Closed njzjz closed 1 month ago
Hi, after checking on this bug, there are a few things that I noted here:
torch.cross
is more flexible and allows you for more complex cross products, I believe the expected behavior is usually the one from torch.linalg.cross
implementation. In this discussion it explains why blindly using torch.cross
can lead you to an intersting bug-hunting. torch.svd
is deprecated and will be removed in future pytorch release. I am submitting a PR to attempt to solve this issue and updating the tutorial.
Add Link
https://pytorch.org/tutorials/beginner/introyt/tensors_deeper_tutorial.html
Describe the bug
Describe your environment
I don't actually run it but find the web page has such a warning.