线性回归及其python实现

1. 一元线性回归模型

$$y=f(x)=b+wx$$

(1). 代价函数

$$J(w)=\frac{1}{2m}\sum _{i=1}^m(f^{(i)}(x)-y^{(i)}{)}^2$$

(2) 梯度下降

(i) 偏导

∂ J ∂ w = 1 m ∑ i = 1 m ( f ( i ) ( x ) − y ( i ) ) x ( i ) ∂ J ∂ b = 1 m ∑ i = 1 m ( f ( i ) ( x ) − y ( i ) ) \frac{\partial{J}}{\partial{w}} = \frac{1}{m}\sum_{i=1}^m{(f^{(i)}(x) - y^{(i)})x^{(i)}}\\ \frac{\partial{J}}{\partial{b}} = \frac{1}{m}\sum_{i=1}^m{(f^{(i)}(x) - y^{(i)})} ∂w∂J​=