mlpack::det Namespace Reference

Density Estimation Trees. More...

Classes
class	DTree
	A density estimation tree is similar to both a decision tree and a space partitioning tree (like a kd-tree). More...

Functions
void	PrintLeafMembership (DTree *dtree, const arma::mat &data, const arma::Mat< size_t > &labels, const size_t numClasses, const std::string leafClassMembershipFile="")
	Print the membership of leaves of a density estimation tree given the labels and number of classes. More...
void	PrintVariableImportance (const DTree *dtree, const std::string viFile="")
	Print the variable importance of each dimension of a density estimation tree. More...
DTree *	Trainer (arma::mat &dataset, const size_t folds, const bool useVolumeReg=false, const size_t maxLeafSize=10, const size_t minLeafSize=5, const std::string unprunedTreeOutput="")
	Train the optimal decision tree using cross-validation with the given number of folds. More...

Detailed Description

Density Estimation Trees.

Function Documentation

PrintLeafMembership()

void mlpack::det::PrintLeafMembership	(	DTree *	dtree,
		const arma::mat &	data,
		const arma::Mat< size_t > &	labels,
		const size_t	numClasses,
		const std::string	leafClassMembershipFile = ""
	)

Print the membership of leaves of a density estimation tree given the labels and number of classes.

Optionally, pass the name of a file to print this information to (otherwise stdout is used).

Parameters

dtree	Tree to print membership of.
data	Dataset tree is built upon.
labels	Class labels of dataset.
numClasses	Number of classes in dataset.
leafClassMembershipFile	Name of file to print to (optional).

PrintVariableImportance()

void mlpack::det::PrintVariableImportance	(	const DTree *	dtree,
		const std::string	viFile = ""
	)

Print the variable importance of each dimension of a density estimation tree.

Optionally, pass the name of a file to print this information to (otherwise stdout is used).

Parameters

dtree	Density tree to use.
viFile	Name of file to print to (optional).

Trainer()

DTree* mlpack::det::Trainer	(	arma::mat &	dataset,
		const size_t	folds,
		const bool	useVolumeReg = false,
		const size_t	maxLeafSize = 10,
		const size_t	minLeafSize = 5,
		const std::string	unprunedTreeOutput = ""
	)

Train the optimal decision tree using cross-validation with the given number of folds.

Optionally, give a filename to print the unpruned tree to. This initializes a tree on the heap, so you are responsible for deleting it.

Parameters

dataset	Dataset for the tree to use.
folds	Number of folds to use for cross-validation.
useVolumeReg	If true, use volume regularization.
maxLeafSize	Maximum number of points allowed in a leaf.
minLeafSize	Minimum number of points allowed in a leaf.
unprunedTreeOutput	Filename to print unpruned tree to (optional).