Fix data to allow .shuffle()

NPSC.py

@@ -21,7 +21,6 @@ import json
 import tarfile
 import datasets
 from datasets.tasks import AutomaticSpeechRecognition
-from datasets.utils.streaming_download_manager import xopen
 
 
 _CITATION = """\
@@ -115,60 +114,50 @@ class Npsc(datasets.GeneratorBasedBuilder):
     def _split_generators(self, dl_manager):
         """Returns SplitGenerators."""
         data_urls = {}
-        metadata_urls = {}
         config_name = self.config.name
         for split in ["train", "eval", "test"]:
-            metadata_urls[split] = []
             data_urls[split] = []
             for shard in _SHARDS[split]:
-                metadata_urls[split] += [
-                    _METADATA_URL.format(split=split, shard=shard)
-                ]
-                data_urls[split] += [
+                data_urls[split] += [(
+                    _METADATA_URL.format(split=split, shard=shard),
                     _DATA_URL.format(split=split, shard=shard, config=config_name)
-                ]
-        train_downloaded_metadata = dl_manager.download(metadata_urls["train"])
-        validation_downloaded_metadata = dl_manager.download(metadata_urls["eval"])
-        test_downloaded_metadata = dl_manager.download(metadata_urls["test"])
-        train_downloaded_archives = dl_manager.download(data_urls["train"])
-        validation_downloaded_archives = dl_manager.download(data_urls["eval"])
-        test_downloaded_archives = dl_manager.download(data_urls["test"])
+                )]
+        train_downloaded_data = dl_manager.download(data_urls["train"])
+        validation_downloaded_data = dl_manager.download(data_urls["eval"])
+        test_downloaded_data = dl_manager.download(data_urls["test"])
 
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN, gen_kwargs={
-                    "archives": train_downloaded_archives,
-                    "metadata_paths": train_downloaded_metadata,
+                    "filepaths": train_downloaded_data,
                 }
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.VALIDATION, gen_kwargs={
-                    "archives": validation_downloaded_archives,
-                    "metadata_paths": validation_downloaded_metadata,
+                    "filepaths": validation_downloaded_data,
                 }
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.TEST, gen_kwargs={
-                    "archives": test_downloaded_archives,
-                    "metadata_paths": test_downloaded_metadata,
+                    "filepaths": test_downloaded_data,
                 }
             ),
         ]
 
-    def _generate_examples(self, archives, metadata_paths):
+    def _generate_examples(self, filepaths):
         """Yields examples."""
         data_fields = list(self._info().features.keys())
         data_fields.remove("audio")
-        for archive_path, metadata_path in zip(*[archives, metadata_paths]):
+        for metadata_path, archive_path in filepaths:
             metadata = {}
-            with xopen(metadata_path) as metadata_file:
+            with open(metadata_path) as metadata_file:
                 for line in metadata_file.read().split("\n"):
                     if line:
                         metadata_object = json.loads(line)
                         if "path" in metadata_object:
                             metadata_key = metadata_object["path"].split("/", 1)[-1]
                             metadata[metadata_key] = metadata_object
-            with xopen(archive_path, "rb") as archive_fs:
+            with open(archive_path, "rb") as archive_fs:
                 archive_bytes = io.BytesIO(archive_fs.read())
                 with tarfile.open(fileobj=archive_bytes, mode="r") as tar:
                     for audio_file in tar.getmembers():