"""
Provides FixSymmetry class to preserve spacegroup symmetry during optimisation
"""
import warnings
import numpy as np
from ase.constraints import FixConstraint
from ase.stress import voigt_6_to_full_3x3_stress, full_3x3_to_voigt_6_stress
from ase.utils import atoms_to_spglib_cell
__all__ = ['refine_symmetry', 'check_symmetry', 'FixSymmetry']
def print_symmetry(symprec, dataset):
print("ase.spacegroup.symmetrize: prec", symprec,
"got symmetry group number", dataset["number"],
", international (Hermann-Mauguin)", dataset["international"],
", Hall ", dataset["hall"])
[docs]def refine_symmetry(atoms, symprec=0.01, verbose=False):
"""
Refine symmetry of an Atoms object
Parameters
----------
atoms - input Atoms object
symprec - symmetry precicion
verbose - if True, print out symmetry information before and after
Returns
-------
spglib dataset
"""
import spglib
# test orig config with desired tol
dataset = check_symmetry(atoms, symprec, verbose=verbose)
# set actual cell to symmetrized cell vectors by copying
# transformed and rotated standard cell
std_cell = dataset['std_lattice']
trans_std_cell = dataset['transformation_matrix'].T @ std_cell
rot_trans_std_cell = trans_std_cell @ dataset['std_rotation_matrix']
atoms.set_cell(rot_trans_std_cell, True)
# get new dataset and primitive cell
dataset = check_symmetry(atoms, symprec=symprec, verbose=verbose)
res = spglib.find_primitive(atoms_to_spglib_cell(atoms), symprec=symprec)
prim_cell, prim_scaled_pos, prim_types = res
# calculate offset between standard cell and actual cell
std_cell = dataset['std_lattice']
rot_std_cell = std_cell @ dataset['std_rotation_matrix']
rot_std_pos = dataset['std_positions'] @ rot_std_cell
pos = atoms.get_positions()
dp0 = (pos[list(dataset['mapping_to_primitive']).index(0)] - rot_std_pos[
list(dataset['std_mapping_to_primitive']).index(0)])
# create aligned set of standard cell positions to figure out mapping
rot_prim_cell = prim_cell @ dataset['std_rotation_matrix']
inv_rot_prim_cell = np.linalg.inv(rot_prim_cell)
aligned_std_pos = rot_std_pos + dp0
# find ideal positions from position of corresponding std cell atom +
# integer_vec . primitive cell vectors
# here we are assuming that primitive vectors returned by find_primitive
# are compatible with std_lattice returned by get_symmetry_dataset
mapping_to_primitive = list(dataset['mapping_to_primitive'])
std_mapping_to_primitive = list(dataset['std_mapping_to_primitive'])
pos = atoms.get_positions()
for i_at in range(len(atoms)):
std_i_at = std_mapping_to_primitive.index(mapping_to_primitive[i_at])
dp = aligned_std_pos[std_i_at] - pos[i_at]
dp_s = dp @ inv_rot_prim_cell
pos[i_at] = (aligned_std_pos[std_i_at] - np.round(dp_s) @ rot_prim_cell)
atoms.set_positions(pos)
# test final config with tight tol
return check_symmetry(atoms, symprec=1e-4, verbose=verbose)
[docs]def check_symmetry(atoms, symprec=1.0e-6, verbose=False):
"""
Check symmetry of `atoms` with precision `symprec` using `spglib`
Prints a summary and returns result of `spglib.get_symmetry_dataset()`
"""
import spglib
dataset = spglib.get_symmetry_dataset(atoms_to_spglib_cell(atoms),
symprec=symprec)
if verbose:
print_symmetry(symprec, dataset)
return dataset
def is_subgroup(sup_data, sub_data, tol=1e-10):
"""
Test if spglib dataset `sub_data` is a subgroup of dataset `sup_data`
"""
for rot1, trns1 in zip(sub_data['rotations'], sub_data['translations']):
for rot2, trns2 in zip(sup_data['rotations'], sup_data['translations']):
if np.all(rot1 == rot2) and np.linalg.norm(trns1 - trns2) < tol:
break
else:
return False
return True
def prep_symmetry(atoms, symprec=1.0e-6, verbose=False):
"""
Prepare `at` for symmetry-preserving minimisation at precision `symprec`
Returns a tuple `(rotations, translations, symm_map)`
"""
import spglib
dataset = spglib.get_symmetry_dataset(atoms_to_spglib_cell(atoms),
symprec=symprec)
if verbose:
print_symmetry(symprec, dataset)
rotations = dataset['rotations'].copy()
translations = dataset['translations'].copy()
symm_map = []
scaled_pos = atoms.get_scaled_positions()
for (rot, trans) in zip(rotations, translations):
this_op_map = [-1] * len(atoms)
for i_at in range(len(atoms)):
new_p = rot @ scaled_pos[i_at, :] + trans
dp = scaled_pos - new_p
dp -= np.round(dp)
i_at_map = np.argmin(np.linalg.norm(dp, axis=1))
this_op_map[i_at] = i_at_map
symm_map.append(this_op_map)
return (rotations, translations, symm_map)
def symmetrize_rank1(lattice, inv_lattice, forces, rot, trans, symm_map):
"""
Return symmetrized forces
lattice vectors expected as row vectors (same as ASE get_cell() convention),
inv_lattice is its matrix inverse (reciprocal().T)
"""
scaled_symmetrized_forces_T = np.zeros(forces.T.shape)
scaled_forces_T = np.dot(inv_lattice.T, forces.T)
for (r, t, this_op_map) in zip(rot, trans, symm_map):
transformed_forces_T = np.dot(r, scaled_forces_T)
scaled_symmetrized_forces_T[:, this_op_map] += transformed_forces_T
scaled_symmetrized_forces_T /= len(rot)
symmetrized_forces = (lattice.T @ scaled_symmetrized_forces_T).T
return symmetrized_forces
def symmetrize_rank2(lattice, lattice_inv, stress_3_3, rot):
"""
Return symmetrized stress
lattice vectors expected as row vectors (same as ASE get_cell() convention),
inv_lattice is its matrix inverse (reciprocal().T)
"""
scaled_stress = np.dot(np.dot(lattice, stress_3_3), lattice.T)
symmetrized_scaled_stress = np.zeros((3, 3))
for r in rot:
symmetrized_scaled_stress += np.dot(np.dot(r.T, scaled_stress), r)
symmetrized_scaled_stress /= len(rot)
sym = np.dot(np.dot(lattice_inv, symmetrized_scaled_stress), lattice_inv.T)
return sym
[docs]class FixSymmetry(FixConstraint):
"""
Constraint to preserve spacegroup symmetry during optimisation.
