bart tree More...

#include <BART.h>

Public Member Functions
void	next_gen_tree (const vector< float > &x, const vector< float > &y)
	creating next move in MCMC from current tree using metropolis hasting algorithm

void	clone_tree (bart_tree &tree)
	a function to clone bart_tree from root node deep clone - creating a copy of each of the nodes data

void	set_sigma (double sig)
	setting the parameter sigma - should happen before generation new tree using metropolis hasting

	bart_tree (const bart_tree &other)
	a copy ctor preforming a shallow copy: not allocating memory for all tree nodes again.

	bart_tree ()
	a simple default ctor

Data Fields
bart_node *	root
	the tree root

double	tree_loglikelihood
	the tree likelihood based on tree prior and tree match to data

bart_params	params
	the barat params

Protected Member Functions
void	predict (const vector< float > &x, int nSamples, vector< float > &scores) const

void	predict_on_train (const vector< float > &x, int nSamples, vector< float > &scores) const

Detailed Description

bart tree

Constructor & Destructor Documentation

bart_tree::bart_tree ( const bart_tree & other )

inline

a copy ctor preforming a shallow copy: not allocating memory for all tree nodes again.

usibng the same pointers

creating next move in MCMC from current tree using metropolis hasting algorithm

Parameters

x	the matrix observation by observation in one vector
y	the labels for each observation

The documentation for this class was generated from the following files: