hex.tree.uplift

Class KLDivergence

java.lang.Object

water.Iced<Divergence>

hex.tree.uplift.Divergence

hex.tree.uplift.KLDivergence

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, water.Freezable<Divergence>

public class KLDivergence
extends Divergence

See Also:

Serialized Form

Field Summary

Fields inherited from class hex.tree.uplift.Divergence

ZERO_TO_DIVIDE

Constructor Summary

Constructors

Constructor and Description
KLDivergence()

Method Summary

Modifier and Type	Method and Description
double	metric(double prCT1, double prCT0) Calculate distance divergence metric between two probabilities.
double	norm(double prCT1, double prCT0, double prLCT1, double prLCT0) Calculate normalization factor to normalize gain.

Methods inherited from class hex.tree.uplift.Divergence

gain, node, split, value

Methods inherited from class water.Iced

asBytes, clone, copyOver, frozenType, read, readExternal, readJSON, reloadFromBytes, toJsonBytes, toJsonString, write, writeExternal, writeJSON

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

KLDivergence

public KLDivergence()

Method Detail

metric

public double metric(double prCT1,
                     double prCT0)

Description copied from class: Divergence

Calculate distance divergence metric between two probabilities.

Specified by:

metric in class Divergence

Returns:

distance divergence metric

norm

public double norm(double prCT1,
                   double prCT0,
                   double prLCT1,
                   double prLCT0)

Description copied from class: Divergence

Calculate normalization factor to normalize gain.

Specified by:

norm in class Divergence

Parameters:

prCT1 - probability of treatment group

prCT0 - probability of control group

prLCT1 - probability of treatment group in left node

prLCT0 - probability of control group in left node

Returns:

normalization factor