图像分类丨浅析轻量级网络「SqueezeNet、MobileNet、ShuffleNet」

深度卷积网络除了准确度，计算复杂度也是考虑的重要指标。本文列出了近年主流的轻量级网络，简单地阐述了它们的思想。由于本人水平有限，对这部分的理解还不够深入，还需要继续学习和完善。

最后我参考部分列出来的文章都写的非常棒，建议继续阅读。

复杂度分析

理论计算量（FLOPs）：浮点运算次数（FLoating-point Operation）
参数数量(params)：单位通常为M，用float32表示。

对比

std conv（主要贡献计算量）
- params：𝑘ℎ×𝑘𝑤×𝑐𝑖𝑛×𝑐𝑜𝑢𝑡kh×kw×cin×cout
- FLOPs：𝑘ℎ×𝑘𝑤×𝑐𝑖𝑛×𝑐𝑜𝑢𝑡×𝐻×𝑊kh×kw×cin×cout×H×W
fc（主要贡献参数量）
- params：𝑐𝑖𝑛×𝑐𝑜𝑢𝑡cin×cout
- FLOPs：𝑐𝑖𝑛×𝑐𝑜𝑢𝑡cin×cout
group conv
- params：(𝑘ℎ×𝑘𝑤×𝑐𝑖𝑛/𝑔×𝑐𝑜𝑢𝑡/𝑔)×𝑔=𝑘ℎ×𝑘𝑤×𝑐𝑖𝑛×𝑐𝑜𝑢𝑡/𝑔(kh×kw×cin/g×cout/g)×g=kh×kw×cin×cout/g
- FLOPs：𝑘ℎ×𝑘𝑤×𝑐𝑖𝑛×𝑐𝑜𝑢𝑡×𝐻×𝑊/𝑔kh×kw×cin×cout×H×W/g
depth-wise conv
- params：𝑘ℎ×𝑘𝑤×𝑐𝑖𝑛×𝑐𝑜𝑢𝑡/𝑐𝑖𝑛=𝑘ℎ×𝑘𝑤×𝑐𝑜𝑢𝑡kh×kw×cin×cout/cin=kh×kw×cout
- FLOPs：𝑘ℎ×𝑘𝑤×𝑐𝑜𝑢𝑡×𝐻×𝑊kh×kw×cout×H×W

SqueezeNet#

SqueezeNet：AlexNet-level accuracy with 50x fewer parameters and <0.5MB

核心思想#

提出Fire module，包含两部分：squeeze和expand层。
1. squeeze为1x1卷积，𝑆1<𝑀S1<M，从而压缩
2. Expand层为e1个1x1卷积和e3个3x3卷积，分别输出𝐻×𝑊×𝑒1H×W×e1和𝐻×𝑊×𝑒2H×W×e2。
3. concat得到𝐻×𝑊×(𝑒1+𝑒3)H×W×(e1+e3)

preview

Copy

class Fire(nn.Module):
    def __init__(self, in_channel, out_channel, squzee_channel): super().__init__() self.squeeze = nn.Sequential( nn.Conv2d(in_channel, squzee_channel, 1), nn.BatchNorm2d(squzee_channel), nn.ReLU(inplace=True) ) self.expand_1x1 = nn.Sequential( nn.Conv2d(squzee_channel, int(out_channel / 2), 1), nn.BatchNorm2d(int(out_channel / 2)), nn.ReLU(inplace=True) ) self.expand_3x3 = nn.Sequential( nn.Conv2d(squzee_channel, int(out_channel / 2), 3, padding=1), nn.BatchNorm2d(int(out_channel / 2)), nn.ReLU(inplace=True) ) def forward(self, x): x = self.squeeze(x) x = torch.cat([ self.expand_1x1(x), self.expand_3x3(x) ], 1) return x

