def en_difference_active(self, atoms=None):
"""Returns a list of electronegativity metrics, squared and summed over
adsorbate bonds including those with the surface.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
labels = ['dist_' + s + '_active' for s in electronegativities]
if atoms is None:
return labels
cm = atoms.connectivity
ads = atoms.subsets['ads_atoms']
site = atoms.subsets['site_atoms']
active = ads + site
bonds = cm[ads, :][:, active]
active_numbers = atoms.numbers[active]
ads_numbers = atoms.numbers[ads]
en_active = list_mendeleev_params(active_numbers, electronegativities)
en_ads = list_mendeleev_params(ads_numbers, electronegativities)
delta_en = (en_ads[:, np.newaxis, :] -
en_active[np.newaxis, :, :]) ** 2
en_result = list(np.einsum("ij,ijk->k", bonds, delta_en))
assert len(en_result) == len(labels)
return en_result
def en_difference_chemi(self, atoms=None):
"""Returns a list of electronegativity metrics, squared and summed over
adsorbate-site bonds.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
labels = ['dist_' + s + '_chemi' for s in electronegativities]
if atoms is None:
return labels
cm = atoms.connectivity
chemi = atoms.subsets['chemisorbed_atoms']
site = atoms.subsets['site_atoms']
bonds = cm[chemi, :][:, site]
chemi_numbers = atoms.numbers[chemi]
site_numbers = atoms.numbers[site]
en_chemi = list_mendeleev_params(chemi_numbers, electronegativities)
en_site = list_mendeleev_params(site_numbers, electronegativities)
delta_en = (en_chemi[:, np.newaxis, :] -
en_site[np.newaxis, :, :]) ** 2
en_result = list(np.einsum("ij,ijk->k", bonds, delta_en))
assert len(en_result) == len(labels)
return en_result
def term(self, atoms=None):
"""Return a fingerprint vector with propeties averaged over the
termination atoms.
Parameters
----------
atoms : object
"""
labels = make_labels(self.slab_params, '', '_term')
labels.append('ground_state_magmom_term')
if atoms is None:
return labels
else:
if ('key_value_pairs' in atoms.info and
'term' in atoms.info['key_value_pairs']):
term = atoms.info['key_value_pairs']['term']
numbers = [atomic_numbers[s] for s in string2symbols(term)]
elif 'termination_atoms' in atoms.subsets:
term = atoms.subsets['termination_atoms']
numbers = atoms.numbers[term]
else:
raise NotImplementedError("termination fingerprint.")
dat = list_mendeleev_params(numbers, params=self.slab_params)
result = list(np.nanmean(dat, axis=0))
result += [np.nanmean([gs_magmom[z] for z in numbers])]
check_labels(labels, result, atoms)
return result
def mean_surf_ligands(self, atoms=None):
"""Function that takes an atoms objects and returns a fingerprint
vector containing the count of nearest neighbors and properties of
the nearest neighbors.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
labels = ['nn_surf_ligands', 'identnn_surf_ligands']
labels += make_labels(self.slab_params, '', '_surf_ligands')
labels.append('ground_state_magmom_surf_ligands')
if atoms is None:
return labels
else:
ligand_atoms = atoms.subsets['ligand_atoms']
numbers = atoms.numbers[ligand_atoms]
# Import CatLearn data on that element.
dat = list_mendeleev_params(numbers, params=self.slab_params)
result = list(np.nanmean(dat, axis=0))
result += [np.nanmean([gs_magmom[z] for z in numbers])]
# Append count of ligand atoms.
result = [len(ligand_atoms), len(np.unique(numbers))] + result
check_labels(labels, result, atoms)
return result
def ads_av(self, atoms=None):
"""Function that takes an atoms objects and returns a fingerprint
vector with averages of the atomic properties of the adsorbate.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
ads_params = default_params + ['econf', 'ionenergies']
labels = make_labels(ads_params, '', '_ads_av')
labels.append('ground_state_magmom_ads_av')
if atoms is None:
return labels
else:
numbers = [atoms[j].number for j in atoms.subsets['ads_atoms']]
dat = list_mendeleev_params(numbers, params=ads_params)
result = list(np.nanmean(dat, axis=0))
result += [np.nanmean([gs_magmom[z] for z in numbers])]
check_labels(labels, result, atoms)
return result
def mean_chemisorbed_atoms(self, atoms=None):
"""Function that takes an atoms objects and returns a fingerprint
vector containing properties of the closest add atom to a surface
metal atom.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
extra_ads_params = ['atomic_radius', 'heat_of_formation',
'oxistates', 'block', 'econf', 'ionenergies']
labels = make_labels(default_params + extra_ads_params, '', '_ads1')
labels.append('ground_state_magmom_ads1')
if atoms is None:
return labels
else:
# Get atomic number of alpha adsorbate atom.
chemisorbed_atoms = atoms.subsets['chemisorbed_atoms']
numbers = atoms.numbers[chemisorbed_atoms]
# Import CatLearn data on that element.
dat = list_mendeleev_params(numbers, params=default_params +
extra_ads_params)
result = list(np.nanmean(dat, axis=0))
result += [np.nanmean([gs_magmom[z] for z in numbers])]
check_labels(labels, result, atoms)
return result
def sum_site(self, atoms=None):
"""Function that takes an atoms objects and returns a fingerprint
vector with properties summed over the surface metal atoms
closest to an add atom.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
labels = make_labels(self.slab_params, '', '_site_sum')
labels.append('ground_state_magmom_site_sum')
if atoms is None:
return labels
else:
numbers = [atoms[j].number for j in atoms.subsets['site_atoms']]
dat = list_mendeleev_params(numbers, params=self.slab_params)
result = list(np.nansum(dat, axis=0))
result += [np.nansum([gs_magmom[z] for z in numbers])]
check_labels(labels, result, atoms)
return result
def bulk(self, atoms=None):
"""Return a fingerprint vector with propeties averaged over
the bulk atoms.
Parameters
----------
atoms : object
ASE Atoms object.
Returns
----------
features : list
If None was passed, the elements are strings, naming the feature.
"""
labels = make_labels(self.slab_params, '', '_bulk')
labels.append('ground_state_magmom_bulk')
if atoms is None:
return labels
else:
if ('key_value_pairs' in atoms.info and
'bulk' in atoms.info['key_value_pairs']):
bulk = atoms.info['key_value_pairs']['bulk']
numbers = [atomic_numbers[s] for s in string2symbols(bulk)]
elif 'bulk_atoms' in atoms.subsets:
bulk = atoms.subsets['bulk_atoms']
numbers = atoms.numbers[bulk]
else:
raise NotImplementedError("bulk fingerprint.")
