Source code for qutip.bloch_redfield

__all__ = ['brmesolve', 'bloch_redfield_solve']

import numpy as np
import os
import time
import types
import warnings
import scipy.integrate
from qutip.qobj import Qobj, isket
from qutip.states import ket2dm
from qutip.superoperator import spre, spost, vec2mat, mat2vec
from qutip.expect import expect
from qutip.solver import Options, Result, config, _solver_safety_check
from qutip.cy.spmatfuncs import cy_ode_rhs
from qutip.cy.spconvert import dense2D_to_fastcsr_fmode
from qutip.interpolate import Cubic_Spline
from qutip.cy.br_codegen import BR_Codegen
from qutip.ui.progressbar import BaseProgressBar, TextProgressBar
from qutip.cy.utilities import _cython_build_cleanup
from qutip.expect import expect_rho_vec
from qutip.rhs_generate import _td_format_check
from qutip.cy.openmp.utilities import check_use_openmp
import qutip.settings as qset
from qutip.cy.br_tensor import bloch_redfield_tensor

# -----------------------------------------------------------------------------
# Solve the Bloch-Redfield master equation
#
[docs]def brmesolve(H, psi0, tlist, a_ops=[], e_ops=[], c_ops=[], args={}, use_secular=True, sec_cutoff = 0.1, tol=qset.atol, spectra_cb=None, options=None, progress_bar=None, _safe_mode=True, verbose=False): """ Solves for the dynamics of a system using the Bloch-Redfield master equation, given an input Hamiltonian, Hermitian bath-coupling terms and their associated spectrum functions, as well as possible Lindblad collapse operators. For time-independent systems, the Hamiltonian must be given as a Qobj, whereas the bath-coupling terms (a_ops), must be written as a nested list of operator - spectrum function pairs, where the frequency is specified by the `w` variable. *Example* a_ops = [[a+a.dag(),lambda w: 0.2*(w>=0)]] For time-dependent systems, the Hamiltonian, a_ops, and Lindblad collapse operators (c_ops), can be specified in the QuTiP string-based time-dependent format. For the a_op spectra, the frequency variable must be `w`, and the string cannot contain any other variables other than the possibility of having a time-dependence through the time variable `t`: *Example* a_ops = [[a+a.dag(), '0.2*exp(-t)*(w>=0)']] It is also possible to use Cubic_Spline objects for time-dependence. In the case of a_ops, Cubic_Splines must be passed as a tuple: *Example* a_ops = [ [a+a.dag(), ( f(w), g(t)] ] where f(w) and g(t) are strings or Cubic_spline objects for the bath spectrum and time-dependence, respectively. Finally, if one has bath-couplimg terms of the form H = f(t)*a + conj[f(t)]*a.dag(), then the correct input format is *Example* a_ops = [ [(a,a.dag()), (f(w), g1(t), g2(t))],... ] where f(w) is the spectrum of the operators while g1(t) and g2(t) are the time-dependence of the operators `a` and `a.dag()`, respectively Parameters ---------- H : Qobj / list System Hamiltonian given as a Qobj or nested list in string-based format. psi0: Qobj Initial density matrix or state vector (ket). tlist : array_like List of times for evaluating evolution a_ops : list Nested list of Hermitian system operators that couple to the bath degrees of freedom, along with their associated spectra. e_ops : list List of operators for which to evaluate expectation values. c_ops : list List of system collapse operators, or nested list in string-based format. args : dict Placeholder for future implementation, kept for API consistency. use_secular : bool {True} Use secular approximation when evaluating bath-coupling terms. sec_cutoff : float {0.1} Cutoff for secular approximation. tol : float {qutip.setttings.atol} Tolerance used for removing small values after basis transformation. spectra_cb : list DEPRECIATED. Do not use. options : :class:`qutip.solver.Options` Options for the solver. progress_bar : BaseProgressBar Optional instance of BaseProgressBar, or a subclass thereof, for showing the progress of the simulation. Returns ------- result: :class:`qutip.solver.Result` An instance of the class :class:`qutip.solver.Result`, which contains either an array of expectation values, for operators given in e_ops, or a list of states for the times specified by `tlist`. """ _prep_time = time.time() #This allows for passing a list of time-independent Qobj #as allowed by mesolve if isinstance(H, list): if np.all([isinstance(h,Qobj) for h in H]): H = sum(H) if isinstance(c_ops, Qobj): c_ops = [c_ops] if isinstance(e_ops, Qobj): e_ops = [e_ops] if isinstance(e_ops, dict): e_ops_dict = e_ops e_ops = [e for e in e_ops.values()] else: e_ops_dict = None if not (spectra_cb is None): warnings.warn("The use of spectra_cb is depreciated.", DeprecationWarning) _a_ops = [] for kk, a in enumerate(a_ops): _a_ops.append([a,spectra_cb[kk]]) a_ops = _a_ops if _safe_mode: _solver_safety_check(H, psi0, a_ops+c_ops, e_ops, args) # check for type (if any) of time-dependent inputs _, n_func, n_str = _td_format_check(H, a_ops+c_ops) if progress_bar is None: progress_bar = BaseProgressBar() elif progress_bar is True: progress_bar = TextProgressBar() if options is None: options = Options() if (not options.rhs_reuse) or (not config.tdfunc): # reset config collapse and time-dependence flags to default values config.reset() #check if should use OPENMP check_use_openmp(options) if n_str == 0: R, ekets = bloch_redfield_tensor(H, a_ops, spectra_cb=None, c_ops=c_ops, use_secular=use_secular, sec_cutoff=sec_cutoff) output = Result() output.solver = "brmesolve" output.times = tlist results = bloch_redfield_solve(R, ekets, psi0, tlist, e_ops, options, progress_bar=progress_bar) if e_ops: output.expect = results else: output.states = results return output elif n_str != 0 and n_func == 0: output = _td_brmesolve(H, psi0, tlist, a_ops=a_ops, e_ops=e_ops, c_ops=c_ops, args=args, use_secular=use_secular, sec_cutoff=sec_cutoff, tol=tol, options=options, progress_bar=progress_bar, _safe_mode=_safe_mode, verbose=verbose, _prep_time=_prep_time) return output else: raise Exception('Cannot mix func and str formats.')
# ----------------------------------------------------------------------------- # Evolution of the Bloch-Redfield master equation given the Bloch-Redfield # tensor. #
[docs]def bloch_redfield_solve(R, ekets, rho0, tlist, e_ops=[], options=None, progress_bar=None): """ Evolve the ODEs defined by Bloch-Redfield master equation. The Bloch-Redfield tensor can be calculated by the function :func:`bloch_redfield_tensor`. Parameters ---------- R : :class:`qutip.qobj` Bloch-Redfield tensor. ekets : array of :class:`qutip.qobj` Array of kets that make up a basis tranformation for the eigenbasis. rho0 : :class:`qutip.qobj` Initial density matrix. tlist : *list* / *array* List of times for :math:`t`. e_ops : list of :class:`qutip.qobj` / callback function List of operators for which to evaluate expectation values. options : :class:`qutip.solver.Options` Options for the ODE solver. Returns ------- output: :class:`qutip.solver.Result` An instance of the class :class:`qutip.solver.Result`, which contains either an *array* of expectation values for the times specified by `tlist`. """ if options is None: options = Options() if options.tidy: R.tidyup() if progress_bar is None: progress_bar = BaseProgressBar() elif progress_bar is True: progress_bar = TextProgressBar() # # check initial state # if isket(rho0): # Got a wave function as initial state: convert to density matrix. rho0 = rho0 * rho0.dag() # # prepare output array # n_tsteps = len(tlist) dt = tlist[1] - tlist[0] result_list = [] # # transform the initial density matrix and the e_ops opterators to the # eigenbasis # rho_eb = rho0.transform(ekets) e_eb_ops = [e.transform(ekets) for e in e_ops] for e_eb in e_eb_ops: if e_eb.isherm: result_list.append(np.zeros(n_tsteps, dtype=float)) else: result_list.append(np.zeros(n_tsteps, dtype=complex)) # # setup integrator # initial_vector = mat2vec(rho_eb.full()) r = scipy.integrate.ode(cy_ode_rhs) r.set_f_params(R.data.data, R.data.indices, R.data.indptr) r.set_integrator('zvode', method=options.method, order=options.order, atol=options.atol, rtol=options.rtol, nsteps=options.nsteps, first_step=options.first_step, min_step=options.min_step, max_step=options.max_step) r.set_initial_value(initial_vector, tlist[0]) # # start evolution # dt = np.diff(tlist) progress_bar.start(n_tsteps) for t_idx, _ in enumerate(tlist): progress_bar.update(t_idx) if not r.successful(): raise Exception("ODE integration error: Try to increase " "the allowed number of substeps by increasing " "the nsteps parameter in the Options class.") rho_eb.data = dense2D_to_fastcsr_fmode(vec2mat(r.y), rho0.shape[0], rho0.shape[1]) # calculate all the expectation values, or output rho_eb if no # expectation value operators are given if e_ops: rho_eb_tmp = Qobj(rho_eb) for m, e in enumerate(e_eb_ops): result_list[m][t_idx] = expect(e, rho_eb_tmp) else: result_list.append(rho_eb.transform(ekets, True)) if t_idx < n_tsteps - 1: r.integrate(r.t + dt[t_idx]) progress_bar.finished() return result_list
def _td_brmesolve(H, psi0, tlist, a_ops=[], e_ops=[], c_ops=[], args={}, use_secular=True, sec_cutoff=0.1, tol=qset.atol, options=None, progress_bar=None,_safe_mode=True, verbose=False, _prep_time=0): if isket(psi0): rho0 = ket2dm(psi0) else: rho0 = psi0 nrows = rho0.shape[0] H_terms = [] H_td_terms = [] H_obj = [] A_terms = [] A_td_terms = [] C_terms = [] C_td_terms = [] CA_obj = [] spline_count = [0,0] coupled_ops = [] coupled_lengths = [] coupled_spectra = [] if isinstance(H, Qobj): H_terms.append(H.full('f')) H_td_terms.append('1') else: for kk, h in enumerate(H): if isinstance(h, Qobj): H_terms.append(h.full('f')) H_td_terms.append('1') elif isinstance(h, list): H_terms.append(h[0].full('f')) if isinstance(h[1], Cubic_Spline): H_obj.append(h[1].coeffs) spline_count[0] += 1 H_td_terms.append(h[1]) else: raise Exception('Invalid Hamiltonian specification.') for kk, c in enumerate(c_ops): if isinstance(c, Qobj): C_terms.append(c.full('f')) C_td_terms.append('1') elif isinstance(c, list): C_terms.append(c[0].full('f')) if isinstance(c[1], Cubic_Spline): CA_obj.append(c[1].coeffs) spline_count[0] += 1 C_td_terms.append(c[1]) else: raise Exception('Invalid collapse operator specification.') coupled_offset = 0 for kk, a in enumerate(a_ops): if isinstance(a, list): if isinstance(a[0], Qobj): A_terms.append(a[0].full('f')) A_td_terms.append(a[1]) if isinstance(a[1], tuple): if not len(a[1])==2: raise Exception('Tuple must be len=2.') if isinstance(a[1][0],Cubic_Spline): spline_count[1] += 1 if isinstance(a[1][1],Cubic_Spline): spline_count[1] += 1 elif isinstance(a[0], tuple): if not isinstance(a[1], tuple): raise Exception('Invalid bath-coupling specification.') if (len(a[0])+1) != len(a[1]): raise Exception('BR a_ops tuple lengths not compatible.') coupled_ops.append(kk+coupled_offset) coupled_lengths.append(len(a[0])) coupled_spectra.append(a[1][0]) coupled_offset += len(a[0])-1 if isinstance(a[1][0],Cubic_Spline): spline_count[1] += 1 for nn, _a in enumerate(a[0]): A_terms.append(_a.full('f')) A_td_terms.append(a[1][nn+1]) if isinstance(a[1][nn+1],Cubic_Spline): CA_obj.append(a[1][nn+1].coeffs) spline_count[1] += 1 else: raise Exception('Invalid bath-coupling specification.') string_list = [] for kk,_ in enumerate(H_td_terms): string_list.append("H_terms[{0}]".format(kk)) for kk,_ in enumerate(H_obj): string_list.append("H_obj[{0}]".format(kk)) for kk,_ in enumerate(C_td_terms): string_list.append("C_terms[{0}]".format(kk)) for kk,_ in enumerate(CA_obj): string_list.append("CA_obj[{0}]".format(kk)) for kk,_ in enumerate(A_td_terms): string_list.append("A_terms[{0}]".format(kk)) #Add