Changes default values for empty states and adds warnings for SPHInX

[skatnagallu]

As a part of fixing the issues mentioned here, we thought it would be nice to have some warnings. Also we saw that there was discrepency between default empty states value and documentation of the function.

[skatnagallu]

This is the previous PR

[coveralls]

Pull Request Test Coverage Report for Build 10901773764

Details

• 2 of 6 (33.33%) changed or added relevant lines in 1 file are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.02%) to 70.88%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
pyiron_atomistics/sphinx/base.py	2	6	33.33%

Files with Coverage Reduction	New Missed Lines	%
pyiron_atomistics/sphinx/base.py	1	79.75%

Totals
Change from base Build 10896531720:	-0.02%
Covered Lines:	10778
Relevant Lines:	15206

---

💛 - Coveralls

[samwaseda]

This is what I don’t understand. You’re right that the default may not have been properly chosen, but the new one appears also arbitrary to me. Can you maybe tell why you chose this value?

[skatnagallu]

This is what I don’t understand. You’re right that the default may not have been properly chosen, but the new one appears also arbitrary to me. Can you maybe tell why you chose this value?

The default was already mentioned in the docstring of set_empty_states() function. I am not sure who originally implemented it. But the value I implemented is based on disscusion with @freyso and also what we saw in VASP, to match what was in the docstring.

[skatnagallu]

@samwaseda Are you happy with this?

[samwaseda]

ok I originally implemented it but I didn't realise that there was this discrepancy :D I'm still concerned about the warning, which will keep appearing to everyone, regardless of what kind of material we are talking about. I guess it's slightly unfair regarding other more serious warnings.

In addition, I don't quite understand the difference between the two print statements. Isn't one warning for the filling check enough?

[skatnagallu]

For convergence there are two convergence checks so two print statements. We thought the print statement for scf convergence would be a helpful hint for new users of sphinx.

[samwaseda]

For convergence there are two convergence checks so two print statements. We thought the print statement for scf convergence would be a helpful hint for new users of sphinx.

Sorry what I meant was scf_convergence is set with the nbands_convergence_check, so if the first print statement is given, then the second one is not required, right? Maybe I'm badly mistaken somewhere.

[skatnagallu]

scf_convergence is set from parsing of the log file here. And it doesn't have information on the nbands_convergence_check. The final convergence_check has both informations in it.