-
Notifications
You must be signed in to change notification settings - Fork 420
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
feat: add straight-eval arg in evaluate script #793
Conversation
Codecov Report
@@ Coverage Diff @@
## main #793 +/- ##
==========================================
- Coverage 96.06% 96.00% -0.06%
==========================================
Files 130 131 +1
Lines 4901 4937 +36
==========================================
+ Hits 4708 4740 +32
- Misses 193 197 +4
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
The CI job for the evaluate script is failing :/ |
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.
Looks good! I only added a small question as a comment
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.
Looks good!
I think some of the code of the evaluation script could be refactored, but this could be done in another PR! Either way, we'll need to reharmonize the evaluation script for specific tasks (in references) with this one
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.
Cheers!
This PR adds a straight-eval arg in the evaluate script to allow using the metrics with straight bounding boxes while using a detection predictor working with polygons (assume straight pages can be False).
I also fixed a bug in the predictor to cast to numpy before calling the extraction function (when we use a dataset we have tf/pt tensors otherwise)
Any feedback is welcome!