mlpack::tree::GiniGain Class Reference

The Gini gain, a measure of set purity usable as a fitness function (FitnessFunction) for decision trees. More...

#include <gini_gain.hpp>

Static Public Member Functions
template<bool UseWeights, typename CountType >
static double	EvaluatePtr (const CountType *counts, const size_t countLength, const CountType totalCount)
	Evaluate the Gini impurity given a vector of class weight counts.
template<bool UseWeights, typename RowType , typename WeightVecType >
static double	Evaluate (const RowType &labels, const size_t numClasses, const WeightVecType &weights)
	Evaluate the Gini impurity on the given set of labels. More...
static double	Range (const size_t numClasses)
	Return the range of the Gini impurity for the given number of classes. More...

Detailed Description

The Gini gain, a measure of set purity usable as a fitness function (FitnessFunction) for decision trees.

This is the exact same thing as the well-known Gini impurity, but negated—since the decision tree will be trying to maximize gain (and the Gini impurity would need to be minimized).

Member Function Documentation

Evaluate()

template<bool UseWeights, typename RowType , typename WeightVecType >

static double mlpack::tree::GiniGain::Evaluate ( const RowType &  labels, const size_t  numClasses, const WeightVecType &  weights  )

inlinestatic

Evaluate the Gini impurity on the given set of labels.

RowType should be an Armadillo vector that holds size_t objects.

Note that it is possible that due to floating-point representation issues, it is possible that the gain returned can be very slightly greater than 0! Thus, if you are checking for a perfect fit, be sure to use 'gain >= 0.0' not 'gain == 0.0'.

Parameters

labels	Set of labels to evaluate Gini impurity on.
numClasses	Number of classes in the dataset.
weights	Weight of labels.

Range()

static double mlpack::tree::GiniGain::Range ( const size_t  numClasses )

inlinestatic

Return the range of the Gini impurity for the given number of classes.

(That is, the difference between the maximum possible value and the minimum possible value.)

Parameters

numClasses

Number of classes in the dataset.

---

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

• src/mlpack/methods/decision_tree/gini_gain.hpp