class mp_query():
def __init__(self, api_key):
self.mpr = MPRester(api_key)
def find_access_strings(self, search):
data = self.mpr.get_data(search)
material_ids = [datum['material_id'] for datum in data]
return material_ids
def mp_structure(self, material_id):
struct = self.mpr.get_structure_by_material_id(material_id)
struct = SpacegroupAnalyzer(
struct).get_conventional_standard_structure()
return struct
def make_structures(self, search):
material_ids = self.find_access_strings(search)
structures = []
for ID in material_ids:
struct = self.mp_structure(ID)
structures.append(struct)
return structures
def get_mp_structure(cls, mpid):
"""
Get a structure from MP.
:param mpid: Materials Project id.
:return: Structure
"""
m = MPRester()
return m.get_structure_by_material_id(mpid)
def submit_tests(names=None, params=None):
sma = SubmissionMongoAdapter.auto_load()
# note: TiO2 is duplicated twice purposely, duplicate check should catch this
compounds = {
"Si": 149,
"Al": 134,
"ZnO": 2133,
"FeO": 18905,
"LiCoO2": 601860,
"LiFePO4": 585433,
"GaAs": 2534,
"Ge": 32,
"PbTe": 19717,
"YbO": 1216,
"SiC": 567551,
"Fe3C": 510623,
"SiO2": 547211,
"Na2O": 2352,
"InSb (unstable)": 10148,
"Sb2O5": 1705,
"N2O5": 554368,
"BaTiO3": 5020,
"Rb2O": 1394,
"TiO2": 554278,
"TiO2 (2)": 554278,
'BaNbTePO8': 560794,
"AgCl": 22922,
"AgCl (2)": 570858,
"SiO2 (2)": 555211,
"Mg2SiO4": 2895,
"CO2": 20066,
"PbSO4": 22298,
"SrTiO3": 5532,
"FeAl": 2658,
"AlFeCo2": 10884,
"NaCoO2": 554427,
"ReO3": 547271,
"LaH2": 24153,
"SiH3I": 28538,
"LiBH4": 30209,
"H8S5N2": 28143,
"LiOH": 23856,
"SrO2": 2697,
"Mn": 35,
"Hg4Pt": 2312,
"PdF4": 13868,
"Gd2WO6": 651333,
'MnO2': 19395,
'VO2': 504800
}
mpr = MPRester()
for name, sid in compounds.iteritems():
if not names or name in names:
sid = mpr.get_materials_id_from_task_id("mp-{}".format(sid))
s = mpr.get_structure_by_material_id(sid, final=False)
snl = StructureNL(s, 'Anubhav Jain <*****@*****.**>')
parameters = {'priority': 10} if name == 'Si' else None
if params:
parameters.update(params)
sma.submit_snl(snl, '*****@*****.**', parameters=parameters)
comp = Composition(phase).as_dict()
file_dir = './' + phase + '/'
# checking if path already exists - if not create it
if not os.path.exists(file_dir):
os.makedirs(file_dir)
else:
print(f'{file_dir} already exists\n-----')
# writing materials ID in file
mid_file = open(file_dir + 'material-ID', 'w')
struct = mpr.get_structure_by_material_id(stable_phases[phase],
final=False,
conventional_unit_cell=True)
# getting space group of structure
spacegroup = struct.get_space_group_info()
# prova = mpr.get_entry_by_material_id(stable_phases[phase] , inc_structure = 'final').as_dict()
print(
f'Structure acquired :\n - Reduced formula = {phase} , Space group = {spacegroup}, Materials ID = {stable_phases[phase]} \n --------'
)
#print(struct)
print('')
list_file.write(f'{phase} : {spacegroup} {stable_phases[phase]}\n')
mid_file.write(f'{stable_phases[phase]}')
mid_file.close()
# POSCAR
rest = MPRester('XaCJrv4nBIeuy3kd')
params_dict = {}
structures = {}
psps = os.listdir(
'/home/bcomer3/sparc/ase_sparc/pysparcx/sparc/pseudos/LDA_pseudos/')
available_psps = [a.split('.')[0] for a in psps]
# metals
metals = ['Al', 'Ru', 'Cu', 'Bi']
for metal in metals:
struc_l = rest.get_entries(metal, sort_by_e_above_hull=True)
struc_l = struc_l[0]
struct = rest.get_structure_by_material_id(struc_l.entry_id)
atoms = AseAtomsAdaptor.get_atoms(struct) # switch to ASE
for millers in [(1, 1, 1)]:
slab = surface(atoms, millers, 4, vacuum=6)
slab *= (2, 2, 1)
n_metal = metal
if millers == (1, 1, 1):
if metal == 'Cu':
add_adsorbate(slab, molecule('CO'), 2.5, (0, 1.64))
n_metal += '_COads'
structure = AseAtomsAdaptor.get_structure(slab)
add_to_dict(structure, n_metal, millers)
metals = ['In', 'Rb', 'Sr', 'Te']
class MPData(Dataset):
"""
The MPData dataset is a wrapper for a dataset from Materials Project.
atom_init.json: a JSON file that stores the initialization vector for each
element.
ID.cif: a CIF file that recodes the crystal structure, where ID is the
unique ID for the crystal.
Parameters
----------
criteria: dict
Mongo-like query language for MP structures
atom_init_file_dir: str
dir of the atom_init_file.json
max_num_nbr: int
The maximum number of neighbors while constructing the crystal graph
radius: float
The cutoff radius for searching neighbors
dmin: float
The minimum distance for constructing GaussianDistance
step: float
The step size for constructing GaussianDistance
random_seed: int
Random seed for shuffling the dataset
Returns
-------
atom_fea: torch.Tensor shape (n_i, atom_fea_len)
nbr_fea: torch.Tensor shape (n_i, M, nbr_fea_len)
nbr_fea_idx: torch.LongTensor shape (n_i, M)
target: torch.Tensor shape (1, )
cif_id: str or int
"""
def __init__(self,
criteria,
atom_init_file_dir,
api_key="VIqD4QUxH6wNpyc5",
max_num_nbr=12,
radius=8,
dmin=0,
step=0.2,
random_seed=123):
self.api = MPRester(api_key)
mp_list = [
i['material_id']
for i in self.api.query(criteria=criteria,
properties=['material_id'])
]
self.max_num_nbr, self.radius = max_num_nbr, radius
atom_init_file = atom_init_file_dir
self.ari = AtomCustomJSONInitializer(atom_init_file)
self.gdf = GaussianDistance(dmin=dmin, dmax=self.radius, step=step)
self.id_prop_data = [(target, cif_id)
for target, cif_id in enumerate(mp_list)]
def __len__(self):
return len(self.id_prop_data)
@functools.lru_cache(maxsize=None)
def __getitem__(self, idx):
target, cif_id = self.id_prop_data[idx]
crystal = self.api.get_structure_by_material_id(cif_id)
atom_fea = np.vstack([
self.ari.get_atom_fea(crystal[i].specie.number)
for i in range(len(crystal))
])
atom_fea = torch.Tensor(atom_fea)
all_nbrs = crystal.get_all_neighbors(self.radius, include_index=True)
all_nbrs = [sorted(nbrs, key=lambda x: x[1]) for nbrs in all_nbrs]
nbr_fea_idx = np.array([
list(
map(lambda x: x[2], nbr[:self.max_num_nbr])
for nbr in all_nbrs)
])
nbr_fea = np.array([list(map(lambda x: x[1], nbr[: self.max_num_nbr])) \
for nbr in all_nbrs]
)
atom_fea = torch.Tensor(atom_fea)
nbr_fea = torch.Tensor(nbr_fea)
nbr_fea_idx = torch.LongTensor(nbr_fea_idx)
target = torch.Tensor([float(target)])
return (atom_fea, nbr_fea, nbr_fea_idx), target, cif_id
def cif_lib_build(self, crystal_system, size_limit=None):
''' function to build cif and pdf library based on space group symbol
Parameters
----------
crystal_system: str
name of crystal system. It capitalized, like CUBIC.
space group symbol will be generated by get_symbol_list method
size_list : int
optional. Uppder limit of data pulled out per symbol
'''
self.crystal_system = crystal_system
space_group_symbol = self.get_symbol_list(crystal_system)
if isinstance(space_group_symbol, list):
space_group_symbol_set = space_group_symbol
else:
space_group_symbol_set = list(spac_group_symbol)
## changing dir
data_dir = os.path.join(self.working_dir, crystal_system)
self.data_dir = data_dir
self._makedirs(data_dir)
os.chdir(data_dir)
if os.getcwd() == data_dir:
print('Library will be built at %s' % data_dir)
else:
e = 'Werid, return'
raise RuntimeError(e)
# summary lists
missed_list = [] # reference
m_id_list = [] # reference for searchs have been done in the past
time_info = time.strftime('%Y-%m-%d')
# create dirs, cif and calculated dir
cif_dir = os.path.join(data_dir, 'cif_data')
self._makedirs(cif_dir)
self.cif_dir = cif_dir
# looping
for space_group_symbol in space_group_symbol_set:
print('Building library with space_group symbol: {}'.format(space_group_symbol))
## search query
m = MPRester(self.API_key)
search = m.query(criteria = {"spacegroup.symbol": space_group_symbol},
properties = ["material_id"])
if search:
## crazy looping
if size_limit:
dim = 400 # 400 data sets per symbol
else:
dim = len(search)
print('Pull out %s data sets' % dim)
print('Now, starts to save cif and compute pdf...')
