-
Notifications
You must be signed in to change notification settings - Fork 95
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Fix cox3d_learner ONNX support #372
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for the fixes! Just some minor typo comments.
I assume that files in cox3d/algorithm
, are used as-is and i only added a couple of comments for typos and skipped some other issues in the docstrings and function naming.
src/opendr/perception/activity_recognition/cox3d/algorithm/res.py
Outdated
Show resolved
Hide resolved
src/opendr/perception/activity_recognition/cox3d/algorithm/res.py
Outdated
Show resolved
Hide resolved
src/opendr/perception/activity_recognition/cox3d/cox3d_learner.py
Outdated
Show resolved
Hide resolved
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
Yes I think we don't expect the algorithms to follow all the constraints we impose for our tools, especially considering that these algorithms might evolve so it might end up just complicating keeping them up to date. As long as the test sources passes (or the location is skipped when running the test) it should be fine IMHO. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just a detail, thank you
Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>
…nto fix-co3d-onnx
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you!
* Update CoX3D with continual-inference based impl * Update implementation with proper onnx support * Update CHANGELOG * Fix x3d learner tests * Update src/opendr/perception/activity_recognition/cox3d/algorithm/res.py Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> * Update src/opendr/perception/activity_recognition/cox3d/cox3d_learner.py Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> * Update src/opendr/perception/activity_recognition/cox3d/algorithm/res.py Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> * Update CHANGELOG.md Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com> * Fix typos Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> Co-authored-by: ad-daniel <44834743+ad-daniel@users.noreply.github.com>
This PR is an implementation overhaul for the Continual X3D learner, which uses the continual-inference library to create proper ONNX export and inference for step-wise model operation.