Start here if this is your first time using recipes! You will learn about basic usage, steps, selectors, and checks.
This vignette describes different methods for encoding categorical predictors, with special attention to interaction terms and contrasts.
You can select which variables or features should be used in recipes. This vignette goes over the basics of using selection functions.
In recipes, roles provide a way to select variables for different steps.
In recipes, steps are usually applied to both the training and testing sets. However, in some situations we only want to only apply a step to the training data and we want to skip that step on testing data.
The order in which recipe steps are specified matters, and this vignette gives some general suggestions that you should consider.
Write a new recipe step for data preprocessing.
Improve model performance in imbalanced data sets through undersampling or oversampling.
Build and fit a predictive model with more than one outcome.