org.deeplearning4j.arbiter.data

Class DataSetIteratorFactoryProvider

java.lang.Object

org.deeplearning4j.arbiter.data.DataSetIteratorFactoryProvider

All Implemented Interfaces:

Serializable, DataProvider

public class DataSetIteratorFactoryProvider
extends Object
implements DataProvider

This is a DataProvider for an DataSetIteratorFactory which based on a key of FACTORY_KEY will create DataSetIterator for use with arbiter. This DataProvider is mainly meant for use for command line driven applications.

Author:

Adam Gibson

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static String	FACTORY_KEY

Constructor Summary

Constructors

Constructor and Description
DataSetIteratorFactoryProvider()

Method Summary

Modifier and Type	Method and Description
Class<?>	getDataType()
org.nd4j.linalg.dataset.api.iterator.DataSetIteratorFactory	testData(Map<String,Object> dataParameters) Get training data given some parameters for the data.
org.nd4j.linalg.dataset.api.iterator.DataSetIteratorFactory	trainData(Map<String,Object> dataParameters) Get training data given some parameters for the data.

Methods inherited from class java.lang.Object

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

Field Detail

FACTORY_KEY

public static final String FACTORY_KEY

See Also:

Constant Field Values

Constructor Detail

DataSetIteratorFactoryProvider

public DataSetIteratorFactoryProvider()

Method Detail

trainData

public org.nd4j.linalg.dataset.api.iterator.DataSetIteratorFactory trainData(Map<String,Object> dataParameters)

Get training data given some parameters for the data. Data parameters map is used to specify things like batch size data preprocessing

Specified by:

trainData in interface DataProvider

Parameters:

dataParameters - Parameters for data. May be null or empty for default data

Returns:

training data

testData

public org.nd4j.linalg.dataset.api.iterator.DataSetIteratorFactory testData(Map<String,Object> dataParameters)

Get training data given some parameters for the data. Data parameters map is used to specify things like batch size data preprocessing

Specified by:

testData in interface DataProvider

Parameters:

dataParameters - Parameters for data. May be null or empty for default data

Returns:

training data

getDataType

public Class<?> getDataType()

Specified by:

getDataType in interface DataProvider