首页 > 其他分享> 文章详细

VGG网络及Pytorch实现

2021-03-28 13:01:23 阅读：242 来源： 互联网

标签：nn VGG 网络 Pytorch kwargs progress 512 True pretrained

2014年由牛津大学研究组Visual Geometry Group提出，论文地址Very Deep Convolutional Networks for Large-Scale Image Recognition

在这里插入图片描述

图3 VGG网络结构

文章亮点：通过堆叠多个3×3卷积核来代替大尺度卷积核（减少所需参数）。

论文中提到：堆叠2个3×3的卷积核代替5×5的卷积核，堆叠3个3×3的卷积核代替7×7的卷积核，它们拥有相同的感受野。

基本概念拓展：CNN感受野

在CNN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）。

F ( i ) = ( F ( i + 1 ) − 1 ) × S + k s F(i)=(F(i+1)-1)\times S+ks F(i)=(F(i+1)−1)×S+ks

例如， F = 1 F=1 F=1，经过3个3×3的卷积核后感受野为7×7

C o n v 3 × 3 ( 3 ) : F = ( 1 − 1 ) × 1 + 3 = 3 Conv3\times 3(3):F=(1-1)\times 1+3=3 Conv3×3(3):F=(1−1)×1+3=3
C o n v 3 × 3 ( 2 ) : F = ( 3 − 1 ) × 1 + 3 = 5 Conv3\times 3(2):F=(3-1)\times 1+3=5 Conv3×3(2):F=(3−1)×1+3=5
C o n v 3 × 3 ( 1 ) : F = ( 5 − 1 ) × 1 + 3 = 7 Conv3\times 3(1):F=(5-1)\times 1+3=7 Conv3×3(1):F=(5−1)×1+3=7

pytorch实现

class VGG(nn.Module):

    def __init__(self, features, num_classes=1000, init_weights=True):
        super(VGG, self).__init__()
        self.features = features
        self.avgpool = nn.AdaptiveAvgPool2d((7, 7))
        self.classifier = nn.Sequential(
            nn.Linear(512 * 7 * 7, 4096),
            nn.ReLU(True),
            nn.Dropout(),
            nn.Linear(4096, 4096),
            nn.ReLU(True),
            nn.Dropout(),
            nn.Linear(4096, num_classes),
        )
        if init_weights:
            self._initialize_weights()

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

    def _initialize_weights(self):
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
                if m.bias is not None:
                    nn.init.constant_(m.bias, 0)
            elif isinstance(m, nn.BatchNorm2d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)
            elif isinstance(m, nn.Linear):
                nn.init.normal_(m.weight, 0, 0.01)
                nn.init.constant_(m.bias, 0)


def make_layers(cfg, batch_norm=False):
    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)
            if batch_norm:
                layers += [conv2d, nn.BatchNorm2d(v), nn.ReLU(inplace=True)]
            else:
                layers += [conv2d, nn.ReLU(inplace=True)]
            in_channels = v
    return nn.Sequential(*layers)


cfgs = {
    'A': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
    'B': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
    'D': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'],
    'E': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'],
}


def _vgg(arch, cfg, batch_norm, pretrained, progress, **kwargs):
    if pretrained:
        kwargs['init_weights'] = False
    model = VGG(make_layers(cfgs[cfg], batch_norm=batch_norm), **kwargs)
    if pretrained:
        state_dict = load_state_dict_from_url(model_urls[arch],
                                              progress=progress)
        model.load_state_dict(state_dict)
    return model


def vgg11(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg11', 'A', False, pretrained, progress, **kwargs)


def vgg11_bn(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg11_bn', 'A', True, pretrained, progress, **kwargs)


def vgg13(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg13', 'B', False, pretrained, progress, **kwargs)


def vgg13_bn(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg13_bn', 'B', True, pretrained, progress, **kwargs)


def vgg16(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg16', 'D', False, pretrained, progress, **kwargs)


def vgg16_bn(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg16_bn', 'D', True, pretrained, progress, **kwargs)


def vgg19(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg19', 'E', False, pretrained, progress, **kwargs)


def vgg19_bn(pretrained=False, progress=True, **kwargs):
    return _vgg('vgg19_bn', 'E', True, pretrained, progress, **kwargs)

标签：nn,VGG,网络,Pytorch,kwargs,progress,512,True,pretrained
来源： https://blog.csdn.net/weixin_39750526/article/details/115278427

本站声明： 1. iCode9 技术分享网（下文简称本站）提供的所有内容，仅供技术学习、探讨和分享；
2. 关于本站的所有留言、评论、转载及引用，纯属内容发起人的个人观点，与本站观点和立场无关；
3. 关于本站的所有言论和文字，纯属内容发起人的个人观点，与本站观点和立场无关；
4. 本站文章均是网友提供，不完全保证技术分享内容的完整性、准确性、时效性、风险性和版权归属；如您发现该文章侵犯了您的权益，可联系我们第一时间进行删除；
5. 本站为非盈利性的个人网站，所有内容不会用来进行牟利，也不会利用任何形式的广告来间接获益，纯粹是为了广大技术爱好者提供技术内容和技术思想的分享性交流网站。

ICode9

VGG网络及Pytorch实现