upd: new try

rwth_phoenix_weather_2014.py

@@ -94,8 +94,8 @@ class RWTHPhoenixWeather2014(datasets.GeneratorBasedBuilder):
         )
 
     def _split_generators(self, dl_manager: datasets.DownloadManager):
-        dfs = {}
-        file_ids = {}
+        example_ids = {}
+        annotations = {}
 
         dataDirMapper = {
             datasets.Split.TRAIN: "train",
@@ -113,15 +113,17 @@ class RWTHPhoenixWeather2014(datasets.GeneratorBasedBuilder):
             data_csv = dl_manager.download(
                 f"{base_url}/annotations/manual/{dataDirMapper[split]}{self.config.corpus_file_suffix}")
 
-            dfs[split] = pd.read_csv(data_csv, sep='|')
-            file_ids[split] = dfs[split]['id']
+            df = pd.read_csv(data_csv, sep='|')
+
+            example_ids[split] = df['id']
+            annotations[split] = df['annotation']
 
         return [
             datasets.SplitGenerator(
                 name=split,
                 gen_kwargs={
-                    "df": dfs[split],
-                    "file_ids": file_ids[split],
+                    "example_ids": example_ids[split],
+                    "annotations": annotations[split],
                 },
             )
             for split in [
@@ -131,13 +133,11 @@ class RWTHPhoenixWeather2014(datasets.GeneratorBasedBuilder):
             ]
         ]
 
-    def _generate_examples(self, df, file_ids):
-        for key, (idx) in enumerate(zip(file_ids)):
-            entry = df[df["id"] == idx]
-
+    def _generate_examples(self, example_ids, annotations):
+        for key, (idx, annotation) in enumerate(zip(example_ids, annotations)):
             result = {
                 "id": idx,
-                "transcription": entry["annotation"],
+                "transcription": annotation,
             }
 
             yield key, result