How to prepare custom data for training and testing?

WongKinYiu / yolov7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

GNU General Public License v3.0

13.38k stars 4.22k forks source link

How to prepare custom data for training and testing? #69

Open XiaoJiNu opened 2 years ago

XiaoJiNu commented 2 years ago

Hi, the txt label for training is not the same as yolo label format, what's the format of txt label in coco2017labels-segments.zip? I should I prepare for my custom training data?

XiaoJiNu commented 2 years ago

Another question, why don't you make the training labels format is the same as yolov5 and darknet?

WongKinYiu commented 2 years ago

Support both bbox and segment (polygon) annotation for training. If bbox annotation is used, please set paste_in to zero in hyp.yaml.