search_n_samples: Executes a binary search to find the best # of samples.

In shatteringdt: Provide SLT Tools for 'rpart' and 'tree' to Study Decision Trees

Description

Executes a binary search to find the best # of samples.

Usage

search_n_samples(rpart.tree, lower, upper, epsilon)

Arguments

rpart.tree	rpart.tree. A Decision tree generated by rpart package.
lower	int. The # of samples to be used in lower_delta calculation.
upper	int. The # of samples to be used in upper_delta calculation.
epsilon	float. The epsilon to be used in the delta calculation.

Value

the number of samples needed to ensure learning.

shatteringdt documentation built on March 3, 2021, 9:05 a.m.