Datasets:

---

EricB

/

wikitext2

like 0

Error code:   TransformRowsProcessingError
Exception:    ValueError
Message:      The best audio stream is unknown and there is no specified stream. 
This should never happen. Please report an issue following the steps in
https://github.com/pytorch/torchcodec/issues/new?assignees=&labels=&projects=&template=bug-report.yml.

Traceback:    Traceback (most recent call last):
                File "/src/libs/libapi/src/libapi/rows_utils.py", line 37, in _transform_row
                  transformed_row[featureName] = get_cell_value(
                                                 ~~~~~~~~~~~~~~^
                      dataset=dataset,
                      ^^^^^^^^^^^^^^^^
                  ...<9 lines>...
                      hf_token=hf_token,
                      ^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/libs/libcommon/src/libcommon/viewer_utils/features.py", line 439, in get_cell_value
                  return audio(
                      dataset=dataset,
                  ...<9 lines>...
                      json_path=json_path,
                  )
                File "/src/libs/libcommon/src/libcommon/viewer_utils/features.py", line 205, in audio
                  return create_audio_file(
                      dataset=dataset,
                  ...<8 lines>...
                      filename=f"{append_hash_suffix('audio', json_path)}{target_audio_file_extension}",
                  )
                File "/src/libs/libcommon/src/libcommon/viewer_utils/asset.py", line 136, in create_audio_file
                  audio = Audio().decode_example({"path": tmpfile.name, "bytes": None})
                File "/usr/local/lib/python3.14/site-packages/datasets/features/audio.py", line 196, in decode_example
                  audio = AudioDecoder(
                      path, stream_index=self.stream_index, sample_rate=self.sampling_rate, num_channels=self.num_channels
                  )
                File "/usr/local/lib/python3.14/site-packages/torchcodec/decoders/_audio_decoder.py", line 72, in __init__
                  ) = create_audio_decoder(
                      ~~~~~~~~~~~~~~~~~~~~^
                      source=source,
                      ^^^^^^^^^^^^^^
                  ...<3 lines>...
                      num_channels=num_channels,
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/torchcodec/_core/_decoder_utils.py", line 86, in create_audio_decoder
                  raise ValueError(
                  ...<2 lines>...
                  )
              ValueError: The best audio stream is unknown and there is no specified stream. 
              This should never happen. Please report an issue following the steps in
              https://github.com/pytorch/torchcodec/issues/new?assignees=&labels=&projects=&template=bug-report.yml.