Complex conjugate

I am trying to rewrite a process which was written for numpy arrays to perform using torch tensors. In one point, I need to compute mat.conj().T I was looking around and it seems that bringing complex values to pytorch is a hot topic. I was wondering if there is an equivalent to .conj() in torch or I should simply go back and forth between numpy and torch to do this.

It looks like a complex conjugate operator torch.conj() was added a couple of months ago on master, but isn’t available in v1.3.1:

1 Like