/
data_preprocessing.py
187 lines (154 loc) · 5.98 KB
/
data_preprocessing.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
import os
# Data Directories
CURRENT_WORKING_DIRECTORY = os.getcwd()
DATA_DIRECTORY = CURRENT_WORKING_DIRECTORY + "/NER/"
WRITE_DIRECTORY = DATA_DIRECTORY + "/original_data"
DATA_PATH = CURRENT_WORKING_DIRECTORY + "/crf-ma-datasets/"
WORKER_DATA_PATH = DATA_PATH + "worker_data/"
SENTENCE_DATA_PATH = DATA_PATH + "worker_sentence_data/"
EXTRACTED_DATA = DATA_PATH + "/mturk/extracted_data/"
GROUND_TRUTH_DATA = DATA_PATH + "/mturk/ground_truth/"
DREDZE_DATA = DATA_PATH + "/mturk/dredze_format/"
# Functions
def read_conll(filename):
raw = open(filename, 'r').readlines()
all_x = []
point = []
for line in raw:
stripped_line = line.strip().split(' ')
point.append(stripped_line)
if line == '\n':
if len(point[:-1]) > 0:
all_x.append(point[:-1])
point = []
all_x = all_x
return all_x
def array_to_sentence(array):
sentences = ' '.join(array)
return sentences
#Create Data Directories
if not os.path.exists(DATA_DIRECTORY):
os.mkdir(DATA_DIRECTORY)
if not os.path.exists(DATA_PATH):
os.mkdir(DATA_PATH)
if not os.path.exists(SENTENCE_DATA_PATH):
os.mkdir(SENTENCE_DATA_PATH)
if not os.path.exists(WORKER_DATA_PATH):
os.mkdir(WORKER_DATA_PATH)
if not os.path.exists(WRITE_DIRECTORY):
os.mkdir(WRITE_DIRECTORY)
# Get all the files
files = os.listdir(EXTRACTED_DATA)
#Change the directory
os.chdir(EXTRACTED_DATA)
for file in files:
if file != ".DS_Store":
worker_file = EXTRACTED_DATA + file
worker_files = os.listdir(worker_file)
os.chdir(worker_file)
data_file_path = WORKER_DATA_PATH + file + ".txt"
data_file = open(data_file_path, "w")
for txt_file in worker_files:
read_file = open(txt_file, "r", encoding='windows-1252')
for line in read_file:
data_file.write(line)
read_file.close()
data_file.close()
#
files = os.listdir(GROUND_TRUTH_DATA)
combined_file = open(WRITE_DIRECTORY + "/ground_truth.txt", "w")
os.chdir(GROUND_TRUTH_DATA)
print(os.listdir(GROUND_TRUTH_DATA))
for file in files:
read_file = open(file, "r", encoding='windows-1252')
for line in read_file:
combined_file.write(line)
read_file.close()
combined_file.close()
worker_files = os.listdir(WORKER_DATA_PATH)
for file in worker_files:
file_name = WORKER_DATA_PATH + file
worker_answers = read_conll(file_name)
write_file = open(SENTENCE_DATA_PATH + file, "w")
X_test = [[c[0] for c in x] for x in worker_answers]
for i in range(0, len(X_test)):
output = array_to_sentence(X_test[i])
write_file.write(output)
write_file.write("\n")
write_file.close()
#
#
answers_file_path = WRITE_DIRECTORY + "/answers_sentence.txt"
sentence_files = os.listdir(SENTENCE_DATA_PATH)
answers_file_write = open(answers_file_path, "w")
answers_file_write.close()
for file in sentence_files:
file_name = SENTENCE_DATA_PATH + file
worker_sentence_file = open(file_name, "r")
for line in worker_sentence_file:
answers_file_read = open(answers_file_path, "r")
sentence_list = answers_file_read.readlines()
answers_file_read.close()
found = False
for sentence in sentence_list:
if line in sentence:
found = True
if not found:
answers_file_append = open(answers_file_path, "a")
answers_file_append.write(line)
answers_file_append.close()
worker_sentence_file.close()
trainset = read_conll(DREDZE_DATA + "trainset_mturk.dredze.txt")
X_train = [[c[0] for c in x] for x in trainset]
print("Length of train set: " + str(len(X_train)))
answers_file_train = open(WRITE_DIRECTORY + "/answers_file_train.txt", "w")
for i in range(0,len(trainset)):
for j in range(0,len(trainset[i])):
for k in range(0,len(trainset[i][j])):
if trainset[i][j][k] not in ['NO_POS', 'NO_CHUNK', '1']:
answers_file_train.write(trainset[i][j][k])
answers_file_train.write(" ")
answers_file_train.write("\n")
answers_file_train.write("\n")
answers_file_train.close()
testset = read_conll(DREDZE_DATA + "testset_mturk.dredze.txt")
X_test = [[c[0] for c in x] for x in testset]
print("Length of test set: " + str(len(X_test)))
gt_file_train = open(WRITE_DIRECTORY + "/gt_file_test.txt", "w")
for i in range(0,len(testset)):
for j in range(0,len(testset[i])):
for k in range(0,len(testset[i][j])):
if testset[i][j][k] not in ['NO_POS', 'NO_CHUNK', '1']:
gt_file_train.write(testset[i][j][k])
gt_file_train.write(" ")
gt_file_train.write("\n")
gt_file_train.write("\n")
gt_file_train.close()
answers = read_conll(WRITE_DIRECTORY + "/answers_file_train.txt")
X_answers = [[c[0] for c in x] for x in answers]
ground_truth = read_conll(WRITE_DIRECTORY + "/ground_truth.txt")
X_gt = [[c[0] for c in x] for x in ground_truth]
Y_gt = [[c[1] for c in x] for x in ground_truth]
gt_file_train = open(WRITE_DIRECTORY + "/gt_file_train.txt", "w")
for i in range(0, len(X_answers)):
answers_sentence = array_to_sentence(X_answers[i])
for j in range(0,len(X_gt)):
gt_sentence = array_to_sentence(X_gt[j])
found = False
if str(answers_sentence) == str(gt_sentence):
found = True
if found:
if len(X_answers[i]) == len(X_gt[j]):
for a in range(0,len(X_answers[i])):
gt_file_train.write(X_answers[i][a])
gt_file_train.write(" ")
gt_file_train.write(Y_gt[j][a])
gt_file_train.write("\n")
gt_file_train.write("\n")
gt_file_train.close()
answers = read_conll(WRITE_DIRECTORY + "/answers_file_train.txt")
X_answers = [[c[0] for c in x] for x in answers]
train_gt = read_conll(WRITE_DIRECTORY + "/gt_file_train.txt")
X_train_gt = [[c[0] for c in x] for x in X_answers]
print("X_answers length: " + str(len(X_answers)))
print("X_train_gt length: " + str(len(X_train_gt)))