KSI/preprocessing3.py at master · NatalieSty/KSI · GitHub

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
import numpy as np
from collections import defaultdict
from sklearn.model_selection import train_test_split
import codecs


##########################################################
wikivoc={}
codewiki=defaultdict(list)

file2=codecs.open("wikipedia_knowledge",'r','utf-8')
line=file2.readline()
count=0
while line:
    if line[0:4]=='XXXd':
        line=line.strip('\n')
        line=line.split()
        for i in line:
            if i[0:2]=='d_':
                codewiki[i].append(count)
                wikivoc[i]=1
        count=count+1
    line=file2.readline()

################### four codes have two wikidocuments, correct them
codewiki['d_072']=[214]
codewiki['d_698']=[125]
codewiki['d_305']=[250]
codewiki['d_386']=[219]

np.save('wikivoc',wikivoc)
##################################################
filec=codecs.open("combined_dataset",'r','utf-8')

line=filec.readline()

feature=[]
label=[]

while line:
    line=line.strip('\n')
    line=line.split()

    if line[0]=='codes:':
        temp=line[1:]
        label.append(temp)
        line=filec.readline()
        line=line.strip('\n')
        line=line.split()
        if  line[0]=='notes:':
            tempf=[]
            line=filec.readline()

            while line!='end!\n':
                line=line.strip('\n')
                line=line.split()
                tempf=tempf+line
                line=filec.readline()
            feature.append(tempf)
    line=filec.readline()


prevoc={}
for i in label:
    for j in i:
        if j not in prevoc:
            prevoc[j]=len(prevoc)

##################################
notevec=np.load('notevec.npy')
wikivec=np.load('wikivec.npy')
label_to_ix = {}
ix_to_label={}


for codes in label:
    for code in codes:
        if code not in label_to_ix:
            label_to_ix[code]=len(label_to_ix)
            ix_to_label[label_to_ix[code]]=code

tempwikivec=[]

for i in range(0,len(ix_to_label)):
    if ix_to_label[i] in wikivoc:
        temp=wikivec[codewiki[ix_to_label[i]][0]]
        tempwikivec.append(temp)
    else:
        tempwikivec.append([0.0]*wikivec.shape[1])
wikivec=np.array(tempwikivec)

####################################

data=[]
for i in range(0,len(feature)):
    data.append((feature[i], notevec[i], label[i]))

data=np.array(data)

label_to_ix = {}
ix_to_label={}

for doc, note, codes in data:
    for code in codes:
        if code not in label_to_ix:
            if code in wikivoc:
                label_to_ix[code]=len(label_to_ix)
                ix_to_label[label_to_ix[code]]=code

np.save('label_to_ix',label_to_ix)
np.save('ix_to_label',ix_to_label)

training_data, test_data = train_test_split(data, test_size=0.2, random_state=42)
training_data, val_data = train_test_split(training_data, test_size=0.125, random_state=42)

np.save('training_data',training_data)
np.save('test_data',test_data)
np.save('val_data',val_data)


word_to_ix = {}
ix_to_word={}
ix_to_word[0]='OUT'


for doc, note, codes in training_data:
    for word in doc:
        if word not in word_to_ix:
            word_to_ix[word] = len(word_to_ix)+1
            ix_to_word[word_to_ix[word]]=word

np.save('word_to_ix',word_to_ix)
np.save('ix_to_word',ix_to_word)

newwikivec=[]
for i in range(0,len(ix_to_label)):
    newwikivec.append(wikivec[prevoc[ix_to_label[i]]])
newwikivec=np.array(newwikivec)
np.save('newwikivec',newwikivec)