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 issue with one-cycle logic #728

Merged
merged 2 commits into from
Sep 28, 2022
Merged

Fix issue with one-cycle logic #728

merged 2 commits into from
Sep 28, 2022

Conversation

muellerzr
Copy link
Collaborator

Turns out the logic for one-cycle on the scheduler was backwards so step was never called.

Fixes #690 and adds some tests

@muellerzr muellerzr added the bug Something isn't working label Sep 28, 2022
@muellerzr muellerzr requested a review from sgugger September 28, 2022 20:26
Copy link
Collaborator

@sgugger sgugger 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 adding the tests!

@HuggingFaceDocBuilderDev
Copy link

HuggingFaceDocBuilderDev commented Sep 28, 2022

The documentation is not available anymore as the PR was closed or merged.

@muellerzr muellerzr merged commit 9e4fe78 into main Sep 28, 2022
@muellerzr muellerzr deleted the one-cycle-issue branch September 28, 2022 20:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

Successfully merging this pull request may close these issues.

underlying lr_scheduler.step() never called with OneCycleLR in single GPU
3 participants