Book Image

Python Data Cleaning Cookbook

By : Michael Walker
Book Image

Python Data Cleaning Cookbook

By: Michael Walker

Overview of this book

Getting clean data to reveal insights is essential, as directly jumping into data analysis without proper data cleaning may lead to incorrect results. This book shows you tools and techniques that you can apply to clean and handle data with Python. You'll begin by getting familiar with the shape of data by using practices that can be deployed routinely with most data sources. Then, the book teaches you how to manipulate data to get it into a useful form. You'll also learn how to filter and summarize data to gain insights and better understand what makes sense and what does not, along with discovering how to operate on data to address the issues you've identified. Moving on, you'll perform key tasks, such as handling missing values, validating errors, removing duplicate data, monitoring high volumes of data, and handling outliers and invalid dates. Next, you'll cover recipes on using supervised learning and Naive Bayes analysis to identify unexpected values and classification errors, and generate visualizations for exploratory data analysis (EDA) to visualize unexpected values. Finally, you'll build functions and classes that you can reuse without modification when you have new data. By the end of this Python book, you'll be equipped with all the key skills that you need to clean data and diagnose problems within it.
Table of Contents (12 chapters)

Functions for getting a first look at our data

The first few steps we take after we import our data into a pandas DataFrame are pretty much the same regardless of the characteristics of the data. We almost always want to know the number of columns and rows and the column data types, and see the first few rows. We also might want to view the index and check whether there is a unique identifier for DataFrame rows. These discrete, easily repeatable tasks are good candidates for a collection of functions we can organize into a module.

In this recipe, we will create a module with functions that give us a good first look at any pandas DataFrame. A module is simply a collection of Python code that we can import into another Python program. Modules are easy to reuse because they can be referenced by any program with access to the folder where the module is saved.

Getting ready...

We create two files in this recipe: one with a function we will use to look at our data and another to...