梯度下降法

最新推荐文章于 2024-05-27 20:03:59 发布

半巷巷巷巷

最新推荐文章于 2024-05-27 20:03:59 发布

阅读量283

点赞数

文章标签： numpy python 深度学习

本文链接：https://blog.csdn.net/m0_58945584/article/details/127155737

版权

借用梯度下降求解以下实际问题

python代码：

import numpy as np

def hy(x1,x2):
  return th0+th1*x1+th2*x2
if __name__=='__main__':
 th0 = 0
 th1 = 0
 th2 = 0
 delta1,delta2,delta3=1,1,1
 x1 = np.array([2.104, 1.600, 2.400, 1.416, 3.000])
 x2 = np.array([3, 3, 3, 2, 4])
 y = np.array([4.00, 3.30, 3.69, 2.32, 5.40])
 alpha = 0.01
 for i in range(500):
     th0=th0-alpha*(hy(x1[i%5],x2[i%5])-y[i%5])*1
     delta1=(hy(x1[i%5],x2[i%5])-y[i%5])
     #print(delta1)
     th1=th1-alpha*(hy(x1[i%5],x2[i%5])-y[i%5])*x1[i%5]
     delta2=(hy(x1[i%5],x2[i%5])-y[i%5])*x1[i%5]
     th2=th2-alpha*(hy(x1[i%5],x2[i%5])-y[i%5])*x2[i%5]
     delta3=(hy(x1[i%5],x2[i%5])-y[i%5])*x2[i%5]

     if (abs(delta1) < 0.01 and abs(delta2) < 0.01 and abs(delta3) < 0.01):
        print(th0,th1,th2)

预测结果：