netkit.classifiers.active
Class PickLabelStrategyImp

java.lang.Object
  netkit.classifiers.active.PickLabelStrategyImp

All Implemented Interfaces:

PickLabelStrategy, Configurable

Direct Known Subclasses:

ComparatorLabeler, EmpiricalRiskMinimization, EmpiricalRiskMinimizationHarmonic, ERMHybrid, GraphCentralityLabeling, GreedyTruth, RandomLabeling, UncertaintyLabeling

public abstract class PickLabelStrategyImp
extends java.lang.Object
implements PickLabelStrategy

Nested Class Summary

Nested classes/interfaces inherited from interface netkit.classifiers.active.PickLabelStrategy

PickLabelStrategy.LabelNode

Field Summary

protected int	iteration
protected java.util.logging.Logger	logger

Constructor Summary

PickLabelStrategyImp()

Method Summary

void	configure(Configuration config)
protected double	getAverageRank(java.util.List<? extends PickLabelStrategy.LabelNode> sortedlist, int targetIndex)
protected double	getAverageRank(java.util.List<? extends PickLabelStrategy.LabelNode> list, PickLabelStrategy.LabelNode target)
Configuration	getDefaultConfiguration()
int	getIterationNum()
NetworkLearner	getNetworkLearner()
PickLabelStrategy.LabelNode[]	getNodesToLabel(DataSplit currentSplit, Estimate currentPredictions, int maxPicks) Get the list of nodes to get labels for.
double	getRank(DataSplit currentSplit, Estimate currentPredictions, Node node) Get the rank of the given node if the strategy were to pick the node.
DataSplit	getSplit()
void	initialize(NetworkLearner nl, DataSplit split) Initialize the label strategy by providing a reference to the NetworkLeaner object that calls the strategy, thereby giving it access to all information it is likely to need.
PickLabelStrategy.LabelNode[]	peek(DataSplit currentSplit, Estimate currentPredictions, int numPeek) Get the list of nodes to get labels for...
protected abstract PickLabelStrategy.LabelNode[]	pickNodes(Estimate predictions, int maxPicks) Get the list of nodes to get labels for.
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Methods inherited from interface netkit.classifiers.active.PickLabelStrategy

getDescription, getName, getShortName

Field Detail

logger

protected final java.util.logging.Logger logger

iteration

protected int iteration

Constructor Detail

PickLabelStrategyImp

public PickLabelStrategyImp()

Method Detail

pickNodes

protected abstract PickLabelStrategy.LabelNode[] pickNodes(Estimate predictions,
                                                           int maxPicks)

Get the list of nodes to get labels for.

Parameters:

currentPredictions - Current predictions of the classifier

maxPicks - how many nodes should it pick at maximum (this iteration)

Returns:

An array of Node objects that should receive labels. null is returned if done.

configure

public void configure(Configuration config)

Specified by:

configure in interface Configurable

getDefaultConfiguration

public Configuration getDefaultConfiguration()

Specified by:

getDefaultConfiguration in interface Configurable

initialize

public void initialize(NetworkLearner nl,
                       DataSplit split)

Description copied from interface: PickLabelStrategy

Initialize the label strategy by providing a reference to the NetworkLeaner object that calls the strategy, thereby giving it access to all information it is likely to need. This should re-initialize the strategy if it has any cached values from prior calls.

Specified by:

initialize in interface PickLabelStrategy

Parameters:

nl - The NetworkLearner object that will be calling this strategy.

split - The initial train/test split that this will be used on

getSplit

public final DataSplit getSplit()

getIterationNum

public final int getIterationNum()

getNetworkLearner

public final NetworkLearner getNetworkLearner()

peek

public PickLabelStrategy.LabelNode[] peek(DataSplit currentSplit,
                                          Estimate currentPredictions,
                                          int numPeek)

Description copied from interface: PickLabelStrategy

Get the list of nodes to get labels for... without changing the internal state of the active labeler.

Specified by:

peek in interface PickLabelStrategy

Parameters:

currentSplit - Current datasplit

currentPredictions - Current predictions of the classifier

Returns:

An array of Node objects that should receive labels. null is returned if done.

getRank

public double getRank(DataSplit currentSplit,
                      Estimate currentPredictions,
                      Node node)

Description copied from interface: PickLabelStrategy

Get the rank of the given node if the strategy were to pick the node. If more than one node has the same score, then average their ranks (hence a double is returned)

Specified by:

getRank in interface PickLabelStrategy

Parameters:

currentSplit - Current datasplit

currentPredictions - Current predictions of the classifier

node - the node whose rank is requested

Returns:

the rank of the given node or Double.NaN if the node is not in the rankings. If more than one node has the same score, then average their ranks (hence a double is returned)

getAverageRank

protected double getAverageRank(java.util.List<? extends PickLabelStrategy.LabelNode> list,
                                PickLabelStrategy.LabelNode target)

getAverageRank

protected double getAverageRank(java.util.List<? extends PickLabelStrategy.LabelNode> sortedlist,
                                int targetIndex)

getNodesToLabel

public final PickLabelStrategy.LabelNode[] getNodesToLabel(DataSplit currentSplit,
                                                           Estimate currentPredictions,
                                                           int maxPicks)

Get the list of nodes to get labels for.

Specified by:

getNodesToLabel in interface PickLabelStrategy

Parameters:

currentSplit - Current predictions of the classifier

currentPredictions - Current predictions of the classifier

maxPicks - how many nodes should it pick at maximum (this iteration)

Returns:

An array of Node objects that should receive labels. null is returned if done.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object