Closed souryadey closed 1 year ago
This occurs on torch<=1.12
. It's not an error, it's a warning. Nothing is wrong, so do not worry. Even though Pytorch complains, the matrix shapes are actually perfectly compatible. The results that you get will be perfectly fine.
If you just want to get rid of the message, consider adding the following to your script:
import warnings
warnings.filterwarnings("ignore", category=UserWarning)
Alternatively, upgrading to torch 1.13
should make the message go away.
Running
StatePred.train_net()
results in something like: