Orbital-weight included fitting in TBStudio

[jrhui]

This new feature requires sys.out to also be read, which provides crucial information about atom-orbital and k-paths of the unfold data. I have finished this part.

In the original code, sec30->Adouble1D[4] is designated to store DFT weight. However, the correct weight should be based on comparison with TB weight. So, I need to store the eigenvectors of TB model.

[jrhui]

When SOC is included, the weight is also spin-dependent. I would want to know a general situation of OpenMX output of *.unfold_orbup(dn). Does it name *.unfold_orb when a non-SOC calculation is performed? In what case will it name as *.unfold_orbdn?

[mohammadnakhaee]

I refactored the code in branch tbm_format. The variables are now more readable. You can use sec30->DFTEigValWeight instead of sec30->ArraysOf2DDouble[4]. And define there any varible you need. I will merge this branch to nomad branch.

Concerning the unfold_orbup I need more time to look up a few more things.

[mohammadnakhaee]

sec30->DFTEigValWeight is designed for the user painting weight. I am wondering how we can consider the orbital weights together with this?

Maybe (User painting weight) * (parsed weight) ?

[jrhui]

Yas, we can simpliy multiple the two factors.