Dynamical screening in BSE as perturbation #560
Conversation
|
@JensWehner this is the current WIP situation. Maybe you can have a look already. |
Codecov Report
@@ Coverage Diff @@
## master #560 +/- ##
========================================
+ Coverage 57.0% 57.2% +0.1%
========================================
Files 312 312
Lines 32296 32434 +138
========================================
+ Hits 18440 18565 +125
- Misses 13856 13869 +13
Continue to review full report at Codecov.
|
Co-authored-by: Jens <jenswehner@gmail.com>
Co-authored-by: Jens <jenswehner@gmail.com>
…y eigen::vector, still uses existing analysis function
|
Additional question: The |
|
Yeah that is a good question, I think you could add it to the orbitals file as a new entry. Overwriting the static bse result does not sound like a good idea, because in some sense it is a post-processing step. Storing the new energies is also not a lot of memory. |
include/votca/xtp/bse.h
Outdated
| Eigen::VectorXd Analyze_IndividualContribution(const QMStateType& type, | ||
| const Orbitals& orb, | ||
| const BSE_OPERATOR& H) const; | ||
| Eigen::VectorXd ExpectationValue_Operator(const QMStateType& type, |
There was a problem hiding this comment.
strictly speaking the signature should be something like const tools::EigenSystem,const BSE_OPERATOR& H)`
Working on tests, what about the options? I still have to add them to the default xml but otherwise they are passed through, aren't they? |
|
dftgwbse.xml, iqm, eqm, qmmm |
|
The nested xml packages was not such a bad idea afterall. |
Yeah, for those parts that are reused in different calculators, it has some advantages... |
Implements #3 by evaluating direct part of the BSE Hamiltonian at frequency of the static BSE solution and evaluating
delta Omega_S = <S|Hd(Omega_S)-Hd(0)|S>iteratively.Reuses the existing BSE analysis routines as much as possible.
Still to do:
Analyze_IndividualContributionto perform analysis only on a given state, not all.