Schemashyperparameters/modalitystacking/v1

ModalityStackingHyperparameters

Variant

ModalityStacking: stacked ensemble with one learner per feature group. See `setup_ModalityStacking`.

Raw JSONv1Unknown properties rejected

Properties

base_learners: Hyperparameters[]
meta_learner: Hyperparameters
inner_resampling_config: ResamplerConfig
expand_search_spaces: boolean
Expand a base learner's search space into one library entry per grid combination, rather than tuning it by inner resampling.
ifwtunable: boolean
Inverse Frequency Weighting in classification.
discretetunable: boolean
Keep the single library entry with the lowest cross-validated risk rather than the weighted combination.
feature_groups: object | null
Features each base learner sees, keyed by base learner name. Values must name training features.

Composes

Used by