25,102,6,9,5,1
-2,26,102,6,9,5,1
-2,27,102,6,9,5,1
-2,28,102,6,9,5,1
-2,29,102,6,9,5,1
-2,30,102,6,9,5,1
-2,31,102,6,9,5,1
-2,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv
deleted file mode 100644
index 3bd7fa01..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-3-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-3,3,104,4,9,8,1
-3,4,104,4,9,8,0
-3,5,104,4,9,8,1
-3,6,104,4,9,8,0
-3,7,104,4,9,8,1
-3,8,104,4,9,8,0
-3,9,104,4,9,5,1
-3,10,112,9,8,8,0
-3,11,104,4,1,5,1
-3,12,112,6,1,8,0
-3,13,112,4,1,5,1
-3,14,112,6,5,5,0
-3,15,112,6,1,5,1
-3,16,112,6,5,5,1
-3,17,102,6,5,5,1
-3,18,102,6,1,5,1
-3,19,102,6,1,0,1
-3,20,102,6,1,5,1
-3,21,102,6,1,0,1
-3,22,102,6,1,5,1
-3,23,102,6,1,0,1
-3,24,102,6,1,0,1
-3,25,102,6,1,0,1
-3,26,102,6,1,0,1
-3,27,102,6,1,0,1
-3,28,102,6,1,0,1
-3,29,102,6,1,0,1
-3,30,102,6,1,0,1
-3,31,102,6,9,0,1
-3,32,102,6,9,5,1
-4,3,112,6,1,5,1
-4,4,102,6,5,0,1
-4,5,104,6,1,5,1
-4,6,102,6,5,0,1
-4,7,104,6,1,5,1
-4,8,102,6,1,0,1
-4,9,112,6,1,5,1
-4,10,102,6,5,5,1
-4,11,102,6,9,0,1
-4,12,102,6,9,5,1
-4,13,102,6,9,0,1
-4,14,102,6,9,5,1
-4,15,102,6,9,5,1
-4,16,102,6,9,5,1
-4,17,102,6,9,5,1
-4,18,102,6,9,5,1
-4,19,102,6,9,5,1
-4,20,102,6,9,5,1
-4,21,102,6,9,5,1
-4,22,102,6,9,5,1
-4,23,102,6,9,5,1
-4,24,102,6,9,5,1
-4,25,102,6,9,5,1
-4,26,102,6,9,5,1
-4,27,102,6,9,5,1
-4,28,102,6,9,5,1
-4,29,102,6,9,5,1
-4,30,102,6,9,5,1
-4,31,102,6,9,5,1
-4,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv
deleted file mode 100644
index 083376d3..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-4-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-5,3,102,6,9,2,1
-5,4,102,6,9,7,1
-5,5,102,6,9,2,1
-5,6,102,6,9,7,1
-5,7,102,6,9,2,1
-5,8,102,6,9,7,1
-5,9,102,6,9,8,1
-5,10,102,6,9,5,1
-5,11,102,6,9,5,1
-5,12,102,6,9,5,1
-5,13,102,6,9,5,1
-5,14,102,6,9,5,1
-5,15,102,6,9,5,1
-5,16,102,6,9,5,1
-5,17,102,6,9,5,1
-5,18,102,6,9,5,1
-5,19,102,6,9,5,1
-5,20,102,6,9,5,1
-5,21,102,6,9,5,1
-5,22,102,6,9,5,1
-5,23,102,6,9,5,1
-5,24,102,6,9,5,1
-5,25,102,6,9,5,1
-5,26,102,6,9,5,1
-5,27,102,6,9,5,1
-5,28,102,6,9,5,1
-5,29,102,6,9,5,1
-5,30,102,6,9,5,1
-5,31,102,6,9,5,1
-5,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv
deleted file mode 100644
index 2015e2b4..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-5-predictions.csv
+++ /dev/null
@@ -1,61 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-6,4,121,6,1,5,0
-6,5,102,6,1,5,1
-6,6,102,6,5,5,1
-6,7,102,6,5,5,1
-6,8,102,6,1,5,1
-6,9,102,6,1,5,1
-6,10,102,6,1,5,1
-6,11,102,6,1,5,1
-6,12,102,6,1,5,1
-6,13,102,6,1,0,1
-6,14,102,6,1,0,1
-6,15,102,6,1,0,1
-6,16,102,6,1,0,1
-6,17,102,6,1,0,1
-6,18,102,6,1,0,1
-6,19,102,6,1,0,1
-6,20,102,6,1,0,1
-6,21,102,6,9,0,1
-6,22,102,6,9,5,1
-6,23,102,6,9,5,1
-6,24,102,6,9,5,1
-6,25,102,6,9,5,1
-6,26,102,6,9,5,1
-6,27,102,6,9,5,1
-6,28,102,6,9,5,1
-6,29,102,6,9,5,1
-6,30,102,6,9,5,1
-6,31,102,6,9,5,1
-6,32,102,6,9,5,1
-6,33,102,6,9,5,1
-7,3,112,4,1,5,1
-7,4,112,6,5,5,1
-7,5,112,6,5,5,1
-7,6,102,6,5,5,1
-7,7,102,6,1,5,1
-7,8,102,6,1,5,1
-7,9,102,6,1,5,1
-7,10,102,6,1,0,1
-7,11,102,6,9,0,1
-7,12,102,6,9,5,1
-7,13,102,6,9,5,1
-7,14,102,6,9,5,1
-7,15,102,6,9,5,1
-7,16,102,6,9,5,1
-7,17,102,6,9,5,1
-7,18,102,6,9,5,1
-7,19,102,6,9,5,1
-7,20,102,6,9,5,1
-7,21,102,6,9,5,1
-7,22,102,6,9,5,1
-7,23,102,6,9,5,1
-7,24,102,6,9,5,1
-7,25,102,6,9,5,1
-7,26,102,6,9,5,1
-7,27,102,6,9,5,1
-7,28,102,6,9,5,1
-7,29,102,6,9,5,1
-7,30,102,6,9,5,1
-7,31,102,6,9,5,1
-7,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv
deleted file mode 100644
index 0c53cfc0..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-6-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-8,4,104,6,1,5,1
-8,5,112,6,5,5,1
-8,6,104,6,5,5,1
-8,7,104,6,1,5,1
-8,8,112,6,1,5,1
-8,9,102,6,5,5,1
-8,10,102,6,9,5,1
-8,11,102,6,9,0,1
-8,12,102,6,9,5,1
-8,13,102,6,9,5,1
-8,14,102,6,9,0,1
-8,15,102,6,9,5,1
-8,16,102,6,9,5,1
-8,17,102,6,9,5,1
-8,18,102,6,9,5,1
-8,19,102,6,9,5,1
-8,20,102,6,9,5,1
-8,21,102,6,9,5,1
-8,22,102,6,9,5,1
-8,23,102,6,9,5,1
-8,24,102,6,9,5,1
-8,25,102,6,9,5,1
-8,26,102,6,9,5,1
-8,27,102,6,9,5,1
-8,28,102,6,9,5,1
-8,29,102,6,9,5,1
-8,30,102,6,9,5,1
-8,31,102,6,9,5,1
-8,32,102,6,9,5,1
-8,33,102,6,9,5,1
diff --git a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv b/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv
deleted file mode 100644
index 2d5e6cb9..00000000
--- a/tests/resources/target_outputs/predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-predictions/sequifier-test-hp-search-custom-eval-run-3-best-3-7-predictions.csv
+++ /dev/null
@@ -1,31 +0,0 @@
-sequenceId,itemPosition,itemId,supCat1,supCat2,supCat3,supCat4
-9,3,102,6,9,5,1
-9,4,102,6,9,5,1
-9,5,102,6,9,5,1
-9,6,102,6,9,5,1
-9,7,102,6,9,5,1
-9,8,102,6,9,5,1
-9,9,102,6,9,5,1
-9,10,102,6,9,5,1
-9,11,102,6,9,5,1
-9,12,102,6,9,5,1
-9,13,102,6,9,5,1
-9,14,102,6,9,5,1
-9,15,102,6,9,5,1
-9,16,102,6,9,5,1
-9,17,102,6,9,5,1
-9,18,102,6,9,5,1
-9,19,102,6,9,5,1
-9,20,102,6,9,5,1
-9,21,102,6,9,5,1
-9,22,102,6,9,5,1
-9,23,102,6,9,5,1
-9,24,102,6,9,5,1
-9,25,102,6,9,5,1
-9,26,102,6,9,5,1
-9,27,102,6,9,5,1
-9,28,102,6,9,5,1
-9,29,102,6,9,5,1
-9,30,102,6,9,5,1
-9,31,102,6,9,5,1
-9,32,102,6,9,5,1
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
deleted file mode 100644
index 5cf025b8..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.009402493,0.007112395,0.0062949057,0.02580099,0.005330445,0.05830615,0.09866104,0.031214688,0.053123884,0.008108478,0.0069419877,0.033780657,0.015496235,0.024613172,0.03495121,0.013683665,0.15753023,0.0058497493,0.010813737,0.012419264,0.008993149,0.04336079,0.0240712,0.007038355,0.061834626,0.040662292,0.013590248,0.084627114,0.0076890606,0.026257254,0.025855722,0.011850584,0.024734203
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
deleted file mode 100644
index 92b890cf..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008010796,0.0071721114,0.005879732,0.027109826,0.0056029055,0.055375658,0.075244375,0.029313548,0.04922832,0.0063652983,0.006468322,0.034970116,0.017254872,0.021584291,0.023749208,0.016210178,0.1830067,0.0067967293,0.011571764,0.008914566,0.011331529,0.037273984,0.022148374,0.007266353,0.05995564,0.035404414,0.009834667,0.13455541,0.0071732746,0.021706868,0.022863243,0.011315752,0.019341161
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
deleted file mode 100644
index a64649f4..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012989529,0.008451787,0.008343695,0.025282247,0.0077611892,0.04980413,0.12995923,0.073096894,0.043786965,0.012898268,0.010859504,0.029631373,0.019566048,0.019631185,0.0460368,0.028859274,0.060754,0.012488811,0.016959468,0.00916429,0.014043812,0.013560963,0.049812313,0.011471328,0.055660956,0.03912309,0.016607722,0.044842925,0.010157409,0.03507745,0.033839986,0.01859359,0.030883793
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
deleted file mode 100644
index 9a9a1059..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0084459195,0.005478003,0.005521865,0.0223774,0.0057007815,0.048822243,0.13396621,0.04767435,0.0441009,0.008718918,0.0076107834,0.030250661,0.022718893,0.023471475,0.03372159,0.020928413,0.12131734,0.007938458,0.010315681,0.009893802,0.0094228275,0.02820915,0.035702728,0.00701184,0.05738481,0.034776427,0.009679547,0.10138574,0.0068074698,0.034072928,0.021294992,0.011075441,0.024202444
-0.014952558,0.011281414,0.01131816,0.040481254,0.013168074,0.043056514,0.0334818,0.020567054,0.034637958,0.008031665,0.010289717,0.013072754,0.027047133,0.024557814,0.02275849,0.027457282,0.042199492,0.00863716,0.033859514,0.014415744,0.03214085,0.015577596,0.018920662,0.013239307,0.03609407,0.24212314,0.0075512314,0.030155566,0.013689236,0.021117216,0.06719196,0.03190991,0.015017746
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
deleted file mode 100644
index 6948a5f7..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010768625,0.0071453005,0.0061826278,0.020636555,0.007476882,0.053042125,0.12125908,0.055434898,0.06060058,0.011440382,0.008968601,0.034554392,0.018097715,0.020440182,0.040972427,0.030373385,0.08881222,0.010173463,0.015603303,0.009588735,0.012654849,0.026746185,0.037209474,0.0092568,0.06076114,0.03545333,0.017796088,0.065680444,0.01078829,0.024638936,0.028522924,0.013651196,0.02526882
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
deleted file mode 100644
index e2c34a6e..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008686484,0.005960354,0.008075498,0.039276745,0.009790896,0.05850304,0.052062083,0.018893363,0.048300244,0.0051852576,0.006793672,0.025432982,0.031197459,0.021319004,0.022767233,0.014915387,0.1545811,0.0053153215,0.01660681,0.011906256,0.014398892,0.03866019,0.019179536,0.0076587293,0.052357905,0.098953515,0.005735833,0.112047225,0.008564644,0.016890615,0.025278356,0.02094096,0.013764391
-0.008197728,0.0070594703,0.005574284,0.02309487,0.005791761,0.051699366,0.11453691,0.0425388,0.04338838,0.009572528,0.0075618555,0.034410793,0.019690046,0.022121828,0.033471037,0.017320944,0.13314813,0.0071966257,0.009637701,0.010214874,0.009963171,0.045684554,0.024528876,0.007794729,0.060744863,0.027903598,0.010524435,0.11521132,0.007692336,0.03072128,0.020051876,0.010017549,0.022933504
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
deleted file mode 100644
index b1992ea3..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.011856951,0.0072238427,0.007057155,0.03282652,0.008080111,0.040391415,0.09255859,0.03939192,0.052117724,0.007856198,0.009969188,0.018405346,0.020338742,0.02072426,0.03398535,0.041261356,0.058055628,0.012493185,0.02786714,0.011796777,0.022020405,0.011738096,0.03508339,0.009734918,0.05840191,0.12817924,0.009151513,0.03046122,0.008224325,0.028107053,0.059158992,0.020592093,0.024889484
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
deleted file mode 100644
index 9b35f4f2..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-1-best-3-itemId-probabilities/sequifier-model-categorical-1-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010474039,0.0065215426,0.00741609,0.023902066,0.006605613,0.057496436,0.12374021,0.03752914,0.047728438,0.009753358,0.008065551,0.027190097,0.014648493,0.02641559,0.031456374,0.0145098185,0.13790835,0.0064065787,0.010360852,0.01592333,0.009660522,0.05928386,0.031192506,0.007285326,0.060071684,0.037452076,0.011793552,0.058254838,0.008734744,0.027273806,0.028270485,0.011573874,0.025100792
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
deleted file mode 100644
index b53fb5b7..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0018302883,0.0023094334,0.0024177611,0.009174009,0.0033216283,0.037184585,0.11439202,0.036135826,0.020121714,0.003243756,0.002227362,0.042347167,0.029958114,0.017858125,0.041660704,0.017803762,0.17849426,0.00218419,0.00554009,0.01079341,0.004177411,0.025839655,0.038697787,0.001474616,0.10490975,0.010089096,0.0046706814,0.1171136,0.0022260093,0.067104094,0.010380509,0.011327853,0.022990765
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
deleted file mode 100644
index e16fcc42..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004845992,0.007889446,0.0054141493,0.012189367,0.007869112,0.017851332,0.07520539,0.011662964,0.021426132,0.0076573635,0.0054324535,0.023717629,0.042729106,0.014602124,0.021532323,0.06217893,0.06529054,0.0031855546,0.038700376,0.0100185275,0.034709904,0.009724645,0.06670463,0.015934361,0.021101989,0.11129304,0.006598829,0.06060536,0.006012587,0.026912445,0.14151993,0.0104822675,0.029001156
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
deleted file mode 100644
index 685739bd..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0023620205,0.0032647604,0.002421969,0.0072277756,0.003854628,0.033725373,0.037664812,0.020529818,0.02037918,0.0032004176,0.0029194388,0.019040832,0.0069631515,0.008101235,0.020486154,0.011356625,0.5755077,0.0024824778,0.0064296694,0.010709552,0.0056811264,0.009269812,0.023953447,0.0020496456,0.026671877,0.011577401,0.0042310487,0.043356534,0.0023190514,0.030348403,0.018102089,0.012962391,0.010849504
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
deleted file mode 100644
index f1b2465f..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0066247783,0.009592915,0.006928904,0.018345714,0.008288779,0.028367922,0.13309619,0.009587395,0.023486814,0.0064903456,0.007973895,0.01584719,0.06669925,0.029178161,0.018443769,0.035900217,0.0030158022,0.003753964,0.021668421,0.010438245,0.025283378,0.036490586,0.043768942,0.020859156,0.029089404,0.09840058,0.011953055,0.03282844,0.009378146,0.014542032,0.118312456,0.013809641,0.08155552
-0.006188416,0.008413583,0.006625596,0.034453783,0.009931466,0.030127268,0.022238078,0.0048153163,0.033921164,0.006972913,0.0063575697,0.009609076,0.016657123,0.025657328,0.010869502,0.020110097,0.005996566,0.0045159073,0.036867622,0.01148467,0.03323851,0.0249987,0.016155457,0.03066806,0.015686167,0.36310777,0.008656854,0.013391952,0.008103484,0.0061835432,0.113910325,0.015572561,0.038513523
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
deleted file mode 100644
index 84eebabe..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.004018745,0.0029350165,0.0041301805,0.010802765,0.005631038,0.032027874,0.13366385,0.03408293,0.018921215,0.00460021,0.002905895,0.023617709,0.08428678,0.023087565,0.058966562,0.03335568,0.027337305,0.0037721738,0.0053366604,0.011020286,0.0070758117,0.04848865,0.041665643,0.003843431,0.10164474,0.010063016,0.006108716,0.10480515,0.004982868,0.063377164,0.008564493,0.016542716,0.05833719
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
deleted file mode 100644
index 85d8f3a9..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0027887959,0.004733092,0.0025837973,0.024125917,0.0042522578,0.04005393,0.042278986,0.0055801696,0.051247988,0.0029797198,0.002457765,0.018063435,0.013389846,0.016801085,0.013436619,0.02084646,0.069999106,0.0024221565,0.02295687,0.012420337,0.017399706,0.022620559,0.02542745,0.008995221,0.021923762,0.23663002,0.0059561864,0.06404682,0.0038476116,0.008797948,0.16114126,0.01581707,0.033978023
-0.004896839,0.008109383,0.005606582,0.033393413,0.0066475375,0.021049256,0.049079787,0.0067314794,0.026396584,0.0060873907,0.00528243,0.014969497,0.022778716,0.0222018,0.014368257,0.022405619,0.003741044,0.0030103682,0.029169861,0.010044832,0.025707247,0.019271096,0.026915964,0.025671674,0.018815916,0.27695572,0.008184225,0.022288233,0.0063604494,0.008164105,0.17993744,0.013626392,0.052130867
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
deleted file mode 100644
index 996b347b..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008245247,0.0066397716,0.00823672,0.006531572,0.013031284,0.030800205,0.1454552,0.043889422,0.015470334,0.009597873,0.008879567,0.027317327,0.024919743,0.02190608,0.078744985,0.020652225,0.046914577,0.0051893904,0.0069066836,0.015767885,0.0067592845,0.021571003,0.099289015,0.0069529354,0.07191968,0.008895552,0.012519008,0.016933145,0.00758302,0.11038498,0.012901287,0.040280648,0.03891432
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
deleted file mode 100644
index c46f46fb..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-best-3-itemId-probabilities/sequifier-model-categorical-3-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0014336656,0.002257306,0.0017123552,0.016647093,0.002066613,0.053203385,0.09665847,0.014232341,0.03271058,0.0018544742,0.001527125,0.025984295,0.03937578,0.022107681,0.021644413,0.02069751,0.152486,0.0017975577,0.008073777,0.0071796924,0.007284085,0.053035576,0.028747264,0.0022616726,0.05012752,0.0346748,0.0034840752,0.19593439,0.0019345195,0.02107136,0.041585583,0.010048055,0.02616101
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
deleted file mode 100644
index f47a7ab9..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.013400168,0.013558126,0.010154567,0.02523116,0.019998511,0.053623423,0.021163935,0.025010373,0.029211547,0.012248745,0.01518438,0.046046063,0.03008736,0.030080015,0.01995949,0.022617064,0.03151204,0.014043256,0.017821819,0.06417883,0.02503481,0.059124317,0.085355915,0.030591028,0.049787663,0.023748823,0.0069248127,0.04524373,0.009765551,0.06673543,0.026364677,0.018315857,0.037876457
-0.016588878,0.0097520985,0.008879387,0.044221062,0.015104345,0.03090144,0.10566724,0.03819547,0.02193381,0.012620153,0.01943228,0.031098215,0.05050204,0.026276983,0.033282157,0.03360877,0.030089956,0.0084727695,0.018578662,0.017901855,0.018945092,0.06434126,0.05313283,0.014244754,0.08490602,0.020364854,0.012521942,0.037676815,0.009600906,0.010462474,0.045892905,0.037383612,0.017418982
-0.007843096,0.012338922,0.010931679,0.012195169,0.010070735,0.041175235,0.048801392,0.027925251,0.021403193,0.014369421,0.014202012,0.060972024,0.024229323,0.03443681,0.07798441,0.03546063,0.0761779,0.012006101,0.012582285,0.013498821,0.017434688,0.038529754,0.023783902,0.010678447,0.07325957,0.052973416,0.016931316,0.037563775,0.030175973,0.022628365,0.028945653,0.023621868,0.054868884
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
deleted file mode 100644
index 9cd7eb50..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.019318607,0.008983985,0.018183628,0.021572521,0.036731947,0.06334328,0.03623317,0.017048607,0.045092907,0.009866976,0.015104229,0.037897944,0.038269855,0.024981832,0.032958474,0.025055127,0.035532355,0.018362077,0.027645616,0.016817106,0.034171134,0.014525594,0.031387832,0.03197566,0.0499133,0.10944736,0.012570854,0.058354557,0.009714004,0.035532355,0.039407626,0.011180749,0.012818792
-0.013353927,0.007942907,0.0119237155,0.012398702,0.015250664,0.05385752,0.05385752,0.022849292,0.04121342,0.012842346,0.011466925,0.051592343,0.0370157,0.01989062,0.055134714,0.030552533,0.0328582,0.0134586645,0.0118772285,0.019391943,0.037598606,0.03587684,0.028024603,0.014104559,0.10879518,0.022363625,0.015072988,0.03864088,0.013885888,0.016684277,0.053022537,0.05710752,0.030093549
-0.015880968,0.012368109,0.019156104,0.021812957,0.015819054,0.039846845,0.030019287,0.041434165,0.06087797,0.017612996,0.0115734665,0.094289646,0.0448886,0.01845826,0.05248009,0.019610377,0.05586477,0.022994122,0.02130766,0.014630232,0.028104289,0.031861883,0.013217314,0.019533923,0.04853615,0.059004948,0.014687494,0.05007685,0.023129247,0.019997157,0.019958138,0.019533923,0.021432877
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
deleted file mode 100644
index 4e2e284d..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012296571,0.013664337,0.018971052,0.011825497,0.024949158,0.078940585,0.03707104,0.020234061,0.029225545,0.015066075,0.017891407,0.023130601,0.015666239,0.043425173,0.039078426,0.022331543,0.037581354,0.009765461,0.017545355,0.02581035,0.034893107,0.030523602,0.022266217,0.016709229,0.13533565,0.09301475,0.011825497,0.02400509,0.016386043,0.025968367,0.041598767,0.016840281,0.01616354
-0.011091116,0.008306863,0.0069950796,0.009486735,0.013483369,0.0819529,0.049127843,0.049901493,0.026658228,0.0070499424,0.012324796,0.03199933,0.024921104,0.049320128,0.04104785,0.048177626,0.046151336,0.008051287,0.012421461,0.019089023,0.03210889,0.023594828,0.03528194,0.020479532,0.047245786,0.053119864,0.01741473,0.06137989,0.010177781,0.028621495,0.08655952,0.010257605,0.01620064
-0.015801238,0.013568414,0.009696907,0.008968174,0.017558832,0.065494314,0.037905198,0.038804095,0.030606769,0.00731961,0.013836027,0.02557291,0.017870212,0.035125118,0.06249511,0.09418346,0.030666607,0.0078527955,0.018222671,0.018401498,0.032900464,0.02009202,0.021097422,0.026204934,0.08847717,0.029846711,0.01636663,0.045633186,0.013782086,0.0396468,0.05986665,0.016239265,0.019896762
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
deleted file mode 100644
index fffebf7c..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.014360843,0.01826049,0.018224861,0.011053937,0.030745259,0.07321582,0.037249196,0.018730072,0.03652873,0.01522744,0.015527775,0.020976687,0.02029163,0.02248273,0.037103977,0.029358827,0.034015287,0.013075697,0.019821575,0.02842257,0.050124273,0.02578917,0.01876669,0.019362409,0.14334963,0.056798242,0.010143663,0.0315205,0.01322983,0.017977389,0.05032045,0.014586994,0.03335737
-0.018110253,0.010020984,0.010625695,0.02915285,0.020481568,0.07639609,0.04651793,0.048370995,0.029231247,0.0155208055,0.016716566,0.021952067,0.123038106,0.0518945,0.02795571,0.030791549,0.022826537,0.01009958,0.016265664,0.027199995,0.033147614,0.01734856,0.022102643,0.016139083,0.056551423,0.031475604,0.012667712,0.037159834,0.011310993,0.02592908,0.047250483,0.02221083,0.013537486
-0.01885988,0.014545349,0.01173326,0.02711961,0.03438508,0.08535359,0.038023666,0.026703235,0.023040008,0.01885988,0.025403539,0.013825159,0.023957819,0.11485581,0.02122545,0.043936264,0.0211427,0.019534715,0.019307127,0.024442276,0.02162296,0.017408503,0.021371044,0.017996224,0.02658371,0.06966357,0.020835249,0.066214286,0.014319844,0.038247116,0.020713525,0.01648204,0.02228757
-0.016209159,0.013075401,0.014932577,0.02497076,0.017732874,0.057466473,0.035681743,0.04312454,0.03374982,0.01487436,0.010303169,0.024190675,0.0761307,0.026587727,0.044930097,0.018082622,0.03186021,0.012623705,0.014932577,0.023722786,0.03187577,0.03325904,0.015527423,0.023105456,0.07916341,0.06692318,0.014304532,0.06537291,0.016083017,0.026770102,0.036814403,0.011404567,0.024214312
-0.011247538,0.009215903,0.007290379,0.026253665,0.011424662,0.05734327,0.029875759,0.031475265,0.026511302,0.009361031,0.013727046,0.07868573,0.03687926,0.03061506,0.05040797,0.020727834,0.037901588,0.007882779,0.010986989,0.042948123,0.016884554,0.061761133,0.057119712,0.01920759,0.08408835,0.028166069,0.007011089,0.045097243,0.007944604,0.060566563,0.021427615,0.017420528,0.022543838
-0.017470723,0.01926294,0.011235971,0.03967968,0.018706782,0.08173608,0.03637658,0.032417256,0.028888872,0.014148223,0.018380824,0.028192118,0.035812616,0.036341075,0.021447431,0.03560339,0.05094972,0.011683561,0.03354442,0.023486488,0.024043452,0.01572195,0.02673089,0.050553225,0.078604825,0.035672996,0.005649801,0.04478768,0.014148223,0.04062066,0.028205885,0.01823778,0.021657905
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
deleted file mode 100644
index d521334c..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.010735215,0.008592358,0.008727668,0.021791875,0.0106101455,0.062261418,0.038208604,0.056248505,0.025929088,0.009889743,0.010735215,0.033587515,0.02101853,0.029095972,0.100666255,0.022330452,0.12237905,0.008135081,0.015139079,0.010693362,0.015317533,0.020095231,0.02846363,0.015257817,0.065761164,0.028561084,0.0071791834,0.026119694,0.010486534,0.10305349,0.02200573,0.016757432,0.014166404
-0.009332249,0.014285676,0.009259625,0.03990897,0.015147736,0.035774168,0.034269553,0.10597081,0.024182387,0.012508955,0.010051228,0.065032184,0.027969994,0.02748263,0.029919641,0.015689746,0.1845372,0.009045126,0.017918251,0.02084652,0.013007255,0.023854017,0.029978134,0.0123632215,0.035984397,0.025164125,0.007044351,0.020106722,0.016314754,0.030361144,0.03393652,0.018129466,0.024623245
-0.010900443,0.006874823,0.0077901953,0.03331444,0.01137903,0.014272501,0.043027967,0.029167539,0.029025465,0.011971795,0.015922125,0.04186736,0.070526674,0.025996035,0.048852395,0.056891568,0.064972214,0.008228086,0.014899139,0.010081262,0.01779714,0.044480614,0.061996922,0.016556391,0.091984116,0.01735095,0.0076097352,0.04057924,0.013251569,0.019205749,0.023646556,0.06756041,0.0220195
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
deleted file mode 100644
index 0109cb4c..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.011403347,0.008811824,0.00895059,0.0143030025,0.0126717575,0.08830369,0.04475013,0.040192187,0.019935472,0.008213523,0.014026358,0.034953695,0.020052625,0.07523569,0.03296447,0.041630324,0.031911284,0.008342868,0.014358983,0.024188083,0.029398052,0.049148407,0.036631156,0.027623659,0.055259038,0.03464783,0.015465233,0.052522976,0.012426664,0.048956797,0.049922384,0.016270857,0.016527088
-0.012426136,0.008742878,0.012921137,0.012720813,0.012138285,0.039182864,0.026063208,0.025609069,0.024954835,0.010880688,0.011537277,0.056346525,0.019663937,0.029260758,0.03887794,0.025175132,0.032230943,0.010838268,0.016017923,0.023899434,0.01762664,0.28952438,0.014246635,0.012138285,0.064601615,0.01705099,0.011358408,0.026929962,0.012971709,0.017834418,0.021491416,0.019721631,0.025015835
-0.019027088,0.015468914,0.014760541,0.015959948,0.02570396,0.038775083,0.05162007,0.01921381,0.03617766,0.019746438,0.031370368,0.02615978,0.057698935,0.026032358,0.0391556,0.03479172,0.042711042,0.016022416,0.028106486,0.035253435,0.029751668,0.04115511,0.026937515,0.01172225,0.063369885,0.042295974,0.016725883,0.029570632,0.022683663,0.014084606,0.045822255,0.034656078,0.027468812
-0.012428584,0.012428584,0.011813202,0.044235602,0.012873298,0.0866091,0.03271299,0.07466172,0.020611685,0.01564995,0.016273372,0.03195519,0.090765566,0.05462368,0.034417097,0.031398386,0.0531503,0.010714109,0.016146732,0.01940074,0.014701767,0.02252724,0.012723318,0.013333924,0.06959237,0.032237284,0.009986647,0.033358194,0.018548507,0.042707466,0.01528742,0.013865086,0.018260945
-0.017474385,0.024095498,0.015481445,0.035995435,0.043080896,0.057520542,0.06724832,0.017136399,0.023127198,0.018029079,0.017169902,0.030075314,0.023954727,0.038316943,0.024727201,0.038316943,0.03634868,0.017338397,0.031151453,0.015941713,0.0183488,0.019820416,0.018894313,0.012151526,0.16644123,0.023978129,0.012784532,0.041349653,0.018931253,0.033814583,0.014830317,0.013138932,0.01298586
-0.011372657,0.021455431,0.012637653,0.01984303,0.034102026,0.020917524,0.021330083,0.017821947,0.037692245,0.014949076,0.014659934,0.07215875,0.019920694,0.028604828,0.030124335,0.052586645,0.055978194,0.019611852,0.016611882,0.03873711,0.02567879,0.045244068,0.06595843,0.016482607,0.052586645,0.02252906,0.012153512,0.030390264,0.0261597,0.032948077,0.027361637,0.022266587,0.059124753
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
deleted file mode 100644
index 8c2ab305..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.016558,0.020446401,0.017625898,0.025545549,0.03687931,0.06523284,0.03474655,0.021892983,0.029667743,0.022543868,0.019701554,0.034174465,0.027259566,0.046619814,0.022107832,0.04253081,0.057567783,0.019358264,0.046257015,0.027702365,0.035640083,0.025645532,0.032577675,0.017625898,0.04431151,0.02756743,0.015253944,0.03804998,0.016558,0.02567059,0.045628898,0.022543868,0.018507898
-0.012470433,0.00840502,0.010098864,0.021737188,0.01853841,0.046696678,0.04799116,0.030120214,0.03165833,0.014409557,0.0128161255,0.040570777,0.034064077,0.024897482,0.054593947,0.026907457,0.04033375,0.007895785,0.016201101,0.012967198,0.025172522,0.03225122,0.04761769,0.009598835,0.21009907,0.025056025,0.012916644,0.04153289,0.009903534,0.009017271,0.02899463,0.0196571,0.014809004
-0.009016722,0.0061488706,0.008470427,0.021128936,0.0092305485,0.03788775,0.050884057,0.0316564,0.028093988,0.010217286,0.015397993,0.05068568,0.056987446,0.026745368,0.074908584,0.07260389,0.051483866,0.0075337538,0.011898647,0.010137774,0.02201357,0.03970602,0.028418211,0.019275725,0.09395641,0.027032569,0.007895307,0.03224926,0.011532564,0.02622486,0.028936293,0.05374428,0.017896906
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
deleted file mode 100644
index 9999fe0a..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-itemId-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.012169482,0.016216686,0.019522885,0.03196851,0.019637613,0.0426419,0.030803923,0.03406355,0.04888918,0.015383666,0.010866124,0.071411744,0.027344117,0.018268555,0.069214635,0.016796945,0.05760553,0.015027304,0.014852231,0.022339424,0.023526108,0.06302065,0.01656886,0.01990794,0.05183937,0.050441086,0.013470389,0.05671244,0.02426106,0.030474791,0.01948479,0.02006408,0.015204442
-0.010947577,0.008199356,0.012649901,0.023091303,0.010820034,0.070831634,0.04906364,0.033919077,0.02334354,0.018695341,0.014760376,0.046452522,0.017256627,0.044152495,0.036036216,0.02321566,0.042794067,0.01044625,0.014222666,0.019535234,0.013360959,0.18808013,0.035616383,0.007947088,0.052843623,0.015408434,0.02007677,0.01968845,0.015529283,0.023857038,0.024028016,0.034453224,0.018677093
-0.011479456,0.012030367,0.009369259,0.02470883,0.017098598,0.060148057,0.15359342,0.041582048,0.018272603,0.013792883,0.019641966,0.028929897,0.0232118,0.035706215,0.024361415,0.055627856,0.022084057,0.010330392,0.013632191,0.018272603,0.019641966,0.040381495,0.03815783,0.013058839,0.046297614,0.02357733,0.027896425,0.026547566,0.01587551,0.012412251,0.06580217,0.036911674,0.019565389
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
deleted file mode 100644
index 76bbd353..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.015032878,0.01945399,0.0163819,0.07815367,0.12835126,0.08055583,0.1132696,0.07102088,0.17612204,0.111949965,0.1159557,0.03536449,0.038387794
-0.013001445,0.017090078,0.012601433,0.06812271,0.07749526,0.03989114,0.09329473,0.2801634,0.10306636,0.07069819,0.1018656,0.032813556,0.089896075
-0.025228165,0.03434838,0.03189139,0.042914554,0.058944356,0.05196744,0.067275494,0.112624,0.059668303,0.23935817,0.07264455,0.029379725,0.1737554
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
deleted file mode 100644
index 2e2a6f5c..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034292977,0.03246794,0.029910866,0.1610662,0.043861385,0.09321798,0.11443634,0.09524237,0.02375405,0.107923746,0.08903612,0.067076854,0.10771316
-0.020602653,0.017349172,0.016045362,0.090195745,0.09657709,0.03730658,0.102205165,0.09415581,0.033572294,0.11268975,0.06775169,0.042273894,0.26927477
-0.034788474,0.054092336,0.0319237,0.06609757,0.057021417,0.08601828,0.10508348,0.03451775,0.022725893,0.13335787,0.059699602,0.03345575,0.28121793
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
deleted file mode 100644
index ff861119..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.017545385,0.01622683,0.028927451,0.17582312,0.03502973,0.05546043,0.05766268,0.06001462,0.07401324,0.27232018,0.063121684,0.09522068,0.048633967
-0.01458249,0.015282317,0.021720547,0.11834111,0.027998893,0.098108225,0.122097656,0.1165064,0.06583057,0.12257554,0.13201904,0.062804475,0.08213275
-0.012693802,0.010942741,0.021592885,0.15769221,0.014383962,0.062756255,0.16333468,0.1766069,0.1014168,0.09416231,0.05715417,0.08457157,0.042691693
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
deleted file mode 100644
index 4293f9e1..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.019174946,0.019325335,0.023039225,0.20295402,0.03028449,0.07068914,0.083618164,0.042292938,0.07901361,0.21604344,0.05490511,0.11838204,0.04027746
-0.013898184,0.0113433665,0.015027519,0.16799693,0.04045206,0.06948655,0.2601987,0.13032606,0.05281098,0.07113437,0.049805496,0.07966671,0.03785304
-0.03320181,0.028733771,0.053160068,0.20101622,0.029878395,0.13390566,0.11725172,0.17261124,0.043387983,0.05976947,0.04728154,0.05394451,0.025857594
-0.015875842,0.017032292,0.019604048,0.06506866,0.03764047,0.13930608,0.16607754,0.0550884,0.048237015,0.16935313,0.10211777,0.077763,0.08683565
-0.011254491,0.013575508,0.013575508,0.045037072,0.14090854,0.06673133,0.15235846,0.07019897,0.106779635,0.12981103,0.11681728,0.023640344,0.10931185
-0.018757429,0.02962508,0.041372053,0.19092914,0.034098256,0.11180388,0.07820428,0.15044855,0.06555784,0.08235855,0.07805169,0.099440426,0.019352855
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
deleted file mode 100644
index 6fdeb3cb..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.0173293,0.016026983,0.026840145,0.08275424,0.045745883,0.046285123,0.11103223,0.21781556,0.03755611,0.22124566,0.059722174,0.06348086,0.054165658
-0.025097355,0.02499951,0.034844227,0.056834597,0.074416414,0.054978497,0.049620453,0.1825707,0.07031857,0.15373994,0.16558443,0.048471,0.05852431
-0.0089370245,0.014059912,0.009588027,0.058219835,0.09721465,0.033596482,0.19869386,0.27371234,0.07796174,0.107606806,0.045674965,0.039935794,0.03479862
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
deleted file mode 100644
index c090598e..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.013744185,0.015574202,0.022660334,0.12888199,0.030492809,0.06577828,0.08997413,0.159148,0.10275368,0.09521741,0.1249167,0.04930175,0.10155656
-0.0139527805,0.025663089,0.013629564,0.051437438,0.07935727,0.041736744,0.07068528,0.102295555,0.17131282,0.1051313,0.07982361,0.04468962,0.20028499
-0.01877227,0.024675684,0.015321485,0.057176925,0.06737088,0.03346519,0.07534136,0.19108021,0.1767203,0.12337066,0.10010392,0.09043637,0.026164724
-0.014150181,0.020588392,0.029262004,0.070919394,0.03421075,0.1003548,0.2978537,0.12710838,0.046125602,0.0878736,0.046035603,0.07792775,0.047589783
-0.02630414,0.04422356,0.06218277,0.1501919,0.0382723,0.13202716,0.094079524,0.09811425,0.039952654,0.08803498,0.0712233,0.09927079,0.056122616
-0.03464103,0.031052018,0.03894796,0.060960393,0.14414504,0.06960196,0.059504647,0.102614194,0.18292959,0.14988713,0.05407388,0.04856628,0.023075903
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
deleted file mode 100644
index 116b4471..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.034527026,0.052956816,0.040682778,0.09759292,0.057315916,0.09740249,0.06494744,0.10908562,0.10908562,0.045296367,0.11123714,0.1387075,0.041162338
-0.010824752,0.016249876,0.015997946,0.12682055,0.058006,0.041092265,0.12339978,0.15660247,0.07543277,0.14258817,0.076695524,0.06861528,0.08767462
-0.010323669,0.013255853,0.0137839075,0.0998801,0.08345294,0.05774269,0.24242425,0.14761323,0.056443825,0.16023265,0.024668645,0.03398249,0.05619568
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
deleted file mode 100644
index e1e8df04..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat1-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.0277286,0.038951144,0.032800153,0.049431577,0.069505766,0.087949365,0.03804884,0.030812891,0.030099113,0.18510105,0.070807464,0.021427004,0.31733704
-0.020127492,0.027511021,0.021093432,0.05807045,0.11719104,0.02656064,0.03849484,0.25101787,0.10712128,0.10181735,0.087772034,0.03504995,0.10817252
-0.023074925,0.024276959,0.028717207,0.1020104,0.054923106,0.05251045,0.06695724,0.23304848,0.07640226,0.0833398,0.11627221,0.028382642,0.11008432
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
deleted file mode 100644
index c968105d..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.011098292,0.012674647,0.01627457,0.025603484,0.100475706,0.047441088,0.04359555,0.02958477,0.11252745,0.14505361,0.39122817,0.01773502,0.04670763
-0.013919137,0.01640075,0.02259299,0.06084698,0.08231952,0.060832124,0.09756611,0.069150954,0.16759425,0.10487779,0.100859985,0.14112851,0.061910894
-0.019119935,0.028037531,0.021835651,0.20879574,0.12713669,0.17929146,0.077329114,0.07641887,0.052074987,0.045331985,0.026338823,0.10664244,0.031646825
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
deleted file mode 100644
index 10e0c261..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.028341014,0.033524573,0.04279492,0.06508739,0.20778364,0.08034278,0.12851283,0.036675997,0.08901823,0.095130295,0.027469048,0.06484156,0.10047761
-0.015947836,0.02712818,0.03246818,0.112004794,0.24179083,0.10298231,0.06662037,0.08495948,0.05490756,0.08234555,0.076345675,0.06780178,0.034697466
-0.034501042,0.03056623,0.03477164,0.133815,0.24042203,0.112683535,0.09161106,0.03894239,0.069830276,0.043528304,0.041779082,0.054224778,0.073324576
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
deleted file mode 100644
index 2ee58cc5..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.015611427,0.028826192,0.02494699,0.0594951,0.25793064,0.17866333,0.06114436,0.059611414,0.08167831,0.079112194,0.0540652,0.055131543,0.043783337
-0.013151082,0.019436039,0.017835513,0.16724718,0.14587559,0.07189701,0.05902547,0.032932926,0.17121333,0.038053922,0.05231921,0.16921864,0.041794058
-0.015268664,0.018707559,0.024977867,0.21077739,0.16035098,0.06837931,0.06590439,0.034678336,0.12342657,0.056069046,0.06993255,0.12635355,0.02517377
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
deleted file mode 100644
index 39f27a25..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.016887669,0.02476413,0.026986418,0.071169205,0.21061277,0.16466737,0.04896162,0.025952585,0.05929014,0.12724507,0.16085285,0.031182745,0.031427313
-0.014183183,0.020237265,0.022139635,0.023021577,0.35177594,0.05240823,0.05402566,0.039391156,0.082258716,0.12840404,0.093942404,0.09212539,0.026086865
-0.05163218,0.03439445,0.04831485,0.060128797,0.07838445,0.17196144,0.10228253,0.056375563,0.054321844,0.15596226,0.024484772,0.116810314,0.044946518
-0.019097615,0.02303612,0.014642649,0.044230174,0.4773803,0.0634187,0.023490468,0.019398358,0.11744495,0.038352862,0.09146622,0.0280048,0.040036753
-0.010892983,0.0099959625,0.0115052825,0.037212107,0.3692803,0.028643182,0.035438757,0.026029024,0.12273083,0.05763524,0.22050735,0.023885578,0.04624343
-0.031303268,0.026257228,0.023171922,0.11228792,0.13282534,0.08952374,0.100654244,0.043037932,0.09660937,0.12974845,0.06882482,0.09698748,0.04876837
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
deleted file mode 100644
index bde6b79b..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.016471865,0.014089133,0.014997804,0.11256474,0.45933443,0.05074929,0.040253956,0.0657868,0.054406185,0.04285011,0.01673126,0.09044831,0.021316173
-0.021560203,0.020734245,0.01732411,0.1722547,0.1394961,0.09795627,0.052895073,0.16630408,0.06833443,0.038736675,0.06235607,0.10346244,0.038585655
-0.020330017,0.017389184,0.014529208,0.09567208,0.13893045,0.0472687,0.07885156,0.06191019,0.20214261,0.15199037,0.0521177,0.065722294,0.05314563
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
deleted file mode 100644
index 982b34a3..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,7 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.01163391,0.02277809,0.018302709,0.120284714,0.14737591,0.072344445,0.059545252,0.0682523,0.15935136,0.04692053,0.07198328,0.1538465,0.047380984
-0.016511263,0.022392593,0.019761395,0.07892483,0.21580069,0.08592345,0.03414463,0.0493899,0.16741052,0.042493675,0.094276346,0.05768628,0.1152844
-0.014698291,0.023033507,0.02206475,0.03471272,0.057652313,0.07060261,0.045807593,0.026511405,0.21430898,0.2795116,0.07701353,0.06707407,0.0670086
-0.019513203,0.019975943,0.014786988,0.03957203,0.5769814,0.047176875,0.023537416,0.040175576,0.04321801,0.04915209,0.045597192,0.058028545,0.022284778
-0.04979623,0.031714294,0.031344812,0.07435257,0.1585615,0.17619851,0.089293055,0.055389207,0.042344138,0.16487788,0.021043912,0.059743974,0.045339987
-0.02798013,0.027119271,0.029901266,0.06500812,0.083533235,0.11872466,0.09958642,0.06744114,0.05382787,0.14320928,0.222675,0.029901266,0.031092396
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
deleted file mode 100644
index c2365f59..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.03437901,0.038352557,0.031180408,0.06441689,0.055206407,0.10610183,0.09326949,0.071722016,0.12140899,0.13438751,0.04130733,0.15407604,0.054191515
-0.014236728,0.02205028,0.020234475,0.0990157,0.23338753,0.08585848,0.054574966,0.04250303,0.13350068,0.14777178,0.06443074,0.047880862,0.034554716
-0.015529931,0.014030124,0.013180082,0.09738835,0.33204457,0.034657136,0.064558625,0.04613797,0.14059651,0.06885666,0.033264406,0.10206211,0.03769351
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
deleted file mode 100644
index 52c7d207..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-3-inf-size-best-3-supCat2-probabilities/sequifier-model-categorical-3-inf-size-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,4 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.03764677,0.03245354,0.023466894,0.1687211,0.20753048,0.13243072,0.05847953,0.050265033,0.09745775,0.026591476,0.02841727,0.04208011,0.094459295
-0.0148268975,0.025518704,0.024445422,0.040461432,0.08976761,0.118922755,0.03641135,0.17507124,0.1462775,0.04273579,0.020108255,0.1806286,0.08482443
-0.010265543,0.017462278,0.026731001,0.066939734,0.021817207,0.07453096,0.062030166,0.12653404,0.14006044,0.0301721,0.022775097,0.364711,0.03597049
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
deleted file mode 100644
index 49604740..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005332064,0.0057897666,0.004876065,0.037203588,0.0035652376,0.025075484,0.044439144,0.016188093,0.035687704,0.0061186426,0.005490237,0.02165509,0.025955847,0.017836059,0.021378342,0.014357305,0.21706617,0.0067667514,0.016861672,0.011250842,0.018886117,0.028481785,0.022863349,0.00604765,0.06405484,0.14882804,0.0102097215,0.07615008,0.007247054,0.015511746,0.02183058,0.013150859,0.023844091
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
deleted file mode 100644
index b6b22fe5..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0054142578,0.0058297585,0.0046986197,0.036962792,0.0031884734,0.02565916,0.045016028,0.017804626,0.042272698,0.0064299456,0.0062360885,0.021977011,0.02834539,0.016593434,0.020529408,0.01365076,0.2205314,0.007261174,0.018225364,0.011267549,0.016855016,0.032483444,0.02040384,0.006038299,0.06135676,0.13856284,0.010517521,0.07868336,0.006724646,0.016816795,0.021524182,0.011967942,0.02017145
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
deleted file mode 100644
index be6c1f23..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0102003245,0.008255288,0.008725312,0.028328696,0.0071364064,0.03169596,0.050800964,0.014434744,0.036991633,0.010709721,0.009840102,0.023735257,0.027150547,0.015559097,0.017729346,0.021140099,0.07350095,0.01044429,0.035601992,0.015230984,0.03764622,0.023684606,0.028552298,0.010491399,0.05137351,0.22636302,0.024168057,0.027686903,0.010378668,0.017089726,0.040571056,0.013346022,0.031436786
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
deleted file mode 100644
index 847be8bb..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0067704804,0.0070975586,0.0061097112,0.032388065,0.00469689,0.029014569,0.054458935,0.015219773,0.036924865,0.007705308,0.0074958117,0.023828786,0.02879159,0.017738651,0.018880464,0.020983135,0.1106364,0.008250364,0.025185334,0.012733434,0.027571186,0.027456705,0.025177589,0.008029096,0.059034105,0.21493421,0.017014707,0.051028233,0.0077972026,0.0175814,0.030099008,0.014413131,0.024953345
-0.0064603947,0.007043517,0.0061552613,0.043098435,0.0042703883,0.02461554,0.040515333,0.016898172,0.034789506,0.007203161,0.0067363298,0.023457421,0.025878388,0.01677529,0.01656298,0.01516267,0.20098753,0.008355653,0.024063593,0.011879986,0.021613162,0.027997697,0.024845859,0.007972059,0.05440866,0.17081751,0.011742852,0.048810527,0.008340559,0.015540328,0.028756257,0.0131472815,0.0250977
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
deleted file mode 100644
index 512da4f8..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.005526586,0.0069941585,0.004799951,0.03426365,0.0041857054,0.022670768,0.034019258,0.020106941,0.023731826,0.0055064755,0.0053447275,0.014821659,0.017681288,0.021764155,0.019465124,0.010148311,0.37970722,0.0066347397,0.013183994,0.009544499,0.013646808,0.025851712,0.020261733,0.008503724,0.052374292,0.05979142,0.0068904837,0.08166511,0.008144954,0.0149318455,0.0149261365,0.01588961,0.017021053
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
deleted file mode 100644
index 5cc07cb2..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.007935801,0.008644695,0.008528165,0.050991338,0.0067945397,0.027761009,0.035833094,0.017671425,0.028613951,0.0080224825,0.007939282,0.020514661,0.01836229,0.019954776,0.018463803,0.013146329,0.21864699,0.010478104,0.024880761,0.014835906,0.021379694,0.021088066,0.027753517,0.012069514,0.051489774,0.15231074,0.013349635,0.04405351,0.010880027,0.014509733,0.024076149,0.014572963,0.02444726
-0.0055697993,0.0056607896,0.0050398135,0.03242026,0.003803431,0.024715057,0.048487585,0.019317342,0.07592577,0.0073058023,0.0071936473,0.02850324,0.039357413,0.016763113,0.024640776,0.017066706,0.107044175,0.0064190463,0.02185047,0.009070911,0.016291386,0.042864796,0.0281329,0.005143986,0.0739408,0.13126434,0.012961427,0.079355635,0.007179909,0.021646269,0.034544323,0.011274983,0.02924409
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
deleted file mode 100644
index fb88a77b..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.006583764,0.0068928925,0.0048378515,0.03660268,0.004346068,0.02784286,0.042631034,0.017884584,0.03352133,0.0069473186,0.0063436897,0.018111207,0.024766268,0.018511454,0.01849432,0.013050358,0.25470734,0.007486411,0.018900616,0.010490363,0.018111968,0.020677898,0.024124417,0.009464148,0.06472242,0.13972169,0.010750386,0.05095634,0.00813771,0.01582268,0.02160351,0.016543156,0.020411283
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
deleted file mode 100644
index b2f4d541..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-5-best-3-itemId-probabilities/sequifier-model-categorical-5-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008046485,0.010288931,0.006996726,0.03990828,0.00716229,0.033754513,0.03726218,0.021291913,0.027013965,0.007882266,0.007868176,0.019705186,0.023202414,0.024027875,0.017250976,0.014517649,0.2669638,0.011102698,0.021171337,0.014237667,0.020665446,0.017218808,0.024644135,0.01385889,0.05738821,0.10421709,0.011216665,0.042277344,0.012417349,0.016171632,0.019570941,0.01778471,0.022913428
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
deleted file mode 100644
index 79a63b85..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0012904094,0.0013854031,0.0013756184,0.0049201986,0.0026279963,0.044148285,0.07374064,0.04074432,0.04855693,0.0037117477,0.001404071,0.06342667,0.05690586,0.019039612,0.07667846,0.020172833,0.03751043,0.0011957006,0.0070678433,0.003960929,0.0049003386,0.08352786,0.029856034,0.0016307909,0.09834979,0.00176659,0.013229489,0.1697645,0.0018912563,0.019570507,0.0122591285,0.00250594,0.050883807
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
deleted file mode 100644
index 09d5a2d2..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.015634717,0.01281785,0.009619565,0.048077118,0.012959089,0.015963905,0.028544521,0.007311691,0.007936274,0.007725436,0.014542746,0.0027863763,0.010137666,0.01106823,0.014024269,0.012997272,0.03277003,0.017333027,0.018349424,0.024214767,0.03291807,0.0027870873,0.03269819,0.055223055,0.009176768,0.312866,0.004502388,0.0028817935,0.019294238,0.03272431,0.073402315,0.09117624,0.005535524
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
deleted file mode 100644
index 39f7794f..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0006198204,0.00063342374,0.0004384532,0.006420689,0.0018287954,0.06507862,0.1269804,0.011015099,0.04099147,0.001311511,0.0005743433,0.053394005,0.013784635,0.035449795,0.03551256,0.016842777,0.099894196,0.0006676215,0.013255125,0.008228865,0.0147693455,0.022764573,0.02433358,0.0038868464,0.034915894,0.08704853,0.003976894,0.14972544,0.00076197524,0.026150865,0.054506917,0.018916963,0.025319949
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
deleted file mode 100644
index 84199ce6..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.008406845,0.0100028645,0.004978663,0.028080318,0.0074564624,0.05106336,0.041368242,0.009740752,0.0049058977,0.0059933676,0.007738937,0.017632268,0.006884857,0.034210872,0.021888284,0.011264261,0.012693914,0.009070253,0.011593515,0.026868774,0.041217517,0.010789327,0.05191277,0.029620975,0.016518576,0.22696489,0.006126781,0.0071672136,0.01255229,0.037649535,0.10909438,0.1091829,0.009360133
-0.00081456243,0.00051110506,0.000592047,0.00687621,0.0025706633,0.021814596,0.050301585,0.009817175,0.11709266,0.0014886729,0.00065051304,0.01347012,0.02308426,0.012487394,0.021035971,0.016027665,0.18970953,0.00078846415,0.018496549,0.005921485,0.009317239,0.015121425,0.008347157,0.0038045023,0.022821637,0.034895573,0.003979794,0.3231347,0.0008225577,0.014405082,0.024048084,0.0045162523,0.02123482
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
deleted file mode 100644
index 9d17464e..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.000644279,0.00054107426,0.0006426197,0.007318598,0.0016342783,0.046498753,0.114201136,0.03403976,0.04745785,0.0018703459,0.00065563683,0.033165775,0.049821507,0.015586319,0.048468433,0.019384295,0.11359834,0.0006508778,0.006752685,0.0037247261,0.0035874853,0.040763292,0.021660022,0.0012675843,0.087549336,0.003357906,0.006457923,0.2118147,0.0010242914,0.023060877,0.0100330515,0.0032326481,0.03953359
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
deleted file mode 100644
index 6de20a20..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.00171548,0.001514297,0.0009926202,0.01718312,0.00527013,0.029487936,0.05267179,0.005046223,0.118690774,0.0020176421,0.0016863676,0.011463078,0.016580433,0.016751137,0.024944035,0.0152443405,0.24363711,0.0021403863,0.033990223,0.010999503,0.035781205,0.0076473923,0.015846951,0.0128892865,0.020958621,0.13015136,0.0035353715,0.062455773,0.001783925,0.017721385,0.050996166,0.016302522,0.011903471
-0.027323829,0.037167955,0.029990472,0.07488815,0.025425775,0.060669463,0.025825078,0.022680491,0.008052186,0.018953843,0.03934911,0.011748633,0.017262118,0.02899022,0.01647767,0.0139917005,0.021608276,0.05273841,0.012544975,0.029721929,0.036775358,0.00949796,0.053365026,0.036835194,0.033656303,0.029669568,0.0132582355,0.00075166044,0.05680184,0.031587865,0.026841847,0.075078,0.020470945
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
deleted file mode 100644
index f362f47d..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.0008632209,0.00079384743,0.0006646036,0.018937923,0.0021681492,0.06119901,0.1648909,0.011469957,0.03668653,0.0013531677,0.0008869396,0.012405908,0.020405136,0.021181801,0.037859116,0.016523449,0.1998736,0.0012055563,0.012034784,0.008868842,0.012623,0.00803282,0.025626624,0.0059644775,0.03410578,0.09410486,0.0027197315,0.04003143,0.001474787,0.04476594,0.049414117,0.03120969,0.019654365
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
deleted file mode 100644
index 9cbae14a..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-50-best-3-itemId-probabilities/sequifier-model-categorical-50-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.006902314,0.009175283,0.010000225,0.015625658,0.0108586475,0.09034072,0.04221178,0.06655271,0.033279005,0.011762234,0.008651375,0.061778508,0.059007354,0.030211158,0.05453736,0.023306027,0.035477407,0.007880052,0.008436053,0.014697746,0.008330109,0.07010769,0.029269386,0.0033454658,0.15206087,0.0012518687,0.02848025,0.0133461775,0.010879269,0.022960618,0.005361438,0.006997196,0.046918076
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
deleted file mode 100644
index 65ce62b1..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-0-probabilities.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.035109174,0.032228634,0.029122373,0.027463987,0.027945641,0.030163735,0.032845262,0.02594842,0.034922395,0.030951612,0.027564056,0.027116977,0.027191272,0.029049408,0.031170875,0.030246926,0.030047692,0.031019162,0.02844935,0.035145335,0.032860417,0.033282854,0.03376338,0.030274218,0.03223822,0.030229334,0.027426753,0.03153387,0.026364023,0.030059563,0.03067464,0.02712991,0.030460587
-0.027818047,0.029806498,0.034336902,0.030778736,0.026007194,0.031295177,0.036927246,0.030886251,0.03125275,0.02985114,0.028995993,0.02902339,0.02976673,0.027265698,0.028691404,0.025839522,0.03046781,0.029810123,0.03192401,0.030833513,0.027267322,0.027360896,0.026245909,0.028813887,0.031931926,0.032841697,0.03109479,0.031397782,0.03992314,0.030624682,0.032517962,0.029487668,0.028914195
-0.030183403,0.03242052,0.029843858,0.0296451,0.026645405,0.031996164,0.028702747,0.030074446,0.028383184,0.030225128,0.032427616,0.029124603,0.0312291,0.030315014,0.033579975,0.03126286,0.02904425,0.03404973,0.030180257,0.03460662,0.029601347,0.034505613,0.02989275,0.03159202,0.030262558,0.030775197,0.025916839,0.030558,0.026761474,0.02995052,0.025229465,0.028859543,0.032154743
-0.025296634,0.026471082,0.026577652,0.033839412,0.02780684,0.039299954,0.03074986,0.029498447,0.0331007,0.030452378,0.02612987,0.029916089,0.026885556,0.031654015,0.03073531,0.03307961,0.027617231,0.029398216,0.02377331,0.028508658,0.0275827,0.033749986,0.036714476,0.034851376,0.03230926,0.035462867,0.02979035,0.031253695,0.02989457,0.027552374,0.027788708,0.028714927,0.033543948
-0.025478277,0.028292565,0.028266197,0.03619709,0.029252974,0.0347753,0.030366326,0.034015596,0.027419873,0.030975945,0.029481627,0.03149804,0.029024819,0.02893344,0.03221479,0.033274993,0.029894393,0.03219735,0.02582195,0.029421443,0.026916182,0.034794923,0.028513875,0.03288691,0.031072868,0.031177122,0.031510886,0.029790463,0.029789902,0.029741006,0.025122572,0.0299503,0.031930022
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
deleted file mode 100644
index f2aa4f3b..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-1-probabilities.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.026077993,0.025926517,0.030068072,0.03636011,0.027121866,0.034626346,0.031620685,0.0315067,0.031596735,0.032528404,0.03069347,0.0314288,0.027027652,0.029328331,0.032343224,0.029408194,0.028625358,0.024174193,0.029095232,0.026811505,0.02688699,0.03139536,0.029974222,0.033895213,0.032119796,0.030733231,0.028192906,0.033536673,0.03282051,0.031133441,0.02963972,0.031351656,0.031950876
-0.028449364,0.024803886,0.0294986,0.03234279,0.029346153,0.03284062,0.029287698,0.031351723,0.03181133,0.02964178,0.031152815,0.030545823,0.028929776,0.032174207,0.032390404,0.029105593,0.031171152,0.027026553,0.02852814,0.027785255,0.029967785,0.031725638,0.03196683,0.033753444,0.030419005,0.026873102,0.030337531,0.033957142,0.027139246,0.032121036,0.03149567,0.03064888,0.031410974
-0.02473332,0.025805464,0.026985694,0.034197494,0.02668728,0.04076238,0.030618208,0.029560212,0.0342524,0.031228403,0.026219497,0.029502872,0.026915431,0.03167128,0.030253854,0.031405352,0.028886192,0.028909547,0.023818407,0.028952114,0.0280342,0.033739235,0.035629865,0.03535153,0.033023417,0.034959845,0.027452167,0.031207168,0.030123003,0.029052667,0.026981011,0.029005291,0.034075227
-0.027612552,0.029588765,0.028897526,0.028167427,0.03146175,0.030992467,0.030953089,0.03307828,0.028132018,0.028136177,0.030538997,0.032958314,0.03215302,0.032910448,0.03135702,0.03136277,0.032575402,0.030259186,0.031658836,0.031609565,0.029420199,0.029490033,0.034439214,0.03181292,0.02608727,0.026992736,0.033126123,0.028185237,0.027118426,0.031353865,0.029262694,0.028808676,0.029499011
-0.028027277,0.029319167,0.029869592,0.028233605,0.0284659,0.03679354,0.030892001,0.03222718,0.030400075,0.029845333,0.028747197,0.028966922,0.030681144,0.0330812,0.02865518,0.030550085,0.030910296,0.03369551,0.029822486,0.032989994,0.029237293,0.031726234,0.03142848,0.03187399,0.030198226,0.033343483,0.026842874,0.03070044,0.026204878,0.029943591,0.027344182,0.027224164,0.031758487
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
deleted file mode 100644
index 39b55eb7..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-2-probabilities.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.034084663,0.031583164,0.032546442,0.032138847,0.029565305,0.02397673,0.032969736,0.031043336,0.027734615,0.02915201,0.031556893,0.02899239,0.030178683,0.024926245,0.032692198,0.02895135,0.031203741,0.03237466,0.029557178,0.030954232,0.029987698,0.030188313,0.023178574,0.026828399,0.03162781,0.025829652,0.03609925,0.031925663,0.034451466,0.03171625,0.032841537,0.031190302,0.027952647
-0.025305754,0.0299227,0.030633897,0.03035362,0.030023782,0.031172622,0.034432754,0.032215968,0.02902592,0.0293114,0.029836485,0.03252568,0.03256719,0.02886391,0.029192211,0.027415669,0.0338695,0.027007982,0.034716416,0.02912123,0.02748679,0.027996587,0.02957499,0.029614456,0.03000929,0.03120561,0.03207879,0.02746605,0.034783393,0.03193119,0.03044039,0.030926643,0.028971093
-0.028770477,0.027449202,0.030062556,0.03365554,0.029413687,0.029721614,0.035221003,0.03161255,0.033253502,0.030102383,0.026639005,0.028710429,0.027786555,0.02702573,0.028572798,0.029887274,0.032881945,0.030976614,0.022857795,0.029277911,0.02882068,0.030132292,0.026810376,0.030622266,0.032237146,0.028753674,0.035257693,0.031009149,0.036623254,0.032417532,0.031938296,0.031099904,0.03039909
-0.028727695,0.029837677,0.029908122,0.02894165,0.033573724,0.030687451,0.028665278,0.033794753,0.027919773,0.029874934,0.030719016,0.03329227,0.03228232,0.032287028,0.029487554,0.031092698,0.03433863,0.031789694,0.032016642,0.030929402,0.031242827,0.030201329,0.03151861,0.030687474,0.026339065,0.026228543,0.031674992,0.02779673,0.025773538,0.031888627,0.028663844,0.029518012,0.028300088
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
deleted file mode 100644
index b31d6767..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-3-probabilities.csv
+++ /dev/null
@@ -1,9 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.033855878,0.031555194,0.032104813,0.032544944,0.02899993,0.024226334,0.033384256,0.031535964,0.027571268,0.028917968,0.03153802,0.029317284,0.029884605,0.02487106,0.03317497,0.029042145,0.031352926,0.031901088,0.028874978,0.031184962,0.029574865,0.030010147,0.02339816,0.027419098,0.031692605,0.02540574,0.03633683,0.03199998,0.034317236,0.032391068,0.03252312,0.031074857,0.028017707
-0.032645375,0.029211296,0.029622218,0.029929172,0.026831938,0.033350952,0.02760981,0.026967196,0.03171539,0.030520461,0.03234894,0.028386977,0.02834992,0.032421347,0.03404085,0.028682854,0.028301395,0.03028663,0.030858088,0.032663256,0.03316226,0.03464805,0.033819176,0.03297321,0.03245527,0.029456258,0.02407616,0.032931916,0.023553552,0.029962027,0.027696542,0.027935032,0.03258643
-0.03079726,0.02836858,0.026739504,0.029305056,0.029472884,0.034308825,0.030819302,0.026103878,0.033304565,0.029323498,0.027538827,0.028037373,0.027434818,0.030544518,0.032991886,0.030903487,0.029249443,0.029962836,0.028534466,0.030887995,0.030986693,0.037431307,0.03750989,0.032438066,0.033674967,0.032570437,0.029678013,0.032340832,0.022803176,0.027646143,0.029873798,0.027064433,0.03135321
-0.025392696,0.02627636,0.027077502,0.035034917,0.027847782,0.04042039,0.029661613,0.030495398,0.033150274,0.03138564,0.026224535,0.030131755,0.02675575,0.031256,0.029583888,0.032415524,0.02878367,0.03011321,0.023064954,0.028201777,0.027973127,0.03411287,0.03447896,0.035180874,0.033063572,0.03497268,0.028549135,0.031398144,0.029217428,0.028573478,0.026990656,0.029208178,0.03300724
-0.024685439,0.030674905,0.029863117,0.031178279,0.02778255,0.035222333,0.02968134,0.030667255,0.030530095,0.03223433,0.028045597,0.030551089,0.03133845,0.029112432,0.02964785,0.033652943,0.027950848,0.031787325,0.028992387,0.029821344,0.025748067,0.032986067,0.031084312,0.030607037,0.02886095,0.037558336,0.02887098,0.029779462,0.035294015,0.026569769,0.027097931,0.031046059,0.031077052
-0.027818462,0.029526941,0.034052044,0.030185841,0.025486842,0.032498095,0.036262725,0.02962454,0.032617006,0.02988826,0.02899451,0.028541261,0.029611835,0.028164327,0.028638976,0.025260951,0.0305133,0.029842326,0.03165989,0.031313885,0.02806211,0.02798404,0.027728057,0.029662656,0.032396797,0.033439204,0.029236948,0.03131323,0.03833288,0.030617509,0.03189458,0.029205821,0.029624088
-0.02698204,0.029599056,0.030312235,0.033456706,0.03267925,0.028223274,0.027507698,0.03228416,0.028564464,0.032395225,0.030720642,0.031880267,0.03154193,0.028109167,0.030944703,0.035500936,0.02953083,0.032430664,0.027058449,0.027667807,0.027628645,0.033845417,0.027544307,0.029797006,0.02666573,0.029446175,0.033847805,0.029539702,0.034211446,0.027991222,0.028181324,0.034006014,0.029905727
-0.030370781,0.027035689,0.028008454,0.031036004,0.030973144,0.028833287,0.033738982,0.03354538,0.027871111,0.027916575,0.029257156,0.029191185,0.030842572,0.02881783,0.03315752,0.02846031,0.0370996,0.029838745,0.03092998,0.032305267,0.03198498,0.030699644,0.026844112,0.028627403,0.032551426,0.02384607,0.03394906,0.030086126,0.026287593,0.03627954,0.030584106,0.028412709,0.030617632
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
deleted file mode 100644
index d25ff96a..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-4-probabilities.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.028437903,0.027254757,0.030338282,0.034053758,0.028367294,0.029867936,0.035484016,0.032350704,0.033063136,0.030184561,0.027064085,0.028652815,0.027901977,0.026792774,0.028899506,0.029313952,0.033530645,0.030812234,0.022711875,0.029815352,0.028521638,0.029946608,0.02587863,0.030986773,0.032278765,0.027954705,0.034373548,0.031195903,0.037098233,0.033722892,0.031238012,0.031281143,0.030625613
-0.025746314,0.024933858,0.026921023,0.035529517,0.029480062,0.033978727,0.031168824,0.03179294,0.029984899,0.028650545,0.027421633,0.030246083,0.029631928,0.030445006,0.032250527,0.029986437,0.032097112,0.028758565,0.026263552,0.02810667,0.030151488,0.030298227,0.03089507,0.030969998,0.033610586,0.028924711,0.03355842,0.02988841,0.032800015,0.03190985,0.02990864,0.030362934,0.033327445
-0.032267183,0.028918488,0.028129185,0.026897648,0.027366351,0.030486971,0.029913964,0.02686379,0.03348956,0.029299267,0.028501146,0.027437123,0.031400625,0.030622374,0.033201296,0.028855711,0.031837236,0.027683068,0.03356016,0.03306354,0.033076447,0.030550973,0.03371906,0.02831449,0.032340366,0.028730428,0.027243681,0.032467235,0.027846765,0.031897455,0.033190466,0.029134385,0.031693567
-0.025270436,0.02990884,0.02897759,0.03144431,0.028191876,0.03697099,0.02852898,0.02989229,0.031260435,0.03201398,0.027516115,0.03016531,0.030635659,0.029858166,0.029304778,0.033509642,0.027898662,0.031974133,0.02800069,0.029125461,0.02662284,0.034055386,0.032390516,0.03126936,0.030411147,0.038490403,0.028171586,0.030406484,0.03229996,0.02624073,0.02718416,0.030607712,0.031401414
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
deleted file mode 100644
index a425fc01..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-5-probabilities.csv
+++ /dev/null
@@ -1,10 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.029452197,0.0306485,0.03368184,0.029208787,0.02547852,0.031809207,0.035802938,0.028849022,0.033015896,0.029569034,0.028465796,0.027966324,0.029913383,0.027351242,0.028488118,0.025190182,0.031235317,0.030739307,0.03148216,0.032006137,0.028662467,0.028530112,0.027585056,0.028937826,0.03342983,0.03354061,0.029346963,0.03168956,0.03664937,0.030737981,0.032591004,0.029114723,0.028830616
-0.025410624,0.030470952,0.030369842,0.029661402,0.02980678,0.030244453,0.035259236,0.033240262,0.028276877,0.02860871,0.029643638,0.032424532,0.03325316,0.028354872,0.029587941,0.028144129,0.034343008,0.02772666,0.034166075,0.029962657,0.026728328,0.027766544,0.028914964,0.029249499,0.029259473,0.030429421,0.033704724,0.027519733,0.035062674,0.032374986,0.030571992,0.030798515,0.028663335
-0.025337975,0.027083492,0.02744322,0.030657839,0.028513115,0.031966254,0.031017954,0.03185362,0.033549584,0.032333627,0.027234409,0.027745046,0.033427104,0.028093094,0.029163105,0.030709269,0.03713305,0.027277596,0.028337648,0.03032646,0.028630411,0.03181621,0.026347104,0.028802983,0.032475777,0.030587494,0.026387088,0.029571716,0.034264978,0.03554306,0.02840355,0.033852767,0.034113403
-0.025487956,0.028920492,0.031584144,0.032168623,0.030936489,0.03206438,0.032932498,0.032587145,0.030902065,0.0329446,0.030976553,0.03291904,0.028920155,0.029618418,0.028930362,0.031141395,0.029884664,0.026178582,0.030916898,0.027372908,0.025814181,0.03133012,0.030211058,0.03310097,0.02752377,0.032027237,0.030076312,0.030099913,0.03261934,0.029556777,0.02907321,0.031551532,0.029628236
-0.028115489,0.029519554,0.031406675,0.033021085,0.02769064,0.03581125,0.02951925,0.030510264,0.02980014,0.029648412,0.029389903,0.031621452,0.02773173,0.030956693,0.030655386,0.030680252,0.026869236,0.03531161,0.025839085,0.030112183,0.028472967,0.032457538,0.033420917,0.033944823,0.029430335,0.03254275,0.03216424,0.031369843,0.030058293,0.026615452,0.02813199,0.028111547,0.029069062
-0.032825507,0.029050106,0.029823868,0.028476052,0.02723421,0.0323307,0.02776565,0.02706704,0.0309637,0.02922705,0.03263243,0.02747781,0.029745135,0.03299441,0.03421459,0.028740628,0.02831149,0.03199856,0.031706,0.03320294,0.03318537,0.034523427,0.033132866,0.031837754,0.03197639,0.029589491,0.025036542,0.0330899,0.023499442,0.029383754,0.02828902,0.02767611,0.032992084
-0.024933483,0.027506273,0.029821714,0.029868472,0.0295652,0.03086805,0.030575937,0.032271694,0.031803988,0.02926964,0.02913596,0.029194605,0.034982435,0.030616261,0.027791448,0.028632607,0.0363351,0.03158228,0.027264304,0.030072883,0.029574117,0.029052928,0.027467903,0.02984582,0.029604388,0.029606704,0.029627847,0.02741072,0.036741782,0.033972777,0.028591022,0.033564672,0.032846995
-0.024271613,0.02529622,0.025262209,0.031130273,0.029015483,0.032995865,0.03252496,0.032081917,0.030489897,0.027709953,0.02744618,0.028730193,0.033057757,0.030686473,0.03244001,0.032020606,0.032457124,0.025812166,0.02857281,0.02809348,0.026829114,0.030221019,0.031125693,0.03006681,0.032895982,0.031971052,0.032191124,0.030761,0.033361543,0.032130904,0.030909894,0.031800143,0.035640564
-0.025240514,0.02986517,0.029861702,0.02963675,0.030136624,0.031917684,0.03455104,0.032634728,0.028964166,0.02846413,0.02896291,0.03225899,0.033151124,0.02891383,0.028778404,0.02759961,0.034724258,0.027591249,0.034086443,0.029194983,0.02715989,0.028105231,0.029908778,0.029540123,0.030567242,0.03184583,0.03296974,0.02765829,0.03352342,0.032018468,0.03083127,0.03059918,0.028738184
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
deleted file mode 100644
index e15e21d2..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-6-probabilities.csv
+++ /dev/null
@@ -1,6 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.02990466,0.027794937,0.030200135,0.033223048,0.028840246,0.029801996,0.03408861,0.031201469,0.034209393,0.030435495,0.02688698,0.028210292,0.027972383,0.026755435,0.028162813,0.028899178,0.034516744,0.031520512,0.022439493,0.029979838,0.029862743,0.030803071,0.026144683,0.030880546,0.033226147,0.02804711,0.033052064,0.03126565,0.034700904,0.03395223,0.03138371,0.03145586,0.030181592
-0.02977129,0.027457537,0.028779993,0.030530233,0.0316698,0.027341226,0.034179598,0.033910844,0.027016249,0.027471438,0.030157587,0.029443314,0.031970903,0.02910385,0.032963887,0.028755877,0.03607527,0.030105237,0.031945523,0.031880952,0.031306166,0.029473325,0.026267372,0.027917987,0.031049537,0.024086313,0.035278693,0.02937533,0.028586993,0.035309106,0.03102215,0.028993413,0.030802958
-0.034256108,0.031212777,0.032661483,0.032562193,0.028868396,0.02445675,0.032526903,0.030098625,0.028534148,0.029497765,0.031607904,0.028765576,0.029720742,0.025040075,0.032955986,0.0282345,0.030974323,0.031867743,0.029543443,0.030924428,0.030652856,0.030405918,0.023620382,0.027121766,0.032386478,0.025952253,0.0347015,0.032217287,0.03453304,0.031794257,0.032801624,0.031243311,0.02825948
-0.032598216,0.029093012,0.028690943,0.026613811,0.028034385,0.029812267,0.029320821,0.027035141,0.033263125,0.02949445,0.028751273,0.027457636,0.032119725,0.030603293,0.0324208,0.028617961,0.03232818,0.02807421,0.03443888,0.03276456,0.033602335,0.030092882,0.032640934,0.027553203,0.03204509,0.028555753,0.027167292,0.032216128,0.028183337,0.03194732,0.033609107,0.029577114,0.03127676
-0.028709698,0.029625166,0.029874118,0.028837366,0.033504315,0.030749444,0.028552493,0.033821665,0.027835265,0.029518366,0.030826217,0.033214457,0.0324431,0.032692183,0.02955219,0.031034948,0.034116004,0.03191638,0.03192628,0.030853745,0.03129503,0.029935427,0.03172751,0.030749017,0.026323091,0.026236577,0.03182046,0.027901417,0.02581471,0.031797033,0.028829262,0.029471148,0.028495917
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
deleted file mode 100644
index 938c691e..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-bert-best-1-itemId-probabilities/sequifier-model-categorical-bert-best-1-7-probabilities.csv
+++ /dev/null
@@ -1,5 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.034894887,0.032275356,0.02996399,0.026949542,0.027929796,0.029617915,0.033669554,0.026341518,0.034484576,0.030552784,0.027683964,0.026792545,0.027649531,0.029394316,0.030769546,0.030085139,0.029768215,0.032148227,0.028801225,0.035756513,0.03274599,0.032206167,0.032878507,0.029650897,0.03140312,0.03026901,0.027987143,0.03128506,0.027762262,0.029794872,0.030969823,0.026882874,0.03063518
-0.026114888,0.027567258,0.027771655,0.029752698,0.02820829,0.031714298,0.030308545,0.031291846,0.033901554,0.032065872,0.02763642,0.027419902,0.033871558,0.028658088,0.0289209,0.030442988,0.03671656,0.027796298,0.028483596,0.030759042,0.029196337,0.031400263,0.026616627,0.028743278,0.032370545,0.030774299,0.02558757,0.029802978,0.033976573,0.03539298,0.028634975,0.033863362,0.034237936
-0.03136811,0.032610927,0.028663022,0.02766169,0.027849054,0.030900279,0.029014288,0.029312909,0.027264105,0.027272008,0.032112505,0.028268036,0.032192186,0.031883806,0.034192175,0.032434538,0.027403167,0.03600785,0.030057115,0.03440504,0.029654121,0.03388859,0.032026663,0.030806169,0.030339846,0.032073505,0.028873272,0.03086741,0.025530437,0.02776002,0.026841227,0.027687589,0.032778434
-0.026323346,0.02814542,0.029942133,0.029598957,0.030931026,0.030908601,0.028679792,0.033140037,0.031168977,0.028879615,0.028810753,0.029077675,0.03557963,0.030610638,0.02624658,0.029175146,0.03712324,0.034798477,0.025707724,0.029618949,0.030265389,0.029436663,0.026252486,0.029421646,0.030187503,0.029937852,0.03027628,0.027772948,0.03426053,0.033577643,0.02875072,0.033727318,0.031666294
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
deleted file mode 100644
index 3c45b01b..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.031594455,0.029204236,0.028718323,0.032224495,0.03337781,0.035745747,0.030596118,0.03137629,0.029209606,0.03408813,0.031277202,0.027920168,0.03155741,0.028168924,0.030145012,0.03245209,0.031797096,0.030348834,0.027226841,0.02847952,0.03093982,0.026849763,0.033478472,0.028779963,0.031998247,0.030514607,0.025465587,0.0325864,0.033844102,0.029311249,0.025630878,0.027453667,0.027638951
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
deleted file mode 100644
index 84d81025..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027228992,0.030756999,0.031926684,0.03520301,0.030777134,0.032955393,0.03167837,0.029797489,0.031220395,0.03152831,0.033776756,0.027809406,0.028926183,0.027881108,0.025017528,0.031061377,0.031814765,0.03505755,0.0288942,0.03108295,0.028965265,0.028036237,0.032550003,0.026335774,0.030168375,0.029126244,0.031778075,0.03171285,0.031987205,0.028292153,0.029634615,0.026793996,0.030224688
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
deleted file mode 100644
index 0aaba66d..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.027887529,0.031016335,0.030412342,0.03501881,0.03126559,0.032115854,0.03112963,0.0304681,0.031105764,0.031803418,0.0336682,0.02792235,0.027942816,0.02709955,0.025281327,0.031618994,0.032682337,0.034516595,0.029082842,0.03194304,0.029754547,0.02831144,0.032550737,0.026138999,0.029740384,0.029294712,0.030734966,0.031810347,0.033461045,0.028520357,0.028774502,0.026385749,0.03054078
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
deleted file mode 100644
index 9c566baf..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.032903668,0.028901948,0.030422168,0.02729401,0.029996833,0.031081805,0.029144669,0.029184328,0.028582947,0.030838203,0.02802886,0.03143278,0.0348301,0.03218194,0.036367875,0.029662728,0.028250722,0.026152506,0.029222546,0.027613314,0.03228014,0.03035405,0.029751362,0.03398891,0.032090932,0.030233484,0.027714087,0.029350067,0.028187223,0.031162726,0.03029153,0.03414756,0.02835399
-0.031707615,0.028367328,0.03251506,0.02744744,0.029457696,0.030506311,0.029133363,0.028074272,0.028364692,0.03026175,0.028753564,0.031309176,0.034676917,0.033179097,0.034860846,0.029045649,0.02786958,0.026573405,0.030232163,0.027840735,0.032212436,0.03031208,0.028782444,0.03392059,0.0323571,0.030005956,0.030180851,0.028813053,0.02663781,0.030342521,0.03171807,0.035299603,0.029240826
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
deleted file mode 100644
index c967ac22..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.028136749,0.03145643,0.032249775,0.035381794,0.031226981,0.033643033,0.031521108,0.02954274,0.030872017,0.033072893,0.033556398,0.027660204,0.030459512,0.027515223,0.027106846,0.030692818,0.030569497,0.032710854,0.027416352,0.030361334,0.029780567,0.027815046,0.033936046,0.027293991,0.03050649,0.027694564,0.028838307,0.03159304,0.03238606,0.02936506,0.029868718,0.027438898,0.028330646
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
deleted file mode 100644
index 9a40a897..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.032555487,0.026670558,0.035933323,0.028395625,0.032177225,0.030180544,0.029458955,0.029028589,0.029231155,0.03132255,0.030461755,0.031708784,0.03087897,0.035742663,0.030636476,0.028290126,0.02821839,0.027709417,0.033597182,0.028345522,0.02946698,0.028262552,0.028240243,0.032148767,0.031331882,0.029351693,0.032904956,0.028570952,0.025903815,0.028767973,0.032443482,0.031798936,0.03026446
-0.035266466,0.027003983,0.029438915,0.029231507,0.03491313,0.035511877,0.029532434,0.031505685,0.027652096,0.033847447,0.0306067,0.029304575,0.030613605,0.03229601,0.030978212,0.0330794,0.03178621,0.02973907,0.02895497,0.02695284,0.03035813,0.025822245,0.03114871,0.029226782,0.033070102,0.03168039,0.026520967,0.032253344,0.032956786,0.027481772,0.024973137,0.027664494,0.02862798
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
deleted file mode 100644
index 9fc8ce15..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.030240778,0.026598437,0.036707304,0.030852366,0.030847387,0.030217325,0.028302671,0.027445653,0.028811673,0.03210606,0.032096457,0.030536423,0.032865256,0.031971786,0.031098066,0.028391903,0.030326098,0.026407817,0.031497795,0.028856413,0.031845562,0.02874827,0.030271817,0.030664615,0.031373702,0.028012546,0.032444347,0.029172558,0.025456056,0.028003585,0.032420244,0.035377927,0.030031078
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
deleted file mode 100644
index d3fd32e1..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-itemId-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129
-0.03120083,0.03140158,0.02597218,0.030870566,0.030339655,0.032766707,0.029818222,0.031150384,0.029739713,0.03175302,0.029487183,0.029177235,0.032368988,0.026637957,0.03345298,0.032503065,0.031507123,0.029519858,0.026424674,0.029624622,0.03226006,0.030312635,0.032872517,0.02997535,0.030994944,0.030409798,0.02527944,0.03171313,0.034694858,0.031350695,0.027044604,0.02949311,0.02788231
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
deleted file mode 100644
index bc5e64b0..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-0-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.074083805,0.087974995,0.075273834,0.06646733,0.07345789,0.06847239,0.0818435,0.08279163,0.07455243,0.08277229,0.07101856,0.07774641,0.08354494
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
deleted file mode 100644
index b27b3ac2..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-1-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.084869295,0.07880388,0.08157886,0.06951565,0.08075639,0.07072125,0.0760694,0.068276174,0.07958265,0.07747237,0.073025174,0.07977048,0.07955835
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
deleted file mode 100644
index 83d9348d..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-2-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.083191805,0.0784117,0.08124768,0.07007233,0.07956042,0.06936452,0.07542335,0.067135476,0.07857058,0.08090579,0.071468234,0.081839085,0.08280907
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
deleted file mode 100644
index 6d556ab8..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-3-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07036531,0.08049916,0.07311549,0.080371074,0.07224095,0.081713885,0.080874875,0.08775907,0.07437307,0.07501077,0.080234654,0.071844436,0.07159725
-0.07462252,0.07882556,0.076731205,0.08212352,0.0746616,0.08412205,0.08021181,0.08402447,0.07401236,0.071274504,0.08248981,0.069406666,0.067493975
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
deleted file mode 100644
index 7894c70f..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-4-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07968078,0.08148541,0.07910197,0.06712172,0.07808477,0.06944092,0.076500565,0.069036186,0.0806076,0.080700286,0.075026445,0.08142645,0.081786945
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
deleted file mode 100644
index f8bde07a..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-5-probabilities.csv
+++ /dev/null
@@ -1,3 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07887929,0.07922841,0.0837173,0.07808471,0.08080305,0.08518249,0.0778711,0.080227315,0.073275805,0.06477759,0.08631688,0.063998096,0.067637995
-0.075657144,0.091537185,0.07574736,0.070048064,0.07345481,0.07445074,0.086469546,0.08901579,0.06910504,0.07258692,0.07215753,0.07019448,0.079575315
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
deleted file mode 100644
index 15fe6887..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-6-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.07799491,0.078715876,0.08464348,0.07995696,0.07754624,0.08058426,0.07884206,0.075935386,0.074297816,0.07091199,0.08250889,0.070520476,0.067541696
diff --git a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv b/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
deleted file mode 100644
index 59e7177d..00000000
--- a/tests/resources/target_outputs/probabilities/sequifier-model-categorical-multitarget-5-best-3-supCat1-probabilities/sequifier-model-categorical-multitarget-5-best-3-7-probabilities.csv
+++ /dev/null
@@ -1,2 +0,0 @@
-[unknown],[other],[mask],0,1,2,3,4,5,6,7,8,9
-0.069228604,0.08208278,0.06885877,0.0737444,0.06933344,0.069567144,0.08054654,0.081645325,0.077265635,0.08891717,0.06935114,0.08529186,0.08416724
diff --git a/tests/unit/data/empty.parquet b/tests/unit/data/empty.parquet
deleted file mode 100644
index e69de29b..00000000
diff --git a/tests/unit/io/test_batch.py b/tests/unit/io/test_batch.py
deleted file mode 100644
index e8b24ff9..00000000
--- a/tests/unit/io/test_batch.py
+++ /dev/null
@@ -1,32 +0,0 @@
-import pytest
-import torch
-
-from sequifier.io.batch import SequifierBatch
-
-
-def _metadata():
-    valid_mask = torch.ones(1, 2, dtype=torch.bool)
-    return {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-
-def test_sequifier_batch_requires_metadata_masks():
-    with pytest.raises(ValueError, match="target_valid_mask"):
-        SequifierBatch(
-            inputs={"item": torch.ones(1, 2)},
-            targets={"item": torch.ones(1, 2)},
-            metadata={"attention_valid_mask": torch.ones(1, 2, dtype=torch.bool)},
-        )
-
-
-def test_sequifier_batch_does_not_support_tuple_unpacking():
-    batch = SequifierBatch(
-        inputs={"item": torch.ones(1, 2)},
-        targets={"item": torch.ones(1, 2)},
-        metadata=_metadata(),
-    )
-
-    with pytest.raises(TypeError):
-        tuple(batch)  # type: ignore
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
deleted file mode 100644
index 6be5d40e..00000000
--- a/tests/unit/io/test_sequifier_dataset_from_folder_parquet_lazy.py
+++ /dev/null
@@ -1,372 +0,0 @@
-import json
-from unittest.mock import MagicMock, patch
-
-import polars as pl
-import pytest
-import torch
-
-from sequifier.helpers import ModelWindowView, StoredWindowLayout
-from sequifier.io.sequifier_dataset_from_folder_parquet_lazy import (
-    SequifierDatasetFromFolderParquetLazy,
-)
-
-CONTEXT_LENGTH = 2
-FUTURE_CAPACITY = 1
-STORED_WIDTH = CONTEXT_LENGTH + FUTURE_CAPACITY
-
-
-def _folder_metadata(total_samples, batch_files):
-    return {
-        "total_samples": total_samples,
-        "batch_files": batch_files,
-        "stored_context_width": STORED_WIDTH,
-        "max_target_offset": FUTURE_CAPACITY,
-        "stored_window_layout_version": 2,
-    }
-
-
-@pytest.fixture
-def mock_config():
-    config = MagicMock()
-    config.project_root = "."
-    config.seed = 42
-    config.storage_layout = StoredWindowLayout(
-        stored_context_width=STORED_WIDTH, max_target_offset=FUTURE_CAPACITY, version=2
-    )
-    config.window_view = ModelWindowView(
-        context_length=CONTEXT_LENGTH, objective="causal", target_offset=1
-    )
-    config.column_types = {"item": "Float64"}
-    config.training_spec.batch_size = 5
-    config.training_spec.num_workers = 0
-    config.training_spec.sampling_strategy = "exact"
-    config.training_spec.training_objective = "causal"
-    config.input_columns = ["item"]
-    config.target_columns = ["item"]
-
-    return config
-
-
-@pytest.fixture
-def dataset_path(tmp_path):
-    data_dir = tmp_path / "parquet_data"
-    data_dir.mkdir()
-
-    # Layout matches the long-format extraction logic
-    schema = {
-        "sequenceId": pl.Int64,
-        "subsequenceId": pl.Int64,
-        "startItemPosition": pl.Int64,
-        "leftPadLength": pl.Int64,
-        "inputCol": pl.String,
-        "2": pl.Float64,
-        "1": pl.Float64,
-        "0": pl.Float64,
-    }
-
-    # Populate 4 files with distinct row markers so sharding can be observed.
-    batch_files = []
-    for i in range(1, 5):
-        filename = f"file_{i}.parquet"
-        rows = []
-        for s in range(10):
-            marker = i * 100 + s
-            rows.append(
-                (
-                    s,
-                    0,
-                    s * 2,
-                    0,
-                    "item",
-                    float(marker),
-                    float(marker + 1),
-                    float(marker + 2),
-                )
-            )
-
-        df = pl.DataFrame(rows, schema=schema, orient="row")
-        df.write_parquet(data_dir / filename)
-        batch_files.append({"path": filename, "samples": 10})
-
-    metadata = _folder_metadata(40, batch_files)
-
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    return str(data_dir)
-
-
-def _input_markers(batches):
-    return torch.cat([batch.inputs["item"][:, 0] for batch in batches]).tolist()
-
-
-def _sample_valid_mask(batches):
-    return torch.cat(
-        [batch.metadata["sample_valid_mask"] for batch in batches]
-    ).tolist()
-
-
-def test_initialization(mock_config, dataset_path):
-    """Metadata-backed batch length."""
-    dataset = SequifierDatasetFromFolderParquetLazy(dataset_path, mock_config)
-
-    # 40 total samples / batch size of 5 = 8 batches
-    assert len(dataset) == 8
-    assert dataset.total_samples == 40
-    assert dataset.target_samples == 40
-
-
-def test_iteration_yields_correct_batches(mock_config, dataset_path):
-    """Structured tensors from file iteration."""
-    dataset = SequifierDatasetFromFolderParquetLazy(
-        dataset_path, mock_config, shuffle=False
-    )
-
-    batches = list(dataset)
-    assert len(batches) == 8  # 40 samples / batch_size 5
-
-    # Check structural integrity of first batch
-    batch = batches[0]
-    seq_batch = batch.inputs
-    tgt_batch = batch.targets
-    assert "item" in seq_batch
-    assert "item" in tgt_batch
-    assert isinstance(seq_batch["item"], torch.Tensor)
-    assert seq_batch["item"].shape == (5, 2)  # batch_size=5, seq_len=2
-    assert tgt_batch["item"].shape == (5, 2)
-
-
-def test_iteration_attaches_explicit_padding_masks(mock_config, tmp_path):
-    data_dir = tmp_path / "parquet_masks"
-    data_dir.mkdir()
-
-    schema = {
-        "sequenceId": pl.Int64,
-        "subsequenceId": pl.Int64,
-        "startItemPosition": pl.Int64,
-        "leftPadLength": pl.Int64,
-        "inputCol": pl.String,
-        "2": pl.Float64,
-        "1": pl.Float64,
-        "0": pl.Float64,
-    }
-    rows = [
-        (0, 0, 0, 0, "item", 0.0, 1.0, 2.0),
-        (1, 0, 0, 1, "item", 0.0, 0.0, 2.0),
-        (2, 0, 0, 2, "item", 0.0, 0.0, 2.0),
-        (3, 0, 0, 3, "item", 0.0, 0.0, 2.0),
-        (4, 0, 0, 0, "item", 0.0, 1.0, 2.0),
-    ]
-    pl.DataFrame(rows, schema=schema, orient="row").write_parquet(
-        data_dir / "file.parquet"
-    )
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(
-            _folder_metadata(5, [{"path": "file.parquet", "samples": 5}]),
-            f,
-        )
-
-    dataset = SequifierDatasetFromFolderParquetLazy(
-        str(data_dir), mock_config, shuffle=False
-    )
-    batch = next(iter(dataset))
-    metadata_batch = batch.metadata
-
-    assert torch.equal(
-        metadata_batch["attention_valid_mask"],
-        torch.tensor(
-            [
-                [True, True],
-                [False, True],
-                [False, False],
-                [False, False],
-                [True, True],
-            ]
-        ),
-    )
-    assert torch.equal(
-        metadata_batch["target_valid_mask"],
-        torch.tensor(
-            [
-                [True, True],
-                [True, True],
-                [False, True],
-                [False, False],
-                [True, True],
-            ]
-        ),
-    )
-
-
-@patch("torch.distributed.is_initialized", return_value=True)
-@patch("torch.distributed.get_rank", return_value=0)
-@patch("torch.distributed.get_world_size", return_value=2)
-def test_distributed_sharding(mock_ws, mock_rank, mock_init, mock_config, dataset_path):
-    """Distributed file sharding."""
-    dataset = SequifierDatasetFromFolderParquetLazy(
-        dataset_path, mock_config, shuffle=False
-    )
-
-    # World size = 2, Total files = 4
-    # Rank 0 gets file index 0 and 2 (file_1, file_3) -> 20 samples total -> 4 batches
-    batches = list(dataset)
-    assert len(batches) == 4
-
-    # Verify input mapping structures
-    for batch in batches:
-        assert batch.inputs["item"].shape[0] == 5
-
-    assert _input_markers(batches) == [
-        *[float(100 + i) for i in range(10)],
-        *[float(300 + i) for i in range(10)],
-    ]
-    assert _sample_valid_mask(batches) == [True] * 20
-
-
-def test_exact_strategy_uneven_files_exception(mock_config, tmp_path):
-    """Exact distributed mode rejects uneven ranks."""
-    data_dir = tmp_path / "uneven_parquet_data"
-    data_dir.mkdir()
-
-    # Write asymmetrical sample quotas across files
-    batch_files = [
-        {"path": "file_1.parquet", "samples": 15},
-        {"path": "file_2.parquet", "samples": 10},
-    ]
-    metadata = _folder_metadata(25, batch_files)
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    with patch("torch.distributed.is_initialized", return_value=True), patch(
-        "torch.distributed.get_world_size", return_value=2
-    ):
-        with pytest.raises(Exception) as exc_info:
-            SequifierDatasetFromFolderParquetLazy(str(data_dir), mock_config)
-
-        assert "different number of samples per rank/GPU" in str(exc_info.value)
-
-
-@patch("torch.distributed.get_rank", return_value=1)
-def test_oversampling_strategy(mock_rank, mock_config, tmp_path):
-    """Oversampling pads short ranks."""
-    data_dir = tmp_path / "oversample_parquet_data"
-    data_dir.mkdir()
-
-    schema = {
-        "sequenceId": pl.Int64,
-        "subsequenceId": pl.Int64,
-        "startItemPosition": pl.Int64,
-        "leftPadLength": pl.Int64,
-        "inputCol": pl.String,
-        "2": pl.Float64,
-        "1": pl.Float64,
-        "0": pl.Float64,
-    }
-
-    # File 1 has 15 rows, File 2 has 10 rows
-    for i, num_rows in [(1, 15), (2, 10)]:
-        rows = [
-            (
-                s,
-                0,
-                s * 2,
-                0,
-                "item",
-                float(i * 100 + s),
-                float(i * 100 + s + 1),
-                float(i * 100 + s + 2),
-            )
-            for s in range(num_rows)
-        ]
-        pl.DataFrame(rows, schema=schema, orient="row").write_parquet(
-            data_dir / f"file_{i}.parquet"
-        )
-
-    batch_files = [
-        {"path": "file_1.parquet", "samples": 15},
-        {"path": "file_2.parquet", "samples": 10},
-    ]
-    metadata = _folder_metadata(25, batch_files)
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    mock_config.training_spec.sampling_strategy = "oversampling"
-
-    with patch("torch.distributed.is_initialized", return_value=True), patch(
-        "torch.distributed.get_world_size", return_value=2
-    ):
-        dataset = SequifierDatasetFromFolderParquetLazy(
-            str(data_dir), mock_config, shuffle=False
-        )
-        # Should match max(15, 10)
-        assert dataset.target_samples == 15
-        assert len(dataset) == 3
-
-        batches = list(dataset)
-        assert len(batches) == 3
-        assert _input_markers(batches) == [
-            *[float(200 + i) for i in range(10)],
-            *[float(200 + i) for i in range(5)],
-        ]
-        assert _sample_valid_mask(batches) == [True] * 10 + [False] * 5
-
-
-@patch("torch.distributed.get_rank", return_value=0)
-def test_undersampling_strategy(mock_rank, mock_config, tmp_path):
-    """Undersampling truncates long ranks."""
-    data_dir = tmp_path / "undersample_parquet_data"
-    data_dir.mkdir()
-
-    schema = {
-        "sequenceId": pl.Int64,
-        "subsequenceId": pl.Int64,
-        "startItemPosition": pl.Int64,
-        "leftPadLength": pl.Int64,
-        "inputCol": pl.String,
-        "2": pl.Float64,
-        "1": pl.Float64,
-        "0": pl.Float64,
-    }
-
-    for i, num_rows in [(1, 15), (2, 10)]:
-        rows = [
-            (
-                s,
-                0,
-                s * 2,
-                0,
-                "item",
-                float(i * 100 + s),
-                float(i * 100 + s + 1),
-                float(i * 100 + s + 2),
-            )
-            for s in range(num_rows)
-        ]
-        pl.DataFrame(rows, schema=schema, orient="row").write_parquet(
-            data_dir / f"file_{i}.parquet"
-        )
-
-    batch_files = [
-        {"path": "file_1.parquet", "samples": 15},
-        {"path": "file_2.parquet", "samples": 10},
-    ]
-    metadata = _folder_metadata(25, batch_files)
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    mock_config.training_spec.sampling_strategy = "undersampling"
-
-    with patch("torch.distributed.is_initialized", return_value=True), patch(
-        "torch.distributed.get_world_size", return_value=2
-    ):
-        dataset = SequifierDatasetFromFolderParquetLazy(
-            str(data_dir), mock_config, shuffle=False
-        )
-        # Should match min(15, 10)
-        assert dataset.target_samples == 10
-        assert len(dataset) == 2
-
-        batches = list(dataset)
-        assert len(batches) == 2
-        assert _input_markers(batches) == [float(100 + i) for i in range(10)]
-        assert _sample_valid_mask(batches) == [True] * 10
diff --git a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py b/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
deleted file mode 100644
index 6f3c8c98..00000000
--- a/tests/unit/io/test_sequifier_dataset_from_folder_pt_lazy.py
+++ /dev/null
@@ -1,421 +0,0 @@
-import json
-from unittest.mock import MagicMock, patch
-
-import pytest
-import torch
-
-from sequifier.helpers import ModelWindowView, StoredWindowLayout
-from sequifier.io.sequifier_dataset_from_folder_pt_lazy import (
-    SequifierDatasetFromFolderPtLazy,
-)
-
-CONTEXT_LENGTH = 5
-FUTURE_CAPACITY = 1
-STORED_WIDTH = CONTEXT_LENGTH + FUTURE_CAPACITY
-
-
-def _folder_metadata(total_samples, batch_files):
-    return {
-        "total_samples": total_samples,
-        "batch_files": batch_files,
-        "stored_context_width": STORED_WIDTH,
-        "max_target_offset": FUTURE_CAPACITY,
-        "stored_window_layout_version": 2,
-    }
-
-
-# --- Fixtures ---
-@pytest.fixture
-def mock_config(tmp_path):
-    """Mock dataset config."""
-    config = MagicMock()
-    config.project_root = str(tmp_path)
-    config.training_spec.batch_size = 5
-    config.training_spec.sampling_strategy = "exact"
-    config.training_spec.num_workers = 0
-    config.seed = 42
-    config.storage_layout = StoredWindowLayout(
-        stored_context_width=STORED_WIDTH, max_target_offset=FUTURE_CAPACITY, version=2
-    )
-    config.window_view = ModelWindowView(
-        context_length=CONTEXT_LENGTH, objective="causal", target_offset=1
-    )
-    config.input_columns = ["col1"]
-    config.target_columns = ["col1", "tgt1"]
-    config.training_spec.training_objective = "causal"
-    return config
-
-
-@pytest.fixture
-def dataset_path(tmp_path):
-    """Sets up a temporary data directory with metadata.json."""
-    data_dir = tmp_path / "data"
-    data_dir.mkdir()
-
-    # Create dummy metadata.json
-    # We define 4 files with 10 samples each = 40 total samples
-    metadata = _folder_metadata(
-        40,
-        [
-            {"path": "file1.pt", "samples": 10},
-            {"path": "file2.pt", "samples": 10},
-            {"path": "file3.pt", "samples": 10},
-            {"path": "file4.pt", "samples": 10},
-        ],
-    )
-
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    return str(data_dir)
-
-
-@pytest.fixture
-def mock_torch_load():
-    """Mocks torch.load to return dummy tensors matching the preprocessed format."""
-    with patch("torch.load") as mock_load:
-
-        def side_effect(path, map_location, weights_only):
-            dummy_seq = {"col1": torch.ones((10, 6)), "tgt1": torch.zeros((10, 6))}
-            left_pad_lengths = torch.zeros(10, dtype=torch.int64)
-
-            return (
-                dummy_seq,
-                None,
-                None,
-                None,
-                left_pad_lengths,
-            )
-
-        mock_load.side_effect = side_effect
-        yield mock_load
-
-
-# --- Tests ---
-
-
-def test_initialization(mock_config, dataset_path):
-    """Metadata-backed batch length."""
-    dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config)
-
-    # 40 total samples / batch size of 5 = 8 batches
-    assert len(dataset) == 8
-    assert len(dataset.batch_files_info) == 4
-    assert dataset.total_samples == 40
-
-
-def test_iteration_yields_correct_batches(mock_config, dataset_path, mock_torch_load):
-    """Tensor slices from file iteration."""
-    dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config, shuffle=False)
-
-    # Consume the generator
-    batches = list(dataset)
-
-    # 4 files * 10 samples = 40 total samples. 40 / 5 batch_size = 8 batches
-    assert len(batches) == 8
-
-    # Each file has 10 samples, so torch.load should be called 4 times
-    assert mock_torch_load.call_count == 4
-
-    # Verify the structure of a yielded batch
-    batch = batches[0]
-    seq_dict = batch.inputs
-    tgt_dict = batch.targets
-
-    assert "col1" in seq_dict, f"{seq_dict = }"
-
-    assert "tgt1" in tgt_dict, f"{tgt_dict = }"
-    # Check that batch size and sequence length truncation works properly
-    assert seq_dict["col1"].shape == (5, 5)
-    assert tgt_dict["tgt1"].shape == (5, 5)
-
-
-def test_iteration_attaches_explicit_padding_masks(mock_config, dataset_path):
-    with patch("torch.load") as mock_load:
-
-        def side_effect(path, map_location, weights_only):
-            dummy_seq = {
-                "col1": torch.ones((10, 6)),
-                "tgt1": torch.ones((10, 6)),
-            }
-            left_pad_lengths = torch.tensor([0, 1, 2, 3, 4, 5, 0, 0, 0, 0])
-            return (
-                dummy_seq,
-                torch.arange(10),
-                torch.zeros(10, dtype=torch.int64),
-                torch.zeros(10, dtype=torch.int64),
-                left_pad_lengths,
-            )
-
-        mock_load.side_effect = side_effect
-        dataset = SequifierDatasetFromFolderPtLazy(
-            dataset_path, mock_config, shuffle=False
-        )
-        batch = next(iter(dataset))
-        metadata_dict = batch.metadata
-
-    assert torch.equal(
-        metadata_dict["attention_valid_mask"],
-        torch.tensor(
-            [
-                [True, True, True, True, True],
-                [False, True, True, True, True],
-                [False, False, True, True, True],
-                [False, False, False, True, True],
-                [False, False, False, False, True],
-            ]
-        ),
-    )
-    assert torch.equal(
-        metadata_dict["target_valid_mask"],
-        torch.tensor(
-            [
-                [True, True, True, True, True],
-                [True, True, True, True, True],
-                [False, True, True, True, True],
-                [False, False, True, True, True],
-                [False, False, False, True, True],
-            ]
-        ),
-    )
-
-
-@patch("torch.distributed.is_initialized", return_value=True)
-@patch("torch.distributed.get_world_size", return_value=2)
-@patch("torch.distributed.get_rank", return_value=0)
-def test_distributed_sharding(
-    mock_rank, mock_ws, mock_init, mock_config, dataset_path, mock_torch_load
-):
-    """Distributed file sharding."""
-    dataset = SequifierDatasetFromFolderPtLazy(dataset_path, mock_config, shuffle=False)
-
-    # World size = 2, Total files = 4
-    # Rank 0 gets file index 0 and 2 (file1.pt, file3.pt) -> Total 20 samples
-    # 20 samples / 5 batch_size = 4 expected batches
-    assert len(dataset) == 4
-
-    batches = list(dataset)
-
-    assert len(batches) == 4
-    assert mock_torch_load.call_count == 2
-
-    # Verify it loaded the correct specific files
-    loaded_files = [call.args[0] for call in mock_torch_load.call_args_list]
-    assert any("file1.pt" in f for f in loaded_files)
-    assert any("file3.pt" in f for f in loaded_files)
-    assert not any("file2.pt" in f for f in loaded_files)
-
-
-@patch("sequifier.io.sequifier_dataset_from_folder_pt_lazy.get_worker_info")
-def test_dataloader_worker_sharding_continuous_boundaries(
-    mock_worker_info, mock_config, tmp_path
-):
-    """Worker boundaries can start and stop inside PT files."""
-    data_dir = tmp_path / "data_uneven_worker"
-    data_dir.mkdir()
-    file_specs = [
-        ("file1.pt", 7, 0),
-        ("file2.pt", 11, 100),
-        ("file3.pt", 13, 200),
-        ("file4.pt", 9, 300),
-    ]
-    metadata = _folder_metadata(
-        sum(samples for _, samples, _ in file_specs),
-        [{"path": path, "samples": samples} for path, samples, _ in file_specs],
-    )
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    mock_info = MagicMock()
-    mock_info.id = 1
-    mock_info.num_workers = 2
-    mock_worker_info.return_value = mock_info
-
-    mock_config.training_spec.num_workers = 2
-
-    file_offsets = {path: (samples, offset) for path, samples, offset in file_specs}
-
-    def load_uneven_file(path, map_location, weights_only):
-        file_name = path.split("/")[-1]
-        samples, offset = file_offsets[file_name]
-        values = torch.arange(offset, offset + samples * STORED_WIDTH).reshape(
-            samples, STORED_WIDTH
-        )
-        return (
-            {"col1": values, "tgt1": values.clone()},
-            None,
-            None,
-            None,
-            torch.zeros(samples, dtype=torch.int64),
-        )
-
-    with patch("torch.load", side_effect=load_uneven_file) as mock_load:
-        dataset = SequifierDatasetFromFolderPtLazy(
-            str(data_dir), mock_config, shuffle=False
-        )
-        batches = list(dataset)
-
-    assert len(batches) == 4
-    assert mock_load.call_count == 2
-
-    loaded_files = [call.args[0] for call in mock_load.call_args_list]
-    assert any("file3.pt" in f for f in loaded_files)
-    assert any("file4.pt" in f for f in loaded_files)
-    assert not any("file1.pt" in f for f in loaded_files)
-    assert not any("file2.pt" in f for f in loaded_files)
-
-    sample_ids = []
-    for batch in batches:
-        sample_ids.extend(batch.inputs["col1"][:, 0].tolist())
-
-    assert sample_ids == [
-        200 + 2 * STORED_WIDTH,
-        200 + 3 * STORED_WIDTH,
-        200 + 4 * STORED_WIDTH,
-        200 + 5 * STORED_WIDTH,
-        200 + 6 * STORED_WIDTH,
-        200 + 7 * STORED_WIDTH,
-        200 + 8 * STORED_WIDTH,
-        200 + 9 * STORED_WIDTH,
-        200 + 10 * STORED_WIDTH,
-        200 + 11 * STORED_WIDTH,
-        200 + 12 * STORED_WIDTH,
-        300,
-        300 + STORED_WIDTH,
-        300 + 2 * STORED_WIDTH,
-        300 + 3 * STORED_WIDTH,
-        300 + 4 * STORED_WIDTH,
-        300 + 5 * STORED_WIDTH,
-        300 + 6 * STORED_WIDTH,
-        300 + 7 * STORED_WIDTH,
-        300 + 8 * STORED_WIDTH,
-    ]
-
-
-@patch("torch.distributed.is_initialized", return_value=True)
-@patch("torch.distributed.get_world_size", return_value=2)
-@patch("torch.distributed.get_rank", return_value=0)
-def test_exact_strategy_uneven_files_exception(
-    mock_rank, mock_ws, mock_init, mock_config, tmp_path
-):
-    """Exact mode rejects uneven rank samples."""
-
-    data_dir = tmp_path / "data_uneven"
-    data_dir.mkdir()
-
-    # Rank 0 will get file1 (10) + file3 (10) = 20 samples
-    # Rank 1 will get file2 (10) + file4 (5) = 15 samples
-    metadata = _folder_metadata(
-        35,
-        [
-            {"path": "file1.pt", "samples": 10},
-            {"path": "file2.pt", "samples": 10},
-            {"path": "file3.pt", "samples": 10},
-            {"path": "file4.pt", "samples": 5},
-        ],
-    )
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    mock_config.training_spec.sampling_strategy = "exact"
-
-    # The dataset initialization calls _get_target_samples(), which should raise the Exception
-    with pytest.raises(Exception) as exc_info:
-        SequifierDatasetFromFolderPtLazy(str(data_dir), mock_config)
-
-    error_msg = str(exc_info.value)
-
-    # Assert the core error text is present
-    assert "Found 2 different number of samples per rank/GPU" in error_msg
-
-
-@patch("torch.distributed.is_initialized", return_value=True)
-@patch("torch.distributed.get_world_size", return_value=2)
-@patch("torch.distributed.get_rank", return_value=1)
-def test_oversampling_strategy(
-    mock_rank, mock_ws, mock_init, mock_config, tmp_path, mock_torch_load
-):
-    """Oversampling loops short ranks."""
-    data_dir = tmp_path / "data_oversample"
-    data_dir.mkdir()
-
-    # Rank 0 gets file1 (10) + file3 (5) = 15 samples
-    # Rank 1 gets file2 (10) = 10 samples
-    metadata = _folder_metadata(
-        25,
-        [
-            {"path": "file1.pt", "samples": 10},
-            {"path": "file2.pt", "samples": 10},
-            {"path": "file3.pt", "samples": 5},
-        ],
-    )
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    mock_config.training_spec.sampling_strategy = "oversampling"
-    mock_config.training_spec.batch_size = 5
-    mock_config.training_spec.num_workers = 0
-
-    dataset = SequifierDatasetFromFolderPtLazy(
-        str(data_dir), mock_config, shuffle=False
-    )
-
-    # Max samples across ranks is 15. Rank 1 must pad its 10 samples up to 15.
-    assert dataset.target_samples == 15
-    assert len(dataset) == 3  # 15 total samples / batch_size 5
-
-    batches = list(dataset)
-    assert len(batches) == 3
-
-    # Rank 1 only has file2 assigned. To get 15 samples, it must load file2,
-    # run out of data, loop back, and load file2 again.
-    assert mock_torch_load.call_count == 2
-    loaded_files = [call.args[0] for call in mock_torch_load.call_args_list]
-    assert "file2.pt" in loaded_files[0]
-    assert "file2.pt" in loaded_files[1]
-
-
-@patch("torch.distributed.is_initialized", return_value=True)
-@patch("torch.distributed.get_world_size", return_value=2)
-@patch("torch.distributed.get_rank", return_value=0)
-def test_undersampling_strategy(
-    mock_rank, mock_ws, mock_init, mock_config, tmp_path, mock_torch_load
-):
-    """Undersampling truncates heavy ranks."""
-    data_dir = tmp_path / "data_undersample"
-    data_dir.mkdir()
-
-    # Rank 0 gets file1 (10) + file3 (5) = 15 samples
-    # Rank 1 gets file2 (10) = 10 samples
-    metadata = _folder_metadata(
-        25,
-        [
-            {"path": "file1.pt", "samples": 10},
-            {"path": "file2.pt", "samples": 10},
-            {"path": "file3.pt", "samples": 5},
-        ],
-    )
-    with open(data_dir / "metadata.json", "w") as f:
-        json.dump(metadata, f)
-
-    mock_config.training_spec.sampling_strategy = "undersampling"
-    mock_config.training_spec.batch_size = 5
-    mock_config.training_spec.num_workers = 0
-
-    dataset = SequifierDatasetFromFolderPtLazy(
-        str(data_dir), mock_config, shuffle=False
-    )
-
-    # Min samples across ranks is 10. Rank 0 must truncate its 15 samples down to 10.
-    assert dataset.target_samples == 10
-    assert len(dataset) == 2  # 10 total samples / batch_size 5
-
-    batches = list(dataset)
-    assert len(batches) == 2
-
-    # Rank 0 should only need file1 (10 samples) to meet its truncated quota of 10.
-    # It should never attempt to load file3.
-    assert mock_torch_load.call_count == 1
-    loaded_files = [call.args[0] for call in mock_torch_load.call_args_list]
-    assert "file1.pt" in loaded_files[0]
-    assert not any("file3.pt" in f for f in loaded_files)
diff --git a/tests/unit/test_helpers.py b/tests/unit/test_helpers.py
deleted file mode 100644
index 360f04df..00000000
--- a/tests/unit/test_helpers.py
+++ /dev/null
@@ -1,617 +0,0 @@
-from types import SimpleNamespace
-
-import polars as pl
-import pytest
-import torch
-
-from sequifier.helpers import (
-    ModelWindowView,
-    StoredWindowLayout,
-    apply_bert_masking,
-    build_valid_mask,
-    construct_index_maps,
-    numpy_to_pytorch,
-    resolve_window_view,
-)
-from sequifier.special_tokens import SPECIAL_TOKEN_IDS
-
-
-def test_construct_index_maps_string():
-    """Reverse string ID map."""
-    id_maps: dict[str, dict[str | int, int]] = {"itemId": {"apple": 3, "banana": 4}}
-    target_cols = ["itemId"]
-
-    result = construct_index_maps(id_maps, target_cols, map_to_id=True)
-
-    # Check standard reversal
-    assert result["itemId"][3] == "apple"
-    assert result["itemId"][4] == "banana"
-
-    # Check the special 0 index for strings
-    assert result["itemId"][0] == "[unknown]"
-    assert result["itemId"][1] == "[other]"
-    assert result["itemId"][2] == "[mask]"
-
-
-def test_construct_index_maps_integer():
-    """Reverse int ID map with sentinels."""
-    id_maps: dict[str, dict[str | int, int]] = {"storeId": {100: 3, 101: 4}}
-    target_cols = ["storeId"]
-
-    result = construct_index_maps(id_maps, target_cols, map_to_id=True)
-
-    assert result["storeId"][3] == 100
-    assert result["storeId"][4] == 101
-
-    assert result["storeId"][0] == "[unknown]"
-    assert result["storeId"][1] == "[other]"
-    assert result["storeId"][2] == "[mask]"
-
-
-def test_construct_index_maps_flag_false():
-    """Empty maps when map_to_id is false."""
-    id_maps: dict[str, dict[str | int, int]] = {"itemId": {"a": 1}}
-    result = construct_index_maps(id_maps, ["itemId"], map_to_id=False)
-    assert result == {}
-
-
-def test_numpy_to_pytorch_shapes_and_shifting():
-    """Check causal input and shifted target tensors."""
-
-    data = pl.DataFrame(
-        {
-            "sequenceId": [1, 2],
-            "subsequenceId": [0, 0],
-            "startItemPosition": [0, 0],
-            "leftPadLength": [0, 0],
-            "inputCol": ["A", "A"],
-            "3": [10, 11],
-            "2": [20, 21],
-            "1": [30, 31],
-            "0": [40, 41],
-        }
-    )
-
-    column_types = {"A": torch.float32}
-    all_columns = ["A"]
-    resolved_view = resolve_window_view(
-        StoredWindowLayout(stored_context_width=4, max_target_offset=1, version=2),
-        ModelWindowView(context_length=3, objective="causal", target_offset=1),
-    )
-
-    tensors, metadata = numpy_to_pytorch(
-        data,
-        column_types,
-        all_columns,
-        resolved_view,
-    )
-
-    assert "A" in tensors
-    assert "A_target" in tensors
-    assert torch.equal(
-        metadata["attention_valid_mask"],
-        torch.tensor([[True, True, True], [True, True, True]]),
-    )
-    assert torch.equal(
-        metadata["target_valid_mask"],
-        torch.tensor([[True, True, True], [True, True, True]]),
-    )
-
-    expected_input = torch.tensor([[10, 20, 30], [11, 21, 31]], dtype=torch.float32)
-    assert torch.equal(tensors["A"], expected_input)
-
-    expected_target = torch.tensor([[20, 30, 40], [21, 31, 41]], dtype=torch.float32)
-    assert torch.equal(tensors["A_target"], expected_target)
-
-
-def test_numpy_to_pytorch_dtypes():
-    """Check column_types controls tensor dtypes."""
-
-    # Note: numpy_to_pytorch filters by inputCol, so we need to handle how
-    # it selects rows. It assumes all rows matching "inputCol" have valid data
-    # for that type.
-
-    # We'll test one type at a time to avoid schema conflicts in the Polars DF creation
-    # (Polars columns usually have a single type).
-
-    # Case 1: Integer
-    data_int = pl.DataFrame(
-        {
-            "sequenceId": [1],
-            "subsequenceId": [0],
-            "startItemPosition": [0],
-            "leftPadLength": [0],
-            "inputCol": ["int_col"],
-            "1": [10],
-            "0": [20],
-        }
-    )
-    tensors_int, _ = numpy_to_pytorch(
-        data_int,
-        {"int_col": torch.int64},
-        ["int_col"],
-        resolve_window_view(
-            StoredWindowLayout(stored_context_width=2, max_target_offset=1, version=2),
-            ModelWindowView(context_length=1, objective="causal", target_offset=1),
-        ),
-    )
-    assert tensors_int["int_col"].dtype == torch.int64
-
-    # Case 2: Float
-    data_float = pl.DataFrame(
-        {
-            "sequenceId": [1],
-            "subsequenceId": [0],
-            "startItemPosition": [0],
-            "leftPadLength": [0],
-            "inputCol": ["float_col"],
-            "1": [10.5],
-            "0": [20.5],
-        }
-    )
-    tensors_float, _ = numpy_to_pytorch(
-        data_float,
-        {"float_col": torch.float32},
-        ["float_col"],
-        resolve_window_view(
-            StoredWindowLayout(stored_context_width=2, max_target_offset=1, version=2),
-            ModelWindowView(context_length=1, objective="causal", target_offset=1),
-        ),
-    )
-    assert tensors_float["float_col"].dtype == torch.float32
-
-
-def test_build_valid_mask_from_left_pad_lengths():
-    left_pad_lengths = torch.tensor([0, 2, 5], dtype=torch.int64)
-
-    resolved_view = resolve_window_view(
-        StoredWindowLayout(stored_context_width=6, max_target_offset=1, version=2),
-        ModelWindowView(context_length=5, objective="causal", target_offset=1),
-    )
-    input_mask = build_valid_mask(
-        left_pad_lengths, full_length=6, view_slice=resolved_view.input_slice
-    )
-    target_mask = build_valid_mask(
-        left_pad_lengths, full_length=6, view_slice=resolved_view.target_slice
-    )
-
-    assert torch.equal(
-        input_mask,
-        torch.tensor(
-            [
-                [True, True, True, True, True],
-                [False, False, True, True, True],
-                [False, False, False, False, False],
-            ]
-        ),
-    )
-    assert torch.equal(
-        target_mask,
-        torch.tensor(
-            [
-                [True, True, True, True, True],
-                [False, True, True, True, True],
-                [False, False, False, False, True],
-            ]
-        ),
-    )
-
-
-def test_resolve_window_view_slices_tensors():
-    tensor = torch.arange(8).reshape(2, 4)
-    resolved_view = resolve_window_view(
-        StoredWindowLayout(stored_context_width=4, max_target_offset=1, version=2),
-        ModelWindowView(context_length=3, objective="causal", target_offset=1),
-    )
-
-    assert torch.equal(
-        tensor[:, resolved_view.input_slice],
-        torch.tensor([[0, 1, 2], [4, 5, 6]]),
-    )
-    assert torch.equal(
-        tensor[:, resolved_view.target_slice],
-        torch.tensor([[1, 2, 3], [5, 6, 7]]),
-    )
-
-
-def test_numpy_to_pytorch_includes_explicit_padding_masks():
-    data = pl.DataFrame(
-        {
-            "sequenceId": [1, 2],
-            "subsequenceId": [0, 0],
-            "startItemPosition": [0, 0],
-            "leftPadLength": [0, 2],
-            "inputCol": ["A", "A"],
-            "3": [0.0, 0.0],
-            "2": [0.0, 0.0],
-            "1": [1.0, 1.0],
-            "0": [2.0, 2.0],
-        }
-    )
-
-    _, metadata = numpy_to_pytorch(
-        data,
-        {"A": torch.float32},
-        ["A"],
-        resolve_window_view(
-            StoredWindowLayout(stored_context_width=4, max_target_offset=1, version=2),
-            ModelWindowView(context_length=3, objective="causal", target_offset=1),
-        ),
-    )
-
-    assert torch.equal(
-        metadata["attention_valid_mask"],
-        torch.tensor([[True, True, True], [False, False, True]]),
-    )
-    assert torch.equal(
-        metadata["target_valid_mask"],
-        torch.tensor([[True, True, True], [False, True, True]]),
-    )
-
-
-class _OnesSpanMasking:
-    def sample(self, shape, device, generator=None):
-        return torch.ones(shape, dtype=torch.long, device=device)
-
-
-class _VariableSpanMasking:
-    def sample(self, shape, device, generator=None):
-        lengths = torch.tensor([3, 2, 4, 1, 5], dtype=torch.long, device=device)
-        repeats = (shape[1] + lengths.numel() - 1) // lengths.numel()
-        return lengths.repeat(repeats)[: shape[1]].repeat(shape[0], 1)
-
-
-def _bert_masking_config(
-    masking_probability=0.5,
-    *,
-    categorical_columns=None,
-    real_columns=None,
-    n_classes=None,
-    replacement_distribution=None,
-    span_masking=None,
-):
-    return SimpleNamespace(
-        categorical_columns=categorical_columns or [],
-        real_columns=real_columns or [],
-        n_classes=n_classes or {},
-        context_length=6,
-        training_spec=SimpleNamespace(
-            batch_size=4,
-            bert_spec=SimpleNamespace(
-                masking_probability=masking_probability,
-                span_masking=span_masking or _OnesSpanMasking(),
-                replacement_distribution=replacement_distribution
-                or SimpleNamespace(masked=0.0, random=0.0, identical=1.0),
-            ),
-        ),
-    )
-
-
-def test_apply_bert_masking_builds_exact_valid_masks():
-    valid_mask = torch.tensor(
-        [
-            [True, True, True, True, True, True],
-            [False, False, True, True, True, True],
-            [False, False, False, False, False, False],
-            [False, True, False, True, True, True],
-        ]
-    )
-    config = _bert_masking_config(masking_probability=0.5)
-    data_batch = {"passthrough": torch.arange(24).reshape(4, 6)}
-    targets_batch = {"passthrough": torch.arange(24).reshape(4, 6)}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    _, _, masked_metadata = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=11,
-    )
-
-    bert_mask = masked_metadata["bert_mask"]
-    expected_budget = (valid_mask.sum(dim=1) * 0.5).long()
-
-    assert bert_mask.dtype == torch.bool
-    assert bert_mask.device == valid_mask.device
-    assert torch.equal(bert_mask.sum(dim=1), expected_budget)
-    assert not torch.any(bert_mask & ~valid_mask)
-
-
-def test_apply_bert_masking_eval_seed_is_reproducible_without_global_rng_change():
-    valid_mask = torch.tensor(
-        [
-            [True, True, True, True, True, True],
-            [False, True, True, True, True, True],
-        ]
-    )
-    config = _bert_masking_config(masking_probability=0.5)
-    data_batch = {"passthrough": torch.arange(12).reshape(2, 6)}
-    targets_batch = {"passthrough": torch.arange(12).reshape(2, 6)}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    torch.manual_seed(123)
-    rng_state = torch.get_rng_state().clone()
-    _, _, first_metadata = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=99,
-    )
-    after_first = torch.get_rng_state().clone()
-
-    _, _, second_metadata = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=99,
-    )
-    after_second = torch.get_rng_state().clone()
-
-    assert torch.equal(first_metadata["bert_mask"], second_metadata["bert_mask"])
-    assert torch.equal(after_first, rng_state)
-    assert torch.equal(after_second, rng_state)
-
-
-def test_apply_bert_masking_handles_variable_length_spans_and_sparse_masks():
-    valid_mask = torch.tensor(
-        [
-            [True, True, True, True, True, True, True, True],
-            [False, True, False, True, True, False, True, True],
-        ]
-    )
-    config = _bert_masking_config(
-        masking_probability=0.75,
-        span_masking=_VariableSpanMasking(),
-    )
-    data_batch = {"passthrough": torch.arange(16).reshape(2, 8)}
-    targets_batch = {"passthrough": torch.arange(16).reshape(2, 8)}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    _, _, masked_metadata = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=99,
-    )
-
-    assert torch.equal(
-        masked_metadata["bert_mask"],
-        torch.tensor(
-            [
-                [False, True, True, True, True, True, True, False],
-                [False, True, False, True, True, False, False, False],
-            ]
-        ),
-    )
-
-
-@pytest.mark.skipif(not torch.cuda.is_available(), reason="CUDA required")
-def test_apply_bert_masking_cuda_stays_on_device_and_preserves_rng():
-    device = torch.device("cuda")
-    valid_mask = torch.tensor(
-        [
-            [True, True, True, True, True, True],
-            [False, True, True, True, True, True],
-        ],
-        device=device,
-    )
-    config = _bert_masking_config(masking_probability=0.5)
-    data_batch = {"passthrough": torch.arange(12, device=device).reshape(2, 6)}
-    targets_batch = {"passthrough": torch.arange(12, device=device).reshape(2, 6)}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    torch.manual_seed(123)
-    cpu_state = torch.get_rng_state().clone()
-    cuda_state = torch.cuda.get_rng_state(device).clone()
-
-    _, _, masked_metadata = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=99,
-    )
-
-    assert masked_metadata["bert_mask"].device == device
-    assert torch.equal(torch.get_rng_state(), cpu_state)
-    assert torch.equal(torch.cuda.get_rng_state(device), cuda_state)
-
-
-def test_apply_bert_masking_replaces_categorical_with_mask_token_without_mutation():
-    valid_mask = torch.tensor([[True, False, True, True]])
-    config = _bert_masking_config(
-        masking_probability=1.0,
-        categorical_columns=["cat_col"],
-        n_classes={"cat_col": 8},
-        replacement_distribution=SimpleNamespace(
-            masked=1.0,
-            random=0.0,
-            identical=0.0,
-        ),
-    )
-    original = torch.tensor([[3, 4, 5, 6]])
-    data_batch = {"cat_col": original.clone()}
-    targets_batch = {"cat_col": original.clone()}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    masked_data, _, _ = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=7,
-    )
-
-    assert torch.equal(
-        masked_data["cat_col"],
-        torch.tensor(
-            [
-                [
-                    SPECIAL_TOKEN_IDS.mask,
-                    4,
-                    SPECIAL_TOKEN_IDS.mask,
-                    SPECIAL_TOKEN_IDS.mask,
-                ]
-            ]
-        ),
-    )
-    assert torch.equal(data_batch["cat_col"], original)
-
-
-def test_apply_bert_masking_replaces_categorical_with_random_tokens_without_mutation():
-    valid_mask = torch.ones((1, 5), dtype=torch.bool)
-    config = _bert_masking_config(
-        masking_probability=1.0,
-        categorical_columns=["cat_col"],
-        n_classes={"cat_col": 12},
-        replacement_distribution=SimpleNamespace(
-            masked=0.0,
-            random=1.0,
-            identical=0.0,
-        ),
-    )
-    original = torch.tensor([[3, 4, 5, 6, 7]])
-    data_batch = {"cat_col": original.clone()}
-    targets_batch = {"cat_col": original.clone()}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    masked_data, _, _ = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=7,
-    )
-
-    assert torch.all(masked_data["cat_col"] >= SPECIAL_TOKEN_IDS.user_start)
-    assert torch.all(masked_data["cat_col"] < config.n_classes["cat_col"])
-    assert not torch.equal(masked_data["cat_col"], original)
-    assert torch.equal(data_batch["cat_col"], original)
-
-
-def test_apply_bert_masking_replaces_real_with_zero_without_mutation():
-    valid_mask = torch.tensor([[True, False, True, True]])
-    config = _bert_masking_config(
-        masking_probability=1.0,
-        real_columns=["real_col"],
-        replacement_distribution=SimpleNamespace(
-            masked=1.0,
-            random=0.0,
-            identical=0.0,
-        ),
-    )
-    original = torch.tensor([[1.5, 2.5, 3.5, 4.5]])
-    data_batch = {"real_col": original.clone()}
-    targets_batch = {"real_col": original.clone()}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    masked_data, _, _ = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=7,
-    )
-
-    assert torch.equal(
-        masked_data["real_col"],
-        torch.tensor([[0.0, 2.5, 0.0, 0.0]]),
-    )
-    assert torch.equal(data_batch["real_col"], original)
-
-
-def test_apply_bert_masking_replaces_real_with_random_noise_without_mutation():
-    valid_mask = torch.tensor([[True, True, False, True]])
-    config = _bert_masking_config(
-        masking_probability=1.0,
-        real_columns=["real_col"],
-        replacement_distribution=SimpleNamespace(
-            masked=0.0,
-            random=1.0,
-            identical=0.0,
-        ),
-    )
-    original = torch.tensor([[10.0, 11.0, 12.0, 13.0]])
-    data_batch = {"real_col": original.clone()}
-    targets_batch = {"real_col": original.clone()}
-    metadata = {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-    masked_data, _, _ = apply_bert_masking(
-        data_batch,
-        targets_batch,
-        metadata,
-        config,
-        eval_seed=7,
-    )
-
-    assert torch.equal(masked_data["real_col"][~valid_mask], original[~valid_mask])
-    assert torch.all(masked_data["real_col"][valid_mask] != original[valid_mask])
-    assert torch.equal(data_batch["real_col"], original)
-
-
-def test_apply_bert_masking_uses_explicit_valid_mask_for_zero_values():
-    config = SimpleNamespace(
-        categorical_columns=[],
-        real_columns=["real_col"],
-        n_classes={},
-        context_length=4,
-        training_spec=SimpleNamespace(
-            batch_size=1,
-            bert_spec=SimpleNamespace(
-                masking_probability=1.0,
-                span_masking=_OnesSpanMasking(),
-                replacement_distribution=SimpleNamespace(
-                    masked=1.0,
-                    random=0.0,
-                    identical=0.0,
-                ),
-            ),
-        ),
-    )
-    data_batch = {
-        "real_col": torch.tensor([[99.0, 0.0, 1.0, 2.0]]),
-    }
-    targets_batch = {
-        "real_col": torch.tensor([[99.0, 0.0, 1.0, 2.0]]),
-    }
-    metadata = {
-        "attention_valid_mask": torch.tensor([[False, True, True, True]]),
-        "target_valid_mask": torch.tensor([[False, True, True, True]]),
-    }
-
-    _, _, masked_metadata = apply_bert_masking(
-        data_batch, targets_batch, metadata, config
-    )
-
-    assert torch.equal(
-        masked_metadata["bert_mask"],
-        torch.tensor([[False, True, True, True]]),
-    )
diff --git a/tests/unit/test_hyperparameter_search_config.py b/tests/unit/test_hyperparameter_search_config.py
deleted file mode 100644
index 16cfbf7e..00000000
--- a/tests/unit/test_hyperparameter_search_config.py
+++ /dev/null
@@ -1,178 +0,0 @@
-import optuna
-import pytest
-import yaml
-from pydantic import ValidationError
-
-from sequifier.config.hyperparameter_search_config import (
-    TrainingSpecHyperparameterSampling,
-)
-from sequifier.hyperparameter_search import create_sampler
-from sequifier.io.yaml import TrainModelDumper
-
-
-class RecordingTrial:
-    def __init__(self, overrides=None):
-        self.overrides = overrides or {}
-        self.params = {}
-
-    def suggest_categorical(self, name, choices):
-        value = self.overrides.get(name, choices[0])
-        assert value in choices
-        self.params[name] = value
-        return value
-
-    def suggest_float(self, name, low, high, step=None, log=False):
-        value = self.overrides.get(name, low)
-        self.params[name] = value
-        return value
-
-    def suggest_int(self, name, low, high, step=1, log=False):
-        value = self.overrides.get(name, low)
-        self.params[name] = value
-        return value
-
-
-def make_training_sampling(**overrides):
-    config = {
-        "device": "cpu",
-        "epochs": [1],
-        "save_interval_epochs": 1,
-        "batch_size": [2],
-        "learning_rate": [0.001],
-        "criterion": {"itemId": "CrossEntropyLoss"},
-        "accumulation_steps": [1],
-        "dropout": [0.0],
-        "optimizer": [{"name": "Adam"}],
-        "scheduler": [{"name": "StepLR", "step_size": 1, "gamma": 0.99}],
-        "continue_training": False,
-    }
-    config.update(overrides)
-    return TrainingSpecHyperparameterSampling(**config)
-
-
-def bert_spec_sampling_config():
-    return {
-        "masking_probability": [0.15, 0.30],
-        "replacement_distribution": [
-            {"masked": 0.8, "random": 0.1, "identical": 0.1},
-            {"masked": 0.6, "random": 0.2, "identical": 0.2},
-        ],
-        "span_masking": [
-            {"type": "GeometricDistribution", "p": 1.0},
-            {"type": "PoissonDistributionFloor", "rate": 2.0},
-        ],
-    }
-
-
-def sample_training_spec(sampling, trial):
-    return sampling.sample_trial(trial)
-
-
-def test_training_objective_defaults_to_causal_without_bert_spec():
-    sampling = make_training_sampling()
-    trial = RecordingTrial()
-
-    training_spec = sample_training_spec(sampling, trial)
-
-    assert training_spec.training_objective == "causal"
-    assert training_spec.bert_spec is None
-    assert trial.params["training_objective"] == "causal"
-
-
-def test_bert_spec_fields_are_sampled_separately_for_bert_objective():
-    sampling = make_training_sampling(
-        training_objective=["causal", "bert"],
-        bert_spec=bert_spec_sampling_config(),
-    )
-    trial = RecordingTrial(
-        {
-            "training_objective": "bert",
-            "bert_masking_probability": 0.30,
-            "bert_replacement_distribution_index": 1,
-            "bert_span_masking_index": 1,
-        }
-    )
-
-    training_spec = sample_training_spec(sampling, trial)
-
-    assert training_spec.training_objective == "bert"
-    assert training_spec.bert_spec is not None
-    assert training_spec.bert_spec.masking_probability == 0.30
-    assert training_spec.bert_spec.replacement_distribution.masked == 0.6
-    assert training_spec.bert_spec.replacement_distribution.random == 0.2
-    assert training_spec.bert_spec.replacement_distribution.identical == 0.2
-    assert training_spec.bert_spec.span_masking.type == "PoissonDistributionFloor"
-    assert set(trial.params).issuperset(
-        {
-            "training_objective",
-            "bert_masking_probability",
-            "bert_replacement_distribution_index",
-            "bert_span_masking_index",
-        }
-    )
-
-
-def test_causal_objective_does_not_sample_bert_fields():
-    sampling = make_training_sampling(
-        training_objective=["causal", "bert"],
-        bert_spec=bert_spec_sampling_config(),
-    )
-    trial = RecordingTrial({"training_objective": "causal"})
-
-    training_spec = sample_training_spec(sampling, trial)
-
-    assert training_spec.training_objective == "causal"
-    assert training_spec.bert_spec is None
-    assert "bert_masking_probability" not in trial.params
-    assert "bert_replacement_distribution_index" not in trial.params
-    assert "bert_span_masking_index" not in trial.params
-
-
-def test_bert_objective_requires_bert_spec_sampling_config():
-    with pytest.raises(ValidationError, match="bert_spec"):
-        make_training_sampling(training_objective=["bert"])
-
-
-def test_bert_training_spec_dumps_to_plain_yaml():
-    sampling = make_training_sampling(
-        training_objective=["bert"],
-        bert_spec=bert_spec_sampling_config(),
-    )
-    training_spec = sample_training_spec(sampling, RecordingTrial())
-
-    dumped = yaml.dump(training_spec, Dumper=TrainModelDumper, sort_keys=False)
-    loaded = yaml.safe_load(dumped)
-
-    assert loaded["training_objective"] == "bert"
-    assert loaded["bert_spec"]["masking_probability"] == 0.15
-    assert loaded["bert_spec"]["replacement_distribution"] == {
-        "masked": 0.8,
-        "random": 0.1,
-        "identical": 0.1,
-    }
-    assert loaded["bert_spec"]["span_masking"] == {
-        "type": "GeometricDistribution",
-        "p": 1.0,
-    }
-
-
-@pytest.mark.parametrize(
-    ("strategy", "sampler_name"),
-    [
-        ("sample", "RandomSampler"),
-        ("grid", "BruteForceSampler"),
-        ("bayesian", "TPESampler"),
-    ],
-)
-def test_create_sampler_passes_config_seed(monkeypatch, strategy, sampler_name):
-    recorded = {}
-
-    def fake_sampler(*, seed=None):
-        recorded["seed"] = seed
-        return object()
-
-    monkeypatch.setattr(optuna.samplers, sampler_name, fake_sampler)
-
-    create_sampler(type("Config", (), {"search_strategy": strategy, "seed": 123})())
-
-    assert recorded["seed"] == 123
diff --git a/tests/unit/test_infer.py b/tests/unit/test_infer.py
deleted file mode 100644
index b3d836ba..00000000
--- a/tests/unit/test_infer.py
+++ /dev/null
@@ -1,833 +0,0 @@
-import json
-from unittest.mock import MagicMock, patch
-
-import numpy as np
-import polars as pl
-import pytest
-import torch
-import yaml
-
-from sequifier.config.infer_config import InfererModel, load_inferer_config
-from sequifier.helpers import ModelWindowView, StoredWindowLayout
-from sequifier.infer import (
-    Inferer,
-    calculate_item_positions,
-    get_probs_preds_from_dict,
-    infer_embedding,
-    infer_generative,
-    normalize,
-    sample_with_cumsum,
-)
-
-
-def test_normalize():
-    """Softmax-normalized logits."""
-    # Create raw logits
-    # Row 0: exp(0)=1, exp(0)=1 -> probs: 0.5, 0.5
-    # Row 1: exp(1)=e, exp(2)=e^2 -> probs: e/(e+e^2), e^2/(e+e^2)
-    logits = np.array([[0.0, 0.0], [1.0, 2.0], [-1.0, 0.0]])
-    outs = {"target_col": logits}
-
-    probs_dict = normalize(outs)
-
-    assert "target_col" in probs_dict
-    probs = probs_dict["target_col"]
-    assert probs.shape == (3, 2)
-
-    # Assert Row 0 is 50/50
-    np.testing.assert_allclose(probs[0], [0.5, 0.5])
-
-    # Assert Row 1 is proportionally correct
-    e1, e2 = np.exp(1), np.exp(2)
-    np.testing.assert_allclose(probs[1], [e1 / (e1 + e2), e2 / (e1 + e2)])
-
-    # Assert all rows sum to 1.0 (valid probability distribution)
-    np.testing.assert_allclose(probs.sum(axis=1), [1.0, 1.0, 1.0])
-
-
-def test_normalize_is_stable_for_large_logits():
-    """Softmax remains finite for large-magnitude logits."""
-    outs = {
-        "target_col": np.array(
-            [[1000.0, 1001.0], [-1001.0, -1000.0], [1000.0, -1000.0]]
-        )
-    }
-
-    probs = normalize(outs)["target_col"]
-
-    assert np.isfinite(probs).all()
-    np.testing.assert_allclose(probs.sum(axis=1), [1.0, 1.0, 1.0])
-    np.testing.assert_allclose(
-        probs[0],
-        [1.0 / (1.0 + np.e), np.e / (1.0 + np.e)],
-    )
-
-
-@patch("numpy.random.rand")
-def test_sample_with_cumsum(mock_rand):
-    """Inverse-CDF sampling for log-probabilities and probabilities."""
-    # Mock the random thresholds to strictly control the sampling outcome.
-    mock_rand.return_value = np.array([[0.05], [0.90]])
-
-    # Path 1: Test with log-probabilities=True (default)
-    log_probs = np.array([[np.log(0.1), np.log(0.9)], [np.log(0.8), np.log(0.2)]])
-    sampled_from_log_probs = sample_with_cumsum(log_probs, is_log_probs=True)
-    np.testing.assert_array_equal(sampled_from_log_probs, [0, 1])
-
-    # Path 2: Test with is_log_probs=False (pre-normalized probabilities)
-    pure_probs = np.array([[0.1, 0.9], [0.8, 0.2]])
-    sampled_from_probs = sample_with_cumsum(pure_probs, is_log_probs=False)
-    np.testing.assert_array_equal(sampled_from_probs, [0, 1])
-
-
-@patch("numpy.random.rand")
-def test_sample_with_cumsum_clamps_final_cumulative_probability(mock_rand):
-    """Tiny floating-point deficits still allow the final class to be sampled."""
-    mock_rand.return_value = np.array([[0.9999999999999999]])
-    probs = np.array([[0.1, 0.8999999999999998]])
-
-    sampled = sample_with_cumsum(probs, is_log_probs=False)
-
-    np.testing.assert_array_equal(sampled, [1])
-
-
-def test_sample_with_cumsum_rejects_invalid_probability_mass():
-    """Invalid probability rows fail instead of falling through to class zero."""
-    with pytest.raises(ValueError, match="sum to 1.0"):
-        sample_with_cumsum(np.array([[0.4, 0.4]]), is_log_probs=False)
-
-
-@pytest.fixture
-def empty_parquet_path(tmp_path):
-    path = tmp_path / "empty.parquet"
-    pl.DataFrame({"target_col": []}, schema={"target_col": pl.Float64}).write_parquet(
-        path
-    )
-    return str(path)
-
-
-@pytest.fixture
-def mock_inferer():
-    """Sets up an Inferer instance with mocked heavy dependencies (ONNX/PyTorch)."""
-    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
-        "sequifier.infer.load_inference_model"
-    ):
-        inferer = Inferer(
-            model_type="generative",
-            model_path="dummy_model.onnx",
-            project_root=".",
-            id_maps={"cat_col": {"A": 2, "B": 3}},
-            selected_columns_statistics={"real_col": {"mean": 10.0, "std": 2.0}},
-            map_to_id=True,
-            categorical_columns=["cat_col"],
-            real_columns=["real_col"],
-            input_columns=["cat_col", "real_col"],
-            target_columns=["cat_col", "real_col"],
-            target_column_types={"cat_col": "categorical", "real_col": "real"},
-            sample_from_distribution_columns=None,
-            infer_with_dropout=False,
-            prediction_length=1,
-            inference_batch_size=4,
-            device="cpu",
-            args_config={},
-            training_config_path="dummy.yaml",
-        )
-        return inferer
-
-
-def test_inferer_invert_normalization(mock_inferer):
-    """Real outputs denormalize correctly."""
-    # Normalized values
-    values = np.array([-1.0, 0.0, 1.0])
-
-    # Configuration dictates mean = 10.0, std = 2.0
-    # Unnormalized logic: (val * (std - 1e-9)) + mean
-    # Expected approx: [8.0, 10.0, 12.0]
-    unnormalized = mock_inferer.invert_normalization(values, "real_col")
-
-    np.testing.assert_allclose(unnormalized, [8.0, 10.0, 12.0], atol=1e-5)
-
-
-def test_inferer_expand_to_batch_size(mock_inferer):
-    """Strict-batch array padding."""
-    mock_inferer.inference_batch_size = 5
-
-    # Input has 2 samples, batch size needs to be 5
-    x = np.array([[10], [20]])
-
-    # Should repeat the full array twice [10, 20, 10, 20], then append the remainder [10]
-    expanded = mock_inferer.expand_to_batch_size(x)
-
-    assert expanded.shape == (5, 1)
-    np.testing.assert_array_equal(expanded, [[10], [20], [10], [20], [10]])
-
-
-def test_inferer_prepare_inference_batches_pad(mock_inferer):
-    """Chunking with padding."""
-    mock_inferer.inference_batch_size = 4
-    x = {"cat_col": np.array([[1], [2], [3]])}  # 3 total samples
-
-    batches = mock_inferer.prepare_inference_batches(x, pad_to_batch_size=True)
-
-    assert len(batches) == 1
-    # Check that the 3 samples were padded up to the target batch size of 4
-    assert batches[0]["cat_col"].shape == (4, 1)
-    np.testing.assert_array_equal(batches[0]["cat_col"], [[1], [2], [3], [1]])
-
-
-def test_inferer_prepare_inference_batches_no_pad(mock_inferer):
-    """Chunking without padding."""
-    mock_inferer.inference_batch_size = 4
-    x = {"cat_col": np.array([[1], [2], [3]])}  # 3 total samples
-
-    batches = mock_inferer.prepare_inference_batches(x, pad_to_batch_size=False)
-
-    assert len(batches) == 1
-    # Check that it retained its original short size
-    assert batches[0]["cat_col"].shape == (3, 1)
-
-
-def test_inferer_prepare_inference_batches_split(mock_inferer):
-    """Chunking into multiple batches."""
-    mock_inferer.inference_batch_size = 2
-    x = {"cat_col": np.array([[1], [2], [3], [4], [5]])}  # 5 total samples
-
-    batches = mock_inferer.prepare_inference_batches(x, pad_to_batch_size=False)
-
-    assert len(batches) == 3
-    assert batches[0]["cat_col"].shape == (2, 1)
-    assert batches[1]["cat_col"].shape == (2, 1)
-    assert batches[2]["cat_col"].shape == (1, 1)  # The remainder
-    np.testing.assert_array_equal(batches[2]["cat_col"], [[5]])
-
-
-def test_infer_config_defaults_bert_prediction_length_to_context_length(
-    empty_parquet_path,
-):
-    config = InfererModel(
-        project_root=".",
-        metadata_config_path="dummy.json",
-        model_path="dummy.onnx",
-        model_type="generative",
-        training_objective="bert",
-        data_path=empty_parquet_path,
-        input_columns=["target_col"],
-        categorical_columns=[],
-        real_columns=["target_col"],
-        target_columns=["target_col"],
-        column_types={"target_col": "float64"},
-        target_column_types={"target_col": "real"},
-        seed=42,
-        device="cpu",
-        prediction_length=None,
-        storage_layout=StoredWindowLayout(
-            stored_context_width=4, max_target_offset=1, version=2
-        ),
-        window_view=ModelWindowView(
-            context_length=3, objective="bert", target_offset=0
-        ),
-        inference_batch_size=2,
-        output_probabilities=False,
-        map_to_id=False,
-        autoregression=False,
-    )
-
-    assert config.prediction_length == config.window_view.context_length
-
-
-def test_infer_config_defaults_causal_prediction_length_to_one(empty_parquet_path):
-    config = InfererModel(
-        project_root=".",
-        metadata_config_path="dummy.json",
-        model_path="dummy.onnx",
-        model_type="generative",
-        training_objective="causal",
-        data_path=empty_parquet_path,
-        input_columns=["target_col"],
-        categorical_columns=[],
-        real_columns=["target_col"],
-        target_columns=["target_col"],
-        column_types={"target_col": "float64"},
-        target_column_types={"target_col": "real"},
-        seed=42,
-        device="cpu",
-        prediction_length=None,
-        storage_layout=StoredWindowLayout(
-            stored_context_width=4, max_target_offset=1, version=2
-        ),
-        window_view=ModelWindowView(
-            context_length=3, objective="causal", target_offset=1
-        ),
-        inference_batch_size=2,
-        output_probabilities=False,
-        map_to_id=False,
-        autoregression=False,
-    )
-
-    assert config.prediction_length == 1
-
-
-def test_infer_config_rejects_bert_prediction_length_mismatch(empty_parquet_path):
-    with pytest.raises(ValueError, match="prediction_length must be equal"):
-        InfererModel(
-            project_root=".",
-            metadata_config_path="dummy.json",
-            model_path="dummy.onnx",
-            model_type="generative",
-            training_objective="bert",
-            data_path=empty_parquet_path,
-            input_columns=["target_col"],
-            categorical_columns=[],
-            real_columns=["target_col"],
-            target_columns=["target_col"],
-            column_types={"target_col": "float64"},
-            target_column_types={"target_col": "real"},
-            seed=42,
-            device="cpu",
-            prediction_length=1,
-            storage_layout=StoredWindowLayout(
-                stored_context_width=4, max_target_offset=1, version=2
-            ),
-            window_view=ModelWindowView(
-                context_length=3, objective="bert", target_offset=0
-            ),
-            inference_batch_size=2,
-            output_probabilities=False,
-            map_to_id=False,
-            autoregression=False,
-        )
-
-
-def test_load_inferer_config_rejects_mismatched_metadata_special_token_ids(tmp_path):
-    data_path = tmp_path / "data.parquet"
-    data_path.touch()
-    metadata_path = tmp_path / "metadata.json"
-    config_path = tmp_path / "infer.yaml"
-
-    metadata_path.write_text(
-        json.dumps(
-            {
-                "split_paths": [str(data_path)],
-                "stored_context_width": 4,
-                "max_target_offset": 1,
-                "stored_window_layout_version": 2,
-                "column_types": {"target_col": "int64"},
-                "id_maps": {"target_col": {"A": 3}},
-                "selected_columns_statistics": {},
-                "special_token_ids": {
-                    "[unknown]": 10,
-                    "[other]": 11,
-                    "[mask]": 12,
-                },
-            }
-        )
-    )
-    config_path.write_text(
-        yaml.safe_dump(
-            {
-                "project_root": str(tmp_path),
-                "metadata_config_path": metadata_path.name,
-                "model_path": "dummy.onnx",
-                "model_type": "generative",
-                "training_objective": "causal",
-                "data_path": str(data_path),
-                "input_columns": ["target_col"],
-                "categorical_columns": ["target_col"],
-                "real_columns": [],
-                "target_columns": ["target_col"],
-                "column_types": {"target_col": "int64"},
-                "target_column_types": {"target_col": "categorical"},
-                "seed": 42,
-                "device": "cpu",
-                "context_length": 3,
-                "inference_batch_size": 2,
-            }
-        )
-    )
-
-    with pytest.raises(ValueError, match="special_token_ids must match"):
-        load_inferer_config(str(config_path), {}, skip_metadata=False)
-
-
-def test_infer_pure_passes_attention_valid_mask_to_onnx(mock_inferer):
-    class Input:
-        def __init__(self, name):
-            self.name = name
-
-    class Session:
-        def __init__(self):
-            self.ort_inputs = None
-
-        def get_inputs(self):
-            return [
-                Input("cat_col_in"),
-                Input("real_col_in"),
-                Input("attention_valid_mask"),
-            ]
-
-        def run(self, _, ort_inputs):
-            self.ort_inputs = ort_inputs
-            return [np.zeros((1, 2, 3), dtype=np.float32)]
-
-    session = Session()
-    mock_inferer.ort_session = session
-    mock_inferer.inference_batch_size = 2
-
-    x = {
-        "cat_col": np.array([[1, 2], [3, 4]], dtype=np.int64),
-        "real_col": np.array([[1.0, 2.0], [3.0, 4.0]], dtype=np.float32),
-    }
-    metadata = {
-        "attention_valid_mask": np.array([[False, True], [True, True]], dtype=np.bool_)
-    }
-
-    mock_inferer.infer_pure(x, metadata)
-    assert session.ort_inputs is not None
-    np.testing.assert_array_equal(
-        session.ort_inputs["attention_valid_mask"],
-        metadata["attention_valid_mask"],
-    )
-
-
-def test_infer_pure_rejects_unknown_onnx_input(mock_inferer):
-    class Input:
-        def __init__(self, name):
-            self.name = name
-
-    class Session:
-        def get_inputs(self):
-            return [Input("unexpected_input")]
-
-    mock_inferer.ort_session = Session()
-    metadata = {}
-
-    with pytest.raises(ValueError, match="Could not map ONNX input"):
-        mock_inferer.infer_pure(
-            {
-                "cat_col": np.array([[1, 2]], dtype=np.int64),
-                "real_col": np.array([[1.0, 2.0]], dtype=np.float32),
-            },
-            metadata,
-        )
-
-
-def test_calculate_item_positions_bert_uses_full_input_window():
-    positions = calculate_item_positions(
-        np.array([10, 20]),
-        context_length=4,
-        prediction_length=4,
-        training_objective="bert",
-    )
-
-    np.testing.assert_array_equal(positions, [10, 11, 12, 13, 20, 21, 22, 23])
-
-
-def test_calculate_item_positions_causal_uses_future_window_tail():
-    positions = calculate_item_positions(
-        np.array([10, 20]),
-        context_length=4,
-        prediction_length=2,
-        training_objective="causal",
-    )
-
-    np.testing.assert_array_equal(positions, [13, 14, 23, 24])
-
-
-def _bert_inference_config(tmp_path, model_type="generative", input_columns=None):
-    data_path = tmp_path / "data.parquet"
-    data_path.touch()
-    input_columns = input_columns or ["target_col"]
-
-    return InfererModel(
-        project_root=str(tmp_path),
-        metadata_config_path="dummy.json",
-        model_path="dummy.onnx",
-        model_type=model_type,
-        training_objective="bert",
-        data_path=str(data_path),
-        input_columns=input_columns,
-        categorical_columns=input_columns,
-        real_columns=[],
-        target_columns=["target_col"],
-        column_types={col: "int64" for col in input_columns},
-        target_column_types={"target_col": "categorical"},
-        seed=42,
-        device="cpu",
-        prediction_length=None,
-        storage_layout=StoredWindowLayout(
-            stored_context_width=4, max_target_offset=1, version=2
-        ),
-        window_view=ModelWindowView(
-            context_length=3, objective="bert", target_offset=0
-        ),
-        inference_batch_size=4,
-        output_probabilities=False,
-        map_to_id=True,
-        autoregression=False,
-    )
-
-
-def _bert_preprocessed_frame():
-    return pl.DataFrame(
-        {
-            "sequenceId": [7],
-            "subsequenceId": [0],
-            "startItemPosition": [100],
-            "leftPadLength": [2],
-            "inputCol": ["target_col"],
-            "3": [0],
-            "2": [0],
-            "1": [3],
-            "0": [4],
-        }
-    )
-
-
-def _bert_preprocessed_frame_out_of_order():
-    return pl.DataFrame(
-        {
-            "sequenceId": [2, 2, 1, 1],
-            "subsequenceId": [0, 0, 0, 0],
-            "startItemPosition": [200, 200, 100, 100],
-            "leftPadLength": [1, 1, 2, 2],
-            "inputCol": ["aux_col", "target_col", "aux_col", "target_col"],
-            "3": [9, 0, 8, 0],
-            "2": [9, 3, 8, 0],
-            "1": [9, 4, 8, 3],
-            "0": [9, 3, 8, 4],
-        }
-    )
-
-
-def _mock_bert_inferer(config):
-    return Inferer(
-        model_type=config.model_type,
-        model_path=config.model_path,
-        project_root=config.project_root,
-        id_maps={"target_col": {"A": 3, "B": 4}},
-        selected_columns_statistics={},
-        map_to_id=config.map_to_id,
-        categorical_columns=config.categorical_columns,
-        real_columns=config.real_columns,
-        input_columns=config.input_columns,
-        target_columns=config.target_columns,
-        target_column_types=config.target_column_types,
-        sample_from_distribution_columns=config.sample_from_distribution_columns,
-        infer_with_dropout=config.infer_with_dropout,
-        prediction_length=config.prediction_length,
-        inference_batch_size=config.inference_batch_size,
-        device=config.device,
-        args_config={},
-        training_config_path=config.training_config_path,
-    )
-
-
-def _dummy_attention_metadata(batch_size, context_length):
-    return {
-        "attention_valid_mask": torch.ones(
-            (batch_size, context_length), dtype=torch.bool
-        )
-    }
-
-
-def test_bert_generative_inference_filters_left_padded_positions(tmp_path):
-    config = _bert_inference_config(tmp_path)
-    config.output_probabilities = True
-    data = _bert_preprocessed_frame()
-    written = []
-
-    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
-        "sequifier.infer.load_inference_model"
-    ):
-        inferer = _mock_bert_inferer(config)
-
-    def capture_write(dataframe, path, write_format):
-        written.append((path, dataframe, write_format))
-
-    probs = {"target_col": np.full((3, 5), 0.2)}
-    preds = {"target_col": np.array([3, 4, 3])}
-    with patch(
-        "sequifier.infer.get_probs_preds_from_df", return_value=(probs, preds)
-    ), patch("sequifier.infer.write_data", side_effect=capture_write):
-        infer_generative(
-            config, inferer, "bert-model", [data], {"target_col": torch.int64}
-        )
-
-    predictions = next(frame for path, frame, _ in written if "predictions" in path)
-    probabilities = next(frame for path, frame, _ in written if "probabilities" in path)
-
-    assert predictions.height == 1
-    assert predictions.get_column("sequenceId").to_list() == [7]
-    assert predictions.get_column("itemPosition").to_list() == [102]
-    assert predictions.get_column("target_col").to_list() == ["A"]
-    assert probabilities.height == 1
-
-
-def test_bert_generative_inference_uses_dataframe_order_for_padding_masks(tmp_path):
-    config = _bert_inference_config(tmp_path, input_columns=["aux_col", "target_col"])
-    data = _bert_preprocessed_frame_out_of_order()
-    written = []
-
-    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
-        "sequifier.infer.load_inference_model"
-    ):
-        inferer = _mock_bert_inferer(config)
-
-    def capture_write(dataframe, path, write_format):
-        written.append((path, dataframe, write_format))
-
-    preds = {"target_col": np.array([3, 4, 3, 4, 3, 4])}
-    with patch(
-        "sequifier.infer.get_probs_preds_from_df", return_value=(None, preds)
-    ), patch("sequifier.infer.write_data", side_effect=capture_write):
-        infer_generative(
-            config,
-            inferer,
-            "bert-model",
-            [data],
-            {"target_col": torch.int64, "aux_col": torch.int64},
-        )
-
-    predictions = next(frame for path, frame, _ in written if "predictions" in path)
-
-    assert predictions.get_column("sequenceId").to_list() == [2, 2, 1]
-    assert predictions.get_column("itemPosition").to_list() == [201, 202, 102]
-    assert predictions.get_column("target_col").to_list() == ["B", "A", "B"]
-
-
-def test_bert_embedding_inference_filters_left_padded_positions(tmp_path):
-    config = _bert_inference_config(tmp_path, model_type="embedding")
-    data = _bert_preprocessed_frame()
-    written = []
-
-    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
-        "sequifier.infer.load_inference_model"
-    ):
-        inferer = _mock_bert_inferer(config)
-
-    def capture_write(dataframe, path, write_format):
-        written.append((path, dataframe, write_format))
-
-    embeddings = np.array([[0.1, 1.1], [0.2, 1.2], [0.3, 1.3]])
-    with patch("sequifier.infer.get_embeddings", return_value=embeddings), patch(
-        "sequifier.infer.write_data", side_effect=capture_write
-    ):
-        infer_embedding(
-            config, inferer, "bert-model", [data], {"target_col": torch.int64}
-        )
-
-    embeddings_df = next(frame for _, frame, _ in written)
-
-    assert embeddings_df.height == 1
-    assert embeddings_df.get_column("sequenceId").to_list() == [7]
-    assert embeddings_df.get_column("subsequenceId").to_list() == [0]
-    assert embeddings_df.get_column("itemPosition").to_list() == [102]
-    assert embeddings_df.select(["0", "1"]).row(0) == (0.3, 1.3)
-
-
-# ==========================================
-# Test Autoregressive Tensor Inference
-# ==========================================
-
-
-@pytest.fixture
-def ar_config(empty_parquet_path):
-    """Provides an actual InfererModel configuration for autoregressive inference."""
-    return InfererModel(
-        project_root=".",
-        metadata_config_path="dummy.json",
-        model_path="dummy.onnx",
-        model_type="generative",
-        training_objective="causal",
-        data_path=empty_parquet_path,
-        input_columns=["target_col"],
-        categorical_columns=[],
-        real_columns=["target_col"],
-        target_columns=["target_col"],
-        column_types={"target_col": "float64"},
-        target_column_types={"target_col": "real"},
-        seed=42,
-        device="cpu",
-        prediction_length=1,
-        storage_layout=StoredWindowLayout(
-            stored_context_width=4, max_target_offset=1, version=2
-        ),
-        window_view=ModelWindowView(
-            context_length=3, objective="causal", target_offset=1
-        ),
-        inference_batch_size=2,
-        output_probabilities=False,
-        map_to_id=False,  # Set to False to bypass ID mapping requirements
-        autoregression=True,
-        autoregression_total_steps=1,
-    )
-
-
-@pytest.fixture
-def ar_inferer(ar_config):
-    """Sets up an actual Inferer instance with mocked heavy dependencies."""
-    with patch("sequifier.infer.onnxruntime.InferenceSession"), patch(
-        "sequifier.infer.load_inference_model"
-    ):
-        return Inferer(
-            model_type=ar_config.model_type,
-            model_path=ar_config.model_path,
-            project_root=ar_config.project_root,
-            id_maps=None,
-            selected_columns_statistics={"target_col": {"mean": 0.0, "std": 1.0}},
-            map_to_id=ar_config.map_to_id,
-            categorical_columns=ar_config.categorical_columns,
-            real_columns=ar_config.real_columns,
-            input_columns=ar_config.input_columns,
-            target_columns=ar_config.target_columns,
-            target_column_types=ar_config.target_column_types,
-            sample_from_distribution_columns=ar_config.sample_from_distribution_columns,
-            infer_with_dropout=ar_config.infer_with_dropout,
-            prediction_length=ar_config.prediction_length,
-            inference_batch_size=ar_config.inference_batch_size,
-            device=ar_config.device,
-            args_config={},
-            training_config_path=ar_config.training_config_path,
-        )
-
-
-def test_get_probs_preds_from_dict_shifting_and_looping(ar_config, ar_inferer):
-    """Check autoregressive loop count and tensor shifting."""
-    initial_data = {"target_col": torch.tensor([[1.0, 2.0, 3.0], [10.0, 20.0, 30.0]])}
-    metadata = _dummy_attention_metadata(batch_size=2, context_length=3)
-
-    with patch.object(ar_inferer, "infer_generative") as mock_infer:
-        mock_infer.side_effect = [
-            {"target_col": np.array([[4.0], [40.0]])},
-            {"target_col": np.array([[5.0], [50.0]])},
-        ]
-
-        total_steps = 2
-        probs, preds = get_probs_preds_from_dict(
-            ar_config, ar_inferer, initial_data, metadata, total_steps=total_steps
-        )
-
-        assert mock_infer.call_count == 2
-
-        expected_shifted_x = {
-            "target_col": np.array([[2.0, 3.0, 4.0], [20.0, 30.0, 40.0]])
-        }
-        second_call_args = mock_infer.call_args_list[1][0][0]
-
-        np.testing.assert_array_equal(
-            second_call_args["target_col"], expected_shifted_x["target_col"]
-        )
-
-        assert "target_col" in preds
-        np.testing.assert_array_equal(preds["target_col"], [4.0, 5.0, 40.0, 50.0])
-
-
-@patch("sequifier.infer.sample_with_cumsum")
-def test_get_probs_preds_from_dict_with_probabilities(
-    mock_sample, ar_config, ar_inferer
-):
-    """Check probability output drives non-logit sampling."""
-    ar_config.target_column_types["target_col"] = "categorical"
-    ar_inferer.target_column_types["target_col"] = "categorical"
-
-    ar_config.output_probabilities = True
-    ar_config.sample_from_distribution_columns = ["target_col"]
-    ar_inferer.sample_from_distribution_columns = ["target_col"]
-
-    initial_data = {"target_col": torch.tensor([[1.0, 2.0]])}
-
-    dummy_logits = {"target_col": np.array([[np.log(0.2), np.log(0.8)]])}
-    mock_sample.return_value = np.array([1])
-
-    metadata = _dummy_attention_metadata(batch_size=1, context_length=2)
-
-    with patch.object(
-        ar_inferer, "adjust_and_infer_generative", return_value=dummy_logits
-    ) as mock_adjust:
-        probs, preds = get_probs_preds_from_dict(
-            ar_config, ar_inferer, initial_data, metadata, total_steps=1
-        )
-
-        assert mock_adjust.call_count == 1
-
-        mock_sample.assert_called_once()
-        args, kwargs = mock_sample.call_args
-
-        np.testing.assert_allclose(args[0], [[0.2, 0.8]])
-
-        assert kwargs.get("is_log_probs") is False
-
-        assert probs is not None
-        np.testing.assert_allclose(probs["target_col"], [[0.2, 0.8]])
-        np.testing.assert_array_equal(preds["target_col"], [1])
-
-
-def test_get_probs_preds_from_dict_ignores_unselected_columns(ar_config, ar_inferer):
-    """Check inference ignores columns outside config.input_columns."""
-    initial_data = {
-        "target_col": torch.tensor([[1.0, 2.0]]),
-        "ignored_col": torch.tensor([[9.0, 9.0]]),
-    }
-    metadata = _dummy_attention_metadata(batch_size=1, context_length=2)
-
-    with patch.object(ar_inferer, "infer_generative") as mock_infer:
-        mock_infer.return_value = {"target_col": np.array([[3.0]])}
-
-        _, _ = get_probs_preds_from_dict(
-            ar_config, ar_inferer, initial_data, metadata, total_steps=1
-        )
-
-        first_call_args = mock_infer.call_args_list[0][0][0]
-
-        assert "target_col" in first_call_args
-        assert "ignored_col" not in first_call_args
-
-
-def test_adjust_and_infer_embedding_onnx_truncation():
-    """
-    Catches the truncation bug where ONNX embeddings are sliced by `[:size]`
-    instead of `[:size * self.prediction_length]`.
-    """
-    # 1. Mock the Inferer instance to isolate the target method
-    #    We avoid actual __init__ to bypass ONNX runtime initialization.
-    inferer_mock = MagicMock(spec=Inferer)
-    inferer_mock.inference_model_type = "onnx"
-    inferer_mock.inference_batch_size = 2
-
-    # Set prediction_length > 1 to expose the bug (e.g., BERT objective)
-    inferer_mock.prediction_length = 5
-    embedding_dim = 16
-
-    # 2. Mock `prepare_inference_batches` to pass our dummy inputs straight through
-    inferer_mock.prepare_inference_batches.side_effect = (
-        lambda data, pad_to_batch_size: [data]
-    )
-
-    # 3. Mock `infer_pure` to return the shape an ONNX model would actually return.
-    #    `infer_pure` flattens the output to (batch_size * prediction_length, dim).
-    #    For a batch size of 2 and prediction length of 5, it returns (10, 16).
-    flattened_sequence_length = (
-        inferer_mock.inference_batch_size * inferer_mock.prediction_length
-    )
-    dummy_onnx_output = np.ones((flattened_sequence_length, embedding_dim))
-    inferer_mock.infer_pure.return_value = [dummy_onnx_output]
-
-    # 4. Construct dummy inputs
-    size = 2
-    x = {"feature_col": np.zeros((size, 10))}  # 2 sequences, context_length = 10
-    metadata = {"attention_valid_mask": np.ones((size, 10))}
-
-    # 5. Execute the method under test
-    # Passing inferer_mock as `self` to the unbound class method.
-    embeddings = Inferer.adjust_and_infer_embedding(inferer_mock, x, size, metadata)
-
-    # 6. Assert the bug is fixed
-    expected_rows = size * inferer_mock.prediction_length
-
-    assert embeddings.shape[0] == expected_rows, (
-        f"ONNX Embedding truncation bug detected! "
-        f"Expected {expected_rows} rows (size * prediction_length), "
-        f"but got {embeddings.shape[0]} rows. "
-        f"Check the slice at the end of np.concatenate in adjust_and_infer_embedding."
-    )
-    assert embeddings.shape[1] == embedding_dim
diff --git a/tests/unit/test_preprocess.py b/tests/unit/test_preprocess.py
deleted file mode 100644
index 6fb2a5c0..00000000
--- a/tests/unit/test_preprocess.py
+++ /dev/null
@@ -1,787 +0,0 @@
-import json
-
-import numpy as np
-import polars as pl
-import pytest
-import torch
-from pydantic import ValidationError
-
-from sequifier.config.preprocess_config import PreprocessorModel
-from sequifier.helpers import StoredWindowLayout
-from sequifier.preprocess import (
-    Preprocessor,
-    _apply_column_statistics,
-    _apply_mask_column,
-    _get_column_statistics,
-    _get_data_columns,
-    _load_and_preprocess_data,
-    create_id_map,
-    extract_sequences,
-    extract_subsequences,
-    get_batch_limits,
-    get_combined_statistics,
-    load_precomputed_id_maps,
-    process_and_write_data_pt,
-)
-
-RESERVED_MASK_COLUMN = "[mask]"
-
-
-def test_extract_subsequences_basic():
-    """Basic sliding-window extraction."""
-    input_data = {"col1": [10, 11, 12, 13, 14, 15]}
-    context_length = 3
-    stride = 1
-    columns = ["col1"]
-
-    # Expected behavior: Window size is context_length + 1 (history + target)
-    # Windows: [10,11,12,13], [11,12,13,14], [12,13,14,15]
-
-    result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data,
-        context_length + 1,
-        stride,
-        columns,
-        subsequence_start_mode="distribute",
-    )
-
-    assert len(result["col1"]) == 3
-    assert result["col1"][0] == [10, 11, 12, 13]
-    assert result["col1"][2] == [12, 13, 14, 15]
-
-
-def test_extract_subsequences_bert_width():
-    input_data = {"col1": [10, 11, 12, 13, 14, 15]}
-    context_length = 3
-    stride = 1
-    columns = ["col1"]
-
-    result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data,
-        stored_context_width=context_length,
-        stride_for_split=stride,
-        columns=columns,
-        subsequence_start_mode="distribute",
-    )
-
-    assert len(result["col1"]) == 4
-    assert result["col1"][0] == [10, 11, 12]
-    assert result["col1"][3] == [13, 14, 15]
-
-
-def test_extract_subsequences_padding():
-    """Short sequences receive zero padding."""
-    input_data = {"col1": [1, 2]}  # Length 2
-    stored_context_width = 5
-    stride = 1
-    columns = ["col1"]
-
-    # Expected: [0, 0, 0, 1, 2] -> 3 zeroes padding
-
-    result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data,
-        stored_context_width,
-        stride,
-        columns,
-        subsequence_start_mode="distribute",
-    )
-
-    assert len(result["col1"]) == 1
-    assert result["col1"][0] == [0, 0, 0, 1, 2]
-
-
-def test_extract_subsequences_returns_left_pad_lengths_when_requested():
-    input_data = {"col1": [0.0, 1.5]}
-    result, left_pad_lengths, subsequence_starts = extract_subsequences(
-        input_data,
-        stored_context_width=5,
-        stride_for_split=1,
-        columns=["col1"],
-        subsequence_start_mode="distribute",
-    )
-
-    assert result["col1"][0] == [0, 0, 0, 0.0, 1.5]
-    assert left_pad_lengths == [3]
-
-
-def test_extract_sequences_persists_left_pad_length_metadata():
-    schema = {
-        "sequenceId": pl.Int64,
-        "subsequenceId": pl.Int64,
-        "startItemPosition": pl.Int64,
-        "leftPadLength": pl.Int64,
-        "inputCol": pl.String,
-        "4": pl.Float64,
-        "3": pl.Float64,
-        "2": pl.Float64,
-        "1": pl.Float64,
-        "0": pl.Float64,
-    }
-    data = pl.DataFrame(
-        {
-            "sequenceId": [10, 10],
-            "itemPosition": [0, 1],
-            "col1": [0.0, 1.5],
-        }
-    )
-
-    sequences = extract_sequences(
-        data,
-        schema,
-        layout=StoredWindowLayout(
-            stored_context_width=5, max_target_offset=1, version=2
-        ),
-        stride_for_split=1,
-        columns=["col1"],
-        subsequence_start_mode="distribute",
-    )
-
-    assert sequences.get_column("leftPadLength").to_list() == [3]
-
-
-def test_process_and_write_data_pt_persists_left_pad_lengths(tmp_path):
-    data = pl.DataFrame(
-        {
-            "sequenceId": [1, 2],
-            "subsequenceId": [0, 0],
-            "startItemPosition": [0, 0],
-            "leftPadLength": [0, 2],
-            "inputCol": ["col1", "col1"],
-            "3": [1.0, 0.0],
-            "2": [2.0, 0.0],
-            "1": [3.0, 1.0],
-            "0": [4.0, 2.0],
-        }
-    )
-    out_path = tmp_path / "batch.pt"
-
-    process_and_write_data_pt(
-        data,
-        stored_context_width=4,
-        path=str(out_path),
-        column_types={"col1": "Float64"},
-    )
-
-    sequences, _, _, _, left_pad_lengths = torch.load(out_path, weights_only=False)
-
-    assert torch.equal(left_pad_lengths, torch.tensor([0, 2]))
-    assert torch.equal(
-        sequences["col1"],
-        torch.tensor([[1.0, 2.0, 3.0, 4.0], [0.0, 0.0, 1.0, 2.0]]),
-    )
-
-
-def test_mask_column_is_not_a_data_column():
-    data = pl.DataFrame(
-        {
-            "sequenceId": [1],
-            "itemPosition": [0],
-            "itemId": [101],
-            RESERVED_MASK_COLUMN: [1],
-        }
-    )
-
-    assert _get_data_columns(data, RESERVED_MASK_COLUMN) == ["itemId"]
-
-
-def test_apply_mask_column_replaces_values_and_drops_column():
-    data = pl.DataFrame(
-        {
-            "cat_col": [3, 4, 5, 6],
-            "num_col": [-1.0, 2.5, 3.5, 4.5],
-            RESERVED_MASK_COLUMN: [0, 1, 0, 1],
-        }
-    )
-
-    masked = _apply_mask_column(
-        data,
-        ["cat_col", "num_col"],
-        {"cat_col": "Int64", "num_col": "Float64"},
-        RESERVED_MASK_COLUMN,
-    )
-
-    assert RESERVED_MASK_COLUMN not in masked.columns
-    assert masked["cat_col"].to_list() == [3, 2, 5, 2]
-    assert masked["num_col"].to_list() == [-1.0, 0.0, 3.5, 0.0]
-
-
-def test_preprocessor_applies_mask_column_end_to_end(tmp_path):
-    project_root = tmp_path / "project"
-    project_root.mkdir()
-    metadata_dir = project_root / "configs" / "metadata_configs"
-    metadata_dir.mkdir(parents=True)
-    metadata_config_path = "configs/metadata_configs/masked-input-base.json"
-    (project_root / metadata_config_path).write_text(
-        json.dumps(
-            {
-                "n_classes": {"itemId": 6},
-                "id_maps": {"itemId": {"a": 3, "b": 4, "c": 5}},
-                "split_paths": [],
-                "column_types": {"itemId": "Int64", "itemValue": "Float64"},
-                "selected_columns_statistics": {
-                    "itemValue": {"mean": 60.0, "std": 10.0}
-                },
-                "stored_context_width": 3,
-                "max_target_offset": 1,
-                "stored_window_layout_version": 2,
-                "special_token_ids": {"[unknown]": 0, "[other]": 1, "[mask]": 2},
-            }
-        )
-    )
-    data_path = tmp_path / "masked-input.csv"
-    pl.DataFrame(
-        {
-            "sequenceId": [0, 0, 0],
-            "itemPosition": [0, 1, 2],
-            "itemId": ["a", "b", "c"],
-            "itemValue": [10.0, 100.0, 70.0],
-            RESERVED_MASK_COLUMN: [0, 1, 0],
-        }
-    ).write_csv(data_path)
-
-    Preprocessor(
-        project_root=str(project_root),
-        continue_preprocessing=False,
-        data_path=str(data_path),
-        read_format="csv",
-        write_format="parquet",
-        merge_output=True,
-        selected_columns=["itemId", "itemValue"],
-        split_ratios=[1.0],
-        stored_context_width=3,
-        max_target_offset=1,
-        stride_by_split=[1],
-        max_rows=None,
-        seed=1010,
-        n_cores=1,
-        batches_per_file=1024,
-        process_by_file=True,
-        subsequence_start_mode="distribute",
-        use_precomputed_maps=None,
-        metadata_config_path=metadata_config_path,
-        mask_column=RESERVED_MASK_COLUMN,
-    )
-
-    output = pl.read_parquet(project_root / "data" / "masked-input-split0.parquet")
-    metadata_path = project_root / "configs" / "metadata_configs" / "masked-input.json"
-
-    item_sequence = (
-        output.filter(pl.col("inputCol") == "itemId").select(["2", "1", "0"]).row(0)
-    )
-    value_sequence = (
-        output.filter(pl.col("inputCol") == "itemValue").select(["2", "1", "0"]).row(0)
-    )
-
-    assert RESERVED_MASK_COLUMN not in output.columns
-    assert item_sequence == (3.0, 2.0, 5.0)
-    assert value_sequence[1] == 0.0
-    assert value_sequence[0] != 0.0
-    assert value_sequence[2] != 0.0
-
-    metadata = json.loads(metadata_path.read_text())
-    assert RESERVED_MASK_COLUMN not in metadata["column_types"]
-    assert RESERVED_MASK_COLUMN not in metadata["id_maps"]
-    assert RESERVED_MASK_COLUMN not in metadata["selected_columns_statistics"]
-    assert metadata["special_token_ids"] == {
-        "[unknown]": 0,
-        "[other]": 1,
-        "[mask]": 2,
-    }
-
-
-def test_preprocessor_writes_expanded_resume_manifest(tmp_path):
-    project_root = tmp_path / "project"
-    project_root.mkdir()
-    data_path = tmp_path / "manifest-input.csv"
-    pl.DataFrame(
-        {
-            "sequenceId": [0, 0, 0],
-            "itemPosition": [0, 1, 2],
-            "itemId": ["a", "b", "c"],
-        }
-    ).write_csv(data_path)
-
-    Preprocessor(
-        project_root=str(project_root),
-        continue_preprocessing=False,
-        data_path=str(data_path),
-        read_format="csv",
-        write_format="parquet",
-        merge_output=False,
-        selected_columns=["itemId"],
-        split_ratios=[1.0],
-        stored_context_width=3,
-        max_target_offset=1,
-        stride_by_split=[2],
-        max_rows=2,
-        seed=1010,
-        n_cores=1,
-        batches_per_file=1024,
-        process_by_file=True,
-        subsequence_start_mode="exact",
-        use_precomputed_maps=None,
-        metadata_config_path=None,
-        mask_column=None,
-    )
-
-    manifest_path = (
-        project_root / "data" / "manifest-input-temp" / "preprocess-manifest.json"
-    )
-    manifest = json.loads(manifest_path.read_text())
-    config = manifest["preprocessing_config"]
-
-    assert manifest["manifest_version"] == 1
-    assert config["read_format"] == "csv"
-    assert config["write_format"] == "parquet"
-    assert config["merge_output"] is False
-    assert config["selected_columns"] == ["sequenceId", "itemPosition", "itemId"]
-    assert config["data_columns"] == ["itemId"]
-    assert config["split_ratios"] == [1.0]
-    assert config["stride_by_split"] == [2]
-    assert config["max_rows"] == 2
-    assert config["process_by_file"] is True
-    assert config["subsequence_start_mode"] == "exact"
-    assert config["mask_column"] is None
-    assert config["metadata_config_path"] is None
-    assert config["use_precomputed_maps"] is None
-    assert manifest["effective_metadata_digest"]["algorithm"] == "sha256"
-    assert len(manifest["effective_metadata_digest"]["value"]) == 64
-
-
-def test_resume_manifest_rejects_changed_preprocessing_config(tmp_path):
-    project_root = tmp_path / "project"
-    manifest_dir = project_root / "data" / "input-temp"
-    manifest_dir.mkdir(parents=True)
-
-    preprocessor = object.__new__(Preprocessor)
-    preprocessor.project_root = str(project_root)
-    preprocessor.target_dir = "input-temp"
-    preprocessor.continue_preprocessing = False
-    preprocessor.storage_layout = StoredWindowLayout(
-        stored_context_width=3,
-        max_target_offset=1,
-        version=2,
-    )
-    preprocessor.data_path = "data/input.csv"
-    preprocessor.data_name_root = "input"
-    preprocessor.read_format = "csv"
-    preprocessor.merge_output = False
-    preprocessor.split_ratios = [1.0]
-    preprocessor.stride_by_split = [1]
-    preprocessor.max_rows = None
-    preprocessor.process_by_file = True
-    preprocessor.subsequence_start_mode = "distribute"
-    preprocessor.mask_column = None
-    preprocessor.metadata_config_path = None
-    preprocessor.use_precomputed_maps = None
-
-    preprocessor._write_or_validate_resume_manifest(
-        selected_columns=None,
-        write_format="parquet",
-        data_columns=["itemId"],
-        id_maps={"itemId": {"a": 3}},
-        n_classes={"itemId": 4},
-        col_types={"itemId": "Int64"},
-        selected_columns_statistics={},
-    )
-
-    preprocessor.continue_preprocessing = True
-    preprocessor.stride_by_split = [2]
-
-    with pytest.raises(ValueError, match="different preprocessing manifest"):
-        preprocessor._write_or_validate_resume_manifest(
-            selected_columns=None,
-            write_format="parquet",
-            data_columns=["itemId"],
-            id_maps={"itemId": {"a": 3}},
-            n_classes={"itemId": 4},
-            col_types={"itemId": "Int64"},
-            selected_columns_statistics={},
-        )
-
-
-def test_load_and_preprocess_data_requests_mask_column_for_csv_projection(tmp_path):
-    data_path = tmp_path / "masked-input.csv"
-    captured_columns = None
-
-    def projected_read_data(path, read_format, columns=None):
-        nonlocal captured_columns
-        captured_columns = columns
-        assert path == str(data_path)
-        assert read_format == "csv"
-        assert columns is not None
-        assert RESERVED_MASK_COLUMN in columns
-        return pl.DataFrame(
-            {
-                "sequenceId": [0],
-                "itemPosition": [0],
-                "itemId": ["a"],
-                "itemValue": [1.0],
-                RESERVED_MASK_COLUMN: [1],
-            }
-        )
-
-    with pytest.MonkeyPatch.context() as monkeypatch:
-        monkeypatch.setattr("sequifier.preprocess.read_data", projected_read_data)
-        data = _load_and_preprocess_data(
-            str(data_path),
-            "csv",
-            ["itemId", "itemValue"],
-            None,
-            RESERVED_MASK_COLUMN,
-        )
-
-    assert captured_columns == ["itemId", "itemValue", RESERVED_MASK_COLUMN]
-    assert data.columns == [
-        "sequenceId",
-        "itemPosition",
-        "itemId",
-        "itemValue",
-        RESERVED_MASK_COLUMN,
-    ]
-
-
-def test_load_and_preprocess_data_fails_when_mask_column_is_missing(tmp_path):
-    csv_path = tmp_path / "input.csv"
-    pl.DataFrame({"sequenceId": [0], "itemPosition": [0], "itemId": ["a"]}).write_csv(
-        csv_path
-    )
-
-    with pytest.raises(ValueError, match="mask_column '.+' not found"):
-        _load_and_preprocess_data(
-            str(csv_path),
-            "csv",
-            None,
-            None,
-            RESERVED_MASK_COLUMN,
-        )
-
-    parquet_path = tmp_path / "input.parquet"
-    pl.DataFrame(
-        {"sequenceId": [0], "itemPosition": [0], "itemId": ["a"]}
-    ).write_parquet(parquet_path)
-
-    with pytest.raises(ValueError, match="mask_column '.+' not found"):
-        _load_and_preprocess_data(
-            str(parquet_path),
-            "parquet",
-            ["itemId"],
-            None,
-            RESERVED_MASK_COLUMN,
-        )
-
-
-def test_preprocessor_requires_metadata_config_for_mask_column(tmp_path):
-    project_root = tmp_path / "project"
-    project_root.mkdir()
-    data_path = tmp_path / "masked-input.csv"
-    pl.DataFrame(
-        {
-            "sequenceId": [0],
-            "itemPosition": [0],
-            "itemId": ["a"],
-            RESERVED_MASK_COLUMN: [0],
-        }
-    ).write_csv(data_path)
-
-    with pytest.raises(
-        ValueError,
-        match="metadata_config_path must be set when mask_column is set",
-    ):
-        Preprocessor(
-            project_root=str(project_root),
-            continue_preprocessing=False,
-            data_path=str(data_path),
-            read_format="csv",
-            write_format="parquet",
-            merge_output=True,
-            selected_columns=["itemId"],
-            split_ratios=[1.0],
-            stored_context_width=2,
-            max_target_offset=1,
-            stride_by_split=[1],
-            max_rows=None,
-            seed=1010,
-            n_cores=1,
-            batches_per_file=1024,
-            process_by_file=True,
-            subsequence_start_mode="distribute",
-            use_precomputed_maps=None,
-            metadata_config_path=None,
-            mask_column=RESERVED_MASK_COLUMN,
-        )
-
-
-def test_preprocessor_config_defaults_mask_column_to_none(tmp_path):
-    data_path = tmp_path / "input.csv"
-    pl.DataFrame(
-        {
-            "sequenceId": [0],
-            "itemPosition": [0],
-            "itemId": ["a"],
-            RESERVED_MASK_COLUMN: [0],
-        }
-    ).write_csv(data_path)
-
-    config = PreprocessorModel(
-        project_root=str(tmp_path),
-        data_path=str(data_path),
-        split_ratios=[1.0],
-        stored_context_width=2,
-        max_target_offset=1,
-        seed=1010,
-    )
-
-    assert config.mask_column is None
-
-
-def test_preprocessor_config_requires_metadata_config_for_mask_column(
-    tmp_path,
-):
-    data_path = tmp_path / "input.csv"
-    pl.DataFrame(
-        {
-            "sequenceId": [0],
-            "itemPosition": [0],
-            "itemId": ["a"],
-            RESERVED_MASK_COLUMN: [0],
-        }
-    ).write_csv(data_path)
-
-    with pytest.raises(
-        ValidationError,
-        match="metadata_config_path must be set when mask_column is set",
-    ):
-        PreprocessorModel(
-            project_root=str(tmp_path),
-            data_path=str(data_path),
-            split_ratios=[1.0],
-            stored_context_width=2,
-            max_target_offset=1,
-            seed=1010,
-            mask_column=RESERVED_MASK_COLUMN,
-        )
-
-
-@pytest.mark.parametrize("mode", ["distribute", "exact"])
-def test_extract_subsequences_modes(mode):
-    """Distribute vs exact stride modes."""
-    # Length 10. Seq_len 2 (window 3).
-    # distribute: adjusts stride to cover data evenly.
-    # exact: strictly adheres to stride, throws error if misalignment.
-    input_data = {"col1": list(range(10))}
-    context_length = 2
-    stored_context_width = context_length + 1
-    columns = ["col1"]
-
-    if mode == "distribute":
-        stride = 4
-        # distribute might adjust indices to maximize coverage
-        result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data, stored_context_width, stride, columns, mode
-        )
-        assert len(result["col1"]) > 0
-
-    elif mode == "exact":
-        stride = (
-            3  # (10-1) - 2 = 7. 7 % 3 != 0. Should fail or require specific alignment
-        )
-        # Testing a failing exact case
-        with pytest.raises(ValueError):
-            extract_subsequences(input_data, stored_context_width, 4, columns, mode)
-
-        # Testing a passing exact case
-        # (10-1) - 2 = 7. If we change input len to 11: (11-1)-2 = 8. stride 4 works.
-        input_data_exact = {"col1": list(range(11))}
-        result, left_pad_lengths, subsequence_starts = extract_subsequences(
-            input_data_exact, stored_context_width, 4, columns, mode
-        )
-        assert len(result["col1"]) > 0
-
-
-def test_get_batch_limits_perfect_split():
-    """Batch boundaries align with sequence boundaries."""
-    # 3 sequences, each length 10. Total 30 rows.
-    # If we want 3 batches, we should get 3 chunks of 10.
-    data = pl.DataFrame({"sequenceId": np.repeat([1, 2, 3], 10), "val": np.arange(30)})
-
-    limits = get_batch_limits(data, n_batches=3)
-
-    assert len(limits) == 3
-    assert limits == [(0, 10), (10, 20), (20, 30)]
-
-
-def test_get_batch_limits_uneven_split():
-    """Batches never split a sequenceId."""
-    # Seq 1: 5 rows
-    # Seq 2: 15 rows
-    # Seq 3: 5 rows
-    # Total 25 rows. Request 2 batches. Ideal size 12.5.
-    # Split point should occur at index 5 (Seq 1 end) or 20 (Seq 2 end),
-    # NOT at index 12 (middle of Seq 2).
-    data = pl.DataFrame(
-        {"sequenceId": np.repeat([1, 2, 3], [5, 15, 5]), "val": np.arange(25)}
-    )
-
-    limits = get_batch_limits(data, n_batches=2)
-
-    assert len(limits) == 2
-    assert limits[0][0] == 0
-    assert limits[-1][1] == data.height
-    assert all(start < end for start, end in limits)
-    assert all(
-        left_end == right_start
-        for (_, left_end), (right_start, _) in zip(limits, limits[1:])
-    )
-
-    # Check that split points are valid sequence boundaries
-    for start, end in limits:
-        # Start of batch must match start of a sequence (unless 0)
-        if start != 0:
-            prev_id = data["sequenceId"][start - 1]
-            curr_id = data["sequenceId"][start]
-            assert prev_id != curr_id, f"Batch split at {start} broke a sequence"
-        if end != data.height:
-            prev_id = data["sequenceId"][end - 1]
-            curr_id = data["sequenceId"][end]
-            assert prev_id != curr_id, f"Batch split at {end} broke a sequence"
-
-
-def test_get_batch_limits_rejects_too_many_nonempty_batches():
-    """Too many requested chunks fail instead of producing empty batches."""
-    data = pl.DataFrame({"sequenceId": [1, 2, 3, 3], "val": np.arange(4)})
-
-    with pytest.raises(ValueError, match="more non-empty batches"):
-        get_batch_limits(data, n_batches=4)
-
-
-def test_get_combined_statistics_logic():
-    """Merged Welford stats."""
-    # Create two chunks of data
-    chunk1 = np.random.normal(loc=10, scale=2, size=100)
-    chunk2 = np.random.normal(loc=20, scale=5, size=50)
-    full_data = np.concatenate([chunk1, chunk2])
-
-    # Real stats
-    mean1, std1 = np.mean(chunk1), np.std(chunk1, ddof=1)
-    mean2, std2 = np.mean(chunk2), np.std(chunk2, ddof=1)
-
-    # Function output
-    comb_mean, comb_std = get_combined_statistics(
-        len(chunk1), float(mean1), float(std1), len(chunk2), float(mean2), float(std2)
-    )
-
-    # Expected stats
-    expected_mean = np.mean(full_data)
-    expected_std = np.std(full_data, ddof=1)
-
-    np.testing.assert_almost_equal(comb_mean, expected_mean)
-    np.testing.assert_almost_equal(comb_std, expected_std)
-
-
-def test_get_column_statistics_state_accumulation():
-    """Chunked and full-pass stats match."""
-    data_full = pl.DataFrame(
-        {
-            "cat_col": ["a", "b", "a", "c", "b", "d"],
-            "num_col": [1.0, 2.0, 3.0, 4.0, 5.0, 6.0],
-        }
-    )
-
-    # Split into two chunks
-    chunk1 = data_full.slice(0, 3)
-    chunk2 = data_full.slice(3, 3)
-
-    id_maps = {}
-    stats = {}
-    running_count = 0
-
-    # Pass 1
-    id_maps, stats = _get_column_statistics(
-        chunk1, ["cat_col", "num_col"], id_maps, stats, running_count, {}
-    )
-    running_count += len(chunk1)
-
-    # Pass 2
-    id_maps, stats = _get_column_statistics(
-        chunk2, ["cat_col", "num_col"], id_maps, stats, running_count, {}
-    )
-
-    assert len(id_maps["cat_col"]) == 6
-    assert set(id_maps["cat_col"].keys()) == {
-        "[unknown]",
-        "[other]",
-        "a",
-        "b",
-        "c",
-        "d",
-    }
-
-    expected_mean = data_full["num_col"].mean()
-    expected_std = data_full["num_col"].std()
-
-    np.testing.assert_almost_equal(stats["num_col"]["mean"], expected_mean)
-    np.testing.assert_almost_equal(stats["num_col"]["std"], expected_std)
-
-
-def test_get_column_statistics_skips_all_masked_chunks():
-    data = pl.DataFrame(
-        {
-            "cat_col": ["a", "b"],
-            "num_col": [1.0, 2.0],
-            RESERVED_MASK_COLUMN: [1, 1],
-        }
-    )
-
-    id_maps, stats = _get_column_statistics(
-        data,
-        ["cat_col", "num_col"],
-        {},
-        {},
-        0,
-        {},
-        mask_column=RESERVED_MASK_COLUMN,
-    )
-
-    assert id_maps == {}
-    assert stats == {}
-
-
-def test_apply_column_statistics_errors_when_all_rows_masked():
-    data = pl.DataFrame({"cat_col": ["a"], "num_col": [1.0]})
-
-    with pytest.raises(ValueError, match="No unmasked examples"):
-        _apply_column_statistics(data, ["cat_col", "num_col"], {}, {})
-
-
-def test_create_id_map():
-    """Basic ID map creation."""
-    df = pl.DataFrame({"A": ["z", "x", "y", "x"]})
-    mapping = create_id_map(df, "A")
-
-    # Sorted unique values: x, y, z -> 2, 3, 4
-    assert mapping["x"] == 3
-    assert mapping["y"] == 4
-    assert mapping["z"] == 5
-
-
-def test_load_precomputed_id_maps_allows_reserved_entries(tmp_path):
-    project_root = tmp_path / "project"
-    id_map_dir = project_root / "configs" / "id_maps"
-    id_map_dir.mkdir(parents=True)
-    (id_map_dir / "itemId.json").write_text(
-        json.dumps(
-            {
-                "[unknown]": 0,
-                "[other]": 1,
-                "[mask]": 2,
-                "a": 3,
-                "b": 4,
-            }
-        )
-    )
-
-    id_maps = load_precomputed_id_maps(str(project_root), ["itemId"])
-
-    assert id_maps["itemId"]["[mask]"] == 2
-    assert id_maps["itemId"]["a"] == 3
diff --git a/tests/unit/test_probabilities.py b/tests/unit/test_probabilities.py
deleted file mode 100644
index d413d841..00000000
--- a/tests/unit/test_probabilities.py
+++ /dev/null
@@ -1,51 +0,0 @@
-import pytest
-import torch
-
-from sequifier.config.probabilities import (
-    GeometricDistribution,
-    LogNormalDistributionDiscretizedFloor,
-    NormalDistributionDiscretizedFloor,
-    PoissonDistributionFloor,
-)
-
-
-@pytest.mark.parametrize(
-    "distribution",
-    [
-        GeometricDistribution(p=0.35),
-        NormalDistributionDiscretizedFloor(mean=2.0, standard_deviation=0.5),
-        LogNormalDistributionDiscretizedFloor(mean=0.5, standard_deviation=0.4),
-        PoissonDistributionFloor(rate=2.0),
-    ],
-)
-def test_probability_distributions_use_local_generator_without_global_rng(
-    distribution,
-):
-    device = torch.device("cpu")
-    torch.manual_seed(123)
-    rng_state = torch.get_rng_state().clone()
-
-    first_generator = torch.Generator(device=device)
-    first_generator.manual_seed(99)
-    first_samples = distribution.sample(
-        (64,),
-        device=device,
-        generator=first_generator,
-    )
-    after_first = torch.get_rng_state().clone()
-
-    second_generator = torch.Generator(device=device)
-    second_generator.manual_seed(99)
-    second_samples = distribution.sample(
-        (64,),
-        device=device,
-        generator=second_generator,
-    )
-    after_second = torch.get_rng_state().clone()
-
-    assert torch.equal(first_samples, second_samples)
-    assert first_samples.device == device
-    assert first_samples.dtype == torch.long
-    assert torch.all(first_samples >= 1)
-    assert torch.equal(after_first, rng_state)
-    assert torch.equal(after_second, rng_state)
diff --git a/tests/unit/test_train.py b/tests/unit/test_train.py
deleted file mode 100644
index 954076b7..00000000
--- a/tests/unit/test_train.py
+++ /dev/null
@@ -1,1529 +0,0 @@
-import copy
-import json
-from types import SimpleNamespace
-
-import numpy as np
-import pytest
-import torch
-import yaml
-from pydantic import ValidationError
-from torch.utils.data import DataLoader
-
-from sequifier.config.probabilities import PoissonDistributionFloor
-from sequifier.config.train_config import (
-    BERTSpecModel,
-    ModelSpecModel,
-    ReplacementDistribution,
-    TrainingSpecModel,
-    TrainModel,
-    load_train_config,
-)
-from sequifier.helpers import ModelWindowView, StoredWindowLayout
-from sequifier.io.batch import SequifierBatch
-from sequifier.special_tokens import SPECIAL_TOKEN_IDS
-from sequifier.train import (
-    TransformerModel,
-    _get_evaluation_loss_mask,
-    accumulate_class_counts,
-)
-
-
-def _training_spec_kwargs(**overrides):
-    values = {
-        "training_objective": "causal",
-        "device": "cpu",
-        "epochs": 1,
-        "save_interval_epochs": 1,
-        "batch_size": 4,
-        "learning_rate": 0.001,
-        "criterion": {"cat_col": "CrossEntropyLoss", "real_col": "MSELoss"},
-        "optimizer": {"name": "Adam"},
-        "scheduler": {"name": "StepLR", "step_size": 1, "gamma": 0.1},
-        "loss_weights": {"cat_col": 1.0, "real_col": 1.0},
-    }
-    values.update(overrides)
-    return values
-
-
-def _bert_spec():
-    return BERTSpecModel(
-        masking_probability=0.5,
-        replacement_distribution={
-            "masked": 1.0,
-            "random": 0.0,
-            "identical": 0.0,
-        },
-        span_masking={"type": "GeometricDistribution", "p": 1.0},
-    )
-
-
-def test_replacement_distribution_allows_zero_probabilities():
-    distribution = ReplacementDistribution(masked=1.0, random=0.0, identical=0.0)
-
-    assert distribution.masked == 1.0
-    assert distribution.random == 0.0
-    assert distribution.identical == 0.0
-
-
-def test_training_spec_model_requires_bert_spec_for_bert_objective():
-    with pytest.raises(ValidationError, match="BERT hyperparameters must be set"):
-        TrainingSpecModel(**_training_spec_kwargs(training_objective="bert"))
-
-
-def test_training_spec_model_rejects_bert_spec_for_causal_objective():
-    with pytest.raises(
-        ValidationError,
-        match="BERT hyperparameters should only be configured",
-    ):
-        TrainingSpecModel(**_training_spec_kwargs(bert_spec=_bert_spec()))
-
-
-def test_training_spec_model_dump_excludes_runtime_offsets():
-    training_spec = TrainingSpecModel(**_training_spec_kwargs())
-
-    dumped = training_spec.model_dump()
-
-    assert "data_offset" not in dumped
-    assert "target_offset" not in dumped
-    assert "stored_context_width" not in dumped
-    assert "max_target_offset" not in dumped
-
-
-def test_poisson_span_masking_samples_at_least_one_token():
-    distribution = PoissonDistributionFloor(rate=0.1)
-
-    samples = distribution.sample((1000,), device=torch.device("cpu"))
-
-    assert samples.min().item() >= 1
-
-
-@pytest.fixture
-def model_config(tmp_path):
-    """Valid TrainModel config."""
-    project_root = str(tmp_path)
-
-    # Ensure necessary directories exist to avoid init errors (logging)
-    (tmp_path / "logs").mkdir(exist_ok=True)
-
-    model_spec = ModelSpecModel(
-        initial_embedding_dim=16,
-        dim_model=16,
-        n_head=4,
-        dim_feedforward=32,
-        num_layers=2,
-        prediction_length=1,
-        # Embedding dims must sum to dim_model (15 + 1 = 16)
-        feature_embedding_dims={"cat_col": 15, "real_col": 1},
-    )
-
-    training_spec = TrainingSpecModel(
-        training_objective="causal",
-        device="cpu",
-        epochs=1,
-        save_interval_epochs=1,
-        batch_size=4,
-        learning_rate=0.001,
-        criterion={"cat_col": "CrossEntropyLoss", "real_col": "MSELoss"},
-        optimizer={"name": "Adam"},
-        scheduler={"name": "StepLR", "step_size": 1, "gamma": 0.1},
-        loss_weights={"cat_col": 1.0, "real_col": 1.0},
-    )
-
-    config = TrainModel(
-        project_root=project_root,
-        model_name="unit-test-model",
-        metadata_config_path="metadata.json",  # Dummy path
-        training_data_path="data/train.pt",  # Dummy path
-        validation_data_path="data/val.pt",  # Dummy path
-        input_columns=["cat_col", "real_col"],
-        target_columns=["cat_col", "real_col"],
-        target_column_types={"cat_col": "categorical", "real_col": "real"},
-        column_types={"cat_col": "int64", "real_col": "float64"},
-        categorical_columns=["cat_col"],
-        real_columns=["real_col"],
-        # id_maps is needed for constructing index_maps in model init
-        id_maps={"cat_col": {"a": 1, "b": 2, "c": 3, "d": 4}},
-        n_classes={"cat_col": 5},  # 0 + 4 classes
-        storage_layout=StoredWindowLayout(
-            stored_context_width=11, max_target_offset=1, version=2
-        ),
-        window_view=ModelWindowView(
-            context_length=10, objective="causal", target_offset=1
-        ),
-        inference_batch_size=4,
-        seed=42,
-        export_generative_model=True,
-        export_embedding_model=False,
-        model_spec=model_spec,
-        training_spec=training_spec,
-    )
-    return config
-
-
-@pytest.fixture
-def model(model_config):
-    """Instantiates the TransformerModel with the mock config."""
-    return TransformerModel(model_config)
-
-
-@pytest.fixture
-def causal_model(model_config):
-    config = copy.deepcopy(model_config)
-    config.training_spec.training_objective = "causal"
-    return TransformerModel(config)
-
-
-@pytest.fixture
-def bert_model(model_config):
-    config_values = model_config.model_dump()
-    config_values["model_spec"]["prediction_length"] = (
-        model_config.window_view.context_length
-    )
-    config_values["training_spec"] = _training_spec_kwargs(
-        training_objective="bert",
-        bert_spec=_bert_spec(),
-    )
-    config_values["window_view"] = {
-        **config_values["window_view"],
-        "objective": "bert",
-        "target_offset": 0,
-    }
-    config = TrainModel(**config_values)
-    return TransformerModel(config)
-
-
-def _all_valid_metadata(batch_size, seq_len, device=None):
-    valid_mask = torch.ones(
-        batch_size,
-        seq_len,
-        dtype=torch.bool,
-        device=device,
-    )
-    return {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-
-def _loss_shell_model(
-    target_column_types,
-    *,
-    loss_weights=None,
-    class_weights=None,
-):
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_columns = list(target_column_types)
-    model.target_column_types = dict(target_column_types)
-    model.loss_weights = loss_weights
-    model.device = "cpu"
-    model.criterion = {}
-    model.n_classes = {}
-
-    for col, col_type in target_column_types.items():
-        if col_type == "categorical":
-            weight = None
-            n_classes = 3
-            if class_weights is not None and col in class_weights:
-                col_class_weights = class_weights[col]
-                weight = torch.tensor(col_class_weights, dtype=torch.float32)
-                n_classes = len(col_class_weights)
-            model.criterion[col] = torch.nn.CrossEntropyLoss(
-                reduction="none",
-                weight=weight,
-            )
-            model.n_classes[col] = n_classes
-        elif col_type == "real":
-            model.criterion[col] = torch.nn.MSELoss(reduction="none")
-        else:
-            raise ValueError(col_type)
-
-    return model
-
-
-class _IdentityScaler:
-    def scale(self, loss):
-        return loss
-
-    def unscale_(self, optimizer):
-        return None
-
-    def step(self, optimizer):
-        optimizer.step()
-
-    def update(self):
-        return None
-
-
-class _NoopLogger:
-    def info(self, *args, **kwargs):
-        return None
-
-    def warning(self, *args, **kwargs):
-        return None
-
-
-def test_transformer_model_initialization(model, model_config):
-    """Expected model layers."""
-    # Check if encoder dicts were created
-    assert "cat_col" in model.encoder
-    assert model.pos_encoder is None or "cat_col" in model.pos_encoder
-
-    # Check decoder existence
-    assert "cat_col" in model.decoder
-    assert "real_col" in model.decoder
-
-    # Check embedding sizes
-    assert model.dim_model == 16
-    assert model.encoder["cat_col"].embedding_dim == 15
-    # Real column 'real_col' has feature_embedding_dims=1, but it goes through a Linear layer
-    # if dim > 1, or direct if dim == 1. In setup:
-    # if feature_embedding_dims[col] > 1 -> Linear
-    # else -> checked to be 1, then no encoder added to self.encoder for 'direct' columns?
-    # Let's check logic in train.py:
-    #   if self.feature_embedding_dims[col] > 1: ... self.encoder[col] = nn.Linear(...)
-    #   else: self.real_columns_direct.append(col)
-    # Since we set real_col dim to 1, it should NOT be in model.encoder
-    if model.feature_embedding_dims["real_col"] == 1:
-        assert "real_col" not in model.encoder
-    else:
-        assert "real_col" in model.encoder
-
-
-def test_train_model_requires_bert_prediction_length_to_equal_context_length(
-    model_config,
-):
-    config_values = model_config.model_dump()
-    config_values["model_spec"]["prediction_length"] = (
-        model_config.window_view.context_length - 1
-    )
-    config_values["training_spec"] = _training_spec_kwargs(
-        training_objective="bert",
-        bert_spec=_bert_spec(),
-    )
-    config_values["window_view"] = {
-        **config_values["window_view"],
-        "objective": "bert",
-        "target_offset": 0,
-    }
-
-    with pytest.raises(ValidationError, match="prediction_length must be equal"):
-        TrainModel(**config_values)
-
-
-def test_train_model_rejects_mismatched_special_token_ids(model_config):
-    config_values = model_config.model_dump()
-    config_values["special_token_ids"] = {
-        "[unknown]": 10,
-        "[other]": 11,
-        "[mask]": 12,
-    }
-
-    with pytest.raises(ValidationError, match="special_token_ids must match"):
-        TrainModel(**config_values)
-
-
-def test_load_train_config_rejects_mismatched_metadata_special_token_ids(
-    tmp_path, model_config
-):
-    config_path = tmp_path / "train.yaml"
-    metadata_path = tmp_path / "metadata.json"
-    config_values = model_config.model_dump()
-    config_values["project_root"] = str(tmp_path)
-    config_values["metadata_config_path"] = metadata_path.name
-    storage_layout = config_values.pop("storage_layout")
-    config_values.pop("window_view")
-    config_values["context_length"] = model_config.window_view.context_length
-    config_path.write_text(yaml.safe_dump(config_values))
-    metadata_path.write_text(
-        json.dumps(
-            {
-                "split_paths": ["data/train.pt", "data/val.pt"],
-                "stored_context_width": storage_layout["stored_context_width"],
-                "max_target_offset": storage_layout["max_target_offset"],
-                "stored_window_layout_version": storage_layout["version"],
-                "column_types": config_values["column_types"],
-                "n_classes": config_values["n_classes"],
-                "id_maps": config_values["id_maps"],
-                "special_token_ids": {
-                    "[unknown]": 10,
-                    "[other]": 11,
-                    "[mask]": 12,
-                },
-            }
-        )
-    )
-
-    with pytest.raises(ValueError, match="special_token_ids must match"):
-        load_train_config(str(config_path), {}, skip_metadata=False)
-
-
-def test_load_train_config_defaults_missing_metadata_special_token_ids(
-    tmp_path, model_config
-):
-    config_path = tmp_path / "train.yaml"
-    metadata_path = tmp_path / "metadata.json"
-    config_values = model_config.model_dump()
-    config_values["project_root"] = str(tmp_path)
-    config_values["metadata_config_path"] = metadata_path.name
-    storage_layout = config_values.pop("storage_layout")
-    config_values.pop("window_view")
-    config_values["context_length"] = model_config.window_view.context_length
-    config_path.write_text(yaml.safe_dump(config_values))
-    metadata_path.write_text(
-        json.dumps(
-            {
-                "split_paths": ["data/train.pt", "data/val.pt"],
-                "stored_context_width": storage_layout["stored_context_width"],
-                "max_target_offset": storage_layout["max_target_offset"],
-                "stored_window_layout_version": storage_layout["version"],
-                "column_types": config_values["column_types"],
-                "n_classes": config_values["n_classes"],
-                "id_maps": config_values["id_maps"],
-            }
-        )
-    )
-
-    config = load_train_config(str(config_path), {}, skip_metadata=False)
-
-    assert config.special_token_ids == SPECIAL_TOKEN_IDS.ids_by_label
-
-
-@pytest.mark.parametrize(
-    ("class_share_column", "metadata_overrides", "config_overrides", "match"),
-    [
-        (
-            "missing_col",
-            {},
-            {},
-            "Class-share column 'missing_col' must be a target column",
-        ),
-        (
-            "real_col",
-            {},
-            {},
-            "Class-share column 'real_col' must be a categorical target column",
-        ),
-        (
-            "cat_col",
-            {"n_classes": {}},
-            {"drop_n_classes": True},
-            "Class-share column 'cat_col' has no configured class count",
-        ),
-        (
-            "cat_col",
-            {"id_maps": {}},
-            {},
-            "Class-share column 'cat_col' has no index map for logging",
-        ),
-    ],
-)
-def test_load_train_config_validates_class_share_columns_after_metadata_load(
-    tmp_path,
-    model_config,
-    class_share_column,
-    metadata_overrides,
-    config_overrides,
-    match,
-):
-    config_path = tmp_path / "train.yaml"
-    metadata_path = tmp_path / "metadata.json"
-    config_values = model_config.model_dump()
-    config_values["project_root"] = str(tmp_path)
-    config_values["metadata_config_path"] = metadata_path.name
-    config_values["training_spec"]["class_share_log_columns"] = [class_share_column]
-
-    storage_layout = config_values.pop("storage_layout")
-    config_values.pop("window_view")
-    config_values["context_length"] = model_config.window_view.context_length
-
-    if config_overrides.get("drop_n_classes"):
-        config_values.pop("n_classes")
-
-    metadata = {
-        "split_paths": ["data/train.pt", "data/val.pt"],
-        "stored_context_width": storage_layout["stored_context_width"],
-        "max_target_offset": storage_layout["max_target_offset"],
-        "stored_window_layout_version": storage_layout["version"],
-        "column_types": config_values["column_types"],
-        "n_classes": model_config.n_classes,
-        "id_maps": model_config.id_maps,
-        "special_token_ids": SPECIAL_TOKEN_IDS.ids_by_label,
-    }
-    metadata.update(metadata_overrides)
-
-    config_path.write_text(yaml.safe_dump(config_values))
-    metadata_path.write_text(json.dumps(metadata))
-
-    with pytest.raises(ValueError, match=match):
-        load_train_config(str(config_path), {}, skip_metadata=False)
-
-
-def test_forward_train_shapes(model, model_config):
-    """forward_train output shapes."""
-    batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.window_view.context_length
-
-    # Create dummy inputs
-    # Categorical: (batch, seq_len) integers
-    x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
-    # Real: (batch, seq_len) floats
-    x_real = torch.randn(batch_size, seq_len)
-
-    src = {"cat_col": x_cat, "real_col": x_real}
-    metadata = _all_valid_metadata(batch_size, seq_len, device=x_cat.device)
-
-    # forward_train returns a dict of tensors
-    outputs = model.forward_train(src, metadata)
-
-    assert "cat_col" in outputs
-    assert "real_col" in outputs
-
-    # Expected output shape for training: (seq_len, batch_size, n_classes_or_1)
-    # Note: PyTorch Transformer default is (S, N, E) unless batch_first=True.
-    # Sequifier seems to use default, so (S, B, OutputDim)
-
-    out_cat = outputs["cat_col"]
-    assert out_cat.shape == (seq_len, batch_size, model_config.n_classes["cat_col"])
-
-    out_real = outputs["real_col"]
-    assert out_real.shape == (seq_len, batch_size, 1)
-
-
-def test_forward_inference_shapes(model, model_config):
-    """Inference output shapes."""
-    batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.window_view.context_length
-    prediction_length = model_config.model_spec.prediction_length  # 1
-
-    x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
-    x_real = torch.randn(batch_size, seq_len)
-    src = {"cat_col": x_cat, "real_col": x_real}
-    metadata = _all_valid_metadata(batch_size, seq_len, device=x_cat.device)
-
-    # forward returns predictions for the *last* prediction_length tokens
-    # And applies softmax to categorical outputs
-    outputs = model.forward(src, metadata)
-
-    # Expected shape: (prediction_length, batch_size, n_classes_or_1)
-    # If prediction_length is 1, dim 0 is size 1.
-
-    out_cat = outputs["cat_col"]
-    assert out_cat.shape == (
-        prediction_length,
-        batch_size,
-        model_config.n_classes["cat_col"],
-    )
-
-    # Check if Softmax/LogSoftmax was applied (values should be <= 0 for LogSoftmax)
-    # Sequifier uses LogSoftmax for categorical
-    assert (out_cat <= 0).all()
-
-    out_real = outputs["real_col"]
-    assert out_real.shape == (prediction_length, batch_size, 1)
-
-
-def test_calculate_loss(model, model_config):
-    """Scalar loss tensor."""
-    batch_size = model_config.training_spec.batch_size
-    seq_len = model_config.window_view.context_length
-
-    # Inputs
-    x_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
-    x_real = torch.randn(batch_size, seq_len)
-    src = {"cat_col": x_cat, "real_col": x_real}
-
-    # Targets (must match sequence length for training)
-    y_cat = torch.randint(0, model_config.n_classes["cat_col"], (batch_size, seq_len))
-    y_real = torch.randn(batch_size, seq_len)
-    targets = {"cat_col": y_cat, "real_col": y_real}
-    metadata = _all_valid_metadata(batch_size, seq_len, device=x_cat.device)
-
-    # Run forward pass
-    outputs = model.forward_train(src, metadata)
-
-    # Calculate loss
-    total_loss, component_losses = model._calculate_loss(outputs, targets, metadata)
-
-    # Assertions
-    assert total_loss.dim() == 0  # Scalar
-    assert total_loss.item() > 0  # Valid loss value
-    assert "cat_col" in component_losses
-    assert "real_col" in component_losses
-
-
-def test_calculate_loss_uses_explicit_target_mask_for_real_zero_targets():
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_column_types = {"real_col": "real"}
-    model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
-    model.loss_weights = None
-    model.categorical_columns = []
-    outputs = {
-        "real_col": torch.tensor(
-            [
-                [[1.0]],
-                [[2.0]],
-                [[1.0]],
-            ]
-        )
-    }
-    targets = {
-        "real_col": torch.tensor([[0.0, 0.0, 2.0]]),
-    }
-    metadata = {
-        "attention_valid_mask": torch.tensor([[True, True, True]]),
-        "target_valid_mask": torch.tensor([[True, True, True]]),
-    }
-
-    total_loss, component_losses = TransformerModel._calculate_loss(
-        model, outputs, targets, metadata
-    )
-
-    assert torch.isclose(total_loss, torch.tensor(2.0))
-    assert torch.isclose(component_losses["real_col"], torch.tensor(2.0))
-
-
-def test_calculate_loss_uses_explicit_target_mask_for_target_columns():
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_column_types = {"real_target": "real"}
-    model.criterion = {"real_target": torch.nn.MSELoss(reduction="none")}
-    model.loss_weights = None
-    model.categorical_columns = ["context_cat"]
-    outputs = {
-        "real_target": torch.tensor(
-            [
-                [[10.0]],
-                [[20.0]],
-                [[3.0]],
-            ]
-        )
-    }
-    targets = {
-        "real_target": torch.tensor([[0.0, 0.0, 2.0]]),
-    }
-    metadata = {
-        "attention_valid_mask": torch.tensor([[False, False, True]]),
-        "target_valid_mask": torch.tensor([[False, False, True]]),
-    }
-
-    total_loss, component_losses = TransformerModel._calculate_loss(
-        model, outputs, targets, metadata
-    )
-
-    assert torch.isclose(total_loss, torch.tensor(1.0))
-    assert torch.isclose(component_losses["real_target"], torch.tensor(1.0))
-
-
-def test_calculate_local_loss_components_categorical_target_with_padding():
-    model = _loss_shell_model({"cat_col": "categorical"})
-    model.n_classes["cat_col"] = 3
-    logits = torch.tensor(
-        [
-            [[2.0, 0.0, 0.0]],
-            [[0.0, 2.0, 0.0]],
-            [[0.0, 0.0, 2.0]],
-        ]
-    )
-    targets = {"cat_col": torch.tensor([[0, 1, 2]])}
-    valid_mask = torch.tensor([[True, False, True]])
-
-    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
-        model,
-        {"cat_col": logits},
-        targets,
-        valid_mask,
-    )
-    raw_loss = torch.nn.functional.cross_entropy(
-        logits.reshape(-1, 3),
-        targets["cat_col"].T.reshape(-1),
-        reduction="none",
-    )
-
-    assert torch.isclose(loss_sums["cat_col"], raw_loss[[0, 2]].sum())
-    assert torch.equal(token_count, torch.tensor(2))
-
-
-def test_calculate_local_loss_components_real_target_with_padding():
-    model = _loss_shell_model({"real_col": "real"})
-    output = {"real_col": torch.tensor([[[1.0]], [[2.0]], [[4.0]]])}
-    targets = {"real_col": torch.tensor([[0.0, 10.0, 1.0]])}
-    valid_mask = torch.tensor([[True, False, True]])
-
-    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
-        model, output, targets, valid_mask
-    )
-
-    assert torch.isclose(loss_sums["real_col"], torch.tensor(10.0))
-    assert torch.equal(token_count, torch.tensor(2))
-
-
-def test_calculate_training_loss_intersects_target_and_bert_masks():
-    model = _loss_shell_model({"real_col": "real"})
-    output = {"real_col": torch.tensor([[[1.0]], [[2.0]], [[4.0]]])}
-    targets = {"real_col": torch.tensor([[0.0, 10.0, 1.0]])}
-    metadata = {
-        "target_valid_mask": torch.tensor([[True, True, True]]),
-        "bert_mask": torch.tensor([[True, False, True]]),
-    }
-
-    total_loss, component_losses = TransformerModel._calculate_loss(
-        model, output, targets, metadata
-    )
-
-    assert torch.isclose(total_loss, torch.tensor(5.0))
-    assert torch.isclose(component_losses["real_col"], torch.tensor(5.0))
-
-
-def test_calculate_training_loss_applies_unequal_target_loss_weights():
-    model = _loss_shell_model(
-        {"cat_col": "categorical", "real_col": "real"},
-        loss_weights={"cat_col": 0.25, "real_col": 2.0},
-    )
-    model.n_classes["cat_col"] = 3
-    output = {
-        "cat_col": torch.tensor(
-            [
-                [[2.0, 0.0, 0.0]],
-                [[0.0, 2.0, 0.0]],
-            ]
-        ),
-        "real_col": torch.tensor([[[1.0]], [[3.0]]]),
-    }
-    targets = {
-        "cat_col": torch.tensor([[0, 1]]),
-        "real_col": torch.tensor([[0.0, 1.0]]),
-    }
-    metadata = {"target_valid_mask": torch.ones(1, 2, dtype=torch.bool)}
-
-    total_loss, component_losses = TransformerModel._calculate_loss(
-        model, output, targets, metadata
-    )
-    cat_mean = torch.nn.functional.cross_entropy(
-        output["cat_col"].reshape(-1, 3),
-        targets["cat_col"].T.reshape(-1),
-        reduction="mean",
-    )
-    real_mean = torch.nn.functional.mse_loss(
-        output["real_col"].reshape(-1),
-        targets["real_col"].T.reshape(-1),
-        reduction="mean",
-    )
-
-    assert torch.isclose(component_losses["cat_col"], cat_mean * 0.25)
-    assert torch.isclose(component_losses["real_col"], real_mean * 2.0)
-    assert torch.isclose(total_loss, cat_mean * 0.25 + real_mean * 2.0)
-
-
-def test_calculate_local_loss_components_keeps_class_weights_inside_criterion():
-    class_weights = {"cat_col": [1.0, 4.0, 1.0]}
-    model = _loss_shell_model({"cat_col": "categorical"}, class_weights=class_weights)
-    logits = torch.tensor(
-        [
-            [[2.0, 0.0, 0.0]],
-            [[2.0, 0.0, 0.0]],
-        ]
-    )
-    targets = {"cat_col": torch.tensor([[0, 1]])}
-    valid_mask = torch.ones(1, 2, dtype=torch.bool)
-
-    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
-        model,
-        {"cat_col": logits},
-        targets,
-        valid_mask,
-    )
-    expected = torch.nn.functional.cross_entropy(
-        logits.reshape(-1, 3),
-        targets["cat_col"].T.reshape(-1),
-        reduction="none",
-        weight=torch.tensor(class_weights["cat_col"]),
-    ).sum()
-
-    assert torch.isclose(loss_sums["cat_col"], expected)
-    assert torch.equal(token_count, torch.tensor(2))
-
-
-def test_calculate_local_loss_components_zero_selected_tokens_stays_connected():
-    model = _loss_shell_model({"real_col": "real"})
-    output = {"real_col": torch.ones(3, 1, 1, requires_grad=True)}
-    targets = {"real_col": torch.zeros(1, 3)}
-    valid_mask = torch.zeros(1, 3, dtype=torch.bool)
-
-    loss_sums, token_count = TransformerModel._calculate_local_loss_components(
-        model,
-        output,
-        targets,
-        valid_mask,
-    )
-    loss_sums["real_col"].backward()
-
-    assert torch.equal(token_count, torch.tensor(0))
-    assert torch.equal(loss_sums["real_col"].detach(), torch.tensor(0.0))
-    assert torch.equal(output["real_col"].grad, torch.zeros_like(output["real_col"]))
-
-
-def test_calculate_local_loss_components_raises_on_output_mask_mismatch():
-    model = _loss_shell_model({"real_col": "real"})
-    output = {"real_col": torch.zeros(3, 1, 1)}
-    targets = {"real_col": torch.zeros(1, 3)}
-    valid_mask = torch.ones(1, 2, dtype=torch.bool)
-
-    with pytest.raises(RuntimeError, match="Loss/mask size mismatch"):
-        TransformerModel._calculate_local_loss_components(
-            model,
-            output,
-            targets,
-            valid_mask,
-        )
-
-
-def test_training_and_validation_loss_finalization_share_weighted_sum_count_semantics():
-    model = _loss_shell_model(
-        {"cat_col": "categorical", "real_col": "real"},
-        loss_weights={"cat_col": 0.5, "real_col": 3.0},
-    )
-    model.n_classes["cat_col"] = 3
-    output = {
-        "cat_col": torch.tensor(
-            [
-                [[1.0, 0.0, 0.0]],
-                [[0.0, 1.0, 0.0]],
-                [[0.0, 0.0, 1.0]],
-            ]
-        ),
-        "real_col": torch.tensor([[[1.0]], [[2.0]], [[3.0]]]),
-    }
-    targets = {
-        "cat_col": torch.tensor([[0, 1, 2]]),
-        "real_col": torch.tensor([[0.0, 0.0, 0.0]]),
-    }
-    valid_mask = torch.tensor([[True, False, True]])
-    metadata = {"target_valid_mask": valid_mask}
-
-    training_loss, training_components = TransformerModel._calculate_loss(
-        model, output, targets, metadata
-    )
-    sums, count = TransformerModel._calculate_loss_components(
-        model, output, targets, valid_mask
-    )
-    finalized_loss, finalized_components = TransformerModel._finalize_loss_components(
-        model,
-        sums,
-        count.double(),
-        ["cat_col", "real_col"],
-        "test",
-    )
-
-    assert torch.isclose(training_loss.double(), finalized_loss)
-    assert torch.isclose(
-        training_components["cat_col"].double(), finalized_components["cat_col"]
-    )
-    assert torch.isclose(
-        training_components["real_col"].double(), finalized_components["real_col"]
-    )
-
-
-@pytest.mark.parametrize(
-    ("data_parallelism", "expected"),
-    [
-        ("DDP", 7),
-        ("FSDP", 7),
-    ],
-)
-def test_gradient_reduction_factor_uses_active_world_size(
-    data_parallelism,
-    expected,
-    monkeypatch,
-):
-    model = _loss_shell_model({"real_col": "real"})
-    model.hparams = SimpleNamespace(
-        training_spec=SimpleNamespace(data_parallelism=data_parallelism)
-    )
-    group = object()
-    model._data_parallel_process_group = lambda: group
-
-    monkeypatch.setattr("sequifier.train.dist.is_available", lambda: True)
-    monkeypatch.setattr("sequifier.train.dist.is_initialized", lambda: True)
-
-    def fake_get_world_size(group=None):
-        assert group is model._data_parallel_process_group()
-        return 7
-
-    monkeypatch.setattr("sequifier.train.dist.get_world_size", fake_get_world_size)
-
-    assert TransformerModel._gradient_reduction_factor(model) == expected
-
-
-def test_calculate_loss_requires_all_configured_targets():
-    model = _loss_shell_model(
-        {"cat_col": "categorical", "real_col": "real"},
-        loss_weights={"cat_col": 1.0, "real_col": 1.0},
-    )
-    output = {
-        "cat_col": torch.zeros(1, 1, 3),
-        "real_col": torch.zeros(1, 1, 1),
-    }
-    targets = {"cat_col": torch.zeros(1, 1, dtype=torch.long)}
-    metadata = {"target_valid_mask": torch.ones(1, 1, dtype=torch.bool)}
-
-    with pytest.raises(RuntimeError, match="Missing target columns: \\['real_col'\\]"):
-        TransformerModel._calculate_loss(model, output, targets, metadata)
-
-
-def test_training_loss_enables_fsdp_token_weighting(
-    monkeypatch,
-):
-    model = _loss_shell_model({"real_col": "real"})
-    model.hparams = SimpleNamespace(
-        training_spec=SimpleNamespace(data_parallelism="FSDP")
-    )
-    output = {"real_col": torch.zeros(1, 1, 1, requires_grad=True)}
-    targets = {"real_col": torch.ones(1, 1)}
-    metadata = {"target_valid_mask": torch.ones(1, 1, dtype=torch.bool)}
-    process_group = object()
-    model._data_parallel_process_group = lambda: process_group
-
-    monkeypatch.setattr("sequifier.train.dist.is_available", lambda: True)
-    monkeypatch.setattr("sequifier.train.dist.is_initialized", lambda: True)
-
-    def fake_get_world_size(group=None):
-        assert group is process_group
-        return 2
-
-    monkeypatch.setattr("sequifier.train.dist.get_world_size", fake_get_world_size)
-
-    reduce_calls = []
-
-    def fake_all_reduce(tensor, op=None, group=None):
-        reduce_calls.append((tensor.detach().clone(), op, group))
-        tensor.fill_(4)
-
-    monkeypatch.setattr("sequifier.train.dist.all_reduce", fake_all_reduce)
-
-    loss, _ = TransformerModel._calculate_loss(model, output, targets, metadata)
-    loss.backward()
-
-    assert reduce_calls
-    assert reduce_calls[0][2] is process_group
-    assert torch.isclose(loss, torch.tensor(0.5))
-    assert torch.allclose(output["real_col"].grad, torch.tensor([[[-1.0]]]))
-
-
-def test_evaluation_loss_mask_intersects_target_bert_and_sample_masks():
-    metadata = {
-        "target_valid_mask": torch.tensor(
-            [
-                [True, True, False],
-                [True, True, True],
-            ]
-        ),
-        "bert_mask": torch.tensor(
-            [
-                [True, False, True],
-                [True, True, True],
-            ]
-        ),
-        "sample_valid_mask": torch.tensor([True, False]),
-    }
-
-    loss_mask = _get_evaluation_loss_mask(metadata)
-
-    assert torch.equal(
-        loss_mask,
-        torch.tensor(
-            [
-                [True, False, False],
-                [False, False, False],
-            ]
-        ),
-    )
-
-
-def _categorical_logits_from_batch_predictions(batch_predictions, n_classes):
-    prediction_ids = torch.tensor(batch_predictions, dtype=torch.long).T.contiguous()
-    return torch.nn.functional.one_hot(prediction_ids, num_classes=n_classes).float()
-
-
-def test_accumulate_class_counts_all_samples_valid():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [[0, 1], [1, 2]],
-            n_classes=3,
-        )
-    }
-    valid_mask = torch.ones(2, 2, dtype=torch.bool)
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([1, 2, 1]))
-
-
-def test_accumulate_class_counts_excludes_synthetic_samples():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [
-                [0, 1],
-                [2, 2],
-            ],
-            n_classes=3,
-        )
-    }
-    valid_mask = _get_evaluation_loss_mask(
-        {
-            "target_valid_mask": torch.ones(2, 2, dtype=torch.bool),
-            "sample_valid_mask": torch.tensor([True, False]),
-        }
-    )
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([1, 1, 0]))
-
-
-def test_accumulate_class_counts_excludes_target_padding():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [[0, 2]],
-            n_classes=3,
-        )
-    }
-    valid_mask = torch.tensor([[True, False]])
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([1, 0, 0]))
-
-
-def test_accumulate_class_counts_combines_sample_and_token_masks():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [
-                [0, 1],
-                [2, 2],
-            ],
-            n_classes=3,
-        )
-    }
-    valid_mask = _get_evaluation_loss_mask(
-        {
-            "target_valid_mask": torch.tensor(
-                [
-                    [True, False],
-                    [True, True],
-                ]
-            ),
-            "sample_valid_mask": torch.tensor([True, False]),
-        }
-    )
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([1, 0, 0]))
-
-
-def test_accumulate_class_counts_respects_bert_mask():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [
-                [0, 1],
-                [2, 2],
-            ],
-            n_classes=3,
-        )
-    }
-    valid_mask = _get_evaluation_loss_mask(
-        {
-            "target_valid_mask": torch.ones(2, 2, dtype=torch.bool),
-            "bert_mask": torch.tensor(
-                [
-                    [False, True],
-                    [True, True],
-                ]
-            ),
-            "sample_valid_mask": torch.tensor([True, False]),
-        }
-    )
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([0, 1, 0]))
-
-
-def test_accumulate_class_counts_retains_missing_class_slots():
-    counts = {"cat_col": torch.zeros(5, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [[3, 3]],
-            n_classes=5,
-        )
-    }
-    valid_mask = torch.ones(1, 2, dtype=torch.bool)
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 5})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([0, 0, 0, 2, 0]))
-
-
-def test_accumulate_class_counts_all_zero_when_no_positions_are_valid():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [[0, 1]],
-            n_classes=3,
-        )
-    }
-    valid_mask = torch.zeros(1, 2, dtype=torch.bool)
-
-    accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-    assert torch.equal(counts["cat_col"], torch.tensor([0, 0, 0]))
-
-
-def test_accumulate_class_counts_raises_on_shape_mismatch():
-    counts = {"cat_col": torch.zeros(3, dtype=torch.int64)}
-    output = {
-        "cat_col": _categorical_logits_from_batch_predictions(
-            [[0, 1, 2]],
-            n_classes=3,
-        )
-    }
-    valid_mask = torch.ones(1, 2, dtype=torch.bool)
-
-    with pytest.raises(RuntimeError, match="Prediction/mask size mismatch"):
-        accumulate_class_counts(counts, output, valid_mask, {"cat_col": 3})
-
-
-def test_calculate_loss_components_aggregate_by_token_count():
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_column_types = {"real_col": "real"}
-    model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
-    model.loss_weights = {"real_col": 1.0}
-
-    first_sums, first_count = TransformerModel._calculate_loss_components(
-        model,
-        {"real_col": torch.zeros(1, 1, 1)},
-        {"real_col": torch.tensor([[10.0]])},
-        torch.ones(1, 1, dtype=torch.bool),
-    )
-    second_sums, second_count = TransformerModel._calculate_loss_components(
-        model,
-        {"real_col": torch.zeros(100, 1, 1)},
-        {"real_col": torch.ones(1, 100)},
-        torch.ones(1, 100, dtype=torch.bool),
-    )
-
-    aggregate = (first_sums["real_col"] + second_sums["real_col"]) / (
-        first_count + second_count
-    )
-
-    assert torch.isclose(aggregate, torch.tensor(200.0 / 101.0, dtype=torch.float64))
-
-
-class _QueuedEvalModel(torch.nn.Module):
-    def __init__(self, outputs):
-        super().__init__()
-        self.outputs = iter(outputs)
-
-    def forward(self, data, metadata=None, return_logits=False):
-        return next(self.outputs)
-
-
-def _evaluation_shell_model():
-    model = TransformerModel.__new__(TransformerModel)
-    model.input_columns = ["cat_col"]
-    model.target_columns = ["cat_col"]
-    model.target_column_types = {"cat_col": "categorical"}
-    model.n_classes = {"cat_col": 3}
-    model.loss_weights = None
-    model.criterion = {"cat_col": torch.nn.CrossEntropyLoss(reduction="none")}
-    model.device = "cpu"
-    model.class_share_log_columns = ["cat_col"]
-    model.index_maps = {  # type: ignore
-        "cat_col": {
-            0: "[unknown]",
-            1: "a",
-            2: "b",
-        }
-    }  # type: ignore
-    model.decoder = {"cat_col": torch.nn.Linear(1, 1)}
-    model.hparams = SimpleNamespace(
-        seed=42,
-        training_spec=SimpleNamespace(
-            distributed=False,
-            layer_autocast=False,
-            data_parallelism="none",
-            training_objective="causal",
-        ),
-    )
-    return model
-
-
-def _validation_batch(predictions, sample_valid_mask=None):
-    batch_size = len(predictions)
-    seq_len = len(predictions[0])
-    inputs = {
-        "cat_col": torch.zeros(batch_size, seq_len, dtype=torch.long),
-    }
-    targets = {
-        "cat_col": torch.tensor(predictions, dtype=torch.long),
-    }
-    valid_mask = torch.ones(batch_size, seq_len, dtype=torch.bool)
-    metadata = {
-        "attention_valid_mask": valid_mask.clone(),
-        "target_valid_mask": valid_mask,
-    }
-    if sample_valid_mask is not None:
-        metadata["sample_valid_mask"] = torch.tensor(
-            sample_valid_mask,
-            dtype=torch.bool,
-        )
-    return SequifierBatch(inputs=inputs, targets=targets, metadata=metadata)
-
-
-def test_evaluate_returns_class_counts_across_all_validation_batches():
-    model = _evaluation_shell_model()
-    batches = [
-        _validation_batch([[0, 0, 1]]),
-        _validation_batch([[2, 2]]),
-    ]
-    outputs = [
-        {"cat_col": _categorical_logits_from_batch_predictions([[0, 0, 1]], 3)},
-        {"cat_col": _categorical_logits_from_batch_predictions([[2, 2]], 3)},
-    ]
-    eval_model = _QueuedEvalModel(outputs)
-    valid_loader = DataLoader(batches, batch_size=None)
-
-    total_loss, total_losses, class_counts = TransformerModel._evaluate(
-        model,
-        valid_loader,
-        eval_model,
-    )
-
-    assert np.isfinite(total_loss)
-    assert np.isfinite(total_losses["cat_col"])
-    assert torch.equal(class_counts["cat_col"], torch.tensor([2, 1, 2]))
-    assert eval_model.training
-    assert model.baseline_loss >= 0
-
-
-def test_evaluate_excludes_synthetic_final_batch_from_class_counts():
-    model = _evaluation_shell_model()
-    batches = [
-        _validation_batch([[0, 1]]),
-        _validation_batch([[2, 2]], sample_valid_mask=[False]),
-    ]
-    outputs = [
-        {"cat_col": _categorical_logits_from_batch_predictions([[0, 1]], 3)},
-        {"cat_col": _categorical_logits_from_batch_predictions([[2, 2]], 3)},
-    ]
-    valid_loader = DataLoader(batches, batch_size=None)
-
-    _, _, class_counts = TransformerModel._evaluate(
-        model,
-        valid_loader,
-        _QueuedEvalModel(outputs),
-    )
-
-    assert torch.equal(class_counts["cat_col"], torch.tensor([1, 1, 0]))
-
-
-def test_calculate_loss_zero_token_training_batch_is_differentiable():
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_column_types = {"real_col": "real"}
-    model.criterion = {"real_col": torch.nn.MSELoss(reduction="none")}
-    model.loss_weights = None
-
-    output = {"real_col": torch.ones(3, 1, 1, requires_grad=True)}
-    targets = {"real_col": torch.zeros(1, 3)}
-    metadata = {
-        "attention_valid_mask": torch.zeros(1, 3, dtype=torch.bool),
-        "target_valid_mask": torch.zeros(1, 3, dtype=torch.bool),
-    }
-
-    total_loss, component_losses = TransformerModel._calculate_loss(
-        model, output, targets, metadata
-    )
-    total_loss.backward()
-
-    assert torch.equal(total_loss.detach(), torch.tensor(0.0))
-    assert torch.equal(component_losses["real_col"].detach(), torch.tensor(0.0))
-    assert torch.equal(output["real_col"].grad, torch.zeros_like(output["real_col"]))
-
-
-def _train_epoch_test_batch(model, target_valid_mask):
-    seq_len = model.window_view.context_length
-    return SequifierBatch(
-        inputs={
-            "cat_col": torch.ones(1, seq_len, dtype=torch.long),
-            "real_col": torch.ones(1, seq_len, dtype=torch.float32),
-        },
-        targets={
-            "cat_col": torch.ones(1, seq_len, dtype=torch.long),
-            "real_col": torch.ones(1, seq_len, dtype=torch.float32),
-        },
-        metadata={
-            "attention_valid_mask": torch.ones(1, seq_len, dtype=torch.bool),
-            "target_valid_mask": target_valid_mask,
-        },
-    )
-
-
-def test_train_epoch_skips_optimizer_and_batch_scheduler_for_empty_accumulation_window(
-    model,
-):
-    model.rank = 0
-    model.log_interval = 2
-    model.accumulation_steps = 2
-    model.scheduler_step_on = "batch"
-    model.start_batch = 0
-    model.optimizer = torch.optim.AdamW(
-        model.parameters(),
-        lr=0.01,
-        weight_decay=0.1,
-    )
-    model.scheduler = torch.optim.lr_scheduler.StepLR(
-        model.optimizer,
-        step_size=1,
-        gamma=0.1,
-    )
-    seq_len = model.window_view.context_length
-    empty_batch = _train_epoch_test_batch(
-        model,
-        torch.zeros(1, seq_len, dtype=torch.bool),
-    )
-    before = {name: param.detach().clone() for name, param in model.named_parameters()}
-    lr_before = model.scheduler.get_last_lr()[0]
-
-    TransformerModel._train_epoch(
-        model,
-        DataLoader([empty_batch, empty_batch], batch_size=None),
-        DataLoader([], batch_size=None),
-        epoch=1,
-    )
-
-    after = dict(model.named_parameters())
-    assert all(torch.equal(before[name], after[name]) for name in before)
-    assert len(model.optimizer.state) == 0
-    assert model.scheduler.get_last_lr()[0] == lr_before
-
-
-def test_train_epoch_steps_once_for_mixed_empty_and_nonempty_accumulation_window(
-    model,
-):
-    model.rank = 0
-    model.log_interval = 2
-    model.accumulation_steps = 2
-    model.scheduler_step_on = "batch"
-    model.start_batch = 0
-    model.optimizer = torch.optim.AdamW(
-        model.parameters(),
-        lr=0.01,
-        weight_decay=0.1,
-    )
-    model.scheduler = torch.optim.lr_scheduler.StepLR(
-        model.optimizer,
-        step_size=1,
-        gamma=0.1,
-    )
-    seq_len = model.window_view.context_length
-    empty_batch = _train_epoch_test_batch(
-        model,
-        torch.zeros(1, seq_len, dtype=torch.bool),
-    )
-    nonempty_batch = _train_epoch_test_batch(
-        model,
-        torch.ones(1, seq_len, dtype=torch.bool),
-    )
-    before = {name: param.detach().clone() for name, param in model.named_parameters()}
-
-    TransformerModel._train_epoch(
-        model,
-        DataLoader([empty_batch, nonempty_batch], batch_size=None),
