org.encog.ml.model.config

Class FeedforwardConfig

java.lang.Object

org.encog.ml.model.config.FeedforwardConfig

All Implemented Interfaces:

MethodConfig

public class FeedforwardConfig
extends Object
implements MethodConfig

Config class for EncogModel to use a feedforward neural network.

Constructor Summary

Constructors

Constructor and Description
FeedforwardConfig()

Method Summary

Methods

Modifier and Type	Method and Description
int	determineOutputCount(VersatileMLDataSet dataset) Determine the needed output count.
String	getMethodName()
String	suggestModelArchitecture(VersatileMLDataSet dataset) Suggest a model architecture, based on a dataset.
NormalizationStrategy	suggestNormalizationStrategy(VersatileMLDataSet dataset, String architecture) Suggest a normalization strategy based on a dataset.
String	suggestTrainingArgs(String trainingType) Suggest training arguments.
String	suggestTrainingType() Suggest a training type.

Methods inherited from class java.lang.Object

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

Constructor Detail

FeedforwardConfig

public FeedforwardConfig()

Method Detail

getMethodName

public String getMethodName()

Specified by:

getMethodName in interface MethodConfig

Returns:

The method name.

suggestModelArchitecture

public String suggestModelArchitecture(VersatileMLDataSet dataset)

Suggest a model architecture, based on a dataset.

Specified by:

suggestModelArchitecture in interface MethodConfig

Parameters:

dataset - The dataset.

Returns:

The model architecture.

suggestNormalizationStrategy

public NormalizationStrategy suggestNormalizationStrategy(VersatileMLDataSet dataset,
                                                 String architecture)

Suggest a normalization strategy based on a dataset.

Specified by:

suggestNormalizationStrategy in interface MethodConfig

Parameters:

dataset - The dataset.

architecture - The architecture.

Returns:

The strategy.

suggestTrainingType

public String suggestTrainingType()

Suggest a training type.

Specified by:

suggestTrainingType in interface MethodConfig

Returns:

The training type.

suggestTrainingArgs

public String suggestTrainingArgs(String trainingType)

Suggest training arguments.

Specified by:

suggestTrainingArgs in interface MethodConfig

Parameters:

trainingType - The training type.

Returns:

The training arguments.

determineOutputCount

public int determineOutputCount(VersatileMLDataSet dataset)

Determine the needed output count.

Specified by:

determineOutputCount in interface MethodConfig

Parameters:

dataset - The dataset.

Returns:

The needed output count.