-
-
Notifications
You must be signed in to change notification settings - Fork 79
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Speech commands loading broken #383
Labels
help wanted
Extra attention is needed
Comments
Thanks for the issue! Unfortunately, I'm a little time-restricted currently. If I recall correctly, the original implementor was @cpehle. Do you have a minute to check this out? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
When using
norse.datasets.speech_commands.prepare_dataset
, the file lists generated at https://github.com/norse/norse/blob/main/norse/dataset/speech_commands.py#L57 and https://github.com/norse/norse/blob/main/norse/dataset/speech_commands.py#L63 start with ./ e.g../SpeechCommands/speech_commands_v0.02/six/c2d15ea5_nohash_0.wav
but those produced by the walkers at https://github.com/norse/norse/blob/main/norse/dataset/speech_commands.py#L68, https://github.com/norse/norse/blob/main/norse/dataset/speech_commands.py#L74 and https://github.com/norse/norse/blob/main/norse/dataset/speech_commands.py#L77 do not so everything ends up in training set. Afraid I don't know enough about torchaudio to know the root cause of this, otherwise I'd make a PR.Finally, pretty sure it should be
test_filelist
andtest_sc._walker
at https://github.com/norse/norse/blob/main/norse/dataset/speech_commands.py#L77System:
Ubuntu 18.04
Python 3.8
TorchAudio 2.1.2+cu121
Torch 2.1.2+cu121
The text was updated successfully, but these errors were encountered: