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)

Using multiple merge-by columns

The same logic we used to perform one-to-one merges with one merge-by column applies to merges we perform with multiple merge-by columns. Inner, outer, left, and right joins work the same way when you have two or more merge-by columns. We will demonstrate this in this recipe.

Getting ready

We will work with the NLS data in this recipe, specifically weeks worked and college enrollment from 2000 through 2004. Both the weeks worked and college enrollment files contain one row per person, per year.

How to do it...

We will continue this recipe with one-to-one merges, but this time with multiple merge-by columns on each DataFrame. Let's get started:

  1. Import pandas and load the NLS weeks worked and college enrollment data:
    >>> import pandas as pd
    >>> nls97weeksworked = pd.read_csv("data/nls97weeksworked.csv")
    >>> nls97colenr = pd.read_csv("data/nls97colenr.csv")
  2. Look at some of the NLS...