Open yuan-gist opened 1 year ago
Sorry for the delayed reply. I have not used WSWQ before; does this refer to the overlap of the wavefunctions with the wavefunctions of another calculation at some perturbed configuration coordinate? Also, it would be helpful to have a better understanding of the problem. Do you have a minimal reproducing example of VASP and pawpyseed giving different values for what should be equivalent quantities? Also, how large is the error?
I will try my best to help but probably cannot make any significant contributions until March.
Since VASP can provide WSWQ, the overlap matrix elements, could you consider adding unit tests to compare with VASP calculations?
I tried this, but cannot reproduce the VASP WSWQ.