def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for JarvisDft Explorer!')
data, data_jarvis = [], []
general_columns = ['mp-id', 'cid', 'formula']
keys, subkeys = ['NUS', 'JARVIS'], ['id', 'Eₓ', 'CIF']
columns = general_columns + ['##'.join([k, sk]) for k in keys for sk in subkeys]
columns_jarvis = general_columns + ['id', 'E', 'ΔE|optB88vdW', 'ΔE|mbj', 'CIF']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
structures = mpfile.sdata.get(mp_id)
cif_urls = {}
for k in keys:
cif_urls[k] = ''
name = '{}_{}'.format(contrib['formula'], k)
if structures.get(name) is not None:
cif_urls[k] = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', name
])
row = [mp_id, cid_url, contrib['formula']]
for k in keys:
for sk in subkeys:
if sk == subkeys[-1]:
row.append(cif_urls[k])
else:
cell = contrib.get(k, {sk: ''})[sk]
row.append(cell)
data.append((mp_id, row))
row_jarvis = [mp_id, cid_url, contrib['formula']]
for k in columns_jarvis[len(general_columns):]:
if k == columns_jarvis[-1]:
row_jarvis.append(cif_urls[keys[1]])
else:
row_jarvis.append(contrib.get(keys[1], {k: ''}).get(k, ''))
if row_jarvis[3]:
data_jarvis.append((mp_id, row_jarvis))
return [
Table.from_items(data, orient='index', columns=columns),
Table.from_items(data_jarvis, orient='index', columns=columns_jarvis)
]
def get_ionic_radii(self):
data = []
columns = ['mp-id', 'cid', 'species', 'charge', u'rᵢₒₙ', 'HS/LS', 'CN']
docs = self.query_contributions(
criteria={'content.title': 'Ionic Radii'},
projection={'_id': 1, 'identifier': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for RedoxThermoCsp Ionic Radii!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials', doc['_id']
])
nrows = sum(1 for v in contrib.values() if isinstance(v, dict))
rows = [[identifier, cid_url] for i in range(nrows)]
for col in columns[2:]:
for irow, row in enumerate(rows):
val = contrib.get(col)
if val is None:
val = contrib[str(irow)].get(col, '-')
row.append(val)
for row in rows:
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_ionic_radii(self):
data = []
columns = ['mp-id', 'cid', 'species', 'charge', u'rᵢₒₙ', 'HS/LS', 'CN']
docs = self.query_contributions(
criteria={'content.title': 'Ionic Radii'},
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for DlrVieten Ionic Radii!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials', doc['_id']
])
nrows = sum(1 for v in contrib.values() if isinstance(v, dict))
rows = [[identifier, cid_url] for i in range(nrows)]
for col in columns[2:]:
for irow, row in enumerate(rows):
val = contrib.get(col)
if val is None:
val = contrib[str(irow)].get(col, '-')
row.append(val)
for row in rows:
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def read_csv(body, is_data_section=True):
"""run pandas.read_csv on (sub)section body"""
if not body: return None
from mpcontribs.io.core.components import Table
if is_data_section:
options = { 'sep': ',', 'header': 0 }
if body.startswith('\nlevel_'):
options.update({'index_col': [0, 1]})
cur_line = 1
while 1:
first_line = body.split('\n', cur_line)[cur_line-1]
cur_line += 1
if first_line and not first_line.strip().startswith(csv_comment_char):
break
ncols = len(first_line.split(options['sep']))
else:
options = { 'sep': ':', 'header': None, 'index_col': 0 }
ncols = 2
converters = dict((col,strip_converter) for col in range(ncols))
return Table(pandas.read_csv(
StringIO(body), comment=csv_comment_char,
skipinitialspace=True, squeeze=True,
converters=converters, encoding='utf8',
**options
).dropna(how='all'))
def get_contributions(self, doping):
dopings = ['n', 'p']
if doping not in dopings:
raise Exception('doping has to be n or p!')
docs = self.query_contributions(projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for Boltztrap Explorer!')
data = []
columns = [
'##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']
]
keys, subkeys = [u'mₑᶜᵒⁿᵈ', u"Seebeck"], [u"e₁", u"e₂", u"e₃"]
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['pretty_formula']]
row += [
contrib[k].get(doping, {}).get(sk, '') for k in keys
for sk in subkeys
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, limit=20):
docs = self.query_contributions(
projection={'_id': 1, 'identifier': 1, 'content': 1}, limit=limit) # use URL for all data
if not docs:
raise Exception('No contributions found for CarrierTransport Explorer!')
data = []
columns = ['##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']]
keys, subkeys = ['<mₑᶜᵒⁿᵈ>', '<S>', '<σ>', '<S²σ>'], ['n', 'p']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['extra_data']['pretty_formula']]
row += [
contrib['data'].get(k[1:-1], {}).get(sk, {}).get('<ε>', 'n.a. mₑ')
for k in keys for sk in subkeys
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = ['mp-id', 'contribution', 'formula', 'CIF', 'dISO', 'etaQ', 'QCC', 'B']
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for Dibbs Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['formula']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
row += [contrib['data'][col] for col in columns[-4:]]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
projection = {'_id': 1, 'identifier': 1, 'content.data': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception(
'No contributions found for RedoxThermoCsp Explorer!')
