PartitionalClusterer (Cognitive Foundry)

java.lang.Object
- gov.sandia.cognition.util.AbstractCloneableSerializable
- - gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm
  - - gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm<ResultType>
    - - gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType>>
      - gov.sandia.cognition.learning.algorithm.clustering.PartitionalClusterer<DataType,ClusterType>

Type Parameters:

DataType - The type of the data to cluster. This is typically defined by the divergence function used.

ClusterType - The type of Cluster created by the algorithm. This is typically defined by the cluster creator function used.

All Implemented Interfaces:

AnytimeAlgorithm<java.util.Collection<ClusterType>>, IterativeAlgorithm, StoppableAlgorithm, AnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType>>, BatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType>>, BatchClusterer<DataType,ClusterType>, BatchHierarchicalClusterer<DataType,ClusterType>, DivergenceFunctionContainer<ClusterType,DataType>, CloneableSerializable, Randomized, java.io.Serializable, java.lang.Cloneable
```
@CodeReview(reviewer="Justin Basilico",
            date="2011-03-09",
            comments={"Should make do a greedy splitting prioritization.","Should make an interface for incremental cluster creation for this to use."},
            changesNeeded=true)
 @PublicationReference(author={"Ying Zhao","George Karypis"},
                      title="Hierarchical Clustering Algorithms for Document Datasets",
                      type=Journal,
                      year=2005,
                      publication="Data Mining and Knowledge Discovery",
                      pages={141,168},
                      url="http://www.springerlink.com/index/jx3825j42x4333m5.pdf")
public class PartitionalClusterer<DataType,ClusterType extends Cluster<DataType>>
extends AbstractAnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType>>
implements BatchClusterer<DataType,ClusterType>, BatchHierarchicalClusterer<DataType,ClusterType>, Randomized, DivergenceFunctionContainer<ClusterType,DataType>
```
The PartitionalClusterer implements a partitional clustering algorithm, which is a type of hierarchical clustering algorithm. Consider using create(int) or createSpherical(int) to instantiate this clusterer.
Partitional clustering works by creating one cluster for all elements in the collection and then repeatedly partitioning each cluster into two clusters until one of the stopping criteria is satisfied:
- The number of requested clusters is met.
- The minimum number of elements in a each leaf cluster is met.
- The improvement in the criterion function is below some threshold.
- The maximum number of iterations is met.
Note that since there are multiple stopping conditions, it is possible to return fewer than the requested number of clusters.
This implementation supports multiple methods for determining the distance between two clusters. The methods can be supplied by providing a WithinClusterDivergence object or by providing a ClusterDivergenceFunction which is automatically wrapped by WithinClusterDivergenceWrapper into a WithinClusterDivergence.
Since:

3.1.1

Author:

Natasha Singh-Miller

See Also:

Serialized Form

Field Summary

Fields
Modifier and Type	Field and Description
`protected WithinClusterDivergence<? super ClusterType,? super DataType>`	`clusterDivergenceFunction` The divergence function used to find the distance between two clusters.
`protected java.util.ArrayList<ClusterType>`	`clusters` The current set of clusters all clusters created.
`protected java.util.ArrayList<BinaryClusterHierarchyNode<DataType,ClusterType>>`	`clustersHierarchy` The current set of hierarchical clusters created.
`protected IncrementalClusterCreator<ClusterType,DataType>`	`creator` The merger used to merge two clusters into one cluster.
`static double`	`DEFAULT_MAX_CRITERION_DECREASE` The default maximum decrease in the training criterion is 1.0.
`static int`	`DEFAULT_MAX_ITERATIONS` The default maximum number of iterations is 2147483647.
`static int`	`DEFAULT_MIN_CLUSTER_SIZE` The default minimum number of elements per cluster is 1.
`static int`	`DEFAULT_NUM_REQUESTED_CLUSTERS` The default number of requested clusters is 2147483647.
`protected DivergenceFunction<? super ClusterType,? super DataType>`	`divergenceFunction` An optional DivergenceFunction that is used to create a `WithinClusterDivergence` function via a `WithinClusterDivergenceWrapper`.
`protected double`	`maxCriterionDecrease` The maximum decrease in training criterion allowed.
`protected int`	`minClusterSize` The minimum number of elements per cluster allowed.
`protected int`	`numRequestedClusters` The number of clusters requested.
`protected java.util.Random`	`random` The source of randomness used during initial partitioning.
`protected double`	`tolerance` Tolerance for determining when improvement has been made.
`protected boolean`	`useCachedClusters` Whether or not the current learning is using cached cluster results.

Fields inherited from class gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner
data, keepGoing

Fields inherited from class gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm
maxIterations

Fields inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm
DEFAULT_ITERATION, iteration

Constructor Summary

Constructors
Constructor and Description
`PartitionalClusterer(int numRequestedClusters, ClusterDivergenceFunction<ClusterType,DataType> divergenceFunction, IncrementalClusterCreator<ClusterType,DataType> creator)` Creates a new partitional clusterer.
`PartitionalClusterer(int numRequestedClusters, WithinClusterDivergence<ClusterType,DataType> divergenceFunction, IncrementalClusterCreator<ClusterType,DataType> creator)` Creates a new partitional clusterer.

Method Summary

All Methods Static Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`protected void`	`cleanupAlgorithm()` Called to clean up the learning algorithm's state after learning has finished.
`PartitionalClusterer<DataType,ClusterType>`	`clone()` This makes public the clone method on the `Object` class and removes the exception that it throws.
`ClusterHierarchyNode<DataType,ClusterType>`	`clusterHierarchically(java.util.Collection<? extends DataType> data)` Performs hierarchical clustering on the given elements.
`static PartitionalClusterer<Vector,CentroidCluster<Vector>>`	`create(int numRequestedClusters)` Create a partitional clusterer, using Euclidean distance and a vector mean centroid cluster creator.
`static PartitionalClusterer<Vectorizable,NormalizedCentroidCluster<Vectorizable>>`	`createSpherical(int numRequestedClusters)` Create a spherical partitional clusterer, using Cosine distance and a vector mean centroid cluster creator.
`int`	`getClusterCount()` Gets the total number of clusters created.
`java.util.ArrayList<ClusterType>`	`getClusters()` Gets all clusters created.
`java.util.ArrayList<BinaryClusterHierarchyNode<DataType,ClusterType>>`	`getClustersHierarchy()` Gets the hierarchy of clusters.
`IncrementalClusterCreator<ClusterType,DataType>`	`getCreator()` Gets the cluster creator.
`DivergenceFunction<? super ClusterType,? super DataType>`	`getDivergenceFunction()` Gets the stored metric between a cluster and a point.
`double`	`getMaxCriterionDecrease()` Returns the maximum decrease in the training criterion allowed following a bisection.
`int`	`getMinClusterSize()` Returns the minimum number of elements per cluster to allow.
`int`	`getNumRequestedClusters()` Gets the current number of clusters requested.
`java.util.Random`	`getRandom()` Gets the random number generator used by this object.
`java.util.ArrayList<ClusterType>`	`getResult()` Gets the current result of the algorithm.
`double`	`getTolerance()` Gets the current tolerance value.
`WithinClusterDivergence<? super ClusterType,? super DataType>`	`getWithinClusterDivergenceFunction()` Gets the metric on clusters used for partitioning.
`protected boolean`	`initializeAlgorithm()` Called to initialize the learning algorithm's state based on the data that is stored in the data field.
`java.util.Collection<ClusterType>`	`learnUsingCachedClusters(java.util.Collection<? extends DataType> data)` Perform clustering by extending the existing cluster hierarchy, if one exists.
`protected void`	`setClusters(java.util.ArrayList<ClusterType> clusters)` Sets the clusters cache to the provided value.
`protected void`	`setClustersHierarchy(java.util.ArrayList<BinaryClusterHierarchyNode<DataType,ClusterType>> clustersHierarchy)` Sets the hierarchy of clusters.
`void`	`setCreator(IncrementalClusterCreator<ClusterType,DataType> creator)` Sets the cluster creator.
`void`	`setDivergenceFunction(DivergenceFunction<? super ClusterType,? super DataType> divergenceFunction)` Use a metric between a cluster and a point to update the metric on clusters.
`void`	`setMaxCriterionDecrease(double maxCriterionDecrease)` Sets the maximum decrease in the training criterion allowed following a bisection.
`void`	`setMinClusterSize(int minClusterSize)` Sets the minimum number of elements per cluster to allow.
`void`	`setNumRequestedClusters(int numRequestedClusters)` Sets the number of clusters requested.
`void`	`setRandom(java.util.Random random)` Sets the random number generator used by this object.
`void`	`setTolerance(double tolerance)` Sets the tolerance value.
`void`	`setWithinClusterDivergenceFunction(WithinClusterDivergence<? super ClusterType,? super DataType> clusterDivergenceFunction)` Sets the metric on clusters used for partitioning.
`protected boolean`	`step()` Called to take a single step of the learning algorithm.

Methods inherited from class gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner
getData, getKeepGoing, learn, setData, setKeepGoing, stop

Methods inherited from class gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm
getMaxIterations, isResultValid, setMaxIterations

Methods inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm
addIterativeAlgorithmListener, fireAlgorithmEnded, fireAlgorithmStarted, fireStepEnded, fireStepStarted, getIteration, getListeners, removeIterativeAlgorithmListener, setIteration, setListeners

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

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchLearner
learn

Methods inherited from interface gov.sandia.cognition.algorithm.AnytimeAlgorithm
getMaxIterations, setMaxIterations

Methods inherited from interface gov.sandia.cognition.algorithm.IterativeAlgorithm
addIterativeAlgorithmListener, getIteration, removeIterativeAlgorithmListener

Methods inherited from interface gov.sandia.cognition.algorithm.StoppableAlgorithm
isResultValid

- Field Detail
  - DEFAULT_MIN_CLUSTER_SIZE
```
public static final int DEFAULT_MIN_CLUSTER_SIZE
```
    The default minimum number of elements per cluster is 1.
    
    See Also:
    
    Constant Field Values
  - DEFAULT_MAX_CRITERION_DECREASE
```
public static final double DEFAULT_MAX_CRITERION_DECREASE
```
    The default maximum decrease in the training criterion is 1.0.
    
    See Also:
    
    Constant Field Values
  - DEFAULT_MAX_ITERATIONS
```
public static final int DEFAULT_MAX_ITERATIONS
```
    The default maximum number of iterations is 2147483647.
    
    See Also:
    
    Constant Field Values
  - DEFAULT_NUM_REQUESTED_CLUSTERS
```
public static final int DEFAULT_NUM_REQUESTED_CLUSTERS
```
    The default number of requested clusters is 2147483647.
    
    See Also:
    
    Constant Field Values
  - clusterDivergenceFunction
```
protected WithinClusterDivergence<? super ClusterType extends Cluster<DataType>,? super DataType> clusterDivergenceFunction
```
    The divergence function used to find the distance between two clusters.
  - divergenceFunction
```
protected DivergenceFunction<? super ClusterType extends Cluster<DataType>,? super DataType> divergenceFunction
```
    An optional DivergenceFunction that is used to create a WithinClusterDivergence function via a WithinClusterDivergenceWrapper.
  - tolerance
```
protected double tolerance
```
    Tolerance for determining when improvement has been made. The default value is .
  - creator
```
protected IncrementalClusterCreator<ClusterType extends Cluster<DataType>,DataType> creator
```
    The merger used to merge two clusters into one cluster.
  - minClusterSize
```
protected int minClusterSize
```
    The minimum number of elements per cluster allowed.
  - maxCriterionDecrease
```
protected double maxCriterionDecrease
```
    The maximum decrease in training criterion allowed.
  - random
```
protected java.util.Random random
```
    The source of randomness used during initial partitioning.
  - clusters
```
protected transient java.util.ArrayList<ClusterType extends Cluster<DataType>> clusters
```
    The current set of clusters all clusters created.
  - clustersHierarchy
```
protected transient java.util.ArrayList<BinaryClusterHierarchyNode<DataType,ClusterType extends Cluster<DataType>>> clustersHierarchy
```
    The current set of hierarchical clusters created.
  - numRequestedClusters
```
protected int numRequestedClusters
```
    The number of clusters requested.
  - useCachedClusters
```
protected boolean useCachedClusters
```
    Whether or not the current learning is using cached cluster results.
- Constructor Detail
  - PartitionalClusterer
```
public PartitionalClusterer(int numRequestedClusters,
                            ClusterDivergenceFunction<ClusterType,DataType> divergenceFunction,
                            IncrementalClusterCreator<ClusterType,DataType> creator)
```
    Creates a new partitional clusterer.
    
