group Convolution
在普通的卷积中,channels 即同一个卷积对所有的channels操作,然后相加.
而group convolution,即简单的讲就是把 channel 做N等分(N个group)，然后每一份（一个group）分别与上一层的输出的M/N个channel独立连接，之后将每个group的输出叠在一起（concatenate），作为这一层的输出 channel.
group conv最早出现在AlexNet[1]中，因为显卡显存不够，只好把网络分在两块卡里.
mobilenet v1 中的depthwise convolution操作其实是每一个channel都为一个group的特殊情况
shufflenet
Channel Shuffle
介于每个channel都用单独一个卷积 (Pointwise convolution)or 所有通道共用一个卷积(传统卷积).
Channel Shuffle 提出了将channel 分组,然后仅在分组内进行Pointwise卷积.
但是，如果多个组卷积叠加在一起，则会产生一个副作用：某个通道的输出仅来自一小部分输入通道。如图(a)所示.该组的输出仅与该组内的输入有关,阻碍了通道间的信息流.
如果我们允许group convolution 从不同的channel中获取信息(如图b所示),则输入通道和输出通道信息完全相关
shufflenet用 channel shuffle 来实现这一效果(如图c)
image
具体做法:
假设一个卷积层有个输出channel(g 个group),
- (1)先将输出通道reshape维度为
- (2)transpose:将通道信息变为
,通道信息随机变换
- (3)reshape,将通道恢复原来的shape
在transpose过程中进行了通道混乱
pytroch 代码如图:
def shuffle_channels(x, groups):
"""shuffle channels of a 4-D Tensor"""
batch_size, channels, height, width = x.size()
assert channels % groups == 0
channels_per_group = channels // groups
# split into groups
x = x.view(batch_size, groups, channels_per_group,
height, width)
# transpose 1, 2 axis
x = x.transpose(1, 2).contiguous()
# reshape into orignal
x = x.view(batch_size, channels, height, width)
return x
shufflenet v1 bottleneck
shuffle net的组件如图所示:
image
(b)代表了stride 为 1 ,
(c)stride 为2
image
import torch
import torch.nn as nn
import torch.nn.functional as F
def shuffle_channels(x, groups):
"""shuffle channels of a 4-D Tensor"""
batch_size, channels, height, width = x.size()
assert channels % groups == 0
channels_per_group = channels // groups
# split into groups
x = x.view(batch_size, groups, channels_per_group,
height, width)
# transpose 1, 2 axis
x = x.transpose(1, 2).contiguous()
# reshape into orignal
x = x.view(batch_size, channels, height, width)
return x
class ShuffleBottleNeck(nn.Module):
def __init__(self,in_channels,out_channels,stride,groups):
super(ShuffleBottleNeck,self).__init__()
self.stride = stride
self.groups = groups
# bottleneck层中间层的channel数变为输出channel数的1/4
#we set the number of bottleneck channels to 1/4 of the output channels for each ShuffleNetunit.
mid_channels = int(out_channels / 4)
set_groups = groups if in_channels!=24 else 1
# 作者提到不在stage2的第一个pointwise层使用组卷积,因为输入channel数量太少,只有24
self.conv1 = nn.Conv2d(in_channels,mid_channels,kernel_size=1,
groups=set_groups,bias=False)
self.bn1 = nn.BatchNorm2d(mid_channels)
self.conv2 = nn.Conv2d(mid_channels,mid_channels,kernel_size=3,
groups=mid_channels,padding=1,stride=stride,
bias=False)
self.bn2 = nn.BatchNorm2d(mid_channels)
self.conv3 = nn.Conv2d(mid_channels,out_channels,kernel_size=1,
groups=groups,bias=False)
self.bn3 = nn.BatchNorm2d(out_channels)
self.shortcut = nn.Sequential()
if stride ==2:
self.shortcut = nn.Sequential(nn.AvgPool2d(3,stride=2,padding=1))
def forward(self, x):
out = torch.nn.functional.relu(self.bn1(self.conv1(x)))
out = shuffle_channels(out,self.groups)
out = self.bn2(self.conv2(out))
out = self.bn3(self.conv3(out))
res = self.shortcut(x)
out = F.relu(torch.cat([out, res], 1)) if self.stride == 2 else F.relu(out + res)
return out
class ShuffleNet(nn.Module):
def __init__(self, cfg):
super(ShuffleNet, self).__init__()
out_planes = cfg['out_planes']
num_blocks = cfg['num_blocks']
groups = cfg['groups']
self.conv1 = nn.Conv2d(3, 24, kernel_size=1, bias=False)
self.bn1 = nn.BatchNorm2d(24)
self.in_planes = 24
self.layer1 = self._make_layer(out_planes[0], num_blocks[0], groups)
self.layer2 = self._make_layer(out_planes[1], num_blocks[1], groups)
self.layer3 = self._make_layer(out_planes[2], num_blocks[2], groups)
self.linear = nn.Linear(out_planes[2], 10)
def _make_layer(self, out_planes, num_blocks, groups):
layers = []
for i in range(num_blocks):
if i == 0:
layers.append(ShuffleBottleNeck(self.in_planes,
out_planes-self.in_planes,
stride=2, groups=groups))
else:
layers.append(ShuffleBottleNeck(self.in_planes,
out_planes,
stride=1, groups=groups))
self.in_planes = out_planes
return nn.Sequential(*layers)
def forward(self, x):
out = F.relu(self.bn1(self.conv1(x)))
out = self.layer1(out)
out = self.layer2(out)
out = self.layer3(out)
out = F.avg_pool2d(out, 4)
out = out.view(out.size(0), -1)
out = self.linear(out)
return out
def ShuffleNetG2():
cfg = {
'out_planes': [200,400,800],
'num_blocks': [4,8,4],
'groups': 2
}
return ShuffleNet(cfg)
def ShuffleNetG3():
cfg = {
'out_planes': [240,480,960],
'num_blocks': [4,8,4],
'groups': 3
}
return ShuffleNet(cfg)
def test():
net = ShuffleNetG2()
x = torch.randn(1,3,32,32)
y = net(x)
print(y)
if __name__ == '__main__':
test()
