Maniac (meka 1.9.1 API)

java.lang.Object
- weka.classifiers.AbstractClassifier
- - weka.classifiers.SingleClassifierEnhancer
  - - meka.classifiers.multilabel.LabelTransformationClassifier
    - - meka.classifiers.multilabel.Maniac

All Implemented Interfaces:

Serializable, Cloneable, MultiLabelClassifier, weka.classifiers.Classifier, weka.core.BatchPredictor, weka.core.CapabilitiesHandler, weka.core.CapabilitiesIgnorer, weka.core.CommandlineRunnable, weka.core.OptionHandler, weka.core.RevisionHandler, weka.core.TechnicalInformationHandler
```
public class Maniac
extends LabelTransformationClassifier
implements weka.core.TechnicalInformationHandler
```
Maniac - Multi-lAbel classificatioN using AutoenCoders. Transforms the labels using layers of autoencoders.
See: Jörg Wicker, Andrey Tyukin, Stefan Kramer. A Nonlinear Label Compression and Transformation Method for Multi-Label Classification using Autoencoders. The 20th Pacific Asia Conference on Knowledge Discovery and Data Mining (PAKDD), 2016.

Author:

Joerg Wicker (jw@joerg-wicker.org)

See Also:

Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`protected double`	`compression` The compression factor, i.e.
`protected int`	`numberAutoencoders` Number of autoencoders to train, i.e.
`protected boolean`	`optimizeAE` Flag to tell if the number of autoencoders should be optimized.
`protected static long`	`serialVersionUID`

Fields inherited from class weka.classifiers.SingleClassifierEnhancer
m_Classifier

Fields inherited from class weka.classifiers.AbstractClassifier
BATCH_SIZE_DEFAULT, m_BatchSize, m_Debug, m_DoNotCheckCapabilities, m_numDecimalPlaces, NUM_DECIMAL_PLACES_DEFAULT

Constructor Summary

Constructors
Constructor and Description

Maniac()

Constructors
Constructor and Description
`Maniac()`

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`String`	`compressionTipText()` Gives the tiptext for compression.
`String`	`compressionToolTip()` Gives the tooltip for compression.
`double`	`getCompression()` Get the `Compression` value.
`protected weka.classifiers.Classifier`	`getDefaultClassifier()` Change default classifier to CR with Linear Regression as base as this classifier uses numeric values in the compressed labels.
`protected double`	`getDefaultCompression()` Returns the default compression, 0.85 seems to be a good value for most settings.
`protected int`	`getDefaultNumberAutoencoders()` Returns the default number of autoencoders, set to 4, which seems to be good choice for most problems.
`protected boolean`	`getDefaultOptimizeAE()` Tge default setting for optimizing the autoencoders.
`String`	`getModel()` Returns a string representation of the model.
`int`	`getNumberAutoencoders()` Get the `numberAutoencoders` value.
`String[]`	`getOptions()` Returns an array with the options of the classifier.
`weka.core.TechnicalInformation`	`getTechnicalInformation()`
`String`	`globalInfo()` Returns the global information of the classifier.
`boolean`	`isOptimizeAE()` Get the `OptimizeAE` value.
`Enumeration`	`listOptions()` Returns an enumeration of the options.
`static void`	`main(String[] args)` Main method for testing.
`String`	`numberAutoencodersTipText()` Gives the tiptext for numberAutoencoders.
`String`	`numberAutoencodersToolTip()` Gives the tooltip for numberAutoencoders.
`String`	`optimizeAETipText()` Gives the tiptext for optimizeAE.
`String`	`optimizeAEToolTip()` Gives the tooltip for OptimizeAE.
`protected void`	`setAE(org.kramerlab.autoencoder.neuralnet.autoencoder.Autoencoder ae)` Sets the autoencoder (for using a trained one, e.g.
`void`	`setCompression(double compression)` Set the `Compression` value.
`void`	`setNumberAutoencoders(int numberAutoencoders)` Set the `numberAutoencoders` value.
`void`	`setOptimizeAE(boolean optimizeAE)` Set the `OptimizeAE` value.
`void`	`setOptions(String[] options)` Sets the options to the given values in the array.
`String`	`toString()`
`weka.core.Instance`	`transformInstance(weka.core.Instance x)` Transforms the instance in the prediction process before given to the internal multi-label or multi-target classifier.
`weka.core.Instances`	`transformLabels(weka.core.Instances D)` The method to transform the labels into another set of latent labels, typically a compression method is used, e.g., Boolean matrix decomposition in the case of MLC-BMaD, or matrix multiplication based on SVD for PLST.
`double[]`	`transformPredictionsBack(double[] y)` Transforms the predictions of the internal classifier back to the original labels.

