def basic_cis(output):
"""
Parser for CIS/TD-DFT calculations
:param output:
:return:
"""
data_dict = {}
# Molecule
n = output.find('$molecule')
n2 = output[n:].find('$end')
molecule_region = output[n:n + n2 - 1].replace('\t', ' ').split('\n')[1:]
charge, multiplicity = [int(num) for num in molecule_region[0].split()]
coordinates = [[float(l) for l in line.split()[1:4]]
for line in molecule_region[1:]]
symbols = [line.split()[0].capitalize() for line in molecule_region[1:]]
n_atoms = len(symbols)
# structure
structure_input = Structure(coordinates=coordinates,
symbols=symbols,
charge=charge,
multiplicity=multiplicity)
enum = output.find('Standard Nuclear Orientation')
section_structure = output[enum:enum + 200 *
structure_input.get_number_of_atoms()].split(
'\n')
section_structure = section_structure[3:structure_input.
get_number_of_atoms() + 3]
coordinates = [[float(num) for num in s.split()[2:]]
for s in section_structure]
data_dict['structure'] = Structure(coordinates=coordinates,
symbols=symbols,
charge=charge,
multiplicity=multiplicity)
# scf_energy
enum = output.find('Total energy in the final basis set')
try:
data_dict['scf_energy'] = float(output[enum:enum + 100].split()[8])
except IndexError:
pass
enum = output.find('Molecular Point Group')
if enum > 0:
symmetry_data = read_symmetry_info(output[enum:enum + 1000])
enum = output.find('Nuclear Repulsion Energy')
basic_data = read_basic_info(output[enum:enum + 5000])
# CIS excited states
# enum = output.find('CIS Excitation Energies')
try:
enum = list(re.finditer('CIS Excitation Energies', output))[-1].end()
except IndexError:
enum = list(re.finditer('TDDFT/TDA Excitation Energies',
output))[-1].end()
excited_states = []
if enum > 0:
bars = search_bars(output, from_position=enum)
output_cis = output[bars[0]:bars[1]]
for m in re.finditer('Excited state ', output_cis):
state_cis_section = output_cis[m.end():]
state_cis_lines = state_cis_section.split('\n')
exc_energy = float(state_cis_lines[0].split()[5])
exc_energy_units = state_cis_lines[0].split()[3][1:-1]
tot_energy = float(state_cis_lines[1].split()[5])
try:
tot_energy_units = state_cis_lines[1].split()[6]
mul = state_cis_lines[2].split()[-1]
trans_mom = [
float(mom) for mom in [
state_cis_lines[3].split()[2], state_cis_lines[3].
split()[4], state_cis_lines[3].split()[6]
]
]
strength = float(state_cis_lines[4].split()[2])
except ValueError:
# old version of qchem (< 5.01)
state_cis_words = output_cis[m.end():].split()
tot_energy_units = 'au'
mul = state_cis_words[13]
trans_mom = [
float(mom) for mom in [
state_cis_words[16], state_cis_words[18],
state_cis_words[20]
]
]
strength = float(state_cis_words[24])
transitions = []
for line in state_cis_lines[5:]:
if line.find('-->') > 0:
origin = int(
line.split('>')[0].split('(')[1].split(')')[0])
target = int(
line.split('>')[1].split('(')[1].split(')')[0])
amplitude = float(line.split('=')[1].split()[0])
alpha_transitions = []
beta_transitions = []
try:
spin = line[21:].split()[3]
if spin == 'alpha':
alpha_transitions.append({
'origin':
origin,
'target':
target + basic_data['n_alpha']
})
elif spin == 'beta':
beta_transitions.append({
'origin':
origin,
'target':
target + basic_data['n_beta']
})
else:
raise ParserError('basic_cis',
'Error reading configurations')
transitions.append({
'origin':
origin,
'target':
target,
'amplitude':
amplitude,
'occupations':
get_cis_occupations_list(
basic_data['n_basis_functions'],
basic_data['n_alpha'],
basic_data['n_beta'],
alpha_transitions=alpha_transitions,
beta_transitions=beta_transitions)
})
except (IndexError, ParserError):
# This supposes single electron transition
alpha_transitions.append({
'origin':
origin,
'target':
target + basic_data['n_alpha']
})
transitions.append({
'origin':
origin,
'target':
target,
'amplitude':
amplitude / np.sqrt(2),
'occupations':
