windula

How to write a custom dataloader to do fusion from two data folders

2

### Search before asking - [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question I want to write a data loader to train...

Train Yolo9 without bounding box annotation

1

can Yolo9 be trained without bounding box annotation files but just with class folders?, if not possible how to achieve it, like writing a dataloader ├── train │ ├── class-1...

I want to do a thermal and RGB image fusion to feed to another network, can I do it with this library? Kindly provide any examples