With recipes, you can use dplyr-like pipeable sequences of feature engineering steps to get your data ready for modeling. For example, to create a recipe containing an outcome plus two numeric predictors and then center and scale (“normalize”) the predictors:

library(recipes)
data(ad_data, package = "modeldata")

ad_rec <- recipe(Class ~ tau + VEGF, data = ad_data) %>%
  step_normalize(all_numeric_predictors())

ad_rec
#> Data Recipe
#> 
#> Inputs:
#> 
#>       role #variables
#>    outcome          1
#>  predictor          2
#> 
#> Operations:
#> 
#> Centering and scaling for all_numeric_predictors()

More information on recipes can be found at the Get Started page of tidymodels.org.

You may consider recipes as an alternative method for creating and preprocessing design matrices (also known as model matrices) that can be used for modeling or visualization. While R already has long-standing methods for creating such matrices (e.g. formulas and model.matrix), there are some limitations to what the existing infrastructure can do.

There are several ways to install recipes:

# The easiest way to get recipes is to install all of tidymodels:
install.packages("tidymodels")

# Alternatively, install just recipes:
install.packages("recipes")

# Or the development version from GitHub:
# install.packages("devtools")
devtools::install_github("tidymodels/recipes")

This project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

• For questions and discussions about tidymodels packages, modeling, and machine learning, please post on RStudio Community.

• If you think you have encountered a bug, please submit an issue.

• Either way, learn how to create and share a reprex (a minimal, reproducible example), to clearly communicate about your code.

• Check out further details on contributing guidelines for tidymodels packages and how to get help.