get_cis_occupations_list(
basic_data['n_basis_functions'],
basic_data['n_alpha'],
basic_data['n_beta'],
alpha_transitions=alpha_transitions,
beta_transitions=beta_transitions)
})
transitions.append({
'origin':
origin,
'target':
target,
'amplitude':
amplitude /
np.sqrt(2) if mul == 'Singlet' else -amplitude /
np.sqrt(2),
'occupations':
get_cis_occupations_list(
basic_data['n_basis_functions'],
basic_data['n_alpha'],
basic_data['n_beta'],
alpha_transitions=beta_transitions,
beta_transitions=alpha_transitions)
})
if len(line) < 5:
break
excited_states.append({
'total_energy':
tot_energy,
'total_energy_units':
tot_energy_units,
'excitation_energy':
exc_energy,
'excitation_energy_units':
exc_energy_units,
'multiplicity':
mul,
'transition_moment':
standardize_vector(trans_mom),
'strength':
strength,
'configurations':
transitions
})
data_dict['excited_states'] = excited_states
# Spin-Orbit coupling
initial = output.find(
'*********SPIN-ORBIT COUPLING JOB BEGINS HERE*********')
final = output.find('*********SOC CODE ENDS HERE*********')
data_interstate = {}
if initial > 0:
soc_section = output[initial:final]
def label_states(excited_states):
labels = []
ns = 1
nt = 1
for state in excited_states:
if state['multiplicity'].lower() == 'singlet':
labels.append('S{}'.format(ns))
ns += 1
elif state['multiplicity'].lower() == 'triplet':
labels.append('T{}'.format(nt))
nt += 1
else:
try:
m = float(state['multiplicity'])
if abs(m - 1) < 0.1:
labels.append('S{}'.format(ns))
ns += 1
if abs(m - 3) < 0.1:
labels.append('T{}'.format(nt))
nt += 1
state['multiplicity'] = m
except ValueError:
raise ParserError('basic_cis',
'State multiplicity error')
return labels, nt - 1, ns - 1
labels, n_triplet, n_singlet = label_states(excited_states)
for i, label in enumerate(labels):
data_interstate[(i + 1, 0)] = {
'1e_soc_mat': [0j, 0j, 0j],
'soc_units': 'cm-1'
}
data_interstate[(0, i + 1)] = {
'1e_soc_mat': [0j, 0j, 0j],
'soc_units': 'cm-1'
}
for j, label2 in enumerate(labels):
if (label[0] == 'S'
or label2[0] == 'S') and (label[0] != label2[0]):
data_interstate[(i + 1, j + 1)] = {
'1e_soc_mat': [[0j, 0j, 0j]],
'soc_units': 'cm-1'
}
elif label[0] == 'T' and label2[0] == 'T':
data_interstate[(i + 1, j + 1)] = {
'1e_soc_mat': [[0j, 0j, 0j], [0j, 0j, 0j],
[0j, 0j, 0j]],
'soc_units': 'cm-1'
}
elif label[0] == 'S' and label2[0] == 'S':
data_interstate[(i + 1, j + 1)] = {
'1e_soc_mat': [[0j]],
'soc_units': 'cm-1'
}
else:
raise ParserError('basic_cis', 'State multiplicity error')
for i, label in enumerate(labels):
for k2, ms2 in enumerate([-1, 0, 1]):
for j, label2 in enumerate(labels):
if label[0] == 'T':
for k, ms in enumerate([-1, 0, 1]):
enum = soc_section.find(
'SOC between the {} (ms={}) state and excited triplet states (ms={})'
.format(label, ms2, ms))
for line in soc_section[enum:enum + 80 *
(n_triplet + 1)].split(
'\n'):
if len(line.split()) == 0:
break
if line.split()[0] == '{}(ms={})'.format(
label2, ms):
data_interstate[(
i + 1, j + 1
)]['1e_soc_mat'][k2][k] = _list_to_complex(
line.split()[1:4])
data_interstate[(
i + 1, j + 1
)]['1e_soc_mat'][k][k2] = _list_to_complex(
line.split()[1:4])
data_interstate[(
j + 1, i + 1
)]['1e_soc_mat'][k2][k] = _list_to_complex(
line.split()[1:4])
data_interstate[(
j + 1, i + 1
)]['1e_soc_mat'][k][k2] = _list_to_complex(
line.split()[1:4])
break
elif label[0] == 'S':
for k, ms in enumerate([-1, 0, 1]):
enum = soc_section.find(
'SOC between the {} state and excited triplet states (ms={})'
.format(label, ms))
for line in soc_section[enum:enum + 80 *
(n_triplet + 1)].split(
'\n'):
if len(line.split()) == 0:
break
if line.split()[0] == '{}(ms={})'.format(
label2, ms):
data_interstate[(
i + 1, j + 1
)]['1e_soc_mat'][0][k] = _list_to_complex(
line.split()[1:4])
data_interstate[(
j + 1, i + 1
)]['1e_soc_mat'][0][k] = _list_to_complex(
line.split()[1:4])
