Megvii-BaseDetection / Megvii-BaseDetection/YOLOX
How to adjust eval confidence threshold
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 10.6k
- Forks
- 2.5k
- PR merge metrics
- No merged PRs in 30d
Description
当前使用的验证命令如下：
python3 tools/eval.py -n yolox-l -c best_ckpt.pth.tar -b 32 -d 2 --conf 0.06
并且AP50：93.0 AR50：96.2
如何理解--conf 0.06参数，我看默认的是0.01，但是在源码中设置的是如下：
postprocess(prediction, num_classes, conf_thre=0.7, nms_thre=0.45)，
正常理解下是conf正常调整为0.6，0.7，0.8，0.9去调节AP50和AR50，但是默认的--conf 0.01，目前没有看出我们的yolox是如何进行理解和工作的
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with tools/eval.py and trace the --conf argument into postprocess, where the issue cites conf_thre=0.7. Compare the command-line default of 0.01 with the postprocessing default, then document which value is used and how the threshold affects AP50 and AR50.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100