了解最大池化层之后的完全连接层的尺寸 [英] Understanding the dimensions of a fully-connected layer that follows a max-pooling layer

问题描述

在下面的图(体系结构)中，如何从尺寸256x13x13的最后一个最大池层(右侧)派生4096个单元的(完全连接的)密集层?而不是4096，不是256 * 13 * 13 = 43264吗?

In the diagram (architecture) below, how was the (fully-connected) dense layer of 4096 units derived from last max-pool layer (on the right) of dimensions 256x13x13? Instead of 4096, shouldn't it be 256*13*13=43264 ?

推荐答案

如果我是对的，您会问为什么4096x1x1层要小得多.

If I'm correct, you're asking why the 4096x1x1 layer is much smaller.

那是因为它是一个完全连接的层.来自最后一个最大池化层(= 256*13*13=43264神经元)的每个神经元都连接到完全连接层的每个神经元.

That's because it's a fully connected layer. Every neuron from the last max-pooling layer (=256*13*13=43264 neurons) is connectd to every neuron of the fully-connected layer.

这是从ALL到ALL连接的神经网络的示例: 如您所见，layer2大于layer3.这并不意味着他们无法连接.

This is an example of an ALL to ALL connected neural network: As you can see, layer2 is bigger than layer3. That doesn't mean they can't connect.

没有最后一个最大池化层的转换->最大池化层中的所有神经元仅与下一层中的所有4096个神经元相连.

There is no conversion of the last max-pooling layer -> all the neurons in the max-pooling layer are just connected with all the 4096 neurons in the next layer.

密集"操作仅意味着计算所有这些连接(= 4096 * 43264个连接)的权重和偏差，并加上神经元的偏差以计算下一个输出.

The 'dense' operation just means calculate the weights and biases of all these connections (= 4096 * 43264 connections) and add the bias of the neurons to calculate the next output.

它的连接方式与MLP相同.

It's connected the same was an MLP.

但是为什么是4096?没有理由.这只是一个选择.可能是8000，可能是20，这取决于哪种网络最适合您.

But why 4096? There is no reasoning. It's just a choice. It could have been 8000, it could have been 20, it just depends on what works best for the network.

这篇关于了解最大池化层之后的完全连接层的尺寸的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！