Closed miseon119 closed 1 year ago
I think it is a torch version problem. I used pytorch 1.12. My colleague changed code like this, it works:
Maybe such a modification cannot work, since no gradient is propagated from the total loss.
You are right. It's not work.
Hi, I re-run your code in windows 11, I got this error..
if I add
torch.autograd.set_detect_anomaly(True)
in yourmain.py
before I train. I got this .How should I fix this problem.
my environment is: