Feature Engineering for Existing Features.png

Feature engineering is a crucial stage in any machine learning project. It allows you to use data to define features that enable machine learning algorithms to work properly.

In this course, you will learn the techniques that will help you create new features from existing features. You’ll start by diving into label encoding which is crucial for converting categorical features into numerical. You’ll also learn about other various types of encoding such as: one-hot, count, and mean, all of which are important for feature engineering. 

In the remaining chapters, you’ll learn about feature interaction and datetime features. In all, this course will show you the many different ways you can create features from existing ones.

Feature Engineering for Machine Learning

# What is count encoding?

**Count encoding** uses the number of categories in the dataset as the new feature. For example, in the column `Type`, the category value `Sun` appears 3 times, then the new feature for the `Sun` is 3.

**Count encoding** is simple and straightforward and works well with tree-based models, such as `xgboost`. However, it has its own limitations: it’s not friendly to the new features in the test set, and it’s easy to have a conflict if two category values appear the same number of times in the dataset. 

>If you have ever played in Kaggle’s competitions, you may know that count encoding is a prevalent encoding method for tree-based models.

Learn how to use count encoding to encode categorical features.

__default

Count Encoding

What is count encoding?