break
else:
raise ParserError('basic_cis', 'SOC reading error')
enum = soc_section.find(
'SOC between the singlet ground state and excited triplet states (ms={})'
.format(ms2))
for line in soc_section[enum:enum + 80 *
(n_triplet + 1)].split('\n'):
if len(line.split()) == 0:
break
if line.split()[0] == '{}(ms={})'.format(label, ms2):
data_interstate[(
i + 1, 0)]['1e_soc_mat'][k2] = _list_to_complex(
line.split()[1:4])
data_interstate[(
0, i + 1)]['1e_soc_mat'][k2] = _list_to_complex(
line.split()[1:4])
break
data_dict['interstate_properties'] = data_interstate
# diabatization
initial = output.find('Localization Code for CIS excited states')
if initial > 0:
bars = search_bars(output, from_position=initial)
diabat_section = output[initial:bars[0]]
def read_diabatization_matrix(label):
matrix = []
for m in re.finditer(label, diabat_section):
line = diabat_section[m.end():m.end() + 50].split('\n')[0]
matrix.append(float(line.split('=')[1]))
diabat_dim = int(np.sqrt(len(matrix)))
return np.array(matrix).reshape(diabat_dim, diabat_dim).T
r_matrix = read_diabatization_matrix('showmatrix adiabatic R-Matrix')
rot_matrix = read_diabatization_matrix(
'showmatrix final adiabatic -> diabatic RotMatrix')
diabatic_matrix = read_diabatization_matrix(
'showmatrix diabatH') * AU_TO_EV
if len(rot_matrix) == 0:
# Boys diabatization
rot_matrix = read_diabatization_matrix(
'showmatrix Boys adiabatic->diabatic RotMatrix')
diabatic_matrix = read_diabatization_matrix(
'showmatrix Boys diabatH') * AU_TO_EV
adiabatic_matrix = read_diabatization_matrix(
'showmatrix adiabatH') * AU_TO_EV
diabat_data = {
'rot_matrix': rot_matrix,
'adiabatic_matrix': adiabatic_matrix.tolist(),
'diabatic_matrix': diabatic_matrix.tolist()
}
if diabat_section.find('showmatrix Total_Decomposed_H_diabatic'):
tot_decomp_matrix = read_diabatization_matrix(
'showmatrix Total_Decomposed_H_diabatic') * AU_TO_EV
decomp_one_matrix = read_diabatization_matrix(
'showmatrix Decomposed_One_diabatic') * AU_TO_EV
decomp_j_matrix = read_diabatization_matrix(
'showmatrix Decomposed_J_diabatic') * AU_TO_EV
decomp_k_matrix = read_diabatization_matrix(
'showmatrix Decomposed_K_diabatic') * AU_TO_EV
diabat_data.update({
'tot_decomp_matrix': tot_decomp_matrix,
'decomp_one_matrix': decomp_one_matrix.tolist(),
'decomp_j_matrix': decomp_j_matrix.tolist(),
'decomp_k_matrix': decomp_k_matrix.tolist()
})
mulliken_diabatic = []
enum = output.find('Mulliken & Loewdin analysis of')
for m in re.finditer('Mulliken analysis of TDA State', output[enum:]):
section_mulliken = output[m.end() + enum:m.end() + 10000 +
enum] # 10000: assumed to max of section
section_mulliken = section_mulliken[:section_mulliken.find(
'Natural Orbitals stored in FCHK')]
section_attachment = section_mulliken.split('\n')[10 + n_atoms:10 +
n_atoms * 2]
mulliken_diabatic.append({
'attach': [float(l.split()[1]) for l in section_attachment],
'detach': [float(l.split()[2]) for l in section_attachment],
'total': [float(l.split()[3]) for l in section_attachment]
})
diabatic_states = []
for i in range(len(rot_matrix)):
diabat_states_data = {
'excitation_energy': diabatic_matrix[i][i],
'excitation_energy_units': 'eV',
'transition_moment': [],
'dipole_moment_units': 'ua'
}
if len(mulliken_diabatic) > 0:
diabat_states_data['mulliken'] = mulliken_diabatic[i]
diabatic_states.append(diabat_states_data)
diabat_data['diabatic_states'] = diabatic_states
data_dict['diabatization'] = diabat_data
return data_dict
def parser_rasci(output):
"""
Parser for RAS-CI calculations
Include:
- Diabatization scheme data
- Structure
- Adiabatic states
- SOC
:param output:
:return:
"""
data_dict = {}
# Molecule
n = output.find('$molecule')
n2 = output[n:].find('$end')
molecule_region = output[n:n + n2 - 1].replace('\t', ' ').split('\n')[1:]
charge, multiplicity = [int(num) for num in molecule_region[0].split()]
