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
I have installed the environment you mentioned and also downloaded the repository required for the code. But when running train_gen.sh, the following error occurred, and I couldn't find the reason why tensor calculations were performed on multiple devices in the code.
The error is as follows:
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda: 0 and CPU! (When checking argument for argument target in method wrapper_nnll_loss2d_forward)
The text was updated successfully, but these errors were encountered:
I have installed the environment you mentioned and also downloaded the repository required for the code. But when running train_gen.sh, the following error occurred, and I couldn't find the reason why tensor calculations were performed on multiple devices in the code.
The error is as follows:
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda: 0 and CPU! (When checking argument for argument target in method wrapper_nnll_loss2d_forward)
The text was updated successfully, but these errors were encountered: