def fp_oganov(self, delta=0.01, sigma=0.01):
struc_dist_x, struc_dist = self.structure_distances(delta=delta, sigma=sigma)
fp_oganov = struc_dist.copy()
vol = self.structure.volume
for spec_pair in struc_dist:
for i in range(len(struc_dist[spec_pair])):
specie0 = atomic_symbol(spec_pair[0])
specie1 = atomic_symbol(spec_pair[1])
number_atoms0 = self.structure.composition[specie0]
number_atoms1 = self.structure.composition[specie1]
fp_oganov[spec_pair][i] *= vol / (delta * number_atoms0 * number_atoms1)
fp_oganov[spec_pair][i] -= 1
return struc_dist_x, fp_oganov
def fp_oganov(self, delta=0.01, sigma=0.01):
struc_dist_x, struc_dist = self.structure_distances(delta=delta,
sigma=sigma)
fp_oganov = struc_dist.copy()
vol = self.structure.volume
for spec_pair in struc_dist:
for i in range(len(struc_dist[spec_pair])):
specie0 = atomic_symbol(spec_pair[0])
specie1 = atomic_symbol(spec_pair[1])
number_atoms0 = self.structure.composition[specie0]
number_atoms1 = self.structure.composition[specie1]
fp_oganov[spec_pair][i] *= vol / (delta * number_atoms0 *
number_atoms1)
fp_oganov[spec_pair][i] -= 1
return struc_dist_x, fp_oganov
def xyz2input(filename):
"""
Reads a .xyz and return an ABINIT input
as a python dictionary
"""
abiinput = AbinitInput()
atomdict = atomic_symbol()
rf = open(filename, 'r')
natom = int(rf.readline())
typat = []
znucl = []
xangst = []
ntypat = 0
rf.readline()
data = rf.readlines()
for i in range(natom):
atom = data[i].split()
atomnumber = atomdict[atom[0]]
if atomnumber not in znucl:
ntypat += 1
znucl.append(atomnumber)
typat.append(znucl.index(atomnumber) + 1)
xangst += [float(atom[1]), float(atom[2]), float(atom[3])]
abiinput.variables['natom'] = np.array([natom])
abiinput.variables['znucl'] = np.array(znucl)
abiinput.variables['ntypat'] = np.array([ntypat])
abiinput.variables['typat'] = np.array(typat)
abiinput.variables['xcart'] = angstrom_bohr * np.array(xangst)
return abiinput
def xyz2input(filename):
"""
Reads a .xyz and return an ABINIT input
as a python dictionary
"""
abiinput = AbinitInput()
atomdict = atomic_symbol()
rf = open(filename, 'r')
natom = int(rf.readline())
typat = []
znucl = []
xangst = []
ntypat = 0
rf.readline()
data = rf.readlines()
for i in range(natom):
atom = data[i].split()
atomnumber = atomdict[atom[0]]
if atomnumber not in znucl:
ntypat += 1
znucl.append(atomnumber)
typat.append(znucl.index(atomnumber) + 1)
xangst += [float(atom[1]), float(atom[2]), float(atom[3])]
abiinput.variables['natom'] = np.array([natom])
abiinput.variables['znucl'] = np.array(znucl)
abiinput.variables['ntypat'] = np.array([ntypat])
abiinput.variables['typat'] = np.array(typat)
abiinput.variables['xcart'] = angstrom_bohr * np.array(xangst)
return abiinput
def run_one(a):
"""
Take one OQMD 'Entry' object, search all the calculations associated and take the best calculation
in order to insert its data into the PyChemia Database
:param a: OQMD Entry object
:return:
"""
print('Entry: %6d Number of Calculations: %3d Energies: %s' %
(a.id, a.calculation_set.count(),
str([c.energy for c in a.calculation_set.all()])))
energy = 1E10
best_calculation = None
for c in a.calculation_set.all():
if c.energy is not None and c.energy < energy:
best_calculation = c
energy = c.energy
if best_calculation is None:
return None, None
cell = best_calculation.output.cell.T
symbols = atomic_symbol(best_calculation.output.atomic_numbers)
reduced = best_calculation.output.coords
structure = pychemia.Structure(cell=cell, symbols=symbols, reduced=reduced)
structure_id = best_calculation.output_id
entry_id = a.id
calculation_id = best_calculation.id
energy_pa = best_calculation.energy_pa
energy = best_calculation.energy
band_gap = best_calculation.band_gap
settings = best_calculation.settings
try:
spacegroup_number = best_calculation.output.spacegroup.number
except ValueError:
spacegroup_number = None
symm = pychemia.crystal.CrystalSymmetry(structure)
sym2 = symm.number(1E-2)
properties = {
'oqmd': {
'structure_id': structure_id,
'entry_id': entry_id,
'calculation_id': calculation_id,
'energy_pa': energy_pa,
'energy': energy,
'band_gap': band_gap,
'settings': settings,
'spacegroup_number': spacegroup_number
},
'spacegroup_number': {
'value': sym2,
'symprec': 1E-2
}
}
return structure, properties
def atom_name(self, iatom, idtset=''):
"""
Return the name of the atom and the position in the
list of atoms such as H3, N4, etc
"""
atomnumber = self.get_value('znucl', idtset=idtset)
if isinstance(atomnumber, list):
atomnumber = atomnumber[self.get_value('typat', idtset=idtset)[iatom] - 1]
return atomic_symbol(atomnumber) + str(iatom + 1)
def atom_name(self, iatom, idtset=''):
"""
Return the name of the atom and the position in the
list of atoms such as H3, N4, etc
"""
atomnumber = self.get_value('znucl', idtset=idtset)
if isinstance(atomnumber, list):
atomnumber = atomnumber[self.get_value('typat', idtset=idtset)[iatom] - 1]
return atomic_symbol(atomnumber) + str(iatom + 1)
def load(filename):
symbols = loadtxt(filename, skiprows=2, usecols=[0], dtype='|S2', ndmin=1)
symbols = [x.decode('utf-8') for x in symbols]
for i in range(len(symbols)):
if symbols[i].isdigit():
symbols[i] = atomic_symbol(int(symbols[i]))
positions = loadtxt(filename, skiprows=2, usecols=(1, 2, 3), ndmin=2)
natom = len(symbols)
periodicity = 3 * [False]
return Structure(symbols=symbols, positions=positions, periodicity=periodicity, natom=natom)
def read_geometry_bas(filename):
rf = open(filename, 'r')
natom = int(rf.readline())
symbols = []
positions = np.zeros((natom, 3))
for i in range(natom):
line = rf.readline().split()
symbols.append(atomic_symbol(int(line[0])))
positions[i] = np.array(line[1:])
return Structure(symbols=symbols, positions=positions, periodicity=False)
def read_geometry_bas(filename):
rf = open(filename, 'r')
natom = int(rf.readline())
symbols = []
positions = np.zeros((natom, 3))
for i in range(natom):
line = rf.readline().split()
symbols.append(atomic_symbol(int(line[0])))
positions[i] = np.array(line[1:])
return Structure(symbols=symbols, positions=positions, periodicity=False)
def run_one(a):
"""
Take one OQMD 'Entry' object, search all the calculations associated and take the best calculation
in order to insert its data into the PyChemia Database
:param a: OQMD Entry object
:return:
"""
print('Entry: %6d Number of Calculations: %3d Energies: %s' % (a.id, a.calculation_set.count(),
str([c.energy for c in a.calculation_set.all()])))
energy = 1E10
best_calculation = None
for c in a.calculation_set.all():
if c.energy is not None and c.energy < energy:
best_calculation = c
energy = c.energy
if best_calculation is None:
return None, None
cell = best_calculation.output.cell.T
symbols = atomic_symbol(best_calculation.output.atomic_numbers)
reduced = best_calculation.output.coords
structure = pychemia.Structure(cell=cell, symbols=symbols, reduced=reduced)
structure_id = best_calculation.output_id
entry_id = a.id
calculation_id = best_calculation.id
energy_pa = best_calculation.energy_pa
energy = best_calculation.energy
band_gap = best_calculation.band_gap
settings = best_calculation.settings
try:
spacegroup_number = best_calculation.output.spacegroup.number
except ValueError:
spacegroup_number = None
symm = pychemia.symm.StructureSymmetry(structure)
sym2 = symm.number(1E-2)
properties = {'oqmd': {'structure_id': structure_id,
'entry_id': entry_id,
'calculation_id': calculation_id,
'energy_pa': energy_pa,
'energy': energy,
'band_gap': band_gap,
'settings': settings,
'spacegroup_number': spacegroup_number},
'spacegroup_number': {'value': sym2, 'symprec': 1E-2}}
return structure, properties
def load(filename):
symbols = loadtxt(filename, skiprows=2, usecols=[0], dtype='|S2', ndmin=1)
symbols = [x.decode('utf-8') for x in symbols]
for i in range(len(symbols)):
if symbols[i].isdigit():
symbols[i] = atomic_symbol(int(symbols[i]))
positions = loadtxt(filename, skiprows=2, usecols=(1, 2, 3), ndmin=2)
natom = len(symbols)
periodicity = 3 * [False]
return Structure(symbols=symbols,
positions=positions,
periodicity=periodicity,
natom=natom)
def psp_name(atomicnumber, exchange, kind):
"""
Return the filename of a certain PSP given the
atomic number, exchange ('LDA' or 'GGA')
and kind ('FHI','TM')
:param atomicnumber: (int) Atomic number
:param exchange: (str) 'LDA' or 'GGA'
:param kind: (str) Source of Pseudopotentials
:return:
"""
atom_symbol = str(atomic_symbol(atomicnumber))
if kind == 'FHI' and exchange == 'LDA':
filename = str(atomicnumber).zfill(2) + '-' + atom_symbol + '.LDA.fhi'
elif kind == 'FHI' and exchange == 'GGA':
filename = str(atomicnumber).zfill(2) + '-' + atom_symbol + '.GGA.fhi'
elif kind == 'CORE' and exchange == 'LDA':
filename = str(atomicnumber) + atom_symbol.lower() + '.1s_psp.mod'
elif kind == 'GTH' and exchange == 'LDA':
filename = str(atomicnumber).zfill(2) + atom_symbol.lower() + '.pspgth'
elif kind == 'TM' and exchange == 'LDA':
filename = str(atomicnumber) + atom_symbol.lower() + '.pspnc'
elif kind == 'AE' and exchange == 'DEN':
filename = '0.' + str(atomicnumber).zfill(
2) + '-' + atom_symbol + '.8.density.AE'
elif kind == 'FC' and exchange == 'DEN':
filename = str(atomicnumber).zfill(2) + '-' + atom_symbol + '.8.fc'
elif kind == 'PAW' and exchange == 'GGA':
filename = 'JTH-PBE-atomicdata-0.2/' + atom_symbol + '.GGA_PBE-JTH.xml'
elif kind == 'PAW' and exchange == 'LDA':
filename = 'JTH-LDA-atomicdata-0.2/' + atom_symbol + '.LDA_PW-JTH.xml'
elif kind == 'HGH' and exchange == 'GGA':
filename = str(atomicnumber).zfill(
2) + atom_symbol.lower() + '.pbe_hgh'
elif kind == 'ONC' and exchange == 'PBE':
filename = 'pbe_s_sr' + os.sep + atom_symbol + '.psp8'
else:
print('The combination of exchange=%s and kind=%s is not known' %
(exchange, kind))
filename = ''
return filename
def psp_name(atomicnumber, exchange, kind):
"""
Return the filename of a certain PSP given the
atomic number, exchange ('LDA' or 'GGA')
and kind ('FHI','TM')
:param atomicnumber: (int) Atomic number
:param exchange: (str) 'LDA' or 'GGA'
:param kind: (str) Source of Pseudopotentials
:return:
"""
atom_symbol = str(atomic_symbol(atomicnumber))
if kind == 'FHI' and exchange == 'LDA':
filename = str(atomicnumber).zfill(2) + '-' + atom_symbol + '.LDA.fhi'
elif kind == 'FHI' and exchange == 'GGA':
filename = str(atomicnumber).zfill(2) + '-' + atom_symbol + '.GGA.fhi'
elif kind == 'CORE' and exchange == 'LDA':
filename = str(atomicnumber) + atom_symbol.lower() + '.1s_psp.mod'
elif kind == 'GTH' and exchange == 'LDA':
filename = str(atomicnumber).zfill(2) + atom_symbol.lower() + '.pspgth'
elif kind == 'TM' and exchange == 'LDA':
filename = str(atomicnumber) + atom_symbol.lower() + '.pspnc'
elif kind == 'AE' and exchange == 'DEN':
filename = '0.' + str(atomicnumber).zfill(2) + '-' + atom_symbol + '.8.density.AE'
elif kind == 'FC' and exchange == 'DEN':
filename = str(atomicnumber).zfill(2) + '-' + atom_symbol + '.8.fc'
elif kind == 'PAW' and exchange == 'GGA':
filename = 'JTH-PBE-atomicdata-0.2/' + atom_symbol + '.GGA_PBE-JTH.xml'
elif kind == 'PAW' and exchange == 'LDA':
filename = 'JTH-LDA-atomicdata-0.2/' + atom_symbol + '.LDA_PW-JTH.xml'
elif kind == 'HGH' and exchange == 'GGA':
filename = str(atomicnumber).zfill(2) + atom_symbol.lower() + '.pbe_hgh'
elif kind == 'ONC' and exchange == 'PBE':
filename = 'pbe_s_sr' + os.sep + atom_symbol + '.psp8'
else:
print('The combination of exchange=%s and kind=%s is not known' % (exchange, kind))
filename = ''
return filename
def run_one(a):
"""
Take one OQMD 'Entry' object, search all the calculations associated and take the best calculation
in order to insert its data into the PyChemia Database
:param a: OQMD Entry object
:return:
"""
energy = 1E10
best_calculation = None
if a.calculation_set.count() > 0:
if 'standard' in a.calculations:
best_calculation = a.calculations['standard']
calculation_name = 'standard'
elif 'fine_relax' in a.calculations:
best_calculation = a.calculations['fine_relax']
calculation_name = 'fine_relax'
elif 'coarse_relax' in a.calculations:
best_calculation = a.calculations['coarse_relax']
calculation_name = 'coarse_relax'
elif 'static' in a.calculations:
best_calculation = a.calculations['static']
calculation_name = 'static'
elif 'relaxation' in a.calculations:
best_calculation = a.calculations['relaxation']
calculation_name = 'relaxation'
elif len(a.calculations) > 0:
calculations = sorted(a.calculations.keys())
print('Calculations found: %s, using the last one' % calculations)
best_calculation = a.calculations[calculations[-1]]
calculation_name = calculations[-1]
else:
print('ERROR: Count > 0 and no calculation found')
if best_calculation is not None:
structure_name = None
if best_calculation.output is not None:
structure_used = best_calculation.output
structure_id = best_calculation.output_id
from_output = True
elif best_calculation.input is not None:
print(
'WARNING: No data was found from the output of the calculation, using input geometries and leaving energetics empty'
)
structure_used = best_calculation.input
structure_id = best_calculation.input_id
from_output = False
else:
calculation_name = None
if a.structures is not None and len(a.structures) > 0:
struct_keys = sorted(a.structures.keys())
print(
"WARNING: Calculation not found for %s. Structures found: %s using the first one "
% (a, struct_keys))
structure_used = a.structures[struct_keys[0]]
structure_id = None
from_output = False
structure_name = struct_keys[0]
else:
print("ERROR: No calculation and no structure found for %s" % a)
return None, None
cell = structure_used.cell.T
symbols = atomic_symbol(structure_used.atomic_numbers)
reduced = structure_used.coords
structure = pychemia.Structure(cell=cell, symbols=symbols, reduced=reduced)
entry_id = a.id
if best_calculation is not None:
calculation_id = best_calculation.id
energy_pa = best_calculation.energy_pa
energy = best_calculation.energy
band_gap = best_calculation.band_gap
settings = best_calculation.settings
try:
spacegroup_number = best_calculation.output.spacegroup.number
except AttributeError:
spacegroup_number = None
else:
calculation_id = None
energy_pa = None
energy = None
band_gap = None
settings = None
spacegroup_number = None
from_output = False
try:
symm = pychemia.crystal.CrystalSymmetry(structure)
sym2 = symm.number(1E-2)
except ValueError:
sym2 = None
properties = {
'oqmd': {
'structure_id': structure_id,
'entry_id': entry_id,
'calculation_id': calculation_id,
'energy_pa': energy_pa,
'energy': energy,
'band_gap': band_gap,
'settings': settings,
'from_output': from_output,
'calculation_name': calculation_name,
'structure_name': structure_name,
'spacegroup_number': spacegroup_number
},
'spacegroup_number': {
'value': sym2,
'symprec': 1E-2
}
}
return structure, properties
def get_structure(self, idtset=None, units='bohr'):
"""
Return the atomic structure from the input object
for a given dataset (no dataset by default)
"""
# NATOM
natom = self.get_value('natom', idtset)
# SYMBOLS
ntypat = self.get_value('ntypat', idtset)
symbols = []
znucl = self.get_value('znucl', idtset)
typat = self.get_value('typat', idtset)
for i in range(natom):
# NOTE: znucl is a real number in OUT.nc
# Alchemic mixing is not allow here
if ntypat == 1:
symbols.append(atomic_symbol(int(znucl)))
else:
symbols.append(atomic_symbol(int(znucl[typat[i] - 1])))
# POSITIONS
xangst = self.get_value('xangst', idtset)
xcart = self.get_value('xcart', idtset)
xred = self.get_value('xred', idtset)
rprim = self.get_value('rprim', idtset)
acell = np.array(self.get_value('acell', idtset))
# Set rprimd and acell using the default values
# if not found
if rprim is None:
rprim = np.identity(3)
else:
rprim = np.array(rprim).reshape((3, 3))
if acell is None:
acell = np.ones(3)
rprimd = np.zeros((3, 3))
rprimd[0] = rprim[0] * acell
rprimd[1] = rprim[1] * acell
rprimd[2] = rprim[2] * acell
if xangst is not None:
xangst = np.array(xangst)
positions = xangst.reshape((natom, 3))
if units == 'bohr':
positions = positions * angstrom_bohr
elif xcart is not None:
xcart = np.array(xcart)
positions = xcart.reshape((natom, 3))
if units == 'angstrom':
positions = positions * bohr_angstrom
elif xred is not None:
xred = np.array(xred)
xred = xred.reshape((natom, 3))
xcart = np.zeros((natom, 3))
# print rprimd
# print xred
for i in range(natom):
xcart[i] = xred[i, 0] * rprimd[0] + xred[i, 1] * rprimd[1] + xred[i, 2] * rprimd[2]
positions = xcart
if units == 'angstrom':
positions = positions * bohr_angstrom
else:
positions = None
if units == 'angstrom':
rprimd = rprimd * bohr_angstrom
# Create an object atomic_structure
structure = Structure(natom=natom, symbols=symbols, positions=positions, cell=rprimd)
return structure
def get_structure(self, idtset=None, units='bohr'):
"""
Return the atomic structure from the input object
for a given dataset (no dataset by default)
"""
# NATOM
natom = self.get_value('natom', idtset)
# SYMBOLS
ntypat = self.get_value('ntypat', idtset)
symbols = []
znucl = self.get_value('znucl', idtset)
typat = self.get_value('typat', idtset)
for i in range(natom):
# NOTE: znucl is a real number in OUT.nc
# Alchemic mixing is not allow here
if ntypat == 1:
symbols.append(atomic_symbol(int(znucl)))
else:
symbols.append(atomic_symbol(int(znucl[typat[i] - 1])))
# POSITIONS
xangst = self.get_value('xangst', idtset)
xcart = self.get_value('xcart', idtset)
xred = self.get_value('xred', idtset)
rprim = self.get_value('rprim', idtset)
acell = np.array(self.get_value('acell', idtset))
# Set rprimd and acell using the default values
# if not found
if rprim is None:
rprim = np.identity(3)
else:
rprim = np.array(rprim).reshape((3, 3))
if acell is None:
acell = np.ones(3)
rprimd = np.zeros((3, 3))
rprimd[0] = rprim[0] * acell
rprimd[1] = rprim[1] * acell
rprimd[2] = rprim[2] * acell
if xangst is not None:
xangst = np.array(xangst)
positions = xangst.reshape((natom, 3))
if units == 'bohr':
positions = positions * angstrom_bohr
elif xcart is not None:
xcart = np.array(xcart)
positions = xcart.reshape((natom, 3))
if units == 'angstrom':
positions = positions * bohr_angstrom
elif xred is not None:
xred = np.array(xred)
xred = xred.reshape((natom, 3))
xcart = np.zeros((natom, 3))
# print rprimd
# print xred
for i in range(natom):
