inst/doc/recipes.R

In recipes: Preprocessing and Feature Engineering Steps for Modeling

## ----ex_setup, include=FALSE--------------------------------------------------
knitr::opts_chunk$set(
  message = FALSE,
  digits = 3,
  collapse = TRUE,
  comment = "#>",
  eval = requireNamespace("modeldata", quietly = TRUE) && requireNamespace("rsample", quietly = TRUE)
  )
options(digits = 3)

## ----data---------------------------------------------------------------------
library(recipes)
library(rsample)
library(modeldata)

data("credit_data")

set.seed(55)
train_test_split <- initial_split(credit_data)

credit_train <- training(train_test_split)
credit_test <- testing(train_test_split)

## ----missing------------------------------------------------------------------
vapply(credit_train, function(x) mean(!is.na(x)), numeric(1))

## ----first_rec----------------------------------------------------------------
rec_obj <- recipe(Status ~ ., data = credit_train)
rec_obj

## ----step_code, eval = FALSE--------------------------------------------------
#  rec_obj <- step_{X}(rec_obj, arguments)    ## or
#  rec_obj <- rec_obj %>% step_{X}(arguments)

## ----imp-steps----------------------------------------------------------------
grep("impute_", ls("package:recipes"), value = TRUE)

## ----dummy--------------------------------------------------------------------
imputed <- rec_obj %>%
  step_impute_knn(all_predictors()) 
imputed

## ----imputing-----------------------------------------------------------------
ind_vars <- imputed %>%
  step_dummy(all_nominal_predictors()) 
ind_vars

## ----center_scale-------------------------------------------------------------
standardized <- ind_vars %>%
  step_center(all_numeric_predictors()) %>%
  step_scale(all_numeric_predictors()) 
standardized

## ----trained------------------------------------------------------------------
trained_rec <- prep(standardized, training = credit_train)
trained_rec

## ----apply--------------------------------------------------------------------
train_data <- bake(trained_rec, new_data = credit_train)
test_data  <- bake(trained_rec, new_data = credit_test)

## ----tibbles------------------------------------------------------------------
class(test_data)
test_data
vapply(test_data, function(x) mean(!is.na(x)), numeric(1))

## ----step_list, echo = FALSE--------------------------------------------------
grep("^step_", ls("package:recipes"), value = TRUE)

## ----check, eval = FALSE------------------------------------------------------
#  trained_rec <- trained_rec %>%
#    check_missing(contains("Marital"))

## ----check_list, echo = FALSE-------------------------------------------------
grep("^check_", ls("package:recipes"), value = TRUE)