CODEWITHSUNDEEP
pythonmatplotlibpositioncolorbar
pceccon
pceccon

Reputation: 9844

Positioning color bars - Matplotlib

I have a plot in which I merge two datas. Given that, I have to show two different color bars, one for each data. I'm currently plotting my datas as follows:

plt.figure()

# Data 1
fig = plt.imshow(data1, interpolation='nearest', cmap='binary', vmin=0, vmax=1)

# Plotting just the nonzero values of data2
data2_x = numpy.nonzero(data2)[0]
data2_y = numpy.nonzero(data2)[1]

pts = plt.scatter(data2_x, data2_y, marker='s', c=data2[data2_x, data2_y])

plt.colorbar(pts)
plt.colorbar(fig, orientation="horizontal")

And this is the plot that I get:

enter image description here

However, I would like to reposition the color bars to have something like this (made with Photoshop):

enter image description here

Is that possible?

Thank you in advance.

Upvotes: 2

Views: 598

Answers (2)

tacaswell
tacaswell

Reputation: 87376

Probably the 'easiest' way to do this is to lay the axes to be used for the color bars out by hand (via cbax = fig.add_axes([....])). You can then pass that axes to the color bar calls:

Something like:

from matplotlib import pyplot as plt
import numpy as np

fig = plt.figure(figsize=(8, 8))
ax = fig.add_axes([.1, .1, .8, .8])
im = ax.imshow(np.random.rand(150, 150), cmap='gray', interpolation='none')
sc = ax.scatter(2 + 146 * np.random.rand(150), 2 + 146 * np.random.rand(150),
                c=np.random.rand(150), cmap='Accent', s=50, lw=0)

ax_cb1 = fig.add_axes([.1, .05, .8, .02])
ax_cb2 = fig.add_axes([.92, .1, .02, .8])

cb1 = fig.colorbar(im, cax=ax_cb1, orientation='horizontal')
cb1.ax.xaxis.set_label_position('top')
cb2 = fig.colorbar(sc, cax=ax_cb2, orientation='vertical')

enter image description here

Upvotes: 4

MaxNoe
MaxNoe

Reputation: 14997

you can link the colorbar to the axes with the ax-keyword, plt.gca() gives you the current axes:

plt.colorbar(object1, ax=plt.gca())

Upvotes: 0

Related Questions