太监，编写博客太累了，发现网上有很多斯坦福机器学习的学习博客和笔记。比较好的有这两篇。
http://blog.csdn.net/artprog/article/details/51280821
https://mooc.guokr.com/note/16274/
这里贴一个自己写的正规方程的python实现，利用numpy，很短
import numpy as np
# y = 3 + 2 * x1 + 1 * x2
x_train = np.array([[1, 2], [2, 1], [2, 3], [3, 5], [1, 3], [4, 2], [7, 3], [4, 5], [11, 3], [8, 7]])
y_train = np.array([7, 8, 10, 14, 8, 13, 20, 16, 28, 26])
X = np.mat(np.zeros((len(x_train), 3)))
# 形成新向量
def newArray(l):
c = len(l) + 1
newA = np.array(np.zeros(c))
newA[0] = 1
i = 1
while i < c:
newA[i] = l[i - 1]
i += 1
return newA
i = 0
# 化为矩阵
while i < np.shape(X)[0]:
thisArray = newArray(x_train[i])
X[i] = thisArray
# a1[i][1] = x_train[i][0]
# a1[i][2] = x_train[i][1]
i += 1
Y = np.mat(y_train).T
XT = X.T
Theta = (XT * X).I * XT * Y
print(Theta)