dat = list_mendeleev_params(numbers, params=self.slab_params)
result = list(np.nanmean(dat, axis=0))
result += [np.nanmean([gs_magmom[z] for z in numbers])]
check_labels(labels, result, atoms)
return result
def bulk_summation(self, atoms=None):
"""Return a fingerprint vector with propeties of the element name
saved in the atoms.info['key_value_pairs']['bulk']"""
if atoms is None:
return ['atomic_number_sum',
'atomic_volume_sum',
'boiling_point_sum',
'density_sum',
'dipole_polarizability_sum',
'electron_affinity_sum',
'group_id_sum',
'lattice_constant_sum',
'melting_point_sum',
'period_sum',
'vdw_radius_sum',
'covalent_radius_cordero_sum',
'en_allen_sum',
'atomic_weight_sum',
'c6_sum',
'c6_gb_sum',
'en_allred-rochow_sum',
'en_cottrell-sutton_sum',
'en_gordy_sum',
'en_martynov-batsanov_sum',
'en_mulliken_sum',
'en_nagle_sum',
'en_ghosh_sum',
'gas_basicity_sum',
'metallic_radius_sum',
'atomic_radius_sum',
'heat_of_formation_sum',
'dft_sum_modulus_sum',
'dft_rhodensity_sum',
'en_pauling_sum',
'dbcenter_sum',
'dbfilling_sum',
'dbwidth_sum',
'dbskew_sum',
'dbkurtosis_sum',
'sblock_sum',
'pblock_sum',
'dblock_sum',
'fblock_sum',
'ne_outer_sum',
'ne_s_sum',
'ne_p_sum',
'ne_d_sum',
'ne_f_sum',
'ionenergy_sum',
'ground_state_magmom_sum']
else:
numbers = atoms.get_atomic_numbers()
dat = list_mendeleev_params(numbers, params=default_params +
self.extra_params)
result = list(np.nansum(dat, axis=0))
result += [np.nansum([gs_magmom[z] for z in numbers])]
return result
def get_autocorrelation(self, atoms):
"""Return the autocorrelation fingerprint for a molecule."""
connectivity = catlearn_neighborlist(atoms)
G = nx.Graph(connectivity)
distance_matrix = nx.floyd_warshall_numpy(G)
Bm = np.zeros(distance_matrix.shape)
n = len(self.parameters)
W = list_mendeleev_params(atoms.numbers, self.parameters).T
fingerprint = np.zeros(n * (self.dstar + 1))
for dd in range(self.dstar + 1):
B = Bm.copy()
B[distance_matrix == dd] = 1
AC = np.dot(np.dot(W, B), W.T).diagonal()
fingerprint[n * dd:n * (dd + 1)] = AC
return fingerprint
def en_difference_ads(self, atoms=None):
"""Returns a list of electronegativity metrics, squared and summed over
bonds within the adsorbate atoms.
Parameters
----------
atoms : object
"""
labels = ['dist_' + s + '_ads' for s in electronegativities]
if atoms is None:
return labels
cm = atoms.connectivity
ads = atoms.subsets['ads_atoms']
bonds = cm[ads, :][:, ads]
ads_numbers = atoms.numbers[ads]
en_ads = list_mendeleev_params(ads_numbers, electronegativities)
delta_en = (en_ads[:, np.newaxis, :] - en_ads[np.newaxis, :, :])**2
en_result = list(np.einsum("ij,ijk->k", bonds, delta_en))
assert len(en_result) == len(labels)
return en_result
def median_site(self, atoms=None):
"""Function that takes an atoms objects and returns a fingerprint
vector with properties averaged over the surface metal atoms
closest to an add atom.
Parameters
----------
atoms : object
"""
labels = make_labels(self.slab_params, '', '_site_med')
labels.append('ground_state_magmom_site_med')
if atoms is None:
return labels
else:
numbers = [atoms[j].number for j in atoms.subsets['site_atoms']]
dat = list_mendeleev_params(numbers, params=self.slab_params)
result = list(np.nanmedian(dat, axis=0))
result += [np.nanmedian([gs_magmom[z] for z in numbers])]
check_labels(labels, result, atoms)
return result
def conv_bulk(self, atoms=None):
"""Return a fingerprint vector with propeties convoluted over the
bulk atoms.
Parameters
----------
atoms : object
A single atoms object.
