# DetNet结构+代码详细解析

## 与resnet相同的前4个stage

detnet-59是在resnet50的基础上进行改进，前四个stage都是一样的，也就是1+3×3+3×4+3×6=40层是一模一样的。而在代码中，是以Bottleneck类的形式出现：

class Bottleneck(nn.Module):
expansion = 4    #表示一个Bottleneck结构后通道数的扩张倍数

def __init__(self, inplanes, planes, stride=1, downsample=None):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = nn.BatchNorm2d(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
self.bn2 = nn.BatchNorm2d(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride

def forward(self, x):
residual = x

out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)

out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)

out = self.conv3(out)
out = self.bn3(out)

if self.downsample is not None:      #使残差部分和卷积前向出来之后通道一致
residual = self.downsample(x)

out += residual                      #参差部分相加
out = self.relu(out)

return out


    def _make_layer(self, block, planes, blocks, stride=1):  #这里的block就是上文的Bottleneck类
downsample = None
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
nn.Conv2d(self.inplanes, planes * block.expansion,    #如果输入的通道参数不对，做下采样调整
kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(planes * block.expansion),
)

layers = []
layers.append(block(self.inplanes, planes, stride, downsample))
self.inplanes = planes * block.expansion
for i in range(1, blocks):                                    #blocks意味着需要多少个重复的单元
layers.append(block(self.inplanes, planes))

return nn.Sequential(*layers)



        self.layer1 = self._make_layer(block, 64, layers[0])  #layers[:3]=[3,4,6] 对应resnet50中2-4的stage
self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
self.layer3 = self._make_layer(block, 256, layers[2], stride=2)


## 后2个不同的stage

detnet后面给出自己设计的Bottleneck类，首先我们先来看一下结构：

class BottleneckA(nn.Module):
expansion = 4

def __init__(self, inplanes, planes, stride=1, downsample=None):
super(BottleneckA, self).__init__()
assert inplanes == (planes * 4), 'inplanes != planes * 4'
assert stride == 1, 'stride != 1'
assert downsample is None, 'downsample is not None'
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)  # inplanes = 1024, planes = 256
self.bn1 = nn.BatchNorm2d(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, dilation=2,
padding=2, bias=False)  # stride = 1, dilation = 2
self.bn2 = nn.BatchNorm2d(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride

def forward(self, x):
residual = x

out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)

out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)

out = self.conv3(out)
out = self.bn3(out)

if self.downsample is not None:  # downsample always is None, because stride=1 and inplanes=expansion * planes
residual = self.downsample(x)

out += residual
out = self.relu(out)

return out


bottleneckA便是下图这样的结构，主干路上增加了空洞卷积，另一线路并不做处理。整个通道的变化是inplanes–planes–planes–planes*4,而通常设置planes是inplanes的四分之一，因此经过一个bottleneckA结构后,feature的通道数并没有发生改变。

class BottleneckB(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None):
super(BottleneckB, self).__init__()
assert inplanes == (planes * 4), 'inplanes != planes * 4'
assert stride == 1, 'stride != 1'
assert downsample is None, 'downsample is not None'
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)  # inplanes = 1024, planes = 256
self.bn1 = nn.BatchNorm2d(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, dilation=2,
padding=2, bias=False)  # stride = 1, dilation = 2
self.bn2 = nn.BatchNorm2d(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
self.extra_conv = nn.Sequential(     #对残差进行处理的单元
nn.Conv2d(inplanes, planes * 4, kernel_size=1, bias=False),
nn.BatchNorm2d(planes * 4)
)

def forward(self, x):
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)

out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)

out = self.conv3(out)
out = self.bn3(out)

residual = self.extra_conv(x)      #区别就在这一句话

if self.downsample is not None:  # downsample always is None, because stride=1 and inplanes=expansion * planes
residual = self.downsample(x)

out += residual
out = self.relu(out)               #继续相加

return out


stage5和6 便是一个bottleneckB后接俩个bottleneckA结构，也就是3 ×（2+1）× 2 = 18层，加上前面的40层，便是58层，再加上最后的全连接层，也就构成了Detnet-59的全部基本结构了。

## 整体网络生成的代码

class DetNet(nn.Module):
def __init__(self, block, layers, num_classes=1000):  #block就是Bottleneck, 这里layers=[3, 4, 6, 3, 3]
self.inplanes = 64
super(DetNet, self).__init__()
self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
bias=False)     #网络的第一层卷积，卷积核为7*7
self.bn1 = nn.BatchNorm2d(64)
self.relu = nn.ReLU(inplace=True)
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) #第一个池化层         这里layers=[3, 4, 6, 3, 3]，与上文提到的结构相同
self.layer1 = self._make_layer(block, 64, layers[0])
self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
self.layer4 = self._make_new_layer(256, layers[3]) #stage5
self.layer5 = self._make_new_layer(256, layers[4]) #stage6
self.fc = nn.Linear(1024, num_classes)

for m in self.modules():
if isinstance(m, nn.Conv2d):
n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
m.weight.data.normal_(0, math.sqrt(2. / n))
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()

def _make_layer(self, block, planes, blocks, stride=1):#与resnet一样的layer
downsample = None
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
nn.Conv2d(self.inplanes, planes * block.expansion,
kernel_size=1, stride=stride, bias=False),
nn.BatchNorm2d(planes * block.expansion),
)

layers = []
layers.append(block(self.inplanes, planes, stride, downsample))
self.inplanes = planes * block.expansion
for i in range(1, blocks):     #设置了几个这样的层，就创建几个层
layers.append(block(self.inplanes, planes))

return nn.Sequential(*layers)

def _make_new_layer(self, planes, blocks):  #创建新的detnet自己特有的单元
downsample = None
block_b = BottleneckB
block_a = BottleneckA

layers = []
layers.append(block_b(self.inplanes, planes, stride=1, downsample=downsample))   #每个stage中开头都是bottleneckB的结构
self.inplanes = planes * block_b.expansion
for i in range(1, blocks):   #再加上blocks-1个bottleneckA
layers.append(block_a(self.inplanes, planes))

return nn.Sequential(*layers)

def forward(self, x):    #简洁的前向结构
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.maxpool(x)

x = self.layer1(x)
x = self.layer2(x)
x = self.layer3(x)
x = self.layer4(x)
x = self.layer5(x)

x = self.avgpool(x)
x = x.view(x.size(0), -1)
x = self.fc(x)

return x