In an image classification network, you can use aglobalAveragePooling3dLayerbefore the final fully connected layer to reduce the size of the activations without sacrificing performance. The reduced size of the activations means that the downstream fully connected layers will have fewer weights, reducing the size of your network.

You can use aglobalAveragePooling3dLayertowards the end of a classification network instead of afullyConnectedLayer. Since global pooling layers have no learnable parameters, they can be less prone to overfitting and can reduce the size of the network. These networks can also be more robust to spatial translations of input data. You can also replace a fully connected layer with aglobalmaxpooling3dlayer.instead. Whether aglobalmaxpooling3dlayer.or aglobalAveragePooling3dLayer更合适取决于您的数据集。

要使用全局平均池代替完全连接的层，输入的大小globalAveragePooling3dLayermust match the number of classes in the classification problem