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

Fix CUDA launch error #83

Merged
merged 8 commits into from
Feb 14, 2024
Merged

Fix CUDA launch error #83

merged 8 commits into from
Feb 14, 2024

Conversation

lilyminium
Copy link
Collaborator

@lilyminium lilyminium commented Feb 9, 2024

Fixes #81

Changes made in this Pull Request:

  • Adds a CUDA environment for building and testing this bug
  • Removes extraneous kwargs (num_workers, etc) from DataLoader

PR Checklist

  • Tests?
  • Docs?
  • CHANGELOG updated?
  • Issue raised/referenced?

@codecov-commenter
Copy link

codecov-commenter commented Feb 9, 2024

Codecov Report

Merging #83 (084499d) into main (cf8c675) will not change coverage.
Report is 1 commits behind head on main.
The diff coverage is n/a.

Additional details and impacted files

@lilyminium lilyminium force-pushed the fix-cuda-launch-error branch from 009de38 to 72afe76 Compare February 9, 2024 05:22
Comment on lines 112 to 118
main_tests_cuda:
name: CUDA CI (py-${{ matrix.python-version }})
runs-on: ubuntu-latest
strategy:
fail-fast: false
matrix:
python-version: ["3.9", "3.10"]
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Selfish curiosity - is this action meant to be run on GPUs? Or is it just trying to get at something at the API level that doesn't need hardware?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, I think this needs GPUs to run, so this CI might not go anywhere since GitHub doesn't have widely available GPU runners yet -- this test might have to be for local testing only. Although I'm having difficulty reproducing this on the same machine I saw it on in the first place...

@lilyminium
Copy link
Collaborator Author

I can't seem to trigger this with a trimmed down example, although I'm still seeing the error if I try to run a production fit and removing num_workers does fix it in that instance. In general allowing kwargs to set these options seems like a good idea, though.

@lilyminium lilyminium marked this pull request as ready for review February 14, 2024 08:17
@lilyminium lilyminium merged commit 4c6953b into main Feb 14, 2024
46 checks passed
@lilyminium lilyminium deleted the fix-cuda-launch-error branch February 14, 2024 11:46
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

Successfully merging this pull request may close these issues.

CUDA launch error
3 participants