Open gcucurull opened 4 years ago
When I use dice loss for Pytorch 1.3 I get the following error when I try to call loss.backward().
loss.backward()
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn
Any idea why?
Hey @gcucurull, could you post a minimal working example?
When I use dice loss for Pytorch 1.3 I get the following error when I try to call
loss.backward()
.Any idea why?