Reading Delta Lakes into pandas DataFrames
This post explains how to read Delta Lakes into pandas DataFrames. The delta-rs library makes this incredibly easy and doesn’t require any Spark dependencies. Let’s look at some simple examples, […]
2021
Testing Pandas Code
This post explains how to test Pandas code with the built-in test helper methods and with the beavis functions that give more readable error messages. Unit testing helps you write […]
Working with PySpark ArrayType Columns
This post explains how to create DataFrames with ArrayType columns and how to perform common data processing operations. Array columns are one of the most useful column types, but they’re […]
Defining PySpark Schemas with StructType and StructField
This post explains how to define PySpark schemas and when this design pattern is useful. It’ll also explain when defining schemas seems wise, but can actually be safely avoided. Schemas […]
Renaming Columns in Pandas DataFrames
This article explains how to rename a single or multiple columns in a Pandas DataFrame. There are multiple different ways to rename columns and you’ll often want to perform this […]
Adding constant columns with lit and typedLit to PySpark DataFrames
This post explains how to add constant columns to PySpark DataFrames with lit and typedLit. You’ll see examples where these functions are useful and when these functions are invoked implicitly. […]
Navigating None and null in PySpark
This blog post shows you how to gracefully handle null in PySpark and how to avoid null input errors. Mismanaging the null case is a common source of errors and […]
Creating and reusing the SparkSession with PySpark
This post explains how to create a SparkSession with getOrCreate and how to reuse the SparkSession with getActiveSession. You need a SparkSession to read data stored in files, when manually […]
Different ways to write CSV files with Dask
This post explains how to write a Dask DataFrame to CSV files. You’ll see how to write CSV files, customize the filename, change the compression, and append files to an […]
Writing Dask DataFrame to a Single CSV File
Dask DataFrames are composed of multiple partitions and are outputted as multiple files, one per partition, by default. This post explains the different approaches to write a Dask DataFrame to […]