1 import tensorflow as tf 2 from tensorflow.examples.tutorials.mnist import input_data 3
4 #載入數據集
5 mnist = input_data.read_data_sets("MNIST_data",one_hot=True) 6
7 #每個批次100張照片
8 batch_size = 100
9 #計算一共有多少個批次
10 n_batch = mnist.train.num_examples // batch_size 11
12 #定義兩個placeholder
13 x = tf.placeholder(tf.float32,[None,784]) 14 y = tf.placeholder(tf.float32,[None,10]) 15
16 #創建一個簡單的神經網絡，輸入層784個神經元，輸出層10個神經元
17 W = tf.Variable(tf.zeros([784,10])) 18 b = tf.Variable(tf.zeros([10])) 19 prediction = tf.nn.softmax(tf.matmul(x,W)+b) 20
21 #二次代價函數
22 # loss = tf.reduce_mean(tf.square(y-prediction))
23 loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=prediction)) 24 #使用梯度下降法
25 train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss) 26
27 #初始化變量
28 init = tf.global_variables_initializer() 29
30 #結果存放在一個布爾型列表中
31 correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(prediction,1))#argmax返回一維張量中最大的值所在的位置
32 #求准確率
33 accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32)) 34
35 saver = tf.train.Saver() 36
37 with tf.Session() as sess: 38 sess.run(init) 39 for epoch in range(11): 40 for batch in range(n_batch): 41 batch_xs,batch_ys = mnist.train.next_batch(batch_size) 42 sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys}) 43
44 acc = sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels}) 45 print("Iter " + str(epoch) + ",Testing Accuracy " + str(acc)) 46 #保存模型
47 saver.save(sess,'net/my_net.ckpt')
還是以手寫數字識別為例，想要保存模型，首先建立一個saver：
saver = tf.train.Saver()
通過調用save，自動將session中的參數保存起來：
saver.save(sess,'net/my_net.ckpt')
創建路徑為當前路徑下net文件夾，運行之后：
2019-06-19 10:38:19