Package deepnetts.automl

Class HyperParameterSearch

java.lang.Object

deepnetts.automl.HyperParameterSearch

public class HyperParameterSearch
extends Object

The core automl class that performs automated model building and evaluation with specified parameters.

Field Summary

Fields

Modifier and Type	Field	Description
static final String	DIVIDE_AND_CONQUER
static final String	GRID
static final String	RANDOM

Constructor Summary

Constructors

Constructor	Description
HyperParameterSearch()

Method Summary

Modifier and Type	Method	Description
HyperParameterSearch	evaluator(javax.visrec.ml.eval.Evaluator<NeuralNetwork,javax.visrec.ml.data.DataSet<? extends MLDataItem>> evaluator)
final javax.visrec.ml.eval.Evaluator<NeuralNetwork,javax.visrec.ml.data.DataSet<? extends MLDataItem>>	getEvaluator()
final NeuralNetwork	getNetwork()
Parameters	getParameters()
List<TrainingResult>	getResults()
List<Parameters.ParameterCombination>	getSearchSpace()
final javax.visrec.ml.data.DataSet<?>	getTrainingSet()
HyperParameterSearch	networkFactory(NetworkFactory<?> networkFactory)
<T> HyperParameterSearch	paramRange(String name, Range<T> range)
<T> HyperParameterSearch	paramValue(String name, T value)	Set searvh param with specific value.
<T> HyperParameterSearch	paramValues(String name, List<T> values)
HyperParameterSearch	randomSeed(long randomSeed)
void	run()
HyperParameterSearch	testSet(javax.visrec.ml.data.DataSet testSet)
HyperParameterSearch	trainingListener(TrainingListener trainingListener)
HyperParameterSearch	trainingSet(javax.visrec.ml.data.DataSet trainingSet)

Methods inherited from class java.lang.Object

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

Field Details

GRID

public static final String GRID

See Also:

• Constant Field Values

RANDOM

public static final String RANDOM

See Also:

• Constant Field Values

DIVIDE_AND_CONQUER

public static final String DIVIDE_AND_CONQUER

See Also:

• Constant Field Values

Constructor Details

HyperParameterSearch

public HyperParameterSearch()

Method Details

trainingSet

public HyperParameterSearch trainingSet(javax.visrec.ml.data.DataSet trainingSet)

testSet

public HyperParameterSearch testSet(javax.visrec.ml.data.DataSet testSet)

evaluator

public HyperParameterSearch evaluator(javax.visrec.ml.eval.Evaluator<NeuralNetwork,javax.visrec.ml.data.DataSet<? extends MLDataItem>> evaluator)

trainingListener

public HyperParameterSearch trainingListener(TrainingListener trainingListener)

paramValue

public <T> HyperParameterSearch paramValue(String name,
 T value)

Set searvh param with specific value.

Type Parameters:

T -

Parameters:

name -

value -

Returns:

paramValues

public <T> HyperParameterSearch paramValues(String name,
 List<T> values)

paramRange

public <T> HyperParameterSearch paramRange(String name,
 Range<T> range)

getParameters

public Parameters getParameters()

getNetwork

public final NeuralNetwork getNetwork()

getTrainingSet

public final javax.visrec.ml.data.DataSet<?> getTrainingSet()

getEvaluator

public final javax.visrec.ml.eval.Evaluator<NeuralNetwork,javax.visrec.ml.data.DataSet<? extends MLDataItem>> getEvaluator()

run

public void run()

getSearchSpace

public List<Parameters.ParameterCombination> getSearchSpace()

getResults

public List<TrainingResult> getResults()

randomSeed

public HyperParameterSearch randomSeed(long randomSeed)

networkFactory

public HyperParameterSearch networkFactory(NetworkFactory<?> networkFactory)