Automated Code Change

tensorflow_datasets/datasets/dsprites/dsprites_dataset_builder.py

@@ -62,7 +62,7 @@ def _split_generators(self, dl_manager):
     # There is no predefined train/val/test split for this dataset.
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN, gen_kwargs=dict(filepath=filepath)
+            name=tfds.Split.TRAIN, gen_kwargs=dict(filepath=filepath)  # pyrefly: ignore[missing-attribute]
         ),
     ]
 

tensorflow_datasets/datasets/e2e_cleaned/e2e_cleaned_dataset_builder.py

@@ -77,17 +77,17 @@ def _split_generators(self, dl_manager):
     )
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             # These kwargs will be passed to _generate_examples
             gen_kwargs={'csv_path': extracted_path['train_path']},
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             # These kwargs will be passed to _generate_examples
             gen_kwargs={'csv_path': extracted_path['dev_path']},
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             # These kwargs will be passed to _generate_examples
             gen_kwargs={'csv_path': extracted_path['test_path']},
         ),

tensorflow_datasets/datasets/efron_morris75/efron_morris75_dataset_builder.py

@@ -54,7 +54,7 @@ def _info(self) -> tfds.core.DatasetInfo:
   def _split_generators(self, dl_manager: tfds.download.DownloadManager):
     """Returns SplitGenerators."""
     path = dl_manager.download_and_extract(URL)
-    return {tfds.Split.TRAIN: self._generate_examples(path)}
+    return {tfds.Split.TRAIN: self._generate_examples(path)}  # pyrefly: ignore[missing-attribute]
 
   def _generate_examples(self, path):
     """Yields examples."""

tensorflow_datasets/datasets/flic/flic_dataset_builder.py

@@ -99,10 +99,10 @@ def _info(self):
 
   def _split_generators(self, dl_manager):
     """Returns SplitGenerators."""
-    extract_path = dl_manager.download_and_extract(self.builder_config.url)
+    extract_path = dl_manager.download_and_extract(self.builder_config.url)  # pyrefly: ignore[missing-attribute]
 
     mat_path = os.path.join(
-        extract_path, self.builder_config.dir, "examples.mat"
+        extract_path, self.builder_config.dir, "examples.mat"  # pyrefly: ignore[missing-attribute]
     )
     with tf.io.gfile.GFile(mat_path, "rb") as f:
       data = tfds.core.lazy_imports.scipy.io.loadmat(
@@ -111,15 +111,15 @@ def _split_generators(self, dl_manager):
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "extract_path": extract_path,
                 "data": data,
                 "selection_column": 7,  # indicates train split selection
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "extract_path": extract_path,
                 "data": data,
@@ -133,7 +133,7 @@ def _generate_examples(self, extract_path, data, selection_column):
     for u_id, example in enumerate(data["examples"]):
       if example[selection_column]:
         img_path = os.path.join(
-            extract_path, self.builder_config.dir, "images", example[3]
+            extract_path, self.builder_config.dir, "images", example[3]  # pyrefly: ignore[missing-attribute]
         )
         yield u_id, {
             "image": img_path,

tensorflow_datasets/datasets/groove/groove_dataset_builder.py

@@ -130,12 +130,12 @@ def _info(self):
         },
         "midi": tf.string,
     }
-    if self.builder_config.include_audio:
-      features_dict["audio"] = tfds.features.Audio(
-          dtype=np.float32, sample_rate=self.builder_config.audio_rate
+    if self.builder_config.include_audio:  # pyrefly: ignore[missing-attribute]
+      features_dict["audio"] = tfds.features.Audio(  # pyrefly: ignore[bad-assignment]
+          dtype=np.float32, sample_rate=self.builder_config.audio_rate  # pyrefly: ignore[missing-attribute]
       )
     return self.dataset_info_from_configs(
-        features=tfds.features.FeaturesDict(features_dict),
+        features=tfds.features.FeaturesDict(features_dict),  # pyrefly: ignore[bad-argument-type]
         homepage="https://g.co/magenta/groove-dataset",
     )
 

tensorflow_datasets/datasets/imagenet2012/imagenet2012_dataset_builder.py

@@ -101,20 +101,20 @@ def _split_generators(self, dl_manager):
     splits = []
     _add_split_if_exists(
         split_list=splits,
-        split=tfds.Split.TRAIN,
+        split=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
         split_path=train_path,
         dl_manager=dl_manager,
     )
     _add_split_if_exists(
         split_list=splits,
-        split=tfds.Split.VALIDATION,
+        split=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
         split_path=val_path,
         dl_manager=dl_manager,
         validation_labels=imagenet_common.get_validation_labels(val_path),
     )
     _add_split_if_exists(
         split_list=splits,
-        split=tfds.Split.TEST,
+        split=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
         split_path=test_path,
         dl_manager=dl_manager,
         labels_exist=False,

tensorflow_datasets/datasets/imagenet2012_corrupted/imagenet2012_corrupted_dataset_builder.py

