Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

CUDA Out of Memory Error When Running sample.py with a large --num-samples #45

Open
JoshPhd opened this issue Apr 13, 2024 · 2 comments

Comments

@JoshPhd
Copy link

JoshPhd commented Apr 13, 2024

I'm trying to generate a number of sequences with sample.py, so I set the --num-samples parameter to 100.

However, upon execution, I received the following error message:RuntimeError: CUDA out of memory. Tried to allocate 166.00 MiB (GPU 0; 9.77 GiB total capacity; 8.11 GiB already allocated; 149.81 MiB free; 8.15 GiB reserved in total by PyTorch). My GPU seems to run out of memory during the process.

Has anyone else experienced anything similar, or does anyone have a suggestion on how to get around this memory limitation? Thank you in advance for your time and help!

@jamesrgraham
Copy link

I got the same. I found I could run ~15 sequences at a time.

But I'm getting repeat sequences (100% ID to sequences from other runs) when I run multiple sets of 15 to get up to 100.

@JoshPhd
Copy link
Author

JoshPhd commented Apr 16, 2024

I got the same. I found I could run ~15 sequences at a time.

But I'm getting repeat sequences (100% ID to sequences from other runs) when I run multiple sets of 15 to get up to 100.

Got it, Tks for your advice

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants