数学建模——利用模拟数据拟合曲面

 

import numpy as np
import  matplotlib.pyplot as plt
from scipy.interpolate import  interp1d
from mpl_toolkits import  mplot3d
from scipy.interpolate import griddata
from scipy.optimize import  curve_fit


m=200
n=300
x=np.linspace(-6,6,m)
y=np.linspace(-8,8,n)
x2,y2=np.meshgrid(x,y)

x3=np.reshape(x2,(1,-1))
y3=np.reshape(y2,(1,-1))

xy=np.vstack((x3,y3))


def pfun(t,m1,m2,s):
    return np.exp(-((t[0]-m1)**2+(t[1]-m2)**2)/(2*s**2))

z=pfun(xy,1,2,3)
zr=z+0.2*np.random.normal(size=z.shape)


popt,pcov=curve_fit(pfun,xy,zr)

#print(popt)

zn=pfun(xy,*popt)

zn2=np.reshape(zn,x2.shape)

plt.rc('font',size=16)

ax=plt.axes(projection='3d')

ax.plot_surface(x2,y2,zn2,cmap='gist_rainbow')

plt.show()




'''

a = np.array([[1, 2, 3, 4, 5, 6],
              [7, 8, 9, 10, 11, 12],
              [13, 14, 15, 16, 17, 18],
              [19, 20, 21, 22, 23, 24]])


b=np.reshape(a,(6,-1))

print(b)

'''

我还掌握了numpy库中的reshape函数，收获很大

最后结果：