tf2版本识别图中模糊的手写数字（来自深度学习tensorflow入门、原理与进阶实战这本书（李金洪编著））

第五章

#退化学习率
import tensorflow.compat.v1 as tf###
tf.disable_v2_behavior()###tf1  api

global_step = tf.Variable(0, trainable=False)
initial_learning_rate = 0.1
learning_rate = tf.train.exponential_decay(initial_learning_rate,global_step=global_step,decay_steps=10,decay_rate=0.9)

opt = tf.train.GradientDescentOptimizer(learning_rate)
add_global = global_step.assign_add(1) #令global_step加一计步
with tf.Session() as sess:
    #init = tf.global_variables_initializer()
    #sess.run(init)
    tf.global_variables_initializer().run()
    
    print(sess.run(learning_rate))
    for i in range(21):
        i, rate = sess.run([add_global, learning_rate])
        print(i,rate)

import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

print ('输入数据:',mnist.train.images)
print ('输入数据打shape:',mnist.train.images.shape)

import pylab 
im = mnist.train.images[1]
im = im.reshape(-1,28)
pylab.imshow(im)
pylab.show()

print ('输入数据打shape:',mnist.test.images.shape)
print ('输入数据打shape:',mnist.validation.images.shape)

import tensorflow.compat.v1 as tf
tf.disable_v2_behavior()
from tensorflow.examples.tutorials.mnist import input_data
minist = input_data.read_data_sets("MNIST_data/", one_hot=True)
import pylab

tf.reset_default_graph()
x= tf.placeholder(tf.float32, [None, 784])
y= tf.placeholder(tf.float32, [None, 10])

#训练模型并输出中间状态参数
training_epochs=25#迭代次数
batch_size=100#每次训练取出数据条数
display_step=1#每训练一次把具体的中间状态显示出来

saver = tf.train.Saver()
model_path = "log/521model.ckpt"
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    
    for epoch in range(training_epochs):
        avg_cost=0
        total_batch = int(mnist.train.num_examples/batch_size)
        for i in range(total_batch):
            batch_xs,batch_ys = mnist.train.next_batch(batch_size)
            _, c = sess.run([optimizer, cost], feed_dict={x: batch_xs,y: batch_ys})
            avg_cost += c / total_batch ###
            
        if(epoch+1) % display_step ==0:###
            print("Epoch:", "%04d" % (epoch+1), "cost=", "{:.9f}".format(avg_cost))
            
    print("Fished!")
    #测试modle
    correct_prediction = tf.equal(tf.argmax(pred, 1),tf.argmax(y,1))
    #tf.equal(A, B)是对比这两个矩阵或者向量的相等的元素，如果是相等的那就返回True，反正返回False，返回的值的矩阵维度和A是一样的
    #计算准确率
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))#cast：数据类型转换  reduce_mean:均值
    print("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
    
    
    
    #保存模型
    save_path = saver.save(sess, model_path)
    print("Model saved in file: %s" % save_path)

#读取模型
print("Starting 2nd session....")
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    ###恢复模型
    saver.restore(sess, model_path)
    
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
    print("Accuracy:", accuracy.eval({x: mnist.test.images,y:mnist.test.labels}))
    
    output = tf.argmax(pred, 1)
    batch_xs, batch_ys = mnist.train.next_batch(2)
    outputval,predv = sess.run([output,pred], feed_dict={x:batch_xs})
    print(outputval,predv,batch_xs)
    
    im = batch_xs[0]
    im = im.reshape(-1, 28)
    pylab.imshow(im)
    pylab.show()
    
    im = batch_xs[1]
    im = im.reshape(-1, 28)
    pylab.imshow(im)
    pylab.show()