coordinates = [[float(l) for l in line.split()[1:4]]
for line in molecule_region[1:]]
symbols = [line.split()[0].capitalize() for line in molecule_region[1:]]
n_atoms = len(symbols)
# structure
structure_input = Structure(coordinates=coordinates,
symbols=symbols,
charge=charge,
multiplicity=multiplicity)
enum = output.find('Standard Nuclear Orientation')
section_structure = output[enum:enum + 200 *
structure_input.get_number_of_atoms()].split(
'\n')
section_structure = section_structure[3:structure_input.
get_number_of_atoms() + 3]
coordinates = [[float(num) for num in s.split()[2:]]
for s in section_structure]
data_dict['structure'] = Structure(coordinates=coordinates,
symbols=symbols,
charge=charge,
multiplicity=multiplicity)
# basic info
enum = output.find('Nuclear Repulsion Energy')
basic_data = read_basic_info(output[enum:enum + 5000])
# scf_energy
enum = output.find('SCF energy in the final basis set')
scf_energy = float(output[enum:enum + 100].split()[8])
data_dict['scf_energy'] = scf_energy
# total energy
# enum = output.find('Total energy in the final basis set')
# total_energy = float(output[enum:enum+100].split()[8])
# RASCI dimensions
ini_section = output.find('RAS-CI Dimensions')
end_section = search_bars(output, from_position=enum, bar_type='\*\*\*')[1]
dimension_section = output[ini_section:end_section]
enum = dimension_section.find('Doubly Occ')
doubly_occ = int(dimension_section[enum:enum + 50].split()[3])
enum = dimension_section.find('Doubly Vir')
doubly_vir = int(dimension_section[enum:enum + 50].split()[2])
enum = dimension_section.find('Frozen Occ')
frozen_occ = int(dimension_section[enum:enum + 50].split()[3])
enum = dimension_section.find('Frozen Vir')
frozen_vir = int(dimension_section[enum:enum + 50].split()[2])
enum = dimension_section.find('Total CI configurations')
total_conf = int(dimension_section[enum:enum + 50].split()[3])
enum = dimension_section.find('Active configurations')
active_conf = int(dimension_section[enum:enum + 50].split()[2])
enum = dimension_section.find('Hole configurations')
hole_conf = int(dimension_section[enum:enum + 50].split()[2])
enum = dimension_section.find('Particle configurations')
particle_conf = int(dimension_section[enum:enum + 50].split()[2])
rasci_dimensions = {
'doubly_occupied': doubly_occ,
'doubly_virtual': doubly_vir,
'frozen_occupied': frozen_occ,
'frozen_virtual': frozen_vir,
'total_configurations': total_conf,
'active_configurations': active_conf,
'hole_configurations': hole_conf,
'particle_configurations': particle_conf
}
data_dict.update({'rasci_dimensions': rasci_dimensions})
# Diabatization scheme
done_diabat = bool(output.find('RASCI DIABATIZATION') + 1)
if done_diabat:
rot_matrix = _read_simple_matrix(
'showmatrix final adiabatic -> diabatic', output)[-1]
adiabatic_matrix = _read_simple_matrix(
'showing H in adiabatic representation: NO coupling elements',
output)[-1]
diabatic_matrix = _read_simple_matrix(
'showing H in diabatic representation: WITH coupling elements',
output)[-1]
mulliken_adiabatic = []
enum = output.find('Mulliken analysis of Adiabatic State')
for m in re.finditer('Mulliken analysis of Adiabatic State',
output[enum:]):
section_mulliken = output[m.end() + enum:m.end() + 10000 +
enum] # 10000: assumed to max of section
section_mulliken = section_mulliken[:section_mulliken.find(
'Natural Orbitals stored in FCHK')]
section_attachment = section_mulliken.split('\n')[9 + n_atoms:9 +
n_atoms * 2]
mulliken_adiabatic.append({
'attach': [float(l.split()[1]) for l in section_attachment],
'detach': [float(l.split()[2]) for l in section_attachment],
'total': [float(l.split()[3]) for l in section_attachment]
})
mulliken_diabatic = []
enum = output.find('showing H in diabatic representation')
for m in re.finditer('Mulliken Analysis of Diabatic State',