xcart[i] = xred[i, 0] * rprimd[0] + xred[i, 1] * rprimd[1] + xred[i, 2] * rprimd[2]
positions = xcart
if units == 'angstrom':
positions = positions * bohr_angstrom
else:
positions = None
if units == 'angstrom':
rprimd = rprimd * bohr_angstrom
# Create an object atomic_structure
structure = Structure(natom=natom, symbols=symbols, positions=positions, cell=rprimd)
return structure
def run_one(a):
"""
Take one OQMD 'Entry' object, search all the calculations associated and take the best calculation
in order to insert its data into the PyChemia Database
:param a: OQMD Entry object
:return:
"""
energy = 1E10
best_calculation = None
if a.calculation_set.count() > 0:
if 'standard' in a.calculations:
best_calculation = a.calculations['standard']
calculation_name = 'standard'
elif 'fine_relax' in a.calculations:
best_calculation = a.calculations['fine_relax']
calculation_name = 'fine_relax'
elif 'coarse_relax' in a.calculations:
best_calculation = a.calculations['coarse_relax']
calculation_name = 'coarse_relax'
elif 'static' in a.calculations:
best_calculation = a.calculations['static']
calculation_name = 'static'
elif 'relaxation' in a.calculations:
best_calculation = a.calculations['relaxation']
calculation_name = 'relaxation'
elif len(a.calculations) > 0:
calculations = sorted(a.calculations.keys())
print('Calculations found: %s, using the last one' % calculations)
best_calculation = a.calculations[calculations[-1]]
calculation_name = calculations[-1]
else:
print('ERROR: Count > 0 and no calculation found')
if best_calculation is not None:
structure_name = None
if best_calculation.output is not None:
structure_used = best_calculation.output
structure_id = best_calculation.output_id
from_output = True
elif best_calculation.input is not None:
print(
'WARNING: No data was found from the output of the calculation, using input geometries and leaving '
'energetics empty')
structure_used = best_calculation.input
structure_id = best_calculation.input_id
from_output = False
else:
calculation_name = None
if a.structures is not None and len(a.structures) > 0:
struct_keys = sorted(a.structures.keys())
print("WARNING: Calculation not found for %s. Structures found: %s using the first one " % (a, struct_keys))
structure_used = a.structures[struct_keys[0]]
structure_id = None
from_output = False
structure_name = struct_keys[0]
else:
print("ERROR: No calculation and no structure found for %s" % a)
return None, None
cell = structure_used.cell.T
symbols = atomic_symbol(structure_used.atomic_numbers)
reduced = structure_used.coords
structure = pychemia.Structure(cell=cell, symbols=symbols, reduced=reduced)
entry_id = a.id
if best_calculation is not None:
calculation_id = best_calculation.id
energy_pa = best_calculation.energy_pa
energy = best_calculation.energy
band_gap = best_calculation.band_gap
settings = best_calculation.settings
try:
spacegroup_number = best_calculation.output.spacegroup.number
except AttributeError:
spacegroup_number = None
else:
calculation_id = None
energy_pa = None
energy = None
band_gap = None
settings = None
spacegroup_number = None
from_output = False
try:
symm = pychemia.crystal.CrystalSymmetry(structure)
sym2 = symm.number(1E-2)
except ValueError:
sym2 = None
properties = {'oqmd': {'structure_id': structure_id,
'entry_id': entry_id,
'calculation_id': calculation_id,
'energy_pa': energy_pa,
'energy': energy,
'band_gap': band_gap,
'settings': settings,
'from_output': from_output,
'calculation_name': calculation_name,
'structure_name': structure_name,
'spacegroup_number': spacegroup_number},
'spacegroup_number': {'value': sym2, 'symprec': 1E-2}}
return structure, properties
