mobileNet

mobileNetV1

1.背景

• 传统卷积神经网络，内存需求大,运算量大，导致无法在移动设备以及嵌入式设备上运行。
  
• MobileNet网络是由google团队在2017年提出的，专注于移动端或者嵌入式设备中的轻量级CNN网络。相比传统卷积神经网络，在准确率小幅降低的前提下大大减少模型参数与运算量。(相比VGG16准确率减少了0.9%，但模型参数只有VGG的1/32)
  
• 论文原文:
  
  

2.网络中的亮点

• Depthwise Convolution(大大减少运算量和参数数量)
  
• 增加超参数α、β
  

3.传统卷积和DW卷积区别

• 1.传统卷积
  
  

卷积核channel=输入特征矩阵channel

输出特征矩阵channel=卷积核个数

• 2.DW卷积
  
  

卷积核channel=1

输入特征矩阵channel=卷积核个数=输出特征矩阵channel

4.PW卷积和DW卷积

• 1.DW卷积
  
  

卷积核channel=1

输入特征矩阵channel=卷积核个数=输出特征矩阵channel

• 2.PW卷积
  
  

• 3.MobileNet中的DW卷积和PW卷积
  
  

5.MobileNetV1计算量及网络结构

6.MobileNetV1局限性

• 没有残差连接
  
• 很多Depthwise卷积核训练出来是0：
  
• (1)卷积核权重数量和通道数量太少
• (2)太单薄
• (3)RELU
• (4)低精度

# MobileNetV2 # ## 1.网络背景 ## - MobileNet v2网络是由google团队在2018年提出的，相比MobileNet V1网络，准确率更高，模型更小。
- 论文原文:

2.网络亮点

• Inverted Residuals（倒残差结构）
• Linear Bottlenecks
  

第一个1×1卷积表示升维，第二个1×1卷积表示降维

3.网络详解

• 1.残差块

• （1）1×1卷积降维

• （2）3×3卷积

• （3）1×1卷积升维

• 2.倒残差块

• （1）1×1卷积升维

• （2）3×3卷积

• （3）1×1卷积降维
  

• 3.RELU6
  

RELU激活函数对低维特征信息造成大量损失

4.倒残差模块

当stirde=1且输入特征矩阵与输出特征矩阵shape相同时才有shortcut连接。

5.MobileNetV2参数结构

• (1)t是扩展因子
  
• (2)c是输出特征矩阵
  
• (3)n是bottleneck的重复次数
  
• (4)s是步距(针对第一层，其他层均为1)
  

6.MobileNetV2代码详解

from torch import nn
import torch


def _make_divisible(ch, divisor=8, min_ch=None):
    """
    This function is taken from the original tf repo.
    It ensures that all layers have a channel number that is divisible by 8
    It can be seen here:
    https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet/mobilenet.py
    """
    if min_ch is None:
        min_ch = divisor
    new_ch = max(min_ch, int(ch + divisor / 2) // divisor * divisor)
    # Make sure that round down does not go down by more than 10%.
    if new_ch < 0.9 * ch:
        new_ch += divisor
    return new_ch


class ConvBNReLU(nn.Sequential):
    def __init__(self, in_channel, out_channel, kernel_size=3, stride=1, groups=1):
        padding = (kernel_size - 1) // 2
        super(ConvBNReLU, self).__init__(
            nn.Conv2d(in_channel, out_channel, kernel_size, stride, padding, groups=groups, bias=False),
            nn.BatchNorm2d(out_channel),
            nn.ReLU6(inplace=True)
        )


class InvertedResidual(nn.Module):
    def __init__(self, in_channel, out_channel, stride, expand_ratio):
        super(InvertedResidual, self).__init__()
        hidden_channel = in_channel * expand_ratio
        self.use_shortcut = stride == 1 and in_channel == out_channel

        layers = []
        if expand_ratio != 1:
            # 1x1 pointwise conv
            layers.append(ConvBNReLU(in_channel, hidden_channel, kernel_size=1))
        layers.extend([
            # 3x3 depthwise conv
            ConvBNReLU(hidden_channel, hidden_channel, stride=stride, groups=hidden_channel),
            # 1x1 pointwise conv(linear)
            nn.Conv2d(hidden_channel, out_channel, kernel_size=1, bias=False),
            nn.BatchNorm2d(out_channel),
        ])

        self.conv = nn.Sequential(*layers)

    def forward(self, x):
        if self.use_shortcut:
            return x + self.conv(x)
        else:
            return self.conv(x)


class MobileNetV2(nn.Module):
    def __init__(self, num_classes=1000, alpha=1.0, round_nearest=8):
        super(MobileNetV2, self).__init__()
        block = InvertedResidual
        input_channel = _make_divisible(32 * alpha, round_nearest)
        last_channel = _make_divisible(1280 * alpha, round_nearest)

        inverted_residual_setting = [
            # t, c, n, s
            [1, 16, 1, 1],
            [6, 24, 2, 2],
            [6, 32, 3, 2],
            [6, 64, 4, 2],
            [6, 96, 3, 1],
            [6, 160, 3, 2],
            [6, 320, 1, 1],
        ]

        features = []
        # conv1 layer
        features.append(ConvBNReLU(3, input_channel, stride=2))
        # building inverted residual residual blockes
        for t, c, n, s in inverted_residual_setting:
            output_channel = _make_divisible(c * alpha, round_nearest)
            for i in range(n):
                stride = s if i == 0 else 1
                features.append(block(input_channel, output_channel, stride, expand_ratio=t))
                input_channel = output_channel
        # building last several layers
        features.append(ConvBNReLU(input_channel, last_channel, 1))
        # combine feature layers
        self.features = nn.Sequential(*features)

        # building classifier
        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
        self.classifier = nn.Sequential(
            nn.Dropout(0.2),
            nn.Linear(last_channel, num_classes)
        )

        # weight initialization
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight, mode='fan_out')
                if m.bias is not None:
                    nn.init.zeros_(m.bias)
            elif isinstance(m, nn.BatchNorm2d):
                nn.init.ones_(m.weight)
                nn.init.zeros_(m.bias)
            elif isinstance(m, nn.Linear):
                nn.init.normal_(m.weight, 0, 0.01)
                nn.init.zeros_(m.bias)

    def forward(self, x):
        x = self.features(x)
        x = self.avgpool(x)
        x = torch.flatten(x, 1)
        x = self.classifier(x)
        return x