Skip to content

Conversation

@isaacbmiller
Copy link
Collaborator

@isaacbmiller isaacbmiller commented Oct 7, 2024

cc @dilarasoylu

relevant anysacle notebook is anyscale/templates#381

Copy link
Collaborator

@chenmoneygithub chenmoneygithub left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice work!

Agreed with @okhat, the PR can use some refactoring to have the finetuning code in a standalone mode.

def assert_structural_equivalency_for_predictors(
predictor1: object,
predictor2: object,
) -> Optional[AssertionError]:
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes this type hint is a bit confusing, we can omit the type hints here.

Copy link
Collaborator

@chenmoneygithub chenmoneygithub 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 the update! Most comments are on nits.

@okhat
Copy link
Collaborator

okhat commented Oct 17, 2024

Besides the things discussed on Slack, this is good to merge except for lm.py which I think @dilarasoylu is handling.

@okhat
Copy link
Collaborator

okhat commented Oct 17, 2024

Incredible work on this @dilarasoylu & @isaacbmiller !!

@okhat okhat changed the title [DNM] Refactor finetuning implementation to be 2.5 compatible Refactor finetuning implementation to be 2.5 compatible Oct 18, 2024
@isaacbmiller isaacbmiller merged commit 9a952fe into main Oct 18, 2024
4 checks passed
@isaacbmiller isaacbmiller deleted the dev_finetune branch October 18, 2024 22:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants