multimodal-speech-emotion-recognition icon indicating copy to clipboard operation
multimodal-speech-emotion-recognition copied to clipboard

Published 20 hours ago • verified publisher icon Demfier

python -c 'import torch; print(torch.cuda.is_available())' is false

Open ZahraAskarzadeh opened this issue 4 years ago • 3 comments

Hi

torch.cuda.is_available() is false, did you have the same issue?

ZahraAskarzadeh avatar Sep 23 '20 20:09 ZahraAskarzadeh

Hi @ZahraAskarzadeh

I had run my code on a GPU as training wasn't feasible otherwise. Are you unable to run the code on CPU?

Demfier avatar Sep 23 '20 21:09 Demfier

Yes, I cannot run it with CPU (it takes forever). Now I'm switching to GPU but after installing drivers still torch.cuda.is_available() is false

ZahraAskarzadeh avatar Sep 23 '20 21:09 ZahraAskarzadeh

Hi @ZahraAskarzadeh

Sorry for the late reply. If that variable is false, it means that the torch version is not compatible with the CUDA driver on your device. Try looking into it (if you haven't yet)

Demfier avatar Mar 30 '21 20:03 Demfier