Presence of "mini.nw" in reference_data/basis_sets/default slightly impedes development

[jlheflin]

With the file reference_data/basis_sets/default/mini.nw present, it makes it a tad bit harder to do development if only the sto-3g basis set is used in the master branch. With the file present, the resulting development workflow of generate_basis.py results in the mini basis set being generated, which leads to extra cleaning of the mini basis set files that are generated.

My suggestion is to either add the mini basis set data to the master branch, or to remove mini.nw to prevent generating the basis set files. What do you all think?

[jwaldrop107]

I think I lean towards removing mini.nw.

[ryanmrichard]

@jwaldrop107 do you recall why we had it? Wasn't it for the SAD guess?

[jwaldrop107]

It was, yes. It's not the default currently, but the precomputed densities are still in the reference_data. So we could keep it and just add the complete source files for mini to master. I don't have a strong opinion either way.

[ryanmrichard]

Realistically SAD should be the default guess. So I guess I would be in favor of adding it to master.

[jwaldrop107]

I meant MINI isn't the default basis for the SAD guess, it's currently STO-3G. I also don't think SAD is the default guess either, but that's a different point.

[ryanmrichard]

Oh. If STO-3G is the default for SAD I'd vote for deleting mini.