Problemw with python surface plot

Question

I am new to programming in python. I am trying to show a surface plot by modifying an example found on the matplotlib website. Can you tell me what's wrong?

My code is:

import matplotlib.pyplot as plt
from matplotlib import cm
import numpy

def H(n, f, l, delta, H_abs, H_ph):
    c0 = 2.99796e8
    n0 = 1.00027 + 0j
    n1 = complex(n[0], n[1])

    Sum = 0
    for i in range(1, delta+1):
        Sum = Sum + ((n0-n1)*exp(complex(0,-1*2*pi*f*n1*l/c0))/(n1+n0))**i

    H_Theo = 4*n0*n1*exp(complex(0,2*pi*f*l)*(n0-n1)/c0)*(1+Sum)/(n0+n1)

    M = abs(H_Theo) - H_abs
    A = cmath.phase(H_Theo) - H_ph

    return abs(M) + abs(A)

freq = 1213188709257.4814
l_real = 6.77e-4
d = 0
H_abs = 0.798653104536778
H_ph = 2.1845744701729926

n_test = numpy.arange(1.5, 2.5, 0.01).tolist()
k_test = numpy.arange(-0.05, 0.05, 0.001).tolist()

X = n_test
Y = k_test
X, Y = numpy.meshgrid(X, Y)

errore = []
for i in range(len(n_test)):
    errore.append(H([X[i],Y[i]], freq, l_real, d, H_abs, H_ph))

Z = errore

fig2 = plt.figure()
az = fig2.gca(projection='3d')
az.plot_surface(X, Y, Z, rstride=8, cstride=8, alpha=0.3)
cset = az.contour(X, Y, Z, zdir='z', offset=min(Z)-1, cmap=cm.coolwarm)
cset = az.contour(X, Y, Z, zdir='x', offset=min(X)-1, cmap=cm.coolwarm)
cset = az.contour(X, Y, Z, zdir='y', offset=max(Y)+0.05, cmap=cm.coolwarm)

az.set_xlabel('n')
az.set_xlim(min(X)-1, max(X)+1)
az.set_ylabel('k')
az.set_ylim(min(Y)-0.05, max(Y)+0.05)
az.set_zlabel('Err')
az.set_zlim(min(Z)-1, max(Z)+1)

plt.show()

The error I get is:

n1 = complex(n[0], n[1])
TypeError: only length-1 arrays can be converted to Python scalars

I have seen other Q&A on the subject but I was unable to solve the problem, anyway.

Maybe the question is trivial but I will really appreciate any help in order to be able to see the graph that, at the end, should turn out to be similar to this one: http://matplotlib.org/examples/mplot3d/contour3d_demo3.html

Greg Whittier · Accepted Answer · 2013-11-15 15:14:37Z

1

Your problem is that X[i] and Y[i] are 1-d arrays that you pass in to H as n. I.e., n = [X[i], Y[i]]. Then you call complex(n[0], n[1]) which expects scalar arguments, not arrays. If you intend for n to be a 1-d array of complex numbers, you can write n = n[0] + 1j*n[1]. You still need modifications to the rest of your code, however. You're missing some imports and seem to have assumed "from numpy import *"

Anyway, I've made enough modifications to get it to run and plot something, but I really doubt this is what you're looking for. You need to think about what function it is you're trying to plot.

from mpl_toolkits.mplot3d import axes3d
import matplotlib.pyplot as plt
from matplotlib import cm
import numpy

def H(n, f, l, delta, H_abs, H_ph):
    c0 = 2.99796e8
    n0 = 1.00027 + 0j
    n1 = n[0] + 1j * n[1]

    Sum = 0
    for i in range(1, delta+1):
        Sum = Sum + ((n0-n1)*numpy.exp(-1*2*numpy.pi*f*n1*l/c0)/(n1+n0))**i

    H_Theo = 4*n0*n1*numpy.exp(2*numpy.pi*f*l*(n0-n1)/c0)*(1+Sum)/(n0+n1)

    M = numpy.abs(H_Theo) - H_abs
    A = numpy.angle(H_Theo) - H_ph

    return abs(M) + abs(A)

freq = 1213188709257.4814
l_real = 6.77e-4
d = 0
H_abs = 0.798653104536778
H_ph = 2.1845744701729926

n_test = numpy.arange(1.5, 2.5, 0.01).tolist()
k_test = numpy.arange(-0.05, 0.05, 0.001).tolist()

X = n_test
Y = k_test
X, Y = numpy.meshgrid(X, Y)

errore = []
for i in range(len(n_test)):
    errore.append(H([X[i],Y[i]], freq, l_real, d, H_abs, H_ph))

Z = errore

fig2 = plt.figure()
az = fig2.gca(projection='3d')
az.plot_surface(X, Y, Z, rstride=8, cstride=8, alpha=0.3)
cset = az.contour(X, Y, Z, zdir='z', offset=numpy.min(Z)-1, cmap=cm.coolwarm)
cset = az.contour(X, Y, Z, zdir='x', offset=numpy.min(X)-1, cmap=cm.coolwarm)
cset = az.contour(X, Y, Z, zdir='y', offset=numpy.max(Y)+0.05, cmap=cm.coolwarm)

az.set_xlabel('n')
az.set_xlim(numpy.min(X)-1, numpy.max(X)+1)
az.set_ylabel('k')
az.set_ylim(numpy.min(Y)-0.05, numpy.max(Y)+0.05)
az.set_zlabel('Err')
az.set_zlim(numpy.min(Z)-1, numpy.max(Z)+1)

plt.show()

edited Nov 15, 2013 at 15:14

answered Nov 15, 2013 at 14:43

Greg Whittier

3,1951 gold badge22 silver badges14 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

user2996448 Over a year ago

Thank for your suggestion. Now there is a new error: Sum = Sum + ((n0-n1)*exp(complex(0,-1*2*pifn1*l/c0))/(n1+n0))**i TypeError: only length-1 arrays can be converted to Python scalars

user2996448 Over a year ago

Thenk you for your very fast answer. I have modified the code following your indications and a little bit more. Then I will look into the function but, for now, I have one more error: cset = az.contour(X, Y, Z, zdir='z', offset=min(Z)-1, cmap=cm.coolwarm) ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Greg Whittier Over a year ago

Replace set = az.contour(X, Y, Z, zdir='z', offset=min(Z)-1, cmap=cm.coolwarm) with cset = az.contour(X, Y, Z, zdir='z', offset=numpy.min(Z)-1, cmap=cm.coolwarm)

Collectives™ on Stack Overflow

Problemw with python surface plot

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related