pylab 3d scatter plots with 2d projections of plotted data

Question

I am trying to create a simple 3D scatter plot but I want to also show a 2D projection of this data on the same figure. This would allow to show a correlation between two of those 3 variables that might be hard to see in a 3D plot.

I remember seeing this somewhere before but was not able to find it again.

Here is some toy example:

x= np.random.random(100)
y= np.random.random(100)
z= sin(x**2+y**2)

fig= figure()
ax= fig.add_subplot(111, projection= '3d')
ax.scatter(x,y,z)

Answer 1

You can add 2D projections of your 3D scatter data by using the plot method and specifying zdir:

import numpy as np
import matplotlib.pyplot as plt

x= np.random.random(100)
y= np.random.random(100)
z= np.sin(3*x**2+y**2)

fig= plt.figure()
ax= fig.add_subplot(111, projection= '3d')
ax.scatter(x,y,z)

ax.plot(x, z, 'r+', zdir='y', zs=1.5)
ax.plot(y, z, 'g+', zdir='x', zs=-0.5)
ax.plot(x, y, 'k+', zdir='z', zs=-1.5)

ax.set_xlim([-0.5, 1.5])
ax.set_ylim([-0.5, 1.5])
ax.set_zlim([-1.5, 1.5])

plt.show()

Answer 2

The other answer works with matplotlib 0.99, but 1.0 and later versions need something a bit different (this code checked with v1.3.1):

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

x= np.random.random(100)
y= np.random.random(100)
z= np.sin(3*x**2+y**2)

fig= plt.figure()
ax = Axes3D(fig)
ax.scatter(x,y,z)

ax.plot(x, z, 'r+', zdir='y', zs=1.5)
ax.plot(y, z, 'g+', zdir='x', zs=-0.5)
ax.plot(x, y, 'k+', zdir='z', zs=-1.5)

ax.set_xlim([-0.5, 1.5])
ax.set_ylim([-0.5, 1.5])
ax.set_zlim([-1.5, 1.5])

plt.show() 

You can see what version of matplotlib you have by importing it and printing the version string:

import matplotlib
print matplotlib.__version__