Usage code for numpy polynomial fitting function polyfit

import numpy as np
def fit(x,y,m,w):
    if len(x)<=m:
        return False
    xishu = (x,y,m,w=w)
    p = np.poly1d(xishu) # Construct polynomials
    yfit = p(x) # fitted y-values
    yresid = y - yfit # Residuals
    SSresid = sum(pow(yresid, 2)) # Residual sum of squares
    SStotal = len(y) * (y) # Overall average variance
    if SStotal==0:# Horizontal line
        r2=1
    else:
        r2 = 1 - SSresid/SStotal # Goodness of fit
    return (xishu,r2,yfit)
if __name__=="__main__":
    x = (0,5,0.1)
    z = [2+3*x+4*x**2 for x in x]
    y = ([(z,3) for z in z])
    weight=[1 for one in x]
    print(fit(x,y,2,weight))

polyfit(x,y,m,w) The parameter m is the number of polynomials +1 and w is the weight. The meaning of weight is that the error of this data point should be multiplied by the weight, and giving high weight to the low point can reduce the error of the low point and avoid the relative error of the low point is too large.

The return values (xishu,r2,yfit) of the custom function fit(x,y,m,w) are the coefficients, correlation coefficients, and fitted Y-values, respectively.

Crossing the origin can also be achieved by weights.The code is as follows:

if must00:
    x=(x,0)#.append(0)
    y=(y,0)#(0)
    w=(w,100000000)#.append(10000)
fit_r=fit(x,y,m,w)
if fit_r:
    (xishu,r,y_fitted)=fit_r
else:
    #fit error
    xishu=[0,1,0]
    y_fitted=x 
    r=0

summarize

To this article on the use of numpy polynomial fitting function polyfit is introduced to this article, more related numpy polynomial fitting function polyfi content, please search for my previous articles or continue to browse the following related articles I hope you will support me more in the future!