经典卷积模型(四)GoogLeNet-Inception(V3)代码解析

最新推荐文章于 2024-05-15 11:27:48 发布

nooobme

最新推荐文章于 2024-05-15 11:27:48 发布

阅读量3.2k

点赞数 7

分类专栏：深度学习文章标签： pytorch 深度学习 python

本文链接：https://blog.csdn.net/sleepinghm/article/details/120097994

版权

深度学习专栏收录该内容

30 篇文章 7 订阅

订阅专栏

Inception-V3

网络主干依旧由Inception、辅助分类器构成，其中Inception有六类。

BasicConv2d 基本卷积模块

BasicConv2d为带BN+relu的卷积。

class BasicConv2d(nn.Module):
    def __init__(self, in_channels, out_channels, **kwargs):
        super(BasicConv2d, self).__init__()
        self.conv = nn.Conv2d(in_channels, out_channels, bias=False, **kwargs)
        self.bn = nn.BatchNorm2d(out_channels, eps=0.001)
    def forward(self, x):
        x = self.conv(x)
        x = self.bn(x)
        return F.relu(x, inplace=True)

Inception部分

pytorch提供的有六种基本的inception模块，分别是InceptionA——InceptionE。

InceptionA

在这里插入图片描述
得到输入大小不变，通道数为224+pool_features的特征图。 假如输入为(35, 35, 192)的数据：

第一个brach：
- 经过branch1x1为带有64个1*1的卷积核，所以生成第一张特征图(35, 35, 64)；
第二个brach：
- 首先经过branch5x5_1为带有48个1*1的卷积核，所以第二张特征图(35, 35, 48)，
- 然后经过branch5x5_2为带有64个5*5大小且填充为2的卷积核，特征图大小依旧不变，因此第二张特征图最终为(35, 35, 64)；
第三个brach：
- 首先经过branch3x3dbl_1为带有64个1*1的卷积核，所以第三张特征图(35, 35, 64)，
- 然后经过branch3x3dbl_2为带有96个3*3大小且填充为1的卷积核，特征图大小依旧不变，因此进一步生成第三张特征图(35, 35, 96)，
- 最后经过branch3x3dbl_3为带有96个3*3大小且填充为1的卷积核，特征图大小和通道数不变，因此第三张特征图最终为(35, 35, 96)；
第四个brach：
- 首先经过avg_pool2d，其中池化核3*3，步长为1，填充为1，所以第四张特征图大小不变，通道数不变，第四张特征图为(35, 35, 192)，
- 然后经过branch_pool为带有pool_features个的1*1卷积，因此第四张特征图最终为(35, 35, pool_features)；
最后将四张特征图进行拼接，最终得到(35，35，64+64+96+pool_features)的特征图。

class InceptionA(nn.Module):
    def __init__(self, in_channels, pool_features):
        super(InceptionA, self).__init__()
        self.branch1x1 = BasicConv2d(in_channels, 64, kernel_size=1)
        self.branch5x5_1 = BasicConv2d(in_channels, 48, kernel_size=1)
        self.branch5x5_2 = BasicConv2d(48, 64, kernel_size=5, padding=2)
        self.branch3x3dbl_1 = BasicConv2d(in_channels, 64, kernel_size=1)
        self.branch3x3dbl_2 = BasicConv2d(64, 96, kernel_size=3, padding=1)
        self.branch3x3dbl_3 = BasicConv2d(96, 96, kernel_size=3, padding=1)
        self.branch_pool = BasicConv2d(in_channels, pool_features, kernel_size=1)
    def forward(self, x):
        branch1x1 = self.branch1x1(x)
        branch5x5 = self.branch5x5_1(x)
        branch5x5 = self.branch5x5_2(branch5x5)
        branch3x3dbl = self.branch3x3dbl_1(x)
        branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)
        branch3x3dbl = self.branch3x3dbl_3(branch3x3dbl)
        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1)
        branch_pool = self.branch_pool(branch_pool)
        outputs = [branch1x1, branch5x5, branch3x3dbl, branch_pool]
        return torch.cat(outputs, 1)