网络架构#

Copy

class SqueezeNet(nn.Module):
    """mobile net with simple bypass""" def __init__(self, class_num=100): super().__init__() self.stem = nn.Sequential( nn.Conv2d(3, 96, 3, padding=1), nn.BatchNorm2d(96), nn.ReLU(inplace=True), nn.MaxPool2d(2, 2) ) self.fire2 = Fire(96, 128, 16) self.fire3 = Fire(128, 128, 16) self.fire4 = Fire(128, 256, 32) self.fire5 = Fire(256, 256, 32) self.fire6 = Fire(256, 384, 48) self.fire7 = Fire(384, 384, 48) self.fire8 = Fire(384, 512, 64) self.fire9 = Fire(512, 512, 64) self.conv10 = nn.Conv2d(512, class_num, 1) self.avg = nn.AdaptiveAvgPool2d(1) self.maxpool = nn.MaxPool2d(2, 2) def forward(self, x): x = self.stem(x) f2 = self.fire2(x) f3 = self.fire3(f2) + f2 f4 = self.fire4(f3) f4 = self.maxpool(f4) f5 = self.fire5(f4) + f4 f6 = self.fire6(f5) f7 = self.fire7(f6) + f6 f8 = self.fire8(f7) f8 = self.maxpool(f8) f9 = self.fire9(f8) c10 = self.conv10(f9) x = self.avg(c10) x = x.view(x.size(0), -1) return x def squeezenet(class_num=100): return SqueezeNet(class_num=class_num)

实验结果#

注意：0.5MB是模型压缩的结果。

1558486836354

MobileNetV1#

MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

核心思想#

使用了depth-wise separable conv降低了参数和计算量。
提出两个超参数Width Multiplier和Resolution Multiplier来平衡时间和精度。

depth-wise separable conv

Standard Conv

𝐷𝐾DK：kernel size

𝐷𝐹DF：feature map size

𝑀M：input channel number

𝑁N：output channel number

v2-fcdf7e76635e58c7415653521f833a54_b

参数量：𝐷𝐾×𝐷𝐾×𝑀×𝑁(3×3×3×2)DK×DK×M×N(3×3×3×2)

计算量：𝐷𝐾⋅𝐷𝐾⋅𝑀⋅𝑁⋅𝐷𝐹⋅𝐷𝐹DK⋅DK⋅M⋅N⋅DF⋅DF

用depth-wise separable conv来替代std conv，depth-wise conv分解为depthwise conv和pointwise conv。

std conv输出的每个通道的feature包含了输入所有通道的feature，depth-wise separable conv没有办法做到，所以需要用pointwise conv来结合不同通道的feature。

Depthwise Conv

对输入feature的每个通道单独做卷积操作，得到每个通道对应的输出feature。

参数量：𝐷𝐾×𝐷𝐾×𝑀(3×3×3)DK×DK×M(3×3×3)

计算量：𝐷𝐾⋅𝐷𝐾⋅𝑀⋅𝐷𝐹⋅𝐷𝐹DK⋅DK⋅M⋅DF⋅DF

Pointwise Conv

将depthwise conv的输出，即不同通道的feature map结合起来，从而达到和std conv一样的效果。

v2-b2f4c69bc63c2de4728039d409573e6f_r[1]

参数量：1×1×𝑀×𝑁(1×1×3×2)1×1×M×N(1×1×3×2)

计算量：𝑀⋅𝑁⋅𝐷𝐹⋅𝐷𝐹M⋅N⋅DF⋅DF

从而总计算量为𝐷𝐾⋅𝐷𝐾⋅𝑀⋅𝐷𝐹⋅𝐷𝐹+𝑀⋅ 𝑁⋅𝐷𝐹⋅𝐷𝐹DK⋅DK⋅M⋅DF⋅DF+M⋅ N⋅DF⋅DF

通过拆分，相当于将standard conv计算量压缩为：

1558320022812

代码实现

BasicConv2d & DepthSeperableConv2d

Copy

class DepthSeperabelConv2d(nn.Module):
    def __init__(self, input_channels, output_channels, kernel_size, **kwargs): super().__init__() self.depthwise = nn.Sequential( nn.Conv2d( input_channels, input_channels, kernel_size, groups=input_channels, **kwargs), nn.BatchNorm2d(input_channels), nn.ReLU(inplace=True) ) self.pointwise = nn.Sequential( nn.Conv2d(input_channels, output_channels, 1), nn.BatchNorm2d(output_channels), nn.ReLU(inplace=True) ) def forward(self, x): x = self.depthwise(x) x = self.pointwise(x) return x class BasicConv2d(nn.Module): def __init__(self, input_channels, output_channels, kernel_size, **kwargs): super().__init__() self.conv = nn.Conv2d( input_channels, output_channels, kernel_size, **kwargs) self.bn = nn.BatchNorm2d(output_channels) self.relu = nn.ReLU(inplace=True) def forward(self, x): x = self.conv(x) x = self.bn(x) x = self.relu(x) return x

