The Matplotlib Library

Matplotlib is a expansive set of software tools that makes it possible to produce quality figures as easily as possible. The library packages together a suite of modules that enables us to do things like create simple plots, make animations, interact with figures, and save graphics in a variety of formats. The primary module that we will work with directly, pyplot, provides a MatLAB-like set of software tools that can be easily and quickly used to make MatLAB-like line and surface plots, histograms, scatter plots and more. The focus of the next couple lessons will be on learning how to create these things with pyplot.

Anatomy of a Figure

Before we dig into using pyplot, it is really helpful to understand exactly what a figure is from a software perspective. After all, our job as a developer is to create meaningful graphics and to be able to do that in python, we have to know what parts of a figure, plot, axis system, etc to modify. Like everything in software, all the individual pieces of a figure have a specific set of characteristics that we are able to control.

First, let’s look at the important visual components of a figure.

Anatomy of a figure (from the Matplotlib docs)

Undoubtably you are already familiar with most of the different aspects of a figure shown above. I want to make special note of a few things. First, all of the things in blue in the image above are things that we can manipulate. Obviously, we can control the lines and markers that are plotted by changing the data that we are working with. Naturally, we should be able to specify the axis labels, titles, and legends. We also have full control over things like how many major tick marks to use, how many minor ticks, whether either or both of those things should have labels, how the plot is gridded, and how splines are used. We can also control the characteristics of each of these plot elements, from the text used in the labels, to the type of line used to make the grid, to whether or not we want a box around the legend. Generally, everything is customizable.

Additionally, you can see two other objects that can be modified: Figure and Axes. I think it is useful to think about plotting in this way: A Figure can contain one or more sets of Axes and an axis can contain one or more plot types: lines, markers, contour, etc. All of these things are objects in python and they can all be manipulated. For example, let’s say you want to create a line plot. To do so, first you would create a Figure object. On that Figure, you would add an Axis. Then on that Axis, you would add your Line.

It might seem like I’m being a bit pedantic here, but understanding this organization is important because as you try to create interesting graphics, you will have to learn to control these objects and their properties. All that said, because the action of creating a Figure with an Axis and a Line (or scatter or contour, etc) is such a common task, Matplotlib uses several shortcuts to make all this happen without requiring you to do it explicitly. For example, if you use the pyplot.plot() function to plot a line, a Figure and Axis will be created for you automatically behind the scenes. This is nice because it makes it easy to quickly create a line plot, but it can be confusing if you want to do more complex things like make a line plot that has two x-axes or something like that. For now, know that whenever you plot something, even though you may not have explicitly told python to create a Figure or Axis etc. object, those objects do exist and you can modify them.

Matplotlib docs

In the next few lessons we will only scratch the surface of the capabilities of Matplotlib. As typical, if you want to learn more there is expansive documentation available online at https://matplotlib.org/index.html complete with many, many examples. One of the best ways to figure out how to make a particular plot using Matplotlib is to find an example online and start from there. The Matplotlib developers know this and thus have created a beautiful gallery that has all of these examples in one place: https://matplotlib.org/gallery.html. Be sure to visit the gallery to get a glimpse of the many different things you can do with this powerful library and click on one or more of the examples to see the code used to create it.