test
Browse files
run_whisper_finetuning.py
CHANGED
@@ -415,7 +415,6 @@ def main():
|
|
415 |
# Make sure everything is in 16K
|
416 |
train_dataset = train_dataset.cast_column(data_args.audio_column_name, Audio(sampling_rate=16000))
|
417 |
eval_dataset = eval_dataset.cast_column(data_args.audio_column_name, Audio(sampling_rate=16000))
|
418 |
-
|
419 |
|
420 |
# Rename columns
|
421 |
if data_args.audio_column_name != "audio":
|
@@ -430,7 +429,7 @@ def main():
|
|
430 |
data_args.text_column_name, "sentence")
|
431 |
eval_dataset = eval_dataset.rename_column(
|
432 |
data_args.text_column_name, "sentence")
|
433 |
-
|
434 |
|
435 |
# Prepare the dataset
|
436 |
train_dataset = train_dataset.map(prepare_dataset, remove_columns=column_names)
|
|
|
415 |
# Make sure everything is in 16K
|
416 |
train_dataset = train_dataset.cast_column(data_args.audio_column_name, Audio(sampling_rate=16000))
|
417 |
eval_dataset = eval_dataset.cast_column(data_args.audio_column_name, Audio(sampling_rate=16000))
|
|
|
418 |
|
419 |
# Rename columns
|
420 |
if data_args.audio_column_name != "audio":
|
|
|
429 |
data_args.text_column_name, "sentence")
|
430 |
eval_dataset = eval_dataset.rename_column(
|
431 |
data_args.text_column_name, "sentence")
|
432 |
+
column_names.remove(data_args.text_column_name)
|
433 |
|
434 |
# Prepare the dataset
|
435 |
train_dataset = train_dataset.map(prepare_dataset, remove_columns=column_names)
|