sklearn库实现线性回归

第一步 导入库，函数依赖

import numpy as np
import matplotlib.pyplot as plt
import sklearn.linear_model as slm

2.导入数据，简单查看一个导入情况，数据分布情况

points_set = np.genfromtxt('data.csv',delimiter=',')

x_set = points_set[:,0]
y_set = points_set[:,1]

plt.scatter(x_set,y_set)
plt.show()

3.定义损失函数

def compute_SquareCost(w,b,points_set):
    total_cost = 0
    this_cost = 0
    print(f'我进来了',end='\n')
    #开始准点计算损失
    for i in range(len(points_set)):
        x = points_set[i,0]
        y = points_set[i,1]
        this_cost = (y-(w*x+b)) ** 2
        print(f'本次的损失函数值为：{this_cost}',end='\n')
        total_cost+=this_cost
    
    return total_cost/len(points_set)

4.使用sklearn库进行计算

lr = slm.LinearRegression()

x_newSquare = x_set.reshape(-1,1)
y_newSquare = y_set.reshape(-1,1)

lr.fit(x_newSquare,y_newSquare)

5.提取出相关系数，进行验证

w = lr.coef_[0][0]
b = lr.intercept_[0]

print("w is: ", w)
print("b is: ", b)

cost = compute_SquareCost(w,b,points_set)

print("cost is: ", cost)

6.画出最后的图像

plt.scatter(x_set,y_set)

pred_y = w*x_set+b

plt.plot(x_set,pred_y,c='r')
plt.show()