-
Notifications
You must be signed in to change notification settings - Fork 28
/
gcn.py
144 lines (125 loc) · 4.67 KB
/
gcn.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
# coding: utf-8
import math
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch_geometric as tg
# Graph Convolutional Network. For more information, please refer to https://arxiv.org/abs/1609.02907
# We copy and modify GCN code from https://github.com/tkipf/pygcn, and include this method in our graph embedding project framework.
# # Author: jhljx
# # Email: jhljx8918@gmail.com
class GraphConvolution(nn.Module):
input_dim: int
output_dim: int
weight: nn.Parameter
def __init__(self, input_dim, output_dim, bias=True):
super(GraphConvolution, self).__init__()
self.input_dim = input_dim
self.output_dim = output_dim
self.weight = nn.Parameter(torch.FloatTensor(input_dim, output_dim))
if bias:
self.bias = nn.Parameter(torch.FloatTensor(output_dim))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
stdv = 1. / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
if self.bias is not None:
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input, adj):
support = torch.mm(input, self.weight)
output = torch.spmm(adj, support)
del support
if self.bias is not None:
return output + self.bias
else:
return output
def __repr__(self):
return self.__class__.__name__ + ' (' \
+ str(self.input_dim) + ' -> ' \
+ str(self.output_dim) + ')'
# Original version of GCN
class GCN(nn.Module):
input_dim: int
hidden_dim: int
output_dim: int
dropout: float
bias: bool
method_name: str
gc1: GraphConvolution
gc2: GraphConvolution
def __init__(self, input_dim, hidden_dim, output_dim, dropout=0.5, bias=True):
super(GCN, self).__init__()
self.input_dim = input_dim
self.hidden_dim = hidden_dim
self.output_dim = output_dim
self.dropout = dropout
self.bias = bias
self.method_name = 'GCN'
self.gc1 = GraphConvolution(input_dim, hidden_dim, bias=bias)
self.gc2 = GraphConvolution(hidden_dim, output_dim, bias=bias)
def forward(self, x, adj):
# GCN for static embedding
if isinstance(x, list):
timestamp_num = len(x)
output_list = []
for i in range(timestamp_num):
output_list.append(self.gcn(x[i], adj[i]))
return output_list
return self.gcn(x, adj)
def gcn(self, x, adj):
x = F.relu(self.gc1(x, adj))
x = F.dropout(x, self.dropout, training=self.training)
x = self.gc2(x, adj)
return x
# Pytorch-Geometric version of GCN
class TgGCN(torch.nn.Module):
input_dim: int
feature_dim: int
hidden_dim: int
output_dim: int
feature_pre: bool
layer_num: int
dropout: float
bias: bool
method_name: str
def __init__(self, input_dim, feature_dim, hidden_dim, output_dim, feature_pre=True, layer_num=2, dropout=0.5, bias=True, **kwargs):
super(TgGCN, self).__init__()
self.input_dim = input_dim
self.feature_dim = feature_dim
self.hidden_dim = hidden_dim
self.output_dim = output_dim
self.feature_pre = feature_pre
self.layer_num = layer_num
self.dropout = dropout
self.bias = bias
self.method_name = 'TgGCN'
if feature_pre:
self.linear_pre = nn.Linear(input_dim, feature_dim, bias=bias)
self.conv_first = tg.nn.GCNConv(feature_dim, hidden_dim, bias=bias)
else:
self.conv_first = tg.nn.GCNConv(input_dim, hidden_dim, bias=bias)
self.conv_hidden = nn.ModuleList([tg.nn.GCNConv(hidden_dim, hidden_dim, bias=bias) for i in range(layer_num - 2)])
self.conv_out = tg.nn.GCNConv(hidden_dim, output_dim, bias=bias)
def forward(self, x, edge_index):
if isinstance(x, list):
timestamp_num = len(x)
output_list = []
for i in range(timestamp_num):
output_list.append(self.gcn(x[i], edge_index[i]))
return output_list
return self.gcn(x, edge_index)
def gcn(self, x, edge_index):
assert edge_index.shape[0] == 2
if self.feature_pre:
x = self.linear_pre(x)
x = self.conv_first(x, edge_index)
x = F.relu(x)
x = F.dropout(x, self.dropout, training=self.training)
for i in range(self.layer_num-2):
x = self.conv_hidden[i](x, edge_index)
x = F.relu(x)
x = F.dropout(x, self.dropout, training=self.training)
x = self.conv_out(x, edge_index)
return x