output[enum:]):
section_mulliken = output[m.end() + enum:m.end() + 10000 +
enum] # 10000: assumed to max of section
section_mulliken = section_mulliken[:section_mulliken.find(
'Natural Orbitals stored in FCHK')]
section_attachment = section_mulliken.split('\n')[9 + n_atoms:9 +
n_atoms * 2]
mulliken_diabatic.append({
'attach': [float(l.split()[1]) for l in section_attachment],
'detach': [float(l.split()[2]) for l in section_attachment],
'total': [float(l.split()[3]) for l in section_attachment]
})
enum = output.find('Transition dipole moment - diabatic states')
tdm_section = output[enum:enum + 70 * len(rot_matrix)]
diabatic_tdm = []
for m in re.finditer('TDM', tdm_section):
diabatic_tdm.append([
float(n)
for n in tdm_section[m.end():m.end() + 70][14:].split()[:3]
])
diabatic_states = []
for i, tdm in enumerate(diabatic_tdm):
diabatic_states.append({
'excitation_energy': diabatic_matrix[i][i],
'excitation_energy_units': 'eV',
'transition_moment': tdm,
'dipole_moment_units': 'ua',
'mulliken': mulliken_diabatic[i]
})
data_dict['diabatization'] = {
'rot_matrix': rot_matrix,
'adiabatic_matrix': adiabatic_matrix,
'diabatic_matrix': diabatic_matrix,
'diabatic_states': diabatic_states,
'mulliken_adiabatic': mulliken_adiabatic
}
# excited states data
excited_states = []
for m in re.finditer('RAS-CI total energy for state', output):
# print('ll found', m.start(), m.end())
section_state = output[m.end():m.end() +
10000] # 10000: assumed to max of section
section_state = section_state[:section_state.find('********')]
enum = section_state.find('RAS-CI total energy for state')
section_state = section_state[:enum]
# energies
tot_energy = float(section_state.split()[1])
exc_energy_units = section_state.split()[4][1:-1]
exc_energy = float(section_state.split()[6])
state_multiplicity = section_state.split(
)[8] if section_state.split()[8] != ':' else section_state.split()[9]
# dipole moment
enum = section_state.find('Dipole Moment')
dipole_mom = [
float(section_state[enum:].split()[2]) + 0.0,
float(section_state[enum:].split()[4]) + 0.0,
float(section_state[enum:].split()[6]) + 0.0
]
# Transition moment
enum = section_state.find('Trans. Moment')
if enum > -1:
trans_mom = [
float(section_state[enum:].split()[2]) + 0.0,
float(section_state[enum:].split()[4]) + 0.0,
float(section_state[enum:].split()[6]) + 0.0
]
trans_mom = standardize_vector(trans_mom)
strength = float(section_state[enum:].split()[10])
else:
trans_mom = None
strength = None
# configurations table
enum = section_state.find('AMPLITUDE')
enum2 = section_state.find('Contributions')
section_table = section_state[enum:enum2].split('\n')[2:-2]
# ' HOLE | ALPHA | BETA | PART | AMPLITUDE'
table = []
for row in section_table:
table.append({
'hole': row.split('|')[1].strip(),
'alpha': row.split('|')[2].strip(),
'beta': row.split('|')[3].strip(),
'part': row.split('|')[4].strip(),
'amplitude': float(row.split('|')[5]) + 0.0
})
table[-1]['occupations'] = get_rasci_occupations_list(
table[-1], data_dict['structure'],
basic_data['n_basis_functions'])
table = sorted(table,
key=operator.itemgetter('hole', 'alpha', 'beta',
'part'))
# Contributions RASCI wfn
contributions_section = section_state[enum2:]
contributions = {
'active': float(contributions_section.split()[4]),
'hole': float(contributions_section.split()[6]),
'part': float(contributions_section.split()[8])
}
# complete dictionary
tot_energy_units = 'au'
excited_states.append({
'total_energy': tot_energy,
'total_energy_units': tot_energy_units,
'excitation_energy': exc_energy,
'excitation_energy_units': exc_energy_units,
'multiplicity': state_multiplicity,
'dipole_moment': dipole_mom,
'transition_moment': trans_mom,
'dipole_moment_units': 'ua',
'oscillator_strength': strength,
'configurations': table,
'contributions_fwn': contributions
})
data_dict.update({'excited_states': excited_states})