for i in range(dim):
# part 1: grab cif files from data base
m_id = search[i]['material_id']
m_id_list.append(m_id)
m_struc = m.get_structure_by_material_id(m_id)
m_formula = m_struc.formula
m_name = m_formula.replace(' ', '') # material name
cif_w = CifWriter(m_struc)
cif_name = '{}_{}.cif'.format(space_group_symbol, m_name)
cif_w_name = os.path.join(cif_dir, cif_name)
if os.path.isfile(cif_w_name):
print('already have {}, skip'.format(cif_name))
pass # skip files already exist
else:
cif_w.write_file(cif_w_name)
print('{} has been saved'.format(cif_name))
else:
print('Hmm, no reasult. Something wrong')
missed_list.append(space_group_symbol)
pass
m_id_list_name = '{}_{}_material_id.txt'.format(crystal_system, time_info)
m_id_list_w_name = os.path.join(data_dir, m_id_list_name)
np.savetxt(m_id_list_w_name, m_id_list)
print('''SUMMARY: for {} cystsal sytem,
Symbols {} can't be found from data base'''.format(crystal_system, missed_list))
return cif_dir
from pymatgen.io.pwscf import PWInput
from pymatgen import MPRester, Structure
mpr = MPRester()
s: Structure = mpr.get_structure_by_material_id('mp-550893')
s.lattice
s.make_supercell([[-1, -1, 1], [-1, 1, -1], [1, -1, -1]])
pwin = PWInput(s,
pseudo={
'Hf': 'Hf.pbe-spn-kjpaw_psl.1.0.0.UPF',
'O': 'O.pbe-n-kjpaw_psl.1.0.0.UPF'
},
kpoints_grid=(6, 6, 6))
pwin.write_file('Hf-scf.in')
#this is to build a database for cnn to use
import pandas as pd
from pymatgen import Structure
from pymatgen import MPRester
import matplotlib.pyplot as plt
api_key = 'x3NlvC67Z9tPykwGz'
mpr = MPRester(api_key)
df = pd.read_excel(
r'D:\FYP_files\database\data_after_processing\CGCNN_bulkmodulus_traindata.xlsx'
) #this is how to get the cif files
for i in range(
0, 13110
): #this is the number of data in 2019/12/1, there must be more datas
structure = mpr.get_structure_by_material_id("%s" % df.iat[i, 0])
structure.to(
"cif", "D:/FYP_files/Deep_learning/cgcnn-master/data/example/%s.cif" %
df.iat[i, 0])
print(i)
# python main.py --train-ratio 0.6 --val-ratio 0.2 --test-ratio 0.2 data/sample-regression
# python predict.py pre-trained/formation-energy-per-atom.pth.tar data/sample-regression
#python init.py --idx 2
# plot(np.array(bs.bands[sp])[iband-3,:].T-bs.efermi) # from MP
plot(en, color='b') # interpolated by BoltzTraP
show()
if __name__ == "__main__":
# user inputs
PbTe_id = 'mp-19717' # valence_idx = 9
Si_id = 'mp-149' # valence_idx = 4
GaAs_id = 'mp-2534' # valence_idx = ?
SnSe2_id = "mp-665"
Si_bs = api.get_bandstructure_by_material_id(Si_id)
bs = Vasprun('GaAs_28electron_ncsf_line_vasprun.xml').get_band_structure(
kpoints_filename='GaAs_28electron_KPOINTS', line_mode=True)
GaAs_st = api.get_structure_by_material_id(GaAs_id)
bs.structure = GaAs_st
# Si_bs.structure = api.get_structure_by_material_id(Si_id)
Si_bs.structure = Vasprun("../test_files/Si/vasprun.xml").final_structure
print(bs.get_sym_eq_kpoints([0.5, 0.5, 0.5]))
print(bs.get_sym_eq_kpoints([0.5, 0., 0.5]))
# vbm_idx = bs.get_vbm()['band_index'][Spin.up][0]
vbm_idx, _ = get_bindex_bspin(Si_bs.get_vbm(), is_cbm=False)
print('vbm band index (vbm_idx): {}'.format(vbm_idx))
ibands = [1, 2] # in this notation, 1 is the last valence band
ibands = [i + vbm_idx for i in ibands]
PbTe_coeff_file = '../test_files/PbTe/fort.123'
Si_coeff_file = "../test_files/Si/Si_fort.123"
options.dimension = len(system)
for entry in mp_entries:
accept = False
if type(system) is list:
if len(entry.composition) >= options.dimension:
eng = pd.get_e_above_hull(entry)
if eng <= options.cutoff:
accept = True
else:
eng = entry.energy_per_atom
accept = True
if accept:
struc = output_struc(entry, eng, tol=1e-2)
strucs.append(struc)
else:
strucs.append(mpr.get_structure_by_material_id(options.id))
if options.format == 'poscar':
filename = 'MPR.vasp'
else:
filename = 'MPR.cif'
if os.path.isfile(filename):
os.remove(filename)
for struc in strucs:
if options.format == 'poscar':
content = struc.to(fmt='poscar')
else:
content = str(CifWriter(struc, symprec=0.01))
with open(filename, 'a+') as f:
PlotDict = {}
for index in MI_List:
for key in MI_lib.keys():
if any((index == x).all() for x in MI_lib[key]):
if key not in PlotDict.keys():
PlotDict[key] = 0
PlotDict[key] += 1
print(index, MI_lib[key])
import matplotlib.pyplot as plt
plt.bar(PlotDict.keys(), PlotDict.values(), color='g')
plt.show()
def MillerIndexLibrary(structure):
MI_lib = {}
