allow subclasses in dataset wrappers (#7236)

Co-authored-by: Nicolas Hug <[email protected]>

Author: pmeier

test/datasets_utils.py

@@ -596,7 +596,7 @@ def test_transforms_v2_wrapper(self, config):
                 wrapped_sample = wrapped_dataset[0]
                 assert tree_any(lambda item: isinstance(item, (Datapoint, PIL.Image.Image)), wrapped_sample)
         except TypeError as error:
-            if str(error).startswith(f"No wrapper exist for dataset class {type(dataset).__name__}"):
+            if str(error).startswith(f"No wrapper exists for dataset class {type(dataset).__name__}"):
                 return
             raise error
         except RuntimeError as error:

test/test_prototype_datapoints.py

@@ -1,7 +1,11 @@
+import re
+
 import pytest
 import torch
 
 from PIL import Image
+
+from torchvision import datasets
 from torchvision.prototype import datapoints
 
 
@@ -159,3 +163,43 @@ def test_bbox_instance(data, format):
     if isinstance(format, str):
         format = datapoints.BoundingBoxFormat.from_str(format.upper())
     assert bboxes.format == format
+
+
+class TestDatasetWrapper:
+    def test_unknown_type(self):
+        unknown_object = object()
+        with pytest.raises(
+            TypeError, match=re.escape("is meant for subclasses of `torchvision.datasets.VisionDataset`")
+        ):
+            datapoints.wrap_dataset_for_transforms_v2(unknown_object)
+
+    def test_unknown_dataset(self):
+        class MyVisionDataset(datasets.VisionDataset):
+            pass
+
+        dataset = MyVisionDataset("root")
+
+        with pytest.raises(TypeError, match="No wrapper exist"):
+            datapoints.wrap_dataset_for_transforms_v2(dataset)
+
+    def test_missing_wrapper(self):
+        dataset = datasets.FakeData()
+
+        with pytest.raises(TypeError, match="please open an issue"):
+            datapoints.wrap_dataset_for_transforms_v2(dataset)
+
+    def test_subclass(self, mocker):
+        sentinel = object()
+        mocker.patch.dict(
+            datapoints._dataset_wrapper.WRAPPER_FACTORIES,
+            clear=False,
+            values={datasets.FakeData: lambda dataset: lambda idx, sample: sentinel},
+        )
+
+        class MyFakeData(datasets.FakeData):
+            pass
+
+        dataset = MyFakeData()
+        wrapped_dataset = datapoints.wrap_dataset_for_transforms_v2(dataset)
+
+        assert wrapped_dataset[0] is sentinel

torchvision/prototype/datapoints/_dataset_wrapper.py

@@ -39,16 +39,26 @@ def decorator(wrapper_factory):
 class VisionDatasetDatapointWrapper(Dataset):
     def __init__(self, dataset):
         dataset_cls = type(dataset)
-        wrapper_factory = WRAPPER_FACTORIES.get(dataset_cls)
-        if wrapper_factory is None:
-            # TODO: If we have documentation on how to do that, put a link in the error message.
-            msg = f"No wrapper exist for dataset class {dataset_cls.__name__}. Please wrap the output yourself."
-            if dataset_cls in datasets.__dict__.values():
-                msg = (
-                    f"{msg} If an automated wrapper for this dataset would be useful for you, "
-                    f"please open an issue at https://github.com/pytorch/vision/issues."
-                )
-            raise TypeError(msg)
+
+        if not isinstance(dataset, datasets.VisionDataset):
+            raise TypeError(
+                f"This wrapper is meant for subclasses of `torchvision.datasets.VisionDataset`, "
+                f"but got a '{dataset_cls.__name__}' instead."
+            )
+
+        for cls in dataset_cls.mro():
+            if cls in WRAPPER_FACTORIES:
+                wrapper_factory = WRAPPER_FACTORIES[cls]
+                break
+            elif cls is datasets.VisionDataset:
+                # TODO: If we have documentation on how to do that, put a link in the error message.
+                msg = f"No wrapper exists for dataset class {dataset_cls.__name__}. Please wrap the output yourself."
+                if dataset_cls in datasets.__dict__.values():
+                    msg = (
+                        f"{msg} If an automated wrapper for this dataset would be useful for you, "
+                        f"please open an issue at https://github.com/pytorch/vision/issues."
+                    )
+                raise TypeError(msg)
 
         self._dataset = dataset
         self._wrapper = wrapper_factory(dataset)
@@ -98,6 +108,13 @@ def identity(item):
     return item
 
 
+def identity_wrapper_factory(dataset):
+    def wrapper(idx, sample):
+        return sample
+
+    return wrapper
+
+
 def pil_image_to_mask(pil_image):
     return datapoints.Mask(pil_image)
 
@@ -125,10 +142,7 @@ def wrap_target_by_type(target, *, target_types, type_wrappers):
 
 
 def classification_wrapper_factory(dataset):
-    def wrapper(idx, sample):
-        return sample
-
-    return wrapper
+    return identity_wrapper_factory(dataset)
 
 
 for dataset_cls in [
@@ -237,6 +251,9 @@ def wrapper(idx, sample):
     return wrapper
 
 
+WRAPPER_FACTORIES.register(datasets.CocoCaptions)(identity_wrapper_factory)
+
+
 VOC_DETECTION_CATEGORIES = [
     "__background__",
     "aeroplane",