python 一元非线性

import pandas

import matplotlib;

data = pandas.read_csv(

'D:/PDM/4.3/data.csv'

)

x = data[["等级"]]

y = data[["资源"]]

font = {

'family' : 'SimHei'

};

matplotlib.rc('font', **font);

matplotlib.rcParams['axes.unicode_minus'] = False

from pandas.tools.plotting import scatter_matrix;

scatter_matrix(

data[["等级", "资源"]],

alpha=0.8, figsize=(10, 10), diagonal='kde'

)

import numpy;

x_ = numpy.arange(-10, 10, 0.01);

y_ = x_**2

from matplotlib import pyplot as plt;

plt.figure();

plt.title('等级与资源')

plt.xlabel('等级')

plt.ylabel('资源')

plt.grid(True)

plt.plot(x_, y_, 'k.')

plt.show()

from sklearn.linear_model import LinearRegression

from sklearn.preprocessing import PolynomialFeatures

x

pf = PolynomialFeatures(degree=2)

x_2_fit = pf.fit_transform(x)

x_2_fit

lrModel = LinearRegression()

lrModel.fit(x_2_fit, y)

lrModel.score(x_2_fit, y)

0.98438623929845115

x_2_predict = pf.fit_transform([[21], [22], [23]])#预测数值，21,22,23是输入值，但是需要转换以后才能进行预测。

lrModel.predict(x_2_predict)