How to save a plot to a file using Matplotlib

While the feature-list of matplotlib is nearly limitless, we’ll quickly go over how to use the library to generate a basic chart for your own testing purposes.

Like all Python libraries, you’ll need to begin by installing matplotlib. We won’t go through the installation process here, but there’s plenty of information in the official documentation.

Once installed, import the matplotlib library. You’ll likely also want to import the pyplot sub-library, which is what you’ll generally be using to generate your charts and plots when using matplotlib.

Now to create and display a simple chart, we’ll first use the .plot() method and pass in a few arrays of numbers for our values. For this example, we’ll plot the number of books read over the span of a few months.

We can also add a few axis labels:

Finally, we can display the chart by calling .show():

With a simple chart under our belts, now we can opt to output the chart to a file instead of displaying it (or both if desired), by using the .savefig() method.

The .savefig() method requires a filename be specified as the first argument. This filename can be a full path and as seen above, can also include a particular file extension if desired. If no extension is provided, the configuration value of savefig.format is used instead.

In addition to the basic functionality of saving the chart to a file, .savefig() also has a number of useful optional arguments.

• dpi can be used to set the resolution of the file to a numeric value.
• transparent can be set to True, which causes the background of the chart to be transparent.
• bbox_inches can be set to alter the size of the bounding box (whitespace) around the output image. In most cases, if no bounding box is desired, using bbox_inches='tight' is ideal.

• If bbox_inches is set to 'tight', then the pad_inches option specifies the amount of padding around the image.

There are a handful of additional options for specific occasions, but overall this should get you started with easily generating image file outputs from your matplotlib charts.