-
Notifications
You must be signed in to change notification settings - Fork 0
/
stn_coorconv.py
62 lines (50 loc) · 1.96 KB
/
stn_coorconv.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
import torch.nn as nn
import torch.nn.functional as F
from coordconv import *
class CoordNet(nn.Module):
def __init__(self, use_cuda=True):
super(CoordNet, self).__init__()
self.conv1 = CoordConv2d(1, 10, kernel_size=5, use_cuda=use_cuda)
self.conv2 = CoordConv2d(10, 20, kernel_size=5, use_cuda=use_cuda)
self.conv2_drop = nn.Dropout2d()
self.fc1 = nn.Linear(320, 50)
self.fc2 = nn.Linear(50, 10)
# Spatial transformer localization-network
self.localization = nn.Sequential(
CoordConv2d(1, 8, kernel_size=7, use_cuda=use_cuda),
nn.MaxPool2d(2, stride=2),
nn.ReLU(True),
CoordConv2d(8, 10, kernel_size=5, use_cuda=use_cuda),
nn.MaxPool2d(2, stride=2),
nn.ReLU(True),
)
# Regressor for the 3 * 2 affine matrix
self.fc_loc = nn.Sequential(
nn.Linear(10 * 3 * 3, 32), nn.ReLU(True), nn.Linear(32, 3 * 2)
)
transformation_matrix = [1, 0, 0, 0, 1, 0]
# Initialize the weights/bias with the selected transformation matrix
self.fc_loc[2].weight.data.zero_()
self.fc_loc[2].bias.data.copy_(
torch.tensor(transformation_matrix, dtype=torch.float)
)
# Spatial transformer network forward function
def stn(self, x):
xs = self.localization(x)
xs = xs.view(-1, 10 * 3 * 3)
theta = self.fc_loc(xs)
theta = theta.view(-1, 2, 3)
grid = F.affine_grid(theta, x.size())
x = F.grid_sample(x, grid)
return x
def forward(self, x):
# transform the input
x = self.stn(x)
# Perform the usual forward pass
x = F.relu(F.max_pool2d(self.conv1(x), 2))
x = F.relu(F.max_pool2d(self.conv2_drop(self.conv2(x)), 2))
x = x.view(-1, 320)
x = F.relu(self.fc1(x))
x = F.dropout(x, training=self.training)
x = self.fc2(x)
return F.log_softmax(x, dim=1)