In tidymodels/recipes: Preprocessing and Feature Engineering Steps for Modeling

options(cli.width = 70, width = 70, cli.unicode = FALSE)
set.seed(123)

library(dplyr)
library(workflows)
library(recipes)
library(parsnip)

Defining recipes

Creating a recipe comes in two parts:

1. Specifying which variables are used and what roles they should have.
2. Specifying what transformations should be applied to which variables.

The first part is done with recipe() and optionally update_role(), add_role(), and remove_role(). A recipe object can be created in several ways. If an analysis only contains outcomes and predictors, the simplest way to create one is to use a formula (e.g. y ~ x1 + x2) that does not contain inline functions such as log(x3).

recipe(data, formula)
recipe(formula, data)

Variables in recipes can have any type of role, including outcome, predictor, observation ID, case weights, stratification variables, etc. You can instead use the vars and roles argument to specify the variables and roles. vars must be a character vector of names and roles must the the corresponding roles.

recipe(data, vars = vars, roles = roles)

Lastly you can use update_role(), add_role(), and remove_role(). These functions will alter, add, or eliminate roles from the selections. These can be used in combination with the above ways, or by themselves since recipe(data) will consume all the data as undeclared roles. Note that update_role(), add_role(), and remove_role() are applied before steps and checks, regardless of where they are in the pipeline.

recipe(data) %>%
  update_role(class, new_role = "outcome") %>%
  update_role(starts_with("x"), new_role = "predictor")

There are two different types of operations that can be sequentially added to a recipe.

• Steps can include operations like scaling a variable, creating dummy variables or interactions, and so on. More computationally complex actions such as dimension reduction or imputation can also be specified.

• Checks are operations that conduct specific tests of the data. When the test is satisfied, the data are returned without issue or modification. Otherwise, an error is thrown.

If you have defined a recipe and want to see which steps are included, use the [tidy()][tidy.recipe()] method on the recipe object.

Note that the data passed to [recipe()] need not be the complete data that will be used to train the steps (by [prep()]). The recipe only needs to know the names and types of data that will be used. For large data sets, [head()] could be used to pass a smaller data set to save time and memory.

Using recipes

Once a recipe is defined, it needs to be estimated before being applied to data. Most recipe steps have specific quantities that must be calculated or estimated. For example, [step_normalize()] needs to compute the training set's mean for the selected columns, while [step_dummy()] needs to determine the factor levels of selected columns in order to make the appropriate indicator columns.

The two most common application of recipes are modeling and stand-alone preprocessing. How the recipe is estimated depends on how it is being used.

Modeling

The best way to use use a recipe for modeling is via the workflows package. This bundles a model and preprocessor (e.g. a recipe) together and gives the user a fluent way to train the model/recipe and make predictions.

library(dplyr)
library(workflows)
library(recipes)
library(parsnip)

data(biomass, package = "modeldata")

# split data
biomass_tr <- biomass %>% filter(dataset == "Training")
biomass_te <- biomass %>% filter(dataset == "Testing")

# With only predictors and outcomes, use a formula:
rec <- recipe(HHV ~ carbon + hydrogen + oxygen + nitrogen + sulfur,
              data = biomass_tr)

# Now add preprocessing steps to the recipe:
sp_signed <- 
  rec %>%
  step_normalize(all_numeric_predictors()) %>%
  step_spatialsign(all_numeric_predictors())
sp_signed

We can create a parsnip model, and then build a workflow with the model and recipe:

linear_mod <- linear_reg()

linear_sp_sign_wflow <- 
  workflow() %>% 
  add_model(linear_mod) %>% 
  add_recipe(sp_signed)

linear_sp_sign_wflow

To estimate the preprocessing steps and then fit the linear model, a single call to [fit()][parsnip::fit.model_spec()] is used:

linear_sp_sign_fit <- fit(linear_sp_sign_wflow, data = biomass_tr)

When predicting, there is no need to do anything other than call [predict()][parsnip::predict.model_fit()]. This preprocesses the new data in the same manner as the training set, then gives the data to the linear model prediction code:

predict(linear_sp_sign_fit, new_data = head(biomass_te))

Stand-alone use of recipes

When using a recipe to generate data for a visualization or to troubleshoot any problems with the recipe, there are functions that can be used to estimate the recipe and apply it to new data manually.

Once a recipe has been defined, the [prep()] function can be used to estimate quantities required for the operations using a data set (a.k.a. the training data). [prep()] returns a recipe.

As an example of using PCA (perhaps to produce a plot):

# Define the recipe
pca_rec <- 
  rec %>%
  step_normalize(all_numeric_predictors()) %>%
  step_pca(all_numeric_predictors())

Now to estimate the normalization statistics and the PCA loadings:

pca_rec <- prep(pca_rec, training = biomass_tr)
pca_rec

Note that the estimated recipe shows the actual column names captured by the selectors.

You can [tidy.recipe()] a recipe, either when it is prepped or unprepped, to learn more about its components.

tidy(pca_rec)

You can also [tidy()][tidy.recipe()] recipe steps with a number or id argument.

To apply the prepped recipe to a data set, the [bake()] function is used in the same manner that [predict()][parsnip::predict.model_fit()] would be for models. This applies the estimated steps to any data set.

bake(pca_rec, head(biomass_te))

In general, the workflow interface to recipes is recommended for most applications.

Strings and Factors

The primary purpose of a recipe is to facilitate visualization, modeling, and analysis. Because of this, most qualitative data should be encoded as factors instead of character strings (with exceptions for text analysis and related tasks). It is preferred that quantitative data be converted to factors prior to passing the data to the recipe since the number of levels is usually required for steps (e.g., for making dummy indicator columns).

Although it is advisable to create factors before calling [recipe()], that function has a strings_as_factors argument that can do the conversion. This affects the preprocessed training set (when retain = TRUE) as well as the results of both [prep.recipe()] and [bake.recipe()]. This will only affect variables with roles "outcome" and "predictor"

In 1.2.1 and prior versions of the recipes package, this argument was provided via [prep()]. Code that only provides it via [prep()] will continue to work with a once-per-session warning, and in a future version, it will become an error. If provided in both [prep()] and [recipe()], the value in [recipe()] will take precedence. Default to NULL, which will be taken as TRUE.

tidymodels/recipes documentation built on April 17, 2025, 6:17 p.m.