pyscf.tdscf package

Submodules

pyscf.tdscf.dhf module

TDA and TDHF for no-pair DKS Hamiltonian

class pyscf.tdscf.dhf.TDA(mf, frozen=None)

Bases: TDBase

Tamm-Dancoff approximation

Attributes:

conv_tolfloat

Diagonalization convergence tolerance. Default is 1e-9.

nstatesint

Number of TD states to be computed. Default is 3.

Saved results:

convergedbool

Diagonalization converged or not

e1D array

excitation energy for each excited state.

xyA list of two 2D arrays

The two 2D arrays are Excitation coefficients X (shape [nocc,nvir]) and de-excitation coefficients Y (shape [nocc,nvir]) for each excited state. (X,Y) are normalized to 1/2 in RHF/RKS methods and normalized to 1 for UHF/UKS methods. In the TDA calculation, Y = 0.

gen_vind(mf=None)

Generate function to compute Ax

get_init_guess(mf, nstates=None, wfnsym=None)

kernel(x0=None, nstates=None)

TDA diagonalization solver

singlet = None

class pyscf.tdscf.dhf.TDBase(mf, frozen=None)

Bases: TDBase

analyze(verbose=None)

get_ab(mf=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

get_nto(state=1, threshold=0.3, verbose=None)

Natural transition orbital analysis.

The natural transition density matrix between ground state and excited state \(Tia = \langle \Psi_{ex} | i a^\dagger | \Psi_0 \rangle\) can be transformed to diagonal form through SVD \(T = O \sqrt{\lambda} V^\dagger\). O and V are occupied and virtual natural transition orbitals. The diagonal elements \(\lambda\) are the weights of the occupied-virtual orbital pair in the excitation.

Ref: Martin, R. L., JCP, 118, 4775-4777

Note in the TDHF/TDDFT calculations, the excitation part (X) is interpreted as the CIS coefficients and normalized to 1. The de-excitation part (Y) is ignored.

Args:

tdobj : TDA, or TDHF, or TDDFT object

stateint

Excited state ID. state = 1 means the first excited state. If state < 0, state ID is counted from the last excited state.

Kwargs:

thresholdfloat

Above which the NTO coefficients will be printed in the output.

Returns:

A list (weights, NTOs). NTOs are natural orbitals represented in AO basis. The first N_occ NTOs are occupied NTOs and the rest are virtual NTOs.

nuc_grad_method()

class pyscf.tdscf.dhf.TDHF(mf, frozen=None)

Bases: TDBase

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None)

get_precond(hdiag)

kernel(x0=None, nstates=None)

TDHF diagonalization with non-Hermitian eigenvalue solver

singlet = None

pyscf.tdscf.dhf.analyze(tdobj, verbose=None)

pyscf.tdscf.dhf.gen_tda_hop(mf, fock_ao=None, with_nlc=True)

A x

pyscf.tdscf.dhf.gen_tda_operation(mf, fock_ao=None, with_nlc=True)

A x

pyscf.tdscf.dhf.gen_tdhf_operation(mf, fock_ao=None, with_nlc=True)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

pyscf.tdscf.dhf.get_ab(mf, mo_energy=None, mo_coeff=None, mo_occ=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

pyscf.tdscf.dhf.get_nto(tdobj, state=1, threshold=0.3, verbose=None)

pyscf.tdscf.dks module

pyscf.tdscf.dks.RPA

alias of TDDFT

class pyscf.tdscf.dks.TDA(mf, frozen=None)

Bases: TDA

class pyscf.tdscf.dks.TDDFT(mf, frozen=None)

Bases: TDHF

pyscf.tdscf.dks.TDDKS

alias of TDDFT

pyscf.tdscf.ghf module

pyscf.tdscf.ghf.CIS

alias of TDA

pyscf.tdscf.ghf.RPA

alias of TDHF

class pyscf.tdscf.ghf.TDA(mf, frozen=None)

Bases: TDBase

Tamm-Dancoff approximation

Attributes:

conv_tolfloat

Diagonalization convergence tolerance. Default is 1e-9.

nstatesint

Number of TD states to be computed. Default is 3.

Saved results:

convergedbool

Diagonalization converged or not

e1D array

excitation energy for each excited state.

xyA list of two 2D arrays

The two 2D arrays are Excitation coefficients X (shape [nocc,nvir]) and de-excitation coefficients Y (shape [nocc,nvir]) for each excited state. (X,Y) are normalized to 1/2 in RHF/RKS methods and normalized to 1 for UHF/UKS methods. In the TDA calculation, Y = 0.

gen_vind(mf=None)

Generate function to compute Ax

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDA diagonalization solver

singlet = None

class pyscf.tdscf.ghf.TDBase(mf, frozen=None)

Bases: TDBase

analyze(verbose=None)

get_ab(mf=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

get_nto(state=1, threshold=0.3, verbose=None)

Natural transition orbital analysis.

The natural transition density matrix between ground state and excited state \(Tia = \langle \Psi_{ex} | i a^\dagger | \Psi_0 \rangle\) can be transformed to diagonal form through SVD \(T = O \sqrt{\lambda} V^\dagger\). O and V are occupied and virtual natural transition orbitals. The diagonal elements \(\lambda\) are the weights of the occupied-virtual orbital pair in the excitation.

Ref: Martin, R. L., JCP, 118, 4775-4777

Note in the TDHF/TDDFT calculations, the excitation part (X) is interpreted as the CIS coefficients and normalized to 1. The de-excitation part (Y) is ignored.

Args:

tdobj : TDA, or TDHF, or TDDFT object

stateint

Excited state ID. state = 1 means the first excited state. If state < 0, state ID is counted from the last excited state.

Kwargs:

thresholdfloat

Above which the NTO coefficients will be printed in the output.

Returns:

A list (weights, NTOs). NTOs are natural orbitals represented in AO basis. The first N_occ NTOs are occupied NTOs and the rest are virtual NTOs.

nuc_grad_method()

pyscf.tdscf.ghf.TDGHF

alias of TDHF

class pyscf.tdscf.ghf.TDHF(mf, frozen=None)

Bases: TDBase

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDHF diagonalization with non-Hermitian eigenvalue solver

singlet = None

pyscf.tdscf.ghf.analyze(tdobj, verbose=None)

pyscf.tdscf.ghf.gen_tda_hop(mf, fock_ao=None, wfnsym=None, with_nlc=True)

A x

Kwargs:

wfnsymint or str

Point group symmetry irrep symbol or ID for excited CIS wavefunction.

pyscf.tdscf.ghf.gen_tda_operation(mf, fock_ao=None, wfnsym=None, with_nlc=True)

A x

Kwargs:

wfnsymint or str

Point group symmetry irrep symbol or ID for excited CIS wavefunction.

pyscf.tdscf.ghf.gen_tdhf_operation(mf, fock_ao=None, wfnsym=None, with_nlc=True)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

pyscf.tdscf.ghf.get_ab(mf, frozen=None, mo_energy=None, mo_coeff=None, mo_occ=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

pyscf.tdscf.ghf.get_nto(tdobj, state=1, threshold=0.3, verbose=None)

pyscf.tdscf.gks module

class pyscf.tdscf.gks.CasidaTDDFT(mf, frozen=None)

Bases: TDDFT, TDA

Solve the Casida TDDFT formula (A-B)(A+B)(X+Y) = (X+Y)w^2

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDDFT diagonalization solver

nuc_grad_method()

pyscf.tdscf.gks.RPA

alias of TDDFT

class pyscf.tdscf.gks.TDA(mf, frozen=None)

Bases: TDA

class pyscf.tdscf.gks.TDDFT(mf, frozen=None)

Bases: TDHF

pyscf.tdscf.gks.TDDFTNoHybrid

alias of CasidaTDDFT

pyscf.tdscf.gks.TDGKS

alias of TDDFT

pyscf.tdscf.gks.tddft(mf, frozen=None)

Driver to create TDDFT or CasidaTDDFT object

pyscf.tdscf.rhf module

pyscf.tdscf.rhf.CIS

alias of TDA

pyscf.tdscf.rhf.RPA

alias of TDHF

class pyscf.tdscf.rhf.TDA(mf, frozen=None)

Bases: TDBase

Tamm-Dancoff approximation

Attributes:

conv_tolfloat

Diagonalization convergence tolerance. Default is 1e-9.

nstatesint

Number of TD states to be computed. Default is 3.

