Closed tfrederiksen closed 2 years ago
Thanks, about the zak:origin
, this came about from our initial discussion about this, and currently pol is testing its current functionality. If he agrees it works, then I will do the Berry/Zak phase using the R
gauge since that is just way way simpler ;)
In particular when we are dealing with the lowdin transformations as is also being tested now.
This was why I didn't ping you on the zak origin... :)
OK, thought you had already made the switch as it is now in main.
OK, thought you had already made the switch as it is now in main.
Yeah, it was just easier since it was going to be tested soonish, and I had another test that sort of confirmed it. So I think it should be safe, but in case Pol comes back with a negative answer, I'll have to revert. :)
I think this is ready to go in. Everything has been sorted out thanks to @tfrederiksen and @pfebrer :-D
This branch brings
hubbard
up to the current sisl version:parametrize
was changed in https://github.com/zerothi/sisl/commit/873fd40754fadfb845a047dbe67638d53e83b1c4.method='zak:origin'
was removed in https://github.com/zerothi/sisl/commit/57c3c607a0b0b068d41a31be9986dab6ad132b4f