Construct reusable warm-start cross-validation partitions

Usage

make_warm_start_cv_plan(
  ids,
  theta,
  task_id,
  seed = 1L,
  outer_folds = 5L,
  inner_folds = 5L
)

Arguments

ids: Unique item IDs in the same order as theta.
theta: Finite numeric scores from one assessment.
task_id: One nonblank assessment label.
seed: Integer random seed, default 1.
outer_folds, inner_folds: Fold counts, each at least two. Counts are never reduced automatically; every training split needs three nonconstant outcomes.

Value

A portable pairwiseLLM_warm_cv_plan list, format 1. It contains exact ordered IDs and outcomes, task identity, seed/RNG provenance, named outer and inner assignments, and an integrity digest. Save with saveRDS().

Details

Partitions use outcome-ranked blocks with randomized ties and fold labels. Draw order is outer folds, inner folds for outer training sets in fold order, then full-data inner folds. The caller's RNG kind and seed state are preserved. Plans are independent of feature schema and engine. A supplied plan is checked before extraction or fitting and is never regenerated or silently realigned. Its digest detects accidental changes, not authorship. Plans contain outcomes and item IDs; they are development evidence, not anonymized artifacts.

Other adaptive warm start: ensemble_warm_start_models(), extract_warm_start_features(), fit_warm_start_model(), make_warm_start_prior(), pairwiseLLM_warm_model, predict.pairwiseLLM_warm_ensemble(), predict.pairwiseLLM_warm_model(), prepare_warm_start_model(), register_warm_start_model(), save_warm_start_model(), summary.pairwiseLLM_warm_ensemble(), summary.pairwiseLLM_warm_predictions(), warm_start_coefficients(), warm_start_feature_schema(), warm_start_python_status()

Examples

if (requireNamespace("withr", quietly = TRUE)) {
  plan <- make_warm_start_cv_plan(as.character(1:20), seq_len(20), "example")
  plan$outer_foldid
}
#>  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 
#>  1  3  5  4  2  1  2  5  3  4  2  5  1  4  3  1  4  3  2  5