Skip to content

Support Gemma2 checkpoint conversion #1905

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 1 commit into from
Oct 3, 2024
Merged

Support Gemma2 checkpoint conversion #1905

merged 1 commit into from
Oct 3, 2024

Conversation

@jeffcarp
Copy link
Member

@jeffcarp jeffcarp commented Oct 2, 2024

Fixes #1691

  • Adds Gemma 2 presets to the existing Gemma export script
  • Adds gemma_version flag

Tested Gemma 1 & 2 (2B):

$ python tools/gemma/export_gemma_to_hf.py --gemma_version 2 --preset gemma2_2b_en --size 2b --output_dir /tmp/keras_hf_model/

$ python tools/gemma/export_gemma_to_hf.py --gemma_version 1 --preset gemma_2b_en --size 2b --output_dir /tmp/keras_hf_model/

@jeffcarp jeffcarp requested a review from nkovela1 October 2, 2024 23:55
@github-actions github-actions bot added the Gemma Gemma model specific issues label Oct 2, 2024
@divyashreepathihalli divyashreepathihalli added the kokoro:force-run Runs Tests on GPU label Oct 3, 2024
@kokoro-team kokoro-team removed the kokoro:force-run Runs Tests on GPU label Oct 3, 2024
divyashreepathihalli
divyashreepathihalli approved these changes Oct 3, 2024
View reviewed changes
Copy link
Collaborator

@divyashreepathihalli divyashreepathihalli left a comment

Choose a reason for hiding this comment

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

LGTM!

@divyashreepathihalli
Copy link
Collaborator

divyashreepathihalli commented Oct 3, 2024

you would need to run ./shell/format.sh to resolve the code format error.

@jeffcarp
Copy link
Member Author

jeffcarp commented Oct 3, 2024

Done. Thanks!

@jeffcarp
Copy link
Member Author

jeffcarp commented Oct 3, 2024

Can you or someone with permissions merge the PR?

@divyashreepathihalli divyashreepathihalli merged commit f25c8ff into keras-team:master Oct 3, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@divyashreepathihalli divyashreepathihalli divyashreepathihalli approved these changes

@nkovela1 nkovela1 Awaiting requested review from nkovela1

Assignees

No one assigned

Labels

Gemma Gemma model specific issues

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Conversion script modification for Gemma 2 models

3 participants

@jeffcarp @divyashreepathihalli @kokoro-team