Skip to content

Fix Yolos ONNX export test #18606

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

Merged
merged 2 commits into from
Aug 17, 2022
Merged

Fix Yolos ONNX export test #18606

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b106581
fix
ydshieh Aug 12, 2022
bccdc6d
Apply suggestions from code review
ydshieh Aug 16, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 7 additions & 0 deletions tests/onnx/test_onnx_v2.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -284,6 +284,13 @@ def _onnx_export(self, test_name, name, model_name, feature, onnx_config_class_c
model_class = FeaturesManager.get_model_class_for_feature(feature)
config = AutoConfig.from_pretrained(model_name)
model = model_class.from_config(config)

# It seems yolo-like models have problems on ONNX exporting on CUDA.
# See: https://github.com/ultralytics/yolov5/pull/8378
# (Not sure which op causing issue though)
if model.__class__.__name__.startswith("Yolos") and device != "cpu":
return

onnx_config = onnx_config_class_constructor(model.config)

if is_torch_available():
Expand Down