    Parameters:
    
    numRequestedClusters - The number of clusters to create. Note that fewer than the requested number of cluster can be returned due to the algorithm stopping due to one of the other stopping criteria.
    
    divergenceFunction - The distance metric between a cluster and a point. The metric is wrapped by a WithinClusterDivergenceWrapper.
    
    creator - The method for creating clusters
  - PartitionalClusterer
```
public PartitionalClusterer(int numRequestedClusters,
                            WithinClusterDivergence<ClusterType,DataType> divergenceFunction,
                            IncrementalClusterCreator<ClusterType,DataType> creator)
```
    Creates a new partitional clusterer.
    
    Parameters:
    
    numRequestedClusters - The number of clusters to create. Note that fewer than the requested number of cluster can be returned due to the algorithm stopping due to one of the other stopping criteria.
    
    divergenceFunction - The distance metric between a cluster and a point. The metric is wrapped by a WithinClusterDivergenceWrapper.
    
    creator - The method for creating clusters
- Method Detail
  - create
```
public static PartitionalClusterer<Vector,CentroidCluster<Vector>> create(int numRequestedClusters)
```
    Create a partitional clusterer, using Euclidean distance and a vector mean centroid cluster creator. This is probably what you want.
    
    Parameters:
    
    numRequestedClusters -
    
    Returns:
  - createSpherical
```
public static PartitionalClusterer<Vectorizable,NormalizedCentroidCluster<Vectorizable>> createSpherical(int numRequestedClusters)
```
    Create a spherical partitional clusterer, using Cosine distance and a vector mean centroid cluster creator. If you need spherical clustering, this is probably what you want.
    
    Parameters:
    
    numRequestedClusters -
    
    Returns:
  - clone
```
public PartitionalClusterer<DataType,ClusterType> clone()
```
    Description copied from class: AbstractCloneableSerializable
    
    This makes public the clone method on the Object class and removes the exception that it throws. Its default behavior is to automatically create a clone of the exact type of object that the clone is called on and to copy all primitives but to keep all references, which means it is a shallow copy. Extensions of this class may want to override this method (but call super.clone() to implement a "smart copy". That is, to target the most common use case for creating a copy of the object. Because of the default behavior being a shallow copy, extending classes only need to handle fields that need to have a deeper copy (or those that need to be reset). Some of the methods in ObjectUtil may be helpful in implementing a custom clone method. Note: The contract of this method is that you must use super.clone() as the basis for your implementation.
    
    Specified by:
    
    clone in interface CloneableSerializable
    
    Overrides:
    
    clone in class AbstractAnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType extends Cluster<DataType>>>
    
    Returns:
    
    A clone of this object.
  - clusterHierarchically
```
public ClusterHierarchyNode<DataType,ClusterType> clusterHierarchically(java.util.Collection<? extends DataType> data)
```
    Description copied from interface: BatchHierarchicalClusterer
    
    Performs hierarchical clustering on the given elements. It returns the root node of the hierarchy of clusters for the given elements.
    
    Specified by:
    
    clusterHierarchically in interface BatchHierarchicalClusterer<DataType,ClusterType extends Cluster<DataType>>
    
    Parameters:
    
    data - The elements to cluster.
    
    Returns:
    
    The root node of the hierarchical cluster for the given elements.
  - initializeAlgorithm
```
protected boolean initializeAlgorithm()
```
    Description copied from class: AbstractAnytimeBatchLearner
    
    Called to initialize the learning algorithm's state based on the data that is stored in the data field. The return value indicates if the algorithm can be run or not based on the initialization.
    
    Specified by:
    
    initializeAlgorithm in class AbstractAnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType extends Cluster<DataType>>>
    
    Returns:
    
    True if the learning algorithm can be run and false if it cannot.
  - step
```
protected boolean step()
```
    Description copied from class: AbstractAnytimeBatchLearner
    
    Called to take a single step of the learning algorithm.
    
    Specified by:
    
    step in class AbstractAnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType extends Cluster<DataType>>>
    
    Returns:
    
    True if another step can be taken and false it the algorithm should halt.
  - cleanupAlgorithm
```
protected void cleanupAlgorithm()
```
    Description copied from class: AbstractAnytimeBatchLearner
    
    Called to clean up the learning algorithm's state after learning has finished.
    
