Package elki.clustering.uncertain
Class RepresentativeUncertainClustering.Par
- java.lang.Object
-
- elki.clustering.uncertain.RepresentativeUncertainClustering.Par
-
- All Implemented Interfaces:
Parameterizer
- Enclosing class:
- RepresentativeUncertainClustering
public static class RepresentativeUncertainClustering.Par extends java.lang.Object implements Parameterizer
Parameterization class.- Author:
- Alexander Koos, Erich Schubert
-
-
Field Summary
Fields Modifier and Type Field Description static OptionID
ALGORITHM_ID
Parameter to hand an algorithm to be wrapped and run to our instance ofRepresentativeUncertainClustering
.protected double
alpha
Alpha parameter for confidence.static OptionID
ALPHA_ID
Alpha parameter for confidence estimation.static OptionID
CLUSTERDISTANCE_ID
Distance function to measure the similarity of clusterings.static int
DEFAULT_ENSEMBLE_DEPTH
Default number of clusterings to run.protected ClusteringDistanceSimilarity
distance
Distance (dissimilarity) for clusterinogs.protected boolean
keep
Keep all samples (not only the representative results).static OptionID
KEEP_SAMPLES_ID
Flag to keep all samples.static OptionID
META_ALGORITHM_ID
Parameter to hand an algorithm for creating the meta-clustering to our instance ofRepresentativeUncertainClustering
.protected ClusteringAlgorithm<?>
metaAlgorithm
Field to store the inner algorithm for meta-clusteringprotected int
numsamples
Field to store parameter the number of samples.protected RandomFactory
random
Random factory for sampling.static OptionID
RANDOM_ID
Parameter to specify the random generator.static OptionID
SAMPLES_ID
Parameter to specify the amount of clusterings that shall be created and compared.protected ClusteringAlgorithm<?>
samplesAlgorithm
Field to store the algorithm.
-
Constructor Summary
Constructors Constructor Description Par()
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description void
configure(Parameterization config)
Configure the class.RepresentativeUncertainClustering
make()
Make an instance after successful configuration.
-
-
-
Field Detail
-
DEFAULT_ENSEMBLE_DEPTH
public static final int DEFAULT_ENSEMBLE_DEPTH
Default number of clusterings to run.- See Also:
- Constant Field Values
-
CLUSTERDISTANCE_ID
public static final OptionID CLUSTERDISTANCE_ID
Distance function to measure the similarity of clusterings.
-
META_ALGORITHM_ID
public static final OptionID META_ALGORITHM_ID
Parameter to hand an algorithm for creating the meta-clustering to our instance ofRepresentativeUncertainClustering
.It has to use a metric distance function to work on the sample-clusterings.
-
ALGORITHM_ID
public static final OptionID ALGORITHM_ID
Parameter to hand an algorithm to be wrapped and run to our instance ofRepresentativeUncertainClustering
.
-
SAMPLES_ID
public static final OptionID SAMPLES_ID
Parameter to specify the amount of clusterings that shall be created and compared.
-
KEEP_SAMPLES_ID
public static final OptionID KEEP_SAMPLES_ID
Flag to keep all samples.
-
RANDOM_ID
public static final OptionID RANDOM_ID
Parameter to specify the random generator.
-
ALPHA_ID
public static final OptionID ALPHA_ID
Alpha parameter for confidence estimation.
-
distance
protected ClusteringDistanceSimilarity distance
Distance (dissimilarity) for clusterinogs.
-
metaAlgorithm
protected ClusteringAlgorithm<?> metaAlgorithm
Field to store the inner algorithm for meta-clustering
-
samplesAlgorithm
protected ClusteringAlgorithm<?> samplesAlgorithm
Field to store the algorithm.
-
numsamples
protected int numsamples
Field to store parameter the number of samples.
-
random
protected RandomFactory random
Random factory for sampling.
-
alpha
protected double alpha
Alpha parameter for confidence.
-
keep
protected boolean keep
Keep all samples (not only the representative results).
-
-
Method Detail
-
configure
public void configure(Parameterization config)
Description copied from interface:Parameterizer
Configure the class.Note: the status is collected by the parameterization object, so that multiple errors may arise and be reported in one run.
- Specified by:
configure
in interfaceParameterizer
- Parameters:
config
- Parameterization
-
make
public RepresentativeUncertainClustering make()
Description copied from interface:Parameterizer
Make an instance after successful configuration.Note: your class should return the exact type, only this very broad interface should use
Object
as return type.- Specified by:
make
in interfaceParameterizer
- Returns:
- a new instance
-
-