2. Did you make any modifications on the code or config? Did you understand what you have modified?
I just use universalnet_r50 with change the number of classes and change sample per gpu to 4.
3. What dataset did you use?
I just use my custom dataset inherited from coco dataset with 14 class.
**Environment**
follow your requirements
Thanks for your error report and we appreciate it a lot.
Checklist
Describe the bug when I trained with 1 gpu, this process interrupted with the following issue:
Reproduction