Two hyper-parameters

Width Multiplier 𝛼α：以系数1,0.75,0.5和0.251,0.75,0.5和0.25乘以input、output channel

计算量变为𝐷𝐾⋅𝐷𝐾⋅𝛼𝑀⋅𝐷𝐹⋅𝐷𝐹+𝛼𝑀⋅ 𝛼𝑁⋅𝐷𝐹⋅𝐷𝐹DK⋅DK⋅αM⋅DF⋅DF+αM⋅ αN⋅DF⋅DF

Resoltion Multiplier 𝜌ρ：将输入分辨率变为224,192,160或128224,192,160或128。

计算量变为𝐷𝐾⋅𝐷𝐾⋅𝛼𝑀⋅𝜌𝐷𝐹⋅𝜌𝐷𝐹+𝛼𝑀⋅ 𝛼𝑁⋅𝜌𝐷𝐹⋅𝜌𝐷𝐹DK⋅DK⋅αM⋅ρDF⋅ρDF+αM⋅ αN⋅ρDF⋅ρDF

网络架构#

1558488904195

Copy

def mobilenet(alpha=1, class_num=100): return MobileNet(alpha, class_num) class MobileNet(nn.Module): """ Args: width multipler: The role of the width multiplier α is to thin a network uniformly at each layer. For a given layer and width multiplier α, the number of input channels M becomes αM and the number of output channels N becomes αN. """ def __init__(self, width_multiplier=1, class_num=100): super().__init__() alpha = width_multiplier self.stem = nn.Sequential( BasicConv2d(3, int(32 * alpha), 3, padding=1, bias=False), DepthSeperabelConv2d( int(32 * alpha), int(64 * alpha), 3, padding=1, bias=False ) ) #downsample self.conv1 = nn.Sequential( DepthSeperabelConv2d( int(64 * alpha), int(128 * alpha), 3, stride=2, padding=1, bias=False ), DepthSeperabelConv2d( int(128 * alpha), int(128 * alpha), 3, padding=1, bias=False ) ) #downsample self.conv2 = nn.Sequential( DepthSeperabelConv2d( int(128 * alpha), int(256 * alpha), 3, stride=2, padding=1, bias=False ), DepthSeperabelConv2d( int(256 * alpha), int(256 * alpha), 3, padding=1, bias=False ) ) #downsample self.conv3 = nn.Sequential( DepthSeperabelConv2d( int(256 * alpha), int(512 * alpha), 3, stride=2, padding=1, bias=False ), DepthSeperabelConv2d( int(512 * alpha), int(512 * alpha), 3, padding=1, bias=False ), DepthSeperabelConv2d( int(512 * alpha), int(512 * alpha), 3, padding=1, bias=False ), DepthSeperabelConv2d( int(512 * alpha), int(512 * alpha), 3, padding=1, bias=False ), DepthSeperabelConv2d( int(512 * alpha), int(512 * alpha), 3, padding=1, bias=False ), DepthSeperabelConv2d( int(512 * alpha), int(512 * alpha), 3, padding=1, bias=False ) ) #downsample self.conv4 = nn.Sequential( DepthSeperabelConv2d( int(512 * alpha), int(1024 * alpha), 3, stride=2, padding=1, bias=False ), DepthSeperabelConv2d( int(1024 * alpha), int(1024 * alpha), 3, padding=1, bias=False ) ) self.fc = nn.Linear(int(1024 * alpha), class_num) self.avg = nn.AdaptiveAvgPool2d(1) def forward(self, x): x = self.stem(x) x = self.conv1(x) x = self.conv2(x) x = self.conv3(x) x = self.conv4(x) x = self.avg(x) x = x.view(x.size(0), -1) x = self.fc(x) return x

实验结果#

1558490292227

MobileNetV2#

核心思想#

Inverted residual block：引入残差结构和bottleneck层。
Linear Bottlenecks：ReLU会破坏信息，故去掉第二个Conv1x1后的ReLU，改为线性神经元。

