#定义一个计算总的error（即总的loss），其中points为一系列的（x，y）的组合:
def compute_error_for_line_given_points(b, w, points):
    totalError = 0      				#定义总的loss
    for i in range(0, len(points)):
        x = points[i, 0]  				#取该点的x值
        y = points[i, 1]  				#取该点的y值
        totalError += (y - (w * x + b)) ** 2 		#将每一个点的loss累加
    return totalError / float(len(points))			#对总的error求一个平均，返回总的平均error

#定义一个计算梯度的函数：入口参数：b当前值，W当前值，点集合，学习率
def step_gradient(b_current, w_current, points, learningRate):
    b_gradient = 0
    w_gradient = 0
    N = float(len(points))      #总的数据点的个数
    for i in range(0, len(points)):
        x = points[i, 0]
        y = points[i, 1]
        #∂L/∂b = 2（Wx + b - y），所有梯度累加时除以N，以便结果不用再做平均
        b_gradient += -(2/N) * (y - ((w_current * x) + b_current))
        #∂L/∂W = 2（Wx + b - y）x，所有梯度累加时除以N，以便结果不用再做平均
        w_gradient += -(2/N) * x * (y - ((w_current * x) + b_current))
    new_b = b_current - (learningRate * b_gradient)
    new_W = w_current - (learningRate * w_gradient)
    return [new_b, new_W]
#循环迭代W，b，入口参数：（x，y）点集合，初始b,初始W，学习率，迭代次数：
def gradient_descent_runner(points, starting_b, starting_w, learning_rate, num_iterations):
    b = starting_b
    w = starting_w
    for i in range(num_iterations):
        b, w = step_gradient(b, w, np.array(points), learning_rate) 	#np.array（point）为（x，y）的数组
    return [b, w]
def run():
    points = np.genfromtxt("data.csv", delimiter=",")       #取数据
    learning_rate = 0.0001
    initial_b = 0 # initial y-intercept guess
    initial_w = 0 # initial slope guess
    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 b = {1}, m = {2}, error = {3}".
          format(num_iterations, b, w,
                 compute_error_for_line_given_points(b, w, points))
          )

if __name__ == '__main__':
    run()