python梯度下降法

import numpy as np

x1 = np.array([2.104, 1.600, 2.400, 1.416, 3.000])
x2 = np.array([3, 3, 3, 2, 4])
y1 = np.array([4.00, 3.30, 3.69, 2.32, 5.40])  #数据存到矩阵

w0 = 0
w1 = 0
w2 = 0
alpha = 0.01 #赋值

def fx(x1, x2):
    return w0 + w1 * x1 + w2 * x2  # 求和

for i in range(1000):    # 迭代1000次         # 下为梯度下降公式
        w0 = w0 - alpha * (fx(x1[i % 5], x2[i % 5]) - y1[i % 5])
        w1 = w1 - alpha * (fx(x1[i % 5], x2[i % 5]) - y1[i % 5]) * x1[i % 5]
        w2 = w2 - alpha * (fx(x1[i % 5], x2[i % 5]) - y1[i % 5]) * x2[i % 5]

        gred0 = (fx(x1[i % 5], x2[i % 5]) - y1[i % 5])  # 求梯度
        gred1 = (fx(x1[i % 5], x2[i % 5]) - y1[i % 5]) * x1[i % 5]
        gred2 = (fx(x1[i % 5], x2[i % 5]) - y1[i % 5]) * x2[i % 5]

#判断偏导趋近于0
        if (abs(gred0) < 0.01 and abs(gred1) < 0.01 and abs(gred2) < 0.01):
            print(w0, w1, w2)
            print(gred0, gred1, gred2)
            break