1558663130590

Expansion and projection

MobileNetv2与其他网络对比

1558661554255

MobileNetV2 block

代码实现

Copy

class LinearBottleNeck(nn.Module):
    def __init__(self, in_channels, out_channels, stride, t=6, class_num=100): super().__init__() self.residual = nn.Sequential( nn.Conv2d(in_channels, in_channels * t, 1), nn.BatchNorm2d(in_channels * t), nn.ReLU6(inplace=True), nn.Conv2d(in_channels * t, in_channels * t, 3, stride=stride, padding=1, groups=in_channels * t), nn.BatchNorm2d(in_channels * t), nn.ReLU6(inplace=True), nn.Conv2d(in_channels * t, out_channels, 1), nn.BatchNorm2d(out_channels) ) self.stride = stride self.in_channels = in_channels self.out_channels = out_channels def forward(self, x): residual = self.residual(x) if self.stride == 1 and self.in_channels == self.out_channels: residual += x return residual

网络架构#

1558662977335

Copy

class MobileNetV2(nn.Module):
    def __init__(self, class_num=100): super().__init__() self.pre = nn.Sequential( nn.Conv2d(3, 32, 1, padding=1), nn.BatchNorm2d(32), nn.ReLU6(inplace=True) ) self.stage1 = LinearBottleNeck(32, 16, 1, 1) self.stage2 = self._make_stage(2, 16, 24, 2, 6) self.stage3 = self._make_stage(3, 24, 32, 2, 6) self.stage4 = self._make_stage(4, 32, 64, 2, 6) self.stage5 = self._make_stage(3, 64, 96, 1, 6) self.stage6 = self._make_stage(3, 96, 160, 1, 6) self.stage7 = LinearBottleNeck(160, 320, 1, 6) self.conv1 = nn.Sequential( nn.Conv2d(320, 1280, 1), nn.BatchNorm2d(1280), nn.ReLU6(inplace=True) ) self.conv2 = nn.Conv2d(1280, class_num, 1) def forward(self, x): x = self.pre(x) x = self.stage1(x) x = self.stage2(x) x = self.stage3(x) x = self.stage4(x) x = self.stage5(x) x = self.stage6(x) x = self.stage7(x) x = self.conv1(x) x = F.adaptive_avg_pool2d(x, 1) x = self.conv2(x) x = x.view(x.size(0), -1) return x def _make_stage(self, repeat, in_channels, out_channels, stride, t): layers = [] layers.append(LinearBottleNeck(in_channels, out_channels, stride, t)) while repeat - 1: layers.append(LinearBottleNeck(out_channels, out_channels, 1, t)) repeat -= 1 return nn.Sequential(*layers) def mobilenetv2(): return MobileNetV2()

实验结果#

1558662995407

ShuffleNetV1#

核心思想#

利用group convolution和channel shuffle来减少模型参数量。

1558490545399

ShuffleNet unit

从ResNet bottleneck 演化得到shuffleNet unit

(a)带depth-wise conv的bottleneck unit
(b)将1x1conv换成1x1Gconv，并在第一个1x1Gconv后增加一个channel shuffle。
(c)旁路增加AVG pool，减小feature map的分辨率；分辨率小了，最后不采用add而是concat，从而弥补分辨率减小带来的信息损失。

1558490920073

代码实现

Copy

class ChannelShuffle(nn.Module):

