modify scripts download from hub

few_glue.py

@@ -300,6 +300,7 @@ class FewGlueConfig(datasets.BuilderConfig):
         self.features = features
         self.label_classes = label_classes
         self.citation = citation
+        self.data_url = data_url
         self.url = url
 
 
@@ -311,6 +312,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             name="boolq",
             description=_BOOLQ_DESCRIPTION,
             features=["question", "passage"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/boolq.zip",
             citation=_BOOLQ_CITATION,
             url="https://github.com/google-research-datasets/boolean-questions",
         ),
@@ -319,6 +321,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             description=_CB_DESCRIPTION,
             features=["premise", "hypothesis"],
             label_classes=["entailment", "contradiction", "neutral"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/cb.zip",
             citation=_CB_CITATION,
             url="https://github.com/mcdm/CommitmentBank",
         ),
@@ -329,6 +332,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             # Note that question will only be the X in the statement "What's
             # the X for this?".
             features=["premise", "choice1", "choice2", "question"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/copa.zip",
             citation=_COPA_CITATION,
             url="http://people.ict.usc.edu/~gordon/copa.html",
         ),
@@ -336,6 +340,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             name="multirc",
             description=_MULTIRC_DESCRIPTION,
             features=["paragraph", "question", "answer"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/multirc.zip",
             citation=_MULTIRC_CITATION,
             url="https://cogcomp.org/multirc/",
         ),
@@ -346,6 +351,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             # will contain @placeholder as a substring, which represents the word
             # to be substituted in.
             features=["passage", "query", "entities", "answers"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/record.zip",
             citation=_RECORD_CITATION,
             url="https://sheng-z.github.io/ReCoRD-explorer/",
         ),
@@ -354,6 +360,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             description=_RTE_DESCRIPTION,
             features=["premise", "hypothesis"],
             label_classes=["entailment", "not_entailment"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/rte.zip",
             citation=_RTE_CITATION,
             url="https://aclweb.org/aclwiki/Recognizing_Textual_Entailment",
         ),
@@ -363,6 +370,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             # Note that start1, start2, end1, and end2 will be integers stored as
             # datasets.Value('int32').
             features=["word", "sentence1", "sentence2", "start1", "start2", "end1", "end2"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/wic.zip",
             citation=_WIC_CITATION,
             url="https://pilehvar.github.io/wic/",
         ),
@@ -372,6 +380,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             # Note that span1_index and span2_index will be integers stored as
             # datasets.Value('int32').
             features=["text", "span1_index", "span2_index", "span1_text", "span2_text"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/wsc.zip",
             citation=_WSC_CITATION,
             url="https://cs.nyu.edu/faculty/davise/papers/WinogradSchemas/WS.html",
         ),
@@ -384,6 +393,7 @@ class FewGlue(datasets.GeneratorBasedBuilder):
             # Note that span1_index and span2_index will be integers stored as
             # datasets.Value('int32').
             features=["text", "span1_index", "span2_index", "span1_text", "span2_text"],
+            data_url="https://huggingface.co/datasets/juny116/few_glue/resolve/main/FewGLUE_32dev/wsc.zip",
             citation=_WSC_CITATION,
             url="https://cs.nyu.edu/faculty/davise/papers/WinogradSchemas/WS.html",
         ),
@@ -433,9 +443,9 @@ class FewGlue(datasets.GeneratorBasedBuilder):
         )
 
     def _split_generators(self, dl_manager):
-        dl_dir = 'FewGLUE_32dev'
-        dl_dir = os.path.join(dl_dir, self.config.name)
-
+        dl_dir = dl_manager.download_and_extract(self.config.data_url) or ""
+        task_name = _get_task_name_from_data_url(self.config.data_url)
+        dl_dir = os.path.join(dl_dir, task_name)
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,