Source code for pymatgen.analysis.chemenv.coordination_environments.voronoi

# coding: utf-8
# Copyright (c) Pymatgen Development Team.
# Distributed under the terms of the MIT License.


"""
This module contains the object used to describe the possible bonded atoms based on a Voronoi analysis.
"""

__author__ = "David Waroquiers"
__copyright__ = "Copyright 2012, The Materials Project"
__credits__ = "Geoffroy Hautier"
__version__ = "2.0"
__maintainer__ = "David Waroquiers"
__email__ = "david.waroquiers@gmail.com"
__date__ = "Feb 20, 2016"

import logging
import numpy as np
import time
from pymatgen.core.structure import Structure
from pymatgen.core.sites import PeriodicSite
from monty.json import MSONable
from scipy.spatial import Voronoi

from pymatgen.analysis.chemenv.utils.coordination_geometry_utils import my_solid_angle
from pymatgen.analysis.chemenv.utils.coordination_geometry_utils import get_lower_and_upper_f
from pymatgen.analysis.chemenv.utils.coordination_geometry_utils import rectangle_surface_intersection
from pymatgen.analysis.chemenv.utils.defs_utils import AdditionalConditions
from pymatgen.analysis.chemenv.utils.math_utils import normal_cdf_step


[docs]def from_bson_voronoi_list2(bson_nb_voro_list2, structure): """ Returns the voronoi_list needed for the VoronoiContainer object from a bson-encoded voronoi_list. Args: bson_nb_voro_list2: List of periodic sites involved in the Voronoi. structure: Structure object. Returns: The voronoi_list needed for the VoronoiContainer (with PeriodicSites as keys of the dictionary - not allowed in the BSON format). """ voronoi_list = [None] * len(bson_nb_voro_list2) for isite, voro in enumerate(bson_nb_voro_list2): if voro is None or voro == 'None': continue voronoi_list[isite] = [] for psd, dd in voro: struct_site = structure[dd['index']] periodic_site = PeriodicSite(struct_site._species, struct_site.frac_coords + psd[1], struct_site._lattice, properties=struct_site.properties) dd['site'] = periodic_site voronoi_list[isite].append(dd) return voronoi_list
[docs]class DetailedVoronoiContainer(MSONable): """ Class used to store the full Voronoi of a given structure. """ AC = AdditionalConditions() default_voronoi_cutoff = 10.0 default_normalized_distance_tolerance = 1e-5 default_normalized_angle_tolerance = 1e-3 def __init__(self, structure=None, voronoi_list2=None, voronoi_cutoff=default_voronoi_cutoff, isites=None, normalized_distance_tolerance=default_normalized_distance_tolerance, normalized_angle_tolerance=default_normalized_angle_tolerance, additional_conditions=None, valences=None, maximum_distance_factor=None, minimum_angle_factor=None): """ Constructor for the VoronoiContainer object. Either a structure is given, in which case the Voronoi is computed, or the different components of the VoronoiContainer are given (used in the from_dict method). Args: structure: Structure for which the Voronoi is computed. voronoi_list2: List of voronoi polyhedrons for each site. voronoi_cutoff: cutoff used for the voronoi. isites: indices of sites for which the Voronoi has to be computed. normalized_distance_tolerance: Tolerance for two normalized distances to be considered equal. normalized_angle_tolerance:Tolerance for two normalized angles to be considered equal. additional_conditions: Additional conditions to be used. valences: Valences of all the sites in the structure (used when additional conditions require it). maximum_distance_factor: The maximum distance factor to be considered. minimum_angle_factor: The minimum angle factor to be considered. Raises: RuntimeError if the Voronoi cannot be constructed. """ self.normalized_distance_tolerance = normalized_distance_tolerance self.normalized_angle_tolerance = normalized_angle_tolerance if additional_conditions is None: self.additional_conditions = [self.AC.NONE, self.AC.ONLY_ACB] else: self.additional_conditions = additional_conditions self.valences = valences self.maximum_distance_factor = maximum_distance_factor self.minimum_angle_factor = minimum_angle_factor if isites is None: indices = list(range(len(structure))) else: indices = isites self.structure = structure logging.debug('Setting Voronoi list') if voronoi_list2 is not None: self.voronoi_list2 = voronoi_list2 else: self.setup_voronoi_list(indices=indices, voronoi_cutoff=voronoi_cutoff) logging.debug('Setting neighbors distances and angles') t1 = time.process_time() self.setup_neighbors_distances_and_angles(indices=indices) t2 = time.process_time() logging.debug('Neighbors distances and angles set up in {:.2f} seconds'.format(t2 - t1))
[docs] def setup_voronoi_list(self, indices, voronoi_cutoff): """ Set up of the voronoi list of neighbours by calling qhull. Args: indices: indices of the sites for which the Voronoi is needed. voronoi_cutoff: Voronoi cutoff for the search of neighbours. Raises: RuntimeError: If an infinite vertex is found in the voronoi construction. """ self.voronoi_list2 = [None] * len(self.structure) self.voronoi_list_coords = [None] * len(self.structure) logging.debug('Getting all neighbors in structure') struct_neighbors = self.structure.get_all_neighbors(voronoi_cutoff, include_index=True) t1 = time.process_time() logging.debug('Setting up Voronoi list :') for jj, isite in enumerate(indices): logging.debug(' - Voronoi analysis for site #{:d} ({:d}/{:d})'.format(isite, jj + 1, len(indices))) site = self.structure[isite] neighbors1 = [(site, 0.0, isite)] neighbors1.extend(struct_neighbors[isite]) distances = [i[1] for i in sorted(neighbors1, key=lambda s: s[1])] neighbors = [i[0] for i in sorted(neighbors1, key=lambda s: s[1])] qvoronoi_input = [s.coords for s in neighbors] voro = Voronoi(points=qvoronoi_input, qhull_options="o Fv") all_vertices = voro.vertices results2 = [] maxangle = 0.0 mindist = 10000.0 for iridge, ridge_points in enumerate(voro.ridge_points): if 0 in ridge_points: ridge_vertices_indices = voro.ridge_vertices[iridge] if -1 in ridge_vertices_indices: raise RuntimeError("This structure is pathological," " infinite vertex in the voronoi " "construction") ridge_point2 = max(ridge_points) facets = [all_vertices[i] for i in ridge_vertices_indices] sa = my_solid_angle(site.coords, facets) maxangle = max([sa, maxangle]) mindist = min([mindist, distances[ridge_point2]]) for iii, sss in enumerate(self.structure): if neighbors[ridge_point2].is_periodic_image(sss, tolerance=1.0e-6): myindex = iii break results2.append({'site': neighbors[ridge_point2], 'angle': sa, 'distance': distances[ridge_point2], 'index': myindex}) for dd in results2: dd['normalized_angle'] = dd['angle'] / maxangle dd['normalized_distance'] = dd['distance'] / mindist self.voronoi_list2[isite] = results2 self.voronoi_list_coords[isite] = np.array([dd['site'].coords for dd in results2]) t2 = time.process_time() logging.debug('Voronoi list set up in {:.2f} seconds'.format(t2 - t1))
[docs] def setup_neighbors_distances_and_angles(self, indices): """ Initializes the angle and distance separations. Args: indices: Indices of the sites for which the Voronoi is needed. """ self.neighbors_distances = [None] * len(self.structure) self.neighbors_normalized_distances = [None] * len(self.structure) self.neighbors_angles = [None] * len(self.structure) self.neighbors_normalized_angles = [None] * len(self.structure) for isite in indices: results = self.voronoi_list2[isite] if results is None: continue # Initializes neighbors distances and normalized distances groups self.neighbors_distances[isite] = [] self.neighbors_normalized_distances[isite] = [] normalized_distances = [nb_dict['normalized_distance'] for nb_dict in results] isorted_distances = np.argsort(normalized_distances) self.neighbors_normalized_distances[isite].append({'min': normalized_distances[isorted_distances[0]], 'max': normalized_distances[isorted_distances[0]]}) self.neighbors_distances[isite].append({'min': results[isorted_distances[0]]['distance'], 'max': results[isorted_distances[0]]['distance']}) icurrent = 0 nb_indices = {int(isorted_distances[0])} dnb_indices = {int(isorted_distances[0])} for idist in iter(isorted_distances): wd = normalized_distances[idist] if self.maximum_distance_factor is not None: if wd > self.maximum_distance_factor: self.neighbors_normalized_distances[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_distances[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_normalized_distances[isite][icurrent]['dnb_indices'] = list(dnb_indices) self.neighbors_distances[isite][icurrent]['dnb_indices'] = list(dnb_indices) break if np.isclose(wd, self.neighbors_normalized_distances[isite][icurrent]['max'], rtol=0.0, atol=self.normalized_distance_tolerance): self.neighbors_normalized_distances[isite][icurrent]['max'] = wd self.neighbors_distances[isite][icurrent]['max'] = results[idist]['distance'] dnb_indices.add(int(idist)) else: self.neighbors_normalized_distances[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_distances[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_normalized_distances[isite][icurrent]['dnb_indices'] = list(dnb_indices) self.neighbors_distances[isite][icurrent]['dnb_indices'] = list(dnb_indices) dnb_indices = {int(idist)} self.neighbors_normalized_distances[isite].append({'min': wd, 'max': wd}) self.neighbors_distances[isite].append({'min': results[idist]['distance'], 'max': results[idist]['distance']}) icurrent += 1 nb_indices.add(int(idist)) else: self.neighbors_normalized_distances[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_distances[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_normalized_distances[isite][icurrent]['dnb_indices'] = list(dnb_indices) self.neighbors_distances[isite][icurrent]['dnb_indices'] = list(dnb_indices) for idist in range(len(self.neighbors_distances[isite]) - 1): dist_dict = self.neighbors_distances[isite][idist] dist_dict_next = self.neighbors_distances[isite][idist + 1] dist_dict['next'] = dist_dict_next['min'] ndist_dict = self.neighbors_normalized_distances[isite][idist] ndist_dict_next = self.neighbors_normalized_distances[isite][idist + 1] ndist_dict['next'] = ndist_dict_next['min'] if self.maximum_distance_factor is not None: dfact = self.maximum_distance_factor else: dfact = self.default_voronoi_cutoff / self.neighbors_distances[isite][0]['min'] self.neighbors_normalized_distances[isite][-1]['next'] = dfact self.neighbors_distances[isite][-1]['next'] = dfact * self.neighbors_distances[isite][0]['min'] # Initializes neighbors angles and normalized angles groups self.neighbors_angles[isite] = [] self.neighbors_normalized_angles[isite] = [] normalized_angles = [nb_dict['normalized_angle'] for nb_dict in results] isorted_angles = np.argsort(normalized_angles)[::-1] self.neighbors_normalized_angles[isite].append({'max': normalized_angles[isorted_angles[0]], 'min': normalized_angles[isorted_angles[0]]}) self.neighbors_angles[isite].append({'max': results[isorted_angles[0]]['angle'], 'min': results[isorted_angles[0]]['angle']}) icurrent = 0 nb_indices = {int(isorted_angles[0])} dnb_indices = {int(isorted_angles[0])} for iang in iter(isorted_angles): wa = normalized_angles[iang] if self.minimum_angle_factor is not None: if wa < self.minimum_angle_factor: self.neighbors_normalized_angles[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_angles[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_normalized_angles[isite][icurrent]['dnb_indices'] = list(dnb_indices) self.neighbors_angles[isite][icurrent]['dnb_indices'] = list(dnb_indices) break if np.isclose(wa, self.neighbors_normalized_angles[isite][icurrent]['min'], rtol=0.0, atol=self.normalized_angle_tolerance): self.neighbors_normalized_angles[isite][icurrent]['min'] = wa self.neighbors_angles[isite][icurrent]['min'] = results[iang]['angle'] dnb_indices.add(int(iang)) else: self.neighbors_normalized_angles[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_angles[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_normalized_angles[isite][icurrent]['dnb_indices'] = list(dnb_indices) self.neighbors_angles[isite][icurrent]['dnb_indices'] = list(dnb_indices) dnb_indices = {int(iang)} self.neighbors_normalized_angles[isite].append({'max': wa, 'min': wa}) self.neighbors_angles[isite].append({'max': results[iang]['angle'], 'min': results[iang]['angle']}) icurrent += 1 nb_indices.add(int(iang)) else: self.neighbors_normalized_angles[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_angles[isite][icurrent]['nb_indices'] = list(nb_indices) self.neighbors_normalized_angles[isite][icurrent]['dnb_indices'] = list(dnb_indices) self.neighbors_angles[isite][icurrent]['dnb_indices'] = list(dnb_indices) for iang in range(len(self.neighbors_angles[isite]) - 1): ang_dict = self.neighbors_angles[isite][iang] ang_dict_next = self.neighbors_angles[isite][iang + 1] ang_dict['next'] = ang_dict_next['max'] nang_dict = self.neighbors_normalized_angles[isite][iang] nang_dict_next = self.neighbors_normalized_angles[isite][iang + 1] nang_dict['next'] = nang_dict_next['max'] if self.minimum_angle_factor is not None: afact = self.minimum_angle_factor else: afact = 0.0 self.neighbors_normalized_angles[isite][-1]['next'] = afact self.neighbors_angles[isite][-1]['next'] = afact * self.neighbors_angles[isite][0]['max']
def _precompute_additional_conditions(self, ivoronoi, voronoi, valences): additional_conditions = {ac: [] for ac in self.additional_conditions} for ips, (ps, vals) in enumerate(voronoi): for ac in self.additional_conditions: additional_conditions[ac].append(self.AC.check_condition(condition=ac, structure=self.structure, parameters={'valences': valences, 'neighbor_index': vals['index'], 'site_index': ivoronoi})) return additional_conditions def _precompute_distance_conditions(self, ivoronoi, voronoi): distance_conditions = [] for idp, dp_dict in enumerate(self.neighbors_normalized_distances[ivoronoi]): distance_conditions.append([]) dp = dp_dict['max'] for ips, (ps, vals) in enumerate(voronoi): distance_conditions[idp].append(vals['normalized_distance'] <= dp or np.isclose(vals['normalized_distance'], dp, rtol=0.0, atol=self.normalized_distance_tolerance / 2.0)) return distance_conditions def _precompute_angle_conditions(self, ivoronoi, voronoi): angle_conditions = [] for iap, ap_dict in enumerate(self.neighbors_normalized_angles[ivoronoi]): angle_conditions.append([]) ap = ap_dict['max'] for ips, (ps, vals) in enumerate(voronoi): angle_conditions[iap].append(vals['normalized_angle'] >= ap or np.isclose(vals['normalized_angle'], ap, rtol=0.0, atol=self.normalized_angle_tolerance / 2.0)) return angle_conditions # def neighbors_map(self, isite, distfactor, angfactor, additional_condition): # if self.neighbors_normalized_distances[isite] is None: # return None # dist_where = np.argwhere( # np.array([wd['min'] for wd in self.neighbors_normalized_distances[isite]]) <= distfactor) # if len(dist_where) == 0: # return None # idist = dist_where[-1][0] # ang_where = np.argwhere(np.array([wa['max'] for wa in self.neighbors_normalized_angles[isite]]) >= angfactor) # if len(ang_where) == 0: # return None # iang = ang_where[0][0] # if self.additional_conditions.count(additional_condition) != 1: # return None # i_additional_condition = self.additional_conditions.index(additional_condition) # return {'i_distfactor': idist, 'i_angfactor': iang, 'i_additional_condition': i_additional_condition}
[docs] def neighbors_surfaces(self, isite, surface_calculation_type=None, max_dist=2.0): """ Get the different surfaces corresponding to the different distance-angle cutoffs for a given site. Args: isite: Index of the site surface_calculation_type: How to compute the surface. max_dist: The maximum distance factor to be considered. Returns: Surfaces for each distance-angle cutoff. """ if self.voronoi_list2[isite] is None: return None bounds_and_limits = self.voronoi_parameters_bounds_and_limits(isite, surface_calculation_type, max_dist) distance_bounds = bounds_and_limits['distance_bounds'] angle_bounds = bounds_and_limits['angle_bounds'] surfaces = np.zeros((len(distance_bounds), len(angle_bounds)), np.float) for idp in range(len(distance_bounds) - 1): this_dist_plateau = distance_bounds[idp + 1] - distance_bounds[idp] for iap in range(len(angle_bounds) - 1): this_ang_plateau = angle_bounds[iap + 1] - angle_bounds[iap] surfaces[idp][iap] = np.absolute(this_dist_plateau * this_ang_plateau) return surfaces