InceptionB

在这里插入图片描述
得到输入大小减半，通道数+480的特征图，假如输入为(35, 35, 288)的数据：

第一个brach：
- 经过branch1x1为带有384个3*3大小且步长2的卷积核，(35-3+2*0)/2+1=17所以生成第一张特征图(17, 17, 384)；
第二个brach：
- 首先经过branch3x3dbl为带有64个1*1的卷积核，特征图大小不变，即(35, 35, 64)；
- 然后经过branch3x3dbl_2为带有96个3*3大小填充1的卷积核，特征图大小不变，即(35, 35, 96)，
- 再经过branch3x3dbl_3为带有96个3*3大小步长2的卷积核，(35-3+2*0)/2+1=17，即第二张特征图为(17, 17, 96)；
第三个brach：
- 经过max_pool2d，池化核大小3*3，步长为2，所以是二倍最大值下采样，通道数保持不变，第三张特征图为(17, 17, 288)；
最后将三张特征图进行拼接，最终得到(17(即Hin/2)，17(即Win/2)，384+96+288(Cin)=768)的特征图。

class InceptionB(nn.Module):
    def __init__(self, in_channels):
        super(InceptionB, self).__init__()
        self.branch3x3 = BasicConv2d(in_channels, 384, kernel_size=3, stride=2)
        self.branch3x3dbl_1 = BasicConv2d(in_channels, 64, kernel_size=1)
        self.branch3x3dbl_2 = BasicConv2d(64, 96, kernel_size=3, padding=1)
        self.branch3x3dbl_3 = BasicConv2d(96, 96, kernel_size=3, stride=2)
    def forward(self, x):
        branch3x3 = self.branch3x3(x)
        branch3x3dbl = self.branch3x3dbl_1(x)
        branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)
        branch3x3dbl = self.branch3x3dbl_3(branch3x3dbl)
        branch_pool = F.max_pool2d(x, kernel_size=3, stride=2)
        outputs = [branch3x3, branch3x3dbl, branch_pool]
        return torch.cat(outputs, 1)

InceptionC

在这里插入图片描述
最终得到输入大小不变，通道数为768的特征图。 假如输入为(17,17, 768)的数据：

第一个branch1x1为带有192个1*1的卷积核，所以生成第一张特征图(17,17, 192)；
第二个brach：
- 首先经过branch7x7_1为带有c7个1*1的卷积核，所以第二张特征图(17,17, c7)，
- 然后经过branch7x7_2为带有c7个1*7大小且填充为0*3的卷积核，特征图大小不变，进一步生成第二张特征图(17,17, c7)，
- 然后经过branch7x7_3为带有192个7*1大小且填充为3*0的卷积核，特征图大小不变，进一步生成第二张特征图(17,17, 192)，因此第二张特征图最终为(17,17, 192)；
第三个brach：
- 首先经过branch7x7dbl_1为带有c7个1*1的卷积核，所以第三张特征图(17,17, c7)，
- 然后经过branch7x7dbl_2为带有c7个7*1大小且填充为3*0的卷积核，特征图大小不变，进一步生成第三张特征图(17,17, c7)，
- 然后经过branch7x7dbl_3为带有c7个1*7大小且填充为0*3的卷积核，特征图大小不变，进一步生成第三张特征图(17,17, c7)，
- 然后经过branch7x7dbl_4为带有c7个7*1大小且填充为3*0的卷积核，特征图大小不变，进一步生成第三张特征图(17,17, c7)，
- 然后经过branch7x7dbl_5为带有192个1*7大小且填充为0*3的卷积核，特征图大小不变，因此第二张特征图最终为(17,17, 192)；
第四个brach：
- 首先经过avg_pool2d，其中池化核3*3，步长为1，填充为1，所以第四张特征图大小不变，通道数不变，第四张特征图为(17,17, 768)，
- 然后经过branch_pool为带有192个的1*1卷积，因此第四张特征图最终为(17,17, 192)；
最后将四张特征图进行拼接，最终得到(17, 17, 192+192+192+192=768)的特征图。