data, columns = [], ['identifier', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
scope = []
for key, value in contrib.iterate():
level, key = key
level_reduction = bool(level < len(scope))
if level_reduction:
del scope[level:]
if value is None:
scope.append(key)
else:
col = '##'.join(scope + [key]).replace('_', ' ')
if col not in columns:
columns.append(col)
row.append(value)
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
projection = {'_id': 1, 'identifier': 1, 'content.data': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for RedoxThermoCsp Explorer!')
data, columns = [], ['identifier', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
scope = []
for key, value in contrib.iterate():
level, key = key
level_reduction = bool(level < len(scope))
if level_reduction:
del scope[level:]
if value is None:
scope.append(key)
else:
col = '##'.join(scope + [key]).replace('_', ' ')
if col not in columns:
columns.append(col)
row.append(value)
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_concentration_functions(composition_table_dict):
meta = composition_table_dict["meta"]
composition_table = Table.from_dict(composition_table_dict["data"])
elements = [col for col in composition_table.columns if col not in meta]
x = composition_table["X"].values
y = composition_table["Y"].values
cats = composition_table["X"].unique()
concentration, conc, d, y_c, functions = {}, {}, {}, {}, RecursiveDict()
for el in elements:
concentration[el] = to_numeric(composition_table[el].values) / 100.0
conc[el], d[el], y_c[el] = {}, {}, {}
if meta["X"] == "category":
for i in cats:
k = "{:06.2f}".format(float(i))
y_c[el][k] = to_numeric(y[where(x == i)])
conc[el][k] = to_numeric(concentration[el][where(x == i)])
d[el][k] = interp1d(y_c[el][k], conc[el][k])
functions[el] = lambda a, b, el=el: d[el][a](b)
else:
functions[el] = interp2d(float(x), float(y), concentration[el])
return functions
def get_concentration_functions(composition_table_dict):
meta = composition_table_dict['meta']
composition_table = Table.from_dict(composition_table_dict['data'])
elements = [col for col in composition_table.columns if col not in meta]
x = composition_table["X"].values
y = composition_table["Y"].values
cats = composition_table["X"].unique()
concentration, conc, d, y_c, functions = {}, {}, {}, {}, RecursiveDict()
for el in elements:
concentration[el] = to_numeric(composition_table[el].values)/100.
conc[el], d[el], y_c[el] = {}, {}, {}
if meta['X'] == 'category':
for i in cats:
k = '{:06.2f}'.format(float(i))
y_c[el][k] = to_numeric(y[where(x==i)])
conc[el][k] = to_numeric(concentration[el][where(x==i)])
d[el][k] = interp1d(y_c[el][k], conc[el][k])
functions[el] = lambda a, b, el=el: d[el][a](b)
else:
functions[el] = interp2d(float(x), float(y), concentration[el])
return functions
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for SWF Explorer!')
data = []
columns = ['formula', 'contribution']
ncols = 9
for doc in docs:
mpfile = MPFile.from_contribution(doc)
formula = mpfile.ids[0]
contrib = mpfile.hdata[formula].get('data')
if contrib is None:
continue
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [formula, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
n = len(row)
if n < ncols:
row += [''] * (ncols - n)
data.append((formula, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for MpWorkshop2017 Explorer!')
