Nick Papior

Thanks! This actually uncovered a bit more problematic issues, I'll ask for another review once completed. Thanks!

Ok, it got complicated, but the pbc was never really copied over, and this might have inflected problems here and there. I think it should be done now.!

I am thinking whether this would be more appropriate as a regular method: ```python def trace(self, axis1=0, axis2=1, use_overlap=False)... ``` then it can be used for more than 1 thing?...

> Does `np.trace` already work? Well, I haven't tested it, but I guess it *should*. > I don't know if it's a good idea to introduce the matrix multiplication on...

> > Hmm, I think we shouldn't do complicated functions that are allready standardized, e.g. matmul and trace. > > Yes, but the point is that `trace(matmul(X))` doesn't actually require...

This might have the problem that you'll need to bump the sisl version, if that is what you want that is fine, but users who are installing sisl (via pip),...

I am making a commit now that should re-enable the `Geometry.sc` handling, it would be nice if you could check that hubbard works, even with the lattice change.

I have tried some of the examples, and they seem to work just fine! :) Great! They raise a warning which they should! :)

There are still many references to SuperCell...

I would advice you to get this in, at some point (not too far) the `sc` arguments will probably die... ;)