preprocessed images as on device

model-training/configs/yolo11n_bike_with_augmentation_custom_loss_preprocessed.yaml

@@ -0,0 +1,44 @@
+---
+project_name: bike_aug_custom_loss_preprocessed_dataset
+output_dir: runs/
+model: models/yolo11n.pt
+
+# for a full list of training and augmentation arguments see:
+# https://docs.ultralytics.com/modes/train/#train-settings
+train_args:
+  mode: train
+  task: detect
+  # yamllint disable-line rule:line-length
+  data: ./datasets/combined_preprocessed/YOLO/combined_preprocessed_yolo_dataset.yaml
+  epochs: 300
+  patience: 200
+  save: true
+  val: true
+  plots: true
+  rect: false
+  pretrained: true
+  cache: false
+  verbose: true
+  device: cuda
+  batch: 16
+  nbs: 16
+  multi_scale: true
+  workers: 4
+  amp: true
+  dropout: 0.1
+  classes:
+    - 0
+  # Loss function parameters
+  dfl: 2.0
+  kobj: 3.0
+  box: 5.0
+  cls: 0.3
+  # Augmentation Parameters
+  hsv_s: 0.7    # hue
+  hsv_h: 0.015  # saturation
+  hsv_v: 0.5    # brightness
+  degrees: 180
+  scale: 0.1
+  flipud: 0.1
+  fliplr: 0.5
+  mosaic: 0.0

model-training/datasets.dvc

@@ -1,6 +1,6 @@
 outs:
-- md5: 4b3a23bc5a894228cbb81bc71150732d.dir
-  size: 310669866
-  nfiles: 2084
+- md5: c16bc40755d3b2237e70d6fc6cf9b85d.dir
+  size: 346598894
+  nfiles: 2764
   hash: md5
   path: datasets