*** Definition of gaussian distribution ***

*** Sample code for drawing gaussian distribution ***
#!/usr/bin/env python
#coding:utf-8
import pylab as pl
import numpy as np
from enthought.mayavi import mlab
def gauss2d(x):
return 1/np.sqrt(2*np.pi) * np.exp(-(x**2)/2)
def gauss3d(x, y):
return 1/np.sqrt(2*np.pi) * np.exp(-(x**2 + y**2)/2)
def plot_gauss2d():
x = np.mgrid[-4:4:100j]
# matplotlib functions
pl.plot(gauss2d(x), 'bo-')
pl.ylabel('gauss2d(x)')
pl.xlabel('x')
pl.show()
def plot_gauss3d():
x, y = np.mgrid[-4:4:100j, -4:4:100j]
"""
x, y = np.mgrid[-2:2:5j, -2:2:5j]
array([[[-2., -2., -2., -2., -2.],
[-1., -1., -1., -1., -1.],
[ 0., 0., 0., 0., 0.],
[ 1., 1., 1., 1., 1.],
[ 2., 2., 2., 2., 2.]],
[[-2., -1., 0., 1., 2.],
[-2., -1., 0., 1., 2.],
[-2., -1., 0., 1., 2.],
[-2., -1., 0., 1., 2.],
[-2., -1., 0., 1., 2.]]])
x[0,0] ==> -2.0
y[0,0] ==> -2.0
"""
# matplotlib functions
pl.plot(gauss3d(x, y))
pl.ylabel('gauss3d(x)')
pl.xlabel('x')
pl.show()
def plot_gauss3d_mayavi():
x, y = np.mgrid[-4:4:100j, -4:4:100j]
z = gauss3d(x, y)
mlab.surf(z, warp_scale='auto')
mlab.outline()
mlab.axes()
mlab.show()
def main():
#plot_gauss2d()
#plot_gauss3d()
plot_gauss3d_mayavi()
if __name__ == "__main__":
main()
*** Sample code output ***
- output from gauss2d function

- output from gauss3d function

- output from gauss3d_mayavi function

No comments:
Post a Comment