tensorflow
fixes #9927: pascal_voc_detection_metrics low scores for first catategory in label_map
Only add 1 if there is at least one element greater than 0 per row for ground-truth See also issue #9927 This change should reviewed in detail, maybe this could have side effects, I am not aware of.
Description
This bug could results in wrong results and should be fixed
Type of change
For a new feature or function, please create an issue first to discuss it with us before submitting a pull request.
Note: Please delete options that are not relevant.
- [x ] Bug fix (non-breaking change which fixes an issue)
Tests
Evaluate results with metric pascal voc evaluator that has unusual low scores on the first class. you can stop on the changed line and see if there are row entries in the ground truth variable with only zeros
See also #9927
Test Configuration:
Checklist
- [x] I have signed the Contributor License Agreement.
- [x] I have read guidelines for pull request.
- [x] My code follows the coding guidelines.
- [x] I have performed a self code review of my own code.
- [x] I have commented my code, particularly in hard-to-understand areas.
- [x] My changes generate no new warnings.
Hello everyone, I'm Mada, and I'm eager to begin contributing to open-source projects. TensorFlow seems like an excellent place to start. Could someone please help me determine if a specific issue is still open, and if so, guide me on how to proceed with it? I appreciate any assistance provided.
@diaconumadalina sorry for my late response To be honest, I’m not entirely sure. This issue occasionally appears in relation to #9927, and I first encountered it while working on an AI model about two years ago. If I have the opportunity, I could implement some tests to verify this. However, I’m not certain if this is still a requirement, given that the object detection component seems somewhat outdated. The most recent model added is mobilenetv3small.