resolving card bugs

py_legislation.py

@@ -276,16 +276,18 @@ class PYLegislation(datasets.GeneratorBasedBuilder):
 
     def _split_generators(self, dl_manager):
         # TODO: labeled subset has two splits
-        urls_to_download = _URLS
-
-        # root = "fernandoperes/py_legislation/"
+        generators = []
 
-        downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        urls_to_download = _URLS
+        # downloaded_files = dl_manager.download_and_extract(urls_to_download)
 
-        # rt_train = dl_manager.download_and_extract(
-        #     urls_to_download["raw_text"])
+        for k in urls_to_download:
+            fp = dl_manager.download_and_extract(urls_to_download[k])
+            generators.append(
+                datasets.SplitGenerator(
+                    name=datasets.Split.TRAIN,gen_kwargs={"filepath": fp}))
 
-        # us_train = dl_manager.download_and_extract(
+        #us_train = dl_manager.download_and_extract(
         #     urls_to_download["unlabeled_sentences"])
 
         # ls_train = dl_manager.download_and_extract(
@@ -294,21 +296,11 @@ class PYLegislation(datasets.GeneratorBasedBuilder):
         # ls_test = dl_manager.download_and_extract(
         #     urls_to_download["labeled_sentences_test"])
 
-        # generators = [
-        #     datasets.SplitGenerator(
-        #         name=datasets.Split.TRAIN,
-        #         gen_kwargs={
-        #             "filepath": rt_train
-        #         },
-        #     ),
-
-        #     datasets.SplitGenerator(
-        #         name=datasets.Split.TRAIN,
-        #         gen_kwargs={
-        #             # downloaded_files["unlabeled_sentences"],
-        #             "filepath": us_train
-        #         }
-        #     ),
+
+
+
+        #generators.append(
+        #    datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": us_train}))
 
             # datasets.SplitGenerator(
             #     name=datasets.Split.TRAIN,
@@ -329,10 +321,12 @@ class PYLegislation(datasets.GeneratorBasedBuilder):
 
         # return generators
 
-        return datasets.SplitGenerator(
-            name=datasets.Split.TRAIN,
-            gen_kwargs={"filepath": downloaded_files},
-        )
+        return generators
+
+        # datasets.SplitGenerator(
+        #     name=datasets.Split.TRAIN,
+        #     gen_kwargs={"filepath": downloaded_files},
+        # )
 
     def _generate_examples(self, filepath):
         """