Which basis sets should be included in the SDE by default?

[jboschen]

This issue is intended to house discussion about the question in the title.

https://github.com/NWChemEx-Project/SimulationDevelopmentEnvironment/blob/master/reference_data/generate_basis.py will read a list of files containing basis sets and parse them into C++ structures for use in: https://github.com/NWChemEx-Project/SimulationDevelopmentEnvironment/blob/master/SDE/Defaults/DefaultBases.cpp

If we come to some conclusions I will add those defaults in this PR: https://github.com/NWChemEx-Project/SimulationDevelopmentEnvironment/pull/13

[colleeneb]

As many as possible? Or whatever's in the Basis Set Exchange?

[ryanmrichard]

To clarify a bit. Every basis set in EMSL is included in SDE as a file, which can be read in if the user so desires (the same mechanism could be used for user generated basis sets too). The heart of the current issue concerns the fact that I took it one step further and wrote a script that will take some of those files and put them in a giant C++ file.

My two sense: I'd say that at the very least all Pople, Dunning, and Alhrichs basis sets should be in that file. Being blunt I don't think I've seen a modern computational chemistry paper use anything else.

[twindus]

I would include the ECPs (all of them since they all seem to be used in some way or another) and the fitting basis sets (which are mostly captured in the three listed by @ryanmrichard ). Some people like the June basis sets, but I don't have a strong opinion on those.

[ryanmrichard]

@jboschen can this be closed?

[jboschen]

@ryanmrichard sure.