create_gazepoint_group_folds¶
R reference: gp3ml 0.3.0.
Create deterministic group-aware Gazepoint resampling folds¶
Creates repeated grouped assessment folds that preserve the grouping structure implied by an explicit generalization target. A passing feature-provenance manifest is required, and every analysis-assessment pair is evaluated using the leakage audit.
R reference usage¶
create_gazepoint_group_folds(
data,
outcome,
predictors,
feature_manifest,
generalization_target,
participant_id = NULL,
trial_id = NULL,
stimulus_id = NULL,
v = 5L,
repeats = 1L,
seed = 1L,
source_row_id = ".gp3ml_source_row"
)
The Python implementation is exported as gp3mlpy.create_gazepoint_group_folds. See the runtime docstring for Python-specific typing and semantic adaptations.