search

quantumlib / OpenFermion-Psi4

OpenFermion plugin to interface with the electronic structure package Psi4.
GNU Lesser General Public License v3.0
82 stars 45 forks source link

Psi4 calculation converges, Openfermionpsi4 plugin calculation doesn't #31

Closed TariniHardikar closed 6 years ago

TariniHardikar commented 6 years ago

This is a really ridiculous situation, but when I submit a SCF calculation for a very specific molecule (lattice of atoms) with Psi4, the calculation converges. When I submit it through the plugin, it doesn't converge (even though it runs for more iterations) and says maybe there's a seg fault with Psi4. To the best of my knowledge, all my parameters are exactly the same.

Is there anything that is changed when the plugin calls psi4? Or am I just missing something?

jarrodmcc commented 6 years ago

There is an option in the OpenFermion-Psi4 plugin to keep the input file and output file for the Psi4 calculation being run. If you switch this option to retain the input file on, then run the input file generated by the plugin independently, do you get the same error? My feeling is that this problem is most likely to be caused by either the input options differing subtly or the version of Psi4 being called by python is different than the version being called otherwise due to path issues, but hopefully this test will shed more light on the situation.

On Sun, Apr 22, 2018 at 5:15 PM, Tarini Hardikar notifications@github.com wrote:

This is a really ridiculous situation, but when I submit a SCF calculation for a very specific molecule (lattice of atoms) with Psi4, the calculation converges. When I submit it through the plugin, it doesn't converge (even though it runs for more iterations) and says maybe there's a seg fault with Psi4. To the best of my knowledge, all my parameters are exactly the same.

Is there anything that is changed when the plugin calls psi4? Or am I just missing something?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/quantumlib/OpenFermion-Psi4/issues/31, or mute the thread https://github.com/notifications/unsubscribe-auth/AHXFPNKUHFlM69ykIRotbh7f9LcCgqxbks5trR0ZgaJpZM4TfFps .

TariniHardikar commented 6 years ago

It turns out that psi4 geometry parser works differently at different instances?! So even though I had the same input geometry, the "pure" psi4 calculation and the plugin calculation had different parsed geometries (discussed here).

With this fix, I do get convergence using the plugin. Some statistics are still differently weirdly enough, but I think convergence is a good enough start for me.

Thank you!