data = []
columns = ['mp-id', 'contribution']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [mp_id, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append([mp_id, row])
# enforce equal row lengths
ncols = len(columns)
for entry in data:
n = len(entry[1])
if n != ncols:
entry[1] += [''] * (ncols - n)
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
projection = {'_id': 1, 'identifier': 1, 'content': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
data = []
columns = ['formula', 'cid']
keys = RecursiveDict([
('composition', ['Co', 'Cu', 'Ce']),
#('position', ['x', 'y']),
('XAS', ['min', 'max']),
('XMCD', ['min', 'max'])
])
columns += ['##'.join([k, sk]) for k, subkeys in keys.items() for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
row += [contrib[k][sk] for k, subkeys in keys.items() for sk in subkeys]
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
projection = {'_id': 1, 'mp_cat_id': 1, 'content': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
data = []
columns = ['formula', 'cid']
keys = RecursiveDict([
('composition', ['Co', 'Cu', 'Ce']),
#('position', ['x', 'y']),
('XAS', ['min', 'max']),
('XMCD', ['min', 'max'])
])
columns += ['##'.join([k, sk]) for k, subkeys in keys.items() for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]['data']
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url]
row += [contrib[k][sk] for k, subkeys in keys.items() for sk in subkeys]
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = [
'mp-id', 'contribution', 'formula', 'CIF', 'dISO', 'etaQ', 'QCC',
'B'
]
docs = self.query_contributions(projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for Dibbs Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['formula']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif',
structures.keys()[0]
])
row.append(cif_url)
row += [contrib['data'][col] for col in columns[-4:]]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
data = []
columns = ['identifier', 'contribution', 'composition', 'CIF']
docs = self.query_contributions(
criteria={'content.title': {'$ne': 'Ionic Radii'}},
projection={'_id': 1, 'mp_cat_id': 1, 'content': 1}
)
if not docs:
raise Exception('No contributions found for DlrVieten Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
contrib = mpfile.hdata[identifier]
cid_url = self.get_cid_url(doc)
row = [identifier, cid_url, contrib['composition']]
cif_url = ''
structures = mpfile.sdata.get(identifier)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
data.append((identifier, row))
return Table.from_items(data, orient='index', columns=columns)
def iterate(self, nested_dict=None):
"""http://stackoverflow.com/questions/10756427/loop-through-all-nested-dictionary-values"""
from mpcontribs.io.core.components import Table
from pymatgen import Structure
d = self if nested_dict is None else nested_dict
if nested_dict is None:
self.level = 0
for key in list(d.keys()):
value = d[key]
if isinstance(value, _Mapping):
if value.get('@class') == 'Structure':
yield key, Structure.from_dict(value)
continue
yield (self.level, key), None
if value.get('@class') == 'Table':
yield key, Table.from_dict(value)
continue
self.level += 1
for inner_key, inner_value in self.iterate(nested_dict=value):
yield inner_key, inner_value
self.level -= 1
elif isinstance(value, list) and isinstance(value[0], dict):
# index (from archieml parser)
table = ''
for row_dct in value:
table = '\n'.join([table, row_dct['value']])
yield '_'.join([mp_level01_titles[1], key]), table
else:
yield (self.level, key), value
def get_contributions(self, limit=20):
docs = self.query_contributions(projection={
'_id': 1,
'identifier': 1,
'content': 1
},
limit=limit) # use URL for all data
if not docs:
raise Exception(
'No contributions found for CarrierTransport Explorer!')
data = []
columns = [
'##'.join(['general', sk]) for sk in ['mp-id', 'cid', 'formula']
]
keys, subkeys = ['<mₑᶜᵒⁿᵈ>', '<S>', '<σ>', '<S²σ>'], ['n', 'p']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['extra_data']['pretty_formula']]
row += [
contrib['data'].get(k[1:-1], {}).get(sk,
{}).get('<ε>', 'n.a. mₑ')
for k in keys for sk in subkeys
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_table(results, letter):
y = 'Δ{}'.format(letter)
df = Table(
RecursiveDict([('δ', results[0]), (y, results[1]),
(y + 'ₑᵣᵣ', results[2])]))
x0, x1 = map(float, df['δ'].iloc[[0, -1]])
pad = 0.15 * (x1 - x0)
mask = (results[3] > x0 - pad) & (results[3] < x1 + pad)
x, fit = results[3][mask], results[4][mask]
df.set_index('δ', inplace=True)
df2 = pd.DataFrame(RecursiveDict([('δ', x), (y + ' Fit', fit)]))
df2.set_index('δ', inplace=True)
cols = ['δ', y, y + 'ₑᵣᵣ', y + ' Fit']
return pd.concat([df, df2],
sort=True).sort_index().reset_index().rename(columns={
'index': 'δ'
}).fillna('')[cols]
def get_table(results, letter):
