Source code for hiphive.core.orbits

Contains the Orbit class which hold onformation about equivalent clusters.
import pickle
import numpy as np
import tarfile

from .utilities import BiMap
from .atoms import Atom
from import logger
from import (add_items_to_tarfile_hdf5,
logger = logger.getChild('orbits')

# This is the interface accessible for cluster_space
[docs]def get_orbits(cluster_list, atom_list, rotation_matrices, translation_vectors, permutations, prim, symprec): """Generate a list of the orbits for the clusters in a supercell configuration. This method requires as input a list of the clusters in a supercell configuration as well as a set of symmetry operations (rotations and translations). From this information it will generate a list of the orbits, i.e. the set of symmetry inequivalent clusters each associated with its respective set of equivalent clusters. Parameters ---------- cluster_list : BiMap object a list of clusters atom_list : BiMap object a list of atoms in a supercell rotation_matrices : list of NumPy (3,3) arrays rotational symmetries to be imposed (e.g., from spglib) translation_vectors : list of NumPy (3) arrays translational symmetries to be imposed (e.g., from spglib) permutations : list of permutations lookup table for permutations prim : hiPhive Atoms object primitive structure Returns ------- list of Orbits objs orbits associated with the list of input clusters """ logger.debug('Preparing input...') atoms = prepare_atoms(atom_list) clusters = prepare_clusters(cluster_list) rotations = prepare_rotations(rotation_matrices) translations = prepare_translations(translation_vectors) permutations = prepare_permutations(permutations) cell = prim.cell basis = prim.basis logger.debug('Creating permutation map...') permutation_map, extended_atoms = \ get_permutation_map(atoms, rotations, translations, basis, symprec) logger.debug('Creating orbits...') orbits = _get_orbits(permutation_map, extended_atoms, clusters, basis, cell, rotations, permutations) return orbits
# All prepares are in case we changes some interface stuff
[docs]def prepare_permutations(permutations): perms = BiMap() for p in permutations: perms.append(tuple(p)) return perms
[docs]def prepare_atoms(atom_list): atoms = BiMap() for atom in atom_list: atoms.append(atom) return atoms
[docs]def prepare_clusters(cluster_list): clusters = BiMap() for cluster in cluster_list: clusters.append(tuple(cluster)) return clusters
[docs]def prepare_rotations(rotation_matrices): return rotation_matrices
[docs]def prepare_translations(translation_vectors): return translation_vectors
[docs]def get_permutation_map(atoms, rotations, translations, basis, symprec): extended_atoms = atoms.copy() permutation_map = np.zeros((len(atoms), len(rotations)), dtype=int) scaled_positions = [atom.spos(basis) for atom in extended_atoms] for sym_index, (R, T) in enumerate(zip(rotations, translations)): for atom_index, spos in enumerate(scaled_positions): new_spos =, spos) + T new_atom = Atom.spos_to_atom(new_spos, basis, symprec) if new_atom not in extended_atoms: extended_atoms.append(new_atom) mapped_atom_index = extended_atoms.index(new_atom) permutation_map[atom_index, sym_index] = mapped_atom_index return permutation_map, extended_atoms
# The internal function doing the outer loop over orbits def _get_orbits(permutation_map, extended_atoms, clusters, basis, cell, rotations, permutations): cluster_is_found = [False] * len(clusters) orbits = [] for cluster_index, cluster in enumerate(clusters): if cluster_is_found[cluster_index]: continue orbit = Orbit() cluster_atoms = [extended_atoms[i] for i in cluster] positions = [atom.pos(basis, cell) for atom in cluster_atoms] orbit.radius = get_geometrical_radius(positions) orbit.maximum_distance = get_maximum_distance(positions) orbit.order = len(cluster) populate_orbit(orbit, permutations, clusters, cluster, permutation_map, extended_atoms, cluster_is_found) orbits.append(orbit) # bar.tick() return orbits # Takes a cluster and generates all equivalent, translated clusters
[docs]def generate_translated_clusters(cluster, extended_atoms): transformed_cluster_atoms = [extended_atoms[i] for i in cluster] tested_offsets = set() for atom in transformed_cluster_atoms: offset = atom.offset if offset in tested_offsets: continue else: tested_offsets.add(offset) translated_cluster = [] for atom in transformed_cluster_atoms: new_offset = np.subtract(atom.offset, offset) new_atom = Atom(, new_offset) translated_cluster.append(extended_atoms.index(new_atom)) yield tuple(translated_cluster)
# Here is the actual categorization
[docs]def populate_orbit(orbit, permutations, clusters, cluster, permutation_map, extended_atoms, cluster_is_found): for sym_index in range(permutation_map.shape[1]): of = OrientationFamily(sym_index) transformed_cluster = permutation_map[cluster, sym_index] for translated_cluster in generate_translated_clusters( transformed_cluster, extended_atoms): argsort = tuple(np.argsort(translated_cluster)) perm_index = permutations.index(argsort) translated_cluster = tuple(sorted(translated_cluster)) translated_cluster_index = clusters.index(translated_cluster) if cluster == translated_cluster: if (sym_index, perm_index) not in orbit.eigensymmetries: orbit.eigensymmetries.append((sym_index, perm_index)) if not cluster_is_found[translated_cluster_index]: cluster_is_found[translated_cluster_index] = True of.cluster_indices.append(translated_cluster_index) of.permutation_indices.append(perm_index) if len(of.cluster_indices) > 0: orbit.orientation_families.append(of) return orbit
