pytorch深度学习实践（刘二大人）课堂代码&作业——卷积神经网络高级篇

一、课堂代码

1.GoogleNet

主要是加了inception模块，这部分代码如下，把上节课代码的网络换掉就行了

#GoogleNet
class InceptionA(torch.nn.Module):
    def __init__(self, in_channels):
        super(InceptionA, self).__init__()
        self.branch1x1 = torch.nn.Conv2d(in_channels, 16, kernel_size=1)

        self.branch5x5_1 = torch.nn.Conv2d(in_channels, 16,kernel_size=1)
        self.branch5x5_2 = torch.nn.Conv2d(16, 24,kernel_size=5, padding=2)

        self.branch3x3_1 = torch.nn.Conv2d(in_channels, 16, kernel_size=1)
        self.branch3x3_2 = torch.nn.Conv2d(16, 24, kernel_size=3, padding=1)
        self.branch3x3_3 = torch.nn.Conv2d(24, 24, kernel_size=3, padding=1)

        self.branch_pool = torch.nn.Conv2d(in_channels, 24, kernel_size=1)
        
    def forward(self, x):
        branch1x1 = self.branch1x1(x)   #16
        
        branch5x5 = self.branch5x5_1(x)  
        branch5x5 = self.branch5x5_2(branch5x5)  #24

        branch3x3 = self.branch3x3_1(x)
        branch3x3 = self.branch3x3_2(branch3x3)
        branch3x3 = self.branch3x3_3(branch3x3)   #24

        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1)
        branch_pool = self.branch_pool(branch_pool)   #24

        outputs = [branch1x1, branch5x5, branch3x3, branch_pool]
        return torch.cat(outputs, dim=1)  #在channel维度上concatenate, 16+24*3=88


class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = torch.nn.Conv2d(1, 10, kernel_size=5)
        self.conv2 = torch.nn.Conv2d(88, 20, kernel_size=5)

        self.incept1 = InceptionA(in_channels=10)
        self.incept2 = InceptionA(in_channels=20)

        self.mp = torch.nn.MaxPool2d(2)
        self.fc = torch.nn.Linear(1408, 10)
    
    def forward(self, x):
        in_size = x.size(0)
        x = F.relu(self.mp(self.conv1(x)))
        x = self.incept1(x)
        x = F.relu(self.mp(self.conv2(x)))
        x = self.incept2(x)
        x = x.view(in_size,-1)
        x = self.fc(x)
        return x

输出结果：

2.ResNet

和上面一样，换网络就好了。后续有个换ResidualBlock的作业，我把ppt上的两个结合起来直接在这里换了

#conv shortcut
class ResidualBlock(torch.nn.Module):
    def __init__(self, channels):
        super(ResidualBlock, self).__init__()
        self.channels = channels
        self.conv = torch.nn.Conv2d(channels, channels, kernel_size=1)
        self.conv1 = torch.nn.Conv2d(channels, channels, kernel_size=3, padding=1)
        self.conv2 = torch.nn.Conv2d(channels, channels, kernel_size=3, padding=1)

    def forward(self, x):
        y = F.relu(self.conv1(x))
        y = self.conv2(y)
        x_hat = self.conv(x)
        return F.relu(0.5*x_hat+0.5*y)
    
class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = torch.nn.Conv2d(1, 16, kernel_size=5)
        self.conv2 = torch.nn.Conv2d(16, 32, kernel_size=5)
        self.mp = torch.nn.MaxPool2d(2)

        self.rblock1 = ResidualBlock(16)
        self.rblock2 = ResidualBlock(32)

        self.fc = torch.nn.Linear(512, 10)

    def forward(self, x):
        in_size = x.size(0)
        x = self.mp(F.relu(self.conv1(x)))
        x = self.rblock1(x)
        x = self.mp(F.relu(self.conv2(x)))
        x = self.rblock2(x)
        x = x.view(in_size, -1)
        x = self.fc(x)
        return x

结果：

二、课后作业

实现DenseNet

DenseNet理论学习参考链接：DenseNet：比ResNet更优的CNN模型 - 知乎 (zhihu.com)

参照pytorch官方实现搭了一个比较简单的DenseNet用在mnist数据集上，有两个DenseBlock，每个block里有4层Denselayer，两个block中间有一个transition