Saved results:

convergedbool

Diagonalization converged or not

e1D array

excitation energy for each excited state.

xyA list of two 2D arrays

The two 2D arrays are Excitation coefficients X (shape [nocc,nvir]) and de-excitation coefficients Y (shape [nocc,nvir]) for each excited state. (X,Y) are normalized to 1/2 in RHF/RKS methods and normalized to 1 for UHF/UKS methods. In the TDA calculation, Y = 0.

gen_vind(mf=None)

Generate function to compute Ax

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

Generate initial guess for TDA

Kwargs:

nstatesint

The number of initial guess vectors.

wfnsymint or str

The irrep label or ID of the wavefunction.

return_symmetrybool

Whether to return symmetry labels for initial guess vectors.

init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDA diagonalization solver

to_gpu(out=None)

Convert a method to its corresponding GPU variant, and recursively converts all attributes of a method to cupy objects or gpu4pyscf objects.

class pyscf.tdscf.rhf.TDBase(mf, frozen=None)

Bases: StreamObject

analyze(verbose=None)

as_scanner()

Generating a scanner/solver for TDA/TDHF/TDDFT PES.

The returned solver is a function. This function requires one argument “mol” as input and returns total TDA/TDHF/TDDFT energy.

The solver will automatically use the results of last calculation as the initial guess of the new calculation. All parameters assigned in the TDA/TDDFT and the underlying SCF objects (conv_tol, max_memory etc) are automatically applied in the solver.

Note scanner has side effects. It may change many underlying objects (_scf, with_df, with_x2c, …) during calculation.

Examples:

>>> from pyscf import gto, scf, tdscf
>>> mol = gto.M(atom='H 0 0 0; F 0 0 1')
>>> td_scanner = tdscf.TDHF(scf.RHF(mol)).as_scanner()
>>> de = td_scanner(gto.M(atom='H 0 0 0; F 0 0 1.1'))
[ 0.34460866  0.34460866  0.7131453 ]
>>> de = td_scanner(gto.M(atom='H 0 0 0; F 0 0 1.5'))
[ 0.14844013  0.14844013  0.47641829]

check_sanity()

Check input of class/object attributes, check whether a class method is overwritten. It does not check the attributes which are prefixed with “_”. The return value of method set is the object itself. This allows a series of functions/methods to be executed in pipe.

conv_tol = 1e-05

deg_eia_thresh = 0.001

dump_flags(verbose=None)

property e_tot

Excited state energies

exclude_nlc = True

gen_response(*args, **kwargs)

Generate linear response function to compute A*x

gen_vind(mf=None)

get_ab(mf=None, frozen=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

Ref: Chem Phys Lett, 256, 454

get_frozen_mask()

Get boolean mask for the restricted reference orbitals.

In the returned boolean (mask) array of frozen orbital indices, the element is False if it corresponds to the frozen orbital.

See mp2.get_frozen_mask

get_nto(state=1, threshold=0.3, verbose=None)

Natural transition orbital analysis.

The natural transition density matrix between ground state and excited state \(Tia = \langle \Psi_{ex} | i a^\dagger | \Psi_0 \rangle\) can be transformed to diagonal form through SVD \(T = O \sqrt{\lambda} V^\dagger\). O and V are occupied and virtual natural transition orbitals. The diagonal elements \(\lambda\) are the weights of the occupied-virtual orbital pair in the excitation.

Ref: Martin, R. L., JCP, 118, 4775-4777

Note in the TDHF/TDDFT calculations, the excitation part (X) is interpreted as the CIS coefficients and normalized to 1. The de-excitation part (Y) is ignored.

Args:

tdobj : TDA, or TDHF, or TDDFT object

stateint

Excited state ID. state = 1 means the first excited state. If state < 0, state ID is counted from the last excited state.

Kwargs:

thresholdfloat

Above which the NTO coefficients will be printed in the output.

Returns:

A list (weights, NTOs). NTOs are natural orbitals represented in AO basis. The first N_occ NTOs are occupied NTOs and the rest are virtual NTOs.

get_precond(hdiag)

level_shift = 0

lindep = 1e-12

max_cycle = 100

property nroots

nstates = 3

nuc_grad_method()

oscillator_strength(e=None, xy=None, gauge='length', order=0)

positive_eig_threshold = 0.001

reset(mol=None)

set_frozen(method='auto', window=(-1000.0, 1000.0))

singlet = True

to_gpu()

transition_dipole(xy=None)

Transition dipole moments in the length gauge

transition_magnetic_dipole(xy=None)

Transition magnetic dipole moments (imaginary part only)

transition_magnetic_quadrupole(xy=None)

Transition magnetic quadrupole moments (imaginary part only)

transition_octupole(xy=None)

Transition octupole moments in the length gauge

transition_quadrupole(xy=None)

Transition quadrupole moments in the length gauge

transition_velocity_dipole(xy=None)

Transition dipole moments in the velocity gauge (imaginary part only)

transition_velocity_octupole(xy=None)

Transition octupole moments in the velocity gauge (imaginary part only)

transition_velocity_quadrupole(xy=None)

Transition quadrupole moments in the velocity gauge (imaginary part only)

class pyscf.tdscf.rhf.TDHF(mf, frozen=None)

Bases: TDBase

Time-dependent Hartree-Fock

Attributes:

conv_tolfloat

Diagonalization convergence tolerance. Default is 1e-4.

nstatesint

Number of TD states to be computed. Default is 3.

Saved results:

convergedbool

Diagonalization converged or not

e1D array

excitation energy for each excited state.

xyA list of two 2D arrays

The two 2D arrays are Excitation coefficients X (shape [nocc,nvir]) and de-excitation coefficients Y (shape [nocc,nvir]) for each excited state. (X,Y) are normalized to 1/2 in RHF/RKS methods and normalized to 1 for UHF/UKS methods. In the TDA calculation, Y = 0.

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDHF diagonalization with non-Hermitian eigenvalue solver

nuc_grad_method()

to_gpu(out=None)

Convert a method to its corresponding GPU variant, and recursively converts all attributes of a method to cupy objects or gpu4pyscf objects.

pyscf.tdscf.rhf.TDRHF

alias of TDHF

class pyscf.tdscf.rhf.TD_Scanner(td)

Bases: SinglePointScanner

pyscf.tdscf.rhf.analyze(tdobj, verbose=None)

pyscf.tdscf.rhf.as_scanner(td)

Generating a scanner/solver for TDA/TDHF/TDDFT PES.

The returned solver is a function. This function requires one argument “mol” as input and returns total TDA/TDHF/TDDFT energy.

The solver will automatically use the results of last calculation as the initial guess of the new calculation. All parameters assigned in the TDA/TDDFT and the underlying SCF objects (conv_tol, max_memory etc) are automatically applied in the solver.

Note scanner has side effects. It may change many underlying objects (_scf, with_df, with_x2c, …) during calculation.

Examples:

>>> from pyscf import gto, scf, tdscf
>>> mol = gto.M(atom='H 0 0 0; F 0 0 1')
>>> td_scanner = tdscf.TDHF(scf.RHF(mol)).as_scanner()
>>> de = td_scanner(gto.M(atom='H 0 0 0; F 0 0 1.1'))
[ 0.34460866  0.34460866  0.7131453 ]
>>> de = td_scanner(gto.M(atom='H 0 0 0; F 0 0 1.5'))
[ 0.14844013  0.14844013  0.47641829]

pyscf.tdscf.rhf.gen_tda_hop(mf, fock_ao=None, singlet=True, wfnsym=None, with_nlc=True)

Generate function to compute A x

Kwargs:

wfnsymint or str

Point group symmetry irrep symbol or ID for excited CIS wavefunction.

with_nlcboolean

Whether to skip the NLC contribution

pyscf.tdscf.rhf.gen_tda_operation(mf, fock_ao=None, singlet=True, wfnsym=None, with_nlc=True)

Generate function to compute A x

Kwargs:

wfnsymint or str

Point group symmetry irrep symbol or ID for excited CIS wavefunction.

with_nlcboolean

Whether to skip the NLC contribution

pyscf.tdscf.rhf.gen_tdhf_operation(mf, fock_ao=None, singlet=True, wfnsym=None, with_nlc=True)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

pyscf.tdscf.rhf.get_ab(mf, frozen=None, mo_energy=None, mo_coeff=None, mo_occ=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

Ref: Chem Phys Lett, 256, 454

pyscf.tdscf.rhf.get_frozen_mask(td)

Get boolean mask for the restricted reference orbitals.

