Donut charts with Bokeh

Bokeh is a Python library used for creating interactive visualizations in a web browser. It provides powerful tools that offer flexibility, interactivity, and scalability for exploring various data insights.

What is a donut chart?

Donut charts are a variation of pie charts used to illustrate the proportions of categorical data where the size of the segment represents the category's proportion. Different colors represent each sector to create an evident visual difference, and a key is provided against each color to identify the category.

math: To access pi for creating the chart.
pandas: To manipulate data and create a series to hold data.
bokeh.io: To control the output and display of the plots. We specifically import the output_file and save methods.
bokeh.palettes: To assign a color palette to the chart to improve its visual appearance. There are various available themes we use Category10 to improve interface accessibility.
bokeh.plotting: To create and customize plots without working directly with the lower-level Bokeh models. We specifically import the figure and show methods from it.
bokeh.transform: To transform the data by adding visual properties such as colors, sizes, and positions. We specifically import cumsum to calculate the cumulative sum of the data field.

Example code

In this example, we present different teams corresponding to the number of votes they got in a donut chart illustration to show which team is winning.

from math import pi
import pandas as pd
from bokeh.io import output_file, save
from bokeh.palettes import Category10
from bokeh.plotting import figure, show
from bokeh.transform import cumsum

#sample dataset
elecResults = {
    'Tiger team': 106,
    'Lion team': 126,
    'Falcon team': 84,
    'Dolphin team': 130,
    'Iris team': 74,
    'Orchid team': 40,
    'Rio team': 96,
}

#creating data
data = pd.Series(elecResults).reset_index(name='value').rename(columns={'index':'teams'})
data['angle'] = data['value']/data['value'].sum() * 2*pi
data['color'] = Category10[len(elecResults)]

#creating chart
myChart = figure(height=400, 
                 width=650, 
                 title="Election results - based on team votes", 
                 toolbar_location=None,
                 tools="hover", 
                 tooltips="@teams: @value, Angle: @angle{0.0}", 
                 x_range=(-0.5, 0.5))

#creating sectors
myChart.annular_wedge(x=0, y=1,  
                      inner_radius=0.12,  
                      outer_radius=0.25, 
                      direction="anticlock",
                      start_angle=cumsum('angle', include_zero=True), 
                      end_angle=cumsum('angle'),
                      line_color="white", 
                      fill_color='color', 
                      legend_field='teams', 
                      source=data)

#clearing interface
myChart.axis.axis_label=None
myChart.axis.visible=False
myChart.grid.grid_line_color = None

#displaying output
output_file("output.html")
show(myChart)

Creating a donut chart through python bokeh.

Code explanation

Lines 1–6: Import all the necessary libraries and modules.
Lines 9–16: Create a dictionary named elecResults with the team names as a key, and their votes count as the value.
Line 20: Create a series data using pandas and assign each election result's key as an index and the corresponding vote count as value to it. We rename() the data columns to teams and value.
Line 21: Calculate the angles for each elecResults key using the formula.

Note: The formula divides each team's vote count by the sum of total vote counts and then multiply it with the double of pi to normalise the data values.

Line 22: Set a color palette to the chart and pass the length of the electResults dictionary as a parameter. Assign it to color in data.
Lines 25–31: Create a chart figure using figure() and define its dimensions, title, tools and tooltip properties, and range.

Note: Hover over the slices to see the team names, their votes , and angles.

Lines 34–43: Create the pie slices using annular_wedge(), define the properties, and set the source.
- x and y are the center coordinates of the chart.
- inner_radius and outer_radius sets both the radius of the donut chart.
- direction specifies from which point to start creating the chart.
- start-angle and end-angle are specified using cumsum to calculate when each slice starts and ends.
- line_color specifies the border color of the pie chart.
- fill_color specifies the color assigned to each category from the color column.
- legend_field specifies the column we use for each slice label in the legend.
Lines 46–48: Clear the interface by setting the axis gridlines to none and its visibility to false.
Lines 51–52: Set the output to output.html to specify the endpoint where the plot will appear and use show() to display the created plot.

Code output

The donut chart is displayed at the output.html endpoint with the Category10 color palette to represent each wedge of the dataset.

Free Resources

Learn in-demand tech skills in half the time

PRODUCTS

Mock Interview

New

Courses

Skill Paths

Projects

Assessments

Donut charts with Bokeh

What is a donut chart?

Real-life applications

Required imports

Example code

Code explanation

Code output

Common query