bofin package¶

Submodules¶

bofin.heom module¶

This module provides exact solvers for a system-bath setup using the hierarchy equations of motion (HEOM).

class bofin.heom.BosonicHEOMSolver(H_sys, coup_op, ckAR, ckAI, vkAR, vkAI, N_cut, options=None)[source]¶

Bases: object

This is a class for solvers that use the HEOM method for calculating the dynamics evolution. There are many references for this. A good introduction, and perhaps closest to the notation used here is: DOI:10.1103/PhysRevLett.104.250401 A more canonical reference, with full derivation is: DOI: 10.1103/PhysRevA.41.6676 The method can compute open system dynamics without using any Markovian or rotating wave approximation (RWA) for systems where the bath correlations can be approximated to a sum of complex exponentials. The method builds a matrix of linked differential equations, which are then solved used the same ODE solvers as other qutip solvers (e.g. mesolve)/

H_sys¶

System Hamiltonian Or Liouvillian Or QobjEvo Or list of Hamiltonians with time dependence

Format for input (if list): [time_independent_part, [H1, time_dep_function1], [H2, time_dep_function2]]

Type:	Qobj or list

coup_op : Qobj or list: Operator describing the coupling between system and bath. Could also be a list of operators, which needs to be the same length as ck’s and vk’s.
ckAR, ckAI, vkAR, vkAI : lists: Lists containing coefficients for fitting spectral density correlation
N_cut : int: Cutoff parameter for the bath
options : qutip.solver.Options: Generic solver options. If set to None the default options will be used

boson_grad_n(he_n)[source]¶: Get the gradient term for the hierarchy ADM at level n

boson_grad_next(he_n, k, next_he)[source]¶: Get the next gradient

boson_grad_prev(he_n, k, prev_he)[source]¶: Get the previous gradient

boson_rhs()[source]¶: Make the RHS for bosonic case

configure(H_sys, coup_op, ckAR, ckAI, vkAR, vkAI, N_cut, options=None)[source]¶

Configure the solver using the passed parameters The parameters are described in the class attributes, unless there is some specific behaviour

Parameters:	options (`qutip.solver.Options`) – Generic solver options. If set to None the default options will be used

populate(heidx_list)[source]¶: Given a Hierarchy index list, populate the graph of next and previous elements

process_input(H_sys, coup_op, ckAR, ckAI, vkAR, vkAI, N_cut, options=None)[source]¶: Type-checks provided input Merges same gammas

reset()[source]¶: Reset any attributes to default values

run(rho0, tlist)[source]¶

Function to solve for an open quantum system using the HEOM model.

Parameters:	rho0 (Qobj) – Initial state (density matrix) of the system. tlist (list) – Time over which system evolves.
Returns:	results – Object storing all results from the simulation.
Return type:	`qutip.solver.Result`

steady_state(max_iter_refine=100, use_mkl=False, weighted_matching=False)[source]¶

Computes steady state dynamics

max_iter_refine : Int: Parameter for the mkl LU solver. If pardiso errors are returned this should be increased.
use_mkl : Boolean: Optional override default use of mkl if mkl is installed.
weighted_matching : Boolean: Setting this true may increase run time, but reduce stability (pardisio may not converge).

class bofin.heom.FermionicHEOMSolver(H_sys, coup_op, ck, vk, N_cut, options=None)[source]¶

Bases: object

Same as BosonicHEOMSolver, but with Fermionic baths.

H_sys¶

System Hamiltonian Or Liouvillian Or QobjEvo Or list of Hamiltonians with time dependence

Format for input (if list): [time_independent_part, [H1, time_dep_function1], [H2, time_dep_function2]]

Type:	Qobj or list

coup_op¶

Operator describing the coupling between system and bath. Could also be a list of operators, which needs to be the same length as ck’s and vk’s.

Type:	Qobj or list

ck, vk

Lists containing spectral density correlation

Type:	lists

N_cut¶

Cutoff parameter for the bath

Type:	int

options¶

Generic solver options. If set to None the default options will be used

Type:	`qutip.solver.Options`

configure(H_sys, coup_op, ck, vk, N_cut, options=None)[source]¶