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
Came across an issue when using a KANLayer within a pytorch lightning module and training with cuda. Some tensors being built and processed in the KANLayer were always getting sent to 'cpu'. I resolved this issue and put the updated code on my forked repository, pykan-cuda. Would appreciate if you checked it out.
The text was updated successfully, but these errors were encountered:
Came across an issue when using a KANLayer within a pytorch lightning module and training with cuda. Some tensors being built and processed in the KANLayer were always getting sent to 'cpu'. I resolved this issue and put the updated code on my forked repository, pykan-cuda. Would appreciate if you checked it out.
The text was updated successfully, but these errors were encountered: