You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi
Recently, when running your code on GPU I have error: can't optimize a non-leaf Tensor
Tested on local machine, GCP Compute engine with PyTorch and Cuda.
The text was updated successfully, but these errors were encountered:
Bug fixed, thanks for pointing out.
The problem is that we can not set requires_grad=True before .cuda() for PyTorch>0.4.0.
Details can be found in this commit
Hi
Recently, when running your code on GPU I have error: can't optimize a non-leaf Tensor
Tested on local machine, GCP Compute engine with PyTorch and Cuda.
The text was updated successfully, but these errors were encountered: