如何将图例添加到 Mayavi 图中

Question

I can't seem to get a legend to appear on a mayavi plot. I have created a simple 3d plot but I need to provide a legend for the plot. Matplotlib supports adding by:

import matplotlib.pyplot as plt
plt.plot(x, y, z, label="my-label")

I didn't find any working examples of such legends in mayavi.

Answer 1

Since mayavi doesn't have a built in method for creating legends the easiest way to do this would be to create your mavavi figure and then read it into matplotlib and make a legend. This requires you to have programmatic access to the color values used in creating the mayavi plot - it is much easier when you specify the colors explicitly instead of letting the color cycler automatically assign colors.

Here is an example which creates a mayavi figure and then passes it to matplotlib to then create a legend.

import matplotlib.pyplot as plt
from matplotlib.lines import Line2D
import mayavi.mlab as mlab
from mayavi.api import OffScreenEngine
import numpy as np

mlab.options.offscreen = True
e = OffScreenEngine()
e.start()
fig = mlab.figure(engine=e, size=(5000,5000), fgcolor=(0,0,0), bgcolor=(1,1,1))
mlab.view(azimuth=45, elevation=0)

n = 500
d = np.linspace(0, np.pi*8, n)

mlab.plot3d(d, np.cos(d**1.2), np.cos(d), color = colors[0])
mlab.plot3d(d, np.sin(d), np.sin(d), color = colors[1])
mlab.plot3d(d, np.sin(d)*np.cos(d), np.cos(d), color = colors[2])
mlab.plot3d(d, np.sin(d)**2, np.sin(d)**2, color = colors[3])

mlab.savefig("./example.png")

im = plt.imread("./example.png")
labels = ["One", "Two", "Three", "Four"]

elements = [Line2D([0], [0], label = l, color = c) for l, c in zip(labels, colors[:4])]
plt.imshow(im)
plt.legend(handles = elements)
plt.gca().set_axis_off()
plt.subplots_adjust(top = 1, bottom = 0, right = 1, left = 0, 
            hspace = 0, wspace = 0)
plt.margins(0,0)
plt.savefig("./example.png")

colors is an array of RGB values in the range [0.0, 1.0] . The example uses the Tableau 20 color palette mapped into the [0.0, 1.0] space. This generates the following output