marinone94
commited on
Commit
•
b7db389
1
Parent(s):
b9efe48
fix raw dataset features
Browse files
run_speech_recognition_seq2seq_streaming.py
CHANGED
@@ -376,7 +376,7 @@ def load_maybe_streaming_dataset(dataset_names, dataset_config_names, split="tra
|
|
376 |
):
|
377 |
for split_name in split_names.split("+"):
|
378 |
dataset = load_dataset(dataset_name, dataset_config_name, split=split_name, streaming=streaming, **kwargs)
|
379 |
-
raw_datasets_features = list(
|
380 |
if column_names[0] not in raw_datasets_features:
|
381 |
if len(column_names) == 1 or column_names[1] not in raw_datasets_features:
|
382 |
raise ValueError("Column name not found in dataset.")
|
|
|
376 |
):
|
377 |
for split_name in split_names.split("+"):
|
378 |
dataset = load_dataset(dataset_name, dataset_config_name, split=split_name, streaming=streaming, **kwargs)
|
379 |
+
raw_datasets_features = list(dataset.features.keys())
|
380 |
if column_names[0] not in raw_datasets_features:
|
381 |
if len(column_names) == 1 or column_names[1] not in raw_datasets_features:
|
382 |
raise ValueError("Column name not found in dataset.")
|