nrows to parameters string_list.append('nrows') for name, value in args.items(): if isinstance(value, np.ndarray): raise TypeError('NumPy arrays not valid args for BR solver.') else: string_list.append(str(value)) parameter_string = ",".join(string_list) if verbose: print('BR prep time:', time.time()-_prep_time) # # generate and compile new cython code if necessary # if not options.rhs_reuse or config.tdfunc is None: if options.rhs_filename is None: config.tdname = "rhs" + str(os.getpid()) + str(config.cgen_num) else: config.tdname = opt.rhs_filename if verbose: _st = time.time() cgen = BR_Codegen(h_terms=len(H_terms), h_td_terms=H_td_terms, h_obj=H_obj, c_terms=len(C_terms), c_td_terms=C_td_terms, c_obj=CA_obj, a_terms=len(A_terms), a_td_terms=A_td_terms, spline_count=spline_count, coupled_ops = coupled_ops, coupled_lengths = coupled_lengths, coupled_spectra = coupled_spectra, config=config, sparse=False, use_secular = use_secular, sec_cutoff = sec_cutoff, args=args, use_openmp=options.use_openmp, omp_thresh=qset.openmp_thresh if qset.has_openmp else None, omp_threads=options.num_cpus, atol=tol) cgen.generate(config.tdname + ".pyx") code = compile('from ' + config.tdname + ' import cy_td_ode_rhs', '<string>', 'exec') exec(code, globals()) config.tdfunc = cy_td_ode_rhs if verbose: print('BR compile time:', time.time()-_st) initial_vector = mat2vec(rho0.full()).ravel() _ode = scipy.integrate.ode(config.tdfunc) code = compile('_ode.set_f_params(' + parameter_string + ')', '<string>', 'exec') _ode.set_integrator('zvode', method=options.method, order=options.order, atol=options.atol, rtol=options.rtol, nsteps=options.nsteps, first_step=options.first_step, min_step=options.min_step, max_step=options.max_step) _ode.set_initial_value(initial_vector, tlist[0]) exec(code, locals()) # # prepare output array # n_tsteps = len(tlist) e_sops_data = [] output = Result() output.solver = "brmesolve" output.times = tlist if options.store_states: output.states = [] if isinstance(e_ops, types.FunctionType): n_expt_op = 0 expt_callback = True elif isinstance(e_ops, list): n_expt_op = len(e_ops) expt_callback = False if n_expt_op == 0: # fall back on storing states output.states = [] options.store_states = True else: output.expect = [] output.num_expect = n_expt_op for op in e_ops: e_sops_data.append(spre(op).data) if op.isherm: output.expect.append(np.zeros(n_tsteps)) else: output.expect.append(np.zeros(n_tsteps, dtype=complex)) else: raise TypeError("Expectation parameter must be a list or a function") # # start evolution # if type(progress_bar)==BaseProgressBar and verbose: _run_time = time.time() progress_bar.start(n_tsteps) rho = Qobj(rho0) dt = np.diff(tlist) for t_idx, t in enumerate(tlist): progress_bar.update(t_idx) if not _ode.successful(): raise Exception("ODE integration error: Try to increase " "the allowed number of substeps by increasing " "the nsteps parameter in the Options class.") if options.store_states or expt_callback: rho.data = dense2D_to_fastcsr_fmode(vec2mat(_ode.y), rho.shape[0], rho.shape[1]) if options.store_states: output.states.append(Qobj(rho, isherm=True)) if expt_callback: # use callback method e_ops(t, rho) for m in range(n_expt_op): if output.expect[m].dtype == complex: output.expect[m][t_idx] = expect_rho_vec(e_sops_data[m], _ode.y, 0) else: output.expect[m][t_idx] = expect_rho_vec(e_sops_data[m], _ode.y, 1) if t_idx < n_tsteps - 1: _ode.integrate(_ode.t + dt[t_idx]) progress_bar.finished() if type(progress_bar)==BaseProgressBar and verbose: print('BR runtime:', time.time()-_run_time) if (not options.rhs_reuse) and (config.tdname is not None): _cython_build_cleanup(config.tdname) if options.store_final_state: rho.data = dense2D_to_fastcsr_fmode(vec2mat(_ode.y), rho.shape[0], rho.shape[1]) output.final_state = Qobj(rho, dims=rho0.dims, isherm=True) return output