在 Numpy 中获取多项式回归统计

Question

我希望在 Numpy 中使用多项式回归 (polyfit) 来获取系数和截距，但我不确定如何编写脚本来获取多项式函数。

我已经将代码作为线性回归的代码，附在下面：

import matplotlib.pyplot as plt
import sys
from numpy import *
import numpy as np
import numpy.polynomial.polynomial as poly
import pylab
from scipy import stats
from scipy.interpolate import *
from datetime import datetime, timedelta

#Open dataset1,dataset2 data

data1 = np.loadtxt('/home/script/2_columns', delimiter=',', skiprows=0)

data2 = np.loadtxt('/home/script/2_columns_a', delimiter=',', skiprows=0)

#Define first column as dataset1

#Define second column as dataset2

x = data1[:,0]
y = data1[:,1]

#The stuff...
slope, intercept, r_value, p_value, std_err = stats.linregress(x,y)

Answer 1

您需要统计数据吗？如果没有，你只需要系数，使用 numpy.polyfit:

真的很简单

import numpy as np

# from your code
data1 = np.loadtxt('/home/script/2_columns', delimiter=',', skiprows=0)

x = data1[:,0]
y = data1[:,1]

degree = 3
coeffs = np.polyfit(x, y, degree)

# now, coeffs is an array which contains the polynomial coefficients
# in ascending order, i.e. x^0, x^1, x^2
intercept, linear, quadratic, cubic = coeffs

如果您确实需要其他值，请说明您需要什么，因为例如 r_value 是 x 和 y 之间的相关系数，这不是很当您不希望数据是线性的时很有用。

在 Numpy 中获取多项式回归统计

Obtaining Polynomial Regression Stats in Numpy

python

numpy

scipy

polynomials