多元线性回归的实现

代码

import numpy as np
from sklearn.model_selection import train_test_split
from numpy.linalg import inv
import matplotlib.pyplot as plt
#读数据，预处理数据
data = np.loadtxt("D:/机器学习/data/aqi2.csv",delimiter=",",skiprows=1,dtype=float)
#print(data)
index = np.ones((data.shape[0],1))#构造全1矩阵
data = np.hstack((data,index))#合并两个矩阵
#print(data)
y = data[:,0]#提取真实值y
x = data[:,1:]#提取特征向量x
#数据集分训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(
    x, y, test_size=0.33, random_state=42)
weight = np.dot(np.dot(inv(np.dot(X_train.T,X_train)),X_train.T),y_train)#计算权重
print(weight)
#y的预测值
y_prediect = np.dot(X_test,weight)

plt.scatter(range(len(y_test)),y_test,c='red')#真实散点图
plt.plot(range(len(y_test)),y_prediect,c='black')#拟合直线图
plt.show()

真实值与预测值图形的拟合结果结果：