Adding fragment-by-fragment energy evaluation, without all rdms, for …

[lweisburn]

Add a version of oneshot that allows us to calculate the energy of the system on a by-fragment basis, so that we do not have to construct full rdms, for big systems. Tested for oneshot calculations of different BE levels using CCSD, with and without frozen core, in series and parallel, and different molecule types, turning on and off full rdm construction

[mscho527]

@lweisburn Do you want to modify the commit author / email (which I think was not set on the new account on telemachus)? Not super important though :)

[mscho527]

Note to myself: #10 needs to be updated...

1. total energies are not in ebe_tot, but are outputs of oneshot
2. for .misc.be2puffin, call oneshot with frag_energy=True

[lweisburn]

@lweisburn Do you want to modify the commit author / email (which I think was not set on the new account on telemachus)? Not super important though :)

I ~ think ~ I did this, thanks for pointing it out!

[lweisburn]

I believe I made all the changes we discussed. The code now, when running oneshot with calc_frag_energy=True, should return the energy calculated fragment-by-fragment as self.ebe_tot. It should also take the eri_file initialized in pbe

[mscho527]

1cd932193a0e62aa5c70aa5b6a152fed63b1541c has some unresolved merge conflicts and python caches. I can try to tidy up -- please avoid merging it to main now

[mscho527]

I think this changes are great! Nice work on parallelizing the energy computation - we might want to add this to kbe_devel as well at some point. Feel free to merge it with those tiny change.

That's a good point -- I didn't realize that energy evaluation is not parallelized in kbe_devel -- this would be a useful change