Methods inherited from class meka.classifiers.multilabel.LabelTransformationClassifier
buildClassifier, defaultClassifierString, distributionForInstance, extractPart, getRevision, setClassifier, testCapabilities

Methods inherited from class weka.classifiers.SingleClassifierEnhancer
classifierTipText, defaultClassifierOptions, getCapabilities, getClassifier, getClassifierSpec, postExecution, preExecution

Methods inherited from class weka.classifiers.AbstractClassifier
batchSizeTipText, classifyInstance, debugTipText, distributionsForInstances, doNotCheckCapabilitiesTipText, forName, getBatchSize, getDebug, getDoNotCheckCapabilities, getNumDecimalPlaces, implementsMoreEfficientBatchPrediction, makeCopies, makeCopy, numDecimalPlacesTipText, run, runClassifier, setBatchSize, setDebug, setDoNotCheckCapabilities, setNumDecimalPlaces

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface meka.classifiers.multilabel.MultiLabelClassifier
debugTipText, getDebug, setDebug

Methods inherited from interface weka.classifiers.Classifier
classifyInstance, getCapabilities

- Field Detail
  - serialVersionUID
```
protected static final long serialVersionUID
```
    See Also:
    
    Constant Field Values
  - optimizeAE
```
protected boolean optimizeAE
```
    Flag to tell if the number of autoencoders should be optimized. Multiple layers are tested and the best setting is chosen. The maximum number of autoencoders can be set via the numberAutoencoders variable.
  - compression
```
protected double compression
```
    The compression factor, i.e. the compression from one layer to the next. 0.85 means for example, that from one layer with 100 labels, on the next layer 85 are left.
  - numberAutoencoders
```
protected int numberAutoencoders
```
    Number of autoencoders to train, i.e. number of hidden layers + 1. Note that this can be also used as the number of autoencoders to use in the optimization search, autoencoders will be added until this number is reached and then the best configuration in terms of number of layers is selects.
- Constructor Detail
  - Maniac
```
public Maniac()
```
- Method Detail
  - setAE
```
protected void setAE(org.kramerlab.autoencoder.neuralnet.autoencoder.Autoencoder ae)
```
    Sets the autoencoder (for using a trained one, e.g. done in optimization).
    
    Parameters:
    
    ae - The autoencoder
  - getDefaultNumberAutoencoders
```
protected int getDefaultNumberAutoencoders()
```
    Returns the default number of autoencoders, set to 4, which seems to be good choice for most problems.
  - getNumberAutoencoders
```
public final int getNumberAutoencoders()
```
    Get the numberAutoencoders value.
    
    Returns:
    
    an in value
  - setNumberAutoencoders
```
public final void setNumberAutoencoders(int numberAutoencoders)
```
    Set the numberAutoencoders value.
    
    Parameters:
    
    numberAutoencoders - The new NumberAutoencoders value.
  - numberAutoencodersToolTip
```
public String numberAutoencodersToolTip()
```
    Gives the tooltip for numberAutoencoders.
    
    Returns:
    
    the tooltip for numberAutoencoders.
  - numberAutoencodersTipText
```
public String numberAutoencodersTipText()
```
    Gives the tiptext for numberAutoencoders.
    
    Returns:
    
    the tiptext for numberAutoencoders.
  - getCompression
```
public final double getCompression()
```
    Get the Compression value.
    
    Returns:
    
    a double value
  - setCompression
```
public final void setCompression(double compression)
```
    Set the Compression value.
    
    Parameters:
    
    compression - The new Compression value.
  - getDefaultCompression
```
protected double getDefaultCompression()
```
    Returns the default compression, 0.85 seems to be a good value for most settings.
    
    Returns:
    
    The default compression.
  - compressionToolTip
```
public String compressionToolTip()
```
    Gives the tooltip for compression.
    
    Returns:
    
    the tooltip for compression.
  - compressionTipText
```
public String compressionTipText()
```
    Gives the tiptext for compression.
    
    Returns:
    
    the tiptext for compression.
  - optimizeAETipText
```
public String optimizeAETipText()
```
    Gives the tiptext for optimizeAE.
    
    Returns:
    
    the tiptext for optimizeAE.
  - isOptimizeAE
```
public final boolean isOptimizeAE()
```
    Get the OptimizeAE value.
    
    Returns:
    
    a boolean value
  - setOptimizeAE
```
public final void setOptimizeAE(boolean optimizeAE)
```
    Set the OptimizeAE value.
    
    Parameters:
    
    optimizeAE - The new OptimizeAE value.
  - getDefaultOptimizeAE
```
protected boolean getDefaultOptimizeAE()
```
    Tge default setting for optimizing the autoencoders. Set to false, as this is an expensive operation.
    
    Returns:
    
    The default flag for optimizing the autoencoders.
  - optimizeAEToolTip
```
public String optimizeAEToolTip()
```
    Gives the tooltip for OptimizeAE.
    
    Returns:
    
    the tooltip for OptimizeAE.
  - globalInfo
```
public String globalInfo()
```
    Returns the global information of the classifier.
    
    Returns:
    
    Global information of the classfier
  - listOptions
```
public Enumeration listOptions()
```
    Returns an enumeration of the options.
    
    Specified by:
    
    listOptions in interface weka.core.OptionHandler
    
    Overrides:
    
    listOptions in class weka.classifiers.SingleClassifierEnhancer
    
    Returns:
    
    Enumeration of the options.
  - getDefaultClassifier
```
protected weka.classifiers.Classifier getDefaultClassifier()
```
    Change default classifier to CR with Linear Regression as base as this classifier uses numeric values in the compressed labels.
    
    Overrides:
    
    getDefaultClassifier in class LabelTransformationClassifier
  - getOptions
```
public String[] getOptions()
```
    Returns an array with the options of the classifier.
    
    Specified by:
    
    getOptions in interface weka.core.OptionHandler
    
    Overrides:
    
    getOptions in class weka.classifiers.SingleClassifierEnhancer
    
    Returns:
    
    Array of options.
  - setOptions
```
public void setOptions(String[] options)
                throws Exception
```
    Sets the options to the given values in the array.
    
    Specified by:
    
    setOptions in interface weka.core.OptionHandler
    
    Overrides:
    
    setOptions in class weka.classifiers.SingleClassifierEnhancer
    
    Parameters:
    
    options - The options to be set.
    
    Throws:
    
    Exception
  - getTechnicalInformation
```
public weka.core.TechnicalInformation getTechnicalInformation()
```
    Specified by:
    
    getTechnicalInformation in interface weka.core.TechnicalInformationHandler
  - transformInstance
```
public weka.core.Instance transformInstance(weka.core.Instance x)
                                     throws Exception
```
    Description copied from class: LabelTransformationClassifier
    
    Transforms the instance in the prediction process before given to the internal multi-label or multi-target classifier. The instance is passed having the original set of labels, these must be replaced with the transformed labels (attributes) so that the internla classifier can predict them.
    
    Specified by:
    
    transformInstance in class LabelTransformationClassifier
    
    Parameters:
    
    x - The instance to transform. Consists of features and labels.
    
    Returns:
    
    The transformed instance. Consists of features and transformed labels.
    
    Throws:
    
    Exception
  - transformLabels
```
public weka.core.Instances transformLabels(weka.core.Instances D)
                                    throws Exception
```
    Description copied from class: LabelTransformationClassifier
    
    The method to transform the labels into another set of latent labels, typically a compression method is used, e.g., Boolean matrix decomposition in the case of MLC-BMaD, or matrix multiplication based on SVD for PLST.
    
    Specified by:
    
    transformLabels in class LabelTransformationClassifier
    
    Parameters:
    
    D - the instances to transform into new instances with transformed labels. The Instances consist of features and original labels.
    
    Returns:
    
    The resulting instances. Instances consist of features and transformed labels.
    
    Throws:
    
    Exception
  - transformPredictionsBack
```
public double[] transformPredictionsBack(double[] y)
```
    Description copied from class: LabelTransformationClassifier
    
    Transforms the predictions of the internal classifier back to the original labels.
    
    Specified by:
    
    transformPredictionsBack in class LabelTransformationClassifier
    
    Parameters:
    
    y - The predictions that should be transformed back. The array consists only of the predictions as they are returned from the internal classifier.
    
    Returns:
    
    The transformed predictions.
  - getModel
```
public String getModel()
```
    Description copied from interface: MultiLabelClassifier
    
    Returns a string representation of the model.
    
    Specified by:
    
    getModel in interface MultiLabelClassifier
    
    Returns:
    
    the model
  - toString
```
public String toString()
```
    Overrides:
    
    toString in class Object
  - main
```
public static void main(String[] args)
                 throws Exception
```
    Main method for testing.
    
    Parameters:
    
    args - - Arguments passed from the command line
    
    Throws:
    
    Exception

Class Maniac

Field Summary

Fields inherited from class weka.classifiers.SingleClassifierEnhancer

Fields inherited from class weka.classifiers.AbstractClassifier

Constructor Summary

Method Summary

Methods inherited from class meka.classifiers.multilabel.LabelTransformationClassifier

Methods inherited from class weka.classifiers.SingleClassifierEnhancer

Methods inherited from class weka.classifiers.AbstractClassifier

Methods inherited from class java.lang.Object

Methods inherited from interface meka.classifiers.multilabel.MultiLabelClassifier

Methods inherited from interface weka.classifiers.Classifier

Field Detail

serialVersionUID

optimizeAE

compression

numberAutoencoders

Constructor Detail

Maniac

Method Detail

setAE

getDefaultNumberAutoencoders

getNumberAutoencoders

setNumberAutoencoders

numberAutoencodersToolTip

numberAutoencodersTipText

getCompression

setCompression

getDefaultCompression

compressionToolTip

compressionTipText

optimizeAETipText

isOptimizeAE

setOptimizeAE

getDefaultOptimizeAE

optimizeAEToolTip

globalInfo

listOptions

getDefaultClassifier

getOptions

setOptions

getTechnicalInformation

transformInstance

transformLabels

transformPredictionsBack

getModel

toString

main