Before we start plotting, we need to import the data we intend to plot and get familiar with basic plotting commands in Matplotlib. Let's start going through these basic commands!

While working on data visualization projects, we need to ensure that we have basic familiarity and understanding of the tools used for data processing. Before we begin, let's briefly revise the most common data structures you will encounter when handling data with Python.

This is the most basic Python data structure; it stores a collection of values. While you can store any data type as an element in a Python list, for our purpose of data visualization, we mostly handle lists of numerical values as data input, or at, most...

We have just revised the basic ways of data handling with Python. Without further ado, let's create our first "Hello World!" plot example.

To create a pandas DataFrame from objects such as lists and ndarrays, you may call:

import pandas as pd

To start creating Matplotlib plots, we first import the plotting API pyplot by entering this command:

import matplotlib.pyplot as plt

This will start your plotting routine.

We have just learned how to turn numerical values into dots and lines with Matplotlib. By default, Matplotlib optimizes the display by calculating various values in the background, such as the reasonable axis range and font sizes. However, good visualization often requires more design input to suit our custom data visualization needs and purpose. Moreover, text labels are needed to make figures informative in many cases. In the following sections, we will demonstrate the methods to adjust these elements.

While Matplotlib automatically chooses the range of x and y axis limits to spread data onto the whole plotting area, sometimes we want some adjustment...

To get further acquainted with Matplotlib functions, let us plot a multiline plot with axes, labels, title, and legend configured in one single snippet.

In this example, we take real-world data from the World Bank on agriculture. As the world population continues to grow, food security continues to be an important global issue. Let us have a look at the production data of a few major crops in the recent decade by plotting a multiline plot with the following code:

Data source: https://data.oecd.org/agroutput/crop-production.htm
OECD (2017), Crop production (indicator). doi: 10.1787/49a4e677-en (Accessed on 25 December 2017)
# Import relevant modules
import pandas as pd
import matplotlib.pyplot as plt

# Import dataset
crop_prod = pd.read_csv('OECD-THND_TONNES.txt',delimiter='\t')
years = crop_prod[crop_prod['Crop']=='SOYBEAN...

To save a figure, we put plt.savefig(outputpath) at the end of plotting commands. It can be used in place of plt.show(), to directly save the figure without displaying it.

If you want to save the figure as a file as well as display it on the notebook output, you can call both plt.savefig() and plt.show().

plt.savefig() automatically detects the file extension of the specified output path, and generates the corresponding file format if it is supported. If no file extension is specified in the input, a PNG format file would be obtained as output with...

We have learned to tweak a few major elements in a Matplotlib plot. When we recurrently generate figures of similar style, it would be nice to have a way to store and apply the persistent global settings. Matplotlib offers a few options for configuration.

To keep settings throughout the current session, we can execute matplotlib.rcParams to override configuration file settings.

For instance, we can set the font size of all text in plots to 18 with the following:

matplotlib.rcParams['font.size'] = 18

Alternatively we can call the matplotlib.rc() function. As matplotlib.rc() just changes one property, to change multiple settings, we can use the function matplotlib...

Congratulations! We are now familiar with the basic plotting techniques using Matplotlib syntax! Remember, the success of a data visualization project relies heavily upon making appealing visuals.

In the next chapters, we will learn how to beautify our plots and select the right kind of plot that communicates our data effectively!