-        DataLoader([], batch_size=None),
-        epoch=1,
-    )
-
-    after = dict(model.named_parameters())
-    assert any(not torch.equal(before[name], after[name]) for name in before)
-    assert len(model.optimizer.state) > 0
-    assert model.scheduler.last_epoch == 1
-    assert model.scheduler.get_last_lr()[0] == pytest.approx(0.001)
-
-
-def test_train_epoch_divides_backward_loss_by_accumulation_steps(model, monkeypatch):
-    model.rank = 0
-    model.log_interval = 100
-    model.accumulation_steps = 2
-    model.scheduler_step_on = "batch"
-    model.start_batch = 0
-    model.scaler = _IdentityScaler()
-    model.logger = _NoopLogger()
-    model.save_latest_interval_minutes = None
-    model.save_batch_interval_minutes = None
-    model.save_batch_interval_minutes_val_loss = False
-    model.last_latest_save_time = 0.0
-    model.last_batch_save_time = 0.0
-    tracked_param = next(model.parameters())
-    tracked_value_before = tracked_param.detach().reshape(-1)[0].clone()
-    model.optimizer = torch.optim.SGD([tracked_param], lr=0.1)
-    model.scheduler = torch.optim.lr_scheduler.StepLR(
-        model.optimizer,
-        step_size=1,
-        gamma=0.1,
-    )
-    coefficients = iter([0.2, 0.4])
-
-    def fake_forward(data, metadata=None, return_logits=True):
-        return {}
-
-    def fake_calculate_training_loss(output, targets, metadata):
-        coefficient = next(coefficients)
-        loss = tracked_param.reshape(-1)[0] * coefficient
-        loss_sums = {
-            target_name: loss.detach()
-            for target_name in model._loss_target_names(targets)
-        }
-        count = torch.tensor(1, dtype=torch.int64)
-        return loss, {}, loss_sums, count, count
-
-    monkeypatch.setattr(model, "forward", fake_forward)
-    monkeypatch.setattr(
-        model,
-        "_calculate_training_loss",
-        fake_calculate_training_loss,
-    )
-
-    seq_len = model.window_view.context_length
-    batch = _train_epoch_test_batch(
-        model,
-        torch.ones(1, seq_len, dtype=torch.bool),
-    )
-
-    TransformerModel._train_epoch(
-        model,
-        DataLoader([batch, batch], batch_size=None),
-        DataLoader([], batch_size=None),
-        epoch=1,
-    )
-
-    tracked_value_after = tracked_param.detach().reshape(-1)[0]
-
-    assert torch.allclose(
-        tracked_value_after,
-        tracked_value_before - torch.tensor(0.03),
-        atol=1e-6,
-    )
-
-
-def test_padding_keys_are_masked(bert_model):
-    seq_len = bert_model.window_view.context_length
-
-    valid_mask = torch.ones(
-        2,
-        seq_len,
-        dtype=torch.bool,
-        device=bert_model.src_mask.device,
-    )
-
-    valid_mask[0, :2] = False
-    valid_mask[1, :1] = False
-
-    attn_mask = bert_model._build_attention_mask(valid_mask, dtype=torch.float32)
-
-    assert attn_mask.shape == (2, 1, seq_len, seq_len)
-
-    # Batch 0: keys 0 and 1 are padding.
-    assert torch.all(attn_mask[0, :, :, 0] < -1e20)
-    assert torch.all(attn_mask[0, :, :, 1] < -1e20)
-
-    # Batch 0: keys 2 onward are not padding-masked.
-    assert torch.all(attn_mask[0, :, :, 2:] > -1e20)
-
-    # Batch 1: key 0 is padding.
-    assert torch.all(attn_mask[1, :, :, 0] < -1e20)
-    assert torch.all(attn_mask[1, :, :, 1:] > -1e20)
-
-
-def test_causal_and_padding_masks_are_combined(causal_model):
-    seq_len = causal_model.window_view.context_length
-
-    valid_mask = torch.ones(
-        1,
-        seq_len,
-        dtype=torch.bool,
-        device=causal_model.src_mask.device,
-    )
-
-    valid_mask[0, 0] = False
-
-    attn_mask = causal_model._build_attention_mask(
-        valid_mask,
-        dtype=torch.float32,
-    )
-
-    assert attn_mask.shape == (1, 1, seq_len, seq_len)
-
-    # Padding key 0 is masked for every query.
-    assert torch.all(attn_mask[0, :, :, 0] < -1e20)
-
-    # Causal future positions are masked.
-    assert attn_mask[0, 0, 1, 2] < -1e20
-    assert attn_mask[0, 0, 1, seq_len - 1] < -1e20
-
-    # A past valid key should not be masked by causal masking.
-    assert attn_mask[0, 0, 1, 1] > -1e20
-    assert attn_mask[0, 0, 2, 1] > -1e20
-    assert attn_mask[0, 0, seq_len - 1, seq_len - 1] > -1e20
-
-
-@pytest.fixture
-def batch(model):
-    seq_len = model.context_length
-
-    return {
-        "cat_col": torch.tensor(
-            [
-                [0, 0] + [1] * (seq_len - 2),
-                [0] + [2] * (seq_len - 1),
-            ],
-            dtype=torch.long,
-            device=model.src_mask.device,
-        ),
-        "real_col": torch.tensor(
-            [
-                [0.0, 0.0] + [0.5] * (seq_len - 2),
-                [0.0] + [1.5] * (seq_len - 1),
-            ],
-            dtype=torch.float32,
-            device=model.src_mask.device,
-        ),
-    }
-
-
-@pytest.fixture
-def batch_metadata(model):
-    seq_len = model.context_length
-    valid_mask = torch.ones(
-        2,
-        seq_len,
-        dtype=torch.bool,
-        device=model.src_mask.device,
-    )
-    valid_mask[0, :2] = False
-    valid_mask[1, :1] = False
-
-    return {
-        "attention_valid_mask": valid_mask,
-        "target_valid_mask": valid_mask.clone(),
-    }
-
-
-def test_forward_no_nan_with_padding(model, batch, batch_metadata):
-    out = model.forward_train(batch, batch_metadata)
-
-    for tensor in out.values():
-        assert torch.isfinite(tensor).all()
diff --git a/tests/unit/test_train_distributed_loss.py b/tests/unit/test_train_distributed_loss.py
deleted file mode 100644
index 0e94cf0f..00000000
--- a/tests/unit/test_train_distributed_loss.py
+++ /dev/null
@@ -1,763 +0,0 @@
-import datetime
-import socket
-from types import SimpleNamespace
-
-import pytest
-import torch
-import torch.distributed as dist
-import torch.multiprocessing as mp
-from torch.nn.parallel import DistributedDataParallel as DDP
-from torch.utils.data import DataLoader
-
-from sequifier.io.batch import SequifierBatch
-from sequifier.train import TransformerModel
-
-
-class _TinyRealModel(torch.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.param = torch.nn.Parameter(torch.tensor(0.0))
-
-    def forward(self, seq_len):
-        return {"real_col": self.param.expand(seq_len, 1, 1)}
-
-
-class _TinyTrainEpochRealModel(torch.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.param = torch.nn.Parameter(torch.tensor(0.0))
-
-    def forward(self, data, metadata=None, return_logits=False):
-        seq_len = data["real_col"].shape[1]
-        return {"real_col": self.param.expand(seq_len, 1, 1)}
-
-
-class _TinyMultiTargetModel(torch.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.cat_logits = torch.nn.Parameter(torch.tensor([0.2, -0.1, 0.0]))
-        self.real_param = torch.nn.Parameter(torch.tensor(0.5))
-
-    def forward(self, seq_len):
-        return {
-            "cat_col": self.cat_logits.reshape(1, 1, 3).expand(seq_len, 1, 3),
-            "real_col": self.real_param.expand(seq_len, 1, 1),
-        }
-
-
-def _free_port():
-    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as sock:
-        sock.bind(("127.0.0.1", 0))
-        return sock.getsockname()[1]
-
-
-def _loss_shell(target_column_types, *, data_parallelism=None, loss_weights=None):
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_columns = list(target_column_types)
-    model.target_column_types = dict(target_column_types)
-    model.loss_weights = loss_weights
-    model.device = "cpu"
-    model.hparams = SimpleNamespace(
-        training_spec=SimpleNamespace(
-            data_parallelism=data_parallelism,
-            distributed=data_parallelism is not None,
-            training_objective="causal",
-            layer_autocast=False,
-            layer_type_dtypes=None,
-        )
-    )
-    model.criterion = {}
-    model.n_classes = {}
-
-    for col, col_type in target_column_types.items():
-        if col_type == "real":
-            model.criterion[col] = torch.nn.MSELoss(reduction="none")
-        elif col_type == "categorical":
-            model.criterion[col] = torch.nn.CrossEntropyLoss(reduction="none")
-            model.n_classes[col] = 3
-        else:
-            raise ValueError(col_type)
-
-    return model
-
-
-class _IdentityScaler:
-    def scale(self, loss):
-        return loss
-
-    def unscale_(self, optimizer):
-        return None
-
-    def step(self, optimizer):
-        optimizer.step()
-
-    def update(self):
-        return None
-
-
-class _NoopLogger:
-    def info(self, *args, **kwargs):
-        return None
-
-    def warning(self, *args, **kwargs):
-        return None
-
-
-def _real_batch(seq_len, selected_count, target_value):
-    targets = {"real_col": torch.full((1, seq_len), float(target_value))}
-    mask = torch.zeros(1, seq_len, dtype=torch.bool)
-    mask[:, :selected_count] = True
-    metadata = {"target_valid_mask": mask}
-    return targets, metadata
-
-
-def _real_epoch_batch(seq_len, selected_count, target_value):
-    targets, metadata = _real_batch(seq_len, selected_count, target_value)
-    return SequifierBatch(
-        inputs={"real_col": torch.ones(1, seq_len, dtype=torch.float32)},
-        targets=targets,
-        metadata={
-            "attention_valid_mask": torch.ones(1, seq_len, dtype=torch.bool),
-            **metadata,
-        },
-    )
-
-
-def _concat_real_batch(seq_lens, selected_counts, target_values):
-    target_parts = []
-    mask_parts = []
-    for seq_len, selected_count, target_value in zip(
-        seq_lens, selected_counts, target_values
-    ):
-        target_parts.append(torch.full((1, seq_len), float(target_value)))
-        mask = torch.zeros(1, seq_len, dtype=torch.bool)
-        mask[:, :selected_count] = True
-        mask_parts.append(mask)
-    return (
-        {"real_col": torch.cat(target_parts, dim=1)},
-        {"target_valid_mask": torch.cat(mask_parts, dim=1)},
-    )
-
-
-def _single_process_real_grad(seq_lens, selected_counts, target_values):
-    model = _TinyRealModel()
-    shell = _loss_shell({"real_col": "real"})
-    targets, metadata = _concat_real_batch(seq_lens, selected_counts, target_values)
-    loss, _ = TransformerModel._calculate_loss(
-        shell,
-        model(sum(seq_lens)),
-        targets,
-        metadata,
-    )
-    loss.backward()
-    return model.param.grad.detach().clone()
-
-
-def _single_process_real_update(seq_lens, selected_counts, target_values, lr):
-    model = _TinyRealModel()
-    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
-    shell = _loss_shell({"real_col": "real"})
-    targets, metadata = _concat_real_batch(seq_lens, selected_counts, target_values)
-    loss, _ = TransformerModel._calculate_loss(
-        shell,
-        model(sum(seq_lens)),
-        targets,
-        metadata,
-    )
-    loss.backward()
-    optimizer.step()
-    return model.param.detach().clone()
-
-
-def _single_process_accumulated_real_update(microbatches, lr, accumulation_steps):
-    model = _TinyRealModel()
-    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
-    shell = _loss_shell({"real_col": "real"})
-    for microbatch in microbatches:
-        targets, metadata = _concat_real_batch(
-            microbatch["seq_lens"],
-            microbatch["selected_counts"],
-            microbatch["target_values"],
-        )
-        loss, _ = TransformerModel._calculate_loss(
-            shell,
-            model(sum(microbatch["seq_lens"])),
-            targets,
-            metadata,
-        )
-        (loss / accumulation_steps).backward()
-    optimizer.step()
-    return model.param.detach().clone()
-
-
-def _analytical_accumulated_real_update(microbatches, lr, accumulation_steps):
-    accumulated_grad = 0.0
-    for microbatch in microbatches:
-        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
-        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
-        token_count = selected_counts.sum().item()
-        if token_count == 0:
-            continue
-
-        weighted_target_mean = (
-            selected_counts * target_values
-        ).sum().item() / token_count
-        accumulated_grad += -2.0 * weighted_target_mean / accumulation_steps
-
-    return torch.tensor(-lr * accumulated_grad)
-
-
-def _analytical_unnormalized_accumulated_real_update(microbatches, lr):
-    accumulated_grad = 0.0
-    for microbatch in microbatches:
-        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
-        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
-        token_count = selected_counts.sum().item()
-        if token_count == 0:
-            continue
-
-        weighted_target_mean = (
-            selected_counts * target_values
-        ).sum().item() / token_count
-        accumulated_grad += -2.0 * weighted_target_mean
-
-    return torch.tensor(-lr * accumulated_grad)
-
-
-def _analytical_combined_real_update(microbatches, lr):
-    weighted_target_sum = 0.0
-    token_count = 0.0
-    for microbatch in microbatches:
-        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
-        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
-        weighted_target_sum += (selected_counts * target_values).sum().item()
-        token_count += selected_counts.sum().item()
-
-    if token_count == 0:
-        return torch.tensor(0.0)
-    return torch.tensor(lr * 2.0 * weighted_target_sum / token_count)
-
-
-def _single_process_multi_target_grad(
-    seq_lens, cat_targets, real_targets, loss_weights
-):
-    model = _TinyMultiTargetModel()
-    shell = _loss_shell(
-        {"cat_col": "categorical", "real_col": "real"},
-        loss_weights=loss_weights,
-    )
-    targets = {
-        "cat_col": torch.tensor([sum(cat_targets, [])], dtype=torch.long),
-        "real_col": torch.tensor([sum(real_targets, [])], dtype=torch.float32),
-    }
-    metadata = {"target_valid_mask": torch.ones(1, sum(seq_lens), dtype=torch.bool)}
-    loss, _ = TransformerModel._calculate_loss(
-        shell,
-        model(sum(seq_lens)),
-        targets,
-        metadata,
-    )
-    loss.backward()
-    return torch.cat(
-        [
-            model.cat_logits.grad.detach(),
-            model.real_param.grad.detach().reshape(1),
-        ]
-    )
-
-
-def _ddp_case_worker(rank, world_size, init_method, case, queue):
-    torch.set_num_threads(1)
-    dist.init_process_group(
-        "gloo",
-        rank=rank,
-        world_size=world_size,
-        init_method=init_method,
-        timeout=datetime.timedelta(seconds=30),
-    )
-    try:
-
-        def put_result(tensor):
-            detached = tensor.detach().cpu()
-            if detached.numel() == 1:
-                queue.put(float(detached.item()))
-            else:
-                queue.put(detached.tolist())
-
-        if case["kind"] in {
-            "real_grad",
-            "real_update",
-            "empty_grad",
-            "accumulation_update",
-            "empty_window_state",
-            "train_epoch_empty_window_state",
-        }:
-            if case["kind"] == "train_epoch_empty_window_state":
-                model = _TinyTrainEpochRealModel()
-                ddp_model = DDP(model)
-                shell = _loss_shell({"real_col": "real"}, data_parallelism="DDP")
-                shell.input_columns = ["real_col"]
-                shell.rank = rank
-                shell.device = "cpu"
-                shell.accumulation_steps = case["accumulation_steps"]
-                shell.scheduler_step_on = "batch"
-                shell.start_batch = 0
-                shell.log_interval = case["accumulation_steps"]
-                shell.scaler = _IdentityScaler()
-                shell.logger = _NoopLogger()
-                shell.save_latest_interval_minutes = None
-                shell.save_batch_interval_minutes = None
-                shell.save_batch_interval_minutes_val_loss = False
-                shell.last_latest_save_time = 0.0
-                shell.last_batch_save_time = 0.0
-                shell.parameters = model.parameters
-                shell.optimizer = torch.optim.AdamW(
-                    model.parameters(),
-                    lr=case["lr"],
-                    weight_decay=case["weight_decay"],
-                )
-                shell.scheduler = torch.optim.lr_scheduler.StepLR(
-                    shell.optimizer,
-                    step_size=1,
-                    gamma=0.1,
-                )
-
-                train_loader = [
-                    _real_epoch_batch(
-                        microbatch["seq_lens"][rank],
-                        microbatch["selected_counts"][rank],
-                        microbatch["target_values"][rank],
-                    )
-                    for microbatch in case["microbatches"]
-                ]
-                before = model.param.detach().clone()
-
-                TransformerModel._train_epoch(
-                    shell,
-                    DataLoader(train_loader, batch_size=None),
-                    DataLoader([], batch_size=None),
-                    epoch=1,
-                    ddp_model=ddp_model,
-                )
-
-                if rank == 0:
-                    queue.put(
-                        [
-                            float((model.param.detach() - before).abs().item()),
-                            float(shell.scheduler.get_last_lr()[0]),
-                            float(shell.scheduler.last_epoch),
-                            float(len(shell.optimizer.state)),
-                        ]
-                    )
-                return
-
-            model = _TinyRealModel()
-            ddp_model = DDP(model)
-            shell = _loss_shell({"real_col": "real"}, data_parallelism="DDP")
-
-            if case["kind"] == "empty_window_state":
-                optimizer = torch.optim.AdamW(
-                    model.parameters(),
-                    lr=case["lr"],
-                    weight_decay=case["weight_decay"],
-                )
-                scheduler = torch.optim.lr_scheduler.StepLR(
-                    optimizer,
-                    step_size=1,
-                    gamma=0.1,
-                )
-                before = model.param.detach().clone()
-                accumulated_global_token_count = torch.zeros((), dtype=torch.int64)
-
-                for batch_idx, microbatch in enumerate(case["microbatches"]):
-                    seq_len = microbatch["seq_lens"][rank]
-                    targets, metadata = _real_batch(
-                        seq_len,
-                        microbatch["selected_counts"][rank],
-                        microbatch["target_values"][rank],
-                    )
-                    loss, _, _, _, global_count = (
-                        TransformerModel._calculate_training_loss(
-                            shell,
-                            ddp_model(seq_len),
-                            targets,
-                            metadata,
-                        )
-                    )
-                    (loss / case["accumulation_steps"]).backward()
-                    accumulated_global_token_count += global_count.detach()
-                    optimizer_step_due = (batch_idx + 1) % case[
-                        "accumulation_steps"
-                    ] == 0 or (batch_idx + 1) == len(case["microbatches"])
-                    optimizer_step_performed = False
-                    if (
-                        optimizer_step_due
-                        and accumulated_global_token_count.detach().cpu().item() > 0
-                    ):
-                        optimizer.step()
-                        optimizer.zero_grad()
-                        optimizer_step_performed = True
-
-                    if optimizer_step_due:
-                        if not optimizer_step_performed:
-                            optimizer.zero_grad()
-                        accumulated_global_token_count.zero_()
-
-                    if optimizer_step_performed:
-                        scheduler.step()
-
-                if rank == 0:
-                    queue.put(
-                        [
-                            float((model.param.detach() - before).abs().item()),
-                            float(scheduler.get_last_lr()[0]),
-                            float(scheduler.last_epoch),
-                            float(len(optimizer.state)),
-                        ]
-                    )
-                return
-
-            if case["kind"] == "accumulation_update":
-                optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
-                for microbatch in case["microbatches"]:
-                    seq_len = microbatch["seq_lens"][rank]
-                    targets, metadata = _real_batch(
-                        seq_len,
-                        microbatch["selected_counts"][rank],
-                        microbatch["target_values"][rank],
-                    )
-                    loss, _, _, _, _ = TransformerModel._calculate_training_loss(
-                        shell,
-                        ddp_model(seq_len),
-                        targets,
-                        metadata,
-                    )
-                    (loss / case["accumulation_steps"]).backward()
-                optimizer.step()
-                if rank == 0:
-                    put_result(model.param)
-                return
-
-            seq_len = case["seq_lens"][rank]
-            targets, metadata = _real_batch(
-                seq_len,
-                case["selected_counts"][rank],
-                case["target_values"][rank],
-            )
-            loss, _, _, _, _ = TransformerModel._calculate_training_loss(
-                shell,
-                ddp_model(seq_len),
-                targets,
-                metadata,
-            )
-            loss.backward()
-
-            if case["kind"] == "real_update":
-                optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
-                optimizer.step()
-                result = model.param.detach().clone()
-            else:
-                result = model.param.grad.detach().clone()
-
-            if rank == 0:
-                put_result(result)
-            return
-
-        if case["kind"] == "metrics":
-            shell = _loss_shell(
-                {"cat_col": "categorical", "real_col": "real"},
-                data_parallelism="DDP",
-                loss_weights=case["loss_weights"],
-            )
-            sums = {
-                "cat_col": torch.tensor(case["cat_sums"][rank], dtype=torch.float64),
-                "real_col": torch.tensor(case["real_sums"][rank], dtype=torch.float64),
-            }
-            count = torch.tensor(case["counts"][rank], dtype=torch.float64)
-            total, losses = TransformerModel._finalize_loss_components(
-                shell,
-                sums,
-                count,
-                ["cat_col", "real_col"],
-                "training",
-            )
-            if rank == 0:
-                queue.put(
-                    [
-                        float(total.detach().cpu().item()),
-                        float(losses["cat_col"].detach().cpu().item()),
-                        float(losses["real_col"].detach().cpu().item()),
-                    ]
-                )
-            return
-
-        if case["kind"] == "multi_target_grad":
-            model = _TinyMultiTargetModel()
-            ddp_model = DDP(model)
-            shell = _loss_shell(
-                {"cat_col": "categorical", "real_col": "real"},
-                data_parallelism="DDP",
-                loss_weights=case["loss_weights"],
-            )
-            seq_len = case["seq_lens"][rank]
-            targets = {
-                "cat_col": torch.tensor([case["cat_targets"][rank]], dtype=torch.long),
-                "real_col": torch.tensor(
-                    [case["real_targets"][rank]], dtype=torch.float32
-                ),
-            }
-            metadata = {"target_valid_mask": torch.ones(1, seq_len, dtype=torch.bool)}
-            loss, _, _, _, _ = TransformerModel._calculate_training_loss(
-                shell,
-                ddp_model(seq_len),
-                targets,
-                metadata,
-            )
-            loss.backward()
-            if rank == 0:
-                put_result(
-                    torch.cat(
-                        [
-                            model.cat_logits.grad.detach(),
-                            model.real_param.grad.detach().reshape(1),
-                        ]
-                    )
-                )
-            return
-
-        raise ValueError(case["kind"])
-    finally:
-        dist.destroy_process_group()
-
-
-def _run_ddp_case(case):
-    ctx = mp.get_context("spawn")
-    queue = ctx.SimpleQueue()
-    init_method = f"tcp://127.0.0.1:{_free_port()}"
-    mp.spawn(
-        _ddp_case_worker,
-        args=(2, init_method, case, queue),
-        nprocs=2,
-        join=True,
-    )
-    return torch.tensor(queue.get())
-
-
-pytestmark = pytest.mark.skipif(
-    not dist.is_gloo_available(),
-    reason="torch.is_gloo_available is False",
-)
-
-
-def test_ddp_unequal_token_counts_match_single_process_reference():
-    case = {
-        "kind": "real_grad",
-        "seq_lens": [100, 10],
-        "selected_counts": [100, 10],
-        "target_values": [2.0, 8.0],
-    }
-
-    ddp_grad = _run_ddp_case(case)
-    reference_grad = _single_process_real_grad(
-        case["seq_lens"], case["selected_counts"], case["target_values"]
-    )
-    old_equal_rank_mean = torch.tensor((-2.0 * 2.0 + -2.0 * 8.0) / 2.0)
-
-    assert torch.allclose(ddp_grad, reference_grad, atol=1e-6)
-    assert not torch.allclose(ddp_grad, old_equal_rank_mean, atol=1e-3)
-
-
-def test_ddp_zero_token_rank_does_not_attenuate_populated_rank():
-    case = {
-        "kind": "real_grad",
-        "seq_lens": [5, 5],
-        "selected_counts": [5, 0],
-        "target_values": [3.0, 100.0],
-    }
-
-    ddp_grad = _run_ddp_case(case)
-    reference_grad = _single_process_real_grad(
-        case["seq_lens"], case["selected_counts"], case["target_values"]
-    )
-
-    assert torch.allclose(ddp_grad, reference_grad, atol=1e-6)
-    assert torch.allclose(ddp_grad, torch.tensor(-6.0), atol=1e-6)
-
-
-def test_ddp_equal_token_counts_retain_equal_rank_mean_behavior():
-    case = {
-        "kind": "real_grad",
-        "seq_lens": [4, 4],
-        "selected_counts": [4, 4],
-        "target_values": [1.0, 5.0],
-    }
-
-    ddp_grad = _run_ddp_case(case)
-    equal_rank_mean = torch.tensor((-2.0 * 1.0 + -2.0 * 5.0) / 2.0)
-
-    assert torch.allclose(ddp_grad, equal_rank_mean, atol=1e-6)
-
-
-def test_ddp_multi_target_weighting_matches_single_process_reference():
-    case = {
-        "kind": "multi_target_grad",
-        "seq_lens": [3, 1],
-        "cat_targets": [[0, 1, 2], [1]],
-        "real_targets": [[1.0, 2.0, 3.0], [4.0]],
-        "loss_weights": {"cat_col": 0.25, "real_col": 2.0},
-    }
-
-    ddp_grad = _run_ddp_case(case)
-    reference_grad = _single_process_multi_target_grad(
-        case["seq_lens"],
-        case["cat_targets"],
-        case["real_targets"],
-        case["loss_weights"],
-    )
-
-    assert torch.allclose(ddp_grad, reference_grad, atol=1e-6)
-
-
-def test_ddp_world_size_two_optimizer_update_matches_single_process_reference():
-    case = {
-        "kind": "real_update",
-        "seq_lens": [6, 4],
-        "selected_counts": [6, 4],
-        "target_values": [1.0, 3.0],
-        "lr": 0.1,
-    }
-
-    ddp_param = _run_ddp_case(case)
-    reference_param = _single_process_real_update(
-        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
-    )
-
-    assert torch.allclose(ddp_param, reference_param, atol=1e-6)
-
-
-def test_ddp_globally_empty_batch_gradients_are_finite_and_zero():
-    case = {
-        "kind": "empty_grad",
-        "seq_lens": [3, 4],
-        "selected_counts": [0, 0],
-        "target_values": [1.0, 3.0],
-    }
-
-    ddp_grad = _run_ddp_case(case)
-
-    assert torch.isfinite(ddp_grad)
-    assert torch.equal(ddp_grad, torch.tensor(0.0))
-
-
-def test_ddp_globally_empty_accumulation_window_leaves_state_unchanged():
-    case = {
-        "kind": "empty_window_state",
-        "microbatches": [
-            {
-                "seq_lens": [3, 4],
-                "selected_counts": [0, 0],
-                "target_values": [1.0, 3.0],
-            },
-            {
-                "seq_lens": [2, 5],
-                "selected_counts": [0, 0],
-                "target_values": [2.0, 4.0],
-            },
-        ],
-        "accumulation_steps": 2,
-        "lr": 0.01,
-        "weight_decay": 0.1,
-    }
-
-    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_ddp_case(case)
-
-    assert param_delta == 0
-    assert lr == pytest.approx(case["lr"])
-    assert scheduler_epoch == 0
-    assert optimizer_state_len == 0
-
-
-def test_ddp_train_epoch_globally_empty_accumulation_window_leaves_state_unchanged():
-    case = {
-        "kind": "train_epoch_empty_window_state",
-        "microbatches": [
-            {
-                "seq_lens": [3, 4],
-                "selected_counts": [0, 0],
-                "target_values": [1.0, 3.0],
-            },
-            {
-                "seq_lens": [2, 5],
-                "selected_counts": [0, 0],
-                "target_values": [2.0, 4.0],
-            },
-        ],
-        "accumulation_steps": 2,
-        "lr": 0.01,
-        "weight_decay": 0.1,
-    }
-
-    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_ddp_case(case)
-
-    assert param_delta == 0
-    assert lr == pytest.approx(case["lr"])
-    assert scheduler_epoch == 0
-    assert optimizer_state_len == 0
-
-
-def test_ddp_gradient_accumulation_averages_per_microbatch_weighting():
-    microbatches = [
-        {
-            "seq_lens": [1, 1],
-            "selected_counts": [1, 1],
-            "target_values": [1.0, 3.0],
-        },
-        {
-            "seq_lens": [5, 5],
-            "selected_counts": [5, 5],
-            "target_values": [10.0, 2.0],
-        },
-    ]
-    case = {
-        "kind": "accumulation_update",
-        "microbatches": microbatches,
-        "accumulation_steps": 2,
-        "lr": 0.05,
-    }
-
-    ddp_param = _run_ddp_case(case)
-    reference_param = _analytical_accumulated_real_update(
-        microbatches, case["lr"], case["accumulation_steps"]
-    )
-    unnormalized_param = _analytical_unnormalized_accumulated_real_update(
-        microbatches, case["lr"]
-    )
-    combined_batch_param = _analytical_combined_real_update(microbatches, case["lr"])
-
-    assert torch.allclose(ddp_param, reference_param, atol=1e-6)
-    assert not torch.allclose(ddp_param, unnormalized_param, atol=1e-3)
-    assert not torch.allclose(ddp_param, combined_batch_param, atol=1e-3)
-
-
-def test_ddp_global_training_metric_finalization_matches_manual_sum_count():
-    case = {
-        "kind": "metrics",
-        "cat_sums": [10.0, 5.0],
-        "real_sums": [2.0, 8.0],
-        "counts": [4.0, 6.0],
-        "loss_weights": {"cat_col": 0.5, "real_col": 2.0},
-    }
-
-    total, cat_loss, real_loss = _run_ddp_case(case)
-
-    global_count = sum(case["counts"])
-    expected_cat = (
-        sum(case["cat_sums"]) / global_count * case["loss_weights"]["cat_col"]
-    )
-    expected_real = (
-        sum(case["real_sums"]) / global_count * case["loss_weights"]["real_col"]
-    )
-
-    assert cat_loss == pytest.approx(expected_cat)
-    assert real_loss == pytest.approx(expected_real)
-    assert total == pytest.approx(expected_cat + expected_real)
diff --git a/tests/unit/test_train_fsdp_loss.py b/tests/unit/test_train_fsdp_loss.py
deleted file mode 100644
index 5c092d68..00000000
--- a/tests/unit/test_train_fsdp_loss.py
+++ /dev/null
@@ -1,754 +0,0 @@
-import datetime
-import importlib
-import socket
-from types import SimpleNamespace
-
-import pytest
-import torch
-import torch.distributed as dist
-import torch.multiprocessing as mp
-from torch.distributed.checkpoint.state_dict import (
-    StateDictOptions,
-    get_model_state_dict,
-)
-from torch.utils.data import DataLoader
-
-from sequifier.io.batch import SequifierBatch
-from sequifier.train import TransformerModel
-
-_train_module = importlib.import_module("sequifier.train")
-MixedPrecisionPolicy = getattr(_train_module, "MixedPrecisionPolicy")
-fully_shard = getattr(_train_module, "fully_shard")
-init_device_mesh = getattr(_train_module, "init_device_mesh")
-
-pytestmark = pytest.mark.skipif(
-    not dist.is_available()
-    or not dist.is_nccl_available()
-    or not torch.cuda.is_available()
-    or torch.cuda.device_count() < 2,
-    reason="FSDP2 loss tests require two CUDA devices and NCCL",
-)
-
-
-class _TinyRealModel(torch.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.param = torch.nn.Parameter(torch.zeros(2))
-
-    def forward(self, seq_len):
-        return {"real_col": self.param[0].expand(seq_len, 1, 1)}
-
-
-class _TinyTrainEpochRealModel(torch.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.param = torch.nn.Parameter(torch.zeros(2))
-
-    def forward(self, data, metadata=None, return_logits=False):
-        seq_len = data["real_col"].shape[1]
-        return {"real_col": self.param[0].expand(seq_len, 1, 1)}
-
-
-class _TinyMultiTargetModel(torch.nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.cat_logits = torch.nn.Parameter(torch.tensor([0.2, -0.1, 0.0]))
-        self.real_param = torch.nn.Parameter(torch.tensor([0.5, 0.0]))
-
-    def forward(self, seq_len):
-        return {
-            "cat_col": self.cat_logits.reshape(1, 1, 3).expand(seq_len, 1, 3),
-            "real_col": self.real_param[0].expand(seq_len, 1, 1),
-        }
-
-
-def _free_port():
-    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as sock:
-        sock.bind(("127.0.0.1", 0))
-        return sock.getsockname()[1]
-
-
-def _loss_shell(target_column_types, *, device, loss_weights=None):
-    model = TransformerModel.__new__(TransformerModel)
-    model.target_columns = list(target_column_types)
-    model.target_column_types = dict(target_column_types)
-    model.loss_weights = loss_weights
-    model.device = device
-    model.hparams = SimpleNamespace(
-        training_spec=SimpleNamespace(
-            data_parallelism="FSDP",
-            distributed=True,
-            training_objective="causal",
-            layer_autocast=False,
-            layer_type_dtypes=None,
-        )
-    )
-    model.criterion = {}
-    model.n_classes = {}
-
-    for col, col_type in target_column_types.items():
-        if col_type == "real":
-            model.criterion[col] = torch.nn.MSELoss(reduction="none")
-        elif col_type == "categorical":
-            model.criterion[col] = torch.nn.CrossEntropyLoss(reduction="none")
-            model.n_classes[col] = 3
-        else:
-            raise ValueError(col_type)
-
-    return model
-
-
-class _IdentityScaler:
-    def scale(self, loss):
-        return loss
-
-    def unscale_(self, optimizer):
-        return None
-
-    def step(self, optimizer):
-        optimizer.step()
-
-    def update(self):
-        return None
-
-
-class _NoopLogger:
-    def info(self, *args, **kwargs):
-        return None
-
-    def warning(self, *args, **kwargs):
-        return None
-
-
-def _real_batch(seq_len, selected_count, target_value, device):
-    targets = {"real_col": torch.full((1, seq_len), float(target_value), device=device)}
-    mask = torch.zeros(1, seq_len, dtype=torch.bool, device=device)
-    mask[:, :selected_count] = True
-    metadata = {"target_valid_mask": mask}
-    return targets, metadata
-
-
-def _real_epoch_batch(seq_len, selected_count, target_value):
-    targets = {"real_col": torch.full((1, seq_len), float(target_value))}
-    mask = torch.zeros(1, seq_len, dtype=torch.bool)
-    mask[:, :selected_count] = True
-    return SequifierBatch(
-        inputs={"real_col": torch.ones(1, seq_len, dtype=torch.float32)},
-        targets=targets,
-        metadata={
-            "attention_valid_mask": torch.ones(1, seq_len, dtype=torch.bool),
-            "target_valid_mask": mask,
-        },
-    )
-
-
-def _concat_real_batch(seq_lens, selected_counts, target_values):
-    target_parts = []
-    mask_parts = []
-    for seq_len, selected_count, target_value in zip(
-        seq_lens, selected_counts, target_values
-    ):
-        target_parts.append(torch.full((1, seq_len), float(target_value)))
-        mask = torch.zeros(1, seq_len, dtype=torch.bool)
-        mask[:, :selected_count] = True
-        mask_parts.append(mask)
-    return (
-        {"real_col": torch.cat(target_parts, dim=1)},
-        {"target_valid_mask": torch.cat(mask_parts, dim=1)},
-    )
-
-
-def _single_process_real_update(seq_lens, selected_counts, target_values, lr):
-    model = _TinyRealModel()
-    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
-    shell = _loss_shell({"real_col": "real"}, device="cpu")
-    targets, metadata = _concat_real_batch(seq_lens, selected_counts, target_values)
-    loss, _ = TransformerModel._calculate_loss(
-        shell,
-        model(sum(seq_lens)),
-        targets,
-        metadata,
-    )
-    loss.backward()
-    optimizer.step()
-    return model.param.detach().clone()
-
-
-def _single_process_accumulated_real_update(microbatches, lr, accumulation_steps):
-    model = _TinyRealModel()
-    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
-    shell = _loss_shell({"real_col": "real"}, device="cpu")
-    for microbatch in microbatches:
-        targets, metadata = _concat_real_batch(
-            microbatch["seq_lens"],
-            microbatch["selected_counts"],
-            microbatch["target_values"],
-        )
-        loss, _ = TransformerModel._calculate_loss(
-            shell,
-            model(sum(microbatch["seq_lens"])),
-            targets,
-            metadata,
-        )
-        (loss / accumulation_steps).backward()
-    optimizer.step()
-    return model.param.detach().clone()
-
-
-def _analytical_accumulated_real_update(microbatches, lr, accumulation_steps):
-    accumulated_grad = 0.0
-    for microbatch in microbatches:
-        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
-        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
-        token_count = selected_counts.sum().item()
-        if token_count == 0:
-            continue
-
-        weighted_target_mean = (
-            selected_counts * target_values
-        ).sum().item() / token_count
-        accumulated_grad += -2.0 * weighted_target_mean / accumulation_steps
-
-    return torch.tensor([-lr * accumulated_grad, 0.0])
-
-
-def _analytical_unnormalized_accumulated_real_update(microbatches, lr):
-    accumulated_grad = 0.0
-    for microbatch in microbatches:
-        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
-        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
-        token_count = selected_counts.sum().item()
-        if token_count == 0:
-            continue
-
-        weighted_target_mean = (
-            selected_counts * target_values
-        ).sum().item() / token_count
-        accumulated_grad += -2.0 * weighted_target_mean
-
-    return torch.tensor([-lr * accumulated_grad, 0.0])
-
-
-def _analytical_combined_real_update(microbatches, lr):
-    weighted_target_sum = 0.0
-    token_count = 0.0
-    for microbatch in microbatches:
-        selected_counts = torch.tensor(microbatch["selected_counts"], dtype=torch.float)
-        target_values = torch.tensor(microbatch["target_values"], dtype=torch.float)
-        weighted_target_sum += (selected_counts * target_values).sum().item()
-        token_count += selected_counts.sum().item()
-
-    if token_count == 0:
-        return torch.tensor([0.0, 0.0])
-    return torch.tensor([lr * 2.0 * weighted_target_sum / token_count, 0.0])
-
-
-def _single_process_multi_target_update(
-    seq_lens,
-    cat_targets,
-    real_targets,
-    loss_weights,
-    lr,
-):
-    model = _TinyMultiTargetModel()
-    optimizer = torch.optim.SGD(model.parameters(), lr=lr)
-    shell = _loss_shell(
-        {"cat_col": "categorical", "real_col": "real"},
-        device="cpu",
-        loss_weights=loss_weights,
-    )
-    targets = {
-        "cat_col": torch.tensor([sum(cat_targets, [])], dtype=torch.long),
-        "real_col": torch.tensor([sum(real_targets, [])], dtype=torch.float32),
-    }
-    metadata = {"target_valid_mask": torch.ones(1, sum(seq_lens), dtype=torch.bool)}
-    loss, _ = TransformerModel._calculate_loss(
-        shell,
-        model(sum(seq_lens)),
-        targets,
-        metadata,
-    )
-    loss.backward()
-    optimizer.step()
-    return torch.cat(
-        [
-            model.cat_logits.detach().reshape(-1),
-            model.real_param.detach().reshape(-1),
-        ]
-    )
-
-
-def _state_vector(model, keys, rank):
-    options = StateDictOptions(full_state_dict=True, cpu_offload=True)
-    state = get_model_state_dict(model, options=options)
-    if rank != 0:
-        return None
-
-    values = []
-    for key in keys:
-        values.append(state[key].reshape(-1).to(dtype=torch.float32))
-    return torch.cat(values)
-
-
-def _mixed_precision_policy(case):
-    reduce_dtype = case.get("reduce_dtype")
-    if reduce_dtype is None:
-        return None
-
-    return MixedPrecisionPolicy(
-        param_dtype=torch.bfloat16,
-        reduce_dtype=reduce_dtype,
-        output_dtype=torch.bfloat16,
-    )
-
-
-def _fsdp_case_worker(rank, world_size, init_method, case, queue):
-    torch.cuda.set_device(rank)
-    device = torch.device(f"cuda:{rank}")
-    dist.init_process_group(
-        backend="nccl",
-        rank=rank,
-        world_size=world_size,
-        init_method=init_method,
-        timeout=datetime.timedelta(seconds=60),
-    )
-    try:
-        mesh = init_device_mesh("cuda", (world_size,))
-        mp_policy = _mixed_precision_policy(case)
-
-        if case["kind"] in {
-            "real_update",
-            "accumulation_update",
-            "empty_window_state",
-            "train_epoch_empty_window_state",
-        }:
-            if case["kind"] == "train_epoch_empty_window_state":
-                model = _TinyTrainEpochRealModel().to(device)
-                fully_shard(model, mesh=mesh, mp_policy=mp_policy)
-                shell = _loss_shell({"real_col": "real"}, device=str(device))
-                shell._data_parallel_group = mesh.get_group()
-                shell.input_columns = ["real_col"]
-                shell.rank = rank
-                shell.accumulation_steps = case["accumulation_steps"]
-                shell.scheduler_step_on = "batch"
-                shell.start_batch = 0
-                shell.log_interval = case["accumulation_steps"]
-                shell.scaler = _IdentityScaler()
-                shell.logger = _NoopLogger()
-                shell.save_latest_interval_minutes = None
-                shell.save_batch_interval_minutes = None
-                shell.save_batch_interval_minutes_val_loss = False
-                shell.last_latest_save_time = 0.0
-                shell.last_batch_save_time = 0.0
-                shell.parameters = model.parameters
-                shell.optimizer = torch.optim.AdamW(
-                    model.parameters(),
-                    lr=case["lr"],
-                    weight_decay=case["weight_decay"],
-                )
-                shell.scheduler = torch.optim.lr_scheduler.StepLR(
-                    shell.optimizer,
-                    step_size=1,
-                    gamma=0.1,
-                )
-                before = _state_vector(model, ["param"], rank)
-                train_loader = [
-                    _real_epoch_batch(
-                        microbatch["seq_lens"][rank],
-                        microbatch["selected_counts"][rank],
-                        microbatch["target_values"][rank],
-                    )
-                    for microbatch in case["microbatches"]
-                ]
-
-                TransformerModel._train_epoch(
-                    shell,
-                    DataLoader(train_loader, batch_size=None),
-                    DataLoader([], batch_size=None),
-                    epoch=1,
-                    ddp_model=model,
-                )
-
-                after = _state_vector(model, ["param"], rank)
-                if rank == 0:
-                    assert before is not None
-                    assert after is not None
-                    queue.put(
-                        [
-                            float((after - before).abs().max().item()),
-                            float(shell.scheduler.get_last_lr()[0]),
-                            float(shell.scheduler.last_epoch),
-                            float(len(shell.optimizer.state)),
-                        ]
-                    )
-                return
-
-            model = _TinyRealModel().to(device)
-            fully_shard(model, mesh=mesh, mp_policy=mp_policy)
-            shell = _loss_shell({"real_col": "real"}, device=str(device))
-            shell._data_parallel_group = mesh.get_group()
-
-            if case["kind"] == "empty_window_state":
-                optimizer = torch.optim.AdamW(
-                    model.parameters(),
-                    lr=case["lr"],
-                    weight_decay=case["weight_decay"],
-                )
-                scheduler = torch.optim.lr_scheduler.StepLR(
-                    optimizer,
-                    step_size=1,
-                    gamma=0.1,
-                )
-                before = _state_vector(model, ["param"], rank)
-                accumulated_global_token_count = torch.zeros(
-                    (), dtype=torch.int64, device=device
-                )
-
-                for batch_idx, microbatch in enumerate(case["microbatches"]):
-                    seq_len = microbatch["seq_lens"][rank]
-                    targets, metadata = _real_batch(
-                        seq_len,
-                        microbatch["selected_counts"][rank],
-                        microbatch["target_values"][rank],
-                        device,
-                    )
-                    loss, _, _, _, global_count = (
-                        TransformerModel._calculate_training_loss(
-                            shell,
-                            model(seq_len),
-                            targets,
-                            metadata,
-                        )
-                    )
-                    (loss / case["accumulation_steps"]).backward()
-                    accumulated_global_token_count += global_count.detach()
-                    optimizer_step_due = (batch_idx + 1) % case[
-                        "accumulation_steps"
-                    ] == 0 or (batch_idx + 1) == len(case["microbatches"])
-                    optimizer_step_performed = False
-                    if (
-                        optimizer_step_due
-                        and accumulated_global_token_count.detach().cpu().item() > 0
-                    ):
-                        optimizer.step()
-                        optimizer.zero_grad()
-                        optimizer_step_performed = True
-
-                    if optimizer_step_due:
-                        if not optimizer_step_performed:
-                            optimizer.zero_grad()
-                        accumulated_global_token_count.zero_()
-
-                    if optimizer_step_performed:
-                        scheduler.step()
-
-                after = _state_vector(model, ["param"], rank)
-                if rank == 0:
-                    assert before is not None
-                    assert after is not None
-                    queue.put(
-                        [
-                            float((after - before).abs().max().item()),
-                            float(scheduler.get_last_lr()[0]),
-                            float(scheduler.last_epoch),
-                            float(len(optimizer.state)),
-                        ]
-                    )
-                return
-
-            optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
-            if case["kind"] == "accumulation_update":
-                for microbatch in case["microbatches"]:
-                    seq_len = microbatch["seq_lens"][rank]
-                    targets, metadata = _real_batch(
-                        seq_len,
-                        microbatch["selected_counts"][rank],
-                        microbatch["target_values"][rank],
-                        device,
-                    )
-                    loss, _ = TransformerModel._calculate_loss(
-                        shell,
-                        model(seq_len),
-                        targets,
-                        metadata,
-                    )
-                    (loss / case["accumulation_steps"]).backward()
-            else:
-                seq_len = case["seq_lens"][rank]
-                targets, metadata = _real_batch(
-                    seq_len,
-                    case["selected_counts"][rank],
-                    case["target_values"][rank],
-                    device,
-                )
-                loss, _ = TransformerModel._calculate_loss(
-                    shell,
-                    model(seq_len),
-                    targets,
-                    metadata,
-                )
-                loss.backward()
-
-            optimizer.step()
-            state = _state_vector(model, ["param"], rank)
-            if rank == 0:
-                assert state is not None
-                queue.put(state.tolist())
-            return
-
-        if case["kind"] == "multi_target_update":
-            model = _TinyMultiTargetModel().to(device)
-            fully_shard(model, mesh=mesh, mp_policy=mp_policy)
-            shell = _loss_shell(
-                {"cat_col": "categorical", "real_col": "real"},
-                device=str(device),
-                loss_weights=case["loss_weights"],
-            )
-            shell._data_parallel_group = mesh.get_group()
-            optimizer = torch.optim.SGD(model.parameters(), lr=case["lr"])
-            seq_len = case["seq_lens"][rank]
-            targets = {
-                "cat_col": torch.tensor(
-                    [case["cat_targets"][rank]], dtype=torch.long, device=device
-                ),
-                "real_col": torch.tensor(
-                    [case["real_targets"][rank]], dtype=torch.float32, device=device
-                ),
-            }
-            metadata = {
-                "target_valid_mask": torch.ones(
-                    1, seq_len, dtype=torch.bool, device=device
-                )
-            }
-            loss, _ = TransformerModel._calculate_loss(
-                shell,
-                model(seq_len),
-                targets,
-                metadata,
-            )
-            loss.backward()
-            optimizer.step()
-            state = _state_vector(model, ["cat_logits", "real_param"], rank)
-            if rank == 0:
-                assert state is not None
-                queue.put(state.tolist())
-            return
-
-        raise ValueError(case["kind"])
-    finally:
-        dist.destroy_process_group()
-        torch.cuda.empty_cache()
-
-
-def _run_fsdp_case(case):
-    ctx = mp.get_context("spawn")
-    queue = ctx.SimpleQueue()
-    init_method = f"tcp://127.0.0.1:{_free_port()}"
-    mp.spawn(
-        _fsdp_case_worker,
-        args=(2, init_method, case, queue),
-        nprocs=2,
-        join=True,
-    )
-    return torch.tensor(queue.get(), dtype=torch.float32)
-
-
-def test_fsdp_unequal_token_counts_match_single_process_reference():
-    case = {
-        "kind": "real_update",
-        "seq_lens": [100, 10],
-        "selected_counts": [100, 10],
-        "target_values": [2.0, 8.0],
-        "lr": 0.1,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_real_update(
-        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
-    )
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
-
-
-@pytest.mark.parametrize("reduce_dtype", [torch.bfloat16, torch.float32])
-def test_fsdp_mixed_precision_unequal_token_counts_match_reference(reduce_dtype):
-    case = {
-        "kind": "real_update",
-        "seq_lens": [100, 10],
-        "selected_counts": [100, 10],
-        "target_values": [2.0, 8.0],
-        "lr": 0.1,
-        "reduce_dtype": reduce_dtype,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_real_update(
-        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
-    )
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-2, rtol=1e-2)
-
-
-def test_fsdp_zero_token_rank_does_not_attenuate_populated_rank():
-    case = {
-        "kind": "real_update",
-        "seq_lens": [5, 5],
-        "selected_counts": [5, 0],
-        "target_values": [3.0, 100.0],
-        "lr": 0.1,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_real_update(
-        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
-    )
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
-
-
-@pytest.mark.parametrize("reduce_dtype", [torch.bfloat16, torch.float32])
-def test_fsdp_mixed_precision_zero_token_rank_is_not_attenuated(reduce_dtype):
-    case = {
-        "kind": "real_update",
-        "seq_lens": [5, 5],
-        "selected_counts": [5, 0],
-        "target_values": [3.0, 100.0],
-        "lr": 0.1,
-        "reduce_dtype": reduce_dtype,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_real_update(
-        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
-    )
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-2, rtol=1e-2)
-
-
-def test_fsdp_multi_target_weighting_matches_single_process_reference():
-    case = {
-        "kind": "multi_target_update",
-        "seq_lens": [3, 1],
-        "cat_targets": [[0, 1, 2], [1]],
-        "real_targets": [[1.0, 2.0, 3.0], [4.0]],
-        "loss_weights": {"cat_col": 0.25, "real_col": 2.0},
-        "lr": 0.1,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_multi_target_update(
-        case["seq_lens"],
-        case["cat_targets"],
-        case["real_targets"],
-        case["loss_weights"],
-        case["lr"],
-    )
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
-
-
-def test_fsdp_world_size_two_optimizer_update_matches_single_process_reference():
-    case = {
-        "kind": "real_update",
-        "seq_lens": [6, 4],
-        "selected_counts": [6, 4],
-        "target_values": [1.0, 3.0],
-        "lr": 0.1,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _single_process_real_update(
-        case["seq_lens"], case["selected_counts"], case["target_values"], case["lr"]
-    )
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
-
-
-def test_fsdp_gradient_accumulation_averages_per_microbatch_weighting():
-    microbatches = [
-        {
-            "seq_lens": [1, 1],
-            "selected_counts": [1, 1],
-            "target_values": [1.0, 3.0],
-        },
-        {
-            "seq_lens": [5, 5],
-            "selected_counts": [5, 5],
-            "target_values": [10.0, 2.0],
-        },
-    ]
-    case = {
-        "kind": "accumulation_update",
-        "microbatches": microbatches,
-        "accumulation_steps": 2,
-        "lr": 0.05,
-    }
-
-    fsdp_param = _run_fsdp_case(case)
-    reference_param = _analytical_accumulated_real_update(
-        microbatches, case["lr"], case["accumulation_steps"]
-    )
-    unnormalized_param = _analytical_unnormalized_accumulated_real_update(
-        microbatches, case["lr"]
-    )
-    combined_batch_param = _analytical_combined_real_update(microbatches, case["lr"])
-
-    assert torch.allclose(fsdp_param, reference_param, atol=1e-5)
-    assert not torch.allclose(fsdp_param, unnormalized_param, atol=1e-3)
-    assert not torch.allclose(fsdp_param, combined_batch_param, atol=1e-3)
-
-
-def test_fsdp_globally_empty_accumulation_window_leaves_state_unchanged():
-    case = {
-        "kind": "empty_window_state",
-        "microbatches": [
-            {
-                "seq_lens": [3, 4],
-                "selected_counts": [0, 0],
-                "target_values": [1.0, 3.0],
-            },
-            {
-                "seq_lens": [2, 5],
-                "selected_counts": [0, 0],
-                "target_values": [2.0, 4.0],
-            },
-        ],
-        "accumulation_steps": 2,
-        "lr": 0.01,
-        "weight_decay": 0.1,
-    }
-
-    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_fsdp_case(case)
-
-    assert param_delta == 0
-    assert lr == pytest.approx(case["lr"])
-    assert scheduler_epoch == 0
-    assert optimizer_state_len == 0
-
-
-def test_fsdp_train_epoch_globally_empty_accumulation_window_leaves_state_unchanged():
-    case = {
-        "kind": "train_epoch_empty_window_state",
-        "microbatches": [
-            {
-                "seq_lens": [3, 4],
-                "selected_counts": [0, 0],
-                "target_values": [1.0, 3.0],
-            },
-            {
-                "seq_lens": [2, 5],
-                "selected_counts": [0, 0],
-                "target_values": [2.0, 4.0],
-            },
-        ],
-        "accumulation_steps": 2,
-        "lr": 0.01,
-        "weight_decay": 0.1,
-    }
-
-    param_delta, lr, scheduler_epoch, optimizer_state_len = _run_fsdp_case(case)
-
-    assert param_delta == 0
-    assert lr == pytest.approx(case["lr"])
-    assert scheduler_epoch == 0
-    assert optimizer_state_len == 0

