Transformers trainer compatibility #7
Merged
Conversation
This file contains hidden or 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
|
This is great, thanks for the contribution! Would you mind copying it into the two distributed implementations too and I'll go ahead and merge? Also, let me know if you have any questions about hyperparameters for training, happy to help. |
|
Actually nevermind with the copying, I'll do it right now as I have some other things to take care of now that I think of it, thank you for the PR and let me know if you have any questions! |
|
Awesome, thank you! I didn't have much time to go through the entire code, so I'm not entirely sure how it works. However, it seems to have successfully unlocked my SLM pre-training, where I encountered the wall while using AdamW and Moun. Great job! |
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Outdated suggestions cannot be applied.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.
Hey, I'm using your optimizer when training with the Transformers library. My PR allows saving optimizer checkpoints, eliminating the compression error:
AttributeError: Can't pickle local object 'precond_update_prob_schedule.._schedule'