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.

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

Update ort CIs (slow, gpu, train) #2024

Merged
merged 73 commits into from
Jan 29, 2025
Merged

Update ort CIs (slow, gpu, train) #2024

merged 73 commits into from
Jan 29, 2025

Conversation

IlyasMoutawwakil
Copy link
Member

@IlyasMoutawwakil IlyasMoutawwakil commented Sep 14, 2024 •
edited
Loading

What does this PR do?

This is a PR that fixes:

  • Onnxruntime Training testing and examples.
  • Onnruntime GPU testing (CUDA & TRT EPs), and io binding implementation.

I tried my best to make things work with the least change, in some cases that was hard, e.g. io binding was dependant on the input order in the forward signature, which made some model types incompatible with the ORTModel signature.

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?

Who can review?

@IlyasMoutawwakil IlyasMoutawwakil added training gpu-test trigger GPU tests labels Sep 14, 2024
@IlyasMoutawwakil IlyasMoutawwakil mentioned this pull request Sep 26, 2024
Merged
3 tasks
echarlaix
echarlaix approved these changes Jan 21, 2025
View reviewed changes
Copy link
Collaborator

@echarlaix echarlaix left a comment

Choose a reason for hiding this comment

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

Thanks a lot for the great PR @IlyasMoutawwakil !

tests/onnxruntime/test_diffusion.py Show resolved Hide resolved
optimum/onnxruntime/trainer.py
self.state.is_hyper_param_search = trial is not None
self.state.train_batch_size = self._train_batch_size

Copy link
Collaborator

Choose a reason for hiding this comment

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

also cc @JingyaHuang who took care of the ort training integrations

optimum/onnxruntime/modeling_ort.py Outdated Show resolved Hide resolved
optimum/utils/import_utils.py Outdated Show resolved Hide resolved
IlyasMoutawwakil
IlyasMoutawwakil commented Jan 28, 2025
View reviewed changes
setup.py Outdated Show resolved Hide resolved
IlyasMoutawwakil
IlyasMoutawwakil commented Jan 29, 2025
View reviewed changes
setup.py
"accelerate", # ORTTrainer requires it.
"transformers>=4.36,<4.49.0",
],
"onnxruntime-training": [
Copy link
Member Author

Choose a reason for hiding this comment

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

The readme and docs should reflect this change

Copy link
Collaborator

Choose a reason for hiding this comment

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

added in #2173

@echarlaix
Copy link
Collaborator

Thanks a lot @IlyasMoutawwakil for adding many fixes !

@echarlaix echarlaix merged commit b755036 into main Jan 29, 2025
40 checks passed
@echarlaix echarlaix deleted the enable-ort-gpu-tests branch January 29, 2025 17:11
@echarlaix echarlaix mentioned this pull request Jan 29, 2025
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@echarlaix echarlaix echarlaix approved these changes

Assignees
No one assigned
Labels
onnxruntime-gpu onnxruntime-slow onnxruntime-training
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@IlyasMoutawwakil @HuggingFaceDocBuilderDev @echarlaix