A Comprehensive Guide On Pygal – The Next Generation Data Visualization Library In Python

It can be downloaded for Python 3.6, 3.7, 3.8, 3.9, and pypy and its Linux Distribution Packages for the following versions of Linux are:

• Fedora
• Gentoo
• Ubuntu
• Debian

• Arch Linux

You can download its main package that holds all available charts, the config class, and map extensions namespace modules.

When To Use Pygal

Pygal is one Python data visualization library that allows you to create stylish and interactive python plots with minimal coding efforts.

When working with a complex dataset, choose Pygal because it is highly customizable and provides you with different charts (line, bar, Histogram, pie, box, dot, etc.), styles (Built-in style, parametric styles, etc), chart configurations, serie configurations, value configurations (labels, style, node attributes, links), sparklines (spark text, etc), tables (style, total, transposed), and output formats.

Benefits Of Pygal

You must now know that Pygal allows you to create beautiful interactive charts and its completely open-source library. Now, we will emphasize some of the best features, making them worth using in your projects.

• It has three separate map packages to keep the module small in size and easy to use for users.

• It gives interactivity for most graphs that provide you with data exploration, filtering out certain features, zooming in/out, and much more.
• It is optimized with rich support and community to allow users to create and work with SVG images.
• It allows you to be visionary by providing three main concepts- Data, layout, and figure objects. Using these concepts, you can solve multiple complex visualization problems with ease.

Comparison With Other Popular Python Data Visualization Libraries In The Market

Line_Chart = pygal.Line() 
Line_Chart.title = ' Line Chart' 
Line_Chart.x_labels = map(str, range(2005, 2016)) 
Line_Chart.add('A', [None, None, 0, 16.1, 22, 30, 36.9, 45.1, 46.1, 42.33, 37.11]) 
Line_Chart.add('B', [None, None, None, None, None, None, 0, 3.5, 10.1, 23.1, 35.2]) 
Line_Chart.add('C', [85.1, 84.1, 84.1, 74.1, 64, 58.0, 54.1, 44.3, 36.9, 26.5, 20.4]) 
Line_Chart.add('D', [14.1, 15.8, 15.1, 8.8, 8, 10.1, 8.4, 5.1, 6.6, 6.1, 7.4]) 
Line_Chart.render_to_file('Line_Chart.svg')

Line_Chart = pygal.HorizontalLine() 
Line_Chart.title = 'Horizontal Line Bar' 
Line_Chart.x_labels = map(str, range(20005, 2016)) 
Line_Chart.add('A', [None, None, 0, 16.1, 22, 30, 36.9, 45.1, 46.1, 42.33, 37.11]) 
Line_Chart.add('B', [None, None, None, None, None, None, 0, 3.5, 10.1, 23.1, 35.2]) 
Line_Chart.add('C', [85.1, 84.1, 84.1, 74.1, 64, 58.0, 54.1, 44.3, 36.9, 26.5, 20.4]) 
Line_Chart.add('D', [14.1, 15.8, 15.1, 8.8, 8, 10.1, 8.4, 5.1, 6.6, 6.1, 7.4]) 
Line_Chart.range = [0, 100] 
Line_Chart.render_to_file('Line_Chart_hori.svg')

Line_Chart = pygal.StackedLine(fill=True) 
Line_Chart.title = 'Stacked line Chart' 
Line_Chart.x_labels = map(str, range(2005, 2016)) 
Line_Chart.add('A', [None, None, 0, 16.1, 22, 30, 36.9, 45.1, 46.1, 42.33, 37.11]) 
Line_Chart.add('B', [None, None, None, None, None, None, 0, 3.5, 10.1, 23.1, 35.2]) 
Line_Chart.add('C', [85.1, 84.1, 84.1, 74.1, 64, 58.0, 54.1, 44.3, 36.9, 26.5, 20.4]) 
Line_Chart.add('D', [14.1, 15.8, 15.1, 8.8, 8, 10.1, 8.4, 5.1, 6.6, 6.1, 7.4]) 
Line_Chart.render_to_file('Line_Chart_stack.svg')

3. Histogram

Next in the list, we have a unique bar type that takes 3 parameters, named Histogram. It takes height, x-start and x-end as the parameters.

