Update data load script to allow sharding.

edgar-corpus.py

@@ -88,9 +88,9 @@ class EdgarCorpus(datasets.GeneratorBasedBuilder):
         
         # Create the full path to the extracted files as it can be one or multiple
         filepaths = {
-            "test": {k: v for k, v in data_dir.items() if 'test' in k},
-            "train": {k: v for k, v in data_dir.items() if 'train' in k},
-            "validate": {k: v for k, v in data_dir.items() if 'validate' in k},
+            "test": [v for k, v in data_dir.items() if 'test' in k],
+            "train": [v for k, v in data_dir.items() if 'train' in k],
+            "validate": [v for k, v in data_dir.items() if 'validate' in k],
         }
         
         return [
@@ -122,7 +122,7 @@ class EdgarCorpus(datasets.GeneratorBasedBuilder):
 
     # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
     def _generate_examples(self, filepath, split):      
-        for _, path in filepath.items():            
+        for path in filepath:            
             with open(path, encoding="utf-8") as f:
                 for row in f:
                     data = json.loads(row)