    def __init__(self, groups): super().__init__() self.groups = groups def forward(self, x): batchsize, channels, height, width = x.data.size() channels_per_group = int(channels / self.groups) #"""suppose a convolutional layer with g groups whose output has #g x n channels; we first reshape the output channel dimension #into (g, n)""" x = x.view(batchsize, self.groups, channels_per_group, height, width) #"""transposing and then flattening it back as the input of next layer.""" x = x.transpose(1, 2).contiguous() x = x.view(batchsize, -1, height, width) return x class ShuffleNetUnit(nn.Module): def __init__(self, input_channels, output_channels, stage, stride, groups): super().__init__() #"""Similar to [9], we set the number of bottleneck channels to 1/4 #of the output channels for each ShuffleNet unit.""" self.bottlneck = nn.Sequential( PointwiseConv2d( input_channels, int(output_channels / 4), groups=groups ), nn.ReLU(inplace=True) ) #"""Note that for Stage 2, we do not apply group convolution on the first pointwise #layer because the number of input channels is relatively small.""" if stage == 2: self.bottlneck = nn.Sequential( PointwiseConv2d( input_channels, int(output_channels / 4), groups=groups ), nn.ReLU(inplace=True) ) self.channel_shuffle = ChannelShuffle(groups) self.depthwise = DepthwiseConv2d( int(output_channels / 4), int(output_channels / 4), 3, groups=int(output_channels / 4), stride=stride, padding=1 ) self.expand = PointwiseConv2d( int(output_channels / 4), output_channels, groups=groups ) self.relu = nn.ReLU(inplace=True) self.fusion = self._add self.shortcut = nn.Sequential() #"""As for the case where ShuffleNet is applied with stride, #we simply make two modifications (see Fig 2 (c)): #(i) add a 3 × 3 average pooling on the shortcut path; #(ii) replace the element-wise addition with channel concatenation, #which makes it easy to enlarge channel dimension with little extra #computation cost. if stride != 1 or input_channels != output_channels: self.shortcut = nn.AvgPool2d(3, stride=2, padding=1) self.expand = PointwiseConv2d( int(output_channels / 4), output_channels - input_channels, groups=groups ) self.fusion = self._cat def _add(self, x, y): return torch.add(x, y) def _cat(self, x, y): return torch.cat([x, y], dim=1) def forward(self, x): shortcut = self.shortcut(x) shuffled = self.bottlneck(x) shuffled = self.channel_shuffle(shuffled) shuffled = self.depthwise(shuffled) shuffled = self.expand(shuffled) output = self.fusion(shortcut, shuffled) output = self.relu(output) return output

网络架构#

1558490577412

代码实现

Copy

class ShuffleNet(nn.Module):

    def __init__(self, num_blocks, num_classes=100, groups=3): super().__init__() if groups == 1: out_channels = [24, 144, 288, 567] elif groups == 2: out_channels = [24, 200, 400, 800] elif groups == 3: out_channels = [24, 240, 480, 960] elif groups == 4: out_channels = [24, 272, 544, 1088] elif groups == 8: out_channels = [24, 384, 768, 1536] self.conv1 = BasicConv2d(3, out_channels[0], 3, padding=1, stride=1) self.input_channels = out_channels[0] self.stage2 = self._make_stage( ShuffleNetUnit, num_blocks[0], out_channels[1], stride=2, stage=2, groups=groups ) self.stage3 = self._make_stage( ShuffleNetUnit, num_blocks[1], out_channels[2], stride=2, stage=3, groups=groups ) self.stage4 = self._make_stage( ShuffleNetUnit, num_blocks[2], out_channels[3], stride=2, stage=4, groups=groups ) self.avg = nn.AdaptiveAvgPool2d((1, 1)) self.fc = nn.Linear(out_channels[3], num_classes) def forward(self, x): x = self.conv1(x) x = self.stage2(x) x = self.stage3(x) x = self.stage4(x) x = self.avg(x) x = x.view(x.size(0), -1) x = self.fc(x) return x def _make_stage(self, block, num_blocks, output_channels, stride, stage, groups): """make shufflenet stage Args: block: block type, shuffle unit out_channels: output depth channel number of this stage num_blocks: how many blocks per stage stride: the stride of the first block of this stage stage: stage index groups: group number of group convolution Return: return a shuffle net stage """ strides = [stride] + [1] * (num_blocks - 1) stage = [] for stride in strides: stage.append( block( self.input_channels, output_channels, stride=stride, stage=stage, groups=groups ) ) self.input_channels = output_channels return nn.Sequential(*stage) def shufflenet(): return ShuffleNet([4, 8, 4])

