Select and Finalize Best Penalized Cox Workflow — tune_best_glmnet

Runs cross-validation tuning across a grid of elastic net mixture and penalty parameters and finalizes the workflow with the best-performing hyperparameters.

Usage

tune_best_glmnet_workflow(
  formula_or_recipe,
  data,
  resamples = NULL,
  mixture = c(0.2, 0.5, 0.8, 1),
  rule = c("min", "1se"),
  checkpoint_dir = NULL,
  shard = NULL,
  checkpoint_mismatch = c("recompute", "overwrite", "error"),
  ...
)

Arguments

formula_or_recipe: A formula with `Surv()` outcome or a `recipes::recipe`.
data: Data frame containing predictors and survival outcomes.
resamples: An `rsample` resample object (e.g. `vfold_cv` or `group_vfold_cv`).
mixture: Numeric vector of candidate alpha mixture values (default `c(0.2, 0.5, 0.8, 1.0)`).
rule: Hyperparameter selection rule: `"min"` (default) or `"1se"`.
checkpoint_dir: Optional directory path where per-task checkpoints are stored.
shard: Optional shard specification (e.g. `c(1, 2)`) for distributed computation.
checkpoint_mismatch: Action on mismatch: `"recompute"`, `"overwrite"`, or `"error"`.
...: Additional arguments passed to [cv_coxnet()].

Value

A list containing `best_mixture`, `best_penalty`, `model` (fitted on full data), and `cv_results`.