-
Notifications
You must be signed in to change notification settings - Fork 2
/
logistic_link_prediction.py
175 lines (156 loc) · 4.74 KB
/
logistic_link_prediction.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
#coding:utf-8
# from numpy import *
import numpy as np
import sys
from sklearn.cross_validation import train_test_split
# import the LogisticRegression
from sklearn.linear_model import LogisticRegression
train_link_file = ""
test_link_file = ""
result_file = ""
def read_vec(filename,dic):
f = open(filename)
index=0
for cc in f:
ss=cc.strip().split('\t')
if str(index) not in dic:
dic[str(index)] = []
for i in range(100):
dic[str(index)].append(float(ss[i]))
index+=1
return dic
def load_train(filename,entity_vec,relation_vec):
f_i = open(filename,'r')
target=[]
samples = []
for cc in f_i:
ss = cc.strip().split('\t')
target.append(ss[0])
list1=[]
if ss[1] in entity_vec.keys():
list1 += entity_vec[ss[1]]
if ss[2] in entity_vec.keys():
list1 += entity_vec[ss[2]]
if ss[3] in relation_vec.keys():
list1 += relation_vec[ss[3]]
# else:
# print("benlinlin")
samples.append(list1)
# m = len(samples)
# n = len(samples[0])
# smp = np.array([ [samples[x][y] for y in range(n)] for x in range(m)])
# length = len(target)
# tar = np.array(target)
# print("c",len(tar))
# return smp,tar
return samples,target
def LR(samples_train,target_train,samples_test,target_tset):
f=open(result_file,'w')
# ratio=0.1
# while(ratio<=0.9):
# samples,target = load_train()
# # ss
# target = load_target()
# print("ratio=",ratio)
# print("a",len(samples))
# print("b",len(target))
# bound = int(len(samples) * ratio)
X_train = samples_train
X_test = samples_test
y_train = target_train
y_test = target_tset
# m = len(X_test_1)
# n = len(X_test_1[0])
# X_test = np.array([ [X_test_1[x][y] for y in range(n)] for x in range(m)],dtype=np.float)
# y_test = np.array(y_test_1,dtype=np.float)
# X_test = X_test_1
# y_test = y_test_1
# print("X_test",len(X_test))
# print("y_test",len(y_test))
# X_train, X_test, y_train, y_test = train_test_split(samples, target, test_size=0.2, random_state=0)
# print(X_train[0])
#classifier = LogisticRegression()
classifier = LogisticRegression()
classifier.fit(X_train,y_train)
print("fit success")
x = classifier.predict(X_test)
print("predict success")
TP = 0
FP = 0
TN = 0
FN = 0
length = len(x)
num = 0
for i in range(length):
if int(x[i]) == 1 and int(y_test[i]) == 1:
TP += 1
num+=1
elif int(x[i]) == 1 and int(y_test[i]) == -1:
FP += 1
elif int(x[i]) == -1 and int(y_test[i]) == 1:
FN += 1
elif int(x[i]) == -1 and int(y_test[i]) == -1:
TN += 1
num+=1
N = TN + FP
P = TP + FN
Acc = float(TP+TN) / (N+P)
Recall = float(TP) / P
Precision = float(TP) / (TP + FP)
F1 = float(2 * TP) / (2*TP + FN + FP)
print("TP = ",TP)
print("FP = ",FP)
print("FN = ",FN)
print("TN = ",TN)
print("Acc = ",Acc)
print("Recall = ",Recall)
print("Precision = ",Precision)
print("F1 = ",F1)
result=''
result+= "Acc = "+str(Acc)+'\t'+'Recall = '+str(Recall)+'\t'+'Precision = '+str(Precision)+'\t'+'F1 = '+str(F1)+'\n'
f.write(result)
#ratio+=0.1
f.close()
return
if __name__ == '__main__':
if len(sys.argv) == 0:
print("please input parameters:\n")
print("\t-entityVec <entity embedding file>\n")
print("\t-relationVec <relation embedding file>\n")
print("\t-trainLink <link prediction training file>\n")
print("\t-testLink <link prediction test file>\n")
print("\t-result <result output file>\n")
sys.exit()
#ratio = 0.8
entity_vec = {}
relation_vec = {}
for i in range(len(sys.argv)):
if sys.argv[i].startswith('-'):
option = sys.argv[i][1:]
if option == 'entityVec':
entity_file = sys.argv[i+1]
entity_vec = read_vec(entity_file,entity_vec)
# print("entityVec done!\n")
elif option == 'relationVec':
relation_file = sys.argv[i+1]
relation_vec = read_vec(relation_file,relation_vec)
# print("relationVec done!\n")
elif option == 'trainLink':
train_link_file = sys.argv[i+1]
elif option == 'testLink':
test_link_file = sys.argv[i+1]
# print("train link done!\n")
elif option == 'result':
result_file = sys.argv[i+1]
# print("result file done!\n")
# entity_file = input("input entity file : ")
# relation_file = input("input relation file : ")
# entity_vec = read_vec("result_matrix/"+entity_file,entity_vec)
# relation_vec = read_vec("result_matrix/"+relation_file,relation_vec)
print("entity vec num: ",len(entity_vec))
samples_train,target_train = load_train(train_link_file,entity_vec,relation_vec)
samples_test,target_test = load_train(test_link_file,entity_vec,relation_vec)
#print(type(samples))
#print(type(target))
LR(samples_train,target_train,samples_test,target_test)
#print("MSE = ",MSE)