实验结果#

1558490727956

ShuffleNetV2#

核心思想#

基于四条准则，改进了SuffleNetv1

G1）同等通道最小化内存访问量（1x1卷积平衡输入和输出通道大小）

G2）过量使用组卷积增加内存访问量（谨慎使用组卷积）

G3）网络碎片化降低并行度（避免网络碎片化）

G4）不能忽略元素级操作（减少元素级运算）

1558664291379

代码实现

Copy

def channel_split(x, split):
    """split a tensor into two pieces along channel dimension Args: x: input tensor split:(int) channel size for each pieces """ assert x.size(1) == split * 2 return torch.split(x, split, dim=1) def channel_shuffle(x, groups): """channel shuffle operation Args: x: input tensor groups: input branch number """ batch_size, channels, height, width = x.size() channels_per_group = int(channels / groups) x = x.view(batch_size, groups, channels_per_group, height, width) x = x.transpose(1, 2).contiguous() x = x.view(batch_size, -1, height, width) return x class ShuffleUnit(nn.Module): def __init__(self, in_channels, out_channels, stride): super().__init__() self.stride = stride self.in_channels = in_channels self.out_channels = out_channels if stride != 1 or in_channels != out_channels: self.residual = nn.Sequential( nn.Conv2d(in_channels, in_channels, 1), nn.BatchNorm2d(in_channels), nn.ReLU(inplace=True), nn.Conv2d(in_channels, in_channels, 3, stride=stride, padding=1, groups=in_channels), nn.BatchNorm2d(in_channels), nn.Conv2d(in_channels, int(out_channels / 2), 1), nn.BatchNorm2d(int(out_channels / 2)), nn.ReLU(inplace=True) ) self.shortcut = nn.Sequential( nn.Conv2d(in_channels, in_channels, 3, stride=stride, padding=1, groups=in_channels), nn.BatchNorm2d(in_channels), nn.Conv2d(in_channels, int(out_channels / 2), 1), nn.BatchNorm2d(int(out_channels / 2)), nn.ReLU(inplace=True) ) else: self.shortcut = nn.Sequential() in_channels = int(in_channels / 2) self.residual = nn.Sequential( nn.Conv2d(in_channels, in_channels, 1), nn.BatchNorm2d(in_channels), nn.ReLU(inplace=True), nn.Conv2d(in_channels, in_channels, 3, stride=stride, padding=1, groups=in_channels), nn.BatchNorm2d(in_channels), nn.Conv2d(in_channels, in_channels, 1), nn.BatchNorm2d(in_channels), nn.ReLU(inplace=True) ) def forward(self, x): if self.stride == 1 and self.out_channels == self.in_channels: shortcut, residual = channel_split(x, int(self.in_channels / 2)) else: shortcut = x residual = x shortcut = self.shortcut(shortcut) residual = self.residual(residual) x = torch.cat([shortcut, residual], dim=1) x = channel_shuffle(x, 2) return x

网络架构#

1558664321907

Copy

class ShuffleNetV2(nn.Module):

    def __init__(self, ratio=1, class_num=100): super().__init__() if ratio == 0.5: out_channels = [48, 96, 192, 1024] elif ratio == 1: out_channels = [116, 232, 464, 1024] elif ratio == 1.5: out_channels = [176, 352, 704, 1024] elif ratio == 2: out_channels = [244, 488, 976, 2048] else: ValueError('unsupported ratio number') self.pre = nn.Sequential( nn.Conv2d(3, 24, 3, padding=1), nn.BatchNorm2d(24) ) self.stage2 = self._make_stage(24, out_channels[0], 3) self.stage3 = self._make_stage(out_channels[0], out_channels[1], 7) self.stage4 = self._make_stage(out_channels[1], out_channels[2], 3) self.conv5 = nn.Sequential( nn.Conv2d(out_channels[2], out_channels[3], 1), nn.BatchNorm2d(out_channels[3]), nn.ReLU(inplace=True) ) self.fc = nn.Linear(out_channels[3], class_num) def forward(self, x): x = self.pre(x) x = self.stage2(x) x = self.stage3(x) x = self.stage4(x) x = self.conv5(x) x = F.adaptive_avg_pool2d(x, 1) x = x.view(x.size(0), -1) x = self.fc(x) return x def _make_stage(self, in_channels, out_channels, repeat): layers = [] layers.append(ShuffleUnit(in_channels, out_channels, 2)) while repeat: layers.append(ShuffleUnit(out_channels, out_channels, 1)) repeat -= 1 return nn.Sequential(*layers) def shufflenetv2(): return ShuffleNetV2()