Press "Enter" to skip to content

用Numpy实现线性回归

用Numpy实现线性回归

 

现在二维平面上有一系列点point,我们要找到一个一次函数$y=wx+b$,使得所有点到这条直线的距离平方和$\sum(wx+b-y)^2$最小

因此我们可以定义 损失函数
$loss = (wx+b-y)^2$,计算损失的代码如下:

 

# compute loss
def compute_error_for_line_given_points(b, w, points):
    totalError = 0
    for i in range(len(points)):
        x = points[i, 0]
        y = points[i, 1]
        totalError += (y - (w * x + b)) ** 2
    return totalError / float(len(points)) # average

 

然后用梯度下降法更新$w$和$b$

 

$w’ = w – lr*\frac{\partial loss}{\partial w}$,$b’ = b – lr*\frac{\partial loss}{\partial b}$,其中

 

$\frac{\partial loss}{\partial w} = 2 * x * (wx + b – y)$,$\frac{\partial loss}{\partial b} = 2 * (wx + b – y)$

 

# compute gradient
def step_gradient(b_current, w_current, points, learningRate):
    b_gradient = 0
    w_gradient = 0
    N = float(len(points))
    for i in range(len(points)):
        x = points[i, 0]
        y = points[i, 1]
        b_gradient += 2 * ((w_current * x) + b_current - y)
        w_gradient += 2 * x * ((w_current * x) + b_current - y)
    b_gradient = b_gradient / N
    w_gradient = w_gradient / N
    new_b = b_current - (learningRate * b_gradient)
    new_w = w_current - (learningRate * w_gradient)
    return [new_b, new_w]

 

最后只要设定迭代次数,不断的重复更新$w$和$b$就行了

 

def gradient_descent_runner(points, starting_b, starting_w, learning_rate, num_iterations): # num_iteration 迭代次数
    b = starting_b
    w = starting_w
    for i in range(num_iterations):
        b, w = step_gradient(b, w, np.array(points), learning_rate)
    return [b, w]

 

主函数

 

def run():
    points = np.genfromtxt("data.txt", delimiter=",")
    learning_rate = 0.0001
    initial_b = random()
    initial_w = random()
    num_iterations = 1000
    print("Starting gradient descent at b = {0}, w = {1}, error = {2}"
          .format(initial_b, initial_w, 
                  compute_error_for_line_given_points(initial_b, initial_w, points)))
    print("Running...")
    [b, w] = gradient_descent_runner(points, initial_b, initial_w, learning_rate, num_iterations)
    print("After {0} iterations at b = {1}, w = {2}, error = {3}"
          .format(num_iterations, b, w, 
                  compute_error_for_line_given_points(b, w, points)))
run()

 

data.txt数据文件下载

 

Be First to Comment

发表回复

您的电子邮箱地址不会被公开。 必填项已用*标注