buildRecipe()

Specify predictors, outcome, and metadata by building a recipe.

buildRecipe(train_data, use_pca = FALSE, pca_threshold = 0.95)

Arguments

train_data

The part of the feature data designated for ML model training. This can be the output of rsample::training(splitMLInputTibble(ml_input_tibble)).

use_pca

bool Set to TRUE to use PCA instead of all features.

pca_threshold

num The proportion of total variance for which the principle components account

Value

A recipe object

Examples

train <- tibble::tibble(
  genome_id = paste0("g", 1:10),
  genome_drug.resistant_phenotype = rep(c("Resistant", "Susceptible"), each = 5),
  feat_a = rep(c(0L, 1L), 5),
  feat_b = rep(c(1L, 0L), 5)
)
buildRecipe(train, use_pca = FALSE)
#> 
#> ── Recipe ──────────────────────────────────────────────────────────────────────
#> 
#> ── Inputs 
#> Number of variables by role
#> outcome:   1
#> predictor: 2
#> metadata:  1
#> 
#> ── Operations 
#> • Zero variance filter on: recipes::all_predictors()
#> • Centering and scaling for: recipes::all_predictors()