Builder created

py_legislation.py

@@ -282,17 +282,17 @@ class PYLegislation(datasets.GeneratorBasedBuilder):
 
     def _split_generators(self, dl_manager):
         # TODO: labeled subset has two splits
-        generators = []
+        # generators = []
 
-        urls_to_download = _URLS
-        # downloaded_files = dl_manager.download_and_extract(urls_to_download)
+        urls_to_download = _URLS[self.config.name]
+        filepaths = dl_manager.download_and_extract(urls_to_download)
 
-        for k in urls_to_download:
-            fp = dl_manager.download_and_extract(urls_to_download[k])
-            print(f"***************>>>> filepath = {fp}")
-            generators.append(
-                datasets.SplitGenerator(
-                    name=datasets.Split.TRAIN, gen_kwargs={"filepath": fp}))
+        # for k in urls_to_download:
+        #     fp = dl_manager.download_and_extract(urls_to_download[k])
+        #     print(f"***************>>>> filepath = {fp}")
+        #     generators.append(
+        #         datasets.SplitGenerator(
+        #             name=datasets.Split.TRAIN, gen_kwargs={"filepath": fp}))
 
         # us_train = dl_manager.download_and_extract(
         #     urls_to_download["unlabeled_sentences"])
@@ -306,31 +306,29 @@ class PYLegislation(datasets.GeneratorBasedBuilder):
         # generators.append(
         #    datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": us_train}))
 
-            # datasets.SplitGenerator(
-            #     name=datasets.Split.TRAIN,
-            #     gen_kwargs={
-            #         # downloaded_files["labeled_sentences_train"],
-            #         "filepath": ls_train
-            #     },
-            # ),
-
-            # datasets.SplitGenerator(
-            #     name=datasets.Split.TEST,
-            #     gen_kwargs={
-            #         # downloaded_files["labeled_sentences_test"],
-            #         "filepath": ls_test
-            #     },
-            # )
+        # datasets.SplitGenerator(
+        #     name=datasets.Split.TRAIN,
+        #     gen_kwargs={
+        #         # downloaded_files["labeled_sentences_train"],
+        #         "filepath": ls_train
+        #     },
+        # ),
+
+        # datasets.SplitGenerator(
+        #     name=datasets.Split.TEST,
+        #     gen_kwargs={
+        #         # downloaded_files["labeled_sentences_test"],
+        #         "filepath": ls_test
+        #     },
+        # )
         # ]
 
         # return generators
 
-        return generators
-
-        # datasets.SplitGenerator(
-        #     name=datasets.Split.TRAIN,
-        #     gen_kwargs={"filepath": downloaded_files},
-        # )
+        return datasets.SplitGenerator(
+            name=datasets.Split.TRAIN,
+            gen_kwargs={"filepath": filepaths},
+        )
 
     def _generate_examples(self, filepath):
         """