Almost every dataset contains categorical information—and often it’s an unexplored goldmine of information. In this chapter, you’ll learn how pandas handles categorical columns using the data type category.  You’ll also discover how to group data by categories to unearth great summary statistics.

Course introduction

Categorical vs. numerical

Exploring a target variable

Ordinal categorical variables

Categorical data in pandas

Setting dtypes and saving memory

Creating a categorical pandas Series

Setting dtype when reading data

Grouping data by category in pandas

Create lots of groups

Setting up a .groupby() statement

Using pandas functions effectively

Introduction to Categorical Data

Now it’s time to learn how to set, add, and remove categories from a Series. You’ll also explore how to update, rename, collapse, and reorder categories, before applying your new skills to clean and access other data within your DataFrame.

Setting category variables

Setting categories

Adding categories

Removing categories

Updating categories

Collapsing categories knowledge check

Renaming categories

Collapsing categories

Reordering categories

Reordering categories in a Series

Using .groupby() after reordering

Cleaning and accessing data

Cleaning variables

Accessing and filtering data

Categorical pandas Series

In this chapter, you’ll use the seaborn Python library to create informative visualizations using categorical data—including categorical plots (cat-plot), box plots, bar plots, point plots, and count plots. You’ll then learn how to visualize categorical columns and split data across categorical columns to visualize summary statistics of numerical columns.

Introduction to categorical plots using Seaborn

Boxplot understanding

Creating a box plot

Seaborn bar plots

Creating a bar plot

Ordering categories

Bar plot using hue

Point and count plots

Creating a point plot

Creating a count plot

Data Science Process

Review catplot() types

Additional catplot() options

One visualization per group

Updating categorical plots

Visualizing Categorical Data

Lastly, you’ll learn how to overcome the common pitfalls of using categorical data. You’ll also grow your data encoding skills as you are introduced to label encoding and one-hot encoding—perfect for helping you prepare your data for use in machine learning algorithms.

Categorical pitfalls

Memory usage knowledge check

Overcoming pitfalls: string issues

Overcoming pitfalls: using NumPy arrays

Label encoding

Create a label encoding and map

Using saved mappings

Creating a Boolean encoding

One-hot encoding

One-hot knowledge check

One-hot encoding specific columns

Wrap-up video

Pitfalls and Encoding

Adult Census Income

Adoptable Dogs

Tripadvisor Reviews

Used Cars

Being able to understand, use, and summarize non-numerical data—such as a person’s blood type or marital status—is a vital component of being a data scientist. In this course, you’ll learn how to manipulate and visualize categorical data using pandas and seaborn. Through hands-on exercises, you’ll get to grips with pandas' categorical data type, including how to create, delete, and update categorical columns. You’ll also work with a wide range of datasets including the characteristics of adoptable dogs, Las Vegas trip reviews, and census data to develop your skills at working with categorical data.

Data Manipulation with pandas

learn how to manipulate and visualize categorical data using pandas and seaborn.

Working with Categorical Data in Python

Learn how to manipulate and visualize categorical data using pandas and seaborn.

Associate Data Scientist  in Python

Boxplot understanding

Working with Categorical Data in Python

Interaktive Übung