MovingMNIST split fix (#7449)

Co-authored-by: Philip Meier <github.pmeier@posteo.de>

test/test_datasets.py

@@ -1504,14 +1504,16 @@ class MovingMNISTTestCase(datasets_utils.DatasetTestCase):
 
     ADDITIONAL_CONFIGS = combinations_grid(split=(None, "train", "test"), split_ratio=(10, 1, 19))
 
+    _NUM_FRAMES = 20
+
     def inject_fake_data(self, tmpdir, config):
         base_folder = os.path.join(tmpdir, self.DATASET_CLASS.__name__)
         os.makedirs(base_folder, exist_ok=True)
-        num_samples = 20
+        num_samples = 5
         data = np.concatenate(
             [
                 np.zeros((config["split_ratio"], num_samples, 64, 64)),
-                np.ones((20 - config["split_ratio"], num_samples, 64, 64)),
+                np.ones((self._NUM_FRAMES - config["split_ratio"], num_samples, 64, 64)),
             ]
         )
         np.save(os.path.join(base_folder, "mnist_test_seq.npy"), data)
@@ -1519,14 +1521,13 @@ class MovingMNISTTestCase(datasets_utils.DatasetTestCase):
 
     @datasets_utils.test_all_configs
     def test_split(self, config):
-        if config["split"] is None:
-            return
-
-        with self.create_dataset(config) as (dataset, info):
+        with self.create_dataset(config) as (dataset, _):
             if config["split"] == "train":
                 assert (dataset.data == 0).all()
-            else:
+            elif config["split"] == "test":
                 assert (dataset.data == 1).all()
+            else:
+                assert dataset.data.size()[1] == self._NUM_FRAMES
 
 
 class DatasetFolderTestCase(datasets_utils.ImageDatasetTestCase):

torchvision/datasets/moving_mnist.py

@@ -58,7 +58,7 @@ class MovingMNIST(VisionDataset):
         data = torch.from_numpy(np.load(os.path.join(self._base_folder, self._filename)))
         if self.split == "train":
             data = data[: self.split_ratio]
-        else:
+        elif self.split == "test":
             data = data[self.split_ratio :]
         self.data = data.transpose(0, 1).unsqueeze(2).contiguous()