org.deeplearning4j.optimize

Class OutputLayerTrainingEvaluator

java.lang.Object

org.deeplearning4j.optimize.OutputLayerTrainingEvaluator

All Implemented Interfaces:

Serializable, TrainingEvaluator

public class OutputLayerTrainingEvaluator
extends Object
implements TrainingEvaluator

An output layer training evaluator uses a multi layer networks output and score functions to determine if training iterations should continue

Author:

Adam Gibson

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	OutputLayerTrainingEvaluator.Builder

Constructor Summary

Constructors

Constructor and Description
OutputLayerTrainingEvaluator(BaseMultiLayerNetwork network, double patience, double patienceIncrease, double bestLoss, int validationEpochs, int miniBatchSize, org.nd4j.linalg.dataset.DataSet testSet, double improvementThreshold)

Method Summary

Modifier and Type	Method and Description
double	bestLoss() The best validation loss so far
double	improvementThreshold()
int	miniBatchSize()
double	patience()
double	patienceIncrease() Amount patience should be increased when a new best threshold is hit
boolean	shouldStop(int epoch) Whether to terminate or not
int	validationEpochs() The number of epochs to test on

Methods inherited from class java.lang.Object

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

Constructor Detail

OutputLayerTrainingEvaluator

public OutputLayerTrainingEvaluator(BaseMultiLayerNetwork network,
                                    double patience,
                                    double patienceIncrease,
                                    double bestLoss,
                                    int validationEpochs,
                                    int miniBatchSize,
                                    org.nd4j.linalg.dataset.DataSet testSet,
                                    double improvementThreshold)

Method Detail

shouldStop

public boolean shouldStop(int epoch)

Whether to terminate or not

Specified by:

shouldStop in interface TrainingEvaluator

Parameters:

epoch - the current epoch

Returns:

whether to terminate or not on the given epoch

patienceIncrease

public double patienceIncrease()

Amount patience should be increased when a new best threshold is hit

Specified by:

patienceIncrease in interface TrainingEvaluator

Returns:

improvementThreshold

public double improvementThreshold()

Specified by:

improvementThreshold in interface TrainingEvaluator

patience

public double patience()

Specified by:

patience in interface TrainingEvaluator

bestLoss

public double bestLoss()

The best validation loss so far

Specified by:

bestLoss in interface TrainingEvaluator

Returns:

the best validation loss so far

validationEpochs

public int validationEpochs()

The number of epochs to test on

Specified by:

validationEpochs in interface TrainingEvaluator

Returns:

the number of epochs to test on

miniBatchSize

public int miniBatchSize()

Specified by:

miniBatchSize in interface TrainingEvaluator