JarodMica / ai-voice-cloning

GNU General Public License v3.0
567 stars 127 forks source link

Issue Cuda out of memory #122

Open Axienne opened 3 months ago

Axienne commented 3 months ago

Hi. I have this issue. Does anyone have an answer? The most interesting fact is, this going with the very start, I only first time klick on Start.bat. and this error pop. torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB. GPU 0 has a total capacity of 4.00 GiB of which 0 bytes is free. Of the allocated memory 3.41 GiB is allocated by PyTorch, and 53.04 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

maepopi commented 3 months ago

Hey, Strange that it happens right from the start. Have you tried clearing the torch GPU cache? It does the trick for me sometimes. In your console, type "python" while in the virtual environment, then do

import torch

torch.cuda.empty_cache()

exit()

Not sure it will help but there's no harm in trying I guess!