categoryEncodings: Category Variable Encodings

Juraj Szitas · 2020

Simple, fast, and automatic encodings for category data using a data.table backend. Most of the methods are an implementation of "Sufficient Representation for Categorical Variables" by Johannemann, Hadad, Athey, Wager (2019) , particularly their mean, sparse principal component analysis, low rank representation, and multinomial logit encodings.

Read the paper · More papers on PaperTik