Tune base learner

Usage

fit_base_tune(
  recipe,
  model,
  resample,
  tune_mode = c("bayes", "grid"),
  grid = NULL,
  iter_bayes = 10L,
  trim_resamples = TRUE,
  return_best = TRUE,
  workflow = TRUE,
  data_full = NULL,
  metric = "rmse"
)

Arguments

recipe: The recipe object.
model: The model object.
resample: The resample object. It is expected to be generated from the subsamples.
tune_mode: character(1). Hyperparameter tuning mode. Default is "bayes", "grid" is acceptable.
grid: The grid object for hyperparameter tuning.
trim_resamples: logical(1). Default is TRUE, which replaces the actual data.frames in splits column of tune_results object with NA.
return_best: logical(1). If TRUE, the best tuned model is returned.
workflow: logical(1). If TRUE, the best fit model workflow is returned.
data_full: The full data frame to be used for prediction.
metric: character(1). The metric to be used for selecting the best. Must be one of "rmse", "rsq", "mae". Default = "rmse"

Value

List of 3:

base_prediction: data.frame of the best model prediction.
base_parameter: tune_results object of the best model.
best_performance: data.frame of the performance metrics. It includes RMSE, MAPE, R-squared, and MAE for all tuned models.