How to remove a colorbar

Question

I want to remove a color bar from an axes in that way that the axes will return to its default position.

To make it clear, have a look at this code (or better run it):

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import make_axes_locatable

# Init figure
figure, axes = plt.subplots()
axes.set_aspect(1)

# Create some random stuff
image = axes.imshow(np.random.random((10, 10)))
plt.pause(2)

# Add a color bar to the axes
cax = make_axes_locatable(axes).append_axes("right", size="5%", pad=0.05)
colorBar = figure.colorbar(image, cax=cax)
colorBar.ax.tick_params(length=0, labelright=False)
plt.pause(2)

# Remove the color bar
colorBar.remove()

plt.show()

As you see, the following happens: In that moment I add the color bar to the axes, the axes itself is changing its position a litte bit. It is moving to the left to create some space for the color bar. In a more complex case, e.g. in a Gridspec, it might also change its size a bit.

What I want to get: In that moment I am removing the colorbar again, I want to get the axes back in its original position (and its original size).

This does not happen automatically when I remove the color bar. The axes is still on its left position and not back in the center. How do I achieve that?

I need something like the inverse of make_axes_locateable.

This looks like a [XY problem](https://meta.stackexchange.com/questions/66377/what-is-the-xy-problem) to me. Maybe you could tell us more about what you are trying to achieve with this complicated setup. Maybe a better idea is to prevent the colorbar from resizing the Axes in the first place? — Diziet Asahi, Aug 07 '19 at 21:34
Yes, I have a complex case with many axes in a Gridspec (2x4). Some of them will get a colorbar, others will not. I need to change the size of those axes having a colorbar, so that axes+colorbar has the same size like axes without colorbar. Later, I am removing some color bars again and I want the axes beeing back in their original size and position. This does not happen. — Max16hr, Aug 07 '19 at 21:41

score 2 · Accepted Answer · answered Aug 08 '19 at 00:17

make_axes_locatable sets the axes locator to a new locator. To reverse this step you need to keep track of the old locator (original_loc = ax.get_axes_locator()) and reset it to the axes (ax.set_axes_locator(original_loc)) after you removed the colorbar.

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import make_axes_locatable

# Init figure
fig, ax = plt.subplots()
ax.set_aspect(1)

# Create some random stuff
image = ax.imshow(np.random.random((10, 10)))
plt.pause(2)

# Add a color bar to the axes
original_loc = ax.get_axes_locator()
cax = make_axes_locatable(ax).append_axes("right", size="5%", pad=0.05)
colorBar = fig.colorbar(image, cax=cax)
colorBar.ax.tick_params(length=0, labelright=False)
plt.pause(2)

# Remove the color bar
colorBar.remove()
ax.set_axes_locator(original_loc)
plt.pause(0.0001)

plt.show()

How to remove a colorbar

1 Answers1