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.

Sign up for GitHub

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

Jump to bottom

Reproducibility in multi-process data loading #8601

Closed
2 tasks done
UnglvKitDe opened this issue Jul 16, 2022 · 1 comment · Fixed by #8602
Closed
2 tasks done

Reproducibility in multi-process data loading #8601

UnglvKitDe opened this issue Jul 16, 2022 · 1 comment · Fixed by #8602
Labels
enhancement New feature or request

Comments

@UnglvKitDe
Copy link
Contributor

Search before asking

  • I have searched the YOLOv5 issues and found no similar feature requests.

Description

In the official documentation it is recommended to set the rseed for the worker to ensure reproducibility even for multi-process dataloaders.

Use case

No response

Additional

No response

Are you willing to submit a PR?

  • Yes I'd like to help by submitting a PR!
@UnglvKitDe UnglvKitDe added the enhancement New feature or request label Jul 16, 2022
@UnglvKitDe UnglvKitDe mentioned this issue Jul 16, 2022
Merged
@glenn-jocher glenn-jocher linked a pull request Jul 17, 2022 that will close this issue
Add generator and worker seed #8602
Merged
@glenn-jocher
Copy link
Member

@UnglvKitDe I'm seeing fully reproducible results now with master and torch>=1.12.0. Is this not the case for you?
Screen Shot 2022-07-19 at 6 21 53 PM

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Add generator and worker seed UnglvKitDe/yolov5-1
2 participants
@glenn-jocher @UnglvKitDe