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

Added a fix for FP16 overflow issue on GPU/NPU #994

Merged
merged 4 commits into from
Nov 11, 2024
Merged

Conversation

AlexKoff88
Copy link
Collaborator

No description provided.

@AlexKoff88 AlexKoff88 requested a review from eaidova November 11, 2024 10:57
@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.

compress_to_fp16 = ov_config is not None and ov_config.dtype == "fp16"
model = _add_version_info_to_model(model, library_name)

if hasattr(config, "runtime_options"):
model = _add_runtime_options_to_rt_info(model, config.runtime_options)
Copy link
Collaborator

Choose a reason for hiding this comment

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

maybe better to provide runtime options as argument instead of config if it is needed only for providing these options here?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

I looked at the proposed option. It means that I will need to carry this additional argument throughout all the calls between export_models() and _save_model()

@AlexKoff88 AlexKoff88 merged commit b3cbc95 into main Nov 11, 2024
24 of 27 checks passed
@AlexKoff88 AlexKoff88 deleted the ak/fp16_overflow_fix branch November 11, 2024 15:52
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