The Matplotlib package is great for visualizing data. One of its many features is the ability to annotate points on your graph. You can use annotations to explain why a particular data point is significant or interesting.

If you haven’t used Matplotlib before, you should check out my introductory article, Matplotlib – An Intro to Creating Graphs with Python or read the official documentation.

Let’s get started!

Installing Matplotlib

If you don’t have Matplotlib on your computer, you must install it. Fortunately, you can use pip, the Python package manager utility that comes with Python.

Open up your terminal or command prompt and run the following command:

python -m pip install matplotlib

Pip will now install Matplotlib and any dependencies that Matplotlib needs to work properly. Assuming that Matplotlib installs successfully, you are good to go!

Annotating Points on a Graph

Matplotlib comes with a handy annotate()method that you can use. As with most of Matplotlib’s methods, annotate()can take quite a few different parameters.

For this example, you will be using the following parameters:

• text – The label for the annotation
• xy – The x/y coordinate of the point of interest
• arrowprops – A dictionary of arrow properties
• xytext – Where to place the text for the annotation

Now that you know what you’re doing, open up your favorite Python IDE or text editor and create a new Python file. Then enter the following code:

import matplotlib.pylab as plt

import numpy as np



def annotated():

    fig = plt.figure(figsize=(8, 6))



    numbers = list(range(10))

    plt.plot(numbers, np.exp(numbers))

    plt.title("Annotating an Exponential Plot using plt.annotate()")

    plt.xlabel("x-axis")

    plt.ylabel("y-axis")



    plt.annotate("Point 1", xy=(6, 400),

                 arrowprops=dict(arrowstyle="->"),

                 xytext=(4, 600))



    plt.annotate("Point 2", xy=(7, 1150),

                 arrowprops=dict(arrowstyle="->",

                                 connectionstyle="arc3,rad=-.2"),

                 xytext=(4.5, 2000))



    plt.annotate("Point 3", xy=(8, 3000),

                 arrowprops=dict(arrowstyle="->",

                                 connectionstyle="angle,angleA=90,angleB=0"),

                 xytext=(8.5, 2200))



    plt.show()



if __name__ == "__main__":

    annotated()

Here, you are creating a simple line graph. You want to annotate three points on the graph. The arrowprops define the arrowstyleand, in the latter two points, the connectionstyle. These properties tell Matplotlib what type of arrow to use and whether it should be connected to the text as a straight line, an arc, or a 90-degree turn.

When you run this code, you will see the following graph:

You can see how the different points are located and how the arrowprops lines are changed. You should check out the full documentation to learn all the details about the arrows and annotations.

Wrapping Up

Annotating your graph is a great way to make your plots more informative. Matplotlib allows you to add many different labels to your plots, and annotating the interesting data points is quite nice.

You should spend some time experimenting with annotations and learning all the different parameters it takes to fully understand this useful feature.