tensorflow实战（一）TensorFlow实现 softmax Regression 识别手写数字

一、算法思想：

二、操作步骤：

1.在ipython 或者spyder中直接输入以下代码

# -*- coding: utf-8 -*-
"""
Created on Fri Jan 12 16:29:41 2018


@author: Administrator
"""
#下载数据集
from  tensorflow.examples.tutorials.mnist import input_data
mnist=input_data.read_data_sets("D:/ProgramData/Anaconda3/envs/MNIST_data/",one_hot=True)
print(mnist.train.images.shape,mnist.train.labels.shape)
 
import tensorflow as tf


sess=tf.InteractiveSession()
#第一步定义算法公式
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)
#第二步，定义cross function，选定优化器，并指定优化器优化loss
y_=tf.placeholder(tf.float32,[None,10])
#定义交叉熵损失函数
cross_entropy=tf.reduce_mean(-tf.reduce_sum(y_*tf.log(y),reduction_indices=[1]))
 #使用梯度下降法最小化cross_entropy损失函数  
train_step=tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)
#第三步，迭代的对数据进行训练
tf.global_variables_initializer().run()
for i in range(1000):
    batch_xs,batch_ys=mnist.train.next_batch(100)
    train_step.run({x:batch_xs,y_:batch_ys})
#在训练集上或验证集上对准确率进行评测
correct_prediction=tf.equal(tf.arg_max(y,1),tf.arg_max(y_,1))
#转化为float32类型
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
print (accuracy.eval({x:mnist.test.images,y_:mnist.test.labels}))

遇到的错误:

1.URLError:

解决方法：在http://yann.lecun.com/exdb/mnist/   上下载四个文件，放在MNIST_data下。

2.InvalidArgumentError (see above for traceback): You must feed a value for placeholder tensor 'Placeholder_6' with

 for 循环里

把for循环里的 train_step.run({x:batch_xs,y:batch_ys})改为

train_step.run({x:batch_xs,y_:batch_ys})