YahooKID
commented
1 year ago Halo, it seems that you don't set the yaml correctly. You can try to add --yaml data/voc.yaml when you use tools/infer.py because programme uses data/coco.yaml by default. The command you use will be like( you can set other parameters as you like) python tools/infer.py --yaml data/voc.yaml
Before Asking
[X] I have read the README carefully. 我已经仔细阅读了README上的操作指引。
[X] I want to train my custom dataset, and I have read the tutorials for training your custom data carefully and organize my dataset correctly; (FYI: We recommand you to apply the config files of xx_finetune.py.) 我想训练自定义数据集，我已经仔细阅读了训练自定义数据的教程，以及按照正确的目录结构存放数据集。（FYI: 我们推荐使用xx_finetune.py等配置文件训练自定义数据集。）
[X] I have pulled the latest code of main branch to run again and the problem still existed. 我已经拉取了主分支上最新的代码，重新运行之后，问题仍不能解决。
Search before asking
Question
Hello,
I have the following issue (it happens with custom datasets as well as with the VOC used in the repo). After following and executing each chunk given in the yolov6_training notebook (succesfully, and using VOC), i can start training without issues, and it trains correctly. The issue starts when i try to infer on test images(i have tried several): The bounding boxes are very accurate on every image, but the labeling is not working at all, ie: People get labeled as birds, trains as sheep, and so on. I have looked everywhere i can think of but cant seem to find the issue.
I have also tried training with a custom dataset, with the same results (accurate bounding boxes but wrong labeling). Find attached some examples of inferred pictures.
Thank you in advance for your time Best regards.
Additional
No response