[docs] def neighbors_surfaces_bounded(self, isite, surface_calculation_options=None): """ Get the different surfaces (using boundaries) corresponding to the different distance-angle cutoffs for a given site. Args: isite: Index of the site. surface_calculation_options: Options for the boundaries. Returns: Surfaces for each distance-angle cutoff. """ if self.voronoi_list2[isite] is None: return None if surface_calculation_options is None: surface_calculation_options = {'type': 'standard_elliptic', 'distance_bounds': {'lower': 1.2, 'upper': 1.8}, 'angle_bounds': {'lower': 0.1, 'upper': 0.8}} if surface_calculation_options['type'] in ['standard_elliptic', 'standard_diamond', 'standard_spline']: plot_type = {'distance_parameter': ('initial_normalized', None), 'angle_parameter': ('initial_normalized', None)} else: raise ValueError('Type "{}" for the surface calculation in DetailedVoronoiContainer ' 'is invalid'.format(surface_calculation_options['type'])) max_dist = surface_calculation_options['distance_bounds']['upper'] + 0.1 bounds_and_limits = self.voronoi_parameters_bounds_and_limits(isite=isite, plot_type=plot_type, max_dist=max_dist) distance_bounds = bounds_and_limits['distance_bounds'] angle_bounds = bounds_and_limits['angle_bounds'] lower_and_upper_functions = get_lower_and_upper_f(surface_calculation_options=surface_calculation_options) mindist = surface_calculation_options['distance_bounds']['lower'] maxdist = surface_calculation_options['distance_bounds']['upper'] minang = surface_calculation_options['angle_bounds']['lower'] maxang = surface_calculation_options['angle_bounds']['upper'] f_lower = lower_and_upper_functions['lower'] f_upper = lower_and_upper_functions['upper'] surfaces = np.zeros((len(distance_bounds), len(angle_bounds)), np.float) for idp in range(len(distance_bounds) - 1): dp1 = distance_bounds[idp] dp2 = distance_bounds[idp + 1] if dp2 < mindist or dp1 > maxdist: continue if dp1 < mindist: d1 = mindist else: d1 = dp1 if dp2 > maxdist: d2 = maxdist else: d2 = dp2 for iap in range(len(angle_bounds) - 1): ap1 = angle_bounds[iap] ap2 = angle_bounds[iap + 1] if ap1 > ap2: ap1 = angle_bounds[iap + 1] ap2 = angle_bounds[iap] if ap2 < minang or ap1 > maxang: continue intersection, interror = rectangle_surface_intersection(rectangle=((d1, d2), (ap1, ap2)), f_lower=f_lower, f_upper=f_upper, bounds_lower=[mindist, maxdist], bounds_upper=[mindist, maxdist], check=False) surfaces[idp][iap] = intersection return surfaces
@staticmethod def _get_vertices_dist_ang_indices(parameter_indices_list): pp0 = [pp[0] for pp in parameter_indices_list] pp1 = [pp[1] for pp in parameter_indices_list] min_idist = min(pp0) min_iang = min(pp1) max_idist = max(pp0) max_iang = max(pp1) i_min_angs = np.argwhere(np.array(pp1) == min_iang) i_max_dists = np.argwhere(np.array(pp0) == max_idist) pp0_at_min_iang = [pp0[ii[0]] for ii in i_min_angs] pp1_at_max_idist = [pp1[ii[0]] for ii in i_max_dists] max_idist_at_min_iang = max(pp0_at_min_iang) min_iang_at_max_idist = min(pp1_at_max_idist) p1 = (min_idist, min_iang) p2 = (max_idist_at_min_iang, min_iang) p3 = (max_idist_at_min_iang, min_iang_at_max_idist) p4 = (max_idist, min_iang_at_max_idist) p5 = (max_idist, max_iang) p6 = (min_idist, max_iang) return [p1, p2, p3, p4, p5, p6]
[docs] def maps_and_surfaces(self, isite, surface_calculation_type=None, max_dist=2.0, additional_conditions=None): """ Get the different surfaces and their cn_map corresponding to the different distance-angle cutoffs for a given site. Args: isite: Index of the site surface_calculation_type: How to compute the surface. max_dist: The maximum distance factor to be considered. additional_conditions: If additional conditions have to be considered. Returns: Surfaces and cn_map's for each distance-angle cutoff. """ if self.voronoi_list2[isite] is None: return None if additional_conditions is None: additional_conditions = [self.AC.ONLY_ACB] surfaces = self.neighbors_surfaces(isite=isite, surface_calculation_type=surface_calculation_type, max_dist=max_dist) maps_and_surfaces = [] for cn, value in self._unique_coordinated_neighbors_parameters_indices[isite].items(): for imap, list_parameters_indices in enumerate(value): thissurf = 0.0 for (idp, iap, iacb) in list_parameters_indices: if iacb in additional_conditions: thissurf += surfaces[idp, iap] maps_and_surfaces.append({'map': (cn, imap), 'surface': thissurf, 'parameters_indices': list_parameters_indices}) return maps_and_surfaces
