Utility Functions for Random Forest Super Greedy Trees

These are internal utility functions exported for advanced usage. This Rd file is used solely to register aliases.

Value

Return values depend on the helper function used:

filter.rfsgt and filter.custom.rfsgt

Character vector of variables or generated base-learner terms retained by the filtering step.

get.beta

A list containing beta, betaZ, lasso.percent, predicted, and partial. These components summarize forest-averaged local coefficients, splitting scores, the percentage of lasso terminal-node fits, reconstructed predictions, and partial term contributions.

make.baselearner

A data frame or matrix containing the original predictors together with generated polynomial or interaction base-learner terms.

tune.hcut

An object of class "tune.hcut". This is a character vector of selected terms with attributes including all, formula, formula.all, term.map, term.map.all, cv, bsf, hcut, and hcutSeq.

use.tune.hcut

A "tune.hcut" object restricted to the requested hcut, with tuning attributes preserved.

vimp.rfsgt

Variable-importance scores, typically returned as a named numeric vector.