-
Notifications
You must be signed in to change notification settings - Fork 72
/
hg.py
executable file
·184 lines (165 loc) · 6.27 KB
/
hg.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
#coding=utf-8
""" pytorch实现：stacked hourglass network architecture"""
import torch
import torch.nn as nn
from torch.nn import UpsamplingNearest2d,Upsample
from torch.autograd import Variable
class StackedHourGlass(nn.Module):
def __init__(self,nFeats=256,nStack=8,nJoints=18):
"""
输入： 256^2
"""
super(StackedHourGlass,self).__init__()
self._nFeats = nFeats
self._nStack = nStack
self._nJoints = nJoints
self.conv1 = nn.Conv2d(3,64,7,2,3)
self.bn1 = nn.BatchNorm2d(64)
self.relu1 = nn.ReLU(inplace=True)
self.res1 = Residual(64,128)
self.pool1 = nn.MaxPool2d(2,2)
self.res2 = Residual(128,128)
self.res3 = Residual(128,self._nFeats) #cmd:option('-nFeats', 256, 'Number of features in the hourglass')
self._init_stacked_hourglass()
def _init_stacked_hourglass(self):
for i in range(self._nStack):
setattr(self,'hg'+str(i),HourGlass(4,self._nFeats))
setattr(self,'hg'+str(i)+'_res1',Residual(self._nFeats,self._nFeats))
setattr(self,'hg'+str(i)+'_lin1',Lin(self._nFeats,self._nFeats))
setattr(self,'hg'+str(i)+'_conv_pred',nn.Conv2d(self._nFeats,self._nJoints,1))
if i < self._nStack - 1:
setattr(self,'hg'+str(i)+'_conv1',nn.Conv2d(self._nFeats,self._nFeats,1))
setattr(self,'hg'+str(i)+'_conv2',nn.Conv2d(self._nJoints,self._nFeats,1))
def forward(self,x):
# 初始图像处理
x = self.relu1(self.bn1(self.conv1(x))) #(n,64,128,128)
x = self.res1(x) #(n,128,128,128)
x = self.pool1(x) #(n,128,64,64)
x = self.res2(x) #(n,128,64,64)
x = self.res3(x) #(n,256,64,64)
out = []
inter = x
for i in range(self._nStack): #cmd:option('-nStack', 8, 'Number of hourglasses to stack')
hg = eval('self.hg'+str(i))(inter)
# Residual layers at output resolution
ll = hg
ll = eval('self.hg'+str(i)+'_res1')(ll)
# Linear layer to produce first set of predictions
ll = eval('self.hg'+str(i)+'_lin1')(ll)
# Predicted heatmaps
tmpOut = eval('self.hg'+str(i)+'_conv_pred')(ll)
out.append(tmpOut)
# Add predictions back
if i < self._nStack - 1:
ll_ = eval('self.hg'+str(i)+'_conv1')(ll)
tmpOut_ = eval('self.hg'+str(i)+'_conv2')(tmpOut)
inter = inter + ll_ + tmpOut_
return out
class HourGlass(nn.Module):
"""不改变特征图的高宽"""
def __init__(self,n=4,f=256):
"""
:param n: hourglass模块的层级数目
:param f: hourglass模块中的特征图数量
:return:
"""
super(HourGlass,self).__init__()
self._n = n
self._f = f
self._init_layers(self._n,self._f)
def _init_layers(self,n,f):
# 上分支
setattr(self,'res'+str(n)+'_1',Residual(f,f))
# 下分支
setattr(self,'pool'+str(n)+'_1',nn.MaxPool2d(2,2))
setattr(self,'res'+str(n)+'_2',Residual(f,f))
if n > 1:
self._init_layers(n-1,f)
else:
self.res_center = Residual(f,f)
setattr(self,'res'+str(n)+'_3',Residual(f,f))
# setattr(self,'SUSN'+str(n),UpsamplingNearest2d(scale_factor=2))
setattr(self,'SUSN'+str(n),Upsample(scale_factor=2))
def _forward(self,x,n,f):
# 上分支
up1 = x
up1 = eval('self.res'+str(n)+'_1')(up1)
# 下分支
low1 = eval('self.pool'+str(n)+'_1')(x)
low1 = eval('self.res'+str(n)+'_2')(low1)
if n > 1:
low2 = self._forward(low1,n-1,f)
else:
low2 = self.res_center(low1)
low3 = low2
low3 = eval('self.'+'res'+str(n)+'_3')(low3)
up2 = eval('self.'+'SUSN'+str(n)).forward(low3)
return up1+up2
def forward(self,x):
return self._forward(x,self._n,self._f)
class Residual(nn.Module):
"""
残差模块，并不改变特征图的宽高
"""
def __init__(self,ins,outs):
super(Residual,self).__init__()
# 卷积模块
self.convBlock = nn.Sequential(
nn.BatchNorm2d(ins),
nn.ReLU(inplace=True),
nn.Conv2d(ins,outs/2,1),
nn.BatchNorm2d(outs/2),
nn.ReLU(inplace=True),
nn.Conv2d(outs/2,outs/2,3,1,1),
nn.BatchNorm2d(outs/2),
nn.ReLU(inplace=True),
nn.Conv2d(outs/2,outs,1)
)
# 跳层
if ins != outs:
self.skipConv = nn.Conv2d(ins,outs,1)
self.ins = ins
self.outs = outs
def forward(self,x):
residual = x
x = self.convBlock(x)
if self.ins != self.outs:
residual = self.skipConv(residual)
x += residual
return x
class Lin(nn.Module):
def __init__(self,numIn,numout):
super(Lin,self).__init__()
self.conv = nn.Conv2d(numIn,numout,1)
self.bn = nn.BatchNorm2d(numout)
self.relu = nn.ReLU(inplace=True)
def forward(self,x):
return self.relu(self.bn(self.conv(x)))
from torch.utils.data import Dataset,DataLoader
import numpy as np
import torch.optim as optim
class tempDataset(Dataset):
def __init__(self):
self.X = np.random.randn(100,3,512,512)
self.Y = np.random.randn(100,18,128,128)
def __len__(self):
return len(self.X)
def __getitem__(self, item):
# 这里返回的时候不要设置batch_size
return self.X[item],self.Y[item]
if __name__ == '__main__':
from torch.nn import MSELoss
critical = MSELoss()
dataset = tempDataset()
dataLoader = DataLoader(dataset=dataset)
shg = StackedHourGlass()
optimizer = optim.SGD(shg.parameters(), lr=0.01, momentum=0.9,weight_decay=1e-4)
for i,(x,y) in enumerate(dataLoader):
x = Variable(x,requires_grad=True).float()
y = Variable(y).float()
y_pred = shg.forward(x)
loss = critical(y_pred[0],y[0])
print('loss : {}'.format(loss))
optimizer.zero_grad()
loss.backward()
optimizer.step()