Matplotlib - Colorbars

---

---

A colorbar is a visual representation of the color scale used in a plot. It displays the color scale from the minimum to the maximum values in the data, helping us understand the color variations in the plot.

In the following image you can observe a simple colorbar that is highlighted with a red color rectangle −

Colorbars in Matplotlib

The Matplotlib library provides a tool for working with colorbars, including their creation, placement, and customization.

The matplotlib.colorbar module is responsible for creating colorbars, however a colorbar can be created using the Figure.colorbar() or its equivalent pyplot wrapper pyplot.colorbar() functions. These functions are internally uses the Colorbar class along with make_axes_gridspec (for GridSpec-positioned axes) or make_axes (for non-GridSpec-positioned axes).

And a colorbar needs to be a "mappable" (i.e, matplotlib.cm.ScalarMappable) object typically an AxesImage generated via the imshow() function. If you want to create a colorbar without an attached image, you can instead use a ScalarMappable without an associated data.

Example - Usage of Horizontal Colorbar

Here is an simple example that creates a horizontal colorbar without an attached plotusing the ScalarMappable class.

import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt

# Create a figure and axis for the colorbar
fig, ax = plt.subplots(figsize=(6, 1), constrained_layout=True)

# Define a colormap and normalization for the colorbar
cmap = mpl.cm.cool
norm = mpl.colors.Normalize(vmin=5, vmax=10)

# Create a ScalarMappable without associated data using the defined cmap and norm
scalar_mappable = mpl.cm.ScalarMappable(norm=norm, cmap=cmap)

# Add a horizontal colorbar to the figure 
colorbar = fig.colorbar(scalar_mappable, cax=ax, orientation='horizontal', label='Some Units')

# Set the title and display the plot
plt.title('Basic Colorbar')
plt.show()

Output

On executing the above code we will get the following output −

Example - Creating a Simple Colorbar

Here is another example creates a simple colorbar for the plot using the pyplot.colorbar() function with default parameters.

import matplotlib.pyplot as plt
import numpy as np

# Generate sample data
data = np.random.random((10, 10))

# Create a plot with an image and a colorbar
fig, ax = plt.subplots(figsize=(7,4))
im = ax.imshow(data, cmap='viridis')

# Add a colorbar to the right of the image
cbar = plt.colorbar(im, ax=ax)

# Show the plot
plt.show()
print('Successfully drawn the colorbar...')

Output

Successfully drawn the colorbar...

Automatic Colorbar Placement

Automatic placement of colorbars is a straightforward approach. This ensures that each subplot has its own colorbar, providing a clear indication of the quantitative extent of the image data in each subplot.

Example - Automatic Colorbar Placement

This example demonstrates the automatic colorbar placement for multiple subplots.

import matplotlib.pyplot as plt
import numpy as np

# Create a 2x2 subplot grid
fig, axs = plt.subplots(1, 2, figsize=(7,3))
cmaps = ['magma', 'coolwarm']

# Add random data with different colormaps to each subplot
for col in range(2):
   ax = axs[col]
   pcm = ax.pcolormesh(np.random.random((20, 20)) * (col + 1), cmap=cmaps[col])

   # Add a colorbar for the each subplots
   fig.colorbar(pcm, ax=ax, pad=0.03)

plt.show()
print('Successfully drawn the colorbar...')

Output

Successfully placed the colorbar...

Manual Colorbar Placement

This approach allows us to explicitly determine the location and appearance of a colorbar in a plot. Which may be necessary when the automatic placement does not achieve the desired layout.

By creating inset axes, either using inset_axes() or add_axes(), and then assigning it to the colorbar through the cax keyword argument, users can get the desired output.

Example - Manual Colorbar Placement

Here is an example that demonstrates how to determine the colorbar placement manually in a plot.

import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt

# Generate random data points
npoints = 1000
x, y = np.random.normal(10, 2, (2, npoints))

# Create a subplot
fig, ax = plt.subplots(figsize=(7,4))

# Set title
plt.title('Manual Colorbar Placement')

# Draw the plot
hexbin_artist = ax.hexbin(x, y, gridsize=20, cmap='gray_r', edgecolor='white')

# Manually create an inset axes for the colorbar
cax = fig.add_axes([0.8, 0.15, 0.05, 0.3])

# Add a colorbar using the hexbin_artist and the manually created inset axes
colorbar = fig.colorbar(hexbin_artist, cax=cax)

# Display the plot
plt.show()

Output

On executing the above code we will get the following output −

Customizing Colorbars

The appearance of colorbars, including ticks, tick labels, and labels, can be customized to specific requirements. Vertical colorbars typically have these elements on the y-axis, while horizontal colorbars display them on the x-axis. The ticks parameter is used to set the ticks, and the format parameter helps format the tick labels on the visible colorbar axes.

Example - Horizontal and Labelled Colorbar

This example uses the imshow() method to display the data as an image, and places a colorbar horizontally to the image with a specified label.

import matplotlib.pyplot as plt
import numpy as np

# Create a subplot
fig, ax = plt.subplots(figsize=(7, 4))

# Generate random data
data = np.random.normal(size=(250, 250))
data = np.clip(data, -1, 1)

# Display the data using imshow with a specified colormap
cax = ax.imshow(data, cmap='afmhot')
ax.set_title('Horizontal Colorbar with Customizing Tick Labels')

# Add a horizontal colorbar and set its orientation and label
cbar = fig.colorbar(cax, orientation='horizontal', label='A colorbar label')

# Adjust ticks on the colorbar
cbar.set_ticks(ticks=[-1, 0, 1])
cbar.set_ticklabels(['Low', 'Medium', 'High'])

# Show the plot
plt.show()

Output

On executing the above code we will get the following output −

Example - Customizing a Color Bar

This example demonstrates how to customize the position, width, color, number of ticks, font size and more properties of a colorbar.

import numpy as np
from matplotlib import pyplot as plt

# Adjust figure size and autolayout
plt.rcParams["figure.figsize"] = [7.00, 3.50]
plt.rcParams["figure.autolayout"] = True

# Generate random data
data = np.random.randn(4, 4)

# Plot the data with imshow
im = plt.imshow(data, interpolation='nearest', cmap="PuBuGn")

# Add colorbar and adjust its position
# Decrease colorbar width and shift position to the right
clb = plt.colorbar(im, shrink=0.9, pad=0.05)  

# Set the top label for colorbar 
clb.ax.set_title('Color Bar Title')

# Customize color of ticks
clb.ax.set_yticks([0, 1.5, 3, 4.5], labels=["A", "B", "C", "D"])

# Change color and font size of ticks
clb.ax.tick_params(labelcolor='red', labelsize=20)  

plt.show()

Output

On executing the above code you will get the following output −