softmax简单模型处理Mnist分类问题

import tensorflow as tf
import input_data

#导入数据
mnist = input_data.read_data_sets('MNIST_data',one_hot=True)

#权重及偏置设置
weights = tf.Variable(tf.zeros([784,10]))
biases = tf.Variable(tf.zeros([10]))

#输入数据占位符
x=tf.placeholder("float",[None,784])

#softmax函数
y = tf.nn.softmax(tf.matmul(x,weights)+biases)

#真实样本标签
y_real = tf.placeholder("float",[None,10])

#交叉商作为目标函数
cross_entropy = -tf.reduce_sum(y_real*tf.log(y))

#函数优化
train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)
init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init)

#小批量梯度下降优化算法执行
for i in range(1000):

    #随机选取100个样本训练
    batch_xs,batch_ys = mnist.train.next_batch(100)

    #赋值占位变量
    sess.run(train_step,feed_dict={x:batch_xs,y_real:batch_ys})
    if i%100==0:

        #计算正确的样本个数
        correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(y_real,1))

        #计算正确比例
        accuracy = tf.reduce_mean(tf.cast(correct_prediction,"float"))
        print(sess.run(accuracy,feed_dict={x:mnist.test.images,y_real:mnist.test.labels}))