[ase-users] ORCA-calculator for ASE?

[Ask Hjorth Larsen]

Am Di., 7. Jan. 2020 um 18:45 Uhr schrieb Asmus Ougaard Dohn <asod at hi.is>:
>
> Dear Ask,
>
> (sorry for breaking threadedness, my stupid mail client is, well, stupid).
>
> I forgot to link to my modified version of the calculator, where I think I actually already moved the io into ase.io.orca, hurray! https://gitlab.com/asod/ase/blob/orca/ase/calculators/orca.py
>
> What FileIO calc should I be inspired by for making tests? Can I make any meaningful ones without ORCA at all

I recently wrote some tests for Abinit which live in
test/abinit/main.py.  They try to exercise the main output
combinations (spin vs no spin, kpoints vs no kpoints) to cover all the
parsing code but don't really do any verification.  So it's okay but
not all the way there.

Now that we are moving to pytest we can do much better.  I'll try to
get the main pytest changes merged tonight so we can start getting
serious ... argh so much work.  Please feel free to get started and
ping me on Gitlab whenever relevant.

Best regards
Ask


>
> Best,
>
> Asmus
>
> ________________________________
> From: Ask Hjorth Larsen <asklarsen at gmail.com>
> Sent: Tuesday, January 7, 2020 5:40 PM
> To: Asmus Ougaard Dohn <asod at hi.is>
> Cc: ase-users at listserv.fysik.dtu.dk <ase-users at listserv.fysik.dtu.dk>
> Subject: Re: [ase-users] ORCA-calculator for ASE?
>
> Dear Asmus,
>
> Am Di., 7. Jan. 2020 um 18:07 Uhr schrieb Asmus Ougaard Dohn via
> ase-users <ase-users at listserv.fysik.dtu.dk>:
> >
> > Dear ASE users and developers,
> >
> > Is there any interest in an ORCA-calculator in ASE?
> >
> > I discovered some time ago that Ragnar Björnsson has made one (https://sites.google.com/site/orcainputlibrary/qm-mm/orca-ase), so talked to him, and he gave me permission to 'steal' it and put it into ASE.  Respect!
> >
> > I have a branch on my own ASE-fork where a added a bit of QM/MM functionality (as well as an xtb interface which I will get rid of before making a merge request, since they've gotten their own in the meantime), but I've held back just making a merge-request since I'm not quite sure what the best-practice is when it comes to adding entirely new calculators.
> >
> > Any and all comments welcome.
>
> I think the calculator looks good.  At the ASE workshop we discussed
> how "less is more", in the sense that we'd like to keep the
> calculators simple.  This calculator is simple because it is short and
> doesn't do any magical work.  My only comment is that the calculator's
> methods does some reading and writing.  It would be best to move that
> to `ase.io.orca`.
>
> We would need tests of course, as well as updates to the release notes
> and documentation.
>
> I don't know if the 'academic' license is sufficient for us to build a
> docker image.  That would be ideal.
>
> Best regards
> Ask
>
> >
> > Best,
> >
> > Asmus
> >
> >
> >
> > Asmus Ougaard Dohn
> >
> > Postdoc | DTU Physics, Technical University of Denmark
> >
> > Assistant Research Professor | Science Institute, University of Iceland
> >
> > asod at fysik.dtu.dk | asod at hi.is | +45 25 30 12 37
> >
> >
> >
> > _______________________________________________
> > ase-users mailing list
> > ase-users at listserv.fysik.dtu.dk
> > https://listserv.fysik.dtu.dk/mailman/listinfo/ase-users