VGG网络原理与搭建
   
VGG在2014年由牛津大学著名研究组VGG(Visual Geometry Group)提出,斩获该年ImageNet竞赛中Localization Task(定位任务)第一名和Classification Task(分类任务)第二名。
论文中提及的结构
     
   
在卷积神经网络中,决定某一层输出结果中一个元素所对应的输入层的区域大小,被称作感受野(receptive field)。通俗的解释是,输出feature map上的一个单元对应输入层上的区域大小。
    
    
    感受野计算公式:
    
    F(i=(F(i+1)-1)× Stride + Ksize
   
- 
F(i)为第i层感受野 
- 
Stride为第i层的步距 
- 
Ksize为卷积核或池化核尺寸 
Vgg中最常用的是模型D
conv的stride为1,padding为1maxpool的size为2,stride为2
    
    
    使用pytorch搭建VGG网络
   
搭建VGG网络分为两个步骤
- 
提取网络结构 
- 
分类网络结构 
定义一个字典文件
cfgs = {
    'vgg11': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
    'vgg13': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
    'vgg16': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'],
    'vgg19': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'],
}
通过函数提取特征网络结构
def make_features(cfg: list):
    layers = []
    in_channels = 3
    for v in cfg:
        if v == "M":
            layers += [nn.MaxPool2d(kernel_size=2, stride=2)]
        else:
            conv2d = nn.Conv2d(in_channels, v, kernel_size=3, padding=1)
            layers += [conv2d, nn.ReLU(True)]
            in_channels = v
    return nn.Sequential(*layers)
定义Vgg类模型
class VGG(nn.Module):
    def __init__(self, features, num_classes=1000, init_weights=False):
        super(VGG, self).__init__()
        self.features = features
        self.classifier = nn.Sequential(
            nn.Linear(512*7*7, 4096),
            nn.ReLU(True),
            nn.Dropout(p=0.5),
            nn.Linear(4096, 4096),
            nn.ReLU(True),
            nn.Dropout(p=0.5),
            nn.Linear(4096, num_classes)
        )
        if init_weights:
            self._initialize_weights()
设置权重的过程
class VGG(nn.Module):
    def __init__(self, features, num_classes=1000, init_weights=False):
        super(VGG, self).__init__()
        self.features = features
        self.classifier = nn.Sequential(
            nn.Linear(512*7*7, 4096),
            nn.ReLU(True),
            nn.Dropout(p=0.5),
            nn.Linear(4096, 4096),
            nn.ReLU(True),
            nn.Dropout(p=0.5),
            nn.Linear(4096, num_classes)
        )
        if init_weights:
            self._initialize_weights()
VGG网络正向传播的过程
 def forward(self, x):
        # N x 3 x 224 x 224
        x = self.features(x)
        # N x 512 x 7 x 7
        x = torch.flatten(x, start_dim=1)
        # N x 512*7*7
        x = self.classifier(x)
        return x
使用函数构造Vgg模型
def vgg(model_name="vgg16", **kwargs):
    assert model_name in cfgs, "Warning: model number {} not in cfgs dict!".format(model_name)
    cfg = cfgs[model_name]
    model = VGG(make_features(cfg), **kwargs)
    return model
 
版权声明:本文为weixin_46167190原创文章,遵循 CC 4.0 BY-SA 版权协议,转载请附上原文出处链接和本声明。