    Specified by:
    
    cleanupAlgorithm in class AbstractAnytimeBatchLearner<java.util.Collection<? extends DataType>,java.util.Collection<ClusterType extends Cluster<DataType>>>
  - getResult
```
public java.util.ArrayList<ClusterType> getResult()
```
    Description copied from interface: AnytimeAlgorithm
    
    Gets the current result of the algorithm.
    
    Specified by:
    
    getResult in interface AnytimeAlgorithm<java.util.Collection<ClusterType extends Cluster<DataType>>>
    
    Returns:
    
    Current result of the algorithm.
  - learnUsingCachedClusters
```
public java.util.Collection<ClusterType> learnUsingCachedClusters(java.util.Collection<? extends DataType> data)
```
    Perform clustering by extending the existing cluster hierarchy, if one exists. This provides an efficient way to compare clusterings with different numbers of clusters, since the hierarchy does not need to be recreated upon every call to learn().
    
    Parameters:
    
    data - The data to cluster. If the existing clustering was created using a data set other than data, these results will likely be useless.
    
    Returns:
    
    The requested clusters. Note that the number of clusters returned may be less than the requested number of clusters if it is not possible to create the number of clusters requested.
  - getClusterCount
```
public int getClusterCount()
```
    Gets the total number of clusters created.
    
    Returns:
    
    The total number of clusters created.
  - getWithinClusterDivergenceFunction
```
public WithinClusterDivergence<? super ClusterType,? super DataType> getWithinClusterDivergenceFunction()
```
    Gets the metric on clusters used for partitioning.
    
    Returns:
    
    The metric on clusters.
  - getDivergenceFunction
```
public DivergenceFunction<? super ClusterType,? super DataType> getDivergenceFunction()
```
    Gets the stored metric between a cluster and a point. If not null, this metric is wrapped by a WithinClusterDivergenceWrapper.
    
    Specified by:
    
    getDivergenceFunction in interface DivergenceFunctionContainer<ClusterType extends Cluster<DataType>,DataType>
    
    Returns:
    
    A DivergenceFunction providing a distance between a cluster and a point.
  - setDivergenceFunction
```
public void setDivergenceFunction(DivergenceFunction<? super ClusterType,? super DataType> divergenceFunction)
```
    Use a metric between a cluster and a point to update the metric on clusters.
    
    Parameters:
    
    divergenceFunction - The metric between a point and a point used to update the metric on clusters.
  - setWithinClusterDivergenceFunction
```
public void setWithinClusterDivergenceFunction(WithinClusterDivergence<? super ClusterType,? super DataType> clusterDivergenceFunction)
```
    Sets the metric on clusters used for partitioning.
    
    Parameters:
    
    clusterDivergenceFunction - The metric on clusters
  - getTolerance
```
public double getTolerance()
```
    Gets the current tolerance value. The tolerance is used for the error bounds when determining if two floating point numbers are the same.
    
    Returns:
    
    The current tolerance.
  - setTolerance
```
public void setTolerance(double tolerance)
```
    Sets the tolerance value. The tolerance is used for the error bounds when determining if two floating point numbers are the same.
    
    Parameters:
    
    tolerance - The new tolerance to use.
  - getCreator
```
public IncrementalClusterCreator<ClusterType,DataType> getCreator()
```
    Gets the cluster creator.
    
    Returns:
    
    The cluster creator.
  - setCreator
```
public void setCreator(IncrementalClusterCreator<ClusterType,DataType> creator)
```
    Sets the cluster creator.
    
    Parameters:
    
    creator - The creator for clusters.
  - getRandom
```
public java.util.Random getRandom()
```
    Description copied from interface: Randomized
    
    Gets the random number generator used by this object.
    
    Specified by:
    
    getRandom in interface Randomized
    
    Returns:
    
    The random number generator used by this object.
  - setRandom
```
public void setRandom(java.util.Random random)
```
    Description copied from interface: Randomized
    
    Sets the random number generator used by this object.
    
    Specified by:
    
    setRandom in interface Randomized
    
    Parameters:
    
    random - The random number generator for this object to use.
  - getMinClusterSize
```
public int getMinClusterSize()
```
    Returns the minimum number of elements per cluster to allow. If the number of elements in a cluster is less than or equal to this number, it will not be bisected. Must be greater than zero.
    
    Returns:
    
    The minimum number of elements per cluster allowed.
  - setMinClusterSize
```
public void setMinClusterSize(int minClusterSize)
```
    Sets the minimum number of elements per cluster to allow. If the number of elements in a cluster is less than or equal to this number, it will not be bisected.
    
