tensorflow实现CNN—— LeNet5

# -*- coding: utf-8 -*-

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data

minist = input_data.read_data_sets('MNIST_data',one_hot=True)
sess = tf.InteractiveSession()

#参数初始化函数
def weight_variable(shape):
    initial = tf.truncated_normal(shape, stddev=0.1)
    return tf.Variable(initial)
    
def bias_variable(shape):
    initial = tf.constant(0.1,shape=shape)
    return tf.Variable(initial)

#卷积层、池化层
#tf.nn.conv2d(x,w)  x是输入，w是参数，比如[5,5,1,32] 
#  5,5表示卷积核大小，1表示通道数,例如RGB是3通道，32值卷积核的数量
#strides表示卷积核移动的步长，padding=‘SAME’表示输入和输出图像的尺寸保持一致
def conv2d(x,W):
    return tf.nn.conv2d(x, W, strides=[1,1,1,1],padding='SAME')
    
def max_pool_2x2(x):
    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')
    
#由于输入的images图像是1D的，而CNN要求输入具有2D，因而将原始数据784转成28x28
x = tf.placeholder(tf.float32, [None,784])
y_ = tf.placeholder(tf.float32, [None,10])
x_image = tf.reshape(x,[-1,28,28,1])

#定义第一个卷积层
W_conv1 = weight_variable([5,5,1,32])
b_conv1 = bias_variable([32])
h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1) + b_conv1)
h_pool1 = max_pool_2x2(h_conv1)

#定义第二个卷积核
W_conv2 = weight_variable([5,5,32,64])
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1,W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)

#第三层（全连接层）
#这里由于是全连接层，且输出的tensor尺寸为 7x7x64,经过2次池化，图像变成7x7,卷积核为64
#将2D转成1D，输入全连接层，隐含层节点数1024
W_fc1 = weight_variable([7*7*64,1024])
b_fc1 = bias_variable([1024])
h_pool2_flat = tf.reshape(h_pool2, [-1,7*7*64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat,W_fc1) + b_fc1)

#为了减轻过拟合，下面使用一个dropout层，主要控制keep_prob
keep_prob = tf.placeholder(tf.float32)
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob)

#最后将dropout层输出到softmax层，进行多分类，输出概率
W_fc2 = weight_variable([1024,10])
b_fc2 = bias_variable([10])
y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop,W_fc2) + b_fc2)

#定义损失函数
cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ *tf.log(y_conv),
                                              reduction_indices=[1]))
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)

#定义评价准确率
correct_prediction = tf.equal(tf.argmax(y_conv,1),tf.argmax(y_,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))

#开始训练
tf.global_variables_initializer().run()
for i in range(2000):
    batch = minist.train.next_batch(50)
    #每训练100轮就输出一次准确率
    if i%100==0:
        train_accuracy = accuracy.eval(feed_dict={x:batch[0],y_:batch[1],
                                                  keep_prob:1.0})
        print("step %d, training accuracy %g" %(i, train_accuracy))
    train_step.run(feed_dict={x:batch[0],y_:batch[1],keep_prob:0.5})
    
#输出最终的准确率
print("test accuracy %g" %accuracy.eval(feed_dict={x: minist.test.images,
                            y_: minist.test.labels, keep_prob: 1.0})