torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 5.80 GiB total capacity; 5.64 GiB already allocated; 15.69 MiB free; 5.65 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

[yangcopnuli]

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 5.80 GiB total capacity; 5.64 GiB already allocated; 15.69 MiB free; 5.65 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

[praetor29]

If you're running it in an interactive notebook, try restarting the environment. That'll free up memory.

[Knackii]

Try this it worked for me #315 (comment)

[asterocean]

I figured out a solution, load modules on demand rather than load them all at the same time, try this pull request: #531