[docs] def maps_and_surfaces_bounded(self, isite, surface_calculation_options=None, additional_conditions=None): """ Get the different surfaces (using boundaries) and their cn_map corresponding to the different distance-angle cutoffs for a given site. Args: isite: Index of the site surface_calculation_options: Options for the boundaries. additional_conditions: If additional conditions have to be considered. Returns: Surfaces and cn_map's for each distance-angle cutoff. """ if self.voronoi_list2[isite] is None: return None if additional_conditions is None: additional_conditions = [self.AC.ONLY_ACB] surfaces = self.neighbors_surfaces_bounded(isite=isite, surface_calculation_options=surface_calculation_options) maps_and_surfaces = [] for cn, value in self._unique_coordinated_neighbors_parameters_indices[isite].items(): for imap, list_parameters_indices in enumerate(value): thissurf = 0.0 for (idp, iap, iacb) in list_parameters_indices: if iacb in additional_conditions: thissurf += surfaces[idp, iap] maps_and_surfaces.append({'map': (cn, imap), 'surface': thissurf, 'parameters_indices': list_parameters_indices}) return maps_and_surfaces
[docs] def neighbors(self, isite, distfactor, angfactor, additional_condition=None): """ Get the neighbors of a given site corresponding to a given distance and angle factor. Args: isite: Index of the site. distfactor: Distance factor. angfactor: Angle factor. additional_condition: Additional condition to be used (currently not implemented). Returns: List of neighbors of the given site for the given distance and angle factors. """ idist = None dfact = None for iwd, wd in enumerate(self.neighbors_normalized_distances[isite]): if distfactor >= wd['min']: idist = iwd dfact = wd['max'] else: break iang = None afact = None for iwa, wa in enumerate(self.neighbors_normalized_angles[isite]): if angfactor <= wa['max']: iang = iwa afact = wa['min'] else: break if idist is None or iang is None: raise ValueError('Distance or angle parameter not found ...') return [nb for nb in self.voronoi_list2[isite] if nb['normalized_distance'] <= dfact and nb['normalized_angle'] >= afact]
[docs] def voronoi_parameters_bounds_and_limits(self, isite, plot_type, max_dist): """ Get the different boundaries and limits of the distance and angle factors for the given site. Args: isite: Index of the site. plot_type: Types of distance/angle parameters to get. max_dist: Maximum distance factor. Returns: Distance and angle bounds and limits. """ # Initializes the distance and angle parameters if self.voronoi_list2[isite] is None: return None if plot_type is None: plot_type = {'distance_parameter': ('initial_inverse_opposite', None), 'angle_parameter': ('initial_opposite', None)} dd = [dist['min'] for dist in self.neighbors_normalized_distances[isite]] dd[0] = 1.0 if plot_type['distance_parameter'][0] == 'initial_normalized': dd.append(max_dist) distance_bounds = np.array(dd) dist_limits = [1.0, max_dist] elif plot_type['distance_parameter'][0] == 'initial_inverse_opposite': ddinv = [1.0 / dist for dist in dd] ddinv.append(0.0) distance_bounds = np.array([1.0 - invdist for invdist in ddinv]) dist_limits = [0.0, 1.0] elif plot_type['distance_parameter'][0] == 'initial_inverse3_opposite': ddinv = [1.0 / dist ** 3.0 for dist in dd] ddinv.append(0.0) distance_bounds = np.array([1.0 - invdist for invdist in ddinv]) dist_limits = [0.0, 1.0] else: raise NotImplementedError('Plotting type "{}" ' 'for the distance is not implemented'.format(plot_type['distance_parameter'])) if plot_type['angle_parameter'][0] == 'initial_normalized': aa = [0.0] aa.extend([ang['max'] for ang in self.neighbors_normalized_angles[isite]]) angle_bounds = np.array(aa) elif plot_type['angle_parameter'][0] == 'initial_opposite': aa = [0.0] aa.extend([ang['max'] for ang in self.neighbors_normalized_angles[isite]]) aa = [1.0 - ang for ang in aa] angle_bounds = np.array(aa) else: raise NotImplementedError('Plotting type "{}" ' 'for the angle is not implemented'.format(plot_type['angle_parameter'])) ang_limits = [0.0, 1.0] return {'distance_bounds': distance_bounds, 'distance_limits': dist_limits, 'angle_bounds': angle_bounds, 'angle_limits': ang_limits}
