TensorFlow MNIST 手写数字识别 -- TensorFlow 实现

【tensorflow】mnist手写数字识别--tensorflow实现

LeCun最先将反向传播技术用于handwritten postal code recognition，并随后采用convolutional neural networks for the recognition of handwritten digits.接着他把convolutional neural networks applied to the identification of handwritten digits.这正是CNN最初的架构——LeNet-5.

LeNet5包含Input、卷积层1、池化层1、卷积层2、池化层2、全连接层、输出层。

 INPUT: [28x28x1]           weights: 0

    
 CONV5-32: [28x28x32]       weights: (5*5*1+1)*32
    
 POOL2: [14x14x32]          weights: 2*2*1
    
 CONV5-64: [14x14x64]       weights: (5*5*32+1)*64
    
 POOL2: [7x7x64]            weights: 2*2*1
    
 FC: [1x1x1024]