参考:https://ask.csdn.net/questions/7464230.
import numpy as np
# import csv
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib
matplotlib.rcParams['font.sans-serif'] = ['SimHei']
matplotlib.rcParams['font.family']=['sans-serif']
dataframe = pd.read_csv("LR_data_exam.csv", sep=",")
x = dataframe['battery'].tolist()
y = dataframe['KM'].tolist()
# 画出数据的散点图
fig = plt.figure() # 整个绘图区分成一行两列,当前图是第一个。
fig.set_size_inches(10, 10) # 整个绘图区域的宽度10和高度4
# # 画出数据和拟合直线的图
plt.scatter(x, y)
t = np.linspace(0,25,100)
plt.plot(t, 5.5*t) #线性回归线,5.5是斜率
#plt.plot(t, k*t+b) #线性回归线,k是斜率,b截距
plt.xlabel("真实值")
plt.ylabel("预测值")
plt.title("拟合结果")
plt.show()
求斜率和截距:
class MLR:
def __init__(self,x,y,intercept=True):
self.x=x
self.y=y
self.ic=intercept
def fit(self):
if self.ic:
ones=np.ones(len(self.x) )
X=np.c_[ones,self.x]
else:
X=self.x
y=self.y
self.a=np.linalg.inv(X.T@X)@X.T@y
def predict(self,x_new):
if self.ic:
ones=np.ones(x_new.shape)
x=np.c_[ones,x_new]
else:
x=x_new
y=x @ self.a
return y
model = MLR(x,y)
model.fit() #回归
k = float(model.a[1]) #斜率
b = float(model.a[0]) #截距
print("斜率:{}\n截距:{}".format(k,b))