In the returned boolean (mask) array of frozen orbital indices, the element is False if it corresponds to the frozen orbital.

See mp2.get_frozen_mask

pyscf.tdscf.rhf.get_nto(tdobj, state=1, threshold=0.3, verbose=None)

Natural transition orbital analysis.

The natural transition density matrix between ground state and excited state \(Tia = \langle \Psi_{ex} | i a^\dagger | \Psi_0 \rangle\) can be transformed to diagonal form through SVD \(T = O \sqrt{\lambda} V^\dagger\). O and V are occupied and virtual natural transition orbitals. The diagonal elements \(\lambda\) are the weights of the occupied-virtual orbital pair in the excitation.

Ref: Martin, R. L., JCP, 118, 4775-4777

Note in the TDHF/TDDFT calculations, the excitation part (X) is interpreted as the CIS coefficients and normalized to 1. The de-excitation part (Y) is ignored.

Args:

tdobj : TDA, or TDHF, or TDDFT object

stateint

Excited state ID. state = 1 means the first excited state. If state < 0, state ID is counted from the last excited state.

Kwargs:

thresholdfloat

Above which the NTO coefficients will be printed in the output.

Returns:

A list (weights, NTOs). NTOs are natural orbitals represented in AO basis. The first N_occ NTOs are occupied NTOs and the rest are virtual NTOs.

pyscf.tdscf.rhf.oscillator_strength(tdobj, e=None, xy=None, gauge='length', order=0)

pyscf.tdscf.rhf.transition_dipole(tdobj, xy=None)

Transition dipole moments in the length gauge

pyscf.tdscf.rhf.transition_magnetic_dipole(tdobj, xy=None)

Transition magnetic dipole moments (imaginary part only)

pyscf.tdscf.rhf.transition_magnetic_quadrupole(tdobj, xy=None)

Transition magnetic quadrupole moments (imaginary part only)

pyscf.tdscf.rhf.transition_octupole(tdobj, xy=None)

Transition octupole moments in the length gauge

pyscf.tdscf.rhf.transition_quadrupole(tdobj, xy=None)

Transition quadrupole moments in the length gauge

pyscf.tdscf.rhf.transition_velocity_dipole(tdobj, xy=None)

Transition dipole moments in the velocity gauge (imaginary part only)

pyscf.tdscf.rhf.transition_velocity_octupole(tdobj, xy=None)

Transition octupole moments in the velocity gauge (imaginary part only)

pyscf.tdscf.rhf.transition_velocity_quadrupole(tdobj, xy=None)

Transition quadrupole moments in the velocity gauge (imaginary part only)

pyscf.tdscf.rks module

class pyscf.tdscf.rks.CasidaTDDFT(mf, frozen=None)

Bases: TDDFT, TDA

Solve the Casida TDDFT formula (A-B)(A+B)(X+Y) = (X+Y)w^2

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

Generate initial guess for TDA

Kwargs:

nstatesint

The number of initial guess vectors.

wfnsymint or str

The irrep label or ID of the wavefunction.

return_symmetrybool

Whether to return symmetry labels for initial guess vectors.

get_precond(hdiag)

kernel(x0=None, nstates=None)

TDDFT diagonalization solver

nuc_grad_method()

pyscf.tdscf.rks.RPA

alias of TDDFT

class pyscf.tdscf.rks.TDA(mf, frozen=None)

Bases: TDA

nuc_grad_method()

class pyscf.tdscf.rks.TDDFT(mf, frozen=None)

Bases: TDHF

nuc_grad_method()

pyscf.tdscf.rks.TDDFTNoHybrid

alias of CasidaTDDFT

pyscf.tdscf.rks.TDH

alias of dRPA

pyscf.tdscf.rks.TDRKS

alias of TDDFT

class pyscf.tdscf.rks.dRPA(mf, frozen=None)

