-
Notifications
You must be signed in to change notification settings - Fork 0
/
unetpp_AGC.py
184 lines (165 loc) · 7.17 KB
/
unetpp_AGC.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
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
from torch import nn
from torch.nn import functional as F
import torch
from torchvision import models
import torchvision
from AGC import *
class DoubleConv(nn.Module):
def __init__(self, in_ch, out_ch):
super(DoubleConv, self).__init__()
self.conv = nn.Sequential(
nn.Conv2d(in_ch, out_ch, 3, padding=1),
nn.BatchNorm2d(out_ch),
nn.ReLU(inplace=True),
nn.Conv2d(out_ch, out_ch, 3, padding=1),
nn.BatchNorm2d(out_ch),
nn.ReLU(inplace=True)
)
def forward(self, input):
return self.conv(input)
# class VGGBlock(nn.Module):
# def __init__(self, in_channels, middle_channels, out_channels, act_func=nn.ReLU(inplace=True)):
# super(VGGBlock, self).__init__()
# self.act_func = act_func
# self.conv1 = nn.Conv2d(in_channels, middle_channels, 3, padding=1)
# self.bn1 = nn.BatchNorm2d(middle_channels)
# self.conv2 = nn.Conv2d(middle_channels, out_channels, 3, padding=1)
# self.bn2 = nn.BatchNorm2d(out_channels)
#
# def forward(self, x):
# out = self.conv1(x)
# out = self.bn1(out)
# out = self.act_func(out)
#
# out = self.conv2(out)
# out = self.bn2(out)
# out = self.act_func(out)
# return out
class NestedUNet_AGC(nn.Module):
def __init__(self, args,in_channel,out_channel):
super().__init__()
self.args = args
nb_filter = [32, 64, 128, 256, 512]
self.pool = nn.MaxPool2d(2, 2)
self.up = nn.Upsample(scale_factor=2, mode='bilinear', align_corners=True)
self.conv0_0 = DoubleConv(in_channel, nb_filter[0])
self.conv1_0 = DoubleConv(nb_filter[0], nb_filter[1])
"""
for para in self.pool.parameters():
para.requires_grad = False
for para in self.up.parameters():
para.requires_grad = False
for para in self.conv0_0.parameters():
para.requires_grad = False
for para in self.conv1_0.parameters():
para.requires_grad = False
"""
self.AGC0 = AGCnet(nb_filter[1],nb_filter[1])
self.conv2_0 = DoubleConv(nb_filter[1], nb_filter[2])
self.conv3_0 = DoubleConv(nb_filter[2], nb_filter[3])
self.conv4_0 = DoubleConv(nb_filter[3], nb_filter[4])
"""
for para in self.conv2_0.parameters():
para.requires_grad = False
for para in self.conv3_0.parameters():
para.requires_grad = False
for para in self.conv4_0.parameters():
para.requires_grad = False
"""
self.AGC = AGCnet(nb_filter[4],nb_filter[4])
self.conv0_1 = DoubleConv(nb_filter[0]+nb_filter[1], nb_filter[0])
self.conv1_1 = DoubleConv(nb_filter[1]+nb_filter[2], nb_filter[1])
self.conv2_1 = DoubleConv(nb_filter[2]+nb_filter[3], nb_filter[2])
self.conv3_1 = DoubleConv(nb_filter[3]+nb_filter[4], nb_filter[3])
"""
for para in self.conv0_1.parameters():
para.requires_grad = False
for para in self.conv1_1.parameters():
para.requires_grad = False
for para in self.conv2_1.parameters():
para.requires_grad = False
for para in self.conv3_1.parameters():
para.requires_grad = False
"""
