Tensorflow学习教程简单练一波，线性模型

#coding:utf-8

import tensorflow as tf
import numpy as np

#使用numpy 生成100个随机点

x_data = np.random.rand(100)
y_data = x_data*0.1 + 0.2

#构造一个线性模型,k 和 b是我们要求的两个值，利用100个点
#进行不断迭代 然后求出最佳的k和b
b = tf.Variable(0.)
k = tf.Variable(0.)
y = k*x_data + b 

#二次代价函数
loss = tf.reduce_mean(tf.square(y_data-y))
#定义一个训练的优化器:梯度下降法
optimizer = tf.train.GradientDescentOptimizer(0.2) #学习率为0.2
#最小化代价函数
#使用前面定义好的优化器来减小loss值
train = optimizer.minimize(loss) 
init = tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init)
    for step in range(300):
        sess.run(train)
        if step%20 ==0:
            print(step,sess.run([k,b]))

结果

(0, [0.046660472, 0.097824201])
(20, [0.097689927, 0.20109874])
(40, [0.098617099, 0.20065789])
(60, [0.0991721, 0.20039386])
(80, [0.099504374, 0.20023578])
(100, [0.09970329, 0.20014116])
(120, [0.09982238, 0.20008451])
(140, [0.099893652, 0.20005059])
(160, [0.099936351, 0.20003028])
(180, [0.099961899, 0.20001812])
(200, [0.099977195, 0.20001085])
(220, [0.099986345, 0.2000065])
(240, [0.099991821, 0.20000389])
(260, [0.099995099, 0.20000234])
(280, [0.099997051, 0.2000014])

可以看到求得的k和b非常接近于我们定义的k值和b值。