From bb11b238e9cd77b23d809572818cc712997dd781 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 17 Jun 2026 11:51:02 +0200
Subject: [PATCH 68/81] allow_sequence_splitting and device dependent torch
 float type

---
 src/sequifier/config/preprocess_config.py |  1 +
 src/sequifier/preprocess.py               | 57 ++++++++++++++++++++---
 src/sequifier/train.py                    | 36 ++++++++++----
 3 files changed, 78 insertions(+), 16 deletions(-)

diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index bcff237b..ef1cdf4c 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -42,6 +42,7 @@ class PreprocessorModel(BaseModel):
     read_format: str = "csv"
     write_format: str = "parquet"
     merge_output: bool = True
+    allow_sequence_splitting: bool = False
     selected_columns: Optional[list[str]] = None
 
     split_ratios: list[float]
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index c2556d03..0550eb6f 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -81,6 +81,7 @@ def __init__(
         read_format: str,
         write_format: str,
         merge_output: bool,
+        allow_sequence_splitting: bool,
         selected_columns: Optional[list[str]],
         split_ratios: list[float],
         stored_context_width: int,
@@ -114,6 +115,8 @@ def __init__(
                 )
             self.target_dir = f"{self.data_name_root}-temp"
 
+        self.allow_sequence_splitting = allow_sequence_splitting
+
         self.use_precomputed_maps = use_precomputed_maps
         self.metadata_config_path = metadata_config_path
         self.mask_column = mask_column
@@ -257,6 +260,7 @@ def __init__(
                 self.batches_per_file,
                 subsequence_start_mode,
                 self.merge_output,
+                self.allow_sequence_splitting,
             )
 
             if self.merge_output:
@@ -590,6 +594,7 @@ def _process_batches_multiple_files(
                 target_dir=self.target_dir,
                 batches_per_file=self.batches_per_file,
                 merge_output=self.merge_output,
+                allow_sequence_splitting=self.allow_sequence_splitting,
                 continue_preprocessing=self.continue_preprocessing,
                 subsequence_start_mode=subsequence_start_mode,
                 mask_column=mask_column,
@@ -635,6 +640,7 @@ def _process_batches_multiple_files(
                 "target_dir": self.target_dir,
                 "batches_per_file": self.batches_per_file,
                 "merge_output": self.merge_output,
+                "allow_sequence_splitting": self.allow_sequence_splitting,
                 "continue_preprocessing": self.continue_preprocessing,
                 "subsequence_start_mode": subsequence_start_mode,
                 "mask_column": mask_column,
@@ -1310,6 +1316,7 @@ def _process_batches_multiple_files_inner(
     target_dir: str,
     batches_per_file: int,
     merge_output: bool,
+    allow_sequence_splitting: bool,
     continue_preprocessing: bool,
     subsequence_start_mode: str,
     mask_column: Optional[str],
@@ -1402,6 +1409,7 @@ def _process_batches_multiple_files_inner(
                 batches_per_file,
                 subsequence_start_mode,
                 merge_output,
+                allow_sequence_splitting,
             )
 
             if merge_output:
@@ -1449,10 +1457,11 @@ def _process_batches_single_file(
     batches_per_file: int,
     subsequence_start_mode: str,
     merge_output: bool,
+    allow_sequence_splitting: bool,
 ) -> int:
     """Split one file into worker batches and preprocess them."""
     n_cores = n_cores or multiprocessing.cpu_count()
-    batch_limits = get_batch_limits(data, n_cores)
+    batch_limits = get_batch_limits(data, n_cores, allow_sequence_splitting)
     valid_batch_limits = [(s, e) for s, e in batch_limits if (e - s) > 0]
     batches = [
         (
@@ -1551,8 +1560,10 @@ def create_id_map(data: pl.DataFrame, column: str) -> dict[Union[str, int], int]
 
 
 @beartype
-def get_batch_limits(data: pl.DataFrame, n_batches: int) -> list[tuple[int, int]]:
-    """Split rows into batches without crossing sequenceId boundaries."""
+def get_batch_limits(
+    data: pl.DataFrame, n_batches: int, allow_sequence_splitting: bool
+) -> list[tuple[int, int]]:
+    """Split rows into batches without crossing sequenceId boundaries unless allowed."""
     if n_batches <= 0:
         raise ValueError("n_batches must be positive.")
     if data.is_empty():
@@ -1566,10 +1577,42 @@ def get_batch_limits(data: pl.DataFrame, n_batches: int) -> list[tuple[int, int]
     sequence_count = len(sequence_start_indices)
 
     if n_batches > sequence_count:
-        raise ValueError(
-            "Cannot create more non-empty batches than there are sequences without "
-            "splitting a sequence."
-        )
+        if not allow_sequence_splitting:
+            raise ValueError(
+                "Cannot create more non-empty batches than there are sequences without "
+                "splitting a sequence."
+            )
+
+        original_lengths = np.diff(sequence_boundaries)
+        pieces = np.ones(len(original_lengths), dtype=int)
+
+        for _ in range(n_batches - sequence_count):
+            largest_piece_idx = int(np.argmax(original_lengths / pieces))
+            pieces[largest_piece_idx] += 1
+
+        if np.any(pieces > original_lengths):
+            raise ValueError(
+                "Cannot split further: sequences are too short to reach the "
+                "requested number of non-empty batches."
+            )
+
+        new_boundaries = []
+        for start, length, num_pieces in zip(
+            sequence_boundaries[:-1], original_lengths, pieces
+        ):
+            # Calculate evenly spaced boundaries within this sequence
+            splits = start + np.round(np.linspace(0, length, num_pieces + 1)).astype(
+                int
+            )
+
+            if not new_boundaries:
+                new_boundaries.extend(splits)
+            else:
+                new_boundaries.extend(
+                    splits[1:]
+                )  # Avoid duplicating the shared boundary
+
+        sequence_boundaries = np.array(new_boundaries)
 
     interior_boundaries = sequence_boundaries[1:-1]
     ideal_limits = np.linspace(0, data.shape[0], n_batches + 1)[1:-1]
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 3d6123f3..ed8100cb 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1737,10 +1737,20 @@ def _calculate_loss_components(
             output, targets, valid_mask
         )
         return (
-            {col: loss_sum.detach().double() for col, loss_sum in loss_sums.items()},
+            {
+                col: loss_sum.detach().to(dtype=self._metric_float_dtype())
+                for col, loss_sum in loss_sums.items()
+            },
             token_count.detach(),
         )
 
+    @beartype
+    def _metric_float_dtype(self) -> torch.dtype:
+        """Return the highest precision floating dtype supported by this device."""
+        if torch.device(self.device).type == "mps":
+            return torch.float32
+        return torch.float64
+
     @beartype
     def _loss_target_names(
         self, targets: Optional[dict[str, Tensor]] = None
@@ -1795,12 +1805,13 @@ def _new_loss_accumulators(
         self, target_names: list[str]
     ) -> tuple[dict[str, Tensor], Tensor]:
         """Create detached sum/count accumulators for logging or validation."""
+        dtype = self._metric_float_dtype()
         return (
             {
-                col: torch.zeros((), device=self.device, dtype=torch.float64)
+                col: torch.zeros((), device=self.device, dtype=dtype)
                 for col in target_names
             },
-            torch.zeros((), device=self.device, dtype=torch.float64),
+            torch.zeros((), device=self.device, dtype=dtype),
         )
 
     @beartype
@@ -1813,8 +1824,11 @@ def _accumulate_loss_components(
     ) -> None:
         """Accumulate detached local unweighted loss sums and token counts."""
         for col in batch_sums:
-            sums[col] = sums[col] + batch_sums[col].detach().double()
-        count += batch_count.detach().double()
+            sums[col] = sums[col] + batch_sums[col].detach().to(
+                device=sums[col].device,
+                dtype=sums[col].dtype,
+            )
+        count += batch_count.detach().to(device=count.device, dtype=count.dtype)
 
     @beartype
     def _finalize_loss_components(
@@ -1843,14 +1857,15 @@ def _finalize_loss_components(
             if raise_on_empty:
                 raise RuntimeError(f"No valid {label} tokens found.")
 
+            dtype = self._metric_float_dtype()
             losses = {
-                col: torch.zeros((), device=self.device, dtype=torch.float64)
+                col: torch.zeros((), device=self.device, dtype=dtype)
                 for col in target_names
             }
-            return torch.zeros((), device=self.device, dtype=torch.float64), losses
+            return torch.zeros((), device=self.device, dtype=dtype), losses
 
         losses = {}
-        total = torch.zeros((), device=self.device, dtype=torch.float64)
+        total = torch.zeros((), device=self.device, dtype=self._metric_float_dtype())
         for col in target_names:
             losses[col] = reduced_sums[col] / reduced_count * self._loss_weight(col)
             total = total + losses[col]
@@ -2437,7 +2452,10 @@ def _log_epoch_results(
                     )
                     continue
 
-                shares = counts.to(torch.float64) / total
+                share_dtype = (
+                    torch.float32 if counts.device.type == "mps" else torch.float64
+                )
+                shares = counts.to(share_dtype) / total
 
                 value_shares = " | ".join(
                     f"{self.index_maps[categorical_column][class_id]}: "

From d428055ccd8756acd6ac4581d1799ce92f6495a7 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 17 Jun 2026 16:48:41 +0200
Subject: [PATCH 69/81] Correct state saving and loading WIP

---
 src/sequifier/train.py | 554 +++++++++++++++++++++++++++++++----------
 1 file changed, 426 insertions(+), 128 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index ed8100cb..f335b1e1 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1,17 +1,20 @@
 import contextlib
 import copy
 import glob
+import hashlib
 import json
 import logging
 import math
 import os
+import random
+import re
 import sys
 
 os.environ["TORCH_NCCL_ASYNC_ERROR_HANDLING"] = "1"
 import time  # noqa: E402
 import uuid  # noqa: E402
 import warnings  # noqa: E402
-from typing import Any, Optional, Union  # noqa: E402
+from typing import Any, Optional, Union, cast  # noqa: E402
 
 import numpy as np  # noqa: E402
 import torch  # noqa: E402
@@ -52,6 +55,7 @@
 torch._dynamo.config.suppress_errors = True
 
 ClassCounts = dict[str, Tensor]
+CHECKPOINT_FORMAT_VERSION = 2
 
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
@@ -212,15 +216,20 @@ def train_worker(
             checkpoint = torch.load(
                 latest_model_path, map_location="cpu", weights_only=False
             )
+            model._validate_checkpoint_compatibility(checkpoint, len(train_loader))
             model.load_state_dict(checkpoint["model_state_dict"])
             model.optimizer.load_state_dict(checkpoint["optimizer_state_dict"])
             model.scheduler.load_state_dict(checkpoint["scheduler_state_dict"])
-            if checkpoint["batch"] + 1 >= len(train_loader):
-                base_model.start_epoch = checkpoint["epoch"] + 1
-                base_model.start_batch = 0
-            else:
-                base_model.start_epoch = checkpoint["epoch"]
-                base_model.start_batch = checkpoint["batch"] + 1
+            base_model.start_epoch, base_model.start_batch = _checkpoint_start_position(
+                checkpoint, len(train_loader)
+            )
+            model._apply_checkpoint_training_state(
+                checkpoint.get("scaler_state_dict"),
+                checkpoint.get("best_val_loss", float("inf")),
+                checkpoint.get("n_epochs_no_improvement", 0),
+                checkpoint.get("best_model_state_dict"),
+                checkpoint.get("rng_state"),
+            )
         else:
             model.start_epoch = 1
             model.start_batch = 0
@@ -235,6 +244,9 @@ def train_worker(
                 for i in range(len(model.layers)):
                     model.layers[i] = torch.compile(model.layers[i])
 
+        if checkpoint is not None:
+            base_model._restore_rng_state()
+
         model.train_model(train_loader, valid_loader, ddp_model=None)
     elif config.training_spec.data_parallelism == "FSDP":
         mesh = init_device_mesh(
@@ -277,13 +289,20 @@ def train_worker(
                 )
                 full_msd = checkpoint["model_state_dict"]
                 full_osd = checkpoint["optimizer_state_dict"]
-
-                if checkpoint["batch"] + 1 >= len(train_loader):
-                    start_epoch = checkpoint["epoch"] + 1
-                    start_batch = 0
-                else:
-                    start_epoch = checkpoint["epoch"]
-                    start_batch = checkpoint["batch"] + 1
+                start_epoch, start_batch = _checkpoint_start_position(
+                    checkpoint, len(train_loader)
+                )
+                resume_state = {
+                    "scaler_state_dict": checkpoint.get("scaler_state_dict"),
+                    "best_val_loss": checkpoint.get("best_val_loss", float("inf")),
+                    "n_epochs_no_improvement": checkpoint.get(
+                        "n_epochs_no_improvement", 0
+                    ),
+                    "has_best_model_state_dict": checkpoint.get("best_model_state_dict")
+                    is not None,
+                    "rng_state": checkpoint.get("rng_state"),
+                    "checkpoint_metadata": checkpoint.get("checkpoint_metadata"),
+                }
 
                 meta = [
                     start_epoch,
@@ -291,15 +310,33 @@ def train_worker(
                     checkpoint["scheduler_state_dict"],
                     full_msd,
                     full_osd,
+                    resume_state,
                 ]
             else:
-                meta = [None, None, None, None, None]
+                meta = [None, None, None, None, None, None]
 
             # Broadcast the checkpoint data to all ranks simultaneously
             dist.broadcast_object_list(meta, src=0)
 
-            # Unpack on all ranks
-            model.start_epoch, model.start_batch, sched_state, full_msd, full_osd = meta  # type: ignore
+            # Unpack on all ranks. The placeholder Nones are replaced by broadcast.
+            (
+                start_epoch_obj,
+                start_batch_obj,
+                sched_state_obj,
+                full_msd_obj,
+                full_osd_obj,
+                resume_state_obj,
+            ) = meta
+            model.start_epoch = cast(int, start_epoch_obj)
+            model.start_batch = cast(int, start_batch_obj)
+            sched_state = cast(Optional[dict[str, Any]], sched_state_obj)
+            full_msd = cast(dict[str, Tensor], full_msd_obj)
+            full_osd = cast(dict[str, Any], full_osd_obj)
+            resume_state = cast(dict[str, Any], resume_state_obj)
+            model._validate_checkpoint_compatibility(
+                {"checkpoint_metadata": resume_state.get("checkpoint_metadata")},
+                len(train_loader),
+            )
 
             options = StateDictOptions(full_state_dict=True, cpu_offload=True)
 
@@ -318,6 +355,19 @@ def train_worker(
 
             if sched_state is not None:
                 base_model.scheduler.load_state_dict(sched_state)
+            best_model_state_dict = None
+            if resume_state.get("has_best_model_state_dict"):
+                if global_rank == 0 and checkpoint is not None:
+                    best_model_state_dict = checkpoint.get("best_model_state_dict")
+                else:
+                    best_model_state_dict = {}
+            model._apply_checkpoint_training_state(
+                resume_state.get("scaler_state_dict"),
+                resume_state.get("best_val_loss", float("inf")),
+                resume_state.get("n_epochs_no_improvement", 0),
+                best_model_state_dict,
+                resume_state.get("rng_state"),
+            )
 
         else:
             model.start_epoch = 1
@@ -340,22 +390,33 @@ def train_worker(
 
             dist.barrier()
 
+        if checkpoint is not None:
+            base_model._restore_rng_state()
+
         model.train_model(train_loader, valid_loader, ddp_model=base_model)
         cleanup()
     elif config.training_spec.data_parallelism == "DDP":  # DDP
+        params_to_optimize = model.parameters()
+        model.initialize_optimizer(params=params_to_optimize)
+
         if config.training_spec.continue_training and latest_model_path:
             checkpoint = torch.load(
                 latest_model_path, map_location="cpu", weights_only=False
             )
+            base_model._validate_checkpoint_compatibility(checkpoint, len(train_loader))
             base_model.load_state_dict(checkpoint["model_state_dict"])
             base_model.optimizer.load_state_dict(checkpoint["optimizer_state_dict"])
             base_model.scheduler.load_state_dict(checkpoint["scheduler_state_dict"])
-            if checkpoint["batch"] + 1 >= len(train_loader):
-                base_model.start_epoch = checkpoint["epoch"] + 1
-                base_model.start_batch = 0
-            else:
-                base_model.start_epoch = checkpoint["epoch"]
-                base_model.start_batch = checkpoint["batch"] + 1
+            base_model.start_epoch, base_model.start_batch = _checkpoint_start_position(
+                checkpoint, len(train_loader)
+            )
+            base_model._apply_checkpoint_training_state(
+                checkpoint.get("scaler_state_dict"),
+                checkpoint.get("best_val_loss", float("inf")),
+                checkpoint.get("n_epochs_no_improvement", 0),
+                checkpoint.get("best_model_state_dict"),
+                checkpoint.get("rng_state"),
+            )
         else:
             model.start_epoch = 1
             model.start_batch = 0
@@ -363,9 +424,6 @@ def train_worker(
                 f"[INFO] Initializing new model with {format_number(pytorch_total_params)} parameters."
             )
 
-        params_to_optimize = model.parameters()
-        model.initialize_optimizer(params=params_to_optimize)
-
         if config.training_spec.device.startswith("cuda"):
             if torch_compile == "outer":
                 model = torch.compile(model)
@@ -390,6 +448,8 @@ def train_worker(
                     _ = ddp_model(dummy_data, dummy_metadata, False)
 
             dist.barrier()
+        if checkpoint is not None:
+            base_model._restore_rng_state()
         model.train_model(train_loader, valid_loader, ddp_model=ddp_model)
         cleanup()
     else:
@@ -466,17 +526,13 @@ def train(args: Any, args_config: dict[str, Any]) -> None:
 
 
 @beartype
-def format_number(number: Union[int, float, np.float32]) -> str:
-    """Format finite values, zero, and NaN for logs."""
-    if np.isnan(number):
+def format_number(number: int | float | np.float32) -> str:
+    value = float(number)
+    if math.isnan(value):
         return "NaN"
-    elif number == 0:
-        order_of_magnitude = 0
-    else:
-        order_of_magnitude = math.floor(math.log(np.abs(number), 10))
-
-    number_adjusted = number * (10 ** (-order_of_magnitude))
-    return f"{number_adjusted:5.2f}e{order_of_magnitude}"
+    if math.isinf(value):
+        return "Inf" if value > 0 else "-Inf"
+    return f"{value: .2e}"
 
 
 def _get_evaluation_loss_mask(metadata: dict[str, Tensor]) -> Tensor:
@@ -500,6 +556,16 @@ def _get_evaluation_loss_mask(metadata: dict[str, Tensor]) -> Tensor:
     return valid_mask
 
 
+@beartype
+def _checkpoint_start_position(
+    checkpoint: dict[str, Any], num_batches: int
+) -> tuple[int, int]:
+    """Return the next epoch/batch position after a saved checkpoint."""
+    if checkpoint["batch"] + 1 >= num_batches:
+        return checkpoint["epoch"] + 1, 0
+    return checkpoint["epoch"], checkpoint["batch"] + 1
+
+
 @beartype
 def accumulate_class_counts(
     counts: ClassCounts,
@@ -787,6 +853,10 @@ def __init__(
                 use_scaler = True
 
         self.scaler = GradScaler(device=self.device.split(":")[0], enabled=use_scaler)
+        self._resume_best_val_loss = float("inf")
+        self._resume_n_epochs_no_improvement = 0
+        self._resume_best_model_state_dict = None
+        self._resume_rng_state = None
 
         self._apply_layer_dtypes()
 
@@ -1186,6 +1256,182 @@ def _check_and_terminate(self):
 
                 sys.exit(143)
 
+    @beartype
+    def _checkpoint_compatibility_metadata(
+        self, num_batches: Optional[int]
+    ) -> dict[str, Any]:
+        """Return resume-critical settings stored with each new checkpoint."""
+        training_spec = self.hparams.training_spec
+        resume_settings = {
+            "model_name": self.model_name,
+            "num_batches": num_batches,
+            "batch_size": self.batch_size,
+            "accumulation_steps": self.accumulation_steps,
+            "learning_rate": training_spec.learning_rate,
+            "scheduler_step_on": self.scheduler_step_on,
+            "scheduler": dict(training_spec.scheduler),
+            "optimizer": dict(training_spec.optimizer),
+            "distributed": training_spec.distributed,
+            "data_parallelism": training_spec.data_parallelism,
+            "world_size": (
+                dist.get_world_size(group=self._data_parallel_process_group())
+                if self._distributed_is_initialized()
+                else training_spec.world_size
+            ),
+            "training_objective": training_spec.training_objective,
+            "input_columns": self.input_columns,
+            "target_columns": self.target_columns,
+            "target_column_types": self.target_column_types,
+            "model_spec": self.hparams.model_spec.model_dump(mode="json"),
+        }
+        fingerprint_input = json.dumps(
+            resume_settings, sort_keys=True, default=str
+        ).encode("utf-8")
+        return {
+            "format_version": CHECKPOINT_FORMAT_VERSION,
+            "config_fingerprint": hashlib.sha256(fingerprint_input).hexdigest(),
+            "resume_settings": resume_settings,
+        }
+
+    @beartype
+    def _validate_checkpoint_compatibility(
+        self, checkpoint: dict[str, Any], num_batches: int
+    ) -> None:
+        """Reject checkpoints whose resume-critical settings no longer match."""
+        checkpoint_metadata = checkpoint.get("checkpoint_metadata")
+        if checkpoint_metadata is None:
+            self.logger.warning(
+                "[WARNING] Checkpoint has no compatibility metadata; "
+                "continuing with legacy resume behavior."
+            )
+            return
+
+        format_version = checkpoint_metadata.get("format_version")
+        if format_version != CHECKPOINT_FORMAT_VERSION:
+            self.logger.warning(
+                "[WARNING] Checkpoint format version "
+                f"{format_version!r} differs from expected "
+                f"{CHECKPOINT_FORMAT_VERSION!r}."
+            )
+
+        saved_settings = checkpoint_metadata.get("resume_settings")
+        if saved_settings is None:
+            self.logger.warning(
+                "[WARNING] Checkpoint has no resume-settings fingerprint payload; "
+                "continuing without compatibility checks."
+            )
+            return
+
+        current_settings = self._checkpoint_compatibility_metadata(num_batches)[
+            "resume_settings"
+        ]
+        mismatches = []
+        for key, current_value in current_settings.items():
+            saved_value = saved_settings.get(key)
+            if saved_value != current_value:
+                mismatches.append(
+                    f"{key}: checkpoint={saved_value!r}, current={current_value!r}"
+                )
+
+        if mismatches:
+            mismatch_text = "; ".join(mismatches)
+            raise ValueError(
+                "Checkpoint is incompatible with the current training configuration: "
+                f"{mismatch_text}"
+            )
+
+    @beartype
+    def _get_rng_state(self) -> dict[str, Any]:
+        """Capture Python, NumPy, Torch CPU, and CUDA RNG state for this rank."""
+        return {
+            "python": random.getstate(),
+            "numpy": np.random.get_state(),
+            "torch": torch.get_rng_state(),
+            "cuda": torch.cuda.get_rng_state_all()
+            if torch.cuda.is_available()
+            else None,
+        }
+
+    @beartype
+    def _collect_rng_states_for_checkpoint(self) -> Optional[list[Any]]:
+        """Gather per-rank RNG states on rank 0 for checkpointing."""
+        rng_state = self._get_rng_state()
+        if not self.hparams.training_spec.distributed:
+            return [rng_state]
+
+        rng_states = (
+            [None] * dist.get_world_size(group=self._data_parallel_process_group())
+            if self.rank == 0
+            else None
+        )
+        dist.gather_object(
+            rng_state,
+            object_gather_list=rng_states,
+            dst=0,
+            group=self._data_parallel_process_group(),
+        )
+        return rng_states
+
+    @beartype
+    def _select_rng_state_for_rank(self, rng_states: Any) -> Optional[dict[str, Any]]:
+        """Return this rank's saved RNG state from a checkpoint payload."""
+        if rng_states is None:
+            return None
+        if isinstance(rng_states, dict):
+            return rng_states
+        if not isinstance(rng_states, list) or len(rng_states) == 0:
+            return None
+
+        rank = self.rank or 0
+        if rank < len(rng_states):
+            return rng_states[rank]
+        self.logger.warning(
+            "[WARNING] Checkpoint has no RNG state for this rank; "
+            "using rank 0 RNG state as a fallback."
+        )
+        return rng_states[0]
+
+    @beartype
+    def _apply_checkpoint_training_state(
+        self,
+        scaler_state_dict: Optional[dict[str, Any]],
+        best_val_loss: Any,
+        n_epochs_no_improvement: Any,
+        best_model_state_dict: Any,
+        rng_states: Any,
+    ) -> None:
+        """Restore non-model training state from a checkpoint payload."""
+        if scaler_state_dict is not None:
+            self.scaler.load_state_dict(scaler_state_dict)
+        elif self.scaler.is_enabled():
+            self.logger.warning(
+                "[WARNING] Checkpoint has no GradScaler state; "
+                "resuming with a freshly initialized scaler."
+            )
+
+        self._resume_best_val_loss = float(best_val_loss)
+        self._resume_n_epochs_no_improvement = int(n_epochs_no_improvement)
+        self._resume_best_model_state_dict = best_model_state_dict
+        self._resume_rng_state = self._select_rng_state_for_rank(rng_states)
+
+    @beartype
+    def _restore_rng_state(self) -> None:
+        """Apply the checkpoint RNG state after compile/warm-up work is finished."""
+        rng_state = self._resume_rng_state
+        if rng_state is None:
+            self.logger.warning(
+                "[WARNING] Checkpoint has no RNG state; stochastic training will "
+                "continue from the current process RNG state."
+            )
+            return
+
+        random.setstate(rng_state["python"])
+        np.random.set_state(rng_state["numpy"])
+        torch.set_rng_state(rng_state["torch"])
+        cuda_state = rng_state.get("cuda")
+        if cuda_state is not None and torch.cuda.is_available():
+            torch.cuda.set_rng_state_all(cuda_state)
+
     @beartype
     def train_model(
         self,
@@ -1196,10 +1442,10 @@ def train_model(
         """Run epochs, validation, checkpointing, export, and interruption cleanup."""
         self.logger.info(f"--- Starting Training for model: {self.model_name} ---")
 
-        best_val_loss = float("inf")
-        n_epochs_no_improvement = 0
+        best_val_loss: float = float(self._resume_best_val_loss)
+        n_epochs_no_improvement = self._resume_n_epochs_no_improvement
         last_epoch = self.start_epoch - 1
-        best_model_state = None
+        best_model_state = self._resume_best_model_state_dict
 
         try:
             self.last_latest_save_time = time.time()
@@ -1231,7 +1477,15 @@ def train_model(
                     train_loader.dataset.set_epoch(epoch)
                     valid_loader.dataset.set_epoch(epoch)
 
-                    self._train_epoch(train_loader, valid_loader, epoch, ddp_model)
+                    self._train_epoch(
+                        train_loader,
+                        valid_loader,
+                        epoch,
+                        ddp_model,
+                        best_val_loss,
+                        n_epochs_no_improvement,
+                        best_model_state,
+                    )
 
                     total_loss, total_losses, class_counts = self._evaluate(
                         valid_loader, ddp_model
@@ -1250,7 +1504,7 @@ def train_model(
                     )
 
                     if total_loss < best_val_loss:
-                        best_val_loss = total_loss
+                        best_val_loss = float(total_loss)
                         best_model_state = self._get_full_state_dict(ddp_model)
                         n_epochs_no_improvement = 0
                     else:
@@ -1270,6 +1524,10 @@ def train_model(
                             total_loss,
                             ddp_model=ddp_model,
                             suffix=f"epoch-{epoch}",
+                            best_val_loss=best_val_loss,
+                            n_epochs_no_improvement=n_epochs_no_improvement,
+                            best_model_state_dict=best_model_state,
+                            num_batches=len(train_loader),
                         )
 
                     last_epoch = epoch
@@ -1357,6 +1615,9 @@ def _train_epoch(
         valid_loader: DataLoader,
         epoch: int,
         ddp_model: Optional[nn.Module] = None,
+        best_val_loss: float = float("inf"),
+        n_epochs_no_improvement: int = 0,
+        best_model_state: Optional[dict[str, Tensor]] = None,
     ) -> None:
         """Run one train epoch with optional mid-epoch saves."""
         target_names = self._loss_target_names()
@@ -1528,85 +1789,93 @@ def _train_epoch(
                     ):
                         self.scheduler.step()
 
-                should_save_latest = torch.tensor(
-                    [0], dtype=torch.int32, device=self.device
-                )
-                should_save_batch = torch.tensor(
-                    [0], dtype=torch.int32, device=self.device
-                )
-                val_loss_batch = torch.tensor(
-                    [np.float32(np.nan)], dtype=torch.float32, device=self.device
-                )
-
-                current_time = time.time()
-
-                if not self.hparams.training_spec.distributed or self.rank == 0:
-                    if self.save_latest_interval_minutes is not None and (
-                        current_time - self.last_latest_save_time
-                    ) >= (self.save_latest_interval_minutes * 60):
-                        current_time = time.time()
-                        should_save_latest[0] = 1
-                        self.last_latest_save_time = current_time
-
-                    if self.save_batch_interval_minutes is not None and (
-                        current_time - self.last_batch_save_time
-                    ) >= (self.save_batch_interval_minutes * 60):
-                        should_save_batch[0] = 1
-                        self.last_batch_save_time = current_time
-
-                if self.hparams.training_spec.distributed:
-                    dist.broadcast(should_save_latest, src=0)
-                    dist.broadcast(should_save_batch, src=0)
-                    dist.barrier()
-
-                if should_save_batch.item() == 1:
-                    if self.save_batch_interval_minutes_val_loss:
-                        val_loss, val_losses, class_counts = self._evaluate(
-                            valid_loader, ddp_model
-                        )
+                if optimizer_step_due:
+                    should_save_latest = torch.tensor(
+                        [0], dtype=torch.int32, device=self.device
+                    )
+                    should_save_batch = torch.tensor(
+                        [0], dtype=torch.int32, device=self.device
+                    )
+                    val_loss_batch = torch.tensor(
+                        [np.float32(np.nan)], dtype=torch.float32, device=self.device
+                    )
 
-                        if not self.hparams.training_spec.distributed or self.rank == 0:
-                            current_global_step = (epoch - 1) * num_batches + (
-                                batch_count + 1
+                    current_time = time.time()
+                    elapsed_since_batch_save = current_time - self.last_batch_save_time
+
+                    if not self.hparams.training_spec.distributed or self.rank == 0:
+                        if self.save_latest_interval_minutes is not None and (
+                            current_time - self.last_latest_save_time
+                        ) >= (self.save_latest_interval_minutes * 60):
+                            should_save_latest[0] = 1
+
+                        if self.save_batch_interval_minutes is not None and (
+                            elapsed_since_batch_save
+                        ) >= (self.save_batch_interval_minutes * 60):
+                            should_save_batch[0] = 1
+
+                    if self.hparams.training_spec.distributed:
+                        dist.broadcast(should_save_latest, src=0)
+                        dist.broadcast(should_save_batch, src=0)
+                        dist.barrier()
+
+                    if should_save_batch.item() == 1:
+                        if self.save_batch_interval_minutes_val_loss:
+                            val_loss, val_losses, class_counts = self._evaluate(
+                                valid_loader, ddp_model
                             )
-                            self._log_epoch_results(
-                                0,
-                                batch_count + 1,
-                                (current_time - self.last_batch_save_time),
-                                val_loss,
-                                val_losses,
-                                class_counts,
-                                current_global_step,
-                            )
-                            val_loss_batch[0] = float(val_loss)
-                        self._check_and_terminate()
-                    else:
-                        val_loss_batch[0] = np.float32(np.nan)
 
-                if self.hparams.training_spec.distributed:
-                    dist.broadcast(val_loss_batch, src=0)
-
-                if should_save_latest.item() == 1:
-                    self._save(
-                        epoch,
-                        batch_count,
-                        np.float32(np.nan),
-                        ddp_model,
-                        suffix="latest",
-                    )
-                    if self.rank != 0:
+                            if (
+                                not self.hparams.training_spec.distributed
+                                or self.rank == 0
+                            ):
+                                current_global_step = (epoch - 1) * num_batches + (
+                                    batch_count + 1
+                                )
+                                self._log_epoch_results(
+                                    0,
+                                    batch_count + 1,
+                                    elapsed_since_batch_save,
+                                    val_loss,
+                                    val_losses,
+                                    class_counts,
+                                    current_global_step,
+                                )
+                                val_loss_batch[0] = float(val_loss)
+                            self._check_and_terminate()
+                        else:
+                            val_loss_batch.fill_(torch.nan)
+
+                    if self.hparams.training_spec.distributed:
+                        dist.broadcast(val_loss_batch, src=0)
+
+                    if should_save_latest.item() == 1:
+                        self._save(
+                            epoch,
+                            batch_count,
+                            np.float32(np.nan),
+                            ddp_model,
+                            suffix="latest",
+                            best_val_loss=best_val_loss,
+                            n_epochs_no_improvement=n_epochs_no_improvement,
+                            best_model_state_dict=best_model_state,
+                            num_batches=num_batches,
+                        )
                         self.last_latest_save_time = time.time()
 
-                val_loss = np.float32(val_loss_batch.item())
-                if should_save_batch.item() != 0:
-                    self._save(
-                        epoch,
-                        batch_count,
-                        val_loss,  # type: ignore
-                        ddp_model,
-                        suffix=f"epoch-{epoch}-batch-{batch_count + 1}",
-                    )
-                    if self.rank != 0:
+                    val_loss = np.float32(val_loss_batch.item())
+                    if should_save_batch.item() != 0:
+                        self._save(
+                            epoch,
+                            batch_count,
+                            val_loss,  # type: ignore
+                            ddp_model,
+                            suffix=f"epoch-{epoch}-batch-{batch_count + 1}",
+                            best_val_loss=best_val_loss,
+                            n_epochs_no_improvement=n_epochs_no_improvement,
+                            best_model_state_dict=best_model_state,
+                            num_batches=num_batches,
+                        )
                         self.last_batch_save_time = time.time()
 
     @beartype
@@ -2305,6 +2574,10 @@ def _save(
         val_loss: np.float32,
         ddp_model: Optional[nn.Module] = None,
         suffix: Optional[str] = None,
+        best_val_loss: float = float("inf"),
+        n_epochs_no_improvement: int = 0,
+        best_model_state_dict: Optional[dict[str, Tensor]] = None,
+        num_batches: Optional[int] = None,
     ) -> None:
         """Save rank-0 checkpoint state."""
         model_to_extract = ddp_model if ddp_model is not None else self
@@ -2330,6 +2603,8 @@ def _save(
             }
             optim_state_dict = self.optimizer.state_dict()
 
+        rng_state = self._collect_rng_states_for_checkpoint()
+
         if self.rank != 0:
             return
 
@@ -2343,17 +2618,33 @@ def _save(
             file_name,
         )
 
-        torch.save(
-            {
-                "epoch": epoch,
-                "batch": batch,
-                "model_state_dict": model_state_dict,
-                "optimizer_state_dict": optim_state_dict,
-                "scheduler_state_dict": self.scheduler.state_dict(),
-                "loss": val_loss,
-            },
-            output_path,
+        checkpoint = {
+            "checkpoint_metadata": self._checkpoint_compatibility_metadata(num_batches),
+            "epoch": epoch,
+            "batch": batch,
+            "model_state_dict": model_state_dict,
+            "optimizer_state_dict": optim_state_dict,
+            "scheduler_state_dict": self.scheduler.state_dict(),
+            "scaler_state_dict": self.scaler.state_dict(),
+            "rng_state": rng_state,
+            "best_val_loss": float(best_val_loss),
+            "n_epochs_no_improvement": int(n_epochs_no_improvement),
+            "best_model_state_dict": best_model_state_dict,
+            "loss": val_loss,
+        }
+
+        temp_path = os.path.join(
+            self.project_root,
+            "checkpoints",
+            f".{file_name}.{uuid.uuid4().hex}.tmp",
         )
+        try:
+            torch.save(checkpoint, temp_path)
+            os.replace(temp_path, output_path)
+        except Exception:
+            with contextlib.suppress(OSError):
+                os.remove(temp_path)
+            raise
         self.logger.info(f"[INFO] Saved checkpoint to {output_path}")
 
     @beartype
@@ -2384,14 +2675,21 @@ def _initialize_log_file(self):
     @beartype
     def _get_latest_model_name(self) -> Optional[str]:
         """Return the newest checkpoint path for this model name."""
-        checkpoint_path = os.path.join(self.project_root, "checkpoints", "*")
+        checkpoint_path = os.path.join(
+            self.project_root, "checkpoints", f"{self.model_name}-*.pt"
+        )
+        checkpoint_name_re = re.compile(
+            rf"^{re.escape(self.model_name)}-(?:latest|epoch-\d+(?:-batch-\d+)?)\.pt$"
+        )
 
         files = glob.glob(checkpoint_path)
         files = [
-            file for file in files if os.path.split(file)[1].startswith(self.model_name)
+            file
+            for file in files
+            if checkpoint_name_re.fullmatch(os.path.split(file)[1])
         ]
         if files:
-            return max(files, key=os.path.getctime)
+            return max(files, key=os.path.getmtime)
         else:
             return None
 

From 76d6ba57093f276e1685cd8d47dcc3d46c588cb7 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 17 Jun 2026 18:26:18 +0200
Subject: [PATCH 70/81] Correct state saving and loading WIP

---
 src/sequifier/train.py | 154 ++++++++++++++++++++++++++++++++++++-----
 1 file changed, 136 insertions(+), 18 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index f335b1e1..f7dfee2a 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -9,6 +9,7 @@
 import random
 import re
 import sys
+from dataclasses import asdict
 
 os.environ["TORCH_NCCL_ASYNC_ERROR_HANDLING"] = "1"
 import time  # noqa: E402
@@ -56,6 +57,7 @@
 
 ClassCounts = dict[str, Tensor]
 CHECKPOINT_FORMAT_VERSION = 2
+SUPPORTED_CHECKPOINT_FORMAT_VERSIONS = {2}
 
 from sequifier.config.train_config import TrainModel, load_train_config  # noqa: E402
 from sequifier.distributed.env import setup_distributed_env  # noqa: E402
@@ -178,6 +180,13 @@ def train_worker(
         train_dataset = SequifierDatasetFromFile(config.training_data_path, config)
         valid_dataset = SequifierDatasetFromFile(config.validation_data_path, config)
 
+    configure_determinism(config.seed, config.training_spec.enforce_determinism)
+
+    train_loader_generator = torch.Generator()
+    train_loader_generator.manual_seed(config.seed + 10_001)
+    valid_loader_generator = torch.Generator()
+    valid_loader_generator.manual_seed(config.seed + 10_002)
+
     train_loader = DataLoader(
         train_dataset,
         batch_size=None,  # Batching is handled natively by the IterableDataset
@@ -186,6 +195,7 @@ def train_worker(
         pin_memory=config.training_spec.device not in ["mps", "cpu"],
         prefetch_factor=4 if config.training_spec.num_workers > 0 else None,
         persistent_workers=(config.training_spec.num_workers > 0),
+        generator=train_loader_generator,
     )
 
     valid_loader = DataLoader(
@@ -196,10 +206,9 @@ def train_worker(
         pin_memory=config.training_spec.device not in ["mps", "cpu"],
         prefetch_factor=4 if config.training_spec.num_workers > 0 else None,
         persistent_workers=(config.training_spec.num_workers > 0),
+        generator=valid_loader_generator,
     )
 
-    configure_determinism(config.seed, config.training_spec.enforce_determinism)
-
     model = TransformerModel(config, rank=global_rank, local_rank=local_rank)
     base_model = model
 
@@ -282,8 +291,18 @@ def train_worker(
         params_to_optimize = model.parameters()
         model.initialize_optimizer(params=params_to_optimize)
 
-        if config.training_spec.continue_training and latest_model_path:
+        resume_signal = [
+            config.training_spec.continue_training and latest_model_path is not None
+            if global_rank == 0
+            else None
+        ]
+        dist.broadcast_object_list(resume_signal, src=0)
+        did_resume = cast(bool, resume_signal[0])
+
+        if did_resume:
             if global_rank == 0:
+                if latest_model_path is None:
+                    raise RuntimeError("Rank 0 selected resume without a checkpoint.")
                 checkpoint = torch.load(
                     latest_model_path, map_location="cpu", weights_only=False
                 )
@@ -390,7 +409,7 @@ def train_worker(
 
             dist.barrier()
 
-        if checkpoint is not None:
+        if did_resume:
             base_model._restore_rng_state()
 
         model.train_model(train_loader, valid_loader, ddp_model=base_model)
@@ -857,6 +876,7 @@ def __init__(
         self._resume_n_epochs_no_improvement = 0
         self._resume_best_model_state_dict = None
         self._resume_rng_state = None
+        self._checkpoint_path_identity_cache: dict[str, dict[str, Any]] = {}
 
         self._apply_layer_dtypes()
 
@@ -1256,14 +1276,85 @@ def _check_and_terminate(self):
 
                 sys.exit(143)
 
+    @beartype
+    def _path_identity(self, path: str) -> dict[str, Any]:
+        """Return a cached, JSON-safe identity summary for a file or folder."""
+        normalized_path = normalize_path(path, self.project_root)
+        if normalized_path in self._checkpoint_path_identity_cache:
+            return self._checkpoint_path_identity_cache[normalized_path]
+
+        identity: dict[str, Any]
+        if not os.path.exists(normalized_path):
+            identity = {"path": normalized_path, "exists": False}
+        elif os.path.isfile(normalized_path):
+            stat = os.stat(normalized_path)
+            identity = {
+                "path": normalized_path,
+                "exists": True,
+                "type": "file",
+                "size": stat.st_size,
+                "mtime_ns": stat.st_mtime_ns,
+            }
+        else:
+            entries: list[dict[str, Any]] = []
+            for root, dirs, files in os.walk(normalized_path):
+                dirs.sort()
+                files.sort()
+                for file_name in files:
+                    file_path = os.path.join(root, file_name)
+                    with contextlib.suppress(OSError):
+                        stat = os.stat(file_path)
+                        entries.append(
+                            {
+                                "path": os.path.relpath(file_path, normalized_path),
+                                "size": stat.st_size,
+                                "mtime_ns": stat.st_mtime_ns,
+                            }
+                        )
+            manifest = json.dumps(entries, sort_keys=True).encode("utf-8")
+            identity = {
+                "path": normalized_path,
+                "exists": True,
+                "type": "directory",
+                "file_count": len(entries),
+                "manifest_hash": hashlib.sha256(manifest).hexdigest(),
+            }
+
+        self._checkpoint_path_identity_cache[normalized_path] = identity
+        return identity
+
     @beartype
     def _checkpoint_compatibility_metadata(
         self, num_batches: Optional[int]
     ) -> dict[str, Any]:
         """Return resume-critical settings stored with each new checkpoint."""
         training_spec = self.hparams.training_spec
+        bert_spec = (
+            training_spec.bert_spec.model_dump(mode="json")
+            if training_spec.bert_spec is not None
+            else None
+        )
         resume_settings = {
             "model_name": self.model_name,
+            "read_format": self.hparams.read_format,
+            "training_data_path": normalize_path(
+                self.hparams.training_data_path, self.project_root
+            ),
+            "validation_data_path": normalize_path(
+                self.hparams.validation_data_path, self.project_root
+            ),
+            "metadata_config_path": normalize_path(
+                self.hparams.metadata_config_path, self.project_root
+            ),
+            "training_data_identity": self._path_identity(
+                self.hparams.training_data_path
+            ),
+            "validation_data_identity": self._path_identity(
+                self.hparams.validation_data_path
+            ),
+            "metadata_config_identity": self._path_identity(
+                self.hparams.metadata_config_path
+            ),
             "num_batches": num_batches,
             "batch_size": self.batch_size,
             "accumulation_steps": self.accumulation_steps,
@@ -1279,9 +1370,29 @@ def _checkpoint_compatibility_metadata(
                 else training_spec.world_size
             ),
             "training_objective": training_spec.training_objective,
+            "seed": self.hparams.seed,
+            "dropout": training_spec.dropout,
+            "bert_spec": bert_spec,
+            "sampling_strategy": training_spec.sampling_strategy,
+            "criterion": training_spec.criterion,
+            "class_weights": training_spec.class_weights,
+            "loss_weights": training_spec.loss_weights,
+            "layer_type_dtypes": training_spec.layer_type_dtypes,
+            "layer_autocast": training_spec.layer_autocast,
+            "num_workers": training_spec.num_workers,
+            "load_full_data_to_ram": training_spec.load_full_data_to_ram,
+            "fsdp_cpu_offload": training_spec.fsdp_cpu_offload,
+            "storage_layout": asdict(self.storage_layout),
+            "window_view": asdict(self.window_view),
+            "column_types": self.hparams.column_types,
+            "categorical_columns": self.categorical_columns,
+            "real_columns": self.real_columns,
             "input_columns": self.input_columns,
             "target_columns": self.target_columns,
             "target_column_types": self.target_column_types,
+            "n_classes": self.n_classes,
+            "id_maps": self.hparams.id_maps,
+            "special_token_ids": self.hparams.special_token_ids,
             "model_spec": self.hparams.model_spec.model_dump(mode="json"),
         }
         fingerprint_input = json.dumps(
@@ -1305,26 +1416,25 @@ def _validate_checkpoint_compatibility(
                 "continuing with legacy resume behavior."
             )
             return
+        if not isinstance(checkpoint_metadata, dict):
+            raise ValueError("Checkpoint compatibility metadata must be a dictionary.")
 
