When I was running the training of crowdpose dataset, there were 'nan' and 'NaN or Inf found in input tensor' problems. I checked the data. This problem is caused by the disappearance or explosion of the gradient. After I adjusted the learning rate, this problem still can not be solved. Please help me.

Open LiPan123456789 opened this issue 3 years ago • 1 comments

Although it has this problem, it is still running normally. Is it right to continue running like this?

Jan 12 '22 10:01 LiPan123456789

Hi, I am sorry that I have not met this problem. You can reproduce our results in the paper using the cloned code without any change.

Jan 12 '22 13:01 Gengzigang