The code is based on the PySCF quantum chemistry package Author= Milaim Kas Theory and equations: Stasis Chuchurca and Milaim Kas
The method allows to solve the coupled cluster equations for an effective Hamiltonian H+L*Vexp where Vexp is a potential that compares calculated and given one-electron properties. For increasing value of the weight L, the obtained WF gives the best fitted one-electron property, thus minimising ^2.
- for ground state case: L1-ECW-CCS or L1-ECW-CCSD where L1 stands for L1 regularized solution.
- for excited state: ECW-CCS
The following n functionals have to be minimized: J_n = <Psi_nPsi_n> + L*Vexp + _1 Leading to n Schrödinger equations to be solved: EPsi_n> + sum{m} Vexp^{nm}|Psi_m> Different cases can be distinguished:
- Vexp^{nn} potentials contain one-electron properties related to state n
- Vexp^{nm} potentials contain one-electron transition properties related to the n->m transition
The Gs case corresponds to Vexp = Vexp^{00} ES case correspond to Vexp = Vexp^{nm} and Vexp^{nn}
The Couples Cluster formalism is applied to solve the set of couples SE.
See Theory.pdf file for more detailed.
In order to stick with the notation used in the literature, several confusion may occur. These are shown in the following points:
- "rdm1" and "gamma" are used indistinctly for the one-body reduced density matrix. Where "_calc" or "_pred" is added for the calculated one.
- "T1"/"t1" are the singles T equations and the singles t amplitudes, respectively.
- "L1"/"l1" may refer to the singles L or Lambda equations and the singles l or lambda amplitudes, respectively. When prone to confusion, "es_L1" may is used to specify the excited state case and/or "Lam" for the ground state.
- "L1 reg" stands for the L1 regularization method, where "alpha" is the coefficient of the L1 regularization term.
- "L" or "lambda" stands for the weight of the experimental potential.