[docs] def is_close_to(self, other, rtol=0.0, atol=1e-8): """ Whether two DetailedVoronoiContainer objects are close to each other. Args: other: Another DetailedVoronoiContainer to be compared with. rtol: Relative tolerance to compare values. atol: Absolute tolerance to compare values. Returns: True if the two DetailedVoronoiContainer are close to each other. """ isclose = (np.isclose(self.normalized_angle_tolerance, other.normalized_angle_tolerance, rtol=rtol, atol=atol) and np.isclose(self.normalized_distance_tolerance, other.normalized_distance_tolerance, rtol=rtol, atol=atol) and self.additional_conditions == other.additional_conditions and self.valences == other.valences) if not isclose: return isclose for isite, site_voronoi in enumerate(self.voronoi_list2): self_to_other_nbs = {} for inb, nb in enumerate(site_voronoi): if nb is None: if other.voronoi_list2[isite] is None: continue else: return False else: if other.voronoi_list2[isite] is None: return False nb_other = None for inb2, nb2 in enumerate(other.voronoi_list2[isite]): if nb['site'] == nb2['site']: self_to_other_nbs[inb] = inb2 nb_other = nb2 break if nb_other is None: return False if not np.isclose(nb['distance'], nb_other['distance'], rtol=rtol, atol=atol): return False if not np.isclose(nb['angle'], nb_other['angle'], rtol=rtol, atol=atol): return False if not np.isclose(nb['normalized_distance'], nb_other['normalized_distance'], rtol=rtol, atol=atol): return False if not np.isclose(nb['normalized_angle'], nb_other['normalized_angle'], rtol=rtol, atol=atol): return False if nb['index'] != nb_other['index']: return False if nb['site'] != nb_other['site']: return False return True
[docs] def get_rdf_figure(self, isite, normalized=True, figsize=None, step_function=None): """ Get the Radial Distribution Figure for a given site. Args: isite: Index of the site. normalized: Whether to normalize distances. figsize: Size of the figure. step_function: Type of step function to be used for the RDF. Returns: Matplotlib figure. """ def dp_func(dp): return 1.0 - 1.0 / np.power(dp, 3.0) import matplotlib.pyplot as plt if step_function is None: step_function = {'type': 'normal_cdf', 'scale': 0.0001} # Initializes the figure if figsize is None: fig = plt.figure() else: fig = plt.figure(figsize=figsize) subplot = fig.add_subplot(111) if normalized: dists = self.neighbors_normalized_distances[isite] else: dists = self.neighbors_distances[isite] if step_function['type'] == 'step_function': isorted = np.argsort([dd['min'] for dd in dists]) sorted_dists = [dists[ii]['min'] for ii in isorted] dnb_dists = [len(dists[ii]['dnb_indices']) for ii in isorted] xx = [0.0] yy = [0.0] for idist, dist in enumerate(sorted_dists): xx.append(dist) xx.append(dist) yy.append(yy[-1]) yy.append(yy[-1] + dnb_dists[idist]) xx.append(1.1 * xx[-1]) yy.append(yy[-1]) elif step_function['type'] == 'normal_cdf': scale = step_function['scale'] mydists = [dp_func(dd['min']) for dd in dists] mydcns = [len(dd['dnb_indices']) for dd in dists] xx = np.linspace(0.0, 1.1 * max(mydists), num=500) yy = np.zeros_like(xx) for idist, dist in enumerate(mydists): yy += mydcns[idist] * normal_cdf_step(xx, mean=dist, scale=scale) else: raise ValueError('Step function of type "{}" is not allowed'.format(step_function['type'])) subplot.plot(xx, yy) return fig
