-
Notifications
You must be signed in to change notification settings - Fork 2
/
transE.py
executable file
·137 lines (114 loc) · 4.58 KB
/
transE.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
#coding:utf-8
import numpy as np
import tensorflow as tf
import os
import time
import datetime
import ctypes
ll = ctypes.cdll.LoadLibrary
lib = ll("./init.so")
class Config(object):
def __init__(self):
self.L1_flag = True
self.hidden_size = 100
self.nbatches = 100
self.entity = 0
self.relation = 0
self.trainTimes = 3000
self.margin = 1.0
class TransEModel(object):
def __init__(self, config):
entity_total = config.entity
relation_total = config.relation
batch_size = config.batch_size
size = config.hidden_size
margin = config.margin
self.pos_h = tf.placeholder(tf.int32, [None])
self.pos_t = tf.placeholder(tf.int32, [None])
self.pos_r = tf.placeholder(tf.int32, [None])
self.neg_h = tf.placeholder(tf.int32, [None])
self.neg_t = tf.placeholder(tf.int32, [None])
self.neg_r = tf.placeholder(tf.int32, [None])
with tf.name_scope("embedding"):
self.ent_embeddings = tf.get_variable(name = "ent_embedding", shape = [entity_total, size], initializer = tf.contrib.layers.xavier_initializer(uniform = False))
self.rel_embeddings = tf.get_variable(name = "rel_embedding", shape = [relation_total, size], initializer = tf.contrib.layers.xavier_initializer(uniform = False))
pos_h_e = tf.nn.embedding_lookup(self.ent_embeddings, self.pos_h)
pos_t_e = tf.nn.embedding_lookup(self.ent_embeddings, self.pos_t)
pos_r_e = tf.nn.embedding_lookup(self.rel_embeddings, self.pos_r)
neg_h_e = tf.nn.embedding_lookup(self.ent_embeddings, self.neg_h)
neg_t_e = tf.nn.embedding_lookup(self.ent_embeddings, self.neg_t)
neg_r_e = tf.nn.embedding_lookup(self.rel_embeddings, self.neg_r)
if config.L1_flag:
pos = tf.reduce_sum(abs(pos_h_e + pos_r_e - pos_t_e), 1, keep_dims = True)
neg = tf.reduce_sum(abs(neg_h_e + neg_r_e - neg_t_e), 1, keep_dims = True)
else:
pos = tf.reduce_sum((pos_h_e + pos_r_e - pos_t_e) ** 2, 1, keep_dims = True)
neg = tf.reduce_sum((neg_h_e + neg_r_e - neg_t_e) ** 2, 1, keep_dims = True)
with tf.name_scope("output"):
self.loss = tf.reduce_sum(tf.maximum(pos - neg + margin, 0))
def main(_):
lib.init()
config = Config()
config.relation = lib.getRelationTotal()
config.entity = lib.getEntityTotal()
config.batch_size = lib.getTripleTotal() / config.nbatches
with tf.Graph().as_default():
sess = tf.Session()
with sess.as_default():
initializer = tf.contrib.layers.xavier_initializer(uniform = False)
with tf.variable_scope("model", reuse=None, initializer = initializer):
trainModel = TransEModel(config = config)
global_step = tf.Variable(0, name="global_step", trainable=False)
optimizer = tf.train.GradientDescentOptimizer(0.001)
grads_and_vars = optimizer.compute_gradients(trainModel.loss)
train_op = optimizer.apply_gradients(grads_and_vars, global_step=global_step)
sess.run(tf.initialize_all_variables())
def train_step(pos_h_batch, pos_t_batch, pos_r_batch, neg_h_batch, neg_t_batch, neg_r_batch):
feed_dict = {
trainModel.pos_h: pos_h_batch,
trainModel.pos_t: pos_t_batch,
trainModel.pos_r: pos_r_batch,
trainModel.neg_h: neg_h_batch,
trainModel.neg_t: neg_t_batch,
trainModel.neg_r: neg_r_batch
}
_, step, loss = sess.run(
[train_op, global_step, trainModel.loss], feed_dict)
return loss
ph = np.zeros(config.batch_size, dtype = np.int32)
pt = np.zeros(config.batch_size, dtype = np.int32)
pr = np.zeros(config.batch_size, dtype = np.int32)
nh = np.zeros(config.batch_size, dtype = np.int32)
nt = np.zeros(config.batch_size, dtype = np.int32)
nr = np.zeros(config.batch_size, dtype = np.int32)
ph_addr = ph.__array_interface__['data'][0]
pt_addr = pt.__array_interface__['data'][0]
pr_addr = pr.__array_interface__['data'][0]
nh_addr = nh.__array_interface__['data'][0]
nt_addr = nt.__array_interface__['data'][0]
nr_addr = nr.__array_interface__['data'][0]
for times in range(config.trainTimes):
res = 0.0
for batch in range(config.nbatches):
lib.getBatch(ph_addr, pt_addr, pr_addr, nh_addr, nt_addr, nr_addr, config.batch_size)
res += train_step(ph, pt, pr, nh, nt, nr)
current_step = tf.train.global_step(sess, global_step)
print times
print res
if (times % 100 == 0):
f = open("entity2vec.txt", "w")
enb = trainModel.ent_embeddings.eval()
for i in enb:
for j in i:
f.write("%f\t"%(j))
f.write("\n")
f.close()
f = open("relation2vec.txt", "w")
enb = trainModel.rel_embeddings.eval()
for i in enb:
for j in i:
f.write("%f\t"%(j))
f.write("\n")
f.close()
if __name__ == "__main__":
tf.app.run()