Converting Categorical Data into Numbers with Pandas and Scikit-Learn

me@ryanorban.com (Ryan Orban) — Thu, 01 May 2014 00:00:00 +0000

Summary

Most machine learning algorithms require numeric inputs, but real datasets are full of categorical variables — things like city names, product categories, or user types. This FastML post covers the two primary encoding strategies available in Pandas and Scikit-Learn: label encoding (assigning an integer to each category) and one-hot encoding (creating a binary column for each category value).

Data-Preprocessing on Ryan Orban

Converting Categorical Data into Numbers with Pandas and Scikit-Learn

Summary