weka.filters.unsupervised.attribute
Class KernelFilter

java.lang.Object
  extended by weka.filters.Filter
      extended by weka.filters.SimpleFilter
          extended by weka.filters.SimpleBatchFilter
              extended by weka.filters.unsupervised.attribute.KernelFilter
All Implemented Interfaces:
java.io.Serializable, CapabilitiesHandler, OptionHandler, TechnicalInformationHandler, UnsupervisedFilter

public class KernelFilter
extends SimpleBatchFilter
implements UnsupervisedFilter, TechnicalInformationHandler

Converts the given set of predictor variables into a kernel matrix. The class value remains unchangedm, as long as the preprocessing filter doesn't change it.
By default, the data is preprocessed with the Center filter, but the user can choose any filter (NB: one must be careful that the filter does not alter the class attribute unintentionally). With weka.filters.AllFilter the preprocessing gets disabled.

For more information regarding preprocessing the data, see:

K.P. Bennett, M.J. Embrechts: An Optimization Perspective on Kernel Partial Least Squares Regression. In: Advances in Learning Theory: Methods, Models and Applications, 227-249, 2003.

BibTeX:

 @inproceedings{Bennett2003,
    author = {K.P. Bennett and M.J. Embrechts},
    booktitle = {Advances in Learning Theory: Methods, Models and Applications},
    editor = {J. Suykens et al.},
    pages = {227-249},
    publisher = {IOS Press, Amsterdam, The Netherlands},
    series = {NATO Science Series, Series III: Computer and System Sciences},
    title = {An Optimization Perspective on Kernel Partial Least Squares Regression},
    volume = {190},
    year = {2003}
 }
 

Valid options are:

 -D
  Turns on output of debugging information.
 -no-checks
  Turns off all checks - use with caution!
  Turning them off assumes that data is purely numeric, doesn't
  contain any missing values, and has a nominal class. Turning them
  off also means that no header information will be stored if the
  machine is linear. Finally, it also assumes that no instance has
  a weight equal to 0.
  (default: checks on)
 -F <filename>
  The file to initialize the filter with (optional).
 -C <num>
  The class index for the file to initialize with,
  First and last are valid (optional, default: last).
 -K <classname and parameters>
  The Kernel to use.
  (default: weka.classifiers.functions.supportVector.PolyKernel)
 -kernel-factor
  Defines a factor for the kernel.
   - RBFKernel: a factor for gamma
    Standardize: 1/(2*N)
    Normalize..: 6/N
  Available parameters are:
   N for # of instances, A for # of attributes
  (default: 1)
 -P <classname and parameters>
  The Filter used for preprocessing (use weka.filters.AllFilter
  to disable preprocessing).
  (default: weka.filters.unsupervised.attribute.Center)
 
 Options specific to kernel weka.classifiers.functions.supportVector.PolyKernel:
 
 -D
  Enables debugging output (if available) to be printed.
  (default: off)
 -no-checks
  Turns off all checks - use with caution!
  (default: checks on)
 -C <num>
  The size of the cache (a prime number), 0 for full cache and 
  -1 to turn it off.
  (default: 250007)
 -E <num>
  The Exponent to use.
  (default: 1.0)
 -L
  Use lower-order terms.
  (default: no)
 
 Options specific to preprocessing filter weka.filters.unsupervised.attribute.Center:
 
 -unset-class-temporarily
  Unsets the class index temporarily before the filter is
  applied to the data.
  (default: no)

Version:
$Revision: 1.1 $
Author:
Jonathan Miles (jdm18@cs.waikato.ac.nz), FracPete (fracpete at waikato dot ac dot nz)
See Also:
Serialized Form

Constructor Summary
KernelFilter()
           
