recipe: Create a recipe for preprocessing data

In recipes: Preprocessing and Feature Engineering Steps for Modeling

Create a recipe for preprocessing data

Description

A recipe is a description of the steps to be applied to a data set in order to prepare it for data analysis.

Usage

recipe(x, ...)

## Default S3 method:
recipe(x, ...)

## S3 method for class 'data.frame'
recipe(x, formula = NULL, ..., vars = NULL, roles = NULL)

## S3 method for class 'formula'
recipe(formula, data, ...)

## S3 method for class 'matrix'
recipe(x, ...)

Arguments

x, data	A data frame or tibble of the template data set (see below).
...	Further arguments passed to or from other methods (not currently used).
formula	A model formula. No in-line functions should be used here (e.g. log(x), x:y, etc.) and minus signs are not allowed. These types of transformations should be enacted using step functions in this package. Dots are allowed as are simple multivariate outcome terms (i.e. no need for cbind; see Examples). A model formula may not be the best choice for high-dimensional data with many columns, because of problems with memory.
vars	A character string of column names corresponding to variables that will be used in any context (see below)
roles	A character string (the same length of vars) that describes a single role that the variable will take. This value could be anything but common roles are "outcome", "predictor", "case_weight", or "ID"

Details

Defining recipes

Variables in recipes can have any type of role, including outcome, predictor, observation ID, case weights, stratification variables, etc.

recipe objects 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) (see the first example below).

Alternatively, a recipe object can be created by first specifying which variables in a data set should be used and then sequentially defining their roles (see the last example). This alternative is an excellent choice when the number of variables is very high, as the formula method is memory-inefficient with many variables.

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() 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

## 

## -- Recipe ------------------------------------------------------------

## 

## -- Inputs

## Number of variables by role

## outcome:   1
## predictor: 5

## 

## -- Operations

## * Centering and scaling for: all_numeric_predictors()

## * Spatial sign on: all_numeric_predictors()

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

## == Workflow ==========================================================
## Preprocessor: Recipe
## Model: linear_reg()
## 
## -- Preprocessor ------------------------------------------------------
## 2 Recipe Steps
## 
## * step_normalize()
## * step_spatialsign()
## 
## -- Model -------------------------------------------------------------
## Linear Regression Model Specification (regression)
## 
## Computational engine: lm

To estimate the preprocessing steps and then fit the linear model, a single call to fit() 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(). 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))

## # A tibble: 6 x 1
##   .pred
##   <dbl>
## 1  18.1
## 2  17.9
## 3  17.2
## 4  18.8
## 5  19.6
## 6  14.6

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

## 

## -- Recipe ------------------------------------------------------------

## 

## -- Inputs

## Number of variables by role

## outcome:   1
## predictor: 5

## 

## -- Training information

## Training data contained 456 data points and no incomplete rows.

## 

## -- Operations

## * Centering and scaling for: carbon and hydrogen, ... | Trained

## * PCA extraction with: carbon, hydrogen, oxygen, ... | Trained

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)

## # A tibble: 2 x 6
##   number operation type      trained skip  id             
##    <int> <chr>     <chr>     <lgl>   <lgl> <chr>          
## 1      1 step      normalize TRUE    FALSE normalize_AeYA4
## 2      2 step      pca       TRUE    FALSE pca_Zn1yz

You can also tidy() 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() would be for models. This applies the estimated steps to any data set.

bake(pca_rec, head(biomass_te))

## # A tibble: 6 x 6
##     HHV    PC1    PC2     PC3     PC4     PC5
##   <dbl>  <dbl>  <dbl>   <dbl>   <dbl>   <dbl>
## 1  18.3 0.730  -0.412 -0.495   0.333   0.253 
## 2  17.6 0.617   1.41   0.118  -0.466   0.815 
## 3  17.2 0.761   1.10  -0.0550 -0.397   0.747 
## 4  18.9 0.0400  0.950  0.158   0.405  -0.143 
## 5  20.5 0.792  -0.732  0.204   0.465  -0.148 
## 6  18.5 0.433  -0.127 -0.354  -0.0168 -0.0888

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

Value

An object of class recipe with sub-objects:

var_info	A tibble containing information about the original data set columns
term_info	A tibble that contains the current set of terms in the data set. This initially defaults to the same data contained in var_info.
steps	A list of step or check objects that define the sequence of preprocessing operations that will be applied to data. The default value is NULL
template	A tibble of the data. This is initialized to be the same as the data given in the data argument but can be different after the recipe is trained.

Examples

# formula example with single outcome:
data(biomass, package = "modeldata")

# split data
biomass_tr <- biomass[biomass$dataset == "Training", ]
biomass_te <- biomass[biomass$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

# ---------------------------------------------------------------------------
# formula multivariate example:
# no need for `cbind(carbon, hydrogen)` for left-hand side

multi_y <- recipe(carbon + hydrogen ~ oxygen + nitrogen + sulfur,
  data = biomass_tr
)
multi_y <- multi_y %>%
  step_center(all_numeric_predictors()) %>%
  step_scale(all_numeric_predictors())

# ---------------------------------------------------------------------------
# example using `update_role` instead of formula:
# best choice for high-dimensional data

rec <- recipe(biomass_tr) %>%
  update_role(carbon, hydrogen, oxygen, nitrogen, sulfur,
    new_role = "predictor"
  ) %>%
  update_role(HHV, new_role = "outcome") %>%
  update_role(sample, new_role = "id variable") %>%
  update_role(dataset, new_role = "splitting indicator")
rec

recipes documentation built on July 4, 2024, 9:06 a.m.