TensorFlow，十三：模型的保存与载入

一：保存

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

#载入数据集
mnist = input_data.read_data_sets("MNIST_data",one_hot=True)

#每个批次100张照片
batch_size = 100
#计算一共有多少个批次
n_batch = mnist.train.num_examples // batch_size

#定义两个placeholder
x = tf.placeholder(tf.float32,[None,784])
y = tf.placeholder(tf.float32,[None,10])

#创建一个简单的神经网络，输入层784个神经元，输出层10个神经元
W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))
prediction = tf.nn.softmax(tf.matmul(x,W)+b)

#二次代价函数
# loss = tf.reduce_mean(tf.square(y-prediction))
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(labels=y,logits=prediction))
#使用梯度下降法
train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss)

#初始化变量
init = tf.global_variables_initializer()

#结果存放在一个布尔型列表中
correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(prediction,1))#argmax返回一维张量中最大的值所在的位置
#求准确率
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))

saver = tf.train.Saver()

with tf.Session() as sess:
    sess.run(init)
    for epoch in range(11):
        for batch in range(n_batch):
            batch_xs,batch_ys =  mnist.train.next_batch(batch_size)
            sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys})
        
        acc = sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels})
        print("Iter " + str(epoch) + ",Testing Accuracy " + str(acc))
    #保存模型
    saver.save(sess,'net/my_net.ckpt')

结果：

Iter 0,Testing Accuracy 0.8252
Iter 1,Testing Accuracy 0.8916
Iter 2,Testing Accuracy 0.9008
Iter 3,Testing Accuracy 0.906
Iter 4,Testing Accuracy 0.9091
Iter 5,Testing Accuracy 0.9104
Iter 6,Testing Accuracy 0.911
Iter 7,Testing Accuracy 0.9127
Iter 8,Testing Accuracy 0.9145
Iter 9,Testing Accuracy 0.9166
Iter 10,Testing Accuracy 0.9177

二：载入

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

#载入数据集
mnist = input_data.read_data_sets("MNIST_data",one_hot=True)

#每个批次100张照片
batch_size = 100
#计算一共有多少个批次
n_batch = mnist.train.num_examples // batch_size

#定义两个placeholder
x = tf.placeholder(tf.float32,[None,784])
y = tf.placeholder(tf.float32,[None,10])

#创建一个简单的神经网络，输入层784个神经元，输出层10个神经元
W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))
prediction = tf.nn.softmax(tf.matmul(x,W)+b)

#二次代价函数
# loss = tf.reduce_mean(tf.square(y-prediction))
loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(labels=y,logits=prediction))
#使用梯度下降法
train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss)

#初始化变量
init = tf.global_variables_initializer()

#结果存放在一个布尔型列表中
correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(prediction,1))#argmax返回一维张量中最大的值所在的位置
#求准确率
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))

saver = tf.train.Saver()

with tf.Session() as sess:
    sess.run(init)
    # 未载入模型时的识别率
    print('未载入识别率',sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels}))
    saver.restore(sess,'net/my_net.ckpt')
    # 载入模型后的识别率
    print('载入后识别率',sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels}))

结果：

未载入识别率 0.098
INFO:tensorflow:Restoring parameters from net/my_net.ckpt
载入后识别率 0.9177