Bases: CasidaTDDFT

class pyscf.tdscf.rks.dTDA(mf, frozen=None)

Bases: TDA

pyscf.tdscf.rks.tddft(mf, frozen=None)

Driver to create TDDFT or CasidaTDDFT object

pyscf.tdscf.uhf module

pyscf.tdscf.uhf.CIS

alias of TDA

pyscf.tdscf.uhf.RPA

alias of TDHF

class pyscf.tdscf.uhf.TDA(mf, frozen=None)

Bases: TDBase

Tamm-Dancoff approximation

Attributes:

conv_tolfloat

Diagonalization convergence tolerance. Default is 1e-9.

nstatesint

Number of TD states to be computed. Default is 3.

Saved results:

convergedbool

Diagonalization converged or not

e1D array

excitation energy for each excited state.

xyA list of two 2D arrays

The two 2D arrays are Excitation coefficients X (shape [nocc,nvir]) and de-excitation coefficients Y (shape [nocc,nvir]) for each excited state. (X,Y) are normalized to 1/2 in RHF/RKS methods and normalized to 1 for UHF/UKS methods. In the TDA calculation, Y = 0.

gen_vind(mf=None)

Generate function to compute Ax

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDA diagonalization solver

singlet = None

to_gpu(out=None)

Convert a method to its corresponding GPU variant, and recursively converts all attributes of a method to cupy objects or gpu4pyscf objects.

class pyscf.tdscf.uhf.TDBase(mf, frozen=None)

Bases: TDBase

analyze(verbose=None)

get_ab(mf=None, frozen=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

Spin symmetry is considered in the returned A, B lists. List A has three items: (A_aaaa, A_aabb, A_bbbb). A_bbaa = A_aabb.transpose(2,3,0,1). B has three items: (B_aaaa, B_aabb, B_bbbb). B_bbaa = B_aabb.transpose(2,3,0,1).

get_frozen_mask()

Get boolean mask for the unrestricted reference orbitals.

In the returned boolean (mask) array of frozen orbital indices, the element is False if it corresponds to the frozen orbital.

get_nto(state=1, threshold=0.3, verbose=None)

Natural transition orbital analysis.

The natural transition density matrix between ground state and excited state \(Tia = \langle \Psi_{ex} | i a^\dagger | \Psi_0 \rangle\) can be transformed to diagonal form through SVD \(T = O \sqrt{\lambda} V^\dagger\). O and V are occupied and virtual natural transition orbitals. The diagonal elements \(\lambda\) are the weights of the occupied-virtual orbital pair in the excitation.

Ref: Martin, R. L., JCP, 118, 4775-4777

Note in the TDHF/TDDFT calculations, the excitation part (X) is interpreted as the CIS coefficients and normalized to 1. The de-excitation part (Y) is ignored.

Args:

stateint

Excited state ID. state = 1 means the first excited state. If state < 0, state ID is counted from the last excited state.

Kwargs:

thresholdfloat

Above which the NTO coefficients will be printed in the output.

Returns:

A list (weights, NTOs). NTOs are natural orbitals represented in AO basis. The first N_occ NTOs are occupied NTOs and the rest are virtual NTOs.

nuc_grad_method()

class pyscf.tdscf.uhf.TDHF(mf, frozen=None)

Bases: TDBase

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

kernel(x0=None, nstates=None)

TDHF diagonalization with non-Hermitian eigenvalue solver

singlet = None

to_gpu(out=None)

Convert a method to its corresponding GPU variant, and recursively converts all attributes of a method to cupy objects or gpu4pyscf objects.

pyscf.tdscf.uhf.TDUHF

alias of TDHF

pyscf.tdscf.uhf.analyze(tdobj, verbose=None)

pyscf.tdscf.uhf.gen_tda_hop(mf, fock_ao=None, wfnsym=None, with_nlc=True)

A x

Kwargs:

wfnsymint or str

Point group symmetry irrep symbol or ID for excited CIS wavefunction.

pyscf.tdscf.uhf.gen_tda_operation(mf, fock_ao=None, wfnsym=None, with_nlc=True)

A x

Kwargs:

wfnsymint or str

Point group symmetry irrep symbol or ID for excited CIS wavefunction.

pyscf.tdscf.uhf.gen_tdhf_operation(mf, fock_ao=None, wfnsym=None, with_nlc=True)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

pyscf.tdscf.uhf.get_ab(mf, frozen=None, mo_energy=None, mo_coeff=None, mo_occ=None)

A and B matrices for TDDFT response function.