         format_version = checkpoint_metadata.get("format_version")
-        if format_version != CHECKPOINT_FORMAT_VERSION:
-            self.logger.warning(
-                "[WARNING] Checkpoint format version "
-                f"{format_version!r} differs from expected "
-                f"{CHECKPOINT_FORMAT_VERSION!r}."
+        if format_version not in SUPPORTED_CHECKPOINT_FORMAT_VERSIONS:
+            raise ValueError(
+                "Unsupported checkpoint format version "
+                f"{format_version!r}; supported versions are "
+                f"{sorted(SUPPORTED_CHECKPOINT_FORMAT_VERSIONS)!r}."
             )
 
         saved_settings = checkpoint_metadata.get("resume_settings")
-        if saved_settings is None:
-            self.logger.warning(
-                "[WARNING] Checkpoint has no resume-settings fingerprint payload; "
-                "continuing without compatibility checks."
+        if not isinstance(saved_settings, dict):
+            raise ValueError(
+                "Checkpoint compatibility metadata is missing resume_settings."
             )
-            return
 
-        current_settings = self._checkpoint_compatibility_metadata(num_batches)[
-            "resume_settings"
-        ]
+        current_metadata = self._checkpoint_compatibility_metadata(num_batches)
+        current_settings = current_metadata["resume_settings"]
         mismatches = []
         for key, current_value in current_settings.items():
             saved_value = saved_settings.get(key)
@@ -1340,6 +1450,14 @@ def _validate_checkpoint_compatibility(
                 f"{mismatch_text}"
             )
 
+        saved_fingerprint = checkpoint_metadata.get("config_fingerprint")
+        current_fingerprint = current_metadata["config_fingerprint"]
+        if saved_fingerprint != current_fingerprint:
+            raise ValueError(
+                "Checkpoint configuration fingerprint mismatch: "
+                f"checkpoint={saved_fingerprint!r}, current={current_fingerprint!r}"
+            )
+
     @beartype
     def _get_rng_state(self) -> dict[str, Any]:
         """Capture Python, NumPy, Torch CPU, and CUDA RNG state for this rank."""
@@ -2676,7 +2794,7 @@ def _initialize_log_file(self):
     def _get_latest_model_name(self) -> Optional[str]:
         """Return the newest checkpoint path for this model name."""
         checkpoint_path = os.path.join(
-            self.project_root, "checkpoints", f"{self.model_name}-*.pt"
+            self.project_root, "checkpoints", f"{glob.escape(self.model_name)}-*.pt"
         )
         checkpoint_name_re = re.compile(
             rf"^{re.escape(self.model_name)}-(?:latest|epoch-\d+(?:-batch-\d+)?)\.pt$"

From 095ad4af80fcc77b9e66840ca400fe99cbd9a0ab Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Wed, 17 Jun 2026 18:49:40 +0200
Subject: [PATCH 71/81] Correct state saving and loading WIP

---
 src/sequifier/train.py | 129 +++++++++++++++++++----------------------
 1 file changed, 60 insertions(+), 69 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index f7dfee2a..c96cc857 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -210,6 +210,10 @@ def train_worker(
     )
 
     model = TransformerModel(config, rank=global_rank, local_rank=local_rank)
+    model._data_loader_generators = {
+        "train": train_loader_generator,
+        "valid": valid_loader_generator,
+    }
     base_model = model
 
     latest_model_path = model._get_latest_model_name()
@@ -238,6 +242,7 @@ def train_worker(
                 checkpoint.get("n_epochs_no_improvement", 0),
                 checkpoint.get("best_model_state_dict"),
                 checkpoint.get("rng_state"),
+                checkpoint.get("data_loader_generator_states"),
             )
         else:
             model.start_epoch = 1
@@ -255,6 +260,7 @@ def train_worker(
 
         if checkpoint is not None:
             base_model._restore_rng_state()
+            base_model._restore_data_loader_generator_states()
 
         model.train_model(train_loader, valid_loader, ddp_model=None)
     elif config.training_spec.data_parallelism == "FSDP":
@@ -320,6 +326,9 @@ def train_worker(
                     "has_best_model_state_dict": checkpoint.get("best_model_state_dict")
                     is not None,
                     "rng_state": checkpoint.get("rng_state"),
+                    "data_loader_generator_states": checkpoint.get(
+                        "data_loader_generator_states"
+                    ),
                     "checkpoint_metadata": checkpoint.get("checkpoint_metadata"),
                 }
 
@@ -386,6 +395,7 @@ def train_worker(
                 resume_state.get("n_epochs_no_improvement", 0),
                 best_model_state_dict,
                 resume_state.get("rng_state"),
+                resume_state.get("data_loader_generator_states"),
             )
 
         else:
@@ -411,6 +421,7 @@ def train_worker(
 
         if did_resume:
             base_model._restore_rng_state()
+            base_model._restore_data_loader_generator_states()
 
         model.train_model(train_loader, valid_loader, ddp_model=base_model)
         cleanup()
@@ -435,6 +446,7 @@ def train_worker(
                 checkpoint.get("n_epochs_no_improvement", 0),
                 checkpoint.get("best_model_state_dict"),
                 checkpoint.get("rng_state"),
+                checkpoint.get("data_loader_generator_states"),
             )
         else:
             model.start_epoch = 1
@@ -469,6 +481,7 @@ def train_worker(
             dist.barrier()
         if checkpoint is not None:
             base_model._restore_rng_state()
+            base_model._restore_data_loader_generator_states()
         model.train_model(train_loader, valid_loader, ddp_model=ddp_model)
         cleanup()
     else:
@@ -876,7 +889,8 @@ def __init__(
         self._resume_n_epochs_no_improvement = 0
         self._resume_best_model_state_dict = None
         self._resume_rng_state = None
-        self._checkpoint_path_identity_cache: dict[str, dict[str, Any]] = {}
+        self._resume_data_loader_generator_states = None
+        self._data_loader_generators: dict[str, torch.Generator] = {}
 
         self._apply_layer_dtypes()
 
@@ -1276,53 +1290,6 @@ def _check_and_terminate(self):
 
                 sys.exit(143)
 
-    @beartype
-    def _path_identity(self, path: str) -> dict[str, Any]:
-        """Return a cached, JSON-safe identity summary for a file or folder."""
-        normalized_path = normalize_path(path, self.project_root)
-        if normalized_path in self._checkpoint_path_identity_cache:
-            return self._checkpoint_path_identity_cache[normalized_path]
-
-        identity: dict[str, Any]
-        if not os.path.exists(normalized_path):
-            identity = {"path": normalized_path, "exists": False}
-        elif os.path.isfile(normalized_path):
-            stat = os.stat(normalized_path)
-            identity = {
-                "path": normalized_path,
-                "exists": True,
-                "type": "file",
-                "size": stat.st_size,
-                "mtime_ns": stat.st_mtime_ns,
-            }
-        else:
-            entries: list[dict[str, Any]] = []
-            for root, dirs, files in os.walk(normalized_path):
-                dirs.sort()
-                files.sort()
-                for file_name in files:
-                    file_path = os.path.join(root, file_name)
-                    with contextlib.suppress(OSError):
-                        stat = os.stat(file_path)
-                        entries.append(
-                            {
-                                "path": os.path.relpath(file_path, normalized_path),
-                                "size": stat.st_size,
-                                "mtime_ns": stat.st_mtime_ns,
-                            }
-                        )
-            manifest = json.dumps(entries, sort_keys=True).encode("utf-8")
-            identity = {
-                "path": normalized_path,
-                "exists": True,
-                "type": "directory",
-                "file_count": len(entries),
-                "manifest_hash": hashlib.sha256(manifest).hexdigest(),
-            }
-
-        self._checkpoint_path_identity_cache[normalized_path] = identity
-        return identity
-
     @beartype
     def _checkpoint_compatibility_metadata(
         self, num_batches: Optional[int]
@@ -1334,27 +1301,9 @@ def _checkpoint_compatibility_metadata(
             if training_spec.bert_spec is not None
             else None
         )
-        resume_settings = {
+        compatibility_settings = {
             "model_name": self.model_name,
             "read_format": self.hparams.read_format,
-            "training_data_path": normalize_path(
-                self.hparams.training_data_path, self.project_root
-            ),
-            "validation_data_path": normalize_path(
-                self.hparams.validation_data_path, self.project_root
-            ),
-            "metadata_config_path": normalize_path(
-                self.hparams.metadata_config_path, self.project_root
-            ),
-            "training_data_identity": self._path_identity(
-                self.hparams.training_data_path
-            ),
-            "validation_data_identity": self._path_identity(
-                self.hparams.validation_data_path
-            ),
-            "metadata_config_identity": self._path_identity(
-                self.hparams.metadata_config_path
-            ),
             "num_batches": num_batches,
             "batch_size": self.batch_size,
             "accumulation_steps": self.accumulation_steps,
@@ -1395,13 +1344,25 @@ def _checkpoint_compatibility_metadata(
             "special_token_ids": self.hparams.special_token_ids,
             "model_spec": self.hparams.model_spec.model_dump(mode="json"),
         }
+        provenance = {
+            "training_data_path": normalize_path(
+                self.hparams.training_data_path, self.project_root
+            ),
+            "validation_data_path": normalize_path(
+                self.hparams.validation_data_path, self.project_root
+            ),
+            "metadata_config_path": normalize_path(
+                self.hparams.metadata_config_path, self.project_root
+            ),
+        }
         fingerprint_input = json.dumps(
-            resume_settings, sort_keys=True, default=str
+            compatibility_settings, sort_keys=True, default=str
         ).encode("utf-8")
         return {
             "format_version": CHECKPOINT_FORMAT_VERSION,
             "config_fingerprint": hashlib.sha256(fingerprint_input).hexdigest(),
-            "resume_settings": resume_settings,
+            "resume_settings": compatibility_settings,
+            "provenance": provenance,
         }
 
     @beartype
@@ -1509,6 +1470,32 @@ def _select_rng_state_for_rank(self, rng_states: Any) -> Optional[dict[str, Any]
         )
         return rng_states[0]
 
+    @beartype
+    def _get_data_loader_generator_states(self) -> dict[str, Tensor]:
+        """Capture dedicated DataLoader generator states."""
+        return {
+            name: generator.get_state()
+            for name, generator in self._data_loader_generators.items()
+        }
+
+    @beartype
+    def _restore_data_loader_generator_states(self) -> None:
+        """Restore dedicated DataLoader generator states when present."""
+        states = self._resume_data_loader_generator_states
+        if states is None:
+            return
+        if not isinstance(states, dict):
+            self.logger.warning(
+                "[WARNING] Checkpoint DataLoader generator state is not a dictionary; "
+                "using freshly seeded DataLoader generators."
+            )
+            return
+
+        for name, generator in self._data_loader_generators.items():
+            state = states.get(name)
+            if isinstance(state, Tensor):
+                generator.set_state(state)
+
     @beartype
     def _apply_checkpoint_training_state(
         self,
@@ -1517,6 +1504,7 @@ def _apply_checkpoint_training_state(
         n_epochs_no_improvement: Any,
         best_model_state_dict: Any,
         rng_states: Any,
+        data_loader_generator_states: Any,
     ) -> None:
         """Restore non-model training state from a checkpoint payload."""
         if scaler_state_dict is not None:
@@ -1531,6 +1519,7 @@ def _apply_checkpoint_training_state(
         self._resume_n_epochs_no_improvement = int(n_epochs_no_improvement)
         self._resume_best_model_state_dict = best_model_state_dict
         self._resume_rng_state = self._select_rng_state_for_rank(rng_states)
+        self._resume_data_loader_generator_states = data_loader_generator_states
 
     @beartype
     def _restore_rng_state(self) -> None:
@@ -2722,6 +2711,7 @@ def _save(
             optim_state_dict = self.optimizer.state_dict()
 
         rng_state = self._collect_rng_states_for_checkpoint()
+        data_loader_generator_states = self._get_data_loader_generator_states()
 
         if self.rank != 0:
             return
@@ -2745,6 +2735,7 @@ def _save(
             "scheduler_state_dict": self.scheduler.state_dict(),
             "scaler_state_dict": self.scaler.state_dict(),
             "rng_state": rng_state,
+            "data_loader_generator_states": data_loader_generator_states,
             "best_val_loss": float(best_val_loss),
             "n_epochs_no_improvement": int(n_epochs_no_improvement),
             "best_model_state_dict": best_model_state_dict,

From 79b57ed982b8e61f8c8a3961e1f52bd71b8b1f65 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 14:19:59 +0200
Subject: [PATCH 72/81] Refactor checkpoint metadata management

---
 src/sequifier/io/iteration_state.py           | 80 ++++++++++++++++++
 .../io/sequifier_dataset_from_file.py         | 39 +++++++--
 .../sequifier_dataset_from_folder_parquet.py  | 38 ++++++++-
 ...uifier_dataset_from_folder_parquet_lazy.py | 51 +++++++++---
 .../io/sequifier_dataset_from_folder_pt.py    | 38 ++++++++-
 .../sequifier_dataset_from_folder_pt_lazy.py  | 51 +++++++++---
 src/sequifier/train.py                        | 81 ++++++++++++++++---
 7 files changed, 338 insertions(+), 40 deletions(-)
 create mode 100644 src/sequifier/io/iteration_state.py

diff --git a/src/sequifier/io/iteration_state.py b/src/sequifier/io/iteration_state.py
new file mode 100644
index 00000000..d6f80aae
--- /dev/null
+++ b/src/sequifier/io/iteration_state.py
@@ -0,0 +1,80 @@
+from collections.abc import Sequence
+
+import torch
+
+
+def shared_int(value: int = 0) -> torch.Tensor:
+    """Return a shared scalar int tensor visible to DataLoader worker copies."""
+    state = torch.empty((), dtype=torch.int64)
+    state.fill_(int(value))
+    state.share_memory_()
+    return state
+
+
+def read_shared_int(state: torch.Tensor) -> int:
+    return int(state.item())
+
+
+def write_shared_int(state: torch.Tensor, value: int) -> None:
+    state.fill_(int(value))
+
+
+def resolve_resume_worker(
+    start_batch: int,
+    worker_id: int,
+    num_workers: int,
+    worker_batch_counts: Sequence[int],
+) -> tuple[int, int]:
+    """Map a physical worker to the logical worker/batch offset after resume.
+
+    PyTorch requests IterableDataset samples from workers in worker-id order. To
+    resume at a non-zero global batch without yielding skipped batches, worker 0
+    is rotated onto the worker that would have produced start_batch in the
+    uninterrupted iterator. Each logical worker then skips only the batches it
+    owned before start_batch.
+    """
+    if num_workers <= 0:
+        raise ValueError("num_workers must be positive.")
+    if len(worker_batch_counts) != num_workers:
+        raise ValueError("worker_batch_counts must match num_workers.")
+
+    counts = [max(0, int(count)) for count in worker_batch_counts]
+    total_batches = sum(counts)
+    capped_start = min(max(0, int(start_batch)), total_batches)
+    if capped_start == 0 or total_batches == 0:
+        return worker_id, 0
+
+    skips = [0] * num_workers
+    active_workers = [i for i, count in enumerate(counts) if count > 0]
+    consumed = 0
+    start_worker = active_workers[0] if active_workers else 0
+
+    while active_workers and consumed < capped_start:
+        rounds_until_next_exhaustion = min(counts[i] - skips[i] for i in active_workers)
+        block_batches = rounds_until_next_exhaustion * len(active_workers)
+        remaining = capped_start - consumed
+
+        if remaining >= block_batches:
+            for i in active_workers:
+                skips[i] += rounds_until_next_exhaustion
+            consumed += block_batches
+            active_workers = [i for i in active_workers if skips[i] < counts[i]]
+            start_worker = active_workers[0] if active_workers else 0
+            continue
+
+        full_rounds, offset = divmod(remaining, len(active_workers))
+        for i in active_workers:
+            skips[i] += full_rounds
+        for i in active_workers[:offset]:
+            skips[i] += 1
+        start_worker = active_workers[offset]
+        consumed = capped_start
+
+    logical_worker_id = (worker_id + start_worker) % num_workers
+    return logical_worker_id, skips[logical_worker_id]
+
+
+def skip_samples_for_batches(
+    skip_batches: int, batch_size: int, total_samples: int
+) -> int:
+    return min(max(0, int(skip_batches)) * int(batch_size), max(0, int(total_samples)))
diff --git a/src/sequifier/io/sequifier_dataset_from_file.py b/src/sequifier/io/sequifier_dataset_from_file.py
index da4bc2b0..c96358a3 100644
--- a/src/sequifier/io/sequifier_dataset_from_file.py
+++ b/src/sequifier/io/sequifier_dataset_from_file.py
@@ -14,6 +14,13 @@
     resolve_window_view,
 )
 from sequifier.io.batch import SequifierBatch
+from sequifier.io.iteration_state import (
+    read_shared_int,
+    resolve_resume_worker,
+    shared_int,
+    skip_samples_for_batches,
+    write_shared_int,
+)
 
 
 class SequifierDatasetFromFile(IterableDataset):
@@ -24,7 +31,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.config = config
         self.batch_size = config.training_spec.batch_size
         self.shuffle = shuffle
-        self.epoch = 0
+        self._epoch_state = shared_int(0)
+        self._start_batch_state = shared_int(0)
 
         all_columns = sorted(list(set(config.input_columns + config.target_columns)))
 
@@ -70,7 +78,11 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
 
     def set_epoch(self, epoch: int):
         """Set the shuffle epoch."""
-        self.epoch = epoch
+        write_shared_int(self._epoch_state, epoch)
+
+    def set_start_batch(self, start_batch: int):
+        """Set the first global batch to yield on the next iteration."""
+        write_shared_int(self._start_batch_state, start_batch)
 
     def __len__(self) -> int:
         return math.ceil(self.n_samples / self.batch_size)
@@ -84,22 +96,39 @@ def __iter__(
 
         if worker_info is None:
             # Single-process data loading
-            worker_id = 0
+            physical_worker_id = 0
             num_workers = 1
         else:
             # Multi-process data loading
-            worker_id = worker_info.id
+            physical_worker_id = worker_info.id
             num_workers = worker_info.num_workers
+        epoch = read_shared_int(self._epoch_state)
+        start_batch = read_shared_int(self._start_batch_state)
 
         indices = torch.arange(self.n_samples)
         if self.shuffle:
             g = torch.Generator()
             # Use epoch and seed for a different but deterministic shuffle each epoch
-            g.manual_seed(self.config.seed + self.epoch)
+            g.manual_seed(self.config.seed + epoch)
             indices = indices[torch.randperm(self.n_samples, generator=g)]
 
         indices_for_rank = indices[rank::world_size]
+        worker_batch_counts = [
+            len(indices_for_rank[i::num_workers]) // self.batch_size
+            for i in range(num_workers)
+        ]
+        worker_id, skip_batches = resolve_resume_worker(
+            start_batch,
+            physical_worker_id,
+            num_workers,
+            worker_batch_counts,
+        )
+
         indices_for_worker = indices_for_rank[worker_id::num_workers]
+        skipped_samples = skip_samples_for_batches(
+            skip_batches, self.batch_size, len(indices_for_worker)
+        )
+        indices_for_worker = indices_for_worker[skipped_samples:]
 
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_end = i + self.batch_size
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
index 93035822..9e4acca7 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet.py
@@ -19,6 +19,13 @@
     stored_window_layout_from_metadata,
 )
 from sequifier.io.batch import SequifierBatch
+from sequifier.io.iteration_state import (
+    read_shared_int,
+    resolve_resume_worker,
+    shared_int,
+    skip_samples_for_batches,
+    write_shared_int,
+)
 
 
 class SequifierDatasetFromFolderParquet(IterableDataset):
@@ -30,7 +37,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.config = config
         self.batch_size = config.training_spec.batch_size
         self.shuffle = shuffle
-        self.epoch = 0
+        self._epoch_state = shared_int(0)
+        self._start_batch_state = shared_int(0)
         self.sampling_strategy = config.training_spec.sampling_strategy
 
         metadata_path = os.path.join(self.data_dir, "metadata.json")
@@ -142,7 +150,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
 
     def set_epoch(self, epoch: int):
         """Set the shuffle epoch."""
-        self.epoch = epoch
+        write_shared_int(self._epoch_state, epoch)
+
+    def set_start_batch(self, start_batch: int):
+        """Set the first global batch to yield on the next iteration."""
+        write_shared_int(self._start_batch_state, start_batch)
 
     def _get_target_samples(self) -> int:
         """Return per-rank sample count under the configured sampling strategy."""
@@ -171,13 +183,15 @@ def __iter__(
         rank = dist.get_rank() if dist.is_initialized() else 0
 
         worker_info = get_worker_info()
-        worker_id = worker_info.id if worker_info is not None else 0
+        physical_worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
+        epoch = read_shared_int(self._epoch_state)
+        start_batch = read_shared_int(self._start_batch_state)
 
         indices = torch.arange(self.n_samples)
         if self.shuffle:
             g = torch.Generator()
-            g.manual_seed(self.config.seed + self.epoch)
+            g.manual_seed(self.config.seed + epoch)
             indices = indices[torch.randperm(self.n_samples, generator=g)]
 
         indices_for_rank = indices[rank::world_size].tolist()
@@ -202,8 +216,24 @@ def __iter__(
             indices_for_rank = indices_for_rank[: self.target_samples]
             sample_is_real = sample_is_real[: self.target_samples]
 
+        worker_batch_counts = [
+            math.ceil(len(indices_for_rank[i::num_workers]) / self.batch_size)
+            for i in range(num_workers)
+        ]
+        worker_id, skip_batches = resolve_resume_worker(
+            start_batch,
+            physical_worker_id,
+            num_workers,
+            worker_batch_counts,
+        )
+
         indices_for_worker = indices_for_rank[worker_id::num_workers]
         sample_is_real_for_worker = sample_is_real[worker_id::num_workers]
+        skipped_samples = skip_samples_for_batches(
+            skip_batches, self.batch_size, len(indices_for_worker)
+        )
+        indices_for_worker = indices_for_worker[skipped_samples:]
+        sample_is_real_for_worker = sample_is_real_for_worker[skipped_samples:]
 
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
index b45ed0cc..df1a71d1 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_parquet_lazy.py
@@ -20,6 +20,13 @@
     stored_window_layout_from_metadata,
 )
 from sequifier.io.batch import SequifierBatch
+from sequifier.io.iteration_state import (
+    read_shared_int,
+    resolve_resume_worker,
+    shared_int,
+    skip_samples_for_batches,
+    write_shared_int,
+)
 
 
 class SequifierDatasetFromFolderParquetLazy(IterableDataset):
@@ -31,7 +38,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.config = config
         self.batch_size = config.training_spec.batch_size
         self.shuffle = shuffle
-        self.epoch = 0
+        self._epoch_state = shared_int(0)
+        self._start_batch_state = shared_int(0)
 
         metadata_path = os.path.join(self.data_dir, "metadata.json")
         if not os.path.exists(metadata_path):
@@ -75,7 +83,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
 
     def set_epoch(self, epoch: int):
         """Set the shuffle epoch."""
-        self.epoch = epoch
+        write_shared_int(self._epoch_state, epoch)
+
+    def set_start_batch(self, start_batch: int):
+        """Set the first global batch to yield on the next iteration."""
+        write_shared_int(self._start_batch_state, start_batch)
 
     def _get_target_samples(self) -> int:
         """Return per-rank sample count under the configured sampling strategy."""
@@ -144,8 +156,10 @@ def __iter__(
         rank = dist.get_rank() if dist.is_initialized() else 0
 
         worker_info = get_worker_info()
-        worker_id = worker_info.id if worker_info is not None else 0
+        physical_worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
+        epoch = read_shared_int(self._epoch_state)
+        start_batch = read_shared_int(self._start_batch_state)
 
         num_files = len(self.batch_files_info)
         original_files_for_this_rank = list(range(rank, num_files, world_size))
@@ -162,18 +176,37 @@ def __iter__(
 
         base_samples_per_worker = self.target_samples // num_workers
         remainder = self.target_samples % num_workers
+        worker_sample_counts = [
+            base_samples_per_worker + (1 if i < remainder else 0)
+            for i in range(num_workers)
+        ]
+        worker_batch_counts = [
+            math.ceil(sample_count / self.batch_size)
+            for sample_count in worker_sample_counts
+        ]
+        worker_id, skip_batches = resolve_resume_worker(
+            start_batch,
+            physical_worker_id,
+            num_workers,
+            worker_batch_counts,
+        )
 
         worker_start_sample = 0
         for i in range(worker_id):
-            worker_start_sample += base_samples_per_worker + (1 if i < remainder else 0)
+            worker_start_sample += worker_sample_counts[i]
 
-        worker_target_samples = base_samples_per_worker + (
-            1 if worker_id < remainder else 0
-        )
+        worker_target_samples = worker_sample_counts[worker_id]
         worker_end_sample = worker_start_sample + worker_target_samples
+        skipped_samples = skip_samples_for_batches(
+            skip_batches, self.batch_size, worker_target_samples
+        )
+        worker_start_sample += skipped_samples
+        worker_target_samples -= skipped_samples
+        if worker_target_samples <= 0:
+            return
 
         g = torch.Generator()
-        g.manual_seed(self.config.seed + self.epoch)
+        g.manual_seed(self.config.seed + epoch)
 
         if self.shuffle:
             file_order = torch.randperm(len(files_for_this_rank), generator=g).tolist()
@@ -224,7 +257,7 @@ def __iter__(
             indices = torch.arange(file_samples)
             if self.shuffle:
                 g_file = torch.Generator()
-                g_file.manual_seed(self.config.seed + self.epoch + f_id + rank)
+                g_file.manual_seed(self.config.seed + epoch + f_id + rank)
                 indices = indices[torch.randperm(file_samples, generator=g_file)]
 
             worker_file_start_idx = max(0, worker_start_sample - file_start)
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt.py b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
index 7327fe93..469b044e 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt.py
@@ -16,6 +16,13 @@
     validate_stored_window_width,
 )
 from sequifier.io.batch import SequifierBatch
+from sequifier.io.iteration_state import (
+    read_shared_int,
+    resolve_resume_worker,
+    shared_int,
+    skip_samples_for_batches,
+    write_shared_int,
+)
 
 
 class SequifierDatasetFromFolderPt(IterableDataset):
@@ -27,7 +34,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.config = config
         self.batch_size = config.training_spec.batch_size
         self.shuffle = shuffle
-        self.epoch = 0
+        self._epoch_state = shared_int(0)
+        self._start_batch_state = shared_int(0)
         self.sampling_strategy = config.training_spec.sampling_strategy
 
         metadata_path = os.path.join(self.data_dir, "metadata.json")
@@ -101,7 +109,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
 
     def set_epoch(self, epoch: int):
         """Set the shuffle epoch."""
-        self.epoch = epoch
+        write_shared_int(self._epoch_state, epoch)
+
+    def set_start_batch(self, start_batch: int):
+        """Set the first global batch to yield on the next iteration."""
+        write_shared_int(self._start_batch_state, start_batch)
 
     def _get_target_samples(self) -> int:
         """Return per-rank sample count under the configured sampling strategy."""
@@ -130,13 +142,15 @@ def __iter__(
         rank = dist.get_rank() if dist.is_initialized() else 0
 
         worker_info = get_worker_info()
-        worker_id = worker_info.id if worker_info is not None else 0
+        physical_worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
+        epoch = read_shared_int(self._epoch_state)
+        start_batch = read_shared_int(self._start_batch_state)
 
         indices = torch.arange(self.n_samples)
         if self.shuffle:
             g = torch.Generator()
-            g.manual_seed(self.config.seed + self.epoch)
+            g.manual_seed(self.config.seed + epoch)
             indices = indices[torch.randperm(self.n_samples, generator=g)]
 
         indices_for_rank = indices[rank::world_size].tolist()
@@ -161,8 +175,24 @@ def __iter__(
             indices_for_rank = indices_for_rank[: self.target_samples]
             sample_is_real = sample_is_real[: self.target_samples]
 
+        worker_batch_counts = [
+            math.ceil(len(indices_for_rank[i::num_workers]) / self.batch_size)
+            for i in range(num_workers)
+        ]
+        worker_id, skip_batches = resolve_resume_worker(
+            start_batch,
+            physical_worker_id,
+            num_workers,
+            worker_batch_counts,
+        )
+
         indices_for_worker = indices_for_rank[worker_id::num_workers]
         sample_is_real_for_worker = sample_is_real[worker_id::num_workers]
+        skipped_samples = skip_samples_for_batches(
+            skip_batches, self.batch_size, len(indices_for_worker)
+        )
+        indices_for_worker = indices_for_worker[skipped_samples:]
+        sample_is_real_for_worker = sample_is_real_for_worker[skipped_samples:]
 
         for i in range(0, len(indices_for_worker), self.batch_size):
             batch_indices = indices_for_worker[i : i + self.batch_size]
diff --git a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
index 506d12b7..6754f3ec 100644
--- a/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
+++ b/src/sequifier/io/sequifier_dataset_from_folder_pt_lazy.py
@@ -17,6 +17,13 @@
     validate_stored_window_width,
 )
 from sequifier.io.batch import SequifierBatch
+from sequifier.io.iteration_state import (
+    read_shared_int,
+    resolve_resume_worker,
+    shared_int,
+    skip_samples_for_batches,
+    write_shared_int,
+)
 
 
 class SequifierDatasetFromFolderPtLazy(IterableDataset):
@@ -28,7 +35,8 @@ def __init__(self, data_path: str, config: TrainModel, shuffle: bool = True):
         self.config = config
         self.batch_size = config.training_spec.batch_size
         self.shuffle = shuffle
-        self.epoch = 0
+        self._epoch_state = shared_int(0)
+        self._start_batch_state = shared_int(0)
 
         metadata_path = os.path.join(self.data_dir, "metadata.json")
         if not os.path.exists(metadata_path):
@@ -67,7 +75,11 @@ def _calculate_total_batches(self, target_samples: int) -> int:
 
     def set_epoch(self, epoch: int):
         """Set the shuffle epoch."""
-        self.epoch = epoch
+        write_shared_int(self._epoch_state, epoch)
+
+    def set_start_batch(self, start_batch: int):
+        """Set the first global batch to yield on the next iteration."""
+        write_shared_int(self._start_batch_state, start_batch)
 
     def _get_target_samples(self) -> int:
         """Return per-rank sample count under the configured sampling strategy."""
@@ -136,8 +148,10 @@ def __iter__(
         rank = dist.get_rank() if dist.is_initialized() else 0
 
         worker_info = get_worker_info()
-        worker_id = worker_info.id if worker_info is not None else 0
+        physical_worker_id = worker_info.id if worker_info is not None else 0
         num_workers = worker_info.num_workers if worker_info is not None else 1
+        epoch = read_shared_int(self._epoch_state)
+        start_batch = read_shared_int(self._start_batch_state)
 
         num_files = len(self.batch_files_info)
         original_files_for_this_rank = list(range(rank, num_files, world_size))
@@ -154,18 +168,37 @@ def __iter__(
 
         base_samples_per_worker = self.target_samples // num_workers
         remainder = self.target_samples % num_workers
+        worker_sample_counts = [
+            base_samples_per_worker + (1 if i < remainder else 0)
+            for i in range(num_workers)
+        ]
+        worker_batch_counts = [
+            math.ceil(sample_count / self.batch_size)
+            for sample_count in worker_sample_counts
+        ]
+        worker_id, skip_batches = resolve_resume_worker(
+            start_batch,
+            physical_worker_id,
+            num_workers,
+            worker_batch_counts,
+        )
 
         worker_start_sample = 0
         for i in range(worker_id):
-            worker_start_sample += base_samples_per_worker + (1 if i < remainder else 0)
+            worker_start_sample += worker_sample_counts[i]
 
-        worker_target_samples = base_samples_per_worker + (
-            1 if worker_id < remainder else 0
-        )
+        worker_target_samples = worker_sample_counts[worker_id]
         worker_end_sample = worker_start_sample + worker_target_samples
+        skipped_samples = skip_samples_for_batches(
+            skip_batches, self.batch_size, worker_target_samples
+        )
+        worker_start_sample += skipped_samples
+        worker_target_samples -= skipped_samples
+        if worker_target_samples <= 0:
+            return
 
         g = torch.Generator()
-        g.manual_seed(self.config.seed + self.epoch)
+        g.manual_seed(self.config.seed + epoch)
 
         if self.shuffle:
             file_order = torch.randperm(len(files_for_this_rank), generator=g).tolist()
@@ -218,7 +251,7 @@ def __iter__(
             indices = torch.arange(file_samples)
             if self.shuffle:
                 g_file = torch.Generator()
-                g_file.manual_seed(self.config.seed + self.epoch + f_id + rank)
+                g_file.manual_seed(self.config.seed + epoch + f_id + rank)
                 indices = indices[torch.randperm(file_samples, generator=g_file)]
 
             worker_file_start_idx = max(0, worker_start_sample - file_start)
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index c96cc857..9f65c4ce 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -230,9 +230,13 @@ def train_worker(
                 latest_model_path, map_location="cpu", weights_only=False
             )
             model._validate_checkpoint_compatibility(checkpoint, len(train_loader))
-            model.load_state_dict(checkpoint["model_state_dict"])
-            model.optimizer.load_state_dict(checkpoint["optimizer_state_dict"])
-            model.scheduler.load_state_dict(checkpoint["scheduler_state_dict"])
+            model.load_state_dict(copy.deepcopy(checkpoint["model_state_dict"]))
+            model.optimizer.load_state_dict(
+                copy.deepcopy(checkpoint["optimizer_state_dict"])
+            )
+            model.scheduler.load_state_dict(
+                copy.deepcopy(checkpoint["scheduler_state_dict"])
+            )
             base_model.start_epoch, base_model.start_batch = _checkpoint_start_position(
                 checkpoint, len(train_loader)
             )
@@ -377,7 +381,7 @@ def train_worker(
             set_optimizer_state_dict(
                 base_model,
                 base_model.optimizer,
-                optim_state_dict=full_osd,
+                optim_state_dict=copy.deepcopy(full_osd),
                 options=options,
             )
 
@@ -598,6 +602,46 @@ def _checkpoint_start_position(
     return checkpoint["epoch"], checkpoint["batch"] + 1
 
 
+def _update_file_metadata_hash(hasher: Any, file_path: str) -> None:
+    """Hash file identity metadata without reading the file contents."""
+    normalized_path = os.path.abspath(file_path)
+    file_stat = os.stat(normalized_path)
+    hasher.update(normalized_path.encode("utf-8"))
+    hasher.update(str(file_stat.st_size).encode("utf-8"))
+    hasher.update(str(file_stat.st_mtime_ns).encode("utf-8"))
+
+
+def _update_folder_dataset_metadata_hash(hasher: Any, data_dir: str) -> None:
+    """Hash folder dataset metadata and chunk file metadata."""
+    metadata_path = os.path.join(data_dir, "metadata.json")
+    _update_file_metadata_hash(hasher, metadata_path)
+
+    with open(metadata_path, "r") as f:
+        metadata = json.load(f)
+
+    batch_files_info = metadata.get("batch_files")
+    if not isinstance(batch_files_info, list):
+        raise ValueError(f"metadata.json in {data_dir!r} is missing batch_files.")
+
+    for file_info in batch_files_info:
+        relative_path = file_info["path"]
+        file_path = os.path.join(data_dir, relative_path)
+        hasher.update(str(relative_path).encode("utf-8"))
+        hasher.update(str(file_info.get("samples", "")).encode("utf-8"))
+        _update_file_metadata_hash(hasher, file_path)
+
+
+def _dataset_metadata_fingerprint(data_path: str, config: TrainModel) -> str:
+    """Return a fast provenance fingerprint from filesystem metadata."""
+    normalized_path = normalize_path(data_path, config.project_root)
+    hasher = hashlib.sha256()
+    if os.path.isdir(normalized_path):
+        _update_folder_dataset_metadata_hash(hasher, normalized_path)
+    else:
+        _update_file_metadata_hash(hasher, normalized_path)
+    return hasher.hexdigest()
+
+
 @beartype
 def accumulate_class_counts(
     counts: ClassCounts,
@@ -1342,6 +1386,12 @@ def _checkpoint_compatibility_metadata(
             "n_classes": self.n_classes,
             "id_maps": self.hparams.id_maps,
             "special_token_ids": self.hparams.special_token_ids,
+            "training_data_fingerprint": _dataset_metadata_fingerprint(
+                self.hparams.training_data_path, self.hparams
+            ),
+            "validation_data_fingerprint": _dataset_metadata_fingerprint(
+                self.hparams.validation_data_path, self.hparams
+            ),
             "model_spec": self.hparams.model_spec.model_dump(mode="json"),
         }
         provenance = {
@@ -1406,15 +1456,16 @@ def _validate_checkpoint_compatibility(
 
         if mismatches:
             mismatch_text = "; ".join(mismatches)
-            raise ValueError(
-                "Checkpoint is incompatible with the current training configuration: "
+            warnings.warn(
+                "Checkpoint is not identical with the current training configuration. "
+                "Ensure that this is the intended configuration. "
                 f"{mismatch_text}"
             )
 
         saved_fingerprint = checkpoint_metadata.get("config_fingerprint")
         current_fingerprint = current_metadata["config_fingerprint"]
         if saved_fingerprint != current_fingerprint:
-            raise ValueError(
+            warnings.warn(
                 "Checkpoint configuration fingerprint mismatch: "
                 f"checkpoint={saved_fingerprint!r}, current={current_fingerprint!r}"
             )
@@ -1739,17 +1790,26 @@ def _train_epoch(
         num_batches = len(train_loader)
         start_batch = self.start_batch
         self.start_batch = 0
+        set_dataset_start_batch = getattr(train_loader.dataset, "set_start_batch", None)
+        dataset_handles_start_batch = callable(set_dataset_start_batch)
+        if dataset_handles_start_batch:
+            set_dataset_start_batch(start_batch)
 
         model_to_call = ddp_model if ddp_model is not None else self
 
         model_to_call.train()
 
-        for batch_count, batch in enumerate(train_loader):
+        for batch_offset, batch in enumerate(train_loader):
             if not isinstance(batch, SequifierBatch):
                 raise TypeError(
                     "Training DataLoader must yield SequifierBatch objects, "
                     f"got {type(batch).__name__}."
                 )
+            batch_count = (
+                start_batch + batch_offset
+                if dataset_handles_start_batch
+                else batch_offset
+            )
             if batch_count >= start_batch:
                 data = batch.inputs
                 targets = batch.targets
@@ -1985,6 +2045,9 @@ def _train_epoch(
                         )
                         self.last_batch_save_time = time.time()
 
+        if dataset_handles_start_batch:
+            set_dataset_start_batch(0)
+
     @beartype
     def _calculate_loss(
         self,
@@ -2708,7 +2771,7 @@ def _save(
             model_state_dict = {
                 k.replace("_orig_mod.", ""): v for k, v in self.state_dict().items()
             }
-            optim_state_dict = self.optimizer.state_dict()
+            optim_state_dict = copy.deepcopy(self.optimizer.state_dict())
 
         rng_state = self._collect_rng_states_for_checkpoint()
         data_loader_generator_states = self._get_data_loader_generator_states()

From 69024a18357d19e6451706faa9304c70212e5b36 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 14:31:05 +0200
Subject: [PATCH 73/81] WIP

---
 src/sequifier/train.py | 50 +++++-------------------------------------
 1 file changed, 5 insertions(+), 45 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 9f65c4ce..92cc7c56 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -230,13 +230,9 @@ def train_worker(
                 latest_model_path, map_location="cpu", weights_only=False
             )
             model._validate_checkpoint_compatibility(checkpoint, len(train_loader))
-            model.load_state_dict(copy.deepcopy(checkpoint["model_state_dict"]))
-            model.optimizer.load_state_dict(
-                copy.deepcopy(checkpoint["optimizer_state_dict"])
-            )
-            model.scheduler.load_state_dict(
-                copy.deepcopy(checkpoint["scheduler_state_dict"])
-            )
+            model.load_state_dict(checkpoint["model_state_dict"])
+            model.optimizer.load_state_dict(checkpoint["optimizer_state_dict"])
+            model.scheduler.load_state_dict(checkpoint["scheduler_state_dict"])
             base_model.start_epoch, base_model.start_batch = _checkpoint_start_position(
                 checkpoint, len(train_loader)
             )
@@ -381,7 +377,7 @@ def train_worker(
             set_optimizer_state_dict(
                 base_model,
                 base_model.optimizer,
-                optim_state_dict=copy.deepcopy(full_osd),
+                optim_state_dict=full_osd,
                 options=options,
             )
 
@@ -611,37 +607,6 @@ def _update_file_metadata_hash(hasher: Any, file_path: str) -> None:
     hasher.update(str(file_stat.st_mtime_ns).encode("utf-8"))
 
 
-def _update_folder_dataset_metadata_hash(hasher: Any, data_dir: str) -> None:
-    """Hash folder dataset metadata and chunk file metadata."""
-    metadata_path = os.path.join(data_dir, "metadata.json")
-    _update_file_metadata_hash(hasher, metadata_path)
-
-    with open(metadata_path, "r") as f:
-        metadata = json.load(f)
-
-    batch_files_info = metadata.get("batch_files")
-    if not isinstance(batch_files_info, list):
-        raise ValueError(f"metadata.json in {data_dir!r} is missing batch_files.")
-
-    for file_info in batch_files_info:
-        relative_path = file_info["path"]
-        file_path = os.path.join(data_dir, relative_path)
-        hasher.update(str(relative_path).encode("utf-8"))
-        hasher.update(str(file_info.get("samples", "")).encode("utf-8"))
-        _update_file_metadata_hash(hasher, file_path)
-
-
-def _dataset_metadata_fingerprint(data_path: str, config: TrainModel) -> str:
-    """Return a fast provenance fingerprint from filesystem metadata."""
-    normalized_path = normalize_path(data_path, config.project_root)
-    hasher = hashlib.sha256()
-    if os.path.isdir(normalized_path):
-        _update_folder_dataset_metadata_hash(hasher, normalized_path)
-    else:
-        _update_file_metadata_hash(hasher, normalized_path)
-    return hasher.hexdigest()
-
-
 @beartype
 def accumulate_class_counts(
     counts: ClassCounts,
@@ -1386,12 +1351,6 @@ def _checkpoint_compatibility_metadata(
             "n_classes": self.n_classes,
             "id_maps": self.hparams.id_maps,
             "special_token_ids": self.hparams.special_token_ids,
-            "training_data_fingerprint": _dataset_metadata_fingerprint(
-                self.hparams.training_data_path, self.hparams
-            ),
-            "validation_data_fingerprint": _dataset_metadata_fingerprint(
-                self.hparams.validation_data_path, self.hparams
-            ),
             "model_spec": self.hparams.model_spec.model_dump(mode="json"),
         }
         provenance = {
@@ -1896,6 +1855,7 @@ def _train_epoch(
                     or (batch_count + 1) == num_batches
                 )
                 optimizer_step_performed = False
+
                 if (
                     optimizer_step_due
                     and accumulated_global_token_count.detach().cpu().item() > 0

From 49096fbd7baa8ce1f1d0ee8b45a6388dc258a6f1 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 14:32:50 +0200
Subject: [PATCH 74/81] Rename save_batch_interval_minutes to
 save_interval_minutes

---
 documentation/configs/hyperparameter-search.md     |  4 ++--
 documentation/configs/train.md                     |  4 ++--
 documentation/consolidated-docs.md                 |  8 ++++----
 .../config/hyperparameter_search_config.py         |  8 ++++----
 src/sequifier/config/train_config.py               | 10 +++++-----
 src/sequifier/train.py                             | 14 ++++++--------
 6 files changed, 23 insertions(+), 25 deletions(-)

diff --git a/documentation/configs/hyperparameter-search.md b/documentation/configs/hyperparameter-search.md
index 23402080..fa147d06 100644
--- a/documentation/configs/hyperparameter-search.md
+++ b/documentation/configs/hyperparameter-search.md
@@ -134,8 +134,8 @@ Most fields here are lists for sampling, but some are scalar values fixed for al
 | `scheduler` | `list[dict]` | No | `[{'name': 'StepLR'...}]`| List of scheduler configs. `scheduler.step()` is only called if \< total\_steps, so correct configuration is essential. |
 | `scheduler_step_on` | `str` | No | `epoch` | When to step the scheduler: `epoch` or `batch`. |
 | `save_latest_interval_minutes`| `float`| No | `null` | Time interval to overwrite a "latest" checkpoint. |
-| `save_batch_interval_minutes` | `float` | No | `null` | Time interval to save a unique, batch-specific checkpoint. |
-| `save_batch_interval_minutes_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
+| `save_interval_minutes` | `float` | No | `null` | Time interval to save a unique, batch-specific checkpoint. |
+| `save_interval_minutes_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | `false` | Determines if a validation pass runs before epoch 1 begins. |
 | `log_interval` | `int` | No | `10` | Logging frequency (batches). |
 | `class_share_log_columns`| `list[str]`| No | `[]` | Columns for which to log the predicted class distribution in validation. |
diff --git a/documentation/configs/train.md b/documentation/configs/train.md
index 63d986d1..81efc1fb 100644
--- a/documentation/configs/train.md
+++ b/documentation/configs/train.md
@@ -72,8 +72,8 @@ These fields determine the size and complexity of the Transformer.
 | `class_weights` | `dict` | No | `null` | Weights for specific classes (useful for imbalanced datasets). |
 | `save_interval_epochs` | `int` | **Yes** | - | Save a checkpoint every N epochs. |
 | `save_latest_interval_minutes`| `float`| No | Time interval to overwrite a "latest" checkpoint. |
-| `save_batch_interval_minutes` | `float` | No | Time interval to save a unique, batch-specific checkpoint. |
-| `save_batch_interval_minutes_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
+| `save_interval_minutes` | `float` | No | Time interval to save a unique, batch-specific checkpoint. |
+| `save_interval_minutes_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | Determines if a validation pass runs before epoch 1 begins. Defaults to true. |
 | `early_stopping_epochs`| `int` | No | `null` | Stop training if validation loss doesn't improve for N epochs. |
 | `log_interval` | `int` | No | `10` | Print training logs every N batches. |
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index c02aa5bf..20d6f439 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -421,8 +421,8 @@ These fields determine the size and complexity of the Transformer.
 | `class_weights` | `dict` | No | `null` | Weights for specific classes (useful for imbalanced datasets). |
 | `save_interval_epochs` | `int` | **Yes** | - | Save a checkpoint every N epochs. |
 | `save_latest_interval_minutes`| `float`| No | Time interval to overwrite a "latest" checkpoint. |
-| `save_batch_interval_minutes` | `float` | No | Time interval to save a unique, batch-specific checkpoint. |
-| `save_batch_interval_minutes_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
+| `save_interval_minutes` | `float` | No | Time interval to save a unique, batch-specific checkpoint. |
+| `save_interval_minutes_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | Determines if a validation pass runs before epoch 1 begins. Defaults to true. |
 | `early_stopping_epochs`| `int` | No | `null` | Stop training if validation loss doesn't improve for N epochs. |
 | `log_interval` | `int` | No | `10` | Print training logs every N batches. |
@@ -863,8 +863,8 @@ Most fields here are lists for sampling, but some are scalar values fixed for al
 | `scheduler` | `list[dict]` | No | `[{'name': 'StepLR'...}]`| List of scheduler configs. `scheduler.step()` is only called if \< total\_steps, so correct configuration is essential. |
 | `scheduler_step_on` | `str` | No | `epoch` | When to step the scheduler: `epoch` or `batch`. |
 | `save_latest_interval_minutes`| `float`| No | `null` | Time interval to overwrite a "latest" checkpoint. |
-| `save_batch_interval_minutes` | `float` | No | `null` | Time interval to save a unique, batch-specific checkpoint. |
-| `save_batch_interval_minutes_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
+| `save_interval_minutes` | `float` | No | `null` | Time interval to save a unique, batch-specific checkpoint. |
+| `save_interval_minutes_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | `false` | Determines if a validation pass runs before epoch 1 begins. |
 | `log_interval` | `int` | No | `10` | Logging frequency (batches). |
 | `class_share_log_columns`| `list[str]`| No | `[]` | Columns for which to log the predicted class distribution in validation. |
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index 0ca0ccfe..ce1afc65 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -216,8 +216,8 @@ class TrainingSpecHyperparameterSampling(BaseModel):
     early_stopping_epochs: Optional[int] = None
     save_interval_epochs: int
     save_latest_interval_minutes: Optional[float] = None
-    save_batch_interval_minutes: Optional[float] = None
-    save_batch_interval_minutes_val_loss: bool = True
+    save_interval_minutes: Optional[float] = None
+    save_interval_minutes_val_loss: bool = True
     calculate_validation_loss_on_initialization: bool = False
 
     training_objective: list[str] = Field(default_factory=lambda: ["causal"])
@@ -398,8 +398,8 @@ def sample_trial(self, trial: Any) -> TrainingSpecModel:
             early_stopping_epochs=self.early_stopping_epochs,
             save_interval_epochs=self.save_interval_epochs,
             save_latest_interval_minutes=self.save_latest_interval_minutes,
-            save_batch_interval_minutes=self.save_batch_interval_minutes,
-            save_batch_interval_minutes_val_loss=self.save_batch_interval_minutes_val_loss,
+            save_interval_minutes=self.save_interval_minutes,
+            save_interval_minutes_val_loss=self.save_interval_minutes_val_loss,
             calculate_validation_loss_on_initialization=self.calculate_validation_loss_on_initialization,
             batch_size=batch_size,
             learning_rate=learning_rate,
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 21d2457d..9d799a12 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -216,8 +216,8 @@ class TrainingSpecModel(BaseModel):
     early_stopping_epochs: Optional[int] = None
     save_interval_epochs: int
     save_latest_interval_minutes: Optional[float] = None
-    save_batch_interval_minutes: Optional[float] = None
-    save_batch_interval_minutes_val_loss: bool = True
+    save_interval_minutes: Optional[float] = None
+    save_interval_minutes_val_loss: bool = True
     calculate_validation_loss_on_initialization: bool = True
     batch_size: int
     learning_rate: float
@@ -635,11 +635,11 @@ def validate_training_spec(cls, v, info):
             raise ValueError("save_latest_interval_minutes must be larger than 0")
 
         if (
-            v.save_batch_interval_minutes is not None
+            v.save_interval_minutes is not None
             and not os.getenv("SEQUIFIER_TESTING", "0") == "1"
-            and v.save_batch_interval_minutes == 0
+            and v.save_interval_minutes == 0
         ):
-            raise ValueError("save_batch_interval_minutes must be larger than 0")
+            raise ValueError("save_interval_minutes must be larger than 0")
 
         if v.torch_compile not in ["outer", "inner", "none"]:
             raise ValueError(
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 92cc7c56..a1953e56 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -880,11 +880,9 @@ def __init__(
         self.save_latest_interval_minutes = (
             hparams.training_spec.save_latest_interval_minutes
         )
-        self.save_batch_interval_minutes = (
-            hparams.training_spec.save_batch_interval_minutes
-        )
-        self.save_batch_interval_minutes_val_loss = (
-            hparams.training_spec.save_batch_interval_minutes_val_loss
+        self.save_interval_minutes = hparams.training_spec.save_interval_minutes
+        self.save_interval_minutes_val_loss = (
+            hparams.training_spec.save_interval_minutes_val_loss
         )
         self.continue_training = hparams.training_spec.continue_training
 
@@ -1936,9 +1934,9 @@ def _train_epoch(
                         ) >= (self.save_latest_interval_minutes * 60):
                             should_save_latest[0] = 1
 
-                        if self.save_batch_interval_minutes is not None and (
+                        if self.save_interval_minutes is not None and (
                             elapsed_since_batch_save
-                        ) >= (self.save_batch_interval_minutes * 60):
+                        ) >= (self.save_interval_minutes * 60):
                             should_save_batch[0] = 1
 
                     if self.hparams.training_spec.distributed:
@@ -1947,7 +1945,7 @@ def _train_epoch(
                         dist.barrier()
 
                     if should_save_batch.item() == 1:
-                        if self.save_batch_interval_minutes_val_loss:
+                        if self.save_interval_minutes_val_loss:
                             val_loss, val_losses, class_counts = self._evaluate(
                                 valid_loader, ddp_model
                             )

From 9e61362f6a1119f02f7f78d0ab50e4909feed4ff Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 15:01:58 +0200
Subject: [PATCH 75/81] Rename interval checkpointing vars

---
 README.md                                     |  2 +-
 docs/source/conf.py                           |  2 +-
 .../configs/hyperparameter-search.md          |  3 ++-
 documentation/configs/train.md                |  3 ++-
 documentation/consolidated-docs.md            |  8 ++++---
 pyproject.toml                                |  2 +-
 .../config/hyperparameter_search_config.py    |  6 +++--
 src/sequifier/config/train_config.py          | 10 +++++++-
 src/sequifier/train.py                        | 23 +++++++++++++------
 9 files changed, 41 insertions(+), 18 deletions(-)

diff --git a/README.md b/README.md
index 10cb0085..c999411d 100644
--- a/README.md
+++ b/README.md
@@ -199,7 +199,7 @@ Please cite with:
   title = {sequifier - causal transformer models for multivariate sequence modelling},
   year = {2025},
   publisher = {GitHub},
-  version = {v2.0.0.0},
+  version = {v1.9.9.9},
   url = {[https://github.com/0xideas/sequifier](https://github.com/0xideas/sequifier)}
 }
 
diff --git a/docs/source/conf.py b/docs/source/conf.py
index b0b88606..3d5bd0d0 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -15,7 +15,7 @@
 project = 'sequifier'
 copyright = '2025, Leon Luithlen'
 author = 'Leon Luithlen'
-release = 'v2.0.0.0'
+release = 'v1.9.9.9'
 html_baseurl = 'https://www.sequifier.com/'
 
 # -- General configuration ---------------------------------------------------
diff --git a/documentation/configs/hyperparameter-search.md b/documentation/configs/hyperparameter-search.md
index fa147d06..ab453dbe 100644
--- a/documentation/configs/hyperparameter-search.md
+++ b/documentation/configs/hyperparameter-search.md
@@ -135,7 +135,8 @@ Most fields here are lists for sampling, but some are scalar values fixed for al
 | `scheduler_step_on` | `str` | No | `epoch` | When to step the scheduler: `epoch` or `batch`. |
 | `save_latest_interval_minutes`| `float`| No | `null` | Time interval to overwrite a "latest" checkpoint. |
 | `save_interval_minutes` | `float` | No | `null` | Time interval to save a unique, batch-specific checkpoint. |
-| `save_interval_minutes_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
+| `save_interval_batches` | `int` | No | `null` | Batch interval to save a unique, batch-specific checkpoint. |
+| `save_interval_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | `false` | Determines if a validation pass runs before epoch 1 begins. |
 | `log_interval` | `int` | No | `10` | Logging frequency (batches). |
 | `class_share_log_columns`| `list[str]`| No | `[]` | Columns for which to log the predicted class distribution in validation. |
diff --git a/documentation/configs/train.md b/documentation/configs/train.md
index 81efc1fb..a663ed46 100644
--- a/documentation/configs/train.md
+++ b/documentation/configs/train.md
@@ -73,7 +73,8 @@ These fields determine the size and complexity of the Transformer.
 | `save_interval_epochs` | `int` | **Yes** | - | Save a checkpoint every N epochs. |
 | `save_latest_interval_minutes`| `float`| No | Time interval to overwrite a "latest" checkpoint. |
 | `save_interval_minutes` | `float` | No | Time interval to save a unique, batch-specific checkpoint. |
-| `save_interval_minutes_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
+| `save_interval_batches` | `int` | No | Batch interval to save a unique, batch-specific checkpoint. |
+| `save_interval_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | Determines if a validation pass runs before epoch 1 begins. Defaults to true. |
 | `early_stopping_epochs`| `int` | No | `null` | Stop training if validation loss doesn't improve for N epochs. |
 | `log_interval` | `int` | No | `10` | Print training logs every N batches. |
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 20d6f439..96318dab 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -199,7 +199,7 @@ Please cite with:
   title = {sequifier - causal transformer models for multivariate sequence modelling},
   year = {2025},
   publisher = {GitHub},
-  version = {v2.0.0.0},
+  version = {v1.9.9.9},
   url = {[https://github.com/0xideas/sequifier](https://github.com/0xideas/sequifier)}
 }
 