[docs]class Orbit: """ This class serves as a container for storing data pertaining to an orbit. Attributes ---------- orientation_families : list of OrientationFamily objs orientation families of the orbit eigensymmetries : list of tuples each eigensymmetry corresponds to a pair where the first index is the symmetry and the second is the permutation eigentensors : list(numpy.ndarray) decomposition of the force constant into symmetry elements """ # TODO: Make properties of the parameters def __init__(self): self.orientation_families = [] self.eigensymmetries = [] self.eigentensors = [] @property def prototype_index(self): """int : index of cluster that serves as prototype for this orbit In the code the first symmetry is always the identity so the first orientation family should always correspond to the prototype""" return self.orientation_families[0].cluster_indices[0]
[docs] def write(self, f): """Write a Orbit instance to a file. Parameters ---------- f : str or file object name of input file (str) or stream to write to (file object) """ tar_file ='w', fileobj=f) # add eigentensors as NumPy array items_hdf5 = dict(eigentensors=np.array(self.eigentensors), ) add_items_to_tarfile_hdf5(tar_file, items_hdf5, 'eigentensors') # add eigensymmetries as list items_pickle = dict(eigensymmetries=self.eigensymmetries, order=self.order, radius=self.radius, maximum_distance=self.maximum_distance) add_items_to_tarfile_pickle(tar_file, items_pickle, 'attributes') # add orientation families add_list_to_tarfile_custom(tar_file, self.orientation_families, 'orientation_families') tar_file.close()
[docs] @staticmethod def read(f): """Load a ClusterSpace from file Parameters ---------- f : string or file object name of input file (string) or stream to load from (file object) """ orb = Orbit() tar_file ='r', fileobj=f) # read eigentensors hdf5 items_hdf5 = read_items_hdf5(tar_file, 'eigentensors') orb.eigentensors = [et for et in items_hdf5['eigentensors']] # read attributes pickle attributes = read_items_pickle(tar_file, 'attributes') for name, value in attributes.items(): orb.__setattr__(name, value) # read orientation families ofs = read_list_custom(tar_file, 'orientation_families', orb.orientation_families = ofs return orb
[docs]class OrientationFamily: """A container for storing information for a "family of orientations". An orbit contains many clusters. Some of the clusters can be tranlsated onto each other and other must first be rotated. A set of clusters in the orbit which can all be translated onto each other are oriented in the same way and belongs to the same orientation family. The family is haracterized by the symmetry (rotation) which relates it to the prototype structure of the orbit. Since the clusters are generally stored sorted the permutation information must also be stored. Parameters ---------- symmetry_index : int The index of the symmetry corresponding to spglibs symmetry Attributes ---------- symmetry_index : int The index of the symmetry corresponding to spglibs symmetry cluster_indices : list of ints The indices of the clusters belonging to this family permutation_indices : list of ints The indices of the permutation vector """ def __init__(self, symmetry_index=None): self.symmetry_index = symmetry_index self.cluster_indices = [] self.permutation_indices = []
[docs] def write(self, f): """ Write the object to file. Parameters ---------- f : str or file object name of input file (str) or stream to write to (file object) """ try: pickle.dump(self, f) except Exception: raise Exception('Failed writing to file.')
[docs] @staticmethod def read(f): """Load a OrientationFamily object from a pickle file. Parameters ---------- f : str or file object name of input file (str) or stream to load from (file object) Returns ------- OrientationFamily object """ try: return pickle.load(f) except Exception: raise Exception('Failed loading from file.')
[docs]def get_geometrical_radius(positions): """Compute the geometrical size of a 3-dimensional point cloud. The geometrical size is defined as the average distance to the geometric center. Parameters ---------- positions : list of 3-dimensional vectors positions of points in cloud Returns ------- float geometric size of point cloud """ geometric_center = np.mean(positions, axis=0) return np.mean(np.sqrt(np.sum((positions - geometric_center)**2, axis=1)))
[docs]def get_maximum_distance(positions): """Compute the maximum distance between any two points in a 3-dimensional point cloud. This is equivalent to the "size" criterion used when imposing a certain (pair) cutoff criterion during construction of a set of clusters. Parameters ---------- positions : list of 3-dimensional vectors positions of points in cloud Returns ------- float maximum distance betwee any two points """ if len(positions) == 0: return 0.0 max_distance = 0.0 for pt1 in positions[:-1]: for pt2 in positions[1:]: max_distance = max(max_distance, np.linalg.norm(pt1 - pt2)) return max_distance