Fix LR scheduler issue with CPU offload optimizer #1649
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1649
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New FailureAs of commit dc7bbce with merge base 122eb73 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
vkuzo
approved these changes
Feb 1, 2025
gau-nernst
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As @ngc92 pointed out, there should be a synchronization to make sure param H2D finishes before the next forward pass. Hence, I added it. Thank you for the notice!
I also take this opportunity to address an issue regarding PyTorch's built-in LR schedulers (#959, #1209)
Edit: since synchronization issue is fixed by #1650, this PR will only address the LR scheduler issue