def get_chemonly(string=''):
"""
Get only mean chemical descriptors for a chemical formula, say Al2O3
Args:
string: chemical formula, say Al2O3, NaCl etc.
Returns:
mean_chem: average chemical descriptors
"""
comp=Composition(string)
el_dict=comp.get_el_amt_dict()
arr=[]
for k,v in el_dict.items():
des=get_descrp_arr(k)
arr.append(des)
mean_chem=np.mean(arr,axis=0)
return mean_chem
def get_comp_data(un_data):
element_universe = [str(e) for e in Element]
dict_element = {}
for i, j in enumerate(element_universe):
dict_element[str(j)] = i
stoich_array = np.zeros((len(un_data), 3), dtype=float)
at_num_array = np.zeros((len(un_data), 3), dtype=int)
electroneg_array = np.zeros((len(un_data), 3), dtype=float)
comp_array=[]
for index, entry in enumerate(un_data):
comp = Composition(entry[2])
comp_array.append(comp.formula)
temp_dict = dict(comp.get_el_amt_dict())
for count, key in enumerate(temp_dict.keys()):
stoich_array[index][count] = temp_dict[key]
if key not in ['D', 'T']:
at_num_array[index][count] = Element(key).Z
electroneg_array[index][count] = Element(key).X
else:
at_num_array[index][count] = Element('H').Z
electroneg_array[index][count] = Element('H').X
return (comp_array,stoich_array,at_num_array,electroneg_array)
if __name__ == '__main__':
#chemo-structural features
s = Structure.from_file('POSCAR')
x = get_comp_descp(s)
print(len(x))
#only chemical features for a structure
y = get_comp_descp(struct=s,
jcell=False,
jmean_chem=True,
jmean_chg=False,
jrdf=False,
jrdf_adf=False,
print_names=False)
print(len(y))
#chemical features based on composition only
chm = get_chemonly('Al2O3')
print(len(chm))
#charge descriptors
comp = Composition('Al2O3')
el_dict = comp.get_el_amt_dict()
chgarr = []
for k, v in el_dict.items():
chg = get_chgdescrp_arr(k)
chgarr.append(chg)
mean_chg = np.mean(chgarr, axis=0)
print(len(mean_chg))
