forked from Bio503/MHAGP
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathline_embedding.py
118 lines (103 loc) · 3.95 KB
/
line_embedding.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
from cgi import test
from re import A
from model.line import LINE
import torch
import networkx as nx
import numpy as np
print(torch.cuda.is_available())
def get_graph_0():
G = nx.Graph()
all_node = []
all_edge = []
with open("./data/gmd.edgelist") as f_edge:
edge_content = f_edge.readlines()
for edge in edge_content:
node_1 = edge.strip("\n").split("\t")[0]
node_2 = edge.strip("\n").split("\t")[1]
weight = float(edge.strip("\n").split("\t")[2])
if weight >0:
all_edge.append((node_1, node_2))
if node_1 not in all_node:
all_node.append(node_1)
if node_2 not in all_node:
all_node.append(node_2)
G.add_nodes_from(all_node)
G.add_edges_from(all_edge)
print("节点数量：", G.number_of_nodes())
print("边的数量：", G.number_of_edges())
return G
# def get_graph(): # 根据边文件生成图，用于节点嵌入
# print("开始根据边文件生成Graph")
# G = nx.Graph()
# node_list = list(range(4677))
# G.add_nodes_from(node_list) # 首先在图中添加节点，然后再连接边
# with open("../0_data/edge_weight.txt") as f_edge:
# edge_content = f_edge.readlines()
# edge_list = []
# for edge in edge_content:
# head = int(edge.strip("\n").split(" ")[0])
# tail = int(edge.strip("\n").split(" ")[1])
# weight = float(edge.strip("\n").split(" ")[2])
# htw = (head, tail, weight)
# edge_list.append(htw)
# G.add_weighted_edges_from(edge_list)
# print("节点数量：", G.number_of_nodes())
# print("边的数量：", G.number_of_edges())
# return G
# 读取训练集测试集样本
filea = open("data/X_train.txt")
x_train = filea.readlines()
fileb = open("data/y_train.txt")
y_train = fileb.readlines()
filea = open("data/X_test.txt")
X_test = filea.readlines()
fileb = open("data/y_test.txt")
y_test = fileb.readlines()
# 训练集
train = []
for i in range(len(x_train)):
tup1 = (x_train[i].split()[0].split(",")[0],)
tup2 = (x_train[i].split()[0].split(",")[1],)
tup3 = (y_train[i].split()[0],)
tup4 = tup1 + tup2 + tup3
train.append(tup4) # ('G721', 'DO144', '0')
# 测试集
test = []
for i in range(len(X_test)):
tup1 = (X_test[i].split()[0].split(",")[0],)
tup2 = (X_test[i].split()[0].split(",")[1],)
tup3 = (y_test[i].split()[0],)
tup4 = tup1 + tup2 + tup3
test.append(tup4) # ('G721', 'DO144', '0')
if __name__ == "__main__":
G = get_graph_0()
# G = get_graph()
model = LINE(G, embedding_size=64, order='second')
model.train(batch_size=1024, epochs=1, verbose=2)
embeddings = model.get_embeddings()
embeddings_data = np.array(embeddings)
# 训练集
X_train_gd = []
for i in train:
gd_feature = np.concatenate((embeddings[i[0]],embeddings[i[1]]))
X_train_gd.append(gd_feature)
X_train_gd = np.array(X_train_gd)
np.save('data/X_train_gmd_line.npy',X_train_gd)
print(X_train_gd.shape)
# 测试集
X_test_gd = []
for i in test:
gd_feature = np.concatenate((embeddings[i[0]],embeddings[i[1]]))
X_test_gd.append(gd_feature)
X_test_gd = np.array(X_test_gd)
np.save('data/X_test_gmd_line.npy',X_test_gd)
print(X_test_gd.shape)
# with open("./data/gd_feature_line.txt","a") as f:
# print("开始写入文件！！！")
# for node,embed in embeddings.items():
# e = list(embed)
# e.insert(0, node)
# # lambda 冒号前是参数，可以有多个，用逗号隔开，冒号右边的为表达式。其实lambda返回值是一个函数的地址，也就是函数对象。
# e = list(map(lambda x:str(x), e)) # map() 会根据提供的函数对指定序列做映射。
# f.writelines(" ".join(e) + "\n")
print("完成写入！！！")