NVlabs / stylegan2-ada-pytorch

StyleGAN2-ADA - Official PyTorch implementation
https://arxiv.org/abs/2006.06676
Other
4.13k stars 1.17k forks source link

RuntimeError: CUDA out of memory #260

Closed TopKruZ closed 2 years ago

TopKruZ commented 2 years ago

First of all, let me introduce I've trained through Colab before and will resume training on my own computer.

I try to resume but it's not work

I install torch==1.8.1+cu111 , torchvision==0.9.1+cu111 torchaudio==0.8.1 , ninja , click requests tqdm pyspng ninja imageio-ffmpeg==0.4.3

when I try to train it said

"RuntimeError: CUDA out of memory. Tried to allocate 262.00 MiB (GPU 0; 11.00 GiB total capacity; 10.03 GiB already allocated; 0 bytes free; 10.17 GiB reserved in total by PyTorch)"

image

image

Please help me how to slove it

Desktop (information):