YOLO v6 iou_loss dfl_loss一直为0

Question
img record infomation path is:…/mydata/images.train_cache.json
Train: Final numbers of valid images: 1248/ labels: 1248.
0.1s for dataset initialization.
img record infomation path is:…/mydata/images.val_cache.json
Convert to COCO format
100%|██████████| 119/119 [00:00<00:00, 119350.11it/s]
Convert to COCO format finished. Resutls saved in …/mydata\annotations\instances_val.json
Val: Final numbers of valid images: 119/ labels: 119.
0.0s for dataset initialization.
Training start…

第一步：红色框的地方不要用yolov6s.py要用yolov6s_finetune.py

第二步：
In the ‘yolov6/data/datasets.py’ file, find the function ‘def img2label_paths(img_paths)’ and replace it with the below function

def img2label_paths(img_paths):
# Define label paths as a function of image paths
# sa, sb = f'{os.sep}images{os.sep}', f'{os.sep}labels{os.sep}' # /images/, /labels/ substrings
sa, sb = 'images','labels'
return [sb.join(x.rsplit(sa, 1)).rsplit('.', 1)[0] + '.txt' for x in img_paths]

这样问题就解决了！！！！！