@@ -422,7 +422,8 @@ These fields determine the size and complexity of the Transformer.
 | `save_interval_epochs` | `int` | **Yes** | - | Save a checkpoint every N epochs. |
 | `save_latest_interval_minutes`| `float`| No | Time interval to overwrite a "latest" checkpoint. |
 | `save_interval_minutes` | `float` | No | Time interval to save a unique, batch-specific checkpoint. |
-| `save_interval_minutes_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
+| `save_interval_batches` | `int` | No | Batch interval to save a unique, batch-specific checkpoint. |
+| `save_interval_val_loss` | `bool` | No | Whether to calculate validation loss at the moment of the batch interval save. Defaults to true. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | Determines if a validation pass runs before epoch 1 begins. Defaults to true. |
 | `early_stopping_epochs`| `int` | No | `null` | Stop training if validation loss doesn't improve for N epochs. |
 | `log_interval` | `int` | No | `10` | Print training logs every N batches. |
@@ -864,7 +865,8 @@ Most fields here are lists for sampling, but some are scalar values fixed for al
 | `scheduler_step_on` | `str` | No | `epoch` | When to step the scheduler: `epoch` or `batch`. |
 | `save_latest_interval_minutes`| `float`| No | `null` | Time interval to overwrite a "latest" checkpoint. |
 | `save_interval_minutes` | `float` | No | `null` | Time interval to save a unique, batch-specific checkpoint. |
-| `save_interval_minutes_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
+| `save_interval_batches` | `int` | No | `null` | Batch interval to save a unique, batch-specific checkpoint. |
+| `save_interval_val_loss` | `bool` | No | `true` | Whether to calculate validation loss at the moment of the batch interval save. |
 | `calculate_validation_loss_on_initialization` | `bool` | No | `false` | Determines if a validation pass runs before epoch 1 begins. |
 | `log_interval` | `int` | No | `10` | Logging frequency (batches). |
 | `class_share_log_columns`| `list[str]`| No | `[]` | Columns for which to log the predicted class distribution in validation. |
diff --git a/pyproject.toml b/pyproject.toml
index b6550e5c..70f90047 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sequifier"
-version = "v2.0.0.0"
+version = "v1.9.9.9"
 authors = [
     { name = "Leon Luithlen", email = "leontimnaluithlen@gmail.com" },
 ]
diff --git a/src/sequifier/config/hyperparameter_search_config.py b/src/sequifier/config/hyperparameter_search_config.py
index ce1afc65..122b48f8 100644
--- a/src/sequifier/config/hyperparameter_search_config.py
+++ b/src/sequifier/config/hyperparameter_search_config.py
@@ -217,7 +217,8 @@ class TrainingSpecHyperparameterSampling(BaseModel):
     save_interval_epochs: int
     save_latest_interval_minutes: Optional[float] = None
     save_interval_minutes: Optional[float] = None
-    save_interval_minutes_val_loss: bool = True
+    save_interval_val_loss: bool = True
+    save_interval_batches: Optional[int] = None
     calculate_validation_loss_on_initialization: bool = False
 
     training_objective: list[str] = Field(default_factory=lambda: ["causal"])
@@ -399,7 +400,8 @@ def sample_trial(self, trial: Any) -> TrainingSpecModel:
             save_interval_epochs=self.save_interval_epochs,
             save_latest_interval_minutes=self.save_latest_interval_minutes,
             save_interval_minutes=self.save_interval_minutes,
-            save_interval_minutes_val_loss=self.save_interval_minutes_val_loss,
+            save_interval_batches=self.save_interval_batches,
+            save_interval_val_loss=self.save_interval_val_loss,
             calculate_validation_loss_on_initialization=self.calculate_validation_loss_on_initialization,
             batch_size=batch_size,
             learning_rate=learning_rate,
diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index 9d799a12..c5eff0b6 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -217,7 +217,8 @@ class TrainingSpecModel(BaseModel):
     save_interval_epochs: int
     save_latest_interval_minutes: Optional[float] = None
     save_interval_minutes: Optional[float] = None
-    save_interval_minutes_val_loss: bool = True
+    save_interval_batches: Optional[int] = None
+    save_interval_val_loss: bool = True
     calculate_validation_loss_on_initialization: bool = True
     batch_size: int
     learning_rate: float
@@ -641,6 +642,13 @@ def validate_training_spec(cls, v, info):
         ):
             raise ValueError("save_interval_minutes must be larger than 0")
 
+        if (
+            v.save_interval_batches is not None
+            and not os.getenv("SEQUIFIER_TESTING", "0") == "1"
+            and v.save_interval_batches == 0
+        ):
+            raise ValueError("save_interval_batches must be larger than 0")
+
         if v.torch_compile not in ["outer", "inner", "none"]:
             raise ValueError(
                 f'torch_compile {v.torch_compile} invalid, must be one of ["outer", "inner", "none"]'
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index a1953e56..a499db76 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -881,9 +881,7 @@ def __init__(
             hparams.training_spec.save_latest_interval_minutes
         )
         self.save_interval_minutes = hparams.training_spec.save_interval_minutes
-        self.save_interval_minutes_val_loss = (
-            hparams.training_spec.save_interval_minutes_val_loss
-        )
+        self.save_interval_val_loss = hparams.training_spec.save_interval_val_loss
         self.continue_training = hparams.training_spec.continue_training
 
         use_scaler = False
@@ -1565,6 +1563,9 @@ def train_model(
         try:
             self.last_latest_save_time = time.time()
             self.last_batch_save_time = time.time()
+            self.last_batch_save_global_step = (self.start_epoch - 1) * len(
+                train_loader
+            ) + self.start_batch
 
             if (
                 self.start_epoch == 1
@@ -1927,6 +1928,10 @@ def _train_epoch(
 
                     current_time = time.time()
                     elapsed_since_batch_save = current_time - self.last_batch_save_time
+                    current_global_step = (epoch - 1) * num_batches + (batch_count + 1)
+                    batches_since_batch_save = (
+                        current_global_step - self.last_batch_save_global_step
+                    )
 
                     if not self.hparams.training_spec.distributed or self.rank == 0:
                         if self.save_latest_interval_minutes is not None and (
@@ -1939,13 +1944,19 @@ def _train_epoch(
                         ) >= (self.save_interval_minutes * 60):
                             should_save_batch[0] = 1
 
+                        if (
+                            self.save_interval_batches is not None
+                            and batches_since_batch_save >= self.save_interval_batches
+                        ):
+                            should_save_batch[0] = 1
+
                     if self.hparams.training_spec.distributed:
                         dist.broadcast(should_save_latest, src=0)
                         dist.broadcast(should_save_batch, src=0)
                         dist.barrier()
 
                     if should_save_batch.item() == 1:
-                        if self.save_interval_minutes_val_loss:
+                        if self.save_interval_val_loss:
                             val_loss, val_losses, class_counts = self._evaluate(
                                 valid_loader, ddp_model
                             )
@@ -1954,9 +1965,6 @@ def _train_epoch(
                                 not self.hparams.training_spec.distributed
                                 or self.rank == 0
                             ):
-                                current_global_step = (epoch - 1) * num_batches + (
-                                    batch_count + 1
-                                )
                                 self._log_epoch_results(
                                     0,
                                     batch_count + 1,
@@ -2002,6 +2010,7 @@ def _train_epoch(
                             num_batches=num_batches,
                         )
                         self.last_batch_save_time = time.time()
+                        self.last_batch_save_global_step = current_global_step
 
         if dataset_handles_start_batch:
             set_dataset_start_batch(0)

From 1171da5664fead28ef40775f2638ffe27bf8f724 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 16:21:43 +0200
Subject: [PATCH 76/81] Add self.save_interval_batches

---
 src/sequifier/train.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index a499db76..ce8fbad6 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -881,6 +881,7 @@ def __init__(
             hparams.training_spec.save_latest_interval_minutes
         )
         self.save_interval_minutes = hparams.training_spec.save_interval_minutes
+        self.save_interval_batches = hparams.training_spec.save_interval_batches
         self.save_interval_val_loss = hparams.training_spec.save_interval_val_loss
         self.continue_training = hparams.training_spec.continue_training
 

From 89cf33c9c82ad8c7c1ef8fbcf82476aafd942e14 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 16:58:17 +0200
Subject: [PATCH 77/81] remove accumulated_global_token_count,
 empty_global_batches, global_token_count

---
 src/sequifier/train.py | 30 ++++++------------------------
 uv.lock                | 36 ++++++++++++++++++------------------
 2 files changed, 24 insertions(+), 42 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index ce8fbad6..3fc97caf 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -1739,10 +1739,7 @@ def _train_epoch(
         """Run one train epoch with optional mid-epoch saves."""
         target_names = self._loss_target_names()
         train_loss_sums, train_token_count = self._new_loss_accumulators(target_names)
-        empty_global_batches = 0
-        accumulated_global_token_count = torch.zeros(
-            (), device=self.device, dtype=torch.int64
-        )
+
         batches_aggregated = 0
 
         start_time = time.time()
@@ -1814,7 +1811,6 @@ def _train_epoch(
                             backward_components,
                             local_loss_sums,
                             local_token_count,
-                            global_token_count,
                         ) = self._calculate_training_loss(output, targets, metadata)
                 else:
                     output = model_to_call(data, metadata=metadata, return_logits=True)
@@ -1823,7 +1819,6 @@ def _train_epoch(
                         backward_components,
                         local_loss_sums,
                         local_token_count,
-                        global_token_count,
                     ) = self._calculate_training_loss(output, targets, metadata)
 
                 if self.accumulation_steps is None:
@@ -1845,9 +1840,6 @@ def _train_epoch(
                     local_loss_sums,
                     local_token_count,
                 )
-                accumulated_global_token_count += global_token_count.detach()
-                if global_token_count.detach().cpu().item() == 0:
-                    empty_global_batches += 1
 
                 optimizer_step_due = (
                     self.accumulation_steps is None
@@ -1856,10 +1848,7 @@ def _train_epoch(
                 )
                 optimizer_step_performed = False
 
-                if (
-                    optimizer_step_due
-                    and accumulated_global_token_count.detach().cpu().item() > 0
-                ):
+                if optimizer_step_due:
                     self.scaler.unscale_(self.optimizer)
 
                     torch.nn.utils.clip_grad_norm_(self.parameters(), 0.5)
@@ -1872,7 +1861,6 @@ def _train_epoch(
                 if optimizer_step_due:
                     if not optimizer_step_performed:
                         self.optimizer.zero_grad()
-                    accumulated_global_token_count.zero_()
 
                 batches_aggregated += 1
                 if (batch_count + 1) % self.log_interval == 0:
@@ -1891,16 +1879,10 @@ def _train_epoch(
                         self.logger.info(
                             f"[INFO] Epoch {epoch:3d} | Batch {(batch_count+1):5d}/{num_batches:5d} | Loss: {format_number(avg_train_loss.detach().cpu().item())} | LR: {format_number(learning_rate)} | S/Batch {format_number(s_per_batch)}"
                         )
-                        if empty_global_batches > 0:
-                            self.logger.warning(
-                                "[WARNING] "
-                                f"{empty_global_batches} training microbatch(es) "
-                                "in this logging window contained no selected tokens."
-                            )
+
                     train_loss_sums, train_token_count = self._new_loss_accumulators(
                         target_names
                     )
-                    empty_global_batches = 0
                     if self.rank == 0:
                         batches_aggregated = 0
                         self.start_batch = 0
@@ -2024,7 +2006,7 @@ def _calculate_loss(
         metadata: dict[str, Tensor],
     ) -> tuple[Tensor, dict[str, Tensor]]:
         """Return backward-scaled loss and components for the current rank."""
-        loss, backward_components, _, _, _ = self._calculate_training_loss(
+        loss, backward_components, _, _ = self._calculate_training_loss(
             output, targets, metadata
         )
         return loss, backward_components
@@ -2035,7 +2017,7 @@ def _calculate_training_loss(
         output: dict[str, Tensor],
         targets: dict[str, Tensor],
         metadata: dict[str, Tensor],
-    ) -> tuple[Tensor, dict[str, Tensor], dict[str, Tensor], Tensor, Tensor]:
+    ) -> tuple[Tensor, dict[str, Tensor], dict[str, Tensor], Tensor]:
         """Return the normalized backward loss plus local metric primitives."""
         target_names = self._loss_target_names(targets)
         if not target_names:
@@ -2079,7 +2061,7 @@ def _calculate_training_loss(
                 "Loss calculation failed; no loss tensors were generated."
             )
 
-        return loss, backward_components, local_sums, local_count, global_count
+        return loss, backward_components, local_sums, local_count
 
     @beartype
     def _calculate_local_loss_components(
diff --git a/uv.lock b/uv.lock
index f078af13..f4d5481e 100644
--- a/uv.lock
+++ b/uv.lock
@@ -285,7 +285,7 @@ name = "cuda-bindings"
 version = "13.3.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "cuda-pathfinder" },
+    { name = "cuda-pathfinder", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/a9/21/8464d133752951c154feafb3b65c297e7d80f301183d220bec4c830f1441/cuda_bindings-13.3.1-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:120fcc53d57903df529c3486962c56528cba5b7d6c57c99537320ed9922c8b86", size = 6073403, upload-time = "2026-05-29T23:11:36.22Z" },
@@ -320,34 +320,34 @@ wheels = [
 
 [package.optional-dependencies]
 cudart = [
-    { name = "nvidia-cuda-runtime", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-cuda-runtime", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 cufft = [
-    { name = "nvidia-cufft", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-cufft", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 cufile = [
     { name = "nvidia-cufile", marker = "sys_platform == 'linux'" },
 ]
 cupti = [
-    { name = "nvidia-cuda-cupti", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-cuda-cupti", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 curand = [
-    { name = "nvidia-curand", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-curand", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 cusolver = [
-    { name = "nvidia-cusolver", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-cusolver", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 cusparse = [
-    { name = "nvidia-cusparse", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-cusparse", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 nvjitlink = [
-    { name = "nvidia-nvjitlink", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 nvrtc = [
-    { name = "nvidia-cuda-nvrtc", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-cuda-nvrtc", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 nvtx = [
-    { name = "nvidia-nvtx", marker = "sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "nvidia-nvtx", marker = "(python_full_version < '3.11' and sys_platform == 'win32') or sys_platform == 'linux'" },
 ]
 
 [[package]]
@@ -962,7 +962,7 @@ name = "nvidia-cublas"
 version = "13.1.1.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-cuda-nvrtc" },
+    { name = "nvidia-cuda-nvrtc", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/a7/a1/0bd24ee8c8d03adac032fd2909426a00c88f8c57961b1277ded97f91119f/nvidia_cublas-13.1.1.3-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:b7a210458267ac818974c53038fbec2e969d5c99f305ab15c72522fa9f001dd5", size = 542848918, upload-time = "2026-04-08T18:46:22.985Z" },
@@ -1001,7 +1001,7 @@ name = "nvidia-cudnn-cu13"
 version = "9.20.0.48"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-cublas" },
+    { name = "nvidia-cublas", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/56/c5/83384d846b2fd17c44bd499b36c75a45ed4f095fbbb2252294e89cea5c5c/nvidia_cudnn_cu13-9.20.0.48-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:e31454ae00094b0c55319d9d15b6fa2fc50a9e1c0f5c8c80fb75258234e731e1", size = 444574296, upload-time = "2026-03-09T19:28:27.751Z" },
@@ -1013,7 +1013,7 @@ name = "nvidia-cufft"
 version = "12.0.0.61"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-nvjitlink" },
+    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/8b/ae/f417a75c0259e85c1d2f83ca4e960289a5f814ed0cea74d18c353d3e989d/nvidia_cufft-12.0.0.61-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:2708c852ef8cd89d1d2068bdbece0aa188813a0c934db3779b9b1faa8442e5f5", size = 214053554, upload-time = "2025-09-04T08:31:38.196Z" },
@@ -1043,9 +1043,9 @@ name = "nvidia-cusolver"
 version = "12.0.4.66"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-cublas" },
-    { name = "nvidia-cusparse" },
-    { name = "nvidia-nvjitlink" },
+    { name = "nvidia-cublas", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-cusparse", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/c8/c3/b30c9e935fc01e3da443ec0116ed1b2a009bb867f5324d3f2d7e533e776b/nvidia_cusolver-12.0.4.66-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:02c2457eaa9e39de20f880f4bd8820e6a1cfb9f9a34f820eb12a155aa5bc92d2", size = 223467760, upload-time = "2025-09-04T08:33:04.222Z" },
@@ -1057,7 +1057,7 @@ name = "nvidia-cusparse"
 version = "12.6.3.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-nvjitlink" },
+    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.11' and sys_platform == 'emscripten') or (python_full_version < '3.11' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/94/5c26f33738ae35276672f12615a64bd008ed5be6d1ebcb23579285d960a9/nvidia_cusparse-12.6.3.3-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:80bcc4662f23f1054ee334a15c72b8940402975e0eab63178fc7e670aa59472c", size = 162155568, upload-time = "2025-09-04T08:33:42.864Z" },
@@ -1918,7 +1918,7 @@ wheels = [
 
 [[package]]
 name = "sequifier"
-version = "2.0.0.0"
+version = "1.9.9.9"
 source = { editable = "." }
 dependencies = [
     { name = "beartype" },

From 7f08cd4220e23b49193c8bf0dad0779918b5a539 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Sat, 20 Jun 2026 19:08:07 +0200
Subject: [PATCH 78/81] prevent layer_type_dtypes with FSDP

---
 src/sequifier/config/train_config.py | 12 ++++++++++++
 src/sequifier/train.py               | 27 +++++++++++++++++----------
 2 files changed, 29 insertions(+), 10 deletions(-)

diff --git a/src/sequifier/config/train_config.py b/src/sequifier/config/train_config.py
index c5eff0b6..da28e1a8 100644
--- a/src/sequifier/config/train_config.py
+++ b/src/sequifier/config/train_config.py
@@ -654,6 +654,18 @@ def validate_training_spec(cls, v, info):
                 f'torch_compile {v.torch_compile} invalid, must be one of ["outer", "inner", "none"]'
             )
 
+        if v.data_parallelism == "FSDP":
+            if v.layer_type_dtypes is not None:
+                raise ValueError(
+                    "FSDP does not support manual layer pre-casting. Please set "
+                    "'layer_type_dtypes' to null when using FSDP, and rely on "
+                    "'layer_autocast' (MixedPrecisionPolicy) instead."
+                )
+            if v.fsdp_cpu_offload is None:
+                raise ValueError(
+                    "If data_parallelism == 'FSDP', fsdp_cpu_offload cannot be None"
+                )
+
         if v.data_parallelism == "FSDP" and v.torch_compile == "outer":
             raise ValueError(
                 "If data_parallelism is set to 'FSDP' then torch_compile must be one of 'none' and 'inner'"
diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index 3fc97caf..cd22b4fb 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -269,18 +269,19 @@ def train_worker(
         )  # 1D mesh for standard ZeRO-3 full sharding
         model._data_parallel_group = mesh.get_group()
 
-        mp_policy = None
         if config.training_spec.layer_autocast:
             amp_dtype = get_torch_dtype(
                 config.training_spec.layer_type_dtypes.get("linear", "bfloat16")
                 if config.training_spec.layer_type_dtypes
                 else "bfloat16"
             )
-            mp_policy = MixedPrecisionPolicy(
-                param_dtype=amp_dtype,
-                reduce_dtype=amp_dtype,
-                output_dtype=amp_dtype,
-            )
+        else:
+            amp_dtype = torch.float32
+        mp_policy = MixedPrecisionPolicy(
+            param_dtype=amp_dtype,
+            reduce_dtype=amp_dtype,
+            output_dtype=amp_dtype,
+        )
 
         offload_policy = (
             OffloadPolicy() if config.training_spec.fsdp_cpu_offload else None
@@ -1429,12 +1430,13 @@ def _validate_checkpoint_compatibility(
     @beartype
     def _get_rng_state(self) -> dict[str, Any]:
         """Capture Python, NumPy, Torch CPU, and CUDA RNG state for this rank."""
+        device = torch.device(self.device)
         return {
             "python": random.getstate(),
             "numpy": np.random.get_state(),
             "torch": torch.get_rng_state(),
-            "cuda": torch.cuda.get_rng_state_all()
-            if torch.cuda.is_available()
+            "cuda": torch.cuda.get_rng_state(device=device)
+            if device.type == "cuda" and torch.cuda.is_available()
             else None,
         }
 
@@ -1543,8 +1545,13 @@ def _restore_rng_state(self) -> None:
         np.random.set_state(rng_state["numpy"])
         torch.set_rng_state(rng_state["torch"])
         cuda_state = rng_state.get("cuda")
-        if cuda_state is not None and torch.cuda.is_available():
-            torch.cuda.set_rng_state_all(cuda_state)
+        device = torch.device(self.device)
+        if (
+            cuda_state is not None
+            and device.type == "cuda"
+            and torch.cuda.is_available()
+        ):
+            torch.cuda.set_rng_state(cuda_state, device=device)
 
     @beartype
     def train_model(

From e8427806459dc485181e8c754e87cf84556f20d7 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Mon, 22 Jun 2026 15:01:23 +0200
Subject: [PATCH 79/81] Fix FSDP mixed precision policy

---
 src/sequifier/train.py | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)

diff --git a/src/sequifier/train.py b/src/sequifier/train.py
index cd22b4fb..ccdb4e11 100644
--- a/src/sequifier/train.py
+++ b/src/sequifier/train.py
@@ -269,30 +269,28 @@ def train_worker(
         )  # 1D mesh for standard ZeRO-3 full sharding
         model._data_parallel_group = mesh.get_group()
 
+        fsdp_kwargs = {"mesh": mesh}
         if config.training_spec.layer_autocast:
             amp_dtype = get_torch_dtype(
                 config.training_spec.layer_type_dtypes.get("linear", "bfloat16")
                 if config.training_spec.layer_type_dtypes
                 else "bfloat16"
             )
+
+            fsdp_kwargs["mp_policy"] = MixedPrecisionPolicy(
+                param_dtype=amp_dtype,
+                reduce_dtype=amp_dtype,
+                output_dtype=amp_dtype,
+            )
         else:
-            amp_dtype = torch.float32
-        mp_policy = MixedPrecisionPolicy(
-            param_dtype=amp_dtype,
-            reduce_dtype=amp_dtype,
-            output_dtype=amp_dtype,
-        )
+            fsdp_kwargs["mp_policy"] = MixedPrecisionPolicy()
 
-        offload_policy = (
-            OffloadPolicy() if config.training_spec.fsdp_cpu_offload else None
-        )
+        if config.training_spec.fsdp_cpu_offload:
+            fsdp_kwargs["offload_policy"] = OffloadPolicy()
         for layer in model.layers:
-            fully_shard(
-                layer, mesh=mesh, mp_policy=mp_policy, offload_policy=offload_policy
-            )
-        fully_shard(
-            model, mesh=mesh, mp_policy=mp_policy, offload_policy=offload_policy
-        )
+            fully_shard(layer, **fsdp_kwargs)
+
+        fully_shard(model, **fsdp_kwargs)
         dist.barrier()
 
         params_to_optimize = model.parameters()

From c5e452518b01613a2d1419e541400470b88dd6a9 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Tue, 23 Jun 2026 11:54:16 +0200
Subject: [PATCH 80/81] Add type casting to preprocessing

---
 documentation/configs/preprocess.md       |   1 +
 documentation/consolidated-docs.md        |   1 +
 src/sequifier/config/preprocess_config.py |  27 +-
 src/sequifier/helpers.py                  |  67 ++++-
 src/sequifier/make.py                     |   1 +
 src/sequifier/preprocess.py               | 293 ++++++++++++++++++++--
 6 files changed, 354 insertions(+), 36 deletions(-)

diff --git a/documentation/configs/preprocess.md b/documentation/configs/preprocess.md
index 8837d3f0..283af613 100644
--- a/documentation/configs/preprocess.md
+++ b/documentation/configs/preprocess.md
@@ -36,6 +36,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `selected_columns` | `list[str]` | No | `null` | A specific list of columns to process. If `null`, all columns (except metadata) are processed. |
+| `column_types` | `dict[str, str]` | No | `null` | Optional output dtype map for processed columns, such as `Float32`, `Float64`, `Int32`, or `Int64`. If set, every processed column must be included. Parquet uses one unified sequence dtype; `pt` writes each variable to its configured tensor dtype. |
 | `max_rows` | `int` | No | `null` | Limits processing to the first N rows. Useful for rapid debugging. |
 | `metadata_config_path` | `Optional[str]` | No | `null` | use a preexisting metadata config path for tokenizing discrete columns and standardising real-valued columns |
 | `mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
diff --git a/documentation/consolidated-docs.md b/documentation/consolidated-docs.md
index 96318dab..35e0e293 100644
--- a/documentation/consolidated-docs.md
+++ b/documentation/consolidated-docs.md
@@ -244,6 +244,7 @@ The configuration is defined in a YAML file (e.g., `preprocess.yaml`). Below are
 | Field | Type | Mandatory | Default | Description |
 | :--- | :--- | :--- | :--- | :--- |
 | `selected_columns` | `list[str]` | No | `null` | A specific list of columns to process. If `null`, all columns (except metadata) are processed. |
+| `column_types` | `dict[str, str]` | No | `null` | Optional output dtype map for processed columns, such as `Float32`, `Float64`, `Int32`, or `Int64`. If set, every processed column must be included. Parquet uses one unified sequence dtype; `pt` writes each variable to its configured tensor dtype. |
 | `max_rows` | `int` | No | `null` | Limits processing to the first N rows. Useful for rapid debugging. |
 | `metadata_config_path` | `Optional[str]` | No | `null` | use a preexisting metadata config path for tokenizing discrete columns and standardising real-valued columns |
 | `mask_column` | `Optional[str]` | No | `null` | Optional input column used as a row-level mask. If set, `metadata_config_path` must also be set. |
diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index ef1cdf4c..36d1d0f9 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -14,7 +14,7 @@
     model_validator,
 )
 
-from sequifier.helpers import try_catch_excess_keys
+from sequifier.helpers import canonicalize_polars_dtype_name, try_catch_excess_keys
 
 
 @beartype
@@ -44,6 +44,7 @@ class PreprocessorModel(BaseModel):
     merge_output: bool = True
     allow_sequence_splitting: bool = False
     selected_columns: Optional[list[str]] = None
+    column_types: Optional[dict[str, str]] = None
 
     split_ratios: list[float]
     stored_context_width: int = Field(gt=0)
@@ -138,6 +139,30 @@ def validate_batches_per_file(cls, v: int) -> int:
             raise ValueError("batches_per_file must be a positive integer")
         return v
 
+    @field_validator("column_types")
+    @classmethod
+    def validate_column_types(
+        cls, v: Optional[dict[str, str]], info: ValidationInfo
+    ) -> Optional[dict[str, str]]:
+        if v is None:
+            return None
+
+        normalized = {
+            column: canonicalize_polars_dtype_name(dtype) for column, dtype in v.items()
+        }
+        selected_columns = info.data.get("selected_columns")
+        if selected_columns is not None:
+            missing_columns = [
+                column for column in selected_columns if column not in normalized
+            ]
+            if missing_columns:
+                raise ValueError(
+                    "column_types must include every selected column. "
+                    f"Missing: {missing_columns}"
+                )
+
+        return normalized
+
     @field_validator("continue_preprocessing")
     @classmethod
     def validate_continue_preprocessing(cls, v: bool, info: ValidationInfo) -> bool:
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index 4faee46b..b2a846b0 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -19,8 +19,8 @@
 from sequifier.special_tokens import SPECIAL_TOKEN_IDS
 
 PANDAS_TO_TORCH_TYPES = {
-    "Float64": torch.float32,
-    "float64": torch.float32,
+    "Float64": torch.float64,
+    "float64": torch.float64,
     "Float32": torch.float32,
     "float32": torch.float32,
     "Float16": torch.float16,
@@ -33,16 +33,63 @@
     "int16": torch.int16,
     "Int8": torch.int8,
     "int8": torch.int8,
-    "UInt64": torch.int64,
-    "uint64": torch.int64,
-    "UInt32": torch.int64,
-    "uint32": torch.int64,
-    "UInt16": torch.int32,
-    "uint16": torch.int32,
-    "UInt8": torch.int16,
-    "uint8": torch.int16,
+    "UInt64": torch.uint64,
+    "uint64": torch.uint64,
+    "UInt32": torch.uint32,
+    "uint32": torch.uint32,
+    "UInt16": torch.uint16,
+    "uint16": torch.uint16,
+    "UInt8": torch.uint8,
+    "uint8": torch.uint8,
 }
 
+POLARS_NUMERIC_DTYPES = {
+    "Float64": pl.Float64,
+    "Float32": pl.Float32,
+    "Float16": pl.Float16,
+    "Int64": pl.Int64,
+    "Int32": pl.Int32,
+    "Int16": pl.Int16,
+    "Int8": pl.Int8,
+    "UInt64": pl.UInt64,
+    "UInt32": pl.UInt32,
+    "UInt16": pl.UInt16,
+    "UInt8": pl.UInt8,
+}
+
+POLARS_NUMERIC_DTYPE_ALIASES = {
+    alias: canonical
+    for canonical in POLARS_NUMERIC_DTYPES
+    for alias in (canonical, canonical.lower())
+}
+
+
+@beartype
+def canonicalize_polars_dtype_name(dtype_name: str) -> str:
+    dtype_name = dtype_name.strip()
+    if dtype_name not in POLARS_NUMERIC_DTYPE_ALIASES:
+        raise ValueError(
+            f"Unsupported column type '{dtype_name}'. "
+            f"Supported types are: {sorted(POLARS_NUMERIC_DTYPES)}"
+        )
+    return POLARS_NUMERIC_DTYPE_ALIASES[dtype_name]
+
+
+@beartype
+def polars_dtype_from_name(dtype_name: str) -> Any:
+    return POLARS_NUMERIC_DTYPES[canonicalize_polars_dtype_name(dtype_name)]
+
+
+@beartype
+def is_float_dtype_name(dtype_name: str) -> bool:
+    return canonicalize_polars_dtype_name(dtype_name).startswith("Float")
+
+
+@beartype
+def is_integer_dtype_name(dtype_name: str) -> bool:
+    canonical = canonicalize_polars_dtype_name(dtype_name)
+    return canonical.startswith("Int") or canonical.startswith("UInt")
+
 
 @dataclass(frozen=True)
 class StoredWindowLayout:
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index 51d43a5b..ef87be45 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -5,6 +5,7 @@
 read_format: csv
 write_format: parquet
 selected_columns: [EXAMPLE_INPUT_COLUMN_NAME] # should include all target column, can include additional columns
+column_types: null # optional map of selected columns to output dtypes, e.g. {EXAMPLE_INPUT_COLUMN_NAME: Float32}
 mask_column: null
 
 split_ratios:
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 0550eb6f..09fa568f 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -20,6 +20,10 @@
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
     StoredWindowLayout,
+    canonicalize_polars_dtype_name,
+    is_float_dtype_name,
+    is_integer_dtype_name,
+    polars_dtype_from_name,
     read_data,
     write_data,
 )
@@ -34,6 +38,38 @@
 REAL_MASK_VALUE = 0.0
 CURRENT_STORED_WINDOW_LAYOUT_VERSION = 2
 
+FLOAT_TYPE_ORDER = ("Float16", "Float32", "Float64")
+INTEGER_TYPE_ORDER = (
+    "Int8",
+    "UInt8",
+    "Int16",
+    "UInt16",
+    "Int32",
+    "UInt32",
+    "Int64",
+    "UInt64",
+)
+INTEGER_TYPE_INFO = {
+    "Int8": np.iinfo(np.int8),
+    "Int16": np.iinfo(np.int16),
+    "Int32": np.iinfo(np.int32),
+    "Int64": np.iinfo(np.int64),
+    "UInt8": np.iinfo(np.uint8),
+    "UInt16": np.iinfo(np.uint16),
+    "UInt32": np.iinfo(np.uint32),
+    "UInt64": np.iinfo(np.uint64),
+}
+FLOAT_TYPE_INFO = {
+    "Float16": np.finfo(np.float16),
+    "Float32": np.finfo(np.float32),
+    "Float64": np.finfo(np.float64),
+}
+FLOAT_EXACT_INTEGER_LIMITS = {
+    "Float16": 2**11,
+    "Float32": 2**24,
+    "Float64": 2**53,
+}
+
 
 def _stable_json_value(value: Any) -> Any:
     if isinstance(value, dict):
@@ -59,6 +95,182 @@ def _stable_json_digest(value: Any) -> str:
     return hashlib.sha256(encoded).hexdigest()
 
 
+@beartype
+def _normalize_column_types(
+    column_types: Optional[dict[str, str]],
+) -> Optional[dict[str, str]]:
+    if column_types is None:
+        return None
+    return {
+        column: canonicalize_polars_dtype_name(dtype)
+        for column, dtype in column_types.items()
+    }
+
+
+@beartype
+def _configured_column_types_for_data_columns(
+    column_types: Optional[dict[str, str]],
+    data_columns: list[str],
+) -> Optional[dict[str, str]]:
+    if column_types is None:
+        return None
+
+    missing_columns = [column for column in data_columns if column not in column_types]
+    if missing_columns:
+        raise ValueError(
+            "column_types must include every to-be-processed column. "
+            f"Missing: {missing_columns}"
+        )
+
+    return {column: column_types[column] for column in data_columns}
+
+
+@beartype
+def _dtype_is_numeric(dtype: Any) -> bool:
+    return dtype.is_numeric() if hasattr(dtype, "is_numeric") else False
+
+
+@beartype
+def _apply_configured_input_casting(
+    data: pl.DataFrame,
+    data_columns: list[str],
+    column_types: Optional[dict[str, str]],
+) -> pl.DataFrame:
+    """Cast input columns early when the requested type defines processing semantics."""
+    configured = _configured_column_types_for_data_columns(column_types, data_columns)
+    if configured is None:
+        return data
+
+    casts = []
+    for column in data_columns:
+        target_type = configured[column]
+        source_dtype = data.schema[column]
+
+        if is_float_dtype_name(target_type):
+            casts.append(pl.col(column).cast(polars_dtype_from_name(target_type)))
+        elif is_integer_dtype_name(target_type) and _dtype_is_numeric(source_dtype):
+            casts.append(pl.col(column).cast(polars_dtype_from_name(target_type)))
+
+    if not casts:
+        return data
+
+    return data.with_columns(casts)
+
+
+@beartype
+def _apply_output_type_casting(
+    data: pl.DataFrame,
+    data_columns: list[str],
+    col_types: dict[str, str],
+) -> pl.DataFrame:
+    casts = [
+        pl.col(column).cast(polars_dtype_from_name(col_types[column]))
+        for column in data_columns
+    ]
+    if not casts:
+        return data
+    return data.with_columns(casts)
+
+
+@beartype
+def _highest_ranked_type(types: list[str], order: tuple[str, ...]) -> str:
+    return max(types, key=lambda type_: order.index(type_))
+
+
+@beartype
+def _smallest_float_covering_integer_range(integer_type: str) -> str:
+    integer_info = INTEGER_TYPE_INFO[integer_type]
+    largest_magnitude = max(abs(int(integer_info.min)), int(integer_info.max))
+    for float_type in FLOAT_TYPE_ORDER:
+        if (
+            largest_magnitude <= float(FLOAT_TYPE_INFO[float_type].max)
+            and largest_magnitude <= FLOAT_EXACT_INTEGER_LIMITS[float_type]
+        ):
+            return float_type
+    return "Float64"
+
+
+@beartype
+def _resolve_integer_sequence_type(integer_types: list[str]) -> Any:
+    if not integer_types:
+        raise ValueError("Cannot resolve an integer sequence type without integers")
+
+    min_value = min(int(INTEGER_TYPE_INFO[type_].min) for type_ in integer_types)
+    max_value = max(int(INTEGER_TYPE_INFO[type_].max) for type_ in integer_types)
+
+    if min_value >= 0 and all(type_.startswith("UInt") for type_ in integer_types):
+        for dtype_name in ("UInt8", "UInt16", "UInt32", "UInt64"):
+            info = INTEGER_TYPE_INFO[dtype_name]
+            if max_value <= int(info.max):
+                return polars_dtype_from_name(dtype_name)
+
+    for dtype_name in ("Int8", "Int16", "Int32", "Int64"):
+        info = INTEGER_TYPE_INFO[dtype_name]
+        if min_value >= int(info.min) and max_value <= int(info.max):
+            return polars_dtype_from_name(dtype_name)
+
+    raise ValueError(f"Cannot resolve a safe integer dtype for {integer_types}")
+
+
+@beartype
+def _resolve_unified_parquet_type(column_types: dict[str, str]) -> Any:
+    if not column_types:
+        raise ValueError("column_types cannot be empty")
+
+    normalized_types = [
+        canonicalize_polars_dtype_name(type_) for type_ in column_types.values()
+    ]
+    float_types = [type_ for type_ in normalized_types if is_float_dtype_name(type_)]
+    integer_types = [
+        type_ for type_ in normalized_types if is_integer_dtype_name(type_)
+    ]
+
+    if not float_types:
+        if len(set(integer_types)) > 1:
+            logger.warning(
+                "Multiple integer column_types were specified for Parquet output; "
+                "using a unified integer schema."
+            )
+        return _resolve_integer_sequence_type(integer_types)
+
+    resolved_float = _highest_ranked_type(float_types, FLOAT_TYPE_ORDER)
+    if len(set(normalized_types)) > 1:
+        logger.warning(
+            "Multiple column_types were specified for Parquet output; "
+            f"using unified sequence dtype {resolved_float}."
+        )
+
+    if integer_types:
+        required_float = _highest_ranked_type(
+            [
+                _smallest_float_covering_integer_range(integer_type)
+                for integer_type in integer_types
+            ],
+            FLOAT_TYPE_ORDER,
+        )
+        if FLOAT_TYPE_ORDER.index(required_float) > FLOAT_TYPE_ORDER.index(
+            resolved_float
+        ):
+            logger.warning(
+                "An integer column_type has a range exceeding "
+                f"{resolved_float}; upgrading unified Parquet sequence dtype "
+                f"to {required_float}."
+            )
+            resolved_float = required_float
+
+    return polars_dtype_from_name(resolved_float)
+
+
+@beartype
+def _resolve_pt_extraction_type(column_types: dict[str, str]) -> Any:
+    normalized_types = [
+        canonicalize_polars_dtype_name(type_) for type_ in column_types.values()
+    ]
+    if any(is_float_dtype_name(type_) for type_ in normalized_types):
+        return pl.Float64
+    return _resolve_integer_sequence_type(normalized_types)
+
+
 @beartype
 def preprocess(args: Any, args_config: dict[str, Any]) -> None:
     """Load preprocessing config and run preprocessing."""
@@ -96,6 +308,7 @@ def __init__(
         metadata_config_path: Optional[str],
         max_target_offset: int = 1,
         mask_column: Optional[str] = None,
+        column_types: Optional[dict[str, str]] = None,
     ):
         """Initialize and run preprocessing from validated config fields."""
         self.project_root = project_root
@@ -125,6 +338,7 @@ def __init__(
         self.max_rows = max_rows
         self.process_by_file = process_by_file
         self.subsequence_start_mode = subsequence_start_mode
+        self.column_types = _normalize_column_types(column_types)
         if self.mask_column is not None and self.metadata_config_path is None:
             raise ValueError("metadata_config_path must be set when mask_column is set")
 
@@ -176,6 +390,12 @@ def __init__(
                 self.mask_column,
             )
             data_columns = _get_data_columns(data, self.mask_column)
+            configured_col_types = _configured_column_types_for_data_columns(
+                self.column_types, data_columns
+            )
+            data = _apply_configured_input_casting(
+                data, data_columns, configured_col_types
+            )
             if self.metadata_config_path:
                 metadata_path = os.path.join(
                     self.project_root, self.metadata_config_path
@@ -221,9 +441,12 @@ def __init__(
                 id_maps,
                 selected_columns_statistics,
                 n_classes=n_classes,
-                col_types=col_types,
+                col_types=configured_col_types or col_types,
             )
+            if configured_col_types is not None:
+                col_types = configured_col_types
             data = _apply_mask_column(data, data_columns, col_types, self.mask_column)
+            data = _apply_output_type_casting(data, data_columns, col_types)
 
             self._write_or_validate_resume_manifest(
                 selected_columns,
@@ -308,6 +531,11 @@ def __init__(
                     for col in col_types.keys()
                     if col not in _reserved_input_columns(self.mask_column)
                 ]
+                configured_col_types = _configured_column_types_for_data_columns(
+                    self.column_types, data_columns
+                )
+                if configured_col_types is not None:
+                    col_types = configured_col_types
 
                 # We still need to find the files to process
                 files_to_process = []
@@ -324,10 +552,15 @@ def __init__(
                     col_types,
                     data_columns,
                 ) = self._get_column_metadata_across_files(
-                    data_path, read_format, max_rows, selected_columns
+                    data_path,
+                    read_format,
+                    max_rows,
+                    selected_columns,
+                    self.column_types,
                 )
                 for col in id_maps:
-                    col_types[col] = "Int64"
+                    if self.column_types is None:
+                        col_types[col] = "Int64"
 
             self._write_or_validate_resume_manifest(
                 selected_columns,
@@ -380,19 +613,10 @@ def _create_schema(
             "inputCol": pl.String,
         }
 
-        if (np.unique(list(col_types.values())) == np.array(["Int64"]))[0]:
-            sequence_position_type = pl.Int64
+        if self.write_format == "parquet":
+            sequence_position_type = _resolve_unified_parquet_type(col_types)
         else:
-            if not np.all(
-                [
-                    type_.lower().startswith("int")
-                    or type_.lower().startswith("uint")
-                    or type_.lower().startswith("float")
-                    for type_ in col_types.values()
-                ]
-            ):
-                raise ValueError("All column types must start with int or float")
-            sequence_position_type = pl.Float64
+            sequence_position_type = _resolve_pt_extraction_type(col_types)
 
         schema.update(
             {
@@ -410,6 +634,7 @@ def _get_column_metadata_across_files(
         read_format: str,
         max_rows: Optional[int],
         selected_columns: Optional[list[str]],
+        column_types: Optional[dict[str, str]],
     ) -> tuple[
         list[str],
         dict[str, int],
@@ -453,10 +678,20 @@ def _get_column_metadata_across_files(
                     )
 
                     current_file_cols = _get_data_columns(data, self.mask_column)
+                    current_configured_col_types = (
+                        _configured_column_types_for_data_columns(
+                            column_types, current_file_cols
+                        )
+                    )
+                    data = _apply_configured_input_casting(
+                        data, current_file_cols, current_configured_col_types
+                    )
 
                     if col_types is None:
                         data_columns = current_file_cols
-                        col_types = {col: str(data.schema[col]) for col in data_columns}
+                        col_types = current_configured_col_types or {
+                            col: str(data.schema[col]) for col in data_columns
+                        }
 
                         for col in precomputed_id_maps.keys():
                             if col not in data_columns:
@@ -475,12 +710,13 @@ def _get_column_metadata_across_files(
                                 f"Extra: {extra}"
                             )
 
-                        for col in current_file_cols:
-                            if str(data.schema[col]) != col_types[col]:
-                                raise ValueError(
-                                    f"Type mismatch for column '{col}' in file '{file}'. "
-                                    f"Expected {col_types[col]}, got {str(data.schema[col])}"
-                                )
+                        if column_types is None:
+                            for col in current_file_cols:
+                                if str(data.schema[col]) != col_types[col]:
+                                    raise ValueError(
+                                        f"Type mismatch for column '{col}' in file '{file}'. "
+                                        f"Expected {col_types[col]}, got {str(data.schema[col])}"
+                                    )
 
                     if data_columns is None:
                         raise ValueError("data_columns is None")
@@ -967,7 +1203,9 @@ def _apply_mask_column(
     updates = []
     for col in data_columns:
         mask_value = (
-            SPECIAL_TOKEN_IDS.mask if col_types[col] == "Int64" else REAL_MASK_VALUE
+            SPECIAL_TOKEN_IDS.mask
+            if is_integer_dtype_name(col_types[col])
+            else REAL_MASK_VALUE
         )
         updates.append(
             pl.when(mask_expr)
@@ -993,6 +1231,8 @@ def _apply_column_statistics(
     col_types: Optional[dict[str, str]] = None,
 ) -> tuple[pl.DataFrame, dict[str, int], dict[str, str]]:
     """Apply categorical maps and numeric standardization."""
+    col_types_was_provided = col_types is not None
+
     if n_classes is None:
         n_classes = {col: max(id_maps[col].values()) + 1 for col in id_maps}
 
@@ -1013,7 +1253,8 @@ def _apply_column_statistics(
     for col in data_columns:
         if col in id_maps:
             data = data.with_columns(pl.col(col).replace(id_maps[col], default=1))
-            col_types[col] = "Int64"
+            if not col_types_was_provided:
+                col_types[col] = "Int64"
         elif col in selected_columns_statistics:
             data = data.with_columns(
                 (
@@ -1145,7 +1386,7 @@ def _get_column_statistics(
                 id_maps[data_col] = combine_maps(new_id_map, id_maps.get(data_col, {}))
             else:
                 logger.info(f"Applying precomputed map for {data_col}")
-        elif isinstance(dtype, (pl.Float32, pl.Float64)):
+        elif isinstance(dtype, (pl.Float16, pl.Float32, pl.Float64)):
             if data_col in precomputed_id_maps:
                 raise ValueError(
                     f"Column {data_col} is not categorical, precomputed map is invalid."
@@ -1380,6 +1621,7 @@ def _process_batches_multiple_files_inner(
                 max_rows_inner,
                 mask_column,
             )
+            data = _apply_configured_input_casting(data, data_columns, col_types)
             data, _, _ = _apply_column_statistics(
                 data,
                 data_columns,
@@ -1389,6 +1631,7 @@ def _process_batches_multiple_files_inner(
                 col_types,
             )
             data = _apply_mask_column(data, data_columns, col_types, mask_column)
+            data = _apply_output_type_casting(data, data_columns, col_types)
 
             data_name_root_inner = f"{data_name_root}-{process_id}-{file_index_str}"
 

From 94ab5fbaed5186fa4ccf8a254d20040845974429 Mon Sep 17 00:00:00 2001
From: Leon Luithlen <leontimnaluithlen@gmail.com>
Date: Tue, 23 Jun 2026 15:39:44 +0100
Subject: [PATCH 81/81] Add between sequences split

---
 src/sequifier/config/preprocess_config.py |  10 ++
 src/sequifier/helpers.py                  |  14 +++
 src/sequifier/make.py                     |   1 +
 src/sequifier/preprocess.py               | 130 +++++++++++++++++-----
 4 files changed, 126 insertions(+), 29 deletions(-)

diff --git a/src/sequifier/config/preprocess_config.py b/src/sequifier/config/preprocess_config.py
index 36d1d0f9..73d9a42c 100644
--- a/src/sequifier/config/preprocess_config.py
+++ b/src/sequifier/config/preprocess_config.py
@@ -47,6 +47,7 @@ class PreprocessorModel(BaseModel):
     column_types: Optional[dict[str, str]] = None
 
     split_ratios: list[float]
+    split_method: str = Field(default="within_sequence")
     stored_context_width: int = Field(gt=0)
     max_target_offset: int = Field(default=1, ge=0)
     stride_by_split: Optional[list[int]] = None
@@ -111,6 +112,15 @@ def validate_proportions_sum(cls, v: list[float]) -> list[float]:
             raise ValueError(f"All split_ratios must be positive: {v}")
         return v
 
+    @field_validator("split_method")
+    @classmethod
+    def validate_split_method(cls, v: str) -> str:
+        if v not in ["within_sequence", "between_sequence"]:
+            raise ValueError(
+                "split_method must be one of 'within_sequence', 'between_sequence'"
+            )
+        return v
+
     @field_validator("stride_by_split")
     @classmethod
     def validate_step_sizes(
diff --git a/src/sequifier/helpers.py b/src/sequifier/helpers.py
index b2a846b0..c70b60f8 100644
--- a/src/sequifier/helpers.py
+++ b/src/sequifier/helpers.py
@@ -1,4 +1,5 @@
 import glob
+import hashlib
 import math
 import os
 import random
@@ -80,6 +81,19 @@ def polars_dtype_from_name(dtype_name: str) -> Any:
     return POLARS_NUMERIC_DTYPES[canonicalize_polars_dtype_name(dtype_name)]
 
 
+@beartype
+def assign_sequence_to_split(
+    sequence_id: int, split_ratios: list[float], seed: int
+) -> int:
+    """Deterministically assign one sequenceId to a split index."""
+    digest = hashlib.sha256(f"{seed}:{sequence_id}".encode("utf-8")).digest()
+    hash_value = int.from_bytes(digest[:8], byteorder="big", signed=False) / 2**64
+    split_index = int(
+        np.searchsorted(np.cumsum(split_ratios), hash_value, side="right")
+    )
+    return min(split_index, len(split_ratios) - 1)
+
+
 @beartype
 def is_float_dtype_name(dtype_name: str) -> bool:
     return canonicalize_polars_dtype_name(dtype_name).startswith("Float")
diff --git a/src/sequifier/make.py b/src/sequifier/make.py
index ef87be45..938f82ef 100644
--- a/src/sequifier/make.py
+++ b/src/sequifier/make.py
@@ -12,6 +12,7 @@
 - 0.8
 - 0.1
 - 0.1
+split_method: within_sequence # one of within_sequence, between_sequence
 stored_context_width: 49
 max_target_offset: 1
 max_rows: null
diff --git a/src/sequifier/preprocess.py b/src/sequifier/preprocess.py
index 09fa568f..3ad7b9bb 100644
--- a/src/sequifier/preprocess.py
+++ b/src/sequifier/preprocess.py
@@ -20,6 +20,7 @@
 from sequifier.helpers import (
     PANDAS_TO_TORCH_TYPES,
     StoredWindowLayout,
+    assign_sequence_to_split,
     canonicalize_polars_dtype_name,
     is_float_dtype_name,
     is_integer_dtype_name,
@@ -309,6 +310,7 @@ def __init__(
         max_target_offset: int = 1,
         mask_column: Optional[str] = None,
         column_types: Optional[dict[str, str]] = None,
+        split_method: str = "within_sequence",
     ):
         """Initialize and run preprocessing from validated config fields."""
         self.project_root = project_root
@@ -333,6 +335,11 @@ def __init__(
         self.use_precomputed_maps = use_precomputed_maps
         self.metadata_config_path = metadata_config_path
         self.mask_column = mask_column
+        if split_method not in ["within_sequence", "between_sequence"]:
+            raise ValueError(
+                "split_method must be one of 'within_sequence', 'between_sequence'"
+            )
+        self.split_method = split_method
         self.split_ratios = split_ratios
         self.stride_by_split = stride_by_split
         self.max_rows = max_rows
@@ -484,6 +491,8 @@ def __init__(
                 subsequence_start_mode,
                 self.merge_output,
                 self.allow_sequence_splitting,
+                self.split_method,
+                self.seed,
             )
 
             if self.merge_output:
@@ -834,6 +843,8 @@ def _process_batches_multiple_files(
                 continue_preprocessing=self.continue_preprocessing,
                 subsequence_start_mode=subsequence_start_mode,
                 mask_column=mask_column,
+                split_method=self.split_method,
+                seed=self.seed,
             )
             input_files = create_file_paths_for_multiple_files2(
                 self.project_root,
@@ -880,6 +891,8 @@ def _process_batches_multiple_files(
                 "continue_preprocessing": self.continue_preprocessing,
                 "subsequence_start_mode": subsequence_start_mode,
                 "mask_column": mask_column,
+                "split_method": self.split_method,
+                "seed": self.seed,
             }
 
             job_params = [
@@ -988,6 +1001,8 @@ def _write_or_validate_resume_manifest(
                 "selected_columns": selected_columns,
                 "data_columns": data_columns,
                 "split_ratios": self.split_ratios,
+                "split_method": self.split_method,
+                "seed": self.seed,
                 "stride_by_split": self.stride_by_split,
                 "max_rows": self.max_rows,
                 "process_by_file": self.process_by_file,
@@ -1561,6 +1576,8 @@ def _process_batches_multiple_files_inner(
     continue_preprocessing: bool,
     subsequence_start_mode: str,
     mask_column: Optional[str],
+    split_method: str,
+    seed: int,
 ):
     """Process this worker's file shard."""
 
@@ -1653,6 +1670,8 @@ def _process_batches_multiple_files_inner(
                 subsequence_start_mode,
                 merge_output,
                 allow_sequence_splitting,
+                split_method,
+                seed,
             )
 
             if merge_output:
@@ -1701,6 +1720,8 @@ def _process_batches_single_file(
     subsequence_start_mode: str,
     merge_output: bool,
     allow_sequence_splitting: bool,
+    split_method: str = "within_sequence",
+    seed: int = 1010,
 ) -> int:
     """Split one file into worker batches and preprocess them."""
     n_cores = n_cores or multiprocessing.cpu_count()
@@ -1724,6 +1745,8 @@ def _process_batches_single_file(
             batches_per_file,
             subsequence_start_mode,
             merge_output,
+            split_method,
+            seed,
         )
         for process_id, (start, end) in enumerate(valid_batch_limits)
     ]
@@ -2022,6 +2045,56 @@ def _write_accumulated_sequences(
         combined_df.write_parquet(out_path)
 
 
+@beartype
+def _extract_sequences_for_splits(
+    data_subset: pl.DataFrame,
+    sequence_id: int,
+    schema: Any,
+    layout: StoredWindowLayout,
+    stride_by_split: list[int],
+    data_columns: list[str],
+    split_ratios: list[float],
+    subsequence_start_mode: str,
+    split_method: str,
+    seed: int,
+) -> dict[int, pl.DataFrame]:
+    """Return extracted windows for one sequence across configured splits."""
+    if split_method == "within_sequence":
+        group_bounds = get_group_bounds(data_subset, split_ratios)
+        return {
+            i: cast_columns_to_string(
+                extract_sequences(
+                    data_subset.slice(lb, ub - lb),
+                    schema,
+                    layout,
+                    stride_by_split[i],
+                    data_columns,
+                    subsequence_start_mode,
+                )
+            )
+            for i, (lb, ub) in enumerate(group_bounds)
+        }
+
+    if split_method == "between_sequence":
+        assigned_group = assign_sequence_to_split(sequence_id, split_ratios, seed)
+        sequences = {i: pl.DataFrame(schema=schema) for i in range(len(split_ratios))}
+        sequences[assigned_group] = cast_columns_to_string(
+            extract_sequences(
+                data_subset,
+                schema,
+                layout,
+                stride_by_split[assigned_group],
+                data_columns,
+                subsequence_start_mode,
+            )
+        )
+        return sequences
+
+    raise ValueError(
+        "split_method must be one of 'within_sequence', 'between_sequence'"
+    )
+
+
 @beartype
 def preprocess_batch(
     project_root: str,
@@ -2040,6 +2113,8 @@ def preprocess_batch(
     batches_per_file: int,
     subsequence_start_mode: str,
     merge_output: bool,
+    split_method: str = "within_sequence",
+    seed: int = 1010,
 ) -> None:
     """Extract and write all split windows for one batch."""
     sequence_ids = sorted(batch.get_column("sequenceId").unique().to_list())
@@ -2051,20 +2126,18 @@ def preprocess_batch(
         pad_width = len(str(math.ceil(len(sequence_ids) / batches_per_file) + 1))
         for i, sequence_id in enumerate(sequence_ids):
             data_subset = batch.filter(pl.col("sequenceId") == sequence_id)
-            group_bounds = get_group_bounds(data_subset, split_ratios)
-            sequences = {
-                i: cast_columns_to_string(
-                    extract_sequences(
-                        data_subset.slice(lb, ub - lb),
-                        schema,
-                        layout,
-                        stride_by_split[i],
-                        data_columns,
-                        subsequence_start_mode,
-                    )
-                )
-                for i, (lb, ub) in enumerate(group_bounds)
-            }
+            sequences = _extract_sequences_for_splits(
+                data_subset,
+                sequence_id,
+                schema,
+                layout,
+                stride_by_split,
+                data_columns,
+                split_ratios,
+                subsequence_start_mode,
+                split_method,
+                seed,
+            )
 
             for group, split_df in sequences.items():
                 if not split_df.is_empty():
@@ -2103,23 +2176,22 @@ def preprocess_batch(
         written_files: dict[int, list[str]] = {i: [] for i in range(len(split_paths))}
         for i, sequence_id in enumerate(sequence_ids):
             data_subset = batch.filter(pl.col("sequenceId") == sequence_id)
-            group_bounds = get_group_bounds(data_subset, split_ratios)
-            sequences = {
-                j: cast_columns_to_string(
-                    extract_sequences(
-                        data_subset.slice(lb, ub - lb),
-                        schema,
-                        layout,
-                        stride_by_split[j],
-                        data_columns,
-                        subsequence_start_mode,
-                    )
-                )
-                for j, (lb, ub) in enumerate(group_bounds)
-            }
+            sequences = _extract_sequences_for_splits(
+                data_subset,
+                sequence_id,
+                schema,
+                layout,
+                stride_by_split,
+                data_columns,
+                split_ratios,
+                subsequence_start_mode,
+                split_method,
+                seed,
+            )
             post_split_str = f"{process_id}-{i}"
 
-            for split_path, (group, split) in zip(split_paths, sequences.items()):
+            for group, split in sequences.items():
+                split_path = split_paths[group]
                 split_path_batch_seq = split_path.replace(
                     f".{write_format}", f"-{post_split_str}.{write_format}"
                 )