# Interstate transition properties
done_interstate = bool(output.find('Interstate Transition Properties') + 1)
if done_interstate:
ini_section = output.find('Interstate Transition Properties')
end_section = search_bars(output, from_position=ini_section)[1]
interstate_section = output[ini_section:end_section]
interstate_dict = {}
for m in re.finditer('State A: Root', interstate_section):
section_pair = interstate_section[m.start():m.start() + 10000]
section_pair = section_pair[:section_pair.find('********')]
lines = section_pair.split('\n')
state_a = int(lines[0].split()[-1])
state_b = int(lines[1].split()[-1])
pair_dict = {'state_a': state_a, 'state_b': state_b}
s_a = s_b = 0
for i, line in enumerate(lines):
# RAS-CI SOC section
if 'Angular momentum components' in line:
pair_dict['angular_momentum'] = [
complex(lines[i + 1 + k].split()[-1].replace('i', 'j'))
for k in range(3)
]
if '||gamma^AB||_total' in line:
pair_dict['gamma_total'] = float(lines[i + 0].split()[-1])
pair_dict['gamma_sym'] = float(lines[i + 1].split()[-1])
pair_dict['gamma_anti_sym'] = float(lines[i +
2].split()[-1])
if "KET: S',Sz'" in line:
s_a = float(lines[i].split('=')[1].split()[0])
s_b = float(lines[i + 1].split('=')[1].split()[0])
na = int(2 * s_a + 1)
nb = int(2 * s_b + 1)
if 'Spin Matrices' in line:
spinmat_x = _read_soc_matrix(lines[i + 2:], [nb, na])
spinmat_y = _read_soc_matrix(lines[i + 4 + nb:], [nb, na])
spinmat_z = _read_soc_matrix(lines[i + 6 + 2 * nb:],
[nb, na])
pair_dict['spin_matrices'] = [
spinmat_x, spinmat_y, spinmat_z
]
if 'Spin matrices Sx, Sy and Sz for states' in line:
pair_dict['spin_matrices'] = [np.zeros(
(nb, na)).tolist()] * 3
if '1-elec SOC matrix (cm-1)' in line:
pair_dict['1e_soc_mat'] = _read_soc_matrix(
lines[i + 1:], [nb, na])
pair_dict['1e_socc'] = float(lines[i + 2 +
nb].split()[-2:][0])
if '2e-SOMF Reduced matrix elements (cm-1)' in line:
r, c = lines[i + 1].split()[-2:]
pair_dict['hso_l-'] = float(r) + float(c) * 1j
r, c = lines[i + 2].split()[-2:]
pair_dict['hso_l0'] = float(r) + float(c) * 1j
r, c = lines[i + 3].split()[-2:]
pair_dict['hso_l+'] = float(r) + float(c) * 1j
if '2-elec mean-field SOC matrix (cm-1)' in line:
pair_dict['2e_soc_mat'] = _read_soc_matrix(
lines[i + 1:], [nb, na])
if 'Total mean-field SOC matrix (cm-1)' in line:
pair_dict['total_soc_mat'] = _read_soc_matrix(
lines[i + 1:], [nb, na])
if 'Mean-Field SOCC' in line:
pair_dict['mf_socc'] = float(line.split()[-2])
pair_dict['units'] = line.split()[-1]
interstate_dict[(state_a, state_b)] = pair_dict
data_dict.update({'interstate_properties': interstate_dict})
return data_dict
def basic_cis(output):
"""
Parser for CIS/TD-DFT calculations
:param output:
:return:
"""
data_dict = {}
# scf_energy
enum = output.find('Total energy in the final basis set')
try:
data_dict['scf_energy'] = float(output[enum:enum + 100].split()[8])
except IndexError:
pass
enum = output.find('Molecular Point Group')
basic_data = read_basic_info(output[enum:enum + 5000])
# CIS excited states
# enum = output.find('CIS Excitation Energies')
try:
enum = list(re.finditer('CIS Excitation Energies', output))[-1].end()
except IndexError:
enum = list(re.finditer('TDDFT/TDA Excitation Energies',
output))[-1].end()
excited_states = []
if enum > 0:
bars = search_bars(output, from_position=enum)
output_cis = output[bars[0]:bars[1]]
for m in re.finditer('Excited state ', output_cis):
state_cis_section = output_cis[m.end():]
state_cis_lines = state_cis_section.split('\n')
exc_energy = float(state_cis_lines[0].split()[5])
exc_energy_units = state_cis_lines[0].split()[3][1:-1]
tot_energy = float(state_cis_lines[1].split()[5])
try:
tot_energy_units = state_cis_lines[1].split()[6]
mul = state_cis_lines[2].split()[-1]
trans_mom = [
float(mom) for mom in [
state_cis_lines[3].split()[2], state_cis_lines[3].
