How to Sort a DataFrame by Two or More Columns

In this short how-to article, we will learn how to sort DataFrame rows by two or more columns. Rows are sorted by the values in the first column. In the case of equality, the values in the second column are checked, and so on.

Pandas

The sort_values function is used for sorting DataFrame rows. To sort by multiple columns, column names are written in a list.

df = df.sort_values(by=["A","B"])

By default, the index of the rows prior to sorting are kept, which is not an ideal situation. We can change this behavior by using the ignore_index parameter.

df = df.sort_values(by=["A","B"], ignore_index=True)

PySpark

The PySpark equivalent of the sort_values function is orderBy. In the case of sorting by multiple columns, we write the column names in a list.

df = df.orderBy(["Date","Team"])

This question is also being asked as:

• Sort (order) DataFrame rows by multiple columns

People have also asked for:

• How do I sort a dictionary by value?
• Selecting multiple columns in a Pandas DataFrame
• Renaming column names in Pandas

Aporia Team

Sometimes, writing is a joint effort.

building a RAG app?

Read about Aporia’s AI Guardrails

Learn more