Skip to content

fix: fine tuning of models with non-default architecture #581

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
KRiedmiller merged 1 commit into from
Jun 19, 2026
Merged

fix: fine tuning of models with non-default architecture #581

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions panseg/tasks/training_tasks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ def unet_training_task(
max_num_iters: int,
dimensionality: Literal["2D", "3D"],
device: str,
layer_order: str,
modality: str = "",
output_type: str = "",
description: str = "",
Expand Down Expand Up @@ -100,6 +101,7 @@ def unet_training_task(
description=description,
resolution=resolution,
pre_trained=pre_trained,
layer_order=layer_order,
)
except RuntimeError as e:
if "Output size is too small" in str(e):
Expand Down
9 changes: 9 additions & 0 deletions panseg/viewer_napari/widgets/training.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -304,11 +304,19 @@ def factory_unet_training(
)
return

layer_order = "bcr"
pre_model_path = None
if pretrained is not None:
model, model_config, pre_model_path = model_zoo.get_model_by_name(
pretrained
)
pretrained_lo = model_config["layer_order"]
if pretrained_lo != layer_order:
logger.debug(
f"Pretrained model has non-default layer_order, changing from {layer_order}"
)
layer_order = pretrained_lo
logger.info(f"Model architecture: {layer_order}")

widgets_to_reset = [
self.widget_unet_training.pretrained,
Expand Down Expand Up @@ -342,6 +350,7 @@ def factory_unet_training(
"description": description,
"resolution": resolution,
"pre_trained": pre_model_path,
"layer_order": layer_order,
"widgets_to_reset": widgets_to_reset,
"_pbar": pbar,
"_to_hide": [self.widget_unet_training.call_button],
Expand Down
5 changes: 4 additions & 1 deletion tests/widgets/test_training_widget.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -417,10 +417,11 @@ def test_unet_training_feature_maps(training_tab, mocker, tmp_path):

def test_unet_training_pretrained(training_tab, mocker, tmp_path):
m_log = mocker.patch("panseg.viewer_napari.widgets.training.log")
m_logger = mocker.patch("panseg.viewer_napari.widgets.training.logger")
m_get_models = mocker.patch(
"panseg.viewer_napari.widgets.training.model_zoo.get_model_by_name"
)
m_get_models.return_value = [1, 2, mocker.sentinel]
m_get_models.return_value = ["model", {"layer_order": "gcr"}, mocker.sentinel]
m_schedule = mocker.patch("panseg.viewer_napari.widgets.training.schedule_task")
mocker.patch(
"panseg.viewer_napari.widgets.training.PATH_PANSEG_MODELS", new=tmp_path
Expand Down Expand Up @@ -448,6 +449,8 @@ def test_unet_training_pretrained(training_tab, mocker, tmp_path):
pbar=None,
)

m_logger.info.assert_called_once()
m_logger.debug.assert_called_once()
m_log.assert_called_once()
m_get_models.assert_called_with("SOMETHING")
m_schedule.assert_called_once()
Expand Down
Loading