def __transform(self, section):
clazzes = adapter.classes(section)
if clazzes:
ret_val = []
for clazz in clazzes:
ret_val.append(len(clazz['fields']))
return ret_val
else:
return [0]
def __transform(self, section):
clazzes = adapter.classes(section)
if clazzes:
ret_val = []
for clazz in clazzes:
if ((len(clazz['methods']) == 0)
and (len(clazz['fields']) == 0)):
ret_val.append(1)
else:
ret_val.append(0)
return ret_val
else:
return [0]
def __transform(self, section):
stat = get_stat_function(self.stat)
clazzes = adapter.classes(section)
if clazzes:
ret_val = []
for clazz in clazzes:
clazz_values = []
for field in clazz['fields']:
clazz_values.append(len(field['name']))
if not clazz_values:
clazz_values = [0] # if a class does not contain any field
ret_val.append(stat(clazz_values))
return ret_val
else:
return [0]
if score == "RMSE":
return make_scorer(mse, greater_is_better=False)
elif score == 'PC':
return make_scorer(pearson, greater_is_better=True)
else:
raise ValueError('Scoring {} is not supported!'.format(score))
# FEATURES = powerset(FEATURES)
FEATURES = [FEATURES]
X, Y = load(corpus_path=os.path.join(BASEPATH, 'data/training'),
labels=DIMENSIONS)
for x in X:
sections = extract_sections(x)
for section in sections:
classes(section)
recognizer = RECOGNIZER[0]
result = {'recognizer_name': recognizer[0]}
number_features = args.nfeatures
output_filename = os.path.join(
BASEPATH, 'result_{}_{}_{}_{}.json'.format(DIMENSIONS[0], SCORE,
recognizer[0], number_features))
# print(output_filename)
with open(output_filename, 'w') as outfile:
outfile.write('Job started')
print '***Number of features: {}'.format(number_features)
def __transform(self, section):
clazzes = adapter.classes(section)
if clazzes:
return len(clazzes)
else:
return 0