Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Made litellm judge backend more robust. #485

Merged
merged 4 commits into from
Jan 7, 2025

Conversation

JoelNiklaus
Copy link
Contributor

The metric computation fails when a result is None. Therefore retry and otherwise return an error ModelResponse.

Copy link
Member

@clefourrier clefourrier left a comment

Choose a reason for hiding this comment

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

LGTM

Side note: We should add an error flag in the ModelResponse if we want to use it to also report these kind of problems - otherwise, if we implement caching, we won't re-run these even though they are incorrect.
Can you also add a failed attribute to the ModelResponse, False by default, and True here?

You'll be able to use it in #480

src/lighteval/metrics/llm_as_judge.py Outdated Show resolved Hide resolved
Copy link
Member

@clefourrier clefourrier left a comment

Choose a reason for hiding this comment

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

Thanks!

@HuggingFaceDocBuilderDev
Copy link
Collaborator

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@clefourrier clefourrier merged commit fdb12f4 into huggingface:main Jan 7, 2025
3 checks passed
JoelNiklaus added a commit to JoelNiklaus/lighteval that referenced this pull request Jan 7, 2025
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.

3 participants