You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I needed to get the values of the $\alpha$ dependent integrand in ph-AC and to do so ended up modifying the structure of the erpa classmethod in the excitation module.
However, these changes broke the vectorization that function had and are potentially the cause of current inefficiency in the evaluation of ph-AC.
@msricher could you suggest another/nicer way to introduce this feature that is maybe closer to the original structure?
I needed to get the values of the$\alpha$ dependent integrand in ph-AC and to do so ended up modifying the structure of the
erpa
classmethod in the excitation module.However, these changes broke the vectorization that function had and are potentially the cause of current inefficiency in the evaluation of ph-AC.
@msricher could you suggest another/nicer way to introduce this feature that is maybe closer to the original structure?
This is the commit for the changes 5f07062
Though, I think the original structure of the code is seen more clearly here:
https://github.com/gabrielasd/eomee/blob/939571c115e3acfd7d8e493a101dc77c9cf9a5f6/eomee/excitation.py#L89
In case it helps, the equation being implemented corresponds to the last box here:
https://github.com/gabrielasd/eomee/blob/refactor_ph_ac/docs/source/erpa.rst
(notice I have mislabelled the particle-hole equation as particle-particle.)
The text was updated successfully, but these errors were encountered: