[Doc] Add how to use Lion optimizer #152
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
|
The documentation is not available anymore as the PR was closed or merged. |
lvwerra
approved these changes
Feb 21, 2023
Co-authored-by: Leandro von Werra <lvwerra@users.noreply.github.com>
younesbelkada
commented
Feb 21, 2023
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Outdated suggestions cannot be applied.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.
What does this PR do?
This PR adds LION optimizer from: https://arxiv.org/abs/2302.06675 / code source is: https://github.com/google/automl/blob/master/lion/lion_pytorch.py
For now it does not really seem to improve convergence in
gpt-sentiment, but thought it would be nice to showcase that it can be applied fortrland used out of the boxEDIT: it seems to converge nicely after dividing the learning rate by 3, as suggested by the paper
cc @lvwerra @kashif
I think this is a nice-to-have artifact in this lib, as it reduces quite nicely the memory footprint of training (got 10% reduction)