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

Set number of dataloader workers to a sane default #66

Merged
merged 3 commits into from
Apr 21, 2021

Conversation

braun-steven
Copy link
Contributor

The prior default was the number of available CPUs which leads to issues
for systems with a large number of CPUs (e.g. DGX).

This change introduces an optional argument "--num-workers"
which defaults to 8.

Signed-off-by: Steven Lang [email protected]

The prior default was the number of available CPUs which leads to issues
for systems with large number of CPUs (e.g. DGX).

This change introduces an optional argument "--num-workers"
which defaults to 8.

Signed-off-by: Steven Lang <[email protected]>
Copy link
Owner

@mseitzer mseitzer left a comment

Choose a reason for hiding this comment

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

Thanks for the PR. I will merge once the small changes I requested are in.

src/pytorch_fid/fid_score.py Outdated Show resolved Hide resolved
src/pytorch_fid/fid_score.py Outdated Show resolved Hide resolved
src/pytorch_fid/fid_score.py Outdated Show resolved Hide resolved
- Add default (8) to num_workers method arguments
- Fix wrong help string in argument parser

Signed-off-by: Steven Lang <[email protected]>
@braun-steven
Copy link
Contributor Author

Thanks for your fast response, I've incorporated the requested changes.

@mseitzer mseitzer merged commit d99eb12 into mseitzer:master Apr 21, 2021
mseitzer added a commit that referenced this pull request Oct 10, 2021
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.

2 participants