 
Method Summary
 java.lang.String checksTurnedOffTipText()
          Returns the tip text for this property
 Capabilities getCapabilities()
          Returns the Capabilities of this filter.
 boolean getChecksTurnedOff()
          Returns whether the checks are turned off or not.
 java.io.File getInitFile()
          Gets the file to initialize the filter with, can be null.
 java.lang.String getInitFileClassIndex()
          Gets the class index of the file to initialize the filter with.
 Kernel getKernel()
          Gets the kernel to use.
 java.lang.String getKernelFactorExpression()
          Gets the expression for the kernel.
 java.lang.String[] getOptions()
          Gets the current settings of the filter.
 Filter getPreprocessing()
          Gets the filter used for preprocessing
 TechnicalInformation getTechnicalInformation()
          Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.
 java.lang.String globalInfo()
          Returns a string describing this filter.
 java.lang.String initFileClassIndexTipText()
          Returns the tip text for this property
 java.lang.String initFileTipText()
          Returns the tip text for this property
 void initFilter(Instances instances)
          initializes the filter with the given dataset, i.e., the kernel gets built.
 java.lang.String kernelFactorExpressionTipText()
          Returns the tip text for this property
 java.lang.String kernelTipText()
          Returns the tip text for this property
 java.util.Enumeration listOptions()
          Returns an enumeration describing the available options.
static void main(java.lang.String[] args)
          runs the filter with the given arguments
 java.lang.String preprocessingTipText()
          Returns the tip text for this property
 void setChecksTurnedOff(boolean value)
          Disables or enables the checks (which could be time-consuming).
 void setInitFile(java.io.File value)
          Sets the file to initialize the filter with, can be null.
 void setInitFileClassIndex(java.lang.String value)
          Sets class index of the file to initialize the filter with.
 void setKernel(Kernel value)
          Sets the kernel to use.
 void setKernelFactorExpression(java.lang.String value)
          Sets the expression for the kernel.
 void setOptions(java.lang.String[] options)
          Parses a given list of options.
 void setPreprocessing(Filter value)
          Sets the filter to use for preprocessing (use the AllFilter for no preprocessing)
 
Methods inherited from class weka.filters.SimpleBatchFilter
batchFinished, input
 
Methods inherited from class weka.filters.SimpleFilter
debugTipText, getDebug, setDebug, setInputFormat
 
Methods inherited from class weka.filters.Filter
batchFilterFile, filterFile, getCapabilities, getOutputFormat, isFirstBatchDone, isNewBatch, isOutputFormatDefined, makeCopies, makeCopy, numPendingOutput, output, outputPeek, toString, useFilter, wekaStaticWrapper
 
Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, wait, wait, wait
 

Constructor Detail

KernelFilter

public KernelFilter()
Method Detail

globalInfo

public java.lang.String globalInfo()
Returns a string describing this filter.

Specified by:
globalInfo in class SimpleFilter
Returns:
a description of the filter suitable for displaying in the explorer/experimenter gui

getTechnicalInformation

public TechnicalInformation getTechnicalInformation()
Returns an instance of a TechnicalInformation object, containing detailed information about the technical background of this class, e.g., paper reference or book this class is based on.

Specified by:
getTechnicalInformation in interface TechnicalInformationHandler
Returns:
the technical information about this class

listOptions

public java.util.Enumeration listOptions()
Returns an enumeration describing the available options.

Specified by:
listOptions in interface OptionHandler
Overrides:
listOptions in class SimpleFilter
Returns:
an enumeration of all the available options.

getOptions

public java.lang.String[] getOptions()
Gets the current settings of the filter.

Specified by:
getOptions in interface OptionHandler
Overrides:
getOptions in class SimpleFilter
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.

Valid options are:

 -D
  Turns on output of debugging information.
 -no-checks
  Turns off all checks - use with caution!
  Turning them off assumes that data is purely numeric, doesn't
  contain any missing values, and has a nominal class. Turning them
  off also means that no header information will be stored if the
  machine is linear. Finally, it also assumes that no instance has
  a weight equal to 0.
  (default: checks on)
 -F <filename>
  The file to initialize the filter with (optional).
 -C <num>
  The class index for the file to initialize with,
  First and last are valid (optional, default: last).
 -K <classname and parameters>
  The Kernel to use.
  (default: weka.classifiers.functions.supportVector.PolyKernel)
 -kernel-factor
  Defines a factor for the kernel.
   - RBFKernel: a factor for gamma
    Standardize: 1/(2*N)
    Normalize..: 6/N
  Available parameters are:
   N for # of instances, A for # of attributes
  (default: 1)
 -P <classname and parameters>
  The Filter used for preprocessing (use weka.filters.AllFilter
  to disable preprocessing).
  (default: weka.filters.unsupervised.attribute.Center)
 
 Options specific to kernel weka.classifiers.functions.supportVector.PolyKernel:
 