    Parameters:
    
    minClusterSize - The new minimum number of elements per cluster allowed. Must be greater than zero.
  - getMaxCriterionDecrease
```
public double getMaxCriterionDecrease()
```
    Returns the maximum decrease in the training criterion allowed following a bisection. If the decrease in criterion of a bisection step is greater than this value, partitioning will stop. To turn off this stopping criterion set this to 1.0 or greater.
    
    Returns:
    
    The maximum decrease in the training criterion following a bisection.
  - setMaxCriterionDecrease
```
public void setMaxCriterionDecrease(double maxCriterionDecrease)
```
    Sets the maximum decrease in the training criterion allowed following a bisection. If the decrease in criterion of a bisection step is greater than this value, partitioning will stop. To turn off this stopping criterion set this to 1.0 or greater.
    
    Parameters:
    
    maxCriterionDecrease - The new maximum minimum distance.
  - getClusters
```
public java.util.ArrayList<ClusterType> getClusters()
```
    Gets all clusters created.
    
    Returns:
    
    A list of the flatten hierarchy of clusters created.
  - setClusters
```
protected void setClusters(java.util.ArrayList<ClusterType> clusters)
```
    Sets the clusters cache to the provided value. Usually used to clear the list by setting the value to null.
    
    Parameters:
    
    clusters - The clusters.
  - getClustersHierarchy
```
public java.util.ArrayList<BinaryClusterHierarchyNode<DataType,ClusterType>> getClustersHierarchy()
```
    Gets the hierarchy of clusters.
    
    Returns:
    
    The hierarchy of clusters.
  - setClustersHierarchy
```
protected void setClustersHierarchy(java.util.ArrayList<BinaryClusterHierarchyNode<DataType,ClusterType>> clustersHierarchy)
```
    Sets the hierarchy of clusters.
    
    Parameters:
    
    clustersHierarchy - The hierarchy of clusters.
  - getNumRequestedClusters
```
public int getNumRequestedClusters()
```
    Gets the current number of clusters requested. Note that fewer clusters may be returned by partitioning algorithms since partitioning may end due to one of the other stopping criterion.
    
    Returns:
    
    The current number of clusters requested.
  - setNumRequestedClusters
```
public void setNumRequestedClusters(int numRequestedClusters)
```
    Sets the number of clusters requested. Note that fewer clusters may be returned by partitioning algorithms since partitioning may end due to one of the other stopping criterion.
    
    Parameters:
    
    numRequestedClusters - The number of clusters requested.

Class PartitionalClusterer<DataType,ClusterType extends Cluster<DataType>>

Field Summary

Fields inherited from class gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner

Fields inherited from class gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm

Fields inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

Constructor Summary

Method Summary

Methods inherited from class gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner

Methods inherited from class gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm

Methods inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

Methods inherited from class java.lang.Object

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchLearner

Methods inherited from interface gov.sandia.cognition.algorithm.AnytimeAlgorithm

Methods inherited from interface gov.sandia.cognition.algorithm.IterativeAlgorithm

Methods inherited from interface gov.sandia.cognition.algorithm.StoppableAlgorithm

Field Detail

DEFAULT_MIN_CLUSTER_SIZE

DEFAULT_MAX_CRITERION_DECREASE

DEFAULT_MAX_ITERATIONS

DEFAULT_NUM_REQUESTED_CLUSTERS

clusterDivergenceFunction

divergenceFunction

tolerance

creator

minClusterSize

maxCriterionDecrease

random

clusters

clustersHierarchy

numRequestedClusters

useCachedClusters

Constructor Detail

PartitionalClusterer

PartitionalClusterer

Method Detail

create

createSpherical

clone

clusterHierarchically

initializeAlgorithm

step

cleanupAlgorithm

getResult

learnUsingCachedClusters

getClusterCount

getWithinClusterDivergenceFunction

getDivergenceFunction

setDivergenceFunction

setWithinClusterDivergenceFunction

getTolerance

setTolerance

getCreator

setCreator

getRandom

setRandom

getMinClusterSize

setMinClusterSize

getMaxCriterionDecrease

setMaxCriterionDecrease

getClusters

setClusters

getClustersHierarchy

setClustersHierarchy

getNumRequestedClusters

setNumRequestedClusters