self.conv0_2 = DoubleConv(nb_filter[0]*2+nb_filter[1], nb_filter[0])
self.conv1_2 = DoubleConv(nb_filter[1]*2+nb_filter[2], nb_filter[1])
self.conv2_2 = DoubleConv(nb_filter[2]*2+nb_filter[3], nb_filter[2])
"""
for para in self.conv0_2.parameters():
para.requires_grad = False
for para in self.conv1_2.parameters():
para.requires_grad = False
for para in self.conv2_2.parameters():
para.requires_grad = False
"""
self.conv0_3 = DoubleConv(nb_filter[0]*3+nb_filter[1], nb_filter[0])
self.conv1_3 = DoubleConv(nb_filter[1]*3+nb_filter[2], nb_filter[1])
"""
for para in self.conv0_3.parameters():
para.requires_grad = False
for para in self.conv1_3.parameters():
para.requires_grad = False
"""
#self.AGC1 = AGCnet(nb_filter[1],nb_filter[1])
self.conv0_4 = DoubleConv(nb_filter[0]*4+nb_filter[1], nb_filter[0])
self.sigmoid = nn.Sigmoid()
if self.args.deepsupervision:
self.final1 = nn.Conv2d(nb_filter[0], out_channel, kernel_size=1)
self.final2 = nn.Conv2d(nb_filter[0], out_channel, kernel_size=1)
self.final3 = nn.Conv2d(nb_filter[0], out_channel, kernel_size=1)
self.final4 = nn.Conv2d(nb_filter[0], out_channel, kernel_size=1)
"""
for para in self.final1.parameters():
para.requires_grad = False
for para in self.final2.parameters():
para.requires_grad = False
for para in self.final3.parameters():
para.requires_grad = False
for para in self.final4.parameters():
para.requires_grad = False
"""
else:
self.final = nn.Conv2d(nb_filter[0], out_channel, kernel_size=1)
"""
for para in self.final.parameters():
para.requires_grad = False
"""
"""
for para in self.conv0_4.parameters():
para.requires_grad = False
for para in self.sigmoid.parameters():
para.requires_grad = False
"""
def forward(self, input):
x0_0 = self.conv0_0(input)
x1_0 = self.conv1_0(self.pool(x0_0))
x0_1 = self.conv0_1(torch.cat([x0_0, self.up(x1_0)], 1))
x2_0 = self.AGC0(x1_0)
x2_0 = self.conv2_0(self.pool(x1_0))
x1_1 = self.conv1_1(torch.cat([x1_0, self.up(x2_0)], 1))
x0_2 = self.conv0_2(torch.cat([x0_0, x0_1, self.up(x1_1)], 1))
x3_0 = self.conv3_0(self.pool(x2_0))
x2_1 = self.conv2_1(torch.cat([x2_0, self.up(x3_0)], 1))
x1_2 = self.conv1_2(torch.cat([x1_0, x1_1, self.up(x2_1)], 1))
x0_3 = self.conv0_3(torch.cat([x0_0, x0_1, x0_2, self.up(x1_2)], 1))
x4_0 = self.conv4_0(self.pool(x3_0))
x4_0 = self.AGC(x4_0)
x3_1 = self.conv3_1(torch.cat([x3_0, self.up(x4_0)], 1))
x2_2 = self.conv2_2(torch.cat([x2_0, x2_1, self.up(x3_1)], 1))
x1_3 = self.conv1_3(torch.cat([x1_0, x1_1, x1_2, self.up(x2_2)], 1))
#x1_3 = self.AGC1(x1_3)
x0_4 = self.conv0_4(torch.cat([x0_0, x0_1, x0_2, x0_3, self.up(x1_3)], 1))
if self.args.deepsupervision:
output1 = self.final1(x0_1)
output1 = self.sigmoid(output1)
output2 = self.final2(x0_2)
output2 = self.sigmoid(output2)
output3 = self.final3(x0_3)
output3 = self.sigmoid(output3)
output4 = self.final4(x0_4)
output4 = self.sigmoid(output4)
return [output1, output2, output3, output4]
else:
output = self.final(x0_4)
output = self.sigmoid(output)
return output