本文共 4289 字,大约阅读时间需要 14 分钟。
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data",one_hot=True)
#每个批次的大小
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])#样本
y = tf.placeholder(tf.float32,[None,10])#标签
#创建一个简单的神经网络,后面优化可以用多个隐藏层
W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))
prediction = tf.nn.softmax(tf.matmul(x,W)+b)
#定义二次代价函数
loss = tf.reduce_mean(tf.square(y-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))
#求准确率
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
with tf.Session() as sess:
sess.run(init)
for epoch in range(21):
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))
#可以看到最后的预测正确率为:91.356,其实这个程序有许多地方可以进行优化:1增加隐含层,2迭代多次
转载地址:http://zosws.baihongyu.com/