class InceptionC(nn.Module):
    def __init__(self, in_channels, channels_7x7):
        super(InceptionC, self).__init__()
        self.branch1x1 = BasicConv2d(in_channels, 192, kernel_size=1)
        c7 = channels_7x7
        self.branch7x7_1 = BasicConv2d(in_channels, c7, kernel_size=1)
        self.branch7x7_2 = BasicConv2d(c7, c7, kernel_size=(1, 7), padding=(0, 3))
        self.branch7x7_3 = BasicConv2d(c7, 192, kernel_size=(7, 1), padding=(3, 0))
        self.branch7x7dbl_1 = BasicConv2d(in_channels, c7, kernel_size=1)
        self.branch7x7dbl_2 = BasicConv2d(c7, c7, kernel_size=(7, 1), padding=(3, 0))
        self.branch7x7dbl_3 = BasicConv2d(c7, c7, kernel_size=(1, 7), padding=(0, 3))
        self.branch7x7dbl_4 = BasicConv2d(c7, c7, kernel_size=(7, 1), padding=(3, 0))
        self.branch7x7dbl_5 = BasicConv2d(c7, 192, kernel_size=(1, 7), padding=(0, 3))
        self.branch_pool = BasicConv2d(in_channels, 192, kernel_size=1)
    def forward(self, x):
        branch1x1 = self.branch1x1(x)
        branch7x7 = self.branch7x7_1(x)
        branch7x7 = self.branch7x7_2(branch7x7)
        branch7x7 = self.branch7x7_3(branch7x7)
        branch7x7dbl = self.branch7x7dbl_1(x)
        branch7x7dbl = self.branch7x7dbl_2(branch7x7dbl)
        branch7x7dbl = self.branch7x7dbl_3(branch7x7dbl)
        branch7x7dbl = self.branch7x7dbl_4(branch7x7dbl)
        branch7x7dbl = self.branch7x7dbl_5(branch7x7dbl)
        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1)
        branch_pool = self.branch_pool(branch_pool)
        outputs = [branch1x1, branch7x7, branch7x7dbl, branch_pool]
        return torch.cat(outputs, 1)

InceptionD

在这里插入图片描述
得到输入大小减半，通道数+512的特征图，假如输入为(17, 17, 768)的数据：

第一个brach：
- 首先经过branch3x3_1为带有192个1*1的卷积核，所以生成第一张特征图(17, 17, 192)；
- 然后经过branch3x3_2为带有320个3*3大小步长为2的卷积核，(17-3+2*0)/2+1=8，最终第一张特征图(8, 8, 320)；
第二个brach：
- 首先经过branch7x7x3_1为带有192个1*1的卷积核，特征图大小不变，即(17, 17, 192)；
- 然后经过branch7x7x3_2为带有192个1*7大小且填充为0*3的卷积核，特征图大小不变，进一步生成第三张特征图(17,17, 192)；
- 再经过branch7x7x3_3为带有192个7*1大小且填充为3*0的卷积核，特征图大小不变，进一步生成第三张特征图(17,17, 192)；
- 最后经过branch7x7x3_4为带有192个3*3大小步长为2的卷积核，最终第一张特征图(8, 8, 192)；
第三个brach：
- 为max_pool2d，池化核大小3*3，步长为2，所以是二倍最大值下采样，通道数保持不变，第三张特征图为(8, 8, 768)；
最后将三张特征图进行拼接，最终得到(8(即Hin/2)，8(即Win/2)，320+192+768(Cin)=1280)的特征图。

class InceptionD(nn.Module):
    def __init__(self, in_channels):
        super(InceptionD, self).__init__()
        self.branch3x3_1 = BasicConv2d(in_channels, 192, kernel_size=1)
        self.branch3x3_2 = BasicConv2d(192, 320, kernel_size=3, stride=2)
        self.branch7x7x3_1 = BasicConv2d(in_channels, 192, kernel_size=1)
        self.branch7x7x3_2 = BasicConv2d(192, 192, kernel_size=(1, 7), padding=(0, 3))
        self.branch7x7x3_3 = BasicConv2d(192, 192, kernel_size=(7, 1), padding=(3, 0))
        self.branch7x7x3_4 = BasicConv2d(192, 192, kernel_size=3, stride=2)
    def forward(self, x):
        branch3x3 = self.branch3x3_1(x)
        branch3x3 = self.branch3x3_2(branch3x3)
        branch7x7x3 = self.branch7x7x3_1(x)
        branch7x7x3 = self.branch7x7x3_2(branch7x7x3)
        branch7x7x3 = self.branch7x7x3_3(branch7x7x3)
        branch7x7x3 = self.branch7x7x3_4(branch7x7x3)
        branch_pool = F.max_pool2d(x, kernel_size=3, stride=2)
        outputs = [branch3x3, branch7x7x3, branch_pool]
        return torch.cat(outputs, 1)

