Skip to content

Conversation

eaidova
Copy link
Contributor

@eaidova eaidova commented May 16, 2025

What does this PR do?

added possibility to provide ov_export_config to be able compress model to fp16 during export

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Did you make sure to update the documentation with your changes?
  • Did you write any new necessary tests?

@HuggingFaceDocBuilderDev

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.

@eaidova eaidova changed the title allow passing ov_config to from_pretrained allow passing ov_export_config to from_pretrained May 16, 2025
@eaidova eaidova force-pushed the ea/allow_passing_ov_config branch from 4a044f7 to 206cbf9 Compare May 16, 2025 11:26
Copy link
Collaborator

@rkazants rkazants left a comment

Choose a reason for hiding this comment

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

May be it makes sense to add this into documentation.
Also, should not we check the content of this config file like presence of dtype?
Can it be that it contradicts to quantization_config file if it is also specified?

@eaidova
Copy link
Contributor Author

eaidova commented May 16, 2025

From the beginning, I would like to clarify that I'm still working on this PR did not invite anybody to review, and did not ask @rkazants to do that. Please ignore this PR for some time until it is ready

@eaidova eaidova marked this pull request as draft May 16, 2025 12:01
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