weka.core

Class FilteredDistance

java.lang.Object

weka.core.FilteredDistance

All Implemented Interfaces:

Serializable, DistanceFunction, OptionHandler

public class FilteredDistance
extends Object
implements DistanceFunction, OptionHandler, Serializable

Applies the given filter before calling the given distance function.

Valid options are:

 -F
  The filter to use. (default: weka.unsupervised.attribute.RandomProjection

 -E
  The distance function to use. (default: weka.core.EuclideanDistance

 
 Options specific to filter weka.filters.unsupervised.attribute.RandomProjection:
 

 -N <number>
  The number of dimensions (attributes) the data should be reduced to
  (default 10; exclusive of the class attribute, if it is set).

 -D [SPARSE1|SPARSE2|GAUSSIAN]
  The distribution to use for calculating the random matrix.
  Sparse1 is:
    sqrt(3)*{-1 with prob(1/6), 0 with prob(2/3), +1 with prob(1/6)}
  Sparse2 is:
    {-1 with prob(1/2), +1 with prob(1/2)}
 

 -P <percent>
  The percentage of dimensions (attributes) the data should
  be reduced to (exclusive of the class attribute, if it is set). The -N
  option is ignored if this option is present and is greater
  than zero.

 -M
  Replace missing values using the ReplaceMissingValues filter

 -R <num>
  The random seed for the random number generator used for
  calculating the random matrix (default 42).

 
 Options specific to distance function weka.core.EuclideanDistance:
 

 -D
  Turns off the normalization of attribute 
  values in distance calculation.

 -R <col1,col2-col4,...>
  Specifies list of columns to used in the calculation of the 
  distance. 'first' and 'last' are valid indices.
  (default: first-last)

 -V
  Invert matching sense of column indices.

 -R <col1,col2-col4,...>
  Specifies list of columns to used in the calculation of the 
  distance. 'first' and 'last' are valid indices.
  (default: first-last)

 -V
  Invert matching sense of column indices.

Version:

$Revision: 8034 $

Author:

Eibe Frank (eibe@cs.waikato.ac.nz)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
FilteredDistance() Default constructor: need to set up Remove filter.

Method Summary

Methods

Modifier and Type	Method and Description
void	clean() Free any references to training instances
double	distance(Instance first, Instance second) Calculates the distance between two instances.
double	distance(Instance first, Instance second, double cutOffValue) Calculates the distance between two instances.
double	distance(Instance first, Instance second, double cutOffValue, PerformanceStats stats) Calculates the distance between two instances.
double	distance(Instance first, Instance second, PerformanceStats stats) Calculates the distance between two instances.
String	distanceTipText() Returns the tip text for this property
String	filterTipText() Returns the tip text for this property
String	getAttributeIndices() Gets the range of attributes used in the calculation of the distance.
DistanceFunction	getDistance() Gets the distance used.
Filter	getFilter() Gets the filter used.
Instances	getInstances() returns the instances currently set.
boolean	getInvertSelection() Gets whether the matching sense of attribute indices is inverted or not.
String[]	getOptions() Gets the current settings.
String	globalInfo() Returns a string describing this object.
Enumeration<Option>	listOptions() Returns an enumeration describing the available options.
void	postProcessDistances(double[] distances) Does post processing of the distances (if necessary) returned by distance(distance(Instance first, Instance second, double cutOffValue).
void	setAttributeIndices(String value) Sets the range of attributes to use in the calculation of the distance.
void	setDistance(DistanceFunction distance) Sets the distance
void	setFilter(Filter filter) Sets the filter
void	setInstances(Instances insts) Sets the instances.
void	setInvertSelection(boolean value) Sets whether the matching sense of attribute indices is inverted or not.
void	setOptions(String[] options) Parses a given list of options.
void	update(Instance ins) Update the distance function (if necessary) for the newly added instance.

Methods inherited from class java.lang.Object

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

Constructor Detail

FilteredDistance

public FilteredDistance()

Default constructor: need to set up Remove filter.

Method Detail

globalInfo

public String globalInfo()

Returns a string describing this object.

Returns:

a description of the evaluator suitable for displaying in the explorer/experimenter gui

filterTipText

public String filterTipText()

Returns the tip text for this property

Returns:

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

setFilter

public void setFilter(Filter filter)

Sets the filter

Parameters:

filter - the filter with all options set.

getFilter

public Filter getFilter()

Gets the filter used.

