Inheritance diagram for fastreid.layers.non_local.Non_local:
Public Member Functions | |
| __init__ (self, in_channels, bn_norm, reduc_ratio=2) | |
| forward (self, x) | |
Public Attributes | |
| in_channels | |
| inter_channels | |
| g | |
| W | |
| theta | |
| phi | |
Detailed Description
Definition at line 9 of file non_local.py.
Constructor & Destructor Documentation
◆ __init__()
| fastreid.layers.non_local.Non_local.__init__ | ( | self, | |
| in_channels, | |||
| bn_norm, | |||
| reduc_ratio = 2 ) |
Definition at line 10 of file non_local.py.
10 def __init__(self, in_channels, bn_norm, reduc_ratio=2):
11 super(Non_local, self).__init__()
12
13 self.in_channels = in_channels
14 self.inter_channels = in_channels // reduc_ratio
15
16 self.g = nn.Conv2d(in_channels=self.in_channels, out_channels=self.inter_channels,
17 kernel_size=1, stride=1, padding=0)
18
19 self.W = nn.Sequential(
20 nn.Conv2d(in_channels=self.inter_channels, out_channels=self.in_channels,
21 kernel_size=1, stride=1, padding=0),
22 get_norm(bn_norm, self.in_channels),
23 )
24 nn.init.constant_(self.W[1].weight, 0.0)
25 nn.init.constant_(self.W[1].bias, 0.0)
26
27 self.theta = nn.Conv2d(in_channels=self.in_channels, out_channels=self.inter_channels,
28 kernel_size=1, stride=1, padding=0)
29
30 self.phi = nn.Conv2d(in_channels=self.in_channels, out_channels=self.inter_channels,
31 kernel_size=1, stride=1, padding=0)
32
Member Function Documentation
◆ forward()
| fastreid.layers.non_local.Non_local.forward | ( | self, | |
| x ) |
:param x: (b, t, h, w)
:return x: (b, t, h, w)
Definition at line 33 of file non_local.py.
33 def forward(self, x):
34 """
35 :param x: (b, t, h, w)
36 :return x: (b, t, h, w)
37 """
38 batch_size = x.size(0)
39 g_x = self.g(x).view(batch_size, self.inter_channels, -1)
40 g_x = g_x.permute(0, 2, 1)
41
42 theta_x = self.theta(x).view(batch_size, self.inter_channels, -1)
43 theta_x = theta_x.permute(0, 2, 1)
44 phi_x = self.phi(x).view(batch_size, self.inter_channels, -1)
45 f = torch.matmul(theta_x, phi_x)
46 N = f.size(-1)
47 f_div_C = f / N
48
49 y = torch.matmul(f_div_C, g_x)
50 y = y.permute(0, 2, 1).contiguous()
51 y = y.view(batch_size, self.inter_channels, *x.size()[2:])
52 W_y = self.W(y)
53 z = W_y + x
54 return z
Member Data Documentation
◆ g
| fastreid.layers.non_local.Non_local.g |
Definition at line 16 of file non_local.py.
◆ in_channels
| fastreid.layers.non_local.Non_local.in_channels |
Definition at line 13 of file non_local.py.
◆ inter_channels
| fastreid.layers.non_local.Non_local.inter_channels |
Definition at line 14 of file non_local.py.
◆ phi
| fastreid.layers.non_local.Non_local.phi |
Definition at line 30 of file non_local.py.
◆ theta
| fastreid.layers.non_local.Non_local.theta |
Definition at line 27 of file non_local.py.
◆ W
| fastreid.layers.non_local.Non_local.W |
Definition at line 19 of file non_local.py.
The documentation for this class was generated from the following file:
- smreid/fastreid/layers/non_local.py
