Book Image

Python Data Visualization Cookbook (Second Edition)

Book Image

Python Data Visualization Cookbook (Second Edition)

Overview of this book

Python Data Visualization Cookbook will progress the reader from the point of installing and setting up a Python environment for data manipulation and visualization all the way to 3D animations using Python libraries. Readers will benefit from over 60 precise and reproducible recipes that will guide the reader towards a better understanding of data concepts and the building blocks for subsequent and sometimes more advanced concepts. Python Data Visualization Cookbook starts by showing how to set up matplotlib and the related libraries that are required for most parts of the book, before moving on to discuss some of the lesser-used diagrams and charts such as Gantt Charts or Sankey diagrams. Initially it uses simple plots and charts to more advanced ones, to make it easy to understand for readers. As the readers will go through the book, they will get to know about the 3D diagrams and animations. Maps are irreplaceable for displaying geo-spatial data, so this book will also show how to build them. In the last chapter, it includes explanation on how to incorporate matplotlib into different environments, such as a writing system, LaTeX, or how to create Gantt charts using Python.
Table of Contents (16 chapters)
Python Data Visualization Cookbook Second Edition
Credits
About the Authors
About the Reviewer
www.PacktPub.com
Preface
Index

Making error bars


Error bars are useful to display the dispersion of data on a plot. They are relatively simple as a form of visualization; however, they are also a bit problematic because what is shown as an error varies across different sciences and publications. This does not lessen the usefulness of error bars, it just imposes the need to always be careful and explicitly state the nature of the error visualized as an error bar.

Getting ready

To be able to plot an error bar in the raw observed data, we need to compute the mean and the error we want to display.

The error we compute represents the 95% confidence interval that the mean we get from our observation is stable, which means our observations are good estimates of the whole population.

Matplotlib supports these type of plots via matplotlib.pyplot.errorbar function.

It offers several kinds of error bars. They can be vertical (yerr) or horizontal (xerr) and symmetrical or asymmetrical.

How to do it...

In the following code we will:

  1. Use some...