Top label for matplotlib colorbars

Question

By default, matplotlib would position colorbar labels alongside the vertical colorbars. What is the best way to force the label to be on top of a colorbar? Currently my solution needs adjusting labelpad and y values depending on size of the label:

import numpy as np
import matplotlib.pylab as plt 

dat = np.random.randn(10,10)
plt.imshow(dat, interpolation='none')

clb = plt.colorbar()
clb.set_label('label', labelpad=-40, y=1.05, rotation=0)

plt.show()

Is there a better, more generic way to do this?

Worldly

Answer 1

You could set the title of the colorbar axis (which appears above the axis), rather than the label (which appears along the long axis). To access the colorbar's Axes, you can use clb.ax. You can then use set_title, in the same way you can for any other Axes instance.

For example:

import numpy as np
import matplotlib.pylab as plt 

dat = np.random.randn(10,10)
plt.imshow(dat, interpolation='none')

clb = plt.colorbar()
clb.ax.set_title('This is a title')

plt.show()

Pound

You can also clb.ax.set_xlabel('Values') to put it at bottom. – Melidamelilot

Answer 2

Colorbars are stored in the list of Axes in the Figure. So colorbar title/label can be set by accessing the Axes containing the colorbar as well. Since it is an Axes object, set() method is defined on it which can be used to modify numerous properties such as title, xlabel, ylabel, ticks etc.

import numpy as np
import matplotlib.pylab as plt

dat = np.random.randn(10,10)
plt.imshow(dat, interpolation='none')
plt.colorbar()
plt.gcf().axes     # [<Axes: >, <Axes: label='<colorbar>'>]
plt.gcf().axes[1].set(title='This is a title', xlabel='caption', ylabel='colors');

Using subplots, the same plot can be coded as follows.

fig, ax = plt.subplots()
ax.imshow(dat, interpolation='none')
fig.colorbar(ax.images[0], ax=ax)
fig.axes[1].set(title='This is a title', xlabel='caption', ylabel='colors');

Enumerate