split()[4], state_cis_lines[3].split()[6]
]
]
strength = float(state_cis_lines[4].split()[2])
except ValueError:
# old version of qchem (< 5.01)
state_cis_words = output_cis[m.end():].split()
tot_energy_units = 'au'
mul = state_cis_words[13]
trans_mom = [
float(mom) for mom in [
state_cis_words[16], state_cis_words[18],
state_cis_words[20]
]
]
strength = float(state_cis_words[24])
transitions = []
for line in state_cis_lines[5:]:
if line.find('-->') > 0:
origin = int(
line.split('>')[0].split('(')[1].split(')')[0])
target = int(
line.split('>')[1].split('(')[1].split(')')[0])
amplitude = float(line.split('=')[1])
alpha_transitions = []
beta_transitions = []
try:
spin = line[21:].split()[3]
if spin == 'alpha':
alpha_transitions.append({
'origin':
origin,
'target':
target + basic_data['n_alpha']
})
elif spin == 'beta':
beta_transitions.append({
'origin':
origin,
'target':
target + basic_data['n_beta']
})
else:
raise ParserError('basic_cis',
'Error reading configurations')
transitions.append({
'origin':
origin,
'target':
target,
'amplitude':
amplitude,
'occupations':
get_cis_occupations_list(
basic_data['n_basis_functions'],
basic_data['n_alpha'],
basic_data['n_beta'],
alpha_transitions=alpha_transitions,
beta_transitions=beta_transitions)
})
except (IndexError, ParserError):
# This supposes single electron transition
alpha_transitions.append({
'origin':
origin,
'target':
target + basic_data['n_alpha']
})
transitions.append({
'origin':
origin,
'target':
target,
'amplitude':
amplitude / np.sqrt(2),
'occupations':
get_cis_occupations_list(
basic_data['n_basis_functions'],
basic_data['n_alpha'],
basic_data['n_beta'],
alpha_transitions=alpha_transitions,
beta_transitions=beta_transitions)
})
transitions.append({
'origin':
origin,
'target':
target,
'amplitude':
amplitude /
np.sqrt(2) if mul == 'Singlet' else -amplitude /
np.sqrt(2),
'occupations':
get_cis_occupations_list(
basic_data['n_basis_functions'],
basic_data['n_alpha'],
basic_data['n_beta'],
alpha_transitions=beta_transitions,
beta_transitions=alpha_transitions)
})
if len(line) < 5:
break
excited_states.append({
'total_energy':
tot_energy,
'total_energy_units':
tot_energy_units,
'excitation_energy':
exc_energy,
'excitation_energy_units':
exc_energy_units,
'multiplicity':
mul,
'transition_moment':
standardize_vector(trans_mom),
'strength':
strength,
'configurations':
transitions
})
data_dict['excited_states'] = excited_states
# Spin-Orbit coupling
initial = output.find(
'*********SPIN-ORBIT COUPLING JOB BEGINS HERE*********')
final = output.find('*********SOC CODE ENDS HERE*********')
data_interstate = {}
if initial > 0:
soc_section = output[initial:final]
def label_states(excited_states):
labels = []
ns = 1
nt = 1
for state in excited_states:
if state['multiplicity'].lower() == 'singlet':
labels.append('S{}'.format(ns))
ns += 1
elif state['multiplicity'].lower() == 'triplet':
labels.append('T{}'.format(nt))
nt += 1
else:
try:
m = float(state['multiplicity'])
