PCGrad with LR schedules, resume from checkpoint with LR schedules #145
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
Merge new commits from jpata:master
Merge from jpata/particleflow master
Merge jpata/master into master
Merge latest developments
merge jpata/particleflow master
|
|
nice, this looks like a good fix. once you confirm it works, we can merge it! |
|
I confirm it works. Training with PCGrad after this mod gives expected results. Ready for merge. |
jpata
pushed a commit
that referenced
this pull request
Sep 15, 2023
) * feat: PCGrad now works with keras LearningRateSchedule * feat: OneCycle LR and mom scheduler supports resuming from checkpoint * chore: add jobid to pipeline train sbatch script Former-commit-id: fc0e482
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Outdated suggestions cannot be applied.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.
The legacy optimizer, which PCGrad is based on, does not increment the
iterationsattribute of the optimizer whenoptimizer.apply_gradients()is called, causing kerasLearningRateSchedules to not work properly. This PR fixes that, making it possible to use LR schedules with PCGrad as well as to resume from checkpoints while keeping the schedule in order.