"""
labels = [
'atomic_number_bulk_conv_0', 'atomic_volume_bulk_conv_0',
'boiling_point_bulk_conv_0', 'density_bulk_conv_0',
'dipole_polarizability_bulk_conv_0',
'electron_affinity_bulk_conv_0', 'group_id_bulk_conv_0',
'lattice_constant_bulk_conv_0', 'melting_point_bulk_conv_0',
'period_bulk_conv_0', 'vdw_radius_bulk_conv_0',
'covalent_radius_cordero_bulk_conv_0', 'en_allen_bulk_conv_0',
'atomic_weight_bulk_conv_0', 'atomic_radius_bulk_conv_0',
'heat_of_formation_bulk_conv_0', 'dft_bulk_modulus_bulk_conv_0',
'dft_rhodensity_bulk_conv_0', 'dbcenter_bulk_conv_0',
'dbfilling_bulk_conv_0', 'dbwidth_bulk_conv_0',
'dbskew_bulk_conv_0', 'dbkurtosis_bulk_conv_0',
'oxi_min_bulk_conv_0', 'oxi_med_bulk_conv_0',
'oxi_max_bulk_conv_0', 'sblock_bulk_conv_0', 'pblock_bulk_conv_0',
'dblock_bulk_conv_0', 'fblock_bulk_conv_0', 'ne_outer_bulk_conv_0',
'ne_s_bulk_conv_0', 'ne_p_bulk_conv_0', 'ne_d_bulk_conv_0',
'ne_f_bulk_conv_0', 'ionenergy_bulk_conv_0',
'ground_state_magmom_bulk_conv_0', 'atomic_number_bulk_conv_1',
'atomic_volume_bulk_conv_1', 'boiling_point_bulk_conv_1',
'density_bulk_conv_1', 'dipole_polarizability_bulk_conv_1',
'electron_affinity_bulk_conv_1', 'group_id_bulk_conv_1',
'lattice_constant_bulk_conv_1', 'melting_point_bulk_conv_1',
'period_bulk_conv_1', 'vdw_radius_bulk_conv_1',
'covalent_radius_cordero_bulk_conv_1', 'en_allen_bulk_conv_1',
'atomic_weight_bulk_conv_1', 'atomic_radius_bulk_conv_1',
'heat_of_formation_bulk_conv_1', 'dft_bulk_modulus_bulk_conv_1',
'dft_rhodensity_bulk_conv_1', 'dbcenter_bulk_conv_1',
'dbfilling_bulk_conv_1', 'dbwidth_bulk_conv_1',
'dbskew_bulk_conv_1', 'dbkurtosis_bulk_conv_1',
'oxi_min_bulk_conv_1', 'oxi_med_bulk_conv_1',
'oxi_max_bulk_conv_1', 'sblock_bulk_conv_1', 'pblock_bulk_conv_1',
'dblock_bulk_conv_1', 'fblock_bulk_conv_1', 'ne_outer_bulk_conv_1',
'ne_s_bulk_conv_1', 'ne_p_bulk_conv_1', 'ne_d_bulk_conv_1',
'ne_f_bulk_conv_1', 'ionenergy_bulk_conv_1',
'ground_state_magmom_bulk_conv_1'
]
if atoms is None:
return labels
if 'bulk_atoms' in atoms.subsets:
bulk = atoms.subsets['bulk_atoms']
numbers = atoms.numbers[bulk]
connectivity = atoms.connectivity[bulk]
else:
raise NotImplementedError("bulk convoluted fingerprint.")
slab_numbers = atoms.numbers
dat_b = list_mendeleev_params(numbers, params=self.slab_params)
dat = list_mendeleev_params(slab_numbers, params=self.slab_params)
result = list(np.sqrt(np.sum(dat_b * dat_b, axis=0)) / len(bulk))
result += [
np.sqrt(np.sum([gs_magmom[z]**2 for z in numbers])) / len(bulk)
]
for i in range(dat.shape[1]):
result += [
np.sqrt(
np.sum(dat[:, i][:len(connectivity)].reshape(
len(connectivity), 1) * (dat[:, i] * connectivity))) /
len(bulk)
]
gsm = np.array([gs_magmom[z] for z in atoms.numbers])
result += [
np.sqrt(
np.sum(gsm[:len(connectivity)].reshape(len(connectivity), 1) *
(gsm * connectivity))) / len(bulk)
]
check_length(labels, result, atoms)
return result