InceptionE

在这里插入图片描述
最终得到输入大小不变，通道数为2048的特征图。 假如输入为(8,8, 1280)的数据：

第一个brach：
- 首先经过branch1x1为带有320个1*1的卷积核，所以生成第一张特征图(8, 8, 320)；
第二个brach：
- 首先经过branch3x3_1为带有384个1*1的卷积核，所以第二张特征图(8, 8, 384)，
  - 经过分支branch3x3_2a为带有384个1*3大小且填充为0*1的卷积核，特征图大小不变，进一步生成特征图(8,8, 384)，
  - 经过分支branch3x3_2b为带有192个3*1大小且填充为1*0的卷积核，特征图大小不变，进一步生成特征图(8,8, 384)，
- 因此第二张特征图最终为两个分支拼接(8,8, 384+384=768)；
第三个brach：
- 首先经过branch3x3dbl_1为带有448个1*1的卷积核，所以第三张特征图(8,8, 448)，
- 然后经过branch3x3dbl_2为带有384个3*3大小且填充为1的卷积核，特征图大小不变，进一步生成第三张特征图(8,8, 384)，
  - 经过分支branch3x3dbl_3a为带有384个1*3大小且填充为0*1的卷积核，特征图大小不变，进一步生成特征图(8,8, 384)，
  - 经过分支branch3x3dbl_3b为带有384个3*1大小且填充为1*0的卷积核，特征图大小不变，进一步生成特征图(8,8, 384)，
- 因此第三张特征图最终为两个分支拼接(8,8, 384+384=768)；
第四个brach：
- 首先经过avg_pool2d，其中池化核3*3，步长为1，填充为1，所以第四张特征图大小不变，通道数不变，第四张特征图为(8,8, 1280)，
- 然后经过branch_pool为带有192个的1*1卷积，因此第四张特征图最终为(8,8, 192)；
最后将四张特征图进行拼接，最终得到(8, 8, 320+768+768+192=2048)的特征图。

class InceptionE(nn.Module):
    def __init__(self, in_channels):
        super(InceptionE, self).__init__()
        self.branch1x1 = BasicConv2d(in_channels, 320, kernel_size=1)
        self.branch3x3_1 = BasicConv2d(in_channels, 384, kernel_size=1)
        self.branch3x3_2a = BasicConv2d(384, 384, kernel_size=(1, 3), padding=(0, 1))
        self.branch3x3_2b = BasicConv2d(384, 384, kernel_size=(3, 1), padding=(1, 0))
        self.branch3x3dbl_1 = BasicConv2d(in_channels, 448, kernel_size=1)
        self.branch3x3dbl_2 = BasicConv2d(448, 384, kernel_size=3, padding=1)
        self.branch3x3dbl_3a = BasicConv2d(384, 384, kernel_size=(1, 3), padding=(0, 1))
        self.branch3x3dbl_3b = BasicConv2d(384, 384, kernel_size=(3, 1), padding=(1, 0))
        self.branch_pool = BasicConv2d(in_channels, 192, kernel_size=1)
    def forward(self, x):
        branch1x1 = self.branch1x1(x)
        branch3x3 = self.branch3x3_1(x)
        branch3x3 = [
            self.branch3x3_2a(branch3x3),
            self.branch3x3_2b(branch3x3),
        ]
        branch3x3 = torch.cat(branch3x3, 1)
        branch3x3dbl = self.branch3x3dbl_1(x)
        branch3x3dbl = self.branch3x3dbl_2(branch3x3dbl)
        branch3x3dbl = [
            self.branch3x3dbl_3a(branch3x3dbl),
            self.branch3x3dbl_3b(branch3x3dbl),
        ]
        branch3x3dbl = torch.cat(branch3x3dbl, 1)
        branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1)
        branch_pool = self.branch_pool(branch_pool)
        outputs = [branch1x1, branch3x3, branch3x3dbl, branch_pool]
        return torch.cat(outputs, 1)

InceptionAux 辅助分类器

在中间层使用中间层特征+辅助分类器，以便最终的损失函数加入该正则化项，优化参数，以达到提升模型分类效果的作用。
结构：Pool——>1x1Conv——>5x5Conv——>FC

class InceptionAux(nn.Module):
    def __init__(self, in_channels, num_classes):
        super(InceptionAux, self).__init__()
        self.conv0 = BasicConv2d(in_channels, 128, kernel_size=1)
        self.conv1 = BasicConv2d(128, 768, kernel_size=5)
        self.conv1.stddev = 0.01
        self.fc = nn.Linear(768, num_classes)
        self.fc.stddev = 0.001
    def forward(self, x):
        # 17 x 17 x 768
        x = F.avg_pool2d(x, kernel_size=5, stride=3)
        # 5 x 5 x 768
        x = self.conv0(x)
        # 5 x 5 x 128
        x = self.conv1(x)
        # 1 x 1 x 768
        x = x.view(x.size(0), -1)
        # 768
        x = self.fc(x)
        # 1000
        return x