 -D
  Enables debugging output (if available) to be printed.
  (default: off)
 -no-checks
  Turns off all checks - use with caution!
  (default: checks on)
 -C <num>
  The size of the cache (a prime number), 0 for full cache and 
  -1 to turn it off.
  (default: 250007)
 -E <num>
  The Exponent to use.
  (default: 1.0)
 -L
  Use lower-order terms.
  (default: no)
 
 Options specific to preprocessing filter weka.filters.unsupervised.attribute.Center:
 
 -unset-class-temporarily
  Unsets the class index temporarily before the filter is
  applied to the data.
  (default: no)

Specified by:
setOptions in interface OptionHandler
Overrides:
setOptions in class SimpleFilter
Parameters:
options - the list of options as an array of strings
Throws:
java.lang.Exception - if an option is not supported
See Also:
SimpleFilter.reset()

initFileTipText

public java.lang.String initFileTipText()
Returns the tip text for this property

Returns:
tip text for this property suitable for displaying in the explorer/experimenter gui

getInitFile

public java.io.File getInitFile()
Gets the file to initialize the filter with, can be null.

Returns:
the file

setInitFile

public void setInitFile(java.io.File value)
Sets the file to initialize the filter with, can be null.

Parameters:
value - the file

initFileClassIndexTipText

public java.lang.String initFileClassIndexTipText()
Returns the tip text for this property

Returns:
tip text for this property suitable for displaying in the explorer/experimenter gui

getInitFileClassIndex

public java.lang.String getInitFileClassIndex()
Gets the class index of the file to initialize the filter with.

Returns:
the class index

setInitFileClassIndex

public void setInitFileClassIndex(java.lang.String value)
Sets class index of the file to initialize the filter with.

Parameters:
value - the class index

kernelTipText

public java.lang.String kernelTipText()
Returns the tip text for this property

Returns:
tip text for this property suitable for displaying in the explorer/experimenter gui

getKernel

public Kernel getKernel()
Gets the kernel to use.

Returns:
the kernel

setKernel

public void setKernel(Kernel value)
Sets the kernel to use.

Parameters:
value - the kernel

setChecksTurnedOff

public void setChecksTurnedOff(boolean value)
Disables or enables the checks (which could be time-consuming). Use with caution!

Parameters:
value - if true turns off all checks

getChecksTurnedOff

public boolean getChecksTurnedOff()
Returns whether the checks are turned off or not.

Returns:
true if the checks are turned off

checksTurnedOffTipText

public java.lang.String checksTurnedOffTipText()
Returns the tip text for this property

Returns:
tip text for this property suitable for displaying in the explorer/experimenter gui

kernelFactorExpressionTipText

public java.lang.String kernelFactorExpressionTipText()
Returns the tip text for this property

Returns:
tip text for this property suitable for displaying in the explorer/experimenter gui

getKernelFactorExpression

public java.lang.String getKernelFactorExpression()
Gets the expression for the kernel.

Returns:
the expression

setKernelFactorExpression

public void setKernelFactorExpression(java.lang.String value)
Sets the expression for the kernel.

Parameters:
value - the file

preprocessingTipText

public java.lang.String preprocessingTipText()
Returns the tip text for this property

Returns:
tip text for this property suitable for displaying in the explorer/experimenter gui

setPreprocessing

public void setPreprocessing(Filter value)
Sets the filter to use for preprocessing (use the AllFilter for no preprocessing)

Parameters:
value - the preprocessing filter

getPreprocessing

public Filter getPreprocessing()
Gets the filter used for preprocessing

Returns:
the current preprocessing filter.

initFilter

public void initFilter(Instances instances)
                throws java.lang.Exception
initializes the filter with the given dataset, i.e., the kernel gets built. Needs to be called before the first call of Filter.useFilter or batchFinished(), if not the -F option (or setInitFile(File) is used).

Parameters:
instances - the data to initialize with
Throws:
java.lang.Exception - if building of kernel fails

getCapabilities

public Capabilities getCapabilities()
Returns the Capabilities of this filter.

Specified by:
getCapabilities in interface CapabilitiesHandler
Overrides:
getCapabilities in class Filter
Returns:
the capabilities of this object
See Also:
Capabilities

main

public static void main(java.lang.String[] args)
runs the filter with the given arguments

Parameters:
args - the commandline arguments