I have tried to use read_hr function on both windows and mac. It works smoothly on Mac, while it takes forever on windows(>15mins). It should not be related to computer performance as the windows computer should perform better than macbook. When I interrupt the kernal it always stops at this line:

KeyboardInterrupt Traceback (most recent call last)
in
----> 1 HR, Rvectors, Rweights = w90io.read_hr('wannier90_hr.dat')

D:\machine learning\python\lib\site-packages\wannier90_utils-0.1.0-py3.7.egg\w90utils\io_hr.py in read_hr(fname)
23
24 istart_hr = 3+ndegen_lines
---> 25 raw_data = np.fromstring(''.join(contents[istart_hr:]), sep='\n').reshape((-1, 7))
26
27 hr = raw_data[:, 5] + 1j*raw_data[:, 6]

KeyboardInterrupt:

I've encountered a case where the Wannier centers computed by Wannier90 and wannier90-utils are inconsistent. Unfortunately I'm not allowed to share the input files[*], but the root cause is that in

the np.imag(np.log(mii[:, i])) is well-defined only up to multiples of 2π.

This is the same "branch cut" problem that is worked around in Wannier90 by setting the guiding_centres=.true. option.

I'm not sure what the best approach would be here - do you think it would make sense to add a similar option that selects the closest Wannier center to a given "reference"?

[*] I'll try to come up with some other example that reproduces this behavior..