Source code for iDEA.parameters
### Library imports
from __future__ import division
from iDEA.input import InputSection, SystemSection
import numpy as np
### Run parameters
run = InputSection()
run.name = 'run_name' #: Name to identify run. Note: Do not use spaces or any special characters (.~[]{}<>?/\)
run.time_dependence = False #: Run time-dependent calculation
run.verbosity = 'default' #: Output verbosity ('low', 'default', 'high')
run.save = True #: Save results to disk when they are generated
run.module = 'iDEA' #: Specify alternative folder (in this directory) containing modified iDEA module
run.NON = True #: Run Non-Interacting approximation
run.LDA = False #: Run LDA approximation
run.HF = False #: Run Hartree-Fock approximation
run.HYB = False #: Run Hybrid (HF-LDA) calculation
run.EXT = True #: Run Exact Many-Body calculation
### System parameters
sys = SystemSection()
sys.NE = 2 #: Number of electrons
sys.grid = 201 #: Number of grid points (must be odd)
sys.stencil = 3 #: Discretisation of 2nd derivative (3 or 5 or 7)
sys.xmax = 10.0 #: Size of the system
sys.tmax = 1.0 #: Total real time
sys.imax = 1001 #: Number of real time iterations (NB: deltat = tmax/(imax-1))
sys.acon = 1.0 #: Smoothing of the Coloumb interaction
sys.interaction_strength = 1.0 #: Scales the strength of the Coulomb interaction
sys.im = 0 #: Are there imaginary terms in the perturbing potential? (0: no, 1: yes)
[docs]def v_ext(x):
"""Ground-state external potential
"""
return 0.5*(0.25**2)*(x**2)
sys.v_ext = v_ext
[docs]def v_pert(x):
"""Perturbing potential (switched on at t=0)
"""
return -0.01*x
sys.v_pert = v_pert
### Exact parameters
ext = InputSection()
ext.itol = 1e-12 #: Tolerance of imaginary time propagation (Recommended: 1e-12)
ext.itol_solver = 1e-14 #: Tolerance of linear solver in imaginary time propagation (Recommended: 1e-14)
ext.rtol_solver = 1e-12 #: Tolerance of linear solver in real time propagation (Recommended: 1e-12)
ext.itmax = 2000.0 #: Total imaginary time
ext.iimax = 1e5 #: Imaginary time iterations
ext.ideltat = ext.itmax/ext.iimax #: Imaginary time step (DERIVED)
ext.RE = False #: Reverse-engineer ext density to give DFT xc potential
ext.psi_gs = False #: Save the reduced 2 or 3 electron ground-state wavefunction to file
ext.initial_gspsi = 'qho' #: Initial 2 or 3 electron ground-state wavefunction ('qho' by default. 'non' can be selected.
#: 'hf', 'lda1', 'lda2', 'lda3', 'ldaheg' or 'ext' can be selected if the orbitals/wavefunction
#: are available. An ext wavefunction from another run can be used, but specify the run.name
#: instead e.g. 'run_name').
#: WARNING: If no reliable starting guess can be provided e.g. wrong number of electrons per well,
#: then choose 'qho' - this will ensure stable convergence to the true ground-state.)
### NON parameters
non = InputSection()
non.rtol_solver = 1e-13 #: Tolerance of linear solver in real time propagation (Recommended: 1e-13)
non.RE = False #: Reverse-engineer non-interacting density
### LDA parameters
lda = InputSection()
lda.NE = 'heg' #: Number of electrons used in construction of the LDA (1, 2, 3 or 'heg')
lda.scf_type = 'pulay' #: How to perform scf ('pulay', 'linear', 'cg', 'mixh', 'none')
lda.mix = 0.2 #: Mixing parameter for linear & Pulay mixing (float in [0,1])
lda.pulay_order = 20 #: Length of history for Pulay mixing (max: lda.max_iter)
lda.pulay_preconditioner = None #: Preconditioner for pulay mixing (None, 'kerker', rpa')
lda.kerker_length = 0.5 #: Length over which density fluctuations are screened (Kerker only)
lda.tol = 1e-12 #: Convergence tolerance in the density
lda.etol = 1e-12 #: Convergence tolerance in the energy
lda.max_iter = 10000 #: Maximum number of self-consistency iterations
lda.RE = False #: Reverse-engineer LDA density
### HF parameters
hf = InputSection()
hf.fock = 1 #: Include Fock term (0 = Hartree approximation, 1 = Hartree-Fock approximation)
hf.con = 1e-12 #: Tolerance
hf.nu = 0.9 #: Mixing term
hf.RE = False #: Reverse-engineer hf density
### HYB parameters
hyb = InputSection()
hyb.seperate = False #: Seperate Vx and Vc in the hybrid (False: a*F + (1-a)Vxc, True: a*F + (1-a)Vx + Vc)
hyb.functionality = 'o' #: Functionality of hybrid functionals: 'o' for optimal alpha, 'f' for fractional numbers of electrons,
#: 'a' for single alpha run
hyb.of_array = (0.5,1.0,6) #: If finding optimal alpa, this defines an array going from a->b in c steps whose energies are used for
#: optimisation. If fractional run, this defines the numbers of electrons to calculate
hyb.alpha = 1.0 #: If single alpha run, this defines the alpha
hyb.mix = 0.5 #: Mixing parameter for linear mixing (float in [0,1])
hyb.tol = 1e-12 #: Convergence tolerance in the density
hyb.max_iter = 10000 #: Maximum number of self-consistency iterations
hyb.RE = False #: Calculate the external potential for the HYB density
### RE parameters
re = InputSection()
re.stencil = 5 #: Discretisation of 1st derivative (5 or 7)
re.mu = 1.0 #: 1st convergence parameter in the ground-state reverse-engineering algorithm
re.p = 0.05 #: 2nd convergence parameter in the ground-state reverse-engineering algorithm
re.gs_density_tolerance = 1e-12 #: Tolerance of the error in the ground-state density
re.starting_guess = 'extre' #: Starting guess of groud-state Vks (if not available will start with Vxt)
re.nu = 1.0 #: 1st convergence parameter in the time-dependent reverse-engineering algorithm
re.a = 1.0e-6 #: 2nd convergence parameter in the time-dependent reverse-engineering algorithm
re.rtol_solver = 1e-12 #: Tolerance of linear solver in real time propagation (Recommended: 1e-12)
re.td_density_tolerance = 1e-7 #: Tolerance of the error in the time-dependent density
re.cdensity_tolerance = 1e-7 #: Tolerance of the error in the current density
re.max_iterations = 20 #: Maximum number of iterations per time step to find the Kohn-Sham potential
re.damping = True #: Damping term used to filter out the noise in the time-dependent Kohn-Sham vector potential
re.filter_beta = 1.8 #: 1st parameter in the damping term
