How to 4D plot with contour over cube, using matplotlib?

I would like to 4D plot over the cube (x,y,z) vs. q, using the colormap on the 3 surfaces of the cubes, where the color and contour are determined depending on the q variable. Basically, I am looking for a similar image like this:

Any help is appreciated.

Solution

See my example of 3D ABC feild

import pyvista as pv
import numpy as np
from numpy import mgrid
import matplotlib.pyplot as plt

print('initializing domain')
xmin = -800.
xmax = 800.
Lx = xmax-xmin
B0 = 1
k = 1
alpha = 2.0*np.pi*k/Lx
x, y, z = Lx*mgrid[0:1:51j, 0:1:51j, 0:1:51j]


print('initializing 3D B field')
Bx = B0*(np.sin(alpha*z) + np.cos(alpha*y))
By = B0*(np.sin(alpha*x) + np.cos(alpha*z))
Bz = B0*(np.sin(alpha*y) + np.cos(alpha*x))
B = np.column_stack((Bx.ravel(), By.ravel(), Bz.ravel()))
grid = pv.StructuredGrid(x, y, z)
grid["ABC field magnitude"] = np.linalg.norm(B, axis=1)
grid["ABC field vectors"] = B
grid.set_active_vectors("ABC field vectors")
#contours = grid.contour(8, scalars="ABC field magnitude")
#arrows = contours.glyph(orient="ABC field vectors", factor=50.0)

print('plotting')
pv.set_plot_theme('document')
p = pv.Plotter(notebook=0, shape=(1,1))
#p.background_color='white'
#p.window_size
cmap = plt.cm.get_cmap("viridis", 4)
p.add_mesh(grid, cmap=cmap)
p.show_grid()
#p.add_mesh(arrows)
#p.subplot(0,1)
#slices = grid.slice_orthogonal(x=20, y=20, z=30)
#slices = grid.slice_orthogonal()
#p.add_mesh(slices, cmap=cmap)
##p.subplot(1,0)
#p.add_mesh(contours, opacity=1)
#p.subplot(1,1)
#p.add_mesh(arrows)
#single_slice = arrows.slice(normal=[1, 1, 0])
#slices = arrows.slice_orthogonal(x=20, y=20, z=30)
#slices = grid.slice_orthogonal()
#p.add_mesh(single_slice, cmap=cmap)
p.show_grid()

p.link_views()
p.view_isometric()
p.show(screenshot='abc3d_slicing.png')