TensorFlow 实现Softmax Regression识别手写数字

使用TensorFlow实现一个简单的机器学习算法Softmax Regression，一个没有隐含层的最浅的神经网络。

环境：Ubuntu + Anaconda（Spyder） + TensorFlow
数据：MNIST（Mixed National Institute of Standards and Technology database）
步骤：
1）定义算法公司，也就是神经网络forward时的计算。
2）定义loss，选定优化器，并指定优化器优化loss。
3）迭代地对数据进行训练。
4）在测试集或验证集上对准确率进行评测。

---

太多的理论赛不过代码的理解，下面见整个代码和注释。

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu Sep 14 16:33:11 2017
Softmax Regression using TensorFlow
@author: z
"""
##import data
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data  
mnist =input_data.read_data_sets("MNIST_data/", one_hot=True) 

#Print the shape of mist
print('\n','mnist.train.images.shape,mnist.train.labels.shape')
print(mnist.train.images.shape,mnist.train.labels.shape)
#(55000, 784) (55000, 10)     ##28*28=784
print('mnist.test.images.shape, mnist.train.labels.shape')
print(mnist.test.images.shape, mnist.train.labels.shape)
#(10000, 784) (55000, 10)
print('mnist.validation.images.shape, mnist.validation.labels.shape')
print(mnist.validation.images.shape, mnist.validation.labels.shape)
#(5000, 784) (5000, 10)

## 2.Create the model
#create sess
sess=tf.InteractiveSession()
#put-data , first var is datatype, second is tensor shape
x=tf.placeholder(tf.float32,[None,784])
#Weights and bias are initialized to zero
W=tf.Variable(tf.zeros([784,10]))  
b=tf.Variable(tf.zeros([10]))
#realize Softmax Regression: y=sotfmax(Wx+b)
y=tf.nn.softmax(tf.matmul(x,W)+b)

# Define loss
y_ = tf.placeholder(tf.float32, [None, 10])
cross_entropy = tf.reduce_mean(
    -tf.reduce_sum(y_ * tf.log(y), reduction_indices=[1]))
#Define optimizer
train_step=tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy) 
tf.global_variables_initializer().run()  
## 3.Train
for i in range(1000):
    batch_xs, batch_ys = mnist.train.next_batch(100)
    train_step.run({x: batch_xs, y_: batch_ys})
## 4.Test trained model
correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
print(accuracy.eval({x: mnist.test.images,y_: mnist.test.labels}))