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)

Chapter 10: User-Defined Functions and Classes to Automate Data Cleaning

There are a number of great reasons to write code that is reusable. When we step back from the particular data cleaning problem at hand and consider its relationship to very similar problems, we can actually improve our understanding of the key issues involved. We are also more likely to address a task systematically when we set our sights more on solving it for the long term than on the before-lunch solution. This has the additional benefit of helping us to disentangle the substantive issues from the mechanics of data manipulation.

We will create several modules to accomplish routine data cleaning tasks in this chapter. The functions and classes in these modules are examples of code that can be reused across DataFrames, or for one DataFrame over an extended period of time. These functions handle many of the tasks we discussed in the first nine chapters, but in a manner that allows us to reuse our code.

...