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 just installed the exact packages in the tutorial but still get the error of AttributeError: 'EnvWrapper' object has no attribute 'use_cuda' . Thanks for your support.
The text was updated successfully, but these errors were encountered:
Hi @Mshz2
Thank you for trying our code. We recently have a big update for the codebase (and it bumps up to version 2.0). It now has dual support for both CUDA C and Numba, therefore instead of env_wrapper.use_cuda = True or False, it changes to env_wrapper.env_backend = "pycuda", "numba" or "cpu". I believe all other tutorials have been modified accordingly and env_wrapper itself has internal backward compatibility too, but we have not yet update the lightning one. We will update it shortly, thank you for your patience.
Hi, thanks for you great contribution with warp_drive. I am just following the tutorial of tutorial-7-training_with_warp_drive_and_pytorch_lightning. I have no error until the cell which we use EnvWrapper:
I just installed the exact packages in the tutorial but still get the error of
AttributeError: 'EnvWrapper' object has no attribute 'use_cuda'
. Thanks for your support.The text was updated successfully, but these errors were encountered: