def write_features(self):
output_file = open(self.feature_file_path, 'w')
output_file_np_visualization = open("../../np_visual.txt", 'w')
self.question_prop = entity_properties()
for i in range(self.data_start_index, self.data_end_index):
if i == 17 or i == 30 or i == 33 or i == 70 or i == 104 or i == 106 or i == 132:
continue
self.read_data(i)
for np1 in self.question_prop.question_strings_np:
output_file_np_visualization.write(np1 + '\n')
class_finder_obj = class_finder('np_relevant',
self.question_prop)
output_line = class_finder_obj.list_features_toString(np1, np1)
output_file.write(output_line)
output_file.write(" ")
start_index = 0
basic_features = basic_faeture_finder(self.question_prop)
output_line, start_index = basic_features.list_features_toString(
start_index, np1)
output_file.write(output_line)
output_file.write(' ')
useful_np_features = useful_np_feature_finder(
self.question_prop)
output_line, start_index = useful_np_features.list_features_toString(
start_index, np1)
output_file.write(output_line)
output_file.write('\n')
self.question_prop.flush()
def write_features(self):
output_file = open(self.feature_file_path, 'w')
print "hhhhilkjadlksjdaops;jd;alsdj;lajsdo;asd;akls;ldkasd"
self.question_prop = entity_properties()
for i in range(self.data_start_index, self.data_end_index):
print "i is:"
print i
if i == 17 or i == 30 or i == 33 or i == 70:
continue
self.read_data(i)
for np1 in self.question_prop.question_strings_np:
for np2 in self.question_prop.question_strings_np:
if np1 == np2:
continue
start_index = 0
class_finder_obj = class_finder('joint',
self.question_prop)
output_line = class_finder_obj.list_features_toString(
np1, np2)
output_file.write(output_line)
output_file.write(" ")
basic_features = basic_faeture_finder(self.question_prop)
output_line, start_index = basic_features.list_features_toString(
start_index, np1)
output_file.write(output_line)
output_file.write(' ')
output_line, start_index = basic_features.list_features_toString(
start_index, np2)
output_file.write(output_line)
if self.included_disjoint_selection == True:
output_file.write(' ')
disjoint_features = disjoint_feature_finder(
self.question_prop)
output_line, start_index = disjoint_features.list_features_toString(
start_index, np1, np2)
output_file.write(output_line)
if self.included_subset_selection == True:
output_file.write(' ')
subset_faetures = subset_faeture_finder(
self.question_prop)
output_line, start_index = subset_faetures.list_features_toString(
start_index, np1, np2)
output_file.write(output_line)
output_file.write('\n')
self.question_prop.flush()
def write_features(self):
output_file = open(self.feature_file_path, 'w')
self.question_prop = entity_properties()
self.question_prop.find_word_list(self.file_path_refrence.synonym_list_path, self.file_path_refrence.num_of_dimentions)
for i in range(self.data_start_index, self.data_end_index):
print "i is:"
print i
if i in self.file_path_refrence.problematic_indexes:
continue
self.read_data(i)
for np1 in self.question_prop.question_strings_np:
if np1 in self.question_prop.pronoun_list or len(np1.split(' ')) >= 6:
continue
for np2 in self.question_prop.question_strings_np:
if np2 in self.question_prop.pronoun_list or len(np2.split(' ')) >= 6:
continue
if np1 == np2:
continue
if np1.endswith(np2):
if math_modifiers.is_word_number(np1[:np1.index(np2)-1]):
continue
if np2.endswith(np1):
if math_modifiers.is_word_number(np2[:np2.index(np1)-1]):
continue
start_index = 0
class_finder_obj = class_finder('str_disjoint', self.question_prop)
output_line = class_finder_obj.check_for_label(np1, np2)
output_file.write(output_line)
feature_list = []
basic_features = basic_faeture_finder(self.question_prop)
feature_list = basic_features.appropriate_feature_finder_list(feature_list, np1)
feature_list = basic_features.appropriate_feature_finder_list(feature_list, np2)
useful_np_features = useful_np_feature_finder(self.question_prop)
feature_list = useful_np_features.appropriate_feature_finder_list(feature_list, np1, np2)
disjoint_features = disjoint_feature_finder(self.question_prop)
feature_list = disjoint_features.appropriate_feature_finder_list(feature_list, np1, np2)
for i in range(0, len(feature_list)):
output_file.write(' ' + str(i) + ':' + str(feature_list[i]))
output_file.write('\n')
self.question_prop.flush()
def write_features(self):
output_file = open(self.feature_file_path, 'w')
if self.entity_type_mode == 0:
output_file_np_visualization = open(
self.file_path_refrence.np_visualization_file_type, 'w')
else:
output_file_np_visualization = open(
self.file_path_refrence.np_visualization_file_entity, 'w')
self.question_prop = entity_properties()
for i in range(self.data_start_index, self.data_end_index):
if i in self.file_path_refrence.problematic_indexes:
continue
self.read_data(i)
# print self.question_prop.whole_question
# print self.question_prop.relevant_noun_phrases
for np1 in self.question_prop.question_strings_np:
if np1 in self.question_prop.pronoun_list or len(
np1.split(' ')) >= 6:
continue
if (len(np1.split(' ')) == 1) and (
self.question_prop.find_unit_type(np1) != '000'):
continue
output_file_np_visualization.write(np1 + '\t' + str(i) + '\n')
class_finder_obj = class_finder('np_relevant',
self.question_prop)
output_line = class_finder_obj.check_for_label(np1, np1)
output_file.write(output_line)
feature_list = []
basic_features = basic_faeture_finder(self.question_prop)
feature_list = basic_features.appropriate_feature_finder_list(
feature_list, np1)
# useful_np_features = useful_np_feature_finder(self.question_prop)
# feature_list = useful_np_features.appropriate_feature_finder_list(feature_list, np1)
for i in range(0, len(feature_list)):
output_file.write(' ' + str(i) + ':' +
str(feature_list[i]))
output_file.write('\n')
self.question_prop.flush()
def write_features(self):
output_file = open(self.feature_file_path, 'w')
self.question_prop = entity_properties()
for i in range(self.data_start_index, self.data_end_index):
print "i is:"
print i
if i == 17 or i == 30 or i == 33 or i == 70:
continue
self.read_data(i)
for pair in self.question_prop.relevant_pairs:
for str_sample in self.non_relevant_string:
if pair.startswith(str_sample):
continue
parts = pair.split('\t')
np1 = parts[0]
np2 = parts[1]
if np1.endswith(np2):
if math_modifiers.is_word_number(np1[:np1.index(np2)-1]):
continue
if np2.endswith(np1):
if math_modifiers.is_word_number(np2[:np2.index(np1)-1]):
continue
# writing features starts from here
# first:: checking for the label
class_finder_obj = class_finder('eq_noRel', self.question_prop)
output_line = class_finder_obj.check_for_label(np1, np2)
output_file.write(output_line)
feature_list = []
basic_features = basic_faeture_finder(self.question_prop)
feature_list = basic_features.appropriate_feature_finder_list(feature_list, np1)
feature_list = basic_features.appropriate_feature_finder_list(feature_list, np2)
eq_features = equivalence_feature_finder(self.question_prop)
feature_list = eq_features.appropriate_feature_finder_list(feature_list, np1, np2)
for i in range(0, len(feature_list)):
output_file.write(' ' + str(i) + ':' + str(feature_list[i]))
output_file.write('\n')
self.question_prop.flush()
def write_features(self):
output_file = open(self.feature_file_path, 'w')
self.question_prop = entity_properties()
for i in range(self.data_start_index, self.data_end_index):
if i == 17 or i == 30 or i == 33 or i == 70:
continue
self.read_data(i)
for np1 in self.question_prop.question_strings_np:
start_index = 0
basic_features = basic_faeture_finder(self.question_prop)
output_line, start_index = basic_features.list_features_toString(
start_index, np1)
output_file.write(output_line)
# print output_line
if self.included_disjoint_selection == True:
output_file.write(' ')
disjoin_features = disjoint_feature_finder(
self.question_prop)
output_line, start_index = disjoin_features.list_features_toString(
start_index, np1)
output_file.write(output_line)
if self.included_subset_selection == True:
output_file.write(' ')
subset_faetures = subset_faeture_finder(self.question_prop)
output_line, start_index = subset_faetures.list_features_toString(
start_index, np1)
output_file.write(output_line)
if self.include_useful_np_selection == True:
output_file.write(' ')
useful_np_features = useful_np_feature_finder(
self.question_prop)
output_line, start_index = useful_np_features.list_features_toString(
start_index, np1)
output_file.write(output_line)
output_file.write('\n')
self.question_prop.flush()