大数据——机器学习|机器学习之用解析解求解多元线性回归模型线性回归|python|机器学习

代码如下：

import numpy as np from sklearn.model_selection import train_test_split from numpy.linalg import inv import matplotlib.pyplot as plt# 1. 读数据、预处理数据（aqi2.csv读取出来） # 这里的delimiter="," 代表的是用逗号分隔，skiprows=1代表的是前面第一行的数据不用带入。 data = https://www.it610.com/article/np.loadtxt(r"D:\Pycharm\code\MY_machine_learning\mydata\aqi2.csv", delimiter=",", skiprows=1, dtype=np.float32) # print(data) index = np.ones((data.shape[0], 1)) data = https://www.it610.com/article/np.hstack((data, index)) # print(data)# 2. 特征与标签 y = data[:, 0] x = data[:, 1:]# 3. 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) print(weight)# 4. 预测值y_predict y_predict = np.dot(X_test, weight) print('-----------------------------------------') print('打印真实值：\n', y_test) print('打印预测值：\n', y_predict)# 5. 画图分析预测与真实值偏离程度 plt.scatter(range(len(y_test)), y_test, c='red')# 真实值散点图 plt.plot(range(len(y_test)), y_predict, c='green')# 预测值拟合直线图 plt.show()

运行结果如下：