利用sklearn.preprocessing.PolynomialFeatures生成交叉特征

2018​文章源自随机的未知-https://sjdwz.com/11106.html

 当我们使用一次多项式拟合一组数据时，可能不太理想，如下图：文章源自随机的未知-https://sjdwz.com/11106.html

​文章源自随机的未知-https://sjdwz.com/11106.html

如果用直线来进行拟合的话：文章源自随机的未知-https://sjdwz.com/11106.html

​文章源自随机的未知-https://sjdwz.com/11106.html

如果用三次函数来拟合的话：文章源自随机的未知-https://sjdwz.com/11106.html

​文章源自随机的未知-https://sjdwz.com/11106.html

如何用python的sklearn库来做呢？文章源自随机的未知-https://sjdwz.com/11106.html

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression

x = np.linspace(-3,3,num = 50)
#用三次多项式来拟合
from sklearn.pipeline import Pipeline

def PolyLin(degree):
    return Pipeline(
    [
        ("poly",PolynomialFeatures(degree=degree)),
        ("Linearmodel",LinearRegression())
    ])
lin_3 = PolyLin(degree=3)
lin_3.fit(X,y)
y_predict3 = lin_3.predict(X)
plt.scatter(X,y)
y_predict_3 = lin_3.predict(X)

更多关于sklearn.preprocessing.PolynomialFeatures的描述请看：文章源自随机的未知-https://sjdwz.com/11106.html

 文章源自随机的未知-https://sjdwz.com/11106.html

sklearn关于preprocessing.PolynomialFeatures的文档文章源自随机的未知-https://sjdwz.com/11106.html

​文章源自随机的未知-https://sjdwz.com/11106.html