dsa-helpers 1.1.5.dev2__tar.gz → 1.1.5.dev4__tar.gz

{dsa_helpers-1.1.5.dev2/dsa_helpers.egg-info → dsa_helpers-1.1.5.dev4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dsa_helpers
-Version: 1.1.5.dev2
+Version: 1.1.5.dev4
 Summary: Utility functions for working with the DSA girder client.
 Author-email: "Juan C. Vizcarra" <jvizcar@emory.edu>
 Maintainer-email: "Juan C. Vizcarra" <jvizcar@emory.edu>

{dsa_helpers-1.1.5.dev2 → dsa_helpers-1.1.5.dev4}/dsa_helpers/__init__.py

@@ -3,7 +3,7 @@ from .imread import imread
 from .imwrite import imwrite
 
 # Version of the dsa-helpers package
-__version__ = "1.1.5.dev2"
+__version__ = "1.1.5.dev4"
 
 # To avoid slow downs, do not allow from dsa_helpers import * to import anything.
 __all__ = []

{dsa_helpers-1.1.5.dev2 → dsa_helpers-1.1.5.dev4}/dsa_helpers/ml/datasets/SegFormerSegmentationDataset.py

@@ -30,4 +30,7 @@ class SegFormerSegmentationDataset(Dataset):
         mask_path = self.mask_files[idx]
 
         # Return image & mask as PIL images in a dictionary.
-        return {"pixel_values": Image.open(image_path), "label": Image.open(mask_path)}
+        return {
+            "pixel_values": Image.open(image_path),
+            "label": Image.open(mask_path),
+        }

{dsa_helpers-1.1.5.dev2 → dsa_helpers-1.1.5.dev4}/dsa_helpers/ml/datasets/utils.py

@@ -29,7 +29,7 @@ def create_segformer_segmentation_dataset(
     if isinstance(df, str):
         df = pd.read_csv(df, low_memory=low_memory)
 
-    dataset = SegFormerSegmentationDataset(df.copy())
+    dataset = SegFormerSegmentationDataset(df)
 
     dataset = Dataset.from_generator(
         generator=lambda: dataset_generator(dataset),

{dsa_helpers-1.1.5.dev2 → dsa_helpers-1.1.5.dev4}/dsa_helpers/ml/segformer_semantic_segmentation/train.py

@@ -71,6 +71,7 @@ def train_segformer_semantic_segmentation_model(
             will be used. Defaults to None. Note that there is a bug
             that this value must be kept short otherwise a cuda memory
             error will occur.
+
         random_state (int, optional): Random state for the validation
             split. Defaults to 42.
         tile_size (int, optional): Size of the tiles to use for the model.
@@ -90,6 +91,9 @@ def train_segformer_semantic_segmentation_model(
     if test_data is not None and isinstance(test_data, str):
         test_data = pd.read_csv(test_data)
 
+    train_data_copy = train_data.copy()
+    val_data_copy = val_data.copy()
+
     if device is None:
         device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
@@ -127,16 +131,16 @@ def train_segformer_semantic_segmentation_model(
 
     # Create datasets for the model.
     train_dataset = create_segformer_segmentation_dataset(
-        train_data.copy(), transforms=train_transforms
+        train_data_copy, transforms=train_transforms
     )
 
     val_dataset = create_segformer_segmentation_dataset(
         (
-            val_data.copy()
+            val_data_copy
             if max_val_size is None
-            else val_data.sample(
+            else val_data_copy.sample(
                 n=max_val_size, random_state=random_state
-            ).copy()
+            )
         ),
         transforms=val_transforms,
     )
@@ -243,7 +247,7 @@ def train_segformer_semantic_segmentation_model(
     model = trainer.model
 
     dataset = create_segformer_segmentation_dataset(
-        train_data.copy(), transforms=val_transforms
+        train_data, transforms=val_transforms
     )
     results["train"]["metrics"] = per_class_dice_on_dataset(
         model,
@@ -256,7 +260,7 @@ def train_segformer_semantic_segmentation_model(
     )
 
     dataset = create_segformer_segmentation_dataset(
-        val_data.copy(), transforms=val_transforms
+        val_data, transforms=val_transforms
     )
     results["val"]["metrics"] = per_class_dice_on_dataset(
         model,
@@ -272,7 +276,7 @@ def train_segformer_semantic_segmentation_model(
         results["test"] = {}
 
         dataset = create_segformer_segmentation_dataset(
-            test_data.copy(), transforms=val_transforms
+            test_data, transforms=val_transforms
         )
         results["test"]["metrics"] = per_class_dice_on_dataset(
             model,

{dsa_helpers-1.1.5.dev2 → dsa_helpers-1.1.5.dev4/dsa_helpers.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dsa_helpers
-Version: 1.1.5.dev2
+Version: 1.1.5.dev4
 Summary: Utility functions for working with the DSA girder client.
 Author-email: "Juan C. Vizcarra" <jvizcar@emory.edu>
 Maintainer-email: "Juan C. Vizcarra" <jvizcar@emory.edu>

{dsa_helpers-1.1.5.dev2 → dsa_helpers-1.1.5.dev4}/pyproject.toml

@@ -37,7 +37,7 @@ classifiers = [
     "License :: OSI Approved :: MIT License",
 ]
 keywords = ["dsa", "girder", "girder_client"]
-version = "v1.1.5.dev2"
+version = "v1.1.5.dev4"
 
 [project.urls]
 Repository = "https://github.com/Gutman-Lab/dsa-python-utils"