Returns:

the filter

distanceTipText

public String distanceTipText()

Returns the tip text for this property

Returns:

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

setDistance

public void setDistance(DistanceFunction distance)

Sets the distance

Parameters:

distance - the distance with all options set.

getDistance

public DistanceFunction getDistance()

Gets the distance used.

Returns:

the distance

listOptions

public Enumeration<Option> listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface OptionHandler

Returns:

an enumeration of all the available options.

getOptions

public String[] getOptions()

Gets the current settings. Returns empty array.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions()

setOptions

public void setOptions(String[] options)
                throws Exception

Parses a given list of options.

Specified by:

setOptions in interface OptionHandler

Parameters:

options - the list of options as an array of strings

Throws:

Exception - if an option is not supported

setInstances

public void setInstances(Instances insts)

Sets the instances.

Specified by:

setInstances in interface DistanceFunction

Parameters:

insts - the instances to use

getInstances

public Instances getInstances()

returns the instances currently set.

Specified by:

getInstances in interface DistanceFunction

Returns:

the current instances

setAttributeIndices

public void setAttributeIndices(String value)

Sets the range of attributes to use in the calculation of the distance. The indices start from 1, 'first' and 'last' are valid as well. E.g.: first-3,5,6-last

Specified by:

setAttributeIndices in interface DistanceFunction

Parameters:

value - the new attribute index range

getAttributeIndices

public String getAttributeIndices()

Gets the range of attributes used in the calculation of the distance.

Specified by:

getAttributeIndices in interface DistanceFunction

Returns:

the attribute index range

setInvertSelection

public void setInvertSelection(boolean value)

Sets whether the matching sense of attribute indices is inverted or not.

Specified by:

setInvertSelection in interface DistanceFunction

Parameters:

value - if true the matching sense is inverted

getInvertSelection

public boolean getInvertSelection()

Gets whether the matching sense of attribute indices is inverted or not.

Specified by:

getInvertSelection in interface DistanceFunction

Returns:

true if the matching sense is inverted

distance

public double distance(Instance first,
              Instance second)

Calculates the distance between two instances.

Specified by:

distance in interface DistanceFunction

Parameters:

first - the first instance

second - the second instance

Returns:

the distance between the two given instances

distance

public double distance(Instance first,
              Instance second,
              PerformanceStats stats)
                throws Exception

Calculates the distance between two instances.

Specified by:

distance in interface DistanceFunction

Parameters:

first - the first instance

second - the second instance

stats - the performance stats object

Returns:

the distance between the two given instances

Throws:

Exception - if calculation fails

distance

public double distance(Instance first,
              Instance second,
              double cutOffValue)

Calculates the distance between two instances. Offers speed up (if the distance function class in use supports it) in nearest neighbour search by taking into account the cutOff or maximum distance. Depending on the distance function class, post processing of the distances by postProcessDistances(double []) may be required if this function is used.

Specified by:

distance in interface DistanceFunction

Parameters:

first - the first instance

second - the second instance

cutOffValue - If the distance being calculated becomes larger than cutOffValue then the rest of the calculation is discarded.

Returns:

the distance between the two given instances or Double.POSITIVE_INFINITY if the distance being calculated becomes larger than cutOffValue.

distance

public double distance(Instance first,
              Instance second,
              double cutOffValue,
              PerformanceStats stats)

Calculates the distance between two instances. Offers speed up (if the distance function class in use supports it) in nearest neighbour search by taking into account the cutOff or maximum distance. Depending on the distance function class, post processing of the distances by postProcessDistances(double []) may be required if this function is used.

Specified by:

distance in interface DistanceFunction

Parameters:

first - the first instance

second - the second instance

cutOffValue - If the distance being calculated becomes larger than cutOffValue then the rest of the calculation is discarded.

stats - the performance stats object

Returns:

the distance between the two given instances or Double.POSITIVE_INFINITY if the distance being calculated becomes larger than cutOffValue.

postProcessDistances

public void postProcessDistances(double[] distances)

Does post processing of the distances (if necessary) returned by distance(distance(Instance first, Instance second, double cutOffValue). It may be necessary, depending on the distance function, to do post processing to set the distances on the correct scale. Some distance function classes may not return correct distances using the cutOffValue distance function to minimize the inaccuracies resulting from floating point comparison and manipulation.

Specified by:

postProcessDistances in interface DistanceFunction

Parameters:

distances - the distances to post-process

update

public void update(Instance ins)

Update the distance function (if necessary) for the newly added instance.

Specified by:

update in interface DistanceFunction

Parameters:

ins - the instance to add

clean

public void clean()

Free any references to training instances

Specified by:

clean in interface DistanceFunction