Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

bsse_type='cp' does unnecessary monomer-basis calculations #1691

Open
hokru opened this issue Aug 5, 2019 · 6 comments
Open

bsse_type='cp' does unnecessary monomer-basis calculations #1691

hokru opened this issue Aug 5, 2019 · 6 comments

Comments

@hokru
Copy link
Contributor

@hokru hokru commented Aug 5, 2019

Psi4 1.4a2.dev66, Git: Rev {master} 6e67529

see topic. Only recently noticed this. It used to do only dimer-basis calculations.

molecule ne2 {
    0 1
    Ne 0 0 0
    --
    0 1
    Ne 0 0 2
}
set scf_type pk
energy('scf/3-21G', bsse_type="cp")

output:

        BSSE Treatment:                     cp
        Number of 1-body computations:     4
        Number of 2-body computations:     1
@alenaizan

This comment has been minimized.

Copy link
Contributor

@alenaizan alenaizan commented Aug 6, 2019

Yes, I updated this to do monomer-basis calculations. They are needed for correct counterpoise-corrected total energies. Maybe we should not compute them when the user requests interaction quantities with the option return_total_data=False. A new Psi4 driver is coming. It should have better control over which terms of the expansion are computed.

@hokru

This comment has been minimized.

Copy link
Contributor Author

@hokru hokru commented Aug 6, 2019

Not sure I follow. Monomer-basis results are not needed for CP-corrected interaction energies.
Now it seems the monomer-basis energies are calculated but not used at all in the final output.

@alenaizan

This comment has been minimized.

Copy link
Contributor

@alenaizan alenaizan commented Aug 6, 2019

I believe it is used in computing counterpoise-corrected total energies not interaction energies, which is for a dimer E = E^{AB}_{AB} - E^{AB}_{A} - E^{AB}_{B} + E^{A}_{A} + E^{B}_{B}. The table at the end of the output which prints counterpoise corrected energies should have the correct interaction and total quantities.

@hokru

This comment has been minimized.

Copy link
Contributor Author

@hokru hokru commented Aug 7, 2019

The new bsse_type='cp' does what bsse_type=['cp','nocp'] used to do, calculation wise, except the nocp interaction energies are not reported.

To safe computer time it was useful to only calculate the dimer-basis part and get CP-corrected interaction energies directly.

@alenaizan

This comment has been minimized.

Copy link
Contributor

@alenaizan alenaizan commented Aug 7, 2019

I agree we should allow the user to drop the computation of monomer-basis calculation. However, total counterpoise-corrected energies are sometimes needed, e.g. when doing geometry optimization, and so they should be computed when needed.

@dgasmith

This comment has been minimized.

Copy link
Member

@dgasmith dgasmith commented Nov 27, 2019

Before the release of 1.4 we should ensure that these monomer calculations are not included by default and are opt-in only to keep currently released behavior.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
3 participants
You can’t perform that action at this time.