wang_landau¶
Command-line usage
usage: cell wang_landau [-h] [-plfp plat.json] [-mof model.pickle] [-scsh 1]
[-nsubs {0: [1]}] [-ens canonical] [-sli []]
[-cpot None] [-ps True] [-errr None] [-er [-2.0, 2.0]]
[-ebw 0.2]
[-fr [2.718281828459045, 1.0001000050001667]]
[-um square_root]
[-fc [[0.5, 1.1051709180756477], [0.8, 1.0010005001667084], [0.9, 1.00001000005], [0.95, 1.000000100000005], [0.98, 1.00000001]]]
[-id None] [-fn cdos.json] [-rff False] [-phrt True]
[-s 1]
[serialize] [serialize_during_sampling]
[acc_prob_init_structure]
[acc_prob_dist_init_structure] [itmax_init_structure]
[niter_per_sweep] [nproc] [kwargs ...]
positional arguments:
serialize [True]
serialize_during_sampling
[True]
acc_prob_init_structure
[0.001]
acc_prob_dist_init_structure
[gaussian]
itmax_init_structure [100000000]
niter_per_sweep [100000]
nproc [0]
kwargs
options:
-h, --help show this help message and exit
-plfp, --plat-filepath plat.json
Filepath of a serialized ParentLattice object.
-mof, --model-filepath model.pickle
Output Model file path.
-scsh, --sc-shape 1 3x3 integer matrix to specify supercell shape.
-nsubs, --nsubs {0: [1]}
Number of substitutions
-ens, --ensemble canonical
Ensemble type for sampling, at the moment only
'canonical' is supported.
-sli, --sublattice-indices []
Indices of sublattices to be used in the sampling, see
documentation of WangLandau class.
-cpot, --chemical-potentials None
Chemical potentials for each sublattice, e.g. {'Al':
0.0, 'Si': 0.0}, not used at the moment.
-ps, --predict-swap True
Predict swap moves during sampling, default is True.
-errr, --error-reset None
Reset swap prediction after this many iterations,
default is None (no reset).
-er, --energy-range [-2.0, 2.0]
Energy range for sampling, list of maximum and minimum
-ebw, --energy-bin-width 0.2
Energy bin width for sampling.
-fr, --f-range [2.718281828459045, 1.0001000050001667]
Range of modification factors for WL-sampling.
-um, --update-method square_root
Method to update the modification factor, default is
'square_root', Options: 'square_root'.
-fc, --flatness-conditions [[0.5, 1.1051709180756477], [0.8, 1.0010005001667084], [0.9, 1.00001000005], [0.95, 1.000000100000005], [0.98, 1.00000001]]
Flatness conditions for the histogram, list of lists
with flatness and factor pairs.
-id, --initial-decoration None
Initial decoration of the supercell, list of integers,
defaults to random decoration.
-fn, --filename cdos.json
Filename to save the configurational density of states
(CDOS) during sampling.
-rff, --restart-from-file False
Restart sampling and CDOS calculation from a file,
located at 'filename'. Default is False.
-phrt, --plot-hist-real-time True
Plot histogram in real time during sampling, default
is True.
-s, --seed 1 Seed for random number generator, default is 1.
Arguments and behavior
- clusterx.cli.wang_landau.wang_landau(plat_filepath='plat.json', model_filepath='model.pickle', sc_shape=1, nsubs={0: [1]}, ensemble='canonical', sublattice_indices=[], chemical_potentials=None, predict_swap=True, error_reset=None, energy_range=[-2.0, 2.0], energy_bin_width=0.2, f_range=[2.718281828459045, 1.0001000050001667], update_method='square_root', flatness_conditions=[[0.5, 1.1051709180756477], [0.8, 1.0010005001667084], [0.9, 1.00001000005], [0.95, 1.000000100000005], [0.98, 1.00000001]], initial_decoration=None, serialize=True, filename='cdos.json', serialize_during_sampling=True, restart_from_file=False, plot_hist_real_time=True, acc_prob_init_structure=0.001, acc_prob_dist_init_structure='gaussian', itmax_init_structure=100000000, niter_per_sweep=100000, nproc=0, seed=1, **kwargs)¶
Perform Wang-Landau sampling with a linear CE energy model.