UniqueIndices = get_symmetrically_distinct_miller_indices(structure,
max_index=4)
for Index in UniqueIndices:
MI_lib[str(Index)] = get_sym_fam(
structure).get_equivalent_miller_indices(Index)
return MI_lib
MPR = MPRester("2d5wyVmhDCpPMAkq")
S = MPR.get_structure_by_material_id('mp-1143', conventional_unit_cell='True')
print(S)
MI_lib = MillerIndexLibrary(S)
print(MI_lib)
surfscreen(S, MI_lib)
def v_ij(self, ri, rj, arr):
a_ij = 1
beta = 0.33675
n = 22.956
c = 4.8381
d = 2.0417
r_ij = np.linalg.norm(ri, rj)
ksi = lambda: sum([
self.fc(norm(ri, rk)) * g_theta(Functional.get_angle(ri, rj, rk)) *
math.exp(self.lambda2**3 * (norm(ri, rj) - norm(ri, rk))**3)
for rk in arr
])
b_ij = lambda: 1 + beta**n * ksi()**n
g_theta = lambda theta: 1 + c**2 / d**2 - c**2 / (d**2 +
(math.cos(theta))**2)
return self.fc(r_ij) * (a_ij * b_ij() * self.fa(r_ij))
mpr = MPRester('DhmFQPuibZo8JtXn')
s: Structure = mpr.get_structure_by_material_id('mp-149')
for i in s.get_all_neighbors(3, True):
for j in i:
print(j)
#f = Functional()
#x = np.logspace(-5, 2, 1000)
#y = [f.fc(r) for r in x]
#plt.plot(x, y)
#plt.show()
q_pool = []
for x,y in enumerate(q):
q_pool.append([q[x]['energy_per_atom'],q[x]['volume'],q[x]['material_id'],q[x]['pretty_formula']])
q_pool = sorted(q_pool, key = lambda x:x[0], reverse=True)
q_min = []
for m,n in enumerate(q_pool):
min_ = q_pool[-1][0]
if q_pool[m][0] - min_ <= 0.05:
q_min.append(q_pool[m])
q_min = sorted(q_min, key = lambda x:x[1], reverse=True)
print('Min energy struct')
print(q_min[-1])
mp_structure = mpr.get_structure_by_material_id(q_min[-1][2],conventional_unit_cell=True)
ase_atoms = bridge.get_atoms(mp_structure)
ase_atoms.write('POSCAR_'+str(q_min[-1][3])+"_"+str(q_min[-1][2]),format='vasp')
'''
for no, entry in enumerate(pm_atom_list):
ase_atoms = bridge.get_atoms(entry)
mk_dir = str(name)+'-'+str(pm_mat_id[no])
os.makedirs(mk_dir)
os.chdir(mk_dir)
ase_atoms = bridge.get_atoms(entry)
ase_atoms.write('POSCAR',format='vasp')
vol_k = int(15**3/entry.volume)
k=pm_vasp.Kpoints.automatic_density_by_vol(structure=entry,kppvol=vol_k)
k.write_file('KPOINTS')
num_fin = len(df_entries)
createFolder('results/plots/pourbaix')
with open('results/pbx_analysis.log', 'w') as f:
for idx in range(num_ini, num_fin):
formula = df_entries['formula'][idx]
mpid = df_entries['mp_id'][idx]
print('%03d_%s' % (idx + 1.0, formula))
f.writelines('%03d_%s\n' % (idx + 1.0, formula))
try:
struc = mpr.get_structure_by_material_id(mpid)
elements = [str(atom) for atom in struc.species[:2]]
entries = mpr.get_pourbaix_entries(elements)
entry_ids = [e for e in entries if e.entry_id == mpid]
if len(entry_ids) == 1:
pbx = PourbaixDiagram(entries)
plotter = PourbaixPlotter(pbx)
entry = entry_ids[0]
plt = plotter.plot_entry_stability(entry)
plt.savefig('results/plots/pourbaix/%03d_%s.png' %
(idx + 1.0, formula),
dpi=300)
elif len(entry_ids) > 1:
f.writelines('%03d_%s: number of entry > 1 \n' %
(idx + 1.0, formula))
def get_mp_structure(cls, mpid):
m = MPRester()
return m.get_structure_by_material_id(mpid)
def compute_environments(chemenv_configuration):
string_sources = {'cif': {'string': 'a Cif file', 'regexp': '.*\.cif$'},
'mp': {'string': 'the Materials Project database', 'regexp': 'mp-[0-9]+$'}}
questions = {'c': 'cif'}
if chemenv_configuration.has_materials_project_access:
questions['m'] = 'mp'
lgf = LocalGeometryFinder()
lgf.setup_parameters()
allcg = AllCoordinationGeometries()
strategy_class = strategies_class_lookup[chemenv_configuration.package_options['default_strategy']['strategy']]
#TODO: Add the possibility to change the parameters and save them in the chemenv_configuration
default_strategy = strategy_class()
default_strategy.setup_options(chemenv_configuration.package_options['default_strategy']['strategy_options'])
max_dist_factor = chemenv_configuration.package_options['default_max_distance_factor']
firsttime = True
while True:
if len(questions) > 1:
found = False
print('Enter the source from which the structure is coming or <q> to quit :')
for key_character, qq in questions.items():
print(' - <{}> for a structure from {}'.format(key_character, string_sources[qq]['string']))
test = input(' ... ')
if test == 'q':
break
if test not in list(questions.keys()):
for key_character, qq in questions.items():
if re.match(string_sources[qq]['regexp'], str(test)) is not None:
found = True
source_type = qq
if not found:
print('Wrong key, try again ...')
continue
else:
source_type = questions[test]
else:
found = False
source_type = list(questions.values())[0]
if found and len(questions) > 1:
input_source = test
if source_type == 'cif':
if not found:
input_source = input('Enter path to cif file : ')
cp = CifParser(input_source)
structure = cp.get_structures()[0]
elif source_type == 'mp':
if not found:
input_source = input('Enter materials project id (e.g. "mp-1902") : ')
a = MPRester(chemenv_configuration.materials_project_api_key)
structure = a.get_structure_by_material_id(input_source)
lgf.setup_structure(structure)
print('Computing environments for {} ... '.format(structure.composition.reduced_formula))
se = lgf.compute_structure_environments_detailed_voronoi(maximum_distance_factor=max_dist_factor)
print('Computing environments finished')
while True:
test = input('See list of environments determined for each (unequivalent) site ? '
'("y" or "n", "d" with details, "g" to see the grid) : ')
strategy = default_strategy
if test in ['y', 'd', 'g']:
strategy.set_structure_environments(se)
for eqslist in se.equivalent_sites:
site = eqslist[0]
isite = se.structure.index(site)
try:
if strategy.uniquely_determines_coordination_environments:
ces = strategy.get_site_coordination_environments(site)
else:
ces = strategy.get_site_coordination_environments_fractions(site)
except NeighborsNotComputedChemenvError:
continue
if ces is None:
continue
if len(ces) == 0:
continue
comp = site.species_and_occu
#ce = strategy.get_site_coordination_environment(site)
if strategy.uniquely_determines_coordination_environments:
ce = ces[0]
if ce is None:
continue
thecg = allcg.get_geometry_from_mp_symbol(ce[0])
mystring = 'Environment for site #{} {} ({}) : {} ({})\n'.format(str(isite),
comp.get_reduced_formula_and_factor()[0],
str(comp),
thecg.name,
ce[0])
else:
mystring = 'Environments for site #{} {} ({}) : \n'.format(str(isite),
comp.get_reduced_formula_and_factor()[0],
str(comp))
for ce in ces:
cg = allcg.get_geometry_from_mp_symbol(ce[0])
csm = ce[1]['other_symmetry_measures']['csm_wcs_ctwcc']
mystring += ' - {} ({}): {:.2f} % (csm : {:2f})\n'.format(cg.name, cg.mp_symbol,
100.0*ce[2],
csm)
if test in ['d', 'g'] and strategy.uniquely_determines_coordination_environments:
if thecg.mp_symbol != UNCLEAR_ENVIRONMENT_SYMBOL:
mystring += ' <Continuous symmetry measures> '
mingeoms = se.ce_list[isite][thecg.coordination_number][0].minimum_geometries()
for mingeom in mingeoms:
csm = mingeom[1]['other_symmetry_measures']['csm_wcs_ctwcc']
mystring += '{} : {:.2f} '.format(mingeom[0], csm)
print(mystring)
if test == 'g':
test = input('Enter index of site(s) for which you want to see the grid of parameters : ')
indices = list(map(int, test.split()))
print(indices)
for isite in indices:
se.plot_environments(isite, additional_condition=se.AC.ONLY_ACB)
if no_vis:
test = input('Go to next structure ? ("y" to do so)')
if test == 'y':
break
continue
test = input('View structure with environments ? ("y" for the unit cell or "m" for a supercell or "n") : ')
if test in ['y', 'm']:
if test == 'm':
mydeltas = []
test = input('Enter multiplicity (e.g. 3 2 2) : ')
nns = test.split()
for i0 in range(int(nns[0])):
for i1 in range(int(nns[1])):
for i2 in range(int(nns[2])):
mydeltas.append(np.array([1.0*i0, 1.0*i1, 1.0*i2], np.float))
else:
mydeltas = [np.zeros(3, np.float)]
if firsttime:
vis = StructureVis(show_polyhedron=False, show_unit_cell=True)
vis.show_help = False
firsttime = False
vis.set_structure(se.structure)
strategy.set_structure_environments(se)
for isite, site in enumerate(se.structure):
try:
ces = strategy.get_site_coordination_environments(site)
except NeighborsNotComputedChemenvError:
continue
if len(ces) == 0:
continue
ce = strategy.get_site_coordination_environment(site)
if ce is not None and ce[0] != UNCLEAR_ENVIRONMENT_SYMBOL:
for mydelta in mydeltas:
psite = PeriodicSite(site._species, site._fcoords + mydelta, site._lattice,
properties=site._properties)
vis.add_site(psite)
neighbors = strategy.get_site_neighbors(psite)
draw_cg(vis, psite, neighbors, cg=lgf.cg.get_geometry_from_mp_symbol(ce[0]),
perm=ce[1]['permutation'])
vis.show()
test = input('Go to next structure ? ("y" to do so) : ')
if test == 'y':
break
print('')
from pymatgen.io.pwscf import PWInput
from pymatgen import MPRester
mpr = MPRester()
s = mpr.get_structure_by_material_id('mp-20459')
# s.to('POSCAR', 'POSCAR')
PWInput.from_file()
pwin = PWInput(s,
pseudo={
'Pb': 'Pb.pbe-dn-kjpaw_psl.1.0.0.UPF',
'Ti': 'Ti.pbe-spn-kjpaw_psl.1.0.0.UPF',
'O': 'O.pbe-n-kjpaw_psl.1.0.0.UPF'
},
kpoints_grid=(6, 6, 4),
control={
'outdir': 'output',
'prefix': 'scf'
})
pwin.write_file('PTO-scf.in')
def get_mp_structure(cls, mpid):
m = MPRester()
return m.get_structure_by_material_id(mpid)
params_dict = {}
structures = {}
psps = os.listdir('/home/bcomer3/sparc/ase_sparc/pysparcx/sparc/pseudos/LDA_pseudos/')
available_psps = [a.split('.')[0] for a in psps]
# metals
metals = ['Pt', 'Sc', 'La', 'Na', 'Au', 'Os', 'V', 'Zn', 'Bi', 'Co', 'Pd']
sizes = [(4,2,2), (4,3,2), (3,1,1), (3,2,1), (1,1,7), (2,2,2), (2,1,1),(2,2,1), (1,1,1), (2,1,1), (2,2,1)]
for size, metal in zip(sizes, metals):
struc_l = rest.get_entries(metal, sort_by_e_above_hull=True)
struc_l = struc_l[0]
struct = rest.get_structure_by_material_id(struc_l.entry_id)
if metal == 'Pt':
struct.add_site_property('magmom', [0] * len(struct))
struct *= size
if metal in ['Pt', 'Sc']:
del struct[0]
if metal == 'Hf':
del struct[50]
del struct[35]
del struct[100]
#print(len(struct))
if metal == 'Sc':
print(len(struct))
add_to_dict(struct)
# oxides
class CompositionToStructureFromMP(ConversionFeaturizer):
"""
Featurizer to get a Structure object from Materials Project using the
composition alone. The most stable entry from Materials Project is selected,
or NaN if no entry is found in the Materials Project.
Args:
target_col_id (str or None): The column in which the converted data will
be written. If the column already exists then an error will be
thrown unless `overwrite_data` is set to `True`. If `target_col_id`
begins with an underscore the data will be written to the column:
`"{}_{}".format(col_id, target_col_id[1:])`, where `col_id` is the
column being featurized. If `target_col_id` is set to None then
the data will be written "in place" to the `col_id` column (this
will only work if `overwrite_data=True`).
overwrite_data (bool): Overwrite any data in `target_column` if it
exists.
map_key (str): Materials API key
"""
def __init__(self,
target_col_id='structure',
overwrite_data=False,
mapi_key=None):
super().__init__(target_col_id, overwrite_data)
self.mpr = MPRester(mapi_key)
self.set_n_jobs(1)
def featurize(self, comp):
"""
Get the most stable structure from Materials Project
Args:
comp (`pymatgen.core.composition.Composition`): A composition.
Returns:
(`pymatgen.core.structure.Structure`): A Structure object.
"""
entries = self.mpr.get_data(comp.reduced_formula,
prop="energy_per_atom")
if len(entries) > 0:
most_stable_entry = \
sorted(entries, key=lambda e: e['energy_per_atom'])[0]
s = self.mpr.get_structure_by_material_id(
most_stable_entry["material_id"])
return [s]
return [float("nan")]
def citations(self):
return [
"@article{doi:10.1063/1.4812323, author = {Jain,Anubhav and Ong,"
"Shyue Ping and Hautier,Geoffroy and Chen,Wei and Richards, "
"William Davidson and Dacek,Stephen and Cholia,Shreyas "
"and Gunter,Dan and Skinner,David and Ceder,Gerbrand "
"and Persson,Kristin A. }, title = {Commentary: The Materials "
"Project: A materials genome approach to accelerating materials "
"innovation}, journal = {APL Materials}, volume = {1}, number = "
"{1}, pages = {011002}, year = {2013}, doi = {10.1063/1.4812323}, "
"URL = {https://doi.org/10.1063/1.4812323}, "
"eprint = {https://doi.org/10.1063/1.4812323}}",
"@article{Ong2015, author = {Ong, Shyue Ping and Cholia, "
"Shreyas and Jain, Anubhav and Brafman, Miriam and Gunter, Dan "
"and Ceder, Gerbrand and Persson, Kristin a.}, doi = "
"{10.1016/j.commatsci.2014.10.037}, issn = {09270256}, "
"journal = {Computational Materials Science}, month = {feb}, "
"pages = {209--215}, publisher = {Elsevier B.V.}, title = "
"{{The Materials Application Programming Interface (API): A simple, "
"flexible and efficient API for materials data based on "
"REpresentational State Transfer (REST) principles}}, "
"url = {http://linkinghub.elsevier.com/retrieve/pii/S0927025614007113}, "
"volume = {97}, year = {2015} } "
]
def implementors(self):
return ["Anubhav Jain"]
from pymatgen import MPRester, Structure
import sys
mpr = MPRester('DhmFQPuibZo8JtXn')
s: Structure = mpr.get_structure_by_material_id(sys.argv[1])
s.to('POSCAR', 'POSCAR')
from pymatgen import MPRester, Structure
from pymatgen.command_line.gulp_caller import GulpCaller, GulpIO
from matplotlib import pyplot
mpr = MPRester()
mono: Structure = mpr.get_structure_by_material_id('mp-352')
ortho: Structure = mpr.get_structure_by_material_id('mp-685097')
tetra: Structure = mpr.get_structure_by_material_id('mp-1018721')
gio = GulpIO()
gc = GulpCaller()
for i in [mono, ortho, tetra]:
arr = []
arr2 = []
i.apply_strain(-.05)
for j in range(15):
gin = gio.buckingham_input(i, ['conv'], library='morse.lib', alib=True)
gout = gc.run(gin)
arr.append(i.density)
arr2.append(gio.get_energy(gout))
i.apply_strain(-.01)
i.to('POSCAR', i.get_space_group_info()[0].replace('/', '_') + '.vasp')
if i.get_space_group_info()[0] == 'P4_2/nmc':
pyplot.plot(arr, [r * 2 for r in arr2],
label=i.get_space_group_info()[0])
else:
pyplot.plot(arr, arr2, label=i.get_space_group_info()[0])
pyplot.legend()
pyplot.show()
class CompositionToStructureFromMP(ConversionFeaturizer):
"""
Featurizer to get a Structure object from Materials Project using the
composition alone. The most stable entry from Materials Project is selected,
or NaN if no entry is found in the Materials Project.
Args:
target_col_id (str or None): The column in which the converted data will
be written. If the column already exists then an error will be
thrown unless `overwrite_data` is set to `True`. If `target_col_id`
begins with an underscore the data will be written to the column:
`"{}_{}".format(col_id, target_col_id[1:])`, where `col_id` is the
column being featurized. If `target_col_id` is set to None then
the data will be written "in place" to the `col_id` column (this
will only work if `overwrite_data=True`).
overwrite_data (bool): Overwrite any data in `target_column` if it
exists.
map_key (str): Materials API key
"""
def __init__(self, target_col_id='structure', overwrite_data=False,
mapi_key=None):
super().__init__(target_col_id, overwrite_data)
self.mpr = MPRester(mapi_key)
def featurize(self, comp):
"""
Get the most stable structure from Materials Project
Args:
comp (`pymatgen.core.composition.Composition`): A composition.
Returns:
(`pymatgen.core.structure.Structure`): A Structure object.
"""
entries = self.mpr.get_data(comp.reduced_formula, prop="energy_per_atom")
if len(entries) > 0:
most_stable_entry = \
sorted(entries, key=lambda e: e['energy_per_atom'])[0]
s = self.mpr.get_structure_by_material_id(
most_stable_entry["material_id"])
return[s]
return [float("nan")]
def citations(self):
return [
"@article{doi:10.1063/1.4812323, author = {Jain,Anubhav and Ong,"
"Shyue Ping and Hautier,Geoffroy and Chen,Wei and Richards, "
"William Davidson and Dacek,Stephen and Cholia,Shreyas "
"and Gunter,Dan and Skinner,David and Ceder,Gerbrand "
"and Persson,Kristin A. }, title = {Commentary: The Materials "
"Project: A materials genome approach to accelerating materials "
"innovation}, journal = {APL Materials}, volume = {1}, number = "
"{1}, pages = {011002}, year = {2013}, doi = {10.1063/1.4812323}, "
"URL = {https://doi.org/10.1063/1.4812323}, "
"eprint = {https://doi.org/10.1063/1.4812323}}",
"@article{Ong2015, author = {Ong, Shyue Ping and Cholia, "
"Shreyas and Jain, Anubhav and Brafman, Miriam and Gunter, Dan "
"and Ceder, Gerbrand and Persson, Kristin a.}, doi = "
"{10.1016/j.commatsci.2014.10.037}, issn = {09270256}, "
"journal = {Computational Materials Science}, month = {feb}, "
"pages = {209--215}, publisher = {Elsevier B.V.}, title = "
"{{The Materials Application Programming Interface (API): A simple, "
"flexible and efficient API for materials data based on "
"REpresentational State Transfer (REST) principles}}, "
"url = {http://linkinghub.elsevier.com/retrieve/pii/S0927025614007113}, "
"volume = {97}, year = {2015} } "]
def implementors(self):
return ["Anubhav Jain"]
def compute_environments(chemenv_configuration):
string_sources = {
'cif': {
'string': 'a Cif file',
'regexp': '.*\.cif$'
},
'mp': {
'string': 'the Materials Project database',
'regexp': 'mp-[0-9]+$'
}
}
questions = {'c': 'cif'}
if chemenv_configuration.has_materials_project_access:
questions['m'] = 'mp'
lgf = LocalGeometryFinder()
lgf.setup_parameters()
allcg = AllCoordinationGeometries()
strategy_class = strategies_class_lookup[
chemenv_configuration.package_options['default_strategy']['strategy']]
#TODO: Add the possibility to change the parameters and save them in the chemenv_configuration
default_strategy = strategy_class()
default_strategy.setup_options(
chemenv_configuration.package_options['default_strategy']
['strategy_options'])
max_dist_factor = chemenv_configuration.package_options[
'default_max_distance_factor']
firsttime = True
while True:
if len(questions) > 1:
found = False
print(
'Enter the source from which the structure is coming or <q> to quit :'
)
for key_character, qq in questions.items():
print(' - <{}> for a structure from {}'.format(
key_character, string_sources[qq]['string']))
test = input(' ... ')
if test == 'q':
break
if test not in list(questions.keys()):
for key_character, qq in questions.items():
if re.match(string_sources[qq]['regexp'],
str(test)) is not None:
found = True
source_type = qq
if not found:
print('Wrong key, try again ...')
continue
else:
source_type = questions[test]
else:
found = False
source_type = list(questions.values())[0]
if found and len(questions) > 1:
input_source = test
if source_type == 'cif':
if not found:
input_source = input('Enter path to cif file : ')
cp = CifParser(input_source)
structure = cp.get_structures()[0]
elif source_type == 'mp':
if not found:
input_source = input(
'Enter materials project id (e.g. "mp-1902") : ')
a = MPRester(chemenv_configuration.materials_project_api_key)
structure = a.get_structure_by_material_id(input_source)
lgf.setup_structure(structure)
print('Computing environments for {} ... '.format(
structure.composition.reduced_formula))
se = lgf.compute_structure_environments(
maximum_distance_factor=max_dist_factor)
print('Computing environments finished')
while True:
test = input(
'See list of environments determined for each (unequivalent) site ? '
'("y" or "n", "d" with details, "g" to see the grid) : ')
strategy = default_strategy
if test in ['y', 'd', 'g']:
strategy.set_structure_environments(se)
for eqslist in se.equivalent_sites:
site = eqslist[0]
isite = se.structure.index(site)
try:
if strategy.uniquely_determines_coordination_environments:
ces = strategy.get_site_coordination_environments(
site)
else:
ces = strategy.get_site_coordination_environments_fractions(
site)
except NeighborsNotComputedChemenvError:
continue
if ces is None:
continue
if len(ces) == 0:
continue
comp = site.species_and_occu
#ce = strategy.get_site_coordination_environment(site)
if strategy.uniquely_determines_coordination_environments:
ce = ces[0]
if ce is None:
continue
thecg = allcg.get_geometry_from_mp_symbol(ce[0])
mystring = 'Environment for site #{} {} ({}) : {} ({})\n'.format(
str(isite),
comp.get_reduced_formula_and_factor()[0],
str(comp), thecg.name, ce[0])
else:
mystring = 'Environments for site #{} {} ({}) : \n'.format(
str(isite),
comp.get_reduced_formula_and_factor()[0],
str(comp))
for ce in ces:
cg = allcg.get_geometry_from_mp_symbol(ce[0])
csm = ce[1]['other_symmetry_measures'][
'csm_wcs_ctwcc']
mystring += ' - {} ({}): {:.2f} % (csm : {:2f})\n'.format(
cg.name, cg.mp_symbol, 100.0 * ce[2], csm)
if test in [
'd', 'g'
] and strategy.uniquely_determines_coordination_environments:
if thecg.mp_symbol != UNCLEAR_ENVIRONMENT_SYMBOL:
mystring += ' <Continuous symmetry measures> '
mingeoms = se.ce_list[isite][
thecg.
coordination_number][0].minimum_geometries()
for mingeom in mingeoms:
csm = mingeom[1]['other_symmetry_measures'][
'csm_wcs_ctwcc']
mystring += '{} : {:.2f} '.format(
mingeom[0], csm)
print(mystring)
if test == 'g':
test = input(
'Enter index of site(s) for which you want to see the grid of parameters : '
)
indices = list(map(int, test.split()))
print(indices)
for isite in indices:
se.plot_environments(isite,
additional_condition=se.AC.ONLY_ACB)
if no_vis:
test = input('Go to next structure ? ("y" to do so)')
if test == 'y':
break
continue
test = input(
'View structure with environments ? ("y" for the unit cell or "m" for a supercell or "n") : '
)
if test in ['y', 'm']:
if test == 'm':
mydeltas = []
test = input('Enter multiplicity (e.g. 3 2 2) : ')
nns = test.split()
for i0 in range(int(nns[0])):
for i1 in range(int(nns[1])):
for i2 in range(int(nns[2])):
mydeltas.append(
np.array([1.0 * i0, 1.0 * i1, 1.0 * i2],
np.float))
else:
mydeltas = [np.zeros(3, np.float)]
if firsttime:
vis = StructureVis(show_polyhedron=False,
show_unit_cell=True)
vis.show_help = False
firsttime = False
vis.set_structure(se.structure)
strategy.set_structure_environments(se)
for isite, site in enumerate(se.structure):
try:
ces = strategy.get_site_coordination_environments(site)
except NeighborsNotComputedChemenvError:
continue
if len(ces) == 0:
continue
ce = strategy.get_site_coordination_environment(site)
if ce is not None and ce[0] != UNCLEAR_ENVIRONMENT_SYMBOL:
for mydelta in mydeltas:
psite = PeriodicSite(site._species,
site._fcoords + mydelta,
site._lattice,
properties=site._properties)
vis.add_site(psite)
neighbors = strategy.get_site_neighbors(psite)
draw_cg(vis,
psite,
neighbors,
cg=lgf.allcg.get_geometry_from_mp_symbol(
ce[0]),
perm=ce[1]['permutation'])
vis.show()
test = input('Go to next structure ? ("y" to do so) : ')
if test == 'y':
break
print('')
def submit_tests(names=None, params=None):
sma = SubmissionMongoAdapter.auto_load()
# note: TiO2 is duplicated twice purposely, duplicate check should catch this
compounds = {"Si": 149, "Al": 134, "ZnO": 2133, "FeO": 18905,
"LiCoO2": 601860, "LiFePO4": 585433, "GaAs": 2534, "Ge": 32, "PbTe": 19717,
"YbO": 1216, "SiC": 567551, "Fe3C": 510623, "SiO2": 547211, "Na2O": 2352,
"InSb (unstable)": 10148, "Sb2O5": 1705, "N2O5": 554368, "BaTiO3": 5020,
"Rb2O": 1394, "TiO2": 554278, "TiO2 (2)": 554278, 'BaNbTePO8': 560794,
"AgCl": 22922, "AgCl (2)": 570858, "SiO2 (2)": 555211, "Mg2SiO4": 2895, "CO2": 20066,
"PbSO4": 22298, "SrTiO3": 5532, "FeAl": 2658, "AlFeCo2": 10884, "NaCoO2": 554427,
"ReO3": 547271, "LaH2": 24153, "SiH3I": 28538, "LiBH4": 30209, "H8S5N2": 28143,
"LiOH": 23856, "SrO2": 2697, "Mn": 35, "Hg4Pt": 2312,
"PdF4": 13868, "Gd2WO6": 651333, 'MnO2': 19395, 'VO2': 504800}
mpr = MPRester()
for name, sid in compounds.iteritems():
if not names or name in names:
sid = mpr.get_materials_id_from_task_id("mp-{}".format(sid))
s = mpr.get_structure_by_material_id(sid, final=False)
snl = StructureNL(s, 'Anubhav Jain <*****@*****.**>')
parameters = {'priority': 10} if name == 'Si' else None
if params:
parameters.update(params)
sma.submit_snl(snl, '*****@*****.**', parameters=parameters)