⚡ Make WSIPatchDataset Pickleable to Support Windows Multithreading (#947)

This PR makes the WSIPatchDataset class picklable by delaying the creation of the reader object until the first call to `__getitem__`. This enables the use of multiple loader workers on Windows without errors and provides significant performance improvements.

- Delays reader object instantiation to the first `__getitem__` call instead of during initialization
- Extracts reader creation logic into a separate `_get_reader` method
- Stores image path and mode as instance variables for lazy initialization

Speedup for the WSI prediction cell of the patch_prediction example notebook: 
2min 48 sec with 0 loader workers -> 1min 13 sec with 4 workers.

Note: this PR doesn't have any effect for Linux as the multi-threading already works fine there because Linux multithreading doesn't require things to be pickleable

Author: measty

tiatoolbox/models/dataset/classification.py

@@ -163,7 +163,7 @@ class WSIPatchDataset(dataset_abc.PatchDatasetABC):
 
     """
 
-    def __init__(  # skipcq: PY-R1000  # noqa: PLR0915
+    def __init__(  # skipcq: PY-R1000
         self: WSIPatchDataset,
         img_path: str | Path,
         mode: str = "wsi",
@@ -262,40 +262,17 @@ def __init__(  # skipcq: PY-R1000  # noqa: PLR0915
             raise ValueError(msg)
 
         self.preproc_func = preproc_func
-        img_path = Path(img_path)
-        if mode == "wsi":
-            self.reader = WSIReader.open(img_path)
-        else:
-            logger.warning(
-                "WSIPatchDataset only reads image tile at "
-                '`units="baseline"` and `resolution=1.0`.',
-                stacklevel=2,
-            )
-            img = imread(img_path)
-            axes = "YXS"[: len(img.shape)]
-            # initialise metadata for VirtualWSIReader.
-            # here, we simulate a whole-slide image, but with a single level.
-            # ! should we expose this so that use can provide their metadata ?
-            metadata = WSIMeta(
-                mpp=np.array([1.0, 1.0]),
-                axes=axes,
-                objective_power=10,
-                slide_dimensions=np.array(img.shape[:2][::-1]),
-                level_downsamples=[1.0],
-                level_dimensions=[np.array(img.shape[:2][::-1])],
-            )
-            # infer value such that read if mask provided is through
-            # 'mpp' or 'power' as varying 'baseline' is locked atm
+        self.img_path = Path(img_path)
+        self.mode = mode
+        self.reader = None
+        reader = self._get_reader(self.img_path)
+        if mode != "wsi":
             units = "mpp"
             resolution = 1.0
-            self.reader = VirtualWSIReader(
-                img,
-                info=metadata,
-            )
 
         # may decouple into misc ?
         # the scaling factor will scale base level to requested read resolution/units
-        wsi_shape = self.reader.slide_dimensions(resolution=resolution, units=units)
+        wsi_shape = reader.slide_dimensions(resolution=resolution, units=units)
 
         # use all patches, as long as it overlaps source image
         self.inputs = PatchExtractor.get_coordinates(
@@ -316,13 +293,13 @@ def __init__(  # skipcq: PY-R1000  # noqa: PLR0915
             mask = np.array(mask > 0, dtype=np.uint8)
 
             mask_reader = VirtualWSIReader(mask)
-            mask_reader.info = self.reader.info
+            mask_reader.info = reader.info
         elif auto_get_mask and mode == "wsi" and mask_path is None:
             # if no mask provided and `wsi` mode, generate basic tissue
             # mask on the fly
-            mask_reader = self.reader.tissue_mask(resolution=1.25, units="power")
+            mask_reader = reader.tissue_mask(resolution=1.25, units="power")
             # ? will this mess up  ?
-            mask_reader.info = self.reader.info
+            mask_reader.info = reader.info
 
         if mask_reader is not None:
             selected = PatchExtractor.filter_coordinates(
@@ -344,10 +321,44 @@ def __init__(  # skipcq: PY-R1000  # noqa: PLR0915
         # Perform check on the input
         self._check_input_integrity(mode="wsi")
 
+    def _get_reader(self: WSIPatchDataset, img_path: str | Path) -> WSIReader:
+        """Get a reader for the image."""
+        if self.mode == "wsi":
+            reader = WSIReader.open(img_path)
+        else:
+            logger.warning(
+                "WSIPatchDataset only reads image tile at "
+                '`units="baseline"` and `resolution=1.0`.',
+                stacklevel=2,
+            )
+            img = imread(img_path)
+            axes = "YXS"[: len(img.shape)]
+            # initialise metadata for VirtualWSIReader.
+            # here, we simulate a whole-slide image, but with a single level.
+            # ! should we expose this so that use can provide their metadata ?
+            metadata = WSIMeta(
+                mpp=np.array([1.0, 1.0]),
+                axes=axes,
+                objective_power=10,
+                slide_dimensions=np.array(img.shape[:2][::-1]),
+                level_downsamples=[1.0],
+                level_dimensions=[np.array(img.shape[:2][::-1])],
+            )
+            # infer value such that read if mask provided is through
+            # 'mpp' or 'power' as varying 'baseline' is locked atm
+            reader = VirtualWSIReader(
+                img,
+                info=metadata,
+            )
+        return reader
+
     def __getitem__(self: WSIPatchDataset, idx: int) -> dict:
         """Get an item from the dataset."""
         coords = self.inputs[idx]
         # Read image patch from the whole-slide image
+        if self.reader is None:
+            # only set the reader on first call so that it is initially picklable
+            self.reader = self._get_reader(self.img_path)
         patch = self.reader.read_bounds(
             coords,
             resolution=self.resolution,