Create Image Grid with Matplotlib

Introduction

This tutorial will demonstrate how to create a grid of images using Matplotlib's ImageGrid. We will create a 2x2 grid of images and explore various ways to add colorbars to the grid.

VM Tips

After the VM startup is done, click the top left corner to switch to the Notebook tab to access Jupyter Notebook for practice.

Sometimes, you may need to wait a few seconds for Jupyter Notebook to finish loading. The validation of operations cannot be automated because of limitations in Jupyter Notebook.

If you face issues during learning, feel free to ask Labby. Provide feedback after the session, and we will promptly resolve the problem for you.

Import necessary libraries and data

We first need to import the necessary libraries and data to create our grid. We will use matplotlib.pyplot for plotting, cbook to get a sample data set, and ImageGrid to create our grid.

import matplotlib.pyplot as plt
from matplotlib import cbook
from mpl_toolkits.axes_grid1 import ImageGrid

## Get sample data
Z = cbook.get_sample_data("axes_grid/bivariate_normal.npy")  ## 15x15 array
extent = (-3, 4, -4, 3)

Create a grid of 2x2 images with a single colorbar

Our first grid will be a 2x2 grid of images with a single colorbar. We will use the ImageGrid function to create the grid and specify the number of rows and columns we want. We will also specify the location of the colorbar and set share_all to True to share the colorbar across all images.

## Create a grid of 2x2 images with a single colorbar
grid = ImageGrid(
    fig,  ## Figure object
    141,  ## Location of subplot
    nrows_ncols=(2, 2),  ## Number of rows and columns
    axes_pad=0.0,  ## Padding between axes
    label_mode="L",  ## Label mode
    share_all=True,  ## Share colorbar across all images
    cbar_location="top",  ## Location of colorbar
    cbar_mode="single"  ## Colorbar mode
)

## Plot images on grid
for ax in grid:
    im = ax.imshow(Z, extent=extent)

## Add colorbar to grid
grid.cbar_axes[0].colorbar(im)
for cax in grid.cbar_axes:
    cax.tick_params(labeltop=False)

Create a grid of 2x2 images with each image having its own colorbar

Our next grid will be a 2x2 grid of images with each image having its own colorbar. We will use the ImageGrid function again, but this time we will set cbar_mode to "each" to specify that each image should have its own colorbar.

## Create a grid of 2x2 images with each image having its own colorbar
grid = ImageGrid(
    fig,  ## Figure object
    142,  ## Location of subplot
    nrows_ncols=(2, 2),  ## Number of rows and columns
    axes_pad=0.1,  ## Padding between axes
    label_mode="1",  ## Label mode
    share_all=True,  ## Share colorbar across all images
    cbar_location="top",  ## Location of colorbar
    cbar_mode="each",  ## Colorbar mode
    cbar_size="7%",  ## Size of colorbar
    cbar_pad="2%"  ## Padding between colorbar and images
)

## Plot images on grid and add colorbars
for ax, cax in zip(grid, grid.cbar_axes):
    im = ax.imshow(Z, extent=extent)
    cax.colorbar(im)
    cax.tick_params(labeltop=False)

Create a grid of 2x2 images with each image having its own colorbar and a different colorbar range

Our final grid will also be a 2x2 grid of images with each image having its own colorbar, but this time we will use a different colorbar range for each image. We will set the colorbar range using vmin and vmax when plotting each image.

## Create a grid of 2x2 images with each image having its own colorbar and a different colorbar range
grid = ImageGrid(
    fig,  ## Figure object
    143,  ## Location of subplot
    nrows_ncols=(2, 2),  ## Number of rows and columns
    axes_pad=(0.45, 0.15),  ## Padding between axes
    label_mode="1",  ## Label mode
    share_all=True,  ## Share colorbar across all images
    cbar_location="right",  ## Location of colorbar
    cbar_mode="each",  ## Colorbar mode
    cbar_size="7%",  ## Size of colorbar
    cbar_pad="2%"  ## Padding between colorbar and images
)

## Plot images on grid and add colorbars
limits = ((0, 1), (-2, 2), (-1.7, 1.4), (-1.5, 1))  ## Different colorbar ranges
for ax, cax, vlim in zip(grid, grid.cbar_axes, limits):
    im = ax.imshow(Z, extent=extent, vmin=vlim[0], vmax=vlim[1])
    cb = cax.colorbar(im)
    cb.set_ticks((vlim[0], vlim[1]))

Summary

In this tutorial, we learned how to create a grid of images using Matplotlib's ImageGrid. We explored different ways to add colorbars to the grid, including using a single colorbar for all images, giving each image its own colorbar, and giving each image its own colorbar with a different colorbar range.