InceptionV3 主要代码

输入(229,229,3)的数据，首先归一化输入，经过5个卷积，2个最大池化层。

if self.transform_input: # 1
   x = x.clone()
   x[:, 0] = x[:, 0] * (0.229 / 0.5) + (0.485 - 0.5) / 0.5
   x[:, 1] = x[:, 1] * (0.224 / 0.5) + (0.456 - 0.5) / 0.5
   x[:, 2] = x[:, 2] * (0.225 / 0.5) + (0.406 - 0.5) / 0.5
# 299 x 299 x 3
x = self.Conv2d_1a_3x3(x) # BasicConv2d(3, 32, kernel_size=3, stride=2)
# 149 x 149 x 32
x = self.Conv2d_2a_3x3(x) #BasicConv2d(32, 32, kernel_size=3)
# 147 x 147 x 32
x = self.Conv2d_2b_3x3(x) #BasicConv2d(32, 64, kernel_size=3, padding=1)
# 147 x 147 x 64
x = F.max_pool2d(x, kernel_size=3, stride=2)
# 73 x 73 x 64
x = self.Conv2d_3b_1x1(x)# BasicConv2d(64, 80, kernel_size=1)
# 73 x 73 x 80
x = self.Conv2d_4a_3x3(x)# BasicConv2d(80, 192, kernel_size=3)
# 71 x 71 x 192
x = F.max_pool2d(x, kernel_size=3, stride=2)
# 35 x 35 x 192

然后经过3个InceptionA结构，1个InceptionB，3个InceptionC，1个InceptionD，2个InceptionE，其中InceptionA，辅助分类器AuxLogits以经过最后一个InceptionC的输出为输入。

InceptionA：得到输入大小不变，通道数为224+pool_features的特征图。
InceptionB：得到输入大小减半，通道数+480的特征图。
InceptionC：得到输入大小不变，通道数为768的特征图。
InceptionD：得到输入大小减半，通道数+512的特征图。
InceptionE：得到输入大小不变，通道数为2048的特征图。

# 35 x 35 x 192
x = self.Mixed_5b(x) # InceptionA(192, pool_features=32)
# 35 x 35 x 256
x = self.Mixed_5c(x) # InceptionA(256, pool_features=64)
# 35 x 35 x 288
x = self.Mixed_5d(x) # InceptionA(288, pool_features=64)
# 35 x 35 x 288
x = self.Mixed_6a(x) # InceptionB(288)
# 17 x 17 x 768
x = self.Mixed_6b(x) #InceptionC(768, channels_7x7=128)
# 17 x 17 x 768
x = self.Mixed_6c(x) # InceptionC(768, channels_7x7=160)
# 17 x 17 x 768
x = self.Mixed_6d(x) # InceptionC(768, channels_7x7=160)
# 17 x 17 x 768
x = self.Mixed_6e(x) # InceptionC(768, channels_7x7=192)
# 17 x 17 x 768 
if self.training and self.aux_logits:
    aux = self.AuxLogits(x) #InceptionAux(768, num_classes)
# 17 x 17 x 768
x = self.Mixed_7a(x) # InceptionD(768)
# 8 x 8 x 1280
x = self.Mixed_7b(x) # InceptionE(1280)
# 8 x 8 x 2048
x = self.Mixed_7c(x) # InceptionE(2048)
# 8 x 8 x 2048

进入分类部分。经过平均池化层+dropout+打平+全连接层输出。

# 8 x 8 x 2048
x = F.avg_pool2d(x, kernel_size=8)
# 1 x 1 x 2048
x = F.dropout(x, training=self.training)
# 1 x 1 x 2048
x = x.view(x.size(0), -1)
# 2048
x = self.fc(x)
# 1000 (num_classes)
if self.training and self.aux_logits:
    return x, aux
return x

