Open cairuil opened 1 year ago
Thanks for your attention, the code for yolov7 part is in another branch of this repo which is under development. Thanks for your advice about yolov6. We are willing to try YOLOV on different bases but time is limited. After finishing the v7 part, we can add v6 to the TODO list.
I want to train a custom dataset, the yolox part is already done, how should I modify the vid_train part?
I want to train a custom dataset, the yolox part is already done, how should I modify the vid_train part?
For annotation in COCO format, you can refer to https://github.com/YuHengsss/YOLOV/blob/a75e0aea625ffbb81ab20754d94fc067e1ce88c9/yolox/data/datasets/vid.py#L414 and https://github.com/YuHengsss/YOLOV/blob/a75e0aea625ffbb81ab20754d94fc067e1ce88c9/yolox/data/datasets/ovis.py#L238 Otherwise, you may have to rewrite the dataloader pipeline.
Thanks, I will give it a try.
I am working on a project using yolov7 and I believe your solution would be of great help. Do you have an estimation of when the support for yolov7 will be ready?
I saw above that yolov7 was added and didn't find the code for that section, can you tell me which section it is in please? Also, any thoughts on adding a detector for yolov6, v6 is a frameless detector.