facebookresearch / facebookresearch/detectron2
Focal loss for FasterRCNN
Open
enhancement
- Dominant language
- Python
- Stars
- 34.7k
- Forks
- 7.9k
- PR merge metrics
- No merged PRs in 30d
Description
hello
I want to use Focal loss which is used in Retinanet into FasterRCNN.
how can I do it?
thank you
Contributor guide
Research direction
Start by locating the FasterRCNN and RetinaNet implementations in detectron2 and comparing how their loss paths are exposed. The issue names no files or tests, so first clarify whether this is an implementation request or a usage question; done should be a defined, tested way to use focal loss with FasterRCNN.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100