代码：

class InceptionV3(nn.Module):
    def __init__(self, num_classes=1000, aux_logits=True, transform_input=False):
        super(InceptionV3, self).__init__()
        self.aux_logits = aux_logits
        self.transform_input = transform_input
        self.Conv2d_1a_3x3 = BasicConv2d(3, 32, kernel_size=3, stride=2)
        self.Conv2d_2a_3x3 = BasicConv2d(32, 32, kernel_size=3)
        self.Conv2d_2b_3x3 = BasicConv2d(32, 64, kernel_size=3, padding=1)
        self.Conv2d_3b_1x1 = BasicConv2d(64, 80, kernel_size=1)
        self.Conv2d_4a_3x3 = BasicConv2d(80, 192, kernel_size=3)
        self.Mixed_5b = InceptionA(192, pool_features=32)
        self.Mixed_5c = InceptionA(256, pool_features=64)
        self.Mixed_5d = InceptionA(288, pool_features=64)
        self.Mixed_6a = InceptionB(288)
        self.Mixed_6b = InceptionC(768, channels_7x7=128)
        self.Mixed_6c = InceptionC(768, channels_7x7=160)
        self.Mixed_6d = InceptionC(768, channels_7x7=160)
        self.Mixed_6e = InceptionC(768, channels_7x7=192)
        if aux_logits:
            self.AuxLogits = InceptionAux(768, num_classes)
        self.Mixed_7a = InceptionD(768)
        self.Mixed_7b = InceptionE(1280)
        self.Mixed_7c = InceptionE(2048)
        self.fc = nn.Linear(2048, num_classes)
        for m in self.modules():
            if isinstance(m, nn.Conv2d) or isinstance(m, nn.Linear):
                import scipy.stats as stats
                stddev = m.stddev if hasattr(m, 'stddev') else 0.1
                X = stats.truncnorm(-2, 2, scale=stddev)
                values = torch.Tensor(X.rvs(m.weight.data.numel()))
                m.weight.data.copy_(values)
            elif isinstance(m, nn.BatchNorm2d):
                m.weight.data.fill_(1)
                m.bias.data.zero_()
    def forward(self, x):
        if self.transform_input: # 1
            x = x.clone()
            x[:, 0] = x[:, 0] * (0.229 / 0.5) + (0.485 - 0.5) / 0.5
            x[:, 1] = x[:, 1] * (0.224 / 0.5) + (0.456 - 0.5) / 0.5
            x[:, 2] = x[:, 2] * (0.225 / 0.5) + (0.406 - 0.5) / 0.5
        # 299 x 299 x 3
        x = self.Conv2d_1a_3x3(x)
        # 149 x 149 x 32
        x = self.Conv2d_2a_3x3(x)
        # 147 x 147 x 32
        x = self.Conv2d_2b_3x3(x)
        # 147 x 147 x 64
        x = F.max_pool2d(x, kernel_size=3, stride=2)
        # 73 x 73 x 64
        x = self.Conv2d_3b_1x1(x)
        # 73 x 73 x 80
        x = self.Conv2d_4a_3x3(x)
        # 71 x 71 x 192
        x = F.max_pool2d(x, kernel_size=3, stride=2)
        # 35 x 35 x 192
        x = self.Mixed_5b(x)
        # 35 x 35 x 256
        x = self.Mixed_5c(x)
        # 35 x 35 x 288
        x = self.Mixed_5d(x)
        # 35 x 35 x 288
        x = self.Mixed_6a(x)
        # 17 x 17 x 768
        x = self.Mixed_6b(x)
        # 17 x 17 x 768
        x = self.Mixed_6c(x)
        # 17 x 17 x 768
        x = self.Mixed_6d(x)
        # 17 x 17 x 768
        x = self.Mixed_6e(x)
        # 17 x 17 x 768
        if self.training and self.aux_logits:
            aux = self.AuxLogits(x)
        # 17 x 17 x 768
        x = self.Mixed_7a(x)
        # 8 x 8 x 1280
        x = self.Mixed_7b(x)
        # 8 x 8 x 2048
        x = self.Mixed_7c(x)
        # 8 x 8 x 2048
        x = F.avg_pool2d(x, kernel_size=8)
        # 1 x 1 x 2048
        x = F.dropout(x, training=self.training)
        # 1 x 1 x 2048
        x = x.view(x.size(0), -1)
        # 2048
        x = self.fc(x)
        # 1000 (num_classes)
        if self.training and self.aux_logits:
            return x, aux
        return x

参考

https://zhuanlan.zhihu.com/p/30172532

nooobme

关注

7
点赞
踩
50

收藏

觉得还不错? 一键收藏
打赏
1
评论
经典卷积模型(四)GoogLeNet-Inception(V3)代码解析

pytorch提供的有六种基本的inception模块，分别是InceptionA——InceptionE。InceptionAclass InceptionA(nn.Module): def __init__(self, in_channels, pool_features): super(InceptionA, self).__init__() self.branch1x1 = BasicConv2d(in_channels, 64, kernel_siz
复制链接

扫一扫