#DenseNet
class _DenseLayer(nn.Sequential):
    """Basic unit of DenseBlock (using bottleneck layer) """
    def __init__(self, num_input_features, growth_rate, bn_size, drop_rate): #特征的w和h不变，c维度固定为growth_rate
        super(_DenseLayer, self).__init__()                                  # 中间用了一个1*1卷积降低特征数量，用bn_size控制c维度，其实也就是注释里的bottleneck layer
        self.add_module("norm1", nn.BatchNorm2d(num_input_features))
        self.add_module("relu1", nn.ReLU(inplace=True))
        self.add_module("conv1", nn.Conv2d(num_input_features, bn_size*growth_rate,
                                           kernel_size=1, stride=1, bias=False))
        self.add_module("norm2", nn.BatchNorm2d(bn_size*growth_rate))
        self.add_module("relu2", nn.ReLU(inplace=True))
        self.add_module("conv2", nn.Conv2d(bn_size*growth_rate, growth_rate,
                                           kernel_size=3, stride=1, padding=1, bias=False))
        self.drop_rate = drop_rate

    def forward(self, x):   #前向传播，有dropout则dropout，没有就直接把新得到的特征和之前的输入concat到一起，得到下一层的输入
        new_features = super(_DenseLayer, self).forward(x)
        if self.drop_rate > 0:
            new_features = F.dropout(new_features, p=self.drop_rate, training=self.training)
        return torch.cat([x, new_features], 1)
    
class _DenseBlock(nn.Sequential):
    """DenseBlock
    num_layers: block里包含的DenseLayer数量
    num_input_features： 输入特征的c维度，最开始是num_input_features，每多一层多加growth_rate
    """
    def __init__(self, num_layers, num_input_features, bn_size, growth_rate, drop_rate):
        super(_DenseBlock, self).__init__()
        for i in range(num_layers):
            layer = _DenseLayer(num_input_features+i*growth_rate, growth_rate, bn_size,
                                drop_rate)
            self.add_module("denselayer%d" % (i+1,), layer)

class _Transition(nn.Sequential):  #连接两个denseblock，降低特征图大小，压缩channels（通过控制模型里的compression_rate来实现，模块里没有定义压缩的功能）
    """Transition layer between two adjacent DenseBlock"""
    def __init__(self, num_input_feature, num_output_features):
        super(_Transition, self).__init__()
        self.add_module("norm", nn.BatchNorm2d(num_input_feature))
        self.add_module("relu", nn.ReLU(inplace=True))
        self.add_module("conv", nn.Conv2d(num_input_feature, num_output_features,
                                          kernel_size=1, stride=1, bias=False))
        self.add_module("pool", nn.AvgPool2d(2, stride=2))

class DenseNet(nn.Module):
    "DenseNet-BC model"
    def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16), num_init_features=64,
                 bn_size=4, compression_rate=0.5, drop_rate=0, num_classes=1000):
        """
        :param growth_rate: (int) number of filters used in DenseLayer, `k` in the paper
        :param block_config: (list of 4 ints) number of layers in each DenseBlock
        :param num_init_features: (int) number of filters in the first Conv2d
        :param bn_size: (int) the factor using in the bottleneck layer
        :param compression_rate: (float) the compression rate used in Transition Layer
        :param drop_rate: (float) the drop rate after each DenseLayer
        :param num_classes: (int) number of classes for classification
        """
        super(DenseNet, self).__init__()
        # first Conv2d
        self.features = nn.Sequential(OrderedDict([
            ("conv0", nn.Conv2d(1, num_init_features, kernel_size=3, padding=1, bias=False)),
            ("norm0", nn.BatchNorm2d(num_init_features)),
            ("relu0", nn.ReLU(inplace=True)),
            ("pool0", nn.MaxPool2d(2))
        ]))

        # DenseBlock
        num_features = num_init_features
        for i, num_layers in enumerate(block_config):
            block = _DenseBlock(num_layers, num_features, bn_size, growth_rate, drop_rate)
            self.features.add_module("denseblock%d" % (i + 1), block)
            num_features += num_layers*growth_rate
            if i != len(block_config) - 1:  #在denseblock中间插入transition
                transition = _Transition(num_features, int(num_features*compression_rate))
                self.features.add_module("transition%d" % (i + 1), transition)
                num_features = int(num_features * compression_rate)

        # final bn+ReLU
        self.features.add_module("norm5", nn.BatchNorm2d(num_features))
        self.features.add_module("relu5", nn.ReLU(inplace=True))

        # classification layer
        self.classifier = nn.Linear(2600, num_classes)

        # params initialization
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight)
            elif isinstance(m, nn.BatchNorm2d):
                nn.init.constant_(m.bias, 0)
                nn.init.constant_(m.weight, 1)
            elif isinstance(m, nn.Linear):
                nn.init.constant_(m.bias, 0)

    def forward(self, x):
        features = self.features(x)  #1*28*28 →first Conv→ 64*14*14 →denseblock1→ (64+4*12)*14*14 →transtition→ (0.5*102)*7*7 →denseblock2→ (51+4*12)*7*7 →avrgpooling→ 99*5*5
        out = F.avg_pool2d(features, 3, stride=1).view(features.size(0), -1)
        out = self.classifier(out)
        return out
    

model = DenseNet(growth_rate=12, block_config=(4, 4), bn_size=4, compression_rate=0.5, num_classes=10)

结果：