Requires spglib package to be available.
"""
def __init__(self, atoms, symprec=0.01, adjust_positions=True,
adjust_cell=True, verbose=False):
self.verbose = verbose
refine_symmetry(atoms, symprec, self.verbose) # refine initial symmetry
sym = prep_symmetry(atoms, symprec, self.verbose)
self.rotations, self.translations, self.symm_map = sym
self.do_adjust_positions = adjust_positions
self.do_adjust_cell = adjust_cell
def adjust_cell(self, atoms, cell):
if not self.do_adjust_cell:
return
# stress should definitely be symmetrized as a rank 2 tensor
# UnitCellFilter uses deformation gradient as cell DOF with steps
# dF = stress.F^-T quantity that should be symmetrized is therefore dF .
# F^T assume prev F = I, so just symmetrize dF
cur_cell = atoms.get_cell()
cur_cell_inv = atoms.cell.reciprocal().T
# F defined such that cell = cur_cell . F^T
# assume prev F = I, so dF = F - I
delta_deform_grad = np.dot(cur_cell_inv, cell).T - np.eye(3)
# symmetrization doesn't work properly with large steps, since
# it depends on current cell, and cell is being changed by deformation
# gradient
max_delta_deform_grad = np.max(np.abs(delta_deform_grad))
if max_delta_deform_grad > 0.25:
raise RuntimeError('FixSymmetry adjust_cell does not work properly'
' with large deformation gradient step {} > 0.25'
.format(max_delta_deform_grad))
elif max_delta_deform_grad > 0.15:
warnings.warn('FixSymmetry adjust_cell may be ill behaved with'
' large deformation gradient step {}'
.format(max_delta_deform_grad))
symmetrized_delta_deform_grad = symmetrize_rank2(cur_cell, cur_cell_inv,
delta_deform_grad,
self.rotations)
cell[:] = np.dot(cur_cell,
(symmetrized_delta_deform_grad + np.eye(3)).T)
def adjust_positions(self, atoms, new):
if not self.do_adjust_positions:
return
# symmetrize changes in position as rank 1 tensors
step = new - atoms.positions
symmetrized_step = symmetrize_rank1(atoms.get_cell(),
atoms.cell.reciprocal().T, step,
self.rotations, self.translations,
self.symm_map)
new[:] = atoms.positions + symmetrized_step
def adjust_forces(self, atoms, forces):
# symmetrize forces as rank 1 tensors
# print('adjusting forces')
forces[:] = symmetrize_rank1(atoms.get_cell(),
atoms.cell.reciprocal().T, forces,
self.rotations, self.translations,
self.symm_map)
def adjust_stress(self, atoms, stress):
# symmetrize stress as rank 2 tensor
raw_stress = voigt_6_to_full_3x3_stress(stress)
symmetrized_stress = symmetrize_rank2(atoms.get_cell(),
atoms.cell.reciprocal().T,
raw_stress, self.rotations)
stress[:] = full_3x3_to_voigt_6_stress(symmetrized_stress)
def index_shuffle(self, atoms, ind):
if len(atoms) != len(ind) or len(set(ind)) != len(ind):
raise RuntimeError("FixSymmetry can only accomodate atom"
" permutions, and len(Atoms) == {} "
"!= len(ind) == {} or ind has duplicates"
.format(len(atoms), len(ind)))
ind_reversed = np.zeros((len(ind)), dtype=int)
ind_reversed[ind] = range(len(ind))
new_symm_map = []
for sm in self.symm_map:
new_sm = np.array([-1] * len(atoms))
for at_i in range(len(ind)):
new_sm[ind_reversed[at_i]] = ind_reversed[sm[at_i]]
new_symm_map.append(new_sm)
self.symm_map = new_symm_map