@@ -168,7 +168,7 @@ def _split_generators(self, dl_manager):
     splits = super(Builder, self)._split_generators(dl_manager)
 
     corruptions.FROST_FILENAMES = dl_manager.download(_FROST_FILENAMES)
-    return [s for s in splits if s.name != tfds.Split.TRAIN]
+    return [s for s in splits if s.name != tfds.Split.TRAIN]  # pyrefly: ignore[missing-attribute]
 
   def _generate_examples(
       self, archive, validation_labels=None, labels_exist=None
@@ -215,8 +215,8 @@ def _get_corrupted_example(self, x):
     Returns:
       numpy array, corrupted images.
     """
-    corruption_type = self.builder_config.corruption_type
-    severity = self.builder_config.severity
+    corruption_type = self.builder_config.corruption_type  # pyrefly: ignore[missing-attribute]
+    severity = self.builder_config.severity  # pyrefly: ignore[missing-attribute]
     x = np.clip(x, 0, 255)
 
     return {

tensorflow_datasets/datasets/imagenet2012_fewshot/imagenet2012_fewshot_dataset_builder.py

@@ -68,7 +68,7 @@ def _split_generators(self, dl_manager):
       )
 
     # Download and load subset file.
-    subset_file = SUBSET2FILES[self.builder_config.name]
+    subset_file = SUBSET2FILES[self.builder_config.name]  # pyrefly: ignore[missing-attribute]
     if isinstance(subset_file, list):  # it will only be a list during testing,
       subset_file = subset_file[0]  # where the first entry is 1shot.txt.
     subset = set(subset_file.read_text().splitlines())
@@ -77,13 +77,13 @@ def _split_generators(self, dl_manager):
     tuneset = set(TUNE_FILE.read_text().splitlines())
 
     return {
-        tfds.Split.TRAIN: self._generate_examples(
+        tfds.Split.TRAIN: self._generate_examples(  # pyrefly: ignore[missing-attribute]
             archive=dl_manager.iter_archive(train_path), subset=subset
         ),
         tfds.Split('tune'): self._generate_examples(
             archive=dl_manager.iter_archive(train_path), subset=tuneset
         ),
-        tfds.Split.VALIDATION: self._generate_examples(
+        tfds.Split.VALIDATION: self._generate_examples(  # pyrefly: ignore[missing-attribute]
             archive=dl_manager.iter_archive(val_path),
             validation_labels=imagenet_common.get_validation_labels(val_path),
         ),

tensorflow_datasets/datasets/imagenet2012_real/imagenet2012_real_dataset_builder.py

@@ -72,7 +72,7 @@ def _split_generators(self, dl_manager):
       )
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 'archive': dl_manager.iter_archive(val_path),
                 'original_labels': imagenet_common.get_validation_labels(

tensorflow_datasets/datasets/imagenet2012_subset/imagenet2012_subset_dataset_builder.py

@@ -72,22 +72,22 @@ def _split_generators(self, dl_manager):
       )
 
     # Download and load subset file.
-    subset_file = dl_manager.download(SUBSET2FILES[self.builder_config.name])
+    subset_file = dl_manager.download(SUBSET2FILES[self.builder_config.name])  # pyrefly: ignore[missing-attribute]
     if isinstance(subset_file, list):  # it will only be a list during testing,
       subset_file = subset_file[0]  # where the first entry is 1percent.txt.
     with epath.Path(subset_file).open() as fp:
       subset = set(fp.read().splitlines())  # remove trailing `\r` in Windows
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 'archive': dl_manager.iter_archive(train_path),
                 'subset': subset,
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 'archive': dl_manager.iter_archive(val_path),
                 'validation_labels': imagenet_common.get_validation_labels(

tensorflow_datasets/datasets/imagenet_a/imagenet_a_dataset_builder.py

@@ -51,7 +51,7 @@ def _split_generators(self, dl_manager):
     return [
         tfds.core.SplitGenerator(
             # The dataset provides only a test split.
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             # These kwargs will be passed to _generate_examples
             gen_kwargs={'imagenet_a_root': imagenet_a_root},
         ),

tensorflow_datasets/datasets/imagenet_r/imagenet_r_dataset_builder.py

@@ -60,7 +60,7 @@ def _split_generators(self, dl_manager):
     return [
         tfds.core.SplitGenerator(
             # The dataset provides only a test split.
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             # These kwargs will be passed to _generate_examples
             gen_kwargs={'imagenet_r_root': imagenet_r_root},
         ),

tensorflow_datasets/datasets/imagenet_resized/imagenet_resized_dataset_builder.py