[docs] def get_sadf_figure(self, isite, normalized=True, figsize=None, step_function=None): """ Get the Solid Angle Distribution Figure for a given site. Args: isite: Index of the site. normalized: Whether to normalize angles. figsize: Size of the figure. step_function: Type of step function to be used for the SADF. Returns: Matplotlib figure. """ def ap_func(ap): return np.power(ap, -0.1) import matplotlib.pyplot as plt if step_function is None: step_function = {'type': 'step_function', 'scale': 0.0001} # Initializes the figure if figsize is None: fig = plt.figure() else: fig = plt.figure(figsize=figsize) subplot = fig.add_subplot(111) if normalized: angs = self.neighbors_normalized_angles[isite] else: angs = self.neighbors_angles[isite] if step_function['type'] == 'step_function': isorted = np.argsort([ap_func(aa['min']) for aa in angs]) sorted_angs = [ap_func(angs[ii]['min']) for ii in isorted] dnb_angs = [len(angs[ii]['dnb_indices']) for ii in isorted] xx = [0.0] yy = [0.0] for iang, ang in enumerate(sorted_angs): xx.append(ang) xx.append(ang) yy.append(yy[-1]) yy.append(yy[-1] + dnb_angs[iang]) xx.append(1.1 * xx[-1]) yy.append(yy[-1]) elif step_function['type'] == 'normal_cdf': scale = step_function['scale'] myangs = [ap_func(aa['min']) for aa in angs] mydcns = [len(dd['dnb_indices']) for dd in angs] xx = np.linspace(0.0, 1.1 * max(myangs), num=500) yy = np.zeros_like(xx) for iang, ang in enumerate(myangs): yy += mydcns[iang] * normal_cdf_step(xx, mean=ang, scale=scale) else: raise ValueError('Step function of type "{}" is not allowed'.format(step_function['type'])) subplot.plot(xx, yy) return fig
def __eq__(self, other): return (self.normalized_angle_tolerance == other.normalized_angle_tolerance and self.normalized_distance_tolerance == other.normalized_distance_tolerance and self.additional_conditions == other.additional_conditions and self.valences == other.valences and self.voronoi_list2 == other.voronoi_list2 and self.structure == other.structure) def __ne__(self, other): return not self == other
[docs] def to_bson_voronoi_list2(self): """ Transforms the voronoi_list into a vlist + bson_nb_voro_list, that are BSON-encodable. Returns: [vlist, bson_nb_voro_list], to be used in the as_dict method. """ bson_nb_voro_list2 = [None] * len(self.voronoi_list2) for ivoro, voro in enumerate(self.voronoi_list2): if voro is None or voro == 'None': continue site_voro = [] # {'site': neighbors[nn[1]], # 'angle': sa, # 'distance': distances[nn[1]], # 'index': myindex} for nb_dict in voro: site = nb_dict['site'] site_dict = {key: val for key, val in nb_dict.items() if key not in ['site']} # site_voro.append([ps.as_dict(), dd]) [float(c) for c in self.frac_coords] diff = site.frac_coords - self.structure[nb_dict['index']].frac_coords site_voro.append([[nb_dict['index'], [float(c) for c in diff]], site_dict]) bson_nb_voro_list2[ivoro] = site_voro return bson_nb_voro_list2
[docs] def as_dict(self): """ Bson-serializable dict representation of the VoronoiContainer. Returns: dictionary that is BSON-encodable. """ bson_nb_voro_list2 = self.to_bson_voronoi_list2() return {"@module": self.__class__.__module__, "@class": self.__class__.__name__, "bson_nb_voro_list2": bson_nb_voro_list2, # "neighbors_lists": self.neighbors_lists, "structure": self.structure.as_dict(), "normalized_angle_tolerance": self.normalized_angle_tolerance, "normalized_distance_tolerance": self.normalized_distance_tolerance, "additional_conditions": self.additional_conditions, "valences": self.valences, "maximum_distance_factor": self.maximum_distance_factor, "minimum_angle_factor": self.minimum_angle_factor}
[docs] @classmethod def from_dict(cls, d): """ Reconstructs the VoronoiContainer object from a dict representation of the VoronoiContainer created using the as_dict method. Args: d: dict representation of the VoronoiContainer object. Returns: VoronoiContainer object. """ structure = Structure.from_dict(d['structure']) voronoi_list2 = from_bson_voronoi_list2(d['bson_nb_voro_list2'], structure) maximum_distance_factor = d['maximum_distance_factor'] if 'maximum_distance_factor' in d else None minimum_angle_factor = d['minimum_angle_factor'] if 'minimum_angle_factor' in d else None return cls(structure=structure, voronoi_list2=voronoi_list2, # neighbors_lists=neighbors_lists, normalized_angle_tolerance=d['normalized_angle_tolerance'], normalized_distance_tolerance=d['normalized_distance_tolerance'], additional_conditions=d['additional_conditions'], valences=d['valences'], maximum_distance_factor=maximum_distance_factor, minimum_angle_factor=minimum_angle_factor)