weka.dl4j.iterators.instance

Class ConvolutionInstanceIterator

java.lang.Object

weka.dl4j.iterators.instance.AbstractInstanceIterator

weka.dl4j.iterators.instance.DefaultInstanceIterator

weka.dl4j.iterators.instance.ConvolutionInstanceIterator

All Implemented Interfaces:

java.io.Serializable, weka.core.OptionHandler, ConvolutionalIterator

public class ConvolutionInstanceIterator
extends DefaultInstanceIterator
implements ConvolutionalIterator

Converts the given Instances object into a DataSet and then constructs and returns a DefaultDataSetIterator. This iterator is designed for training convolutional networks on data that is represented as standard WEKA instances. It enables specification of filter width and height, and number of channels.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
ConvolutionInstanceIterator()

Method Summary

Modifier and Type	Method and Description
void	enforceValidForZooModel(AbstractZooModel tmpZooModel) The only one-channel zoo model currently implemented is Dl4JLeNet.
org.nd4j.linalg.dataset.api.iterator.DataSetIterator	getDataSetIterator(weka.core.Instances data, int seed, int batchSize) Returns the actual iterator.
int	getHeight()
int	getNumChannels()
java.lang.String[]	getOptions() Gets the current settings of the Classifier.
int	getWidth()
java.lang.String	globalInfo()
java.util.Enumeration<weka.core.Option>	listOptions() Returns an enumeration describing the available options.
void	setHeight(int height)
void	setNumChannels(int numChannels)
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setWidth(int width)

Methods inherited from class weka.dl4j.iterators.instance.DefaultInstanceIterator

validate

Methods inherited from class weka.dl4j.iterators.instance.AbstractInstanceIterator

getDataSetIterator, getTrainBatchSize, initialize, setTrainBatchSize

Methods inherited from class java.lang.Object

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

Methods inherited from interface weka.core.OptionHandler

makeCopy

Constructor Detail

ConvolutionInstanceIterator

public ConvolutionInstanceIterator()

Method Detail

getWidth

@OptionMetadata(displayName="desired width",
                description="The desired width of the images (default = 28).",
                commandLineParamName="width",
                commandLineParamSynopsis="-width <int>",
                displayOrder=1)
public int getWidth()

Specified by:

getWidth in interface ConvolutionalIterator

setWidth

public void setWidth(int width)

getHeight

@OptionMetadata(displayName="desired height",
                description="The desired height of the images (default = 28).",
                commandLineParamName="height",
                commandLineParamSynopsis="-height <int>",
                displayOrder=2)
public int getHeight()

Specified by:

getHeight in interface ConvolutionalIterator

setHeight

public void setHeight(int height)

getNumChannels

@OptionMetadata(displayName="desired number of channels",
                description="The desired number of channels (default = 1).",
                commandLineParamName="numChannels",
                commandLineParamSynopsis="-numChannels <int>",
                displayOrder=3)
public int getNumChannels()

Specified by:

getNumChannels in interface ConvolutionalIterator

setNumChannels

public void setNumChannels(int numChannels)

enforceValidForZooModel

public void enforceValidForZooModel(AbstractZooModel tmpZooModel)
                             throws WrongIteratorException

The only one-channel zoo model currently implemented is Dl4JLeNet. If the user tries using any other zoo models with a 1D ConvolutionInstanceIterator, throw an exception

Parameters:

tmpZooModel - zoo model we need to check against

Throws:

WrongIteratorException

getDataSetIterator

public org.nd4j.linalg.dataset.api.iterator.DataSetIterator getDataSetIterator(weka.core.Instances data,
                                                                               int seed,
                                                                               int batchSize)

Returns the actual iterator.

Overrides:

getDataSetIterator in class DefaultInstanceIterator

Parameters:

data - the dataset to use

seed - the seed for the random number generator

batchSize - the batch size to use

Returns:

the DataSetIterator

listOptions

public java.util.Enumeration<weka.core.Option> listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface weka.core.OptionHandler

Overrides:

listOptions in class DefaultInstanceIterator

Returns:

an enumeration of all the available options.

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the Classifier.

Specified by:

getOptions in interface weka.core.OptionHandler

Overrides:

getOptions in class DefaultInstanceIterator

Returns:

an array of strings suitable for passing to setOptions

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

Parses a given list of options.

Specified by:

setOptions in interface weka.core.OptionHandler

Overrides:

setOptions in class DefaultInstanceIterator

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception - if an option is not supported

globalInfo

public java.lang.String globalInfo()

Overrides:

globalInfo in class DefaultInstanceIterator