if abs(m - 1) < 0.1:
labels.append('S{}'.format(ns))
ns += 1
if abs(m - 3) < 0.1:
labels.append('T{}'.format(nt))
nt += 1
state['multiplicity'] = m
except ValueError:
raise ParserError('basic_cis',
'State multiplicity error')
return labels, nt - 1, ns - 1
labels, n_triplet, n_singlet = label_states(excited_states)
for i, label in enumerate(labels):
data_interstate[(i + 1, 0)] = {
'1e_soc_mat': [0j, 0j, 0j],
'soc_units': 'cm-1'
}
data_interstate[(0, i + 1)] = {
'1e_soc_mat': [0j, 0j, 0j],
'soc_units': 'cm-1'
}
for j, label2 in enumerate(labels):
if (label[0] == 'S'
or label2[0] == 'S') and (label[0] != label2[0]):
data_interstate[(i + 1, j + 1)] = {
'1e_soc_mat': [[0j, 0j, 0j]],
'soc_units': 'cm-1'
}
elif label[0] == 'T' and label2[0] == 'T':
data_interstate[(i + 1, j + 1)] = {
'1e_soc_mat': [[0j, 0j, 0j], [0j, 0j, 0j],
[0j, 0j, 0j]],
'soc_units': 'cm-1'
}
elif label[0] == 'S' and label2[0] == 'S':
data_interstate[(i + 1, j + 1)] = {
'1e_soc_mat': [[0j]],
'soc_units': 'cm-1'
}
else:
raise ParserError('basic_cis', 'State multiplicity error')
for i, label in enumerate(labels):
for k2, ms2 in enumerate([-1, 0, 1]):
for j, label2 in enumerate(labels):
if label[0] == 'T':
for k, ms in enumerate([-1, 0, 1]):
enum = soc_section.find(
'SOC between the {} (ms={}) state and excited triplet states (ms={})'
.format(label, ms2, ms))
for line in soc_section[enum:enum + 50 *
(n_triplet + 1)].split(
'\n'):
if len(line.split()) == 0:
break
if line.split()[0] == '{}(ms={})'.format(
label2, ms):
data_interstate[(
i + 1, j + 1
)]['1e_soc_mat'][k2][k] = _list_to_complex(
line.split()[1:4])
data_interstate[(
i + 1, j + 1
)]['1e_soc_mat'][k][k2] = _list_to_complex(
line.split()[1:4])
data_interstate[(
j + 1, i + 1
)]['1e_soc_mat'][k2][k] = _list_to_complex(
line.split()[1:4])
data_interstate[(
j + 1, i + 1
)]['1e_soc_mat'][k][k2] = _list_to_complex(
line.split()[1:4])
break
elif label[0] == 'S':
for k, ms in enumerate([-1, 0, 1]):
enum = soc_section.find(
'SOC between the {} state and excited triplet states (ms={})'
.format(label, ms))
for line in soc_section[enum:enum + 50 *
(n_triplet + 1)].split(
'\n'):
if len(line.split()) == 0:
break
if line.split()[0] == '{}(ms={})'.format(
label2, ms):
data_interstate[(
i + 1, j + 1
)]['1e_soc_mat'][0][k] = _list_to_complex(
line.split()[1:4])
data_interstate[(
j + 1, i + 1
)]['1e_soc_mat'][0][k] = _list_to_complex(
line.split()[1:4])
break
else:
raise ParserError('basic_cis', 'SOC reading error')
enum = soc_section.find(
'SOC between the singlet ground state and excited triplet states (ms={})'
.format(ms2))
for line in soc_section[enum:enum + 50 *
(n_triplet + 1)].split('\n'):
if len(line.split()) == 0:
break
if line.split()[0] == '{}(ms={})'.format(label, ms2):
data_interstate[(
i + 1, 0)]['1e_soc_mat'][k2] = _list_to_complex(
line.split()[1:4])
data_interstate[(
0, i + 1)]['1e_soc_mat'][k2] = _list_to_complex(
line.split()[1:4])
break
data_dict['interstate_properties'] = data_interstate
return data_dict