apply postprocessor in augment.workflow()

[simonpcouch]

Closes #275.

With this PR:

library(tidymodels)
library(tailor)
library(probably)
#> 
#> Attaching package: 'probably'
#> The following objects are masked from 'package:base':
#> 
#>     as.factor, as.ordered

tidymodels_prefer()
theme_set(theme_bw())
options(pillar.advice = FALSE, pillar.min_title_chars = Inf)

set.seed(900)
reg_train <- sim_regression(5000)
reg_cal <- sim_regression(500)
reg_test <- sim_regression(500)


rf_spec <- rand_forest(trees = 5, min_n = 200) %>%
  set_mode("regression")

cal_post <- tailor() %>%
  adjust_numeric_calibration(method = "isotonic_boot")

rf_wflow <- workflow(outcome ~ ., rf_spec)
rf_cal_wflow <- workflow(outcome ~ ., rf_spec, cal_post)


set.seed(1)
rf_fit <- fit(rf_wflow, reg_train)
rf_pred <- rf_fit %>% augment(reg_test) 

set.seed(1)
rf_cal_fit <- fit(rf_cal_wflow, data = reg_train, calibration = reg_cal)
rf_cal_pred <- rf_cal_fit %>% augment(reg_test) 

all.equal(rf_pred, rf_cal_pred)
#> [1] "Component \".pred\": Mean relative difference: 0.2442093"

^{Created on 2025-01-10 with reprex v2.1.1}

[topepo]

Perfect! Thanks!

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.