CODEWITHSUNDEEP
pythonmatplotlibgraph3dvisualization
xrdty
xrdty

Reputation: 886

Visualising 2 parameters and their results

There's 2 parameters where I want to try different values for:

a = [0.0, 0.5, 0.6]  # len == 3
b = [0.0, 0.02 , 0.05, 0.1]  # len == 4

For each value of a, try each value of b. This comes with 3 * 4 = 12 different results.

My data comes in the format of 

res = [(0.0, 0.0, res1), (0.0, 0.02, res2), ...]

Is there any way I can neatly visualise this? I was thinking of a contour/heat map or 3d plane but sadly I cannot get that to work.

Upvotes: 0

Views: 180

Answers (1)

ImportanceOfBeingErnest
ImportanceOfBeingErnest

Reputation: 339705

There are many different options. The first step in any case needs to be to convert your res list into a numpy array.

For many plots like imshow, pcolor(mesh) or contourf, you need to have three 2D arrays, which you can obtain via reshaping of your input array (given that it is ordered correctly).

The following shows some options you have:

enter image description here

res = [(0.0, 0.0, 0.5), (0.0, 0.02, 0.7), (0.0, 0.05, 0.6), (0.0, 0.10, 0.8),
       (0.5, 0.0, 0.4), (0.5, 0.02, 0.6), (0.5, 0.05, 0.5), (0.5, 0.10, 0.7),
       (0.6, 0.0, 0.3), (0.6, 0.02, 0.5), (0.6, 0.05, 0.4), (0.6, 0.10, 0.6)]

import matplotlib.pyplot as plt
import numpy as np

res = np.array(res)
A = res[:,0].reshape(3,4) #A in y direction
B = res[:,1].reshape(3,4)
Z = res[:,2].reshape(3,4)

fig, ((ax, ax2), (ax3, ax4)) = plt.subplots(2,2)
#imshow
im = ax.imshow(Z, origin="lower")
ax.set_xticks(range(len(Z[0,:])))
ax.set_yticks(range(len(Z[:,0])))
ax.set_xticklabels(B[0,:])
ax.set_yticklabels(A[:,0])

#pcolormesh, first need to extend the grid
bp = np.append(B[0,:], [0.15])
ap = np.append(A[:,0], [0.7])
Bp, Ap = np.meshgrid(bp, ap)
ax2.pcolormesh(Bp, Ap, Z)

#contour
ax3.contourf(B, A, Z, levels=np.linspace(Z.min(), Z.max(),5))
#scatter
ax4.scatter(res[:,1], res[:,0], c=res[:,2], s=121)

ax.set_title("imshow")
ax2.set_title("pcolormesh")
ax3.set_title("contourf")
ax4.set_title("scatter")
plt.tight_layout()
fig.colorbar(im, ax=fig.axes, pad=0.05)
plt.show()

Upvotes: 1

Related Questions