机器学习1-线性回归 

 1 # _*_ coding:utf-8 _*_
 2 """
 3 author:Bevishe
 4 date:2019-05-04
 5 """
 6 
 7 import numpy as np
 8 import pandas as pd
 9 import tensorflow as tf
10 import matplotlib.pyplot as plt
11 '''
12 使用tensorflow来自己实现一个简单的线性回归
13 '''
14 #返回0-1之间100个符合均匀分布的随机数
15 x_data = np.random.rand(100).astype(np.float32)
16 y_data = x_data*0.1 +0.3
17 
18 #create the tensorflow structure
19 Weights = tf.Variable(tf.random_uniform([1],-1.0,1.0))
20 biase = tf.Variable(tf.zeros([1]))
21 
22 y = Weights*x_data +biase
23 
24 loss = tf.reduce_mean(tf.square(y - y_data))
25 
26 optimizer = tf.train.GradientDescentOptimizer(0.5)
27 train = optimizer.minimize(loss)
28 
29 init = tf.initialize_all_variables()
30 
31 #end
32 
33 sess = tf.Session()
34 sess.run(init)
35 
36 for step in range(201):
37     sess.run(train)
38     y_ = x_data*sess.run(Weights) + sess.run(biase)
39     plt.plot(x_data,y_data,color = 'r')
40     plt.plot(x_data,y_,color = 'b')
41     plt.show()
42     if step%20 == 0:
43         print(step,sess.run(Weights),sess.run(biase))

代码地址：https://github.com/bevishe/Leetcode/blob/master/tensorflow/linearRegression/%E7%BA%BF%E6%80%A7%E5%9B%9E%E5%BD%92.py

 

 

 

posted @ 2019-05-04 22:27  BevisHe  阅读(133)  评论(0)  编辑  收藏  举报