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

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.

Trappole con le variabili categoriche

Verifica sulle risorse di memoria

Superare le insidie: problemi con le stringhe

Superare le trappole: usare gli array NumPy

Label encoding

Crea un label encoding e una mappa

Usare le mappature salvate

Creare un'encoding booleana

One-hot encoding

Verifica sulla one-hot encoding

One-hot encoding di colonne specifiche

Video di chiusura

Pitfalls and Encoding

Adult Census Income

Adoptable Dogs

Tripadvisor Reviews

Used Cars

Saper comprendere, usare e riassumere dati non numerici—come il gruppo sanguigno o lo stato civile di una persona—è fondamentale per chi fa data science. In questo corso imparerai a manipolare e visualizzare dati categorici con pandas e seaborn. Con esercizi pratici, prenderai confidenza con il tipo di dato categorico di pandas, inclusa la creazione, l’eliminazione e l’aggiornamento di colonne categoriche. Lavorerai anche con un’ampia gamma di insiemi di dati, come le caratteristiche dei cani adottabili, le recensioni dei viaggi a Las Vegas e i dati del censimento, per sviluppare le tue abilità nel trattare i dati categorici.

Trappole con le variabili categoriche

Create Your Free Account