背景

2014年，VGG分别在定位和分类问题中获得了第一和第二名，在其他数据集上也实现了最好的结果。

结构

VGGNet探索了神经网络的深度与性能之间的关系，表明在结构相似的情况下，网络越深性能越好。

模型中大量使用3*3的卷积核的串联，构造出16到19层的网络。

2个3*3的卷积核的串联相当于5*5的卷积核。 
3个3*3的卷积核的串联相当于7*7的卷积核。

其意义在于7*7所需要的参数为49，3个3*3的卷积核参数为27个，几乎减少了一半。

在C中还是用了1*1的卷积核，而且输出通道和输入通道数并没有发生改变，只是起到了线性变换的作用，其意义在VGG中其实意义不大。

结构图如下：

实现

    data = mx.symbol.Variable(name="data")# group 1conv1_1 = mx.symbol.Convolution(data=data, kernel=(3, 3), pad=(1, 1), num_filter=64, name="conv1_1")relu1_1 = mx.symbol.Activation(data=conv1_1, act_type="relu", name="relu1_1")pool1 = mx.symbol.Pooling(data=relu1_1, pool_type="max", kernel=(2, 2), stride=(2,2), name="pool1")# group 2conv2_1 = mx.symbol.Convolution(data=pool1, kernel=(3, 3), pad=(1, 1), num_filter=128, name="conv2_1")relu2_1 = mx.symbol.Activation(data=conv2_1, act_type="relu", name="relu2_1")pool2 = mx.symbol.Pooling(data=relu2_1, pool_type="max", kernel=(2, 2), stride=(2,2), name="pool2")# group 3conv3_1 = mx.symbol.Convolution(data=pool2, kernel=(3, 3), pad=(1, 1), num_filter=256, name="conv3_1")relu3_1 = mx.symbol.Activation(data=conv3_1, act_type="relu", name="relu3_1")conv3_2 = mx.symbol.Convolution(data=relu3_1, kernel=(3, 3), pad=(1, 1), num_filter=256, name="conv3_2")relu3_2 = mx.symbol.Activation(data=conv3_2, act_type="relu", name="relu3_2")pool3 = mx.symbol.Pooling(data=relu3_2, pool_type="max", kernel=(2, 2), stride=(2,2), name="pool3")# group 4conv4_1 = mx.symbol.Convolution(data=pool3, kernel=(3, 3), pad=(1, 1), num_filter=512, name="conv4_1")relu4_1 = mx.symbol.Activation(data=conv4_1, act_type="relu", name="relu4_1")conv4_2 = mx.symbol.Convolution(data=relu4_1, kernel=(3, 3), pad=(1, 1), num_filter=512, name="conv4_2")relu4_2 = mx.symbol.Activation(data=conv4_2, act_type="relu", name="relu4_2")pool4 = mx.symbol.Pooling(data=relu4_2, pool_type="max", kernel=(2, 2), stride=(2,2), name="pool4")# group 5conv5_1 = mx.symbol.Convolution(data=pool4, kernel=(3, 3), pad=(1, 1), num_filter=512, name="conv5_1")relu5_1 = mx.symbol.Activation(data=conv5_1, act_type="relu", name="relu5_1")conv5_2 = mx.symbol.Convolution(data=relu5_1, kernel=(3, 3), pad=(1, 1), num_filter=512, name="conv5_2")relu5_2 = mx.symbol.Activation(data=conv5_2, act_type="relu", name="conv1_2")pool5 = mx.symbol.Pooling(data=relu5_2, pool_type="max", kernel=(2, 2), stride=(2,2), name="pool5")# group 6flatten = mx.symbol.Flatten(data=pool5, name="flatten")fc6 = mx.symbol.FullyConnected(data=flatten, num_hidden=4096, name="fc6")relu6 = mx.symbol.Activation(data=fc6, act_type="relu", name="relu6")drop6 = mx.symbol.Dropout(data=relu6, p=0.5, name="drop6")# group 7fc7 = mx.symbol.FullyConnected(data=drop6, num_hidden=4096, name="fc7")relu7 = mx.symbol.Activation(data=fc7, act_type="relu", name="relu7")drop7 = mx.symbol.Dropout(data=relu7, p=0.5, name="drop7")# outputfc8 = mx.symbol.FullyConnected(data=drop7, num_hidden=num_classes, name="fc8")softmax = mx.symbol.SoftmaxOutput(data=fc8, name='softmax')return softmax