Let us plot a simple Histogram below:

Hist = pygal.Histogram() 
Hist.add('Wide bars', [(6, 1, 11), (5, 6, 13), (3, 1, 16)]) 
Hist.add('Narrow bars', [(11, 2, 3), (13, 5, 5.6), (9, 12, 14)]) 
Hist.render_to_file('Histogram.svg')

4. XY

You can design a chart in the XY plane with the help of inbuilt-methods. Let’s draw cosinus below:

from math import cos  
XY_chart = pygal.XY()  
XY_chart.title = 'XY Chart'  
XY_chart.add('x = cos(y)', [(cos(x / 20.), x / 20.) for x in range(-40, 40, 4)])  
XY_chart.add('y = cos(x)', [(x / 20., cos(x / 20.)) for x in range(-40, 40, 4)])  
XY_chart.add('x = 1', [(2, -4), (2, 4)])  
XY_chart.add('x = -1', [(-1, -4), (-1, 4)])  
XY_chart.add('y = 1', [(-4, 1), (4, 1)])  
XY_chart.add('y = -1', [(-4, -1), (4, -1)])

For a scatter plot, disable the stroke in the XY object.

XY_chart = pygal.XY(stroke=False)  
XY_chart.title = 'XY Scatter Plot'  
XY_chart.add('A', [(0, 0), (.2, .3), (.5, .3), (.6, 2), (.8, .6), (1, 1.01), (1.4, 1.3), (2, 3.26), (2.44, 2.1)])  
XY_chart.add('B', [(.1, .14), (.11, .22), (.4, .4), (.6, .4), (.22, .22), (.4, .3), (.7, .8), (.7, .8)])  
XY_chart.add('C', [(.06, .02), (.12, .01), (1.6, 1.8), (1.63, 1.7), (1.8, 1.62), (1.5, 1.83), (1.7, 1.24), (2.1, 2.22), (2.19, 1.99)])  
XY_chart.render_to_file('xy_scatter.svg')

6. Radar

Radar charts are outstanding for comparing two or more groups of classes on multiple features or characteristics, and Pygal gives you the option to plot a Kiviat diagram very easily.

Let us plot a simple radar plot below:

Radar_Chart = pygal.Radar() 
Radar_Chart.title = 'Radar' 
Radar_Chart.x_labels = ['R', 'D', 'C', 'X', 'E', 'H', 'S', 'N'] 
Radar_Chart.add('B', [6392, 8211, 7521, 7228, 12434, 1640, 2143, 8637]) 
Radar_Chart.add('A', [7474, 8089, 11701, 2631, 6351, 1042, 3697, 9410]) 
Radar_Chart.add('D', [3471, 2932, 4103, 5219, 5811, 1818, 9011, 4668]) 
Radar_Chart.add('C', [41, 42, 58, 78, 143, 126, 31, 132]) 
Radar_Chart.render_to_file('Radar.svg')

7. Box

A box plot gives you a high-level idea about the data distribution based on five factors: minimum, maximum, median, first quartile (Q1) and third quartile (Q3). In pygal, by default, you can plot a box chart having whiskers as the extremes of the dataset, and the box goes from Q1 to Q3, and the middle line represents the median of the given feature.

Let us plot a simple box chart below:

Box_Plot = pygal.Box() 
Box_Plot.title = 'Box Chart' 
Box_Plot.add('B', [6396, 8222, 7530, 7217, 12364, 1661, 2223, 8617]) 
Box_Plot.add('A', [7471, 8091, 11701, 2631, 6362, 1043, 3787, 9440]) 
Box_Plot.add('D', [3471, 2931, 4201, 5221, 5811, 1821, 9011, 4662]) 
Box_Plot.add('C', [45, 42, 56, 73, 142, 131, 31, 101]) 
Box_Plot.render_to_file('Box.svg')

Pygal has option to change whiskers according to the standard deviation on the given feature.

Box_Plot = pygal.Box(box_mode="stdev") 
Box_Plot.title = 'STD. Deviation Box Chart' 
Box_Plot.add('B', [6396, 8222, 7530, 7217, 12364, 1661, 2223, 8617]) 
Box_Plot.add('A', [7471, 8091, 11701, 2631, 6362, 1043, 3787, 9440]) 
Box_Plot.add('D', [3471, 2931, 4201, 5221, 5811, 1821, 9011, 4662]) 
Box_Plot.add('C', [45, 42, 56, 73, 142, 131, 31, 101]) 
Box_Plot.render_to_file('Box_stdDev.svg')

9. Funnel

Funnel plots are a great way to represent the stages of any meeting or process. Pygal helps you plot a funnel chart most easily.

Let us see how to plot a simple funnel chart below:

Funnel_Chart = pygal.Funnel() 
Funnel_Chart.title = 'Funnel Chart' 
Funnel_Chart.x_labels = ['Ri', 'De', 'Cr', 'Ra', 'Ea', 'Re', 'Sp', 'Na'] 
Funnel_Chart.add('Opera', [3478, 2833, 4233, 5329, 5910, 1838, 9023, 4679]) 
Funnel_Chart.add('A', [7471, 8091, 11701, 2641, 6362, 1042, 3787, 9440]) 
Funnel_Chart.add('B', [6391, 8211, 7521, 7211, 12463, 1661, 2124, 8609]) 
Funnel_Chart.render_to_file('Funnel.svg')

10. Solid Gauge

Solid Gauge charts are the most popular angular Gauge charts used for real-world Gauges or dashboards as they can visualize a number in a range at a glance.

Lets us plot a simple SolidGauge plot using Pygal below:

Gauge = pygal.SolidGauge(inner_radius=0.75) 
per_format = lambda x: '{:.10g}%'.format(x) 
dol_format = lambda x: '{:.10g}$'.format(x) 
Gauge.value_formatter = per_format 
Gauge.add('Series A', [{'value': 226000, 'max_value': 1276000}], 
formatter= dol_format) 
Gauge.add('Series B', [{'value': 160, 'max_value': 150}]) 
Gauge.add('Series C', [{'value': 5}]) 
Gauge.add( 
'Series D', [ 
{'value': 52, 'max_value': 150}, 
{'value': 14, 'max_value': 150}]) 
Gauge.add('Series E', [{'value': 69, 'max_value': 150}]) 
Gauge.add('Series F', 95) 
Gauge.add('Series G', [{'value': 150, 'max_value': 150}]) 
Gauge.render_to_file('SolidGauge.svg')

11. Gauge

A Gauge plot is a type of plot that is most commonly used for a single data value with a quantitative context, and it tracks the progress of a KPI.

Let us plot a simple Gauge chart using Pygal below:

Gauge_chart = pygal.Gauge(human_readable=True) 
Gauge_chart.title = 'Gauge Plot' 
Gauge_chart.range = [0, 15000] 
Gauge_chart.add('X', 8219) 
Gauge_chart.add('Y', 8091) 
Gauge_chart.add('Z', 2953) 
Gauge_chart.add('V', 42) 
Gauge_chart.render_to_file('Gauge.svg')

12. Treemap

Treemap is an essential chart that provides a hierarchical view of the data and often shows sales data.

Let us plot a simple Treemap using Pygal below:

Treemap = pygal.Treemap() 
Treemap.title = 'Treemap Plot' 
Treemap.add('X', [1, 2, 11, 3, 2, 2, 1, 2, 11, 2 , 3, None, 8]) 
Treemap.add('Y', [5, 3, 6, 11, 4, 5, 3, 8, 5, -11, None, 9, 4, 2]) 
Treemap.add('Z', [4, 9, 4, 4, 6, 4, 4, 6, 5, 13]) 
Treemap.add('Q', [24, 19]) 
Treemap.add('E', [2, 1, 2, 1, 4, 4, 2, 1, 4, 
3, 4, 2, 1, 2, 2, 2, 2, 2]) 
Treemap.add('A', [21]) 
Treemap.add('B', [5, 9.2, 8.2, 11, 3, 4, 1]) 
Treemap.add('C', [11, 2, 2]) 
Treemap.render_to_file('Treemap.svg')

13. Maps

Maps display the symbolic representation and relationship between different regions or countries. Pygal provides options to plot World Map, French Map, and Swiss Map. The maps are now packaged separately to keep the pygal package small in size, and below is the link to install and configure them according to your project.

https://www.pygal.org/en/stable/documentation/types/maps/index.html

Some Additional Features

Apart from different styles of charts, it also provides some extra features that we will discuss below.

Conclusion

In this guide, we discussed an open-source data visualization library, Pygal. We have seen different types of charts that can be implemented. Furthermore, the must-known features and comparisons with other Python-based data visualization libraries are also explained in this guide. I hope the underlined benefits and importance of the Pygal Data Visualization library were clear to help you make decisions.

Now, I would love to see how you people will use Pygal in your project. Don’t forget to like, share and comment if you learned something valuable today.

The media shown in this article is not owned by Analytics Vidhya and is used at the Author’s discretion.