intree: Interpreting Tree Ensembles with inTrees

In riAFTBART: A Flexible Approach for Causal Inference with Multiple Treatments and Clustered Survival Outcomes

Interpreting Tree Ensembles with inTrees

Description

The inTrees (interpretable trees) framework that extracts, measures, prunes and selects rules from a tree ensemble. All the codes we use are from the inTrees github repository to act as a work around method since package inTrees was removed from the CRAN repository.

Usage

intree(X, Y, ntree, typeDecay = 2, digits, n_rule)

Arguments

X	A matrix indicating the predictor variables.
Y	A response vector. If a factor, classification is assumed, otherwise regression is assumed.
ntree	Number of trees to grow. This should not be set to too small a number, to ensure that every input row gets predicted at least a few times.
typeDecay	An integer of 1 or 2. 1 representing relative error and 2 representing error. The default is set to 2.
digits	An integer indicating the digits for rounding in Intrees.
n_rule	An integer indicating the minimum number of rules to consider in Intrees.

Value

A matrix including a set of relevant and non-redundant rules, and their metrics

Examples

X <- within(iris,rm("Species")); Y <- iris[,"Species"]
intree_result <- intree(X, Y, ntree=100, digits = 3, n_rule = 2000)

riAFTBART documentation built on May 17, 2022, 1:07 a.m.