According to the yolov8_s_syncbn_fast_8xb16-500e_coco.py configuration file, the custom training data's bounding boxes cannot be read, and the loss_bbox and loss_dfl always remain at 0.0000.

[J-like]

Prerequisite

• I have searched the existing and past issues but cannot get the expected help.
• I have read the FAQ documentation but cannot get the expected help.
• The bug has not been fixed in the latest version.

💬 Describe the reimplementation questions

I trained my own dataset (three classes) using the YOLOv8 configuration file yolov8_s_syncbn_fast_8xb16-500e_coco.py, but the category information and bounding boxes are not being read correctly. The dataset format follows the official documentation guidelines.

Environment

loss_bbox: 0.0000 loss_dfl: 0.0000

Expected results

No response

Additional information

No response