Create Multi-Omics Stacked Ensemble (Convenience Function)

High-level interface for creating and optionally training a stacked ensemble.

Usage

stack_omics(
  data_list,
  learner_list = NULL,
  y = NULL,
  meta_learner = NULL,
  resampling = NULL,
  task_type = "classif"
)

Arguments

data_list

Named list of data matrices (one per modality)

learner_list

Optional named list of learners (or single learner for all)

y

Optional target vector (if provided, trains immediately)

meta_learner

Learner for stacking

resampling

Resampling strategy

task_type

"classif" or "regr"

Value

OmicStackedEnsemble object

Examples

if (FALSE) { # \dontrun{
# Create and train ensemble
ensemble <- stack_omics(
  data_list = list(mRNA = rna_matrix, miRNA = mirna_matrix),
  learner_list = list(mRNA = "classif.ranger", miRNA = "classif.glmnet"),
  y = y_vector,
  meta_learner = "classif.log_reg"
)

# Predict
preds <- ensemble$predict(list(mRNA = new_rna, miRNA = new_mirna))
} # }