weka.classifiers.mi.supportVector
Class MIRBFKernel

java.lang.Object
  extended by weka.classifiers.functions.supportVector.Kernel
      extended by weka.classifiers.functions.supportVector.CachedKernel
          extended by weka.classifiers.functions.supportVector.RBFKernel
              extended by weka.classifiers.mi.supportVector.MIRBFKernel
All Implemented Interfaces:
java.io.Serializable, CapabilitiesHandler, MultiInstanceCapabilitiesHandler, OptionHandler, RevisionHandler

public class MIRBFKernel
extends RBFKernel
implements MultiInstanceCapabilitiesHandler

The RBF kernel. K(x, y) = e^-(gamma * <x-y, x-y>^2)

Valid options are:

 -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)
 -G <num>
  The Gamma parameter.
  (default: 0.01)

Version:
$Revision: 1.4 $
Author:
Eibe Frank (eibe@cs.waikato.ac.nz), Shane Legg (shane@intelligenesis.net) (sparse vector code), Stuart Inglis (stuart@reeltwo.com) (sparse vector code), J. Lindgren (jtlindgr{at}cs.helsinki.fi) (RBF kernel), Lin Dong (ld21@cs.waikato.ac.nz) (MIkernel)
See Also:
Serialized Form

Constructor Summary
MIRBFKernel()
          default constructor - does nothing.
MIRBFKernel(Instances data, int cacheSize, double gamma)
          Constructor.
 
Method Summary
 void buildKernel(Instances data)
          builds the kernel with the given data.
 Capabilities getCapabilities()
          Returns the Capabilities of this kernel.
 Capabilities getMultiInstanceCapabilities()
          Returns the capabilities of this multi-instance kernel for the relational data.
 java.lang.String getRevision()
          Returns the revision string.
 
Methods inherited from class weka.classifiers.functions.supportVector.RBFKernel
gammaTipText, getGamma, getOptions, globalInfo, listOptions, setGamma, setOptions, toString
 
Methods inherited from class weka.classifiers.functions.supportVector.CachedKernel
cacheSizeTipText, clean, eval, getCacheSize, numCacheHits, numEvals, setCacheSize
 
Methods inherited from class weka.classifiers.functions.supportVector.Kernel
checksTurnedOffTipText, debugTipText, forName, getChecksTurnedOff, getDebug, makeCopies, makeCopy, setChecksTurnedOff, setDebug
 
Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, wait, wait, wait
 

Constructor Detail

MIRBFKernel

public MIRBFKernel()
default constructor - does nothing.


MIRBFKernel

public MIRBFKernel(Instances data,
                   int cacheSize,
                   double gamma)
            throws java.lang.Exception
Constructor.

Parameters:
data - the data to use
cacheSize - the size of the cache
gamma - the bandwidth
Throws:
java.lang.Exception - if something goes wrong
Method Detail

getCapabilities

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

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

getMultiInstanceCapabilities

public Capabilities getMultiInstanceCapabilities()
Returns the capabilities of this multi-instance kernel for the relational data.

Specified by:
getMultiInstanceCapabilities in interface MultiInstanceCapabilitiesHandler
Returns:
the capabilities of this object
See Also:
Capabilities

buildKernel

public void buildKernel(Instances data)
                 throws java.lang.Exception
builds the kernel with the given data. Initializes the kernel cache. The actual size of the cache in bytes is (64 * cacheSize).

Overrides:
buildKernel in class RBFKernel
Parameters:
data - the data to base the kernel on
Throws:
java.lang.Exception - if something goes wrong

getRevision

public java.lang.String getRevision()
Returns the revision string.

Specified by:
getRevision in interface RevisionHandler
Overrides:
getRevision in class RBFKernel
Returns:
the revision