meixiangzhang

Results 4 comments of meixiangzhang

I have encountered the same problem. Have you resolved it? How did it solve it? Thank you!

> > During training, the initial loss was effective, but later the loss became nan. I am using the URPC2019 dataset in VOC format, and the environment is configured according...

![Image](https://github.com/user-attachments/assets/40758b19-3cc4-4e31-82c9-6e327505fb8d) torch 2.1.0 cuda 11.8 mmcv 2.1.0 mmdet 3.3.0 Except for the dataset, the rest of the code remains unchanged. How should I solve this problem?