Update SpaCE2021.py

SpaCE2021.py

@@ -116,11 +116,10 @@ _split_name_map = {
 class SpaCE2021Config(datasets.BuilderConfig):
     """BuilderConfig for SpaCE2021."""
 
-    def __init__(self, features, splits, **kwargs):
+    def __init__(self, splits, **kwargs):
         # Version history:
         # 1.0518.0: final version used in SpaCE2021 Eval.
         super().__init__(version=datasets.Version("1.0518.0"), **kwargs)
-        self.features = features
         self.splits = splits
 
 
@@ -145,7 +144,7 @@ class SpaCE2021(datasets.GeneratorBasedBuilder):
     def _info(self):
         return datasets.DatasetInfo(
             description=_DESCRIPTION_DICT[self.config.name],
-            features=_FEATURES_DICT[self.config.name],
+            features=datasets.Features(_FEATURES_DICT[self.config.name]),
             homepage="http://ccl.pku.edu.cn:8084/SpaCE2021/",
             citation=_CITATION,
             license=_LICENSE,
@@ -157,19 +156,27 @@ class SpaCE2021(datasets.GeneratorBasedBuilder):
 
         split_things = []
         for split_name in self.config.splits:
+            # print('')
+            split_data_path = _URLS_DICT[self.config.name][split_name]
+            # print(split_data_path)
+            filepath = dl_manager.download(split_data_path)
+            # print(filepath)
+            # print('')
             split_thing = datasets.SplitGenerator(
                 name=_split_name_map[split_name],
                 gen_kwargs={
-                    "data_file": dl_manager.download(_URLS_DICT[self.config.name][split_name]),
+                    "filepath": filepath,
                     "split": split_name,
                 }
             )
             split_things.append(split_thing)
         return split_things
 
-    def _generate_examples(self, filepath):
-        with open(filepath, encoding="utf-8") as f:
-            for idx, line in enumerate(f):
+    def _generate_examples(self, filepath, split):
+        with open(filepath, encoding="utf-8") as ff:
+            for idx, line in enumerate(ff):
                 example = json.loads(line.strip())
-                yield example.get("qID"), example
+                qid = example.get("qID")
+                if split == qid.split("-")[2]:
+                    yield example.get("qID"), example