Megvii-BaseDetection / Megvii-BaseDetection/YOLOX
The label does not correspond to the box
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 10.6k
- Forks
- 2.5k
- PR merge metrics
- No merged PRs in 30d
Description
My training process refers to the Train Custom Data on the official website .
Instead of the example, I used COCO format dataset, which is about 500 images labeled boat.
In this process, I only changed self.numclasses=80 directly on exps/custom/yolox_s.py.
(An error is reported if it equals 1.)
The categories item of my JSON file is consistent with the coco dataset. Categories id = 9.
Here are the test results of my training model. (fine-tuning)( use demo.py & configuration file used during training---yolox_s.py）
befor fine-tuning
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 exps/custom/yolox_s.py and the demo.py entry point, using the training configuration and COCO-format category details from the report. Reproduce the label-to-box mismatch and trace how numclasses and category IDs are interpreted. Done means identifying and correcting the configuration or dataset-handling issue so detections use the expected label.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100