resetActiveForest: Reset an active forest, either from a specific forest in a...

In stochtree: Stochastic Tree Ensembles (XBART and BART) for Supervised Learning and Causal Inference

Reset an active forest, either from a specific forest in a ForestContainer or to an ensemble of single-node (i.e. root) trees

Description

Reset an active forest, either from a specific forest in a ForestContainer or to an ensemble of single-node (i.e. root) trees

Usage

resetActiveForest(active_forest, forest_samples = NULL, forest_num = NULL)

Arguments

active_forest	Current active forest
forest_samples	(Optional) Container of forest samples from which to re-initialize active forest. If not provided, active forest will be reset to an ensemble of single-node (i.e. root) trees.
forest_num	(Optional) Index of forest samples from which to initialize active forest. If not provided, active forest will be reset to an ensemble of single-node (i.e. root) trees.

Value

None

Examples

num_trees <- 100
leaf_dimension <- 1
is_leaf_constant <- TRUE
is_exponentiated <- FALSE
active_forest <- createForest(num_trees, leaf_dimension, is_leaf_constant, is_exponentiated)
forest_samples <- createForestSamples(num_trees, leaf_dimension, is_leaf_constant, is_exponentiated)
forest_samples$add_forest_with_constant_leaves(0.0)
forest_samples$add_numeric_split_tree(0, 0, 0, 0, 0.5, -1.0, 1.0)
forest_samples$add_numeric_split_tree(0, 1, 0, 1, 0.75, 3.4, 0.75)
active_forest$set_root_leaves(0.1)
resetActiveForest(active_forest, forest_samples, 0)
resetActiveForest(active_forest)

stochtree documentation built on April 4, 2025, 2:11 a.m.