RosettaCommons / RosettaCommons/RFdiffusion

about cuda out of memery issue

Open
#182 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
3.1k
Forks
644
PR merge metrics
No merged PRs in 30d

Description

im doing binder design on A100 with "
python /dssg/home/acct-clswg/clswg/apps/RFdiffusion/scripts/run_inference.py
inference.output_prefix=example_outputs/design_ppi
inference.input_pdb=input.pdb
'contigmap.contigs=[A414-588/0 20-50]'
inference.num_designs=10000
denoiser.noise_scale_ca=0.5
denoiser.noise_scale_frame=0.5"
but always get cuda out of memory after maybe 10 or 20 designs
the input pdb is about 300aa long
is the length of my binder proper or can anything else trigger such issue?
i doubt A100 would go out of memory on this case

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the command and entry point in /dssg/home/acct-clswg/clswg/apps/RFdiffusion/scripts/run_inference.py, then reproduce the run using the 300aa input PDB and the stated A100 settings. Track memory as designs progress and compare the behavior after 10–20 designs. Done means identifying whether the input length or another setting causes the out-of-memory failure, with a reproducible explanation or workaround.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.