ai-forever / ai-forever/ru-dalle
CUDA out of memory with 'ruDALLE-image-prompts-A100.ipynb' & 'ruDALLE-image-prompts-dress-mannequins-V100.ipynb'
- Linguagem predominante
- Jupyter Notebook
- Estrelas
- 1.6k
- Forks
- 242
- Métricas de merge de PRs
- Nenhum PR com merge em 30d
Descrição
Hi I get an OOM error with an (RTX 3060 12 GB) for the following scripts: 'uDALLE-image-prompts-A100.ipynb' , 'ruDALLE-image-prompts-dress-mannequins-V100.ipynb'
Unfortunately I could not make it work.
I tried to look at the other script 'Malevich-3.5GB-vRAM-usage.ipynb' and I have tried to adept the code, but I still get an error:
`RuntimeError: CUDA out of memory. Tried to allocate 1.98 GiB (GPU 0; 11.77 GiB total capacity; 6.63 GiB already allocated; 1.08 GiB free; 11.00 GiB allowed; 7.77 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF`
Any help on how to tweak the code would be appreciated!
Thanks!
Gero
Guia de contribuição
Nenhum guia de contribuição indexado para este repositório
Avaliação
Esta issue ainda não foi avaliada.