CODEWITHSUNDEEP
pythonmatplotlib
FooBar
FooBar

Reputation: 16478

Plotting 4d-data

I have points in 4 dimensions (lets call them, v,w,y,z) , which I would like to visualize.

My plan is to have two squares, (v x w, y x z), next to each other and then just plot each point twice.

Given two points ([1, 1, 1, 3], [2, 2, 2, 2]) I envision something like this:

1337 paint skillz

Given a small set of points, I could use different colors to show which points on the left correspond to the right. With a large set of points, that would be futile. But perhaps heat maps would then be the best to visualize it?

Or is there some alternative established way to visualize data of higher dimensions within python/matplotlib?

Here's some sample data:

>>> resultsArray[:,:4]
array([[ 0.        ,  0.        ,  0.        ,  0.        ],
       [ 0.00495236,  0.03919034,  0.00495287,  0.03919042],
       [ 0.00240293,  0.02667374,  0.00220419,  0.02693434],
       [ 0.0011231 ,  0.0191784 ,  0.00104353,  0.01928256],
       [ 0.00547274,  0.04187615,  0.00657255,  0.04043363],
       [ 0.00291993,  0.0286196 ,  0.00292006,  0.02861962],
       [ 0.00128136,  0.01975574,  0.00121107,  0.01984781],
       [ 0.00591335,  0.04531384,  0.00873814,  0.04160714],
       [ 0.00345499,  0.0310103 ,  0.00396032,  0.03034784],
       [ 0.00149387,  0.02056065,  0.0014939 ,  0.02056065],
       [ 0.00274306,  0.02667374,  0.00220419,  0.02659422],
       [ 0.00123893,  0.01948363,  0.00108284,  0.01952189],
       [ 0.00162006,  0.02379926,  0.00143157,  0.02389168],
       [ 0.00347023,  0.0286196 ,  0.00292006,  0.02806932]])

Upvotes: 2

Views: 10012

Answers (1)

tnknepp
tnknepp

Reputation: 6253

What about a 3-D scatter plot, which actually comes out to be 4-dimensional when the color scale is included?

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

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')

sp = ax.scatter(data[:,0],data[:,1],data[:,2], s=20, c=data[:,3])
plt.colorbar(sp)

You can customize the color scale and projection orientation as you like.

Upvotes: 5

Related Questions