Convenience prediction for subject collections — dkge_predict

Harmonises a variety of beta inputs (matrices, `dkge_subject` objects, or `dkge_data` bundles) before forwarding to [dkge_predict()]. This allows callers to work with tidy inputs without manually assembling `B_list` structures.

Usage

dkge_predict_subjects(
  object,
  betas,
  contrasts,
  ids = NULL,
  return_loadings = TRUE
)

Arguments

object: dkge | dkge_stream | dkge_model.
betas: Subject data. Accepts a matrix, list of matrices, `dkge_subject` objects, or a `dkge_data` bundle.
contrasts: List or matrix accepted by [dkge_predict()].
ids: Optional subject identifiers overriding those inferred from `betas`.
return_loadings: Logical; when TRUE, include projected loadings in the result bundle.

Value

Output from [dkge_predict()] with harmonised subject names.