@@ -64,7 +64,7 @@ class Builder(tfds.core.GeneratorBasedBuilder):
 
   def _info(self):
     names_file = tfds.core.tfds_path(_LABELS_FNAME)
-    size = self.builder_config.size
+    size = self.builder_config.size  # pyrefly: ignore[missing-attribute]
     return self.dataset_info_from_configs(
         features=tfds.features.FeaturesDict({
             'image': tfds.features.Image(shape=(size, size, 3)),
@@ -75,7 +75,7 @@ def _info(self):
     )
 
   def _split_generators(self, dl_manager):
-    size = self.builder_config.size
+    size = self.builder_config.size  # pyrefly: ignore[missing-attribute]
 
     if size in [8, 16, 32]:
       train_path, val_path = dl_manager.download([
@@ -96,7 +96,7 @@ def _split_generators(self, dl_manager):
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 'archive': itertools.chain(
                     *[
@@ -107,7 +107,7 @@ def _split_generators(self, dl_manager):
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 'archive': dl_manager.iter_archive(val_path),
             },
@@ -121,7 +121,7 @@ def _generate_examples(self, archive):
       if content:
         fobj_mem = io.BytesIO(content)
         data = np.load(fobj_mem, allow_pickle=False)
-        size = self.builder_config.size
+        size = self.builder_config.size  # pyrefly: ignore[missing-attribute]
         for i, (image, label) in enumerate(zip(data['data'], data['labels'])):
           record = {
               # The data is packed flat as CHW where as most image datasets

tensorflow_datasets/datasets/imagenet_sketch/imagenet_sketch_dataset_builder.py

@@ -48,7 +48,7 @@ def _split_generators(self, dl_manager: tfds.download.DownloadManager):
     path = dl_manager.download(_IMAGENET_SKETCH_URL)
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 'archive': dl_manager.iter_archive(path),
             },

tensorflow_datasets/datasets/imagenet_v2/imagenet_v2_dataset_builder.py

@@ -100,15 +100,15 @@ def _info(self):
 
   def _split_generators(self, dl_manager):
     """Returns a SplitGenerator for the test set."""
-    variant_url = _IMAGENET_V2_URLS[self.builder_config.variant]
+    variant_url = _IMAGENET_V2_URLS[self.builder_config.variant]  # pyrefly: ignore[missing-attribute]
     imagenet_v2_root = os.path.join(
         dl_manager.download_and_extract(variant_url),
-        _TAR_TOPDIR[self.builder_config.variant],
+        _TAR_TOPDIR[self.builder_config.variant],  # pyrefly: ignore[missing-attribute]
     )
     return [
         tfds.core.SplitGenerator(
             # The dataset provides only a test split.
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             # These kwargs will be passed to _generate_examples
             gen_kwargs={'imagenet_v2_root': imagenet_v2_root},
         ),

tensorflow_datasets/datasets/imagenette/imagenette_dataset_builder.py

@@ -77,21 +77,21 @@ def _info(self):
 
   def _split_generators(self, dl_manager):
     """Returns SplitGenerators."""
-    dirname = self.builder_config.dirname
+    dirname = self.builder_config.dirname  # pyrefly: ignore[missing-attribute]
     url = _URL_PREFIX + "{}.tgz".format(dirname)
     path = dl_manager.download_and_extract(url)
     train_path = os.path.join(path, dirname, "train")
     val_path = os.path.join(path, dirname, "val")
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "datapath": train_path,
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "datapath": val_path,
             },

tensorflow_datasets/datasets/imagewang/imagewang_dataset_builder.py

@@ -73,7 +73,7 @@ def _info(self):
 
   def _split_generators(self, dl_manager):
     """Returns SplitGenerators."""
-    size = self.builder_config.size
+    size = self.builder_config.size  # pyrefly: ignore[missing-attribute]
     if size in _SIZES:
       size_str = "" if size == "full-size" else "-" + size[:-2]
       url = "/".join([_URL_PREFIX, "imagewang%s.tgz" % size_str])
@@ -85,13 +85,13 @@ def _split_generators(self, dl_manager):
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "datapath": train_path,
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "datapath": val_path,
             },

tensorflow_datasets/datasets/imdb_reviews/imdb_reviews_dataset_builder.py

@@ -61,7 +61,7 @@ def _info(self):
     return self.dataset_info_from_configs(
         features=tfds.features.FeaturesDict({
             "text": tfds.features.Text(
-                encoder_config=self.builder_config.text_encoder_config
+                encoder_config=self.builder_config.text_encoder_config  # pyrefly: ignore[missing-attribute]
             ),
             "label": tfds.features.ClassLabel(names=["neg", "pos"]),
         }),
@@ -86,14 +86,14 @@ def _split_generators(self, dl_manager):
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "archive": archive(),
                 "directory": os.path.join("aclImdb", "train"),
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "archive": archive(),
                 "directory": os.path.join("aclImdb", "test"),

tensorflow_datasets/datasets/irc_disentanglement/irc_disentanglement_dataset_builder.py

@@ -167,21 +167,21 @@ def _split_generators(
 
     return [
         tfds.core.SplitGenerator(
-            name=tfds.Split.TRAIN,
+            name=tfds.Split.TRAIN,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "day_to_paths": _get_day_to_paths(
                     os.path.join(data_dir, "train")
                 )
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.VALIDATION,
+            name=tfds.Split.VALIDATION,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "day_to_paths": _get_day_to_paths(os.path.join(data_dir, "dev"))
             },
         ),
         tfds.core.SplitGenerator(
-            name=tfds.Split.TEST,
+            name=tfds.Split.TEST,  # pyrefly: ignore[missing-attribute]
             gen_kwargs={
                 "day_to_paths": _get_day_to_paths(
                     os.path.join(data_dir, "test")