A[i,a,j,b] = delta_{ab}delta_{ij}(E_a - E_i) + (ai||jb) B[i,a,j,b] = (ai||bj)

Spin symmetry is considered in the returned A, B lists. List A has three items: (A_aaaa, A_aabb, A_bbbb). A_bbaa = A_aabb.transpose(2,3,0,1). B has three items: (B_aaaa, B_aabb, B_bbbb). B_bbaa = B_aabb.transpose(2,3,0,1).

pyscf.tdscf.uhf.get_frozen_mask(td)

Get boolean mask for the unrestricted reference orbitals.

In the returned boolean (mask) array of frozen orbital indices, the element is False if it corresponds to the frozen orbital.

pyscf.tdscf.uhf.get_nto(tdobj, state=1, threshold=0.3, verbose=None)

Natural transition orbital analysis.

The natural transition density matrix between ground state and excited state \(Tia = \langle \Psi_{ex} | i a^\dagger | \Psi_0 \rangle\) can be transformed to diagonal form through SVD \(T = O \sqrt{\lambda} V^\dagger\). O and V are occupied and virtual natural transition orbitals. The diagonal elements \(\lambda\) are the weights of the occupied-virtual orbital pair in the excitation.

Ref: Martin, R. L., JCP, 118, 4775-4777

Note in the TDHF/TDDFT calculations, the excitation part (X) is interpreted as the CIS coefficients and normalized to 1. The de-excitation part (Y) is ignored.

Args:

stateint

Excited state ID. state = 1 means the first excited state. If state < 0, state ID is counted from the last excited state.

Kwargs:

thresholdfloat

Above which the NTO coefficients will be printed in the output.

Returns:

A list (weights, NTOs). NTOs are natural orbitals represented in AO basis. The first N_occ NTOs are occupied NTOs and the rest are virtual NTOs.

pyscf.tdscf.uks module

class pyscf.tdscf.uks.CasidaTDDFT(mf, frozen=None)

Bases: TDDFT, TDA

Solve the Casida TDDFT formula (A-B)(A+B)(X+Y) = (X+Y)w^2

gen_vind(mf=None)

Generate function to compute

[ A B ][X] [-B* -A*][Y]

get_init_guess(mf, nstates=None, wfnsym=None, return_symmetry=False)

get_precond(hdiag)

kernel(x0=None, nstates=None)

TDDFT diagonalization solver

nuc_grad_method()

pyscf.tdscf.uks.RPA

alias of TDDFT

class pyscf.tdscf.uks.TDA(mf, frozen=None)

Bases: TDA

nuc_grad_method()

class pyscf.tdscf.uks.TDDFT(mf, frozen=None)

Bases: TDHF

nuc_grad_method()

pyscf.tdscf.uks.TDDFTNoHybrid

alias of CasidaTDDFT

pyscf.tdscf.uks.TDH

alias of dRPA

pyscf.tdscf.uks.TDUKS

alias of TDDFT

class pyscf.tdscf.uks.dRPA(mf, frozen=None)

Bases: CasidaTDDFT

class pyscf.tdscf.uks.dTDA(mf, frozen=None)

Bases: TDA

pyscf.tdscf.uks.tddft(mf, frozen=None)

Driver to create TDDFT or CasidaTDDFT object

Module contents

pyscf.tdscf.RPA(mf, frozen=None)

pyscf.tdscf.TD(mf, frozen=None)

pyscf.tdscf.TDA(mf, frozen=None)

pyscf.tdscf.TDDFT(mf, frozen=None)

pyscf.tdscf.TDHF(mf, frozen=None)

pyscf.tdscf.dRPA(mf, frozen=None)

pyscf.tdscf.dTDA(mf, frozen=None)