Hi, thanks for sharing the DCCA implementation which I found really helpful! Just want to report an issue about moving data tensor to device, e.g. lines like this one, that calling x1.to(device) only returns a new copy of x1 on say GPU without actually overwriting x1 in-place; so should probably change to x1=x1.to(device) instead, to make sure the model and data are on the same device.
Thank you very much!
Hi, thanks for sharing the DCCA implementation which I found really helpful! Just want to report an issue about moving data tensor to device, e.g. lines like this one, that calling
x1.to(device)only returns a new copy of x1 on say GPU without actually overwriting x1 in-place; so should probably change tox1=x1.to(device)instead, to make sure the model and data are on the same device.Thank you very much!