Augment data with predictions

This is a generics::augment() method for a workflow that calls augment() on the underlying parsnip model with new_data.

x must be a trained workflow, resulting in fitted parsnip model to augment() with.

new_data will be preprocessed using the preprocessor in the workflow, and that preprocessed data will be used to generate predictions. The final result will contain the original new_data with new columns containing the prediction information.

Usage

# S3 method for workflow
augment(x, new_data, ...)

Arguments

x

A workflow

new_data

A data frame of predictors

...

Arguments passed on to methods

Value

new_data with new prediction specific columns.

Examples

if (rlang::is_installed("broom")) {

library(parsnip)
library(magrittr)
library(modeldata)

data("attrition")

model <- logistic_reg() %>%
  set_engine("glm")

wf <- workflow() %>%
  add_model(model) %>%
  add_formula(
    Attrition ~ BusinessTravel + YearsSinceLastPromotion + OverTime
  )

wf_fit <- fit(wf, attrition)

augment(wf_fit, attrition)

}