y = 'Δ{}'.format(letter)
df = Table(RecursiveDict([
('δ', results[0]), (y, results[1]), (y+'ₑᵣᵣ', results[2])
]))
x0, x1 = map(float, df['δ'].iloc[[0,-1]])
pad = 0.15 * (x1 - x0)
mask = (results[3] > x0 - pad) & (results[3] < x1 + pad)
x, fit = results[3][mask], results[4][mask]
df.set_index('δ', inplace=True)
df2 = pd.DataFrame(RecursiveDict([
('δ', x), (y+' Fit', fit)
]))
df2.set_index('δ', inplace=True)
cols = ['δ', y, y+'ₑᵣᵣ', y+' Fit']
return pd.concat([df, df2], sort=True).sort_index().reset_index().rename(
columns={'index': 'δ'}).fillna('')[cols]
def get_table(results, letter):
y = "Δ{}".format(letter)
df = Table(
RecursiveDict([("δ", results[0]), (y, results[1]),
(y + "ₑᵣᵣ", results[2])]))
x0, x1 = map(float, df["δ"].iloc[[0, -1]])
pad = 0.15 * (x1 - x0)
mask = (results[3] > x0 - pad) & (results[3] < x1 + pad)
x, fit = results[3][mask], results[4][mask]
df.set_index("δ", inplace=True)
df2 = pd.DataFrame(RecursiveDict([("δ", x), (y + " Fit", fit)]))
df2.set_index("δ", inplace=True)
cols = ["δ", y, y + "ₑᵣᵣ", y + " Fit"]
return (pd.concat([df, df2],
sort=True).sort_index().reset_index().rename(columns={
"index": "δ"
}).fillna("")[cols])
def index(request, cid=None, db_type=None, mdb=None):
try:
response = None
if request.method == 'GET':
axes, dopings = ['<S>', '<σ>', '<S²σ>'], ['n', 'p']
projection = dict(
('content.data.{}'.format(k[1:-1]), 1) for k in axes)
projection.update({'identifier': 1})
docs = mdb.contrib_ad.query_contributions(
{'project': 'carrier_transport'}, projection=projection)
response = {'text': []}
response.update(dict((k, []) for k in axes))
for doc in docs:
d = doc['content']['data']
for doping in dopings:
for idx, k in enumerate(axes):
kk = k[1:-1]
if kk in d and doping in d[kk]:
value = d[kk][doping]['<ε>']
value = float(value.split()[0])
if idx == 2:
value = math.log10(value)
response['text'].append(doc['identifier'])
response[k].append(value)
elif request.method == 'POST':
name = json.loads(request.body)['name']
names = name.split('##')
key, subkey = names[0][1:-1], names[1][0]
table_name = '{}({})'.format(key, subkey)
doc = mdb.contrib_ad.query_contributions(
{'_id': cid},
projection={
'_id': 0,
'content.{}'.format(table_name): 1,
'content.data.{}.{}'.format(key, subkey): 1
})[0]
table = doc['content'].get(table_name)
if table:
table = Table.from_dict(table)
x = [col.split()[0] for col in table.columns[1:]]
y = list(table[table.columns[0]])
z = table[table.columns[1:]].values.tolist()
if not table_name.startswith('S'):
z = [[math.log10(float(c)) for c in r] for r in z]
title = ' '.join([table_name, names[1].split()[-1]])
response = {
'x': x,
'y': y,
'z': z,
'type': 'heatmap',
'colorbar': {
'title': title
}
}
except Exception as ex:
raise ValueError('"REST Error: "{}"'.format(str(ex)))
return {"valid_response": True, 'response': response}
def index(request, cid=None, db_type=None, mdb=None):
try:
response = None
if request.method == 'GET':
axes, dopings = ['<S>', '<σ>', '<S²σ>'], ['n', 'p']
projection = dict(('content.data.{}'.format(k[1:-1]), 1) for k in axes)
projection.update({'identifier': 1})
docs = mdb.contrib_ad.query_contributions(
{'project': 'carrier_transport'}, projection=projection
)
response = {'text': []}
response.update(dict((k, []) for k in axes))
for doc in docs:
d = doc['content']['data']
for doping in dopings:
for idx, k in enumerate(axes):
kk = k[1:-1]
if kk in d and doping in d[kk]:
value = d[kk][doping]['<ε>']
value = float(value.split()[0])
if idx == 2:
value = math.log10(value)
response['text'].append(doc['identifier'])
response[k].append(value)
elif request.method == 'POST':
name = json.loads(request.body)['name']
names = name.split('##')
key, subkey = names[0][1:-1], names[1][0]
table_name = '{}({})'.format(key, subkey)
doc = mdb.contrib_ad.query_contributions(
{'_id': cid}, projection={
'_id': 0, 'content.{}'.format(table_name): 1,
'content.data.{}.{}'.format(key, subkey): 1
}
)[0]
table = doc['content'].get(table_name)
if table:
table = Table.from_dict(table)
x = [col.split()[0] for col in table.columns[1:]]
y = list(table[table.columns[0]])
z = table[table.columns[1:]].values.tolist()
if not table_name.startswith('S'):
z = [[math.log10(float(c)) for c in r] for r in z]
title = ' '.join([table_name, names[1].split()[-1]])
response = {'x': x, 'y': y, 'z': z, 'type': 'heatmap', 'colorbar': {'title': title}}
except Exception as ex:
raise ValueError('"REST Error: "{}"'.format(str(ex)))
return {"valid_response": True, 'response': response}
def get_contributions(self, typ):
types = ['2d', '3d']
if typ not in types:
raise Exception('typ has to be 2d or 3d!')
docs = self.query_contributions(
criteria={'content.data.{}'.format(typ): {
'$exists': 1
}},
projection={
'_id': 1,
'mp_cat_id': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for JarvisDft Explorer!')
data = []
columns = [
'mp-id', 'cid', 'CIF', 'final_energy', 'optB88vDW_bandgap',
'mbj_bandgap', 'bulk_modulus', 'shear_modulus', 'jid'
]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
hdata = mpfile.hdata[mp_id]
contrib = hdata['data'][typ]
cid_url = self.get_cid_url(doc)
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif',
structures.keys()[0]
])
row = [mp_id, cid_url, cif_url
] + [contrib[k] for k in columns[3:-1]]
row.append(hdata['details_url'].format(contrib['jid']))
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, bandgap_range=None):
projection = {
'_id': 1,
'identifier': 1,
'content.formula': 1,
'content.ICSD': 1,
'content.data': 1
}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
data = []
columns = ['mp-id', 'cid', 'formula', 'ICSD', 'C']
keys, subkeys = ['ΔE-KS', 'ΔE-QP'], ['indirect', 'direct']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [
mp_id, cid_url, contrib['formula'], contrib['ICSD'],
contrib['data']['C']
]
row += [contrib['data'][k][sk] for k in keys for sk in subkeys]
if bandgap_range:
in_filter = True
for k, v in bandgap_range.iteritems():
ks = k.split('_')
val = contrib['data'][ks[0]][
ks[1]] if len(ks) > 1 else contrib['data'][k]
dec = float(val.split()[0])
if dec < v[0] or dec > v[1]:
in_filter = False
break
if in_filter:
data.append((mp_id, row))
else:
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1, 'content.abbreviations': 1}
)
if not docs:
raise Exception('No contributions found for PerovskitesDiffusion Explorer!')
data, columns = [], None
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
if columns is None:
columns = ['mp-id', 'contribution'] + contrib.keys()
row = [mp_id, cid_url] + contrib.values()
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_all_spectra(self, typ):
types = ['XAS', 'XMCD']
if typ not in types:
raise Exception('{} not in {}'.format(typ, types))
projection = {'_id': 1, 'identifier': 1, 'content.Co': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for ALS Beamline Explorer!')
table = Table()
for doc in docs:
mpfile = MPFile.from_contribution(doc)
identifier = mpfile.ids[0]
df = mpfile.tdata[identifier]['Co']
if 'Energy' not in table.columns:
table['Energy'] = df['Energy']
table[identifier] = df[typ]
return Plot({'x': 'Energy', 'table': typ, 'showlegend': False}, table)
def get_contributions(self, phase=None):
data = []
phase_query_key = {'$exists': 1} if phase is None else phase
columns = ['mp-id', 'contribution', 'formula']
if phase is None:
columns.append('phase')
columns += ['dH (formation)', 'dH (hydration)', 'GS?', 'CIF']
docs = self.query_contributions(
criteria={
'content.doi': '10.1021/jacs.6b11301',
'content.data.Phase': phase_query_key
},
projection={
'_id': 1, 'mp_cat_id': 1, 'content.data': 1,
'content.{}'.format(mp_level01_titles[3]): 1
}
)
if not docs:
raise Exception('No contributions found for MnO2 Phase Selection Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['Formula']]
if phase is None:
row.append(contrib['Phase'])
row += [contrib['dHf'], contrib['dHh'], contrib['GS']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(
projection={'_id': 1, 'mp_cat_id': 1, 'content.data': 1}
)
if not docs:
raise Exception('No contributions found for PCFC Explorer!')
data = []
columns = ['mp-id', 'cid']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mpid = mpfile.ids[0]
contrib = mpfile.hdata[mpid]['data']
cid_url = self.get_cid_url(doc)
row = [mpid, cid_url]
if len(columns) == 2:
columns += [k for k in contrib.keys()]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append((mpid, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, phase=None):
data = []
phase_query_key = {'$exists': 1} if phase is None else phase
columns = ['mp-id', 'contribution', 'formula']
if phase is None:
columns.append('phase')
columns += ['ΔH', 'ΔHₕ', 'GS?', 'CIF']
docs = self.query_contributions(
criteria={'content.data.Phase': phase_query_key},
projection={
'_id': 1, 'mp_cat_id': 1, 'content.data': 1,
'content.{}'.format(mp_level01_titles[3]): 1
}
)
if not docs:
raise Exception('No contributions found for MnO2 Phase Selection Explorer!')
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['Formula'].replace(' ', '')]
if phase is None:
row.append(contrib['Phase'])
row += [contrib['ΔH'], contrib['ΔHₕ'], contrib['GS']]
cif_url = ''
structures = mpfile.sdata.get(mp_id)
if structures:
cif_url = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer', 'materials',
doc['_id'], 'cif', structures.keys()[0]
])
row.append(cif_url)
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self, bandgap_range=None):
projection = {
'_id': 1, 'mp_cat_id': 1,
'content.formula': 1, 'content.ICSD': 1, 'content.data': 1
}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
data = []
columns = ['mp-id', 'cid', 'formula', 'ICSD', 'C']
keys, subkeys = ['ΔE-KS', 'ΔE-QP'], ['indirect', 'direct']
columns += ['##'.join([k, sk]) for k in keys for sk in subkeys]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]
cid_url = self.get_cid_url(doc)
row = [mp_id, cid_url, contrib['formula'], contrib['ICSD'], contrib['data']['C']]
row += [contrib['data'][k][sk] for k in keys for sk in subkeys]
if bandgap_range:
in_filter = True
for k, v in bandgap_range.iteritems():
ks = k.split('_')
val = contrib['data'][ks[0]][ks[1]] if len(ks) > 1 else contrib['data'][k]
dec = float(val.split()[0])
if dec < v[0] or dec > v[1]:
in_filter = False
break
if in_filter:
data.append((mp_id, row))
else:
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(projection={
'_id': 1,
'identifier': 1,
'content.data': 1
})
if not docs:
raise Exception('No contributions found for SWF Explorer!')
data = []
columns = ['formula', 'contribution']
ncols = 9
for doc in docs:
mpfile = MPFile.from_contribution(doc)
formula = mpfile.ids[0]
contrib = mpfile.hdata[formula].get('data')
if contrib is None:
continue
cid_url = self.get_cid_url(doc)
for k in contrib.keys():
if k not in columns:
columns.append(k)
row = [formula, cid_url]
for col in columns[2:]:
row.append(contrib.get(col, ''))
n = len(row)
if n < ncols:
row += [''] * (ncols - n)
data.append((formula, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
projection = {'_id': 1, 'mp_cat_id': 1, 'content': 1}
docs = self.query_contributions(projection=projection)
if not docs:
raise Exception('No contributions found for DTU Explorer!')
data = []
columns = [
'mp-id', 'cid', 'indirect ΔE-KS', 'direct ΔE-KS', 'C',
'indirect ΔE-QP', 'direct ΔE-QP'
]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
row = [
mp_id, cid_url, contrib['ΔE-KS']['indirect'],
contrib['ΔE-KS']['direct'], contrib['C'],
contrib['ΔE-QP']['indirect'], contrib['ΔE-QP']['direct']
]
data.append((mp_id, row))
return Table.from_items(data, orient='index', columns=columns)
def get_contributions(self):
docs = self.query_contributions(projection={
'_id': 1,
'identifier': 1,
'content.data': 1
})
if not docs:
raise Exception('No contributions found for PCFC Explorer!')
data = []
columns = ['mp-id', 'cid']
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mpid = mpfile.ids[0]
contrib = mpfile.hdata[mpid]['data']
cid_url = self.get_cid_url(doc)
row = [mpid, cid_url]
if len(columns) == 2:
columns += [k for k in contrib.keys()]
for col in columns[2:]:
row.append(contrib.get(col, ''))
data.append((mpid, row))
return Table.from_items(data, orient='index', columns=columns)
def run(mpfile, **kwargs):
# extract data from json files
input_dir = mpfile.hdata.general['input_dir']
for idx, obj in enumerate(scandir(input_dir)):
mpid = obj.name.split('.', 1)[0].rsplit('_', 1)[1]
print(mpid)
input_file = gzip.open(obj.path, 'rb')
try:
data = json.loads(input_file.read())
# filter out metals
if 'GGA' not in data or 'GGA' not in data['gap'] or data['gap']['GGA'] < 0.1:
print('GGA gap < 0.1 -> skip')
continue
# add hierarchical data (nested key-values)
hdata = RecursiveDict()
T, lvl, S2 = '300', '1e+18', None
pf_key = 'S²σ'
hdata['temperature'] = T + ' K'
hdata['doping_level'] = lvl + ' cm⁻³'
variables = [
{'key': 'cond_eff_mass', 'name': 'mₑᶜᵒⁿᵈ', 'unit': 'mₑ'},
{'key': 'seebeck_doping', 'name': 'S', 'unit': 'μV/K'},
{'key': 'cond_doping', 'name': 'σ', 'unit': '(Ωms)⁻¹'},
]
eigs_keys = ['ε₁', 'ε₂', 'ε₃', '<ε>']
for v in variables:
hdata[v['name']] = RecursiveDict()
for doping_type in ['p', 'n']:
if doping_type in data['GGA'][v['key']]:
d = data['GGA'][v['key']][doping_type][T][lvl]
eigs = map(float, d if isinstance(d, list) else d['eigs'])
hdata[v['name']][doping_type] = RecursiveDict(
(eigs_keys[neig], clean_value(eig, v['unit']))
for neig, eig in enumerate(eigs)
)
hdata[v['name']][doping_type][eigs_keys[-1]] = clean_value(np.mean(eigs), v['unit'])
if v['key'] == 'seebeck_doping':
S2 = np.dot(d['tensor'], d['tensor'])
elif v['key'] == 'cond_doping':
pf = np.mean(np.linalg.eigh(np.dot(S2, d['tensor']))[0]) * 1e-8
if pf_key not in hdata:
hdata[pf_key] = RecursiveDict()
hdata[pf_key][doping_type] = {eigs_keys[-1]: clean_value(pf, 'μW/(cmK²s)')}
mpfile_data = nest_dict(hdata, ['data'])
# build data and max values for seebeck, conductivity and kappa
# max/min values computed using numpy. It may be better to code it in pure python.
keys = ['pretty_formula', 'volume']
hdata = RecursiveDict((k, data[k]) for k in keys)
hdata['volume'] = clean_value(hdata['volume'], 'ų')
hdata['bandgap'] = clean_value(data['gap']['GGA'], 'eV')
cols = ['value', 'temperature', 'doping']
tables = RecursiveDict()
props = RecursiveDict()
props['seebeck_doping'] = ['S', 'μV/K']
props['cond_doping'] = ['σ', '(Ωms)⁻¹']
props['kappa_doping'] = ['κₑ', 'W/(mKs)']
for prop_name, (lbl, unit) in props.iteritems():
# TODO install Symbola font if you see squares here (https://fonts2u.com/symbola.font)
# and select it as standard font in your browser (leave other fonts as is, esp. fixed width)
tables[lbl] = RecursiveDict()
hlbl = lbl+'₋' if len(lbl) > 1 else lbl
hlbl += 'ₑₓₜᵣ'
hdata[hlbl] = RecursiveDict()
for doping_type in ['p', 'n']:
prop = data['GGA'][prop_name][doping_type]
prop_averages, dopings, columns = [], None, ['T [K]']
temps = sorted(map(int, prop.keys()))
for temp in temps:
row = [temp]
if dopings is None:
dopings = sorted(map(float, prop[str(temp)].keys()))
for doping in dopings:
doping_str = '%.0e' % doping
if len(columns) <= len(dopings):
columns.append('{} cm⁻³ [{}]'.format(doping_str, unit))
eigs = prop[str(temp)][doping_str]['eigs']
row.append(np.mean(eigs))
prop_averages.append((temp, row))
tables[lbl][doping_type] = Table.from_items(
prop_averages, orient='index', columns=columns
)
arr_prop_avg = np.array([item[1] for item in prop_averages])[:,1:]
max_v = np.max(arr_prop_avg)
if prop_name[0] == 's' and doping_type == 'n':
max_v = np.min(arr_prop_avg)
if prop_name[0] == 'k':
max_v = np.min(arr_prop_avg)
arg_max = np.argwhere(arr_prop_avg==max_v)[0]
vals = [
clean_value(max_v, unit),
clean_value(temps[arg_max[0]], 'K'),
clean_value(dopings[arg_max[1]], 'cm⁻³')
]
hdata[hlbl][doping_type] = RecursiveDict(
(k, v) for k, v in zip(cols, vals)
)
mpfile_data.rec_update(nest_dict(hdata, ['extra_data']))
mpfile.add_hierarchical_data(mpfile_data, identifier=data['mp_id'])
for lbl, dct in tables.iteritems():
for doping_type, table in dct.iteritems():
mpfile.add_data_table(
data['mp_id'], table, name='{}({})'.format(lbl, doping_type)
)
finally:
input_file.close()
def get_contributions(self):
docs = self.query_contributions(projection={
'_id': 1,
'identifier': 1,
'content': 1
})
if not docs:
raise Exception('No contributions found for JarvisDft Explorer!')
data, data_jarvis = [], []
general_columns = ['mp-id', 'cid', 'formula']
keys, subkeys = ['NUS', 'JARVIS'], ['id', 'Eₓ', 'CIF']
columns = general_columns + [
'##'.join([k, sk]) for k in keys for sk in subkeys
]
columns_jarvis = general_columns + [
'id', 'E', 'ΔE|optB88vdW', 'ΔE|mbj', 'CIF'
]
for doc in docs:
mpfile = MPFile.from_contribution(doc)
mp_id = mpfile.ids[0]
contrib = mpfile.hdata[mp_id]['data']
cid_url = self.get_cid_url(doc)
structures = mpfile.sdata.get(mp_id)
cif_urls = {}
for k in keys:
cif_urls[k] = ''
name = '{}_{}'.format(contrib['formula'], k)
if structures.get(name) is not None:
cif_urls[k] = '/'.join([
self.preamble.rsplit('/', 1)[0], 'explorer',
'materials', doc['_id'], 'cif', name
])
row = [mp_id, cid_url, contrib['formula']]
for k in keys:
for sk in subkeys:
if sk == subkeys[-1]:
row.append(cif_urls[k])
else:
cell = contrib.get(k, {sk: ''})[sk]
row.append(cell)
data.append((mp_id, row))
row_jarvis = [mp_id, cid_url, contrib['formula']]
for k in columns_jarvis[len(general_columns):]:
if k == columns_jarvis[-1]:
row_jarvis.append(cif_urls[keys[1]])
else:
row_jarvis.append(contrib.get(keys[1], {k: ''}).get(k, ''))
if row_jarvis[3]:
data_jarvis.append((mp_id, row_jarvis))
return [
Table.from_items(data, orient='index', columns=columns),
Table.from_items(data_jarvis,
orient='index',
columns=columns_jarvis)
]
def run(mpfile, **kwargs):
# extract data from json files
input_dir = mpfile.hdata.general['input_dir']
for idx, obj in enumerate(scandir(input_dir)):
mpid = obj.name.split('.', 1)[0].rsplit('_', 1)[1]
print(mpid)
input_file = gzip.open(obj.path, 'rb')
try:
data = json.loads(input_file.read())
# filter out metals
if 'GGA' not in data or 'GGA' not in data[
'gap'] or data['gap']['GGA'] < 0.1:
print('GGA gap < 0.1 -> skip')
continue
# add hierarchical data (nested key-values)
hdata = RecursiveDict()
T, lvl, S2 = '300', '1e+18', None
pf_key = 'S²σ'
hdata['temperature'] = T + ' K'
hdata['doping_level'] = lvl + ' cm⁻³'
variables = [
{
'key': 'cond_eff_mass',
'name': 'mₑᶜᵒⁿᵈ',
'unit': 'mₑ'
},
{
'key': 'seebeck_doping',
'name': 'S',
'unit': 'μV/K'
},
{
'key': 'cond_doping',
'name': 'σ',
'unit': '(Ωms)⁻¹'
},
]
eigs_keys = ['ε₁', 'ε₂', 'ε₃', '<ε>']
for v in variables:
hdata[v['name']] = RecursiveDict()
for doping_type in ['p', 'n']:
if doping_type in data['GGA'][v['key']]:
d = data['GGA'][v['key']][doping_type][T][lvl]
eigs = map(float,
d if isinstance(d, list) else d['eigs'])
hdata[v['name']][doping_type] = RecursiveDict(
(eigs_keys[neig], clean_value(eig, v['unit']))
for neig, eig in enumerate(eigs))
hdata[v['name']][doping_type][
eigs_keys[-1]] = clean_value(
np.mean(eigs), v['unit'])
if v['key'] == 'seebeck_doping':
S2 = np.dot(d['tensor'], d['tensor'])
elif v['key'] == 'cond_doping':
pf = np.mean(
np.linalg.eigh(np.dot(S2,
d['tensor']))[0]) * 1e-8
if pf_key not in hdata:
hdata[pf_key] = RecursiveDict()
hdata[pf_key][doping_type] = {
eigs_keys[-1]: clean_value(pf, 'μW/(cmK²s)')
}
mpfile_data = nest_dict(hdata, ['data'])
# build data and max values for seebeck, conductivity and kappa
# max/min values computed using numpy. It may be better to code it in pure python.
keys = ['pretty_formula', 'volume']
hdata = RecursiveDict((k, data[k]) for k in keys)
hdata['volume'] = clean_value(hdata['volume'], 'ų')
hdata['bandgap'] = clean_value(data['gap']['GGA'], 'eV')
cols = ['value', 'temperature', 'doping']
tables = RecursiveDict()
props = RecursiveDict()
props['seebeck_doping'] = ['S', 'μV/K']
props['cond_doping'] = ['σ', '(Ωms)⁻¹']
props['kappa_doping'] = ['κₑ', 'W/(mKs)']
for prop_name, (lbl, unit) in props.iteritems():
# TODO install Symbola font if you see squares here (https://fonts2u.com/symbola.font)
# and select it as standard font in your browser (leave other fonts as is, esp. fixed width)
tables[lbl] = RecursiveDict()
hlbl = lbl + '₋' if len(lbl) > 1 else lbl
hlbl += 'ₑₓₜᵣ'
hdata[hlbl] = RecursiveDict()
for doping_type in ['p', 'n']:
prop = data['GGA'][prop_name][doping_type]
prop_averages, dopings, columns = [], None, ['T [K]']
temps = sorted(map(int, prop.keys()))
for temp in temps:
row = [temp]
if dopings is None:
dopings = sorted(map(float,
prop[str(temp)].keys()))
for doping in dopings:
doping_str = '%.0e' % doping
if len(columns) <= len(dopings):
columns.append('{} cm⁻³ [{}]'.format(
doping_str, unit))
eigs = prop[str(temp)][doping_str]['eigs']
row.append(np.mean(eigs))
prop_averages.append((temp, row))
tables[lbl][doping_type] = Table.from_items(
prop_averages, orient='index', columns=columns)
arr_prop_avg = np.array(
[item[1] for item in prop_averages])[:, 1:]
max_v = np.max(arr_prop_avg)
if prop_name[0] == 's' and doping_type == 'n':
max_v = np.min(arr_prop_avg)
if prop_name[0] == 'k':
max_v = np.min(arr_prop_avg)
arg_max = np.argwhere(arr_prop_avg == max_v)[0]
vals = [
clean_value(max_v, unit),
clean_value(temps[arg_max[0]], 'K'),
clean_value(dopings[arg_max[1]], 'cm⁻³')
]
hdata[hlbl][doping_type] = RecursiveDict(
(k, v) for k, v in zip(cols, vals))
mpfile_data.rec_update(nest_dict(hdata, ['extra_data']))
mpfile.add_hierarchical_data(mpfile_data, identifier=data['mp_id'])
for lbl, dct in tables.iteritems():
for doping_type, table in dct.iteritems():
mpfile.add_data_table(data['mp_id'],
table,
name='{}({})'.format(
lbl, doping_type))
finally:
input_file.close()
