第一步:导入头函数
#coding=utf-8
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
第二步:导入数据
#打印数据维度
mnist=input_data.read_data_sets("MNIST_data",one_hot=True)
print(mnist.train.images.shape,mnist.train.labels.shape)
print(mnist.test.images.shape,mnist.test.labels.shape)
print(mnist.validation.images.shape,mnist.validation.labels.shape)
第三步:搭建网络
#构建计算图
x=tf.placeholder(tf.float32,[None,784])
w=tf.Variable(tf.zeros([784,10]))
b=tf.Variable(tf.zeros([10]))
y=tf.nn.softmax(tf.matmul(x,w)+b)
y_=tf.placeholder(tf.float32,[None,10])
第四步:损失函数与优化器
cross_entropy=tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y),reduction_indices=[1]))#计算损失函数loss
train_step=tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)#学习速率为0.5
第五步:训练
#在会话sess中启动图
sess=tf.InteractiveSession()
tf.global_variables_initializer().run()
for i in range(1000):
batch_xs,batch_ys=mnist.train.next_batch(50)
train_step.run({x:batch_xs,y_:batch_ys})
第六步:测试
#测试验证阶段
#沿着第1条轴方向取y和y_的最大值的索引并判断是否相等
correct_prediction=tf.equal(tf.argmax(y,1),tf.argmax(y_,1))
#转换bool型tensor为float32型tensor并求平均即得到正确率
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))#tf.cast函数把correct_prediction的数据类型转float32
print(accuracy.eval({x:mnist.test.images,y_:mnist.test.labels}))#tf.eval把tensor的值输出
总代码:
#coding=utf-8
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
#打印数据维度
mnist=input_data.read_data_sets("MNIST_data",one_hot=True)
print(mnist.train.images.shape,mnist.train.labels.shape)
print(mnist.test.images.shape,mnist.test.labels.shape)
print(mnist.validation.images.shape,mnist.validation.labels.shape)
#构建计算图
x=tf.placeholder(tf.float32,[None,784])
w=tf.Variable(tf.zeros([784,10]))
b=tf.Variable(tf.zeros([10]))
y=tf.nn.softmax(tf.matmul(x,w)+b)
y_=tf.placeholder(tf.float32,[None,10])
cross_entropy=tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y),reduction_indices=[1]))#计算损失函数loss
train_step=tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)#学习速率为0.5
#在会话sess中启动图
sess=tf.InteractiveSession()
tf.global_variables_initializer().run()
for i in range(1000):
batch_xs,batch_ys=mnist.train.next_batch(50)
train_step.run({x:batch_xs,y_:batch_ys})
#测试验证阶段
#沿着第1条轴方向取y和y_的最大值的索引并判断是否相等
correct_prediction=tf.equal(tf.argmax(y,1),tf.argmax(y_,1))
#转换bool型tensor为float32型tensor并求平均即得到正确率
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))#tf.cast函数把correct_prediction的数据类型转float32
print(accuracy.eval({x:mnist.test.images,y_:mnist.test.labels}))#tf.eval把tensor的值输出
结果:0.9114