org.apache.mahout.clustering
Class AbstractCluster

java.lang.Object
  org.apache.mahout.clustering.AbstractCluster

All Implemented Interfaces:

org.apache.hadoop.io.Writable, Cluster, Model<VectorWritable>, Parametered

Direct Known Subclasses:

DistanceMeasureCluster, GaussianCluster

public abstract class AbstractCluster
extends Object
implements Cluster

Nested Class Summary

Nested classes/interfaces inherited from interface org.apache.mahout.common.parameters.Parametered

Parametered.ParameteredGeneralizations

Field Summary

Fields inherited from interface org.apache.mahout.clustering.Cluster

CLUSTERED_POINTS_DIR, CLUSTERS_DIR, FINAL_ITERATION_SUFFIX, INITIAL_CLUSTERS_DIR

Fields inherited from interface org.apache.mahout.common.parameters.Parametered

log

Constructor Summary

protected	AbstractCluster()
protected	AbstractCluster(Vector point, int id2)
protected	AbstractCluster(Vector center2, Vector radius2, int id2)

Method Summary

String	asFormatString(String[] bindings) Produce a custom, human-friendly, printable representation of the Cluster.
Vector	computeCentroid() Compute the centroid by averaging the pointTotals
void	computeParameters() Compute a new set of posterior parameters based upon the Observations that have been observed since my creation
void	configure(org.apache.hadoop.conf.Configuration job)
long	count() Return the number of observations that have been observed by this model
void	createParameters(String prefix, org.apache.hadoop.conf.Configuration jobConf) EXPERT: consumers should never have to call this method.
static String	formatVector(Vector v, String[] bindings) Return a human-readable formatted string representation of the vector, not intended to be complete nor usable as an input/output representation
Vector	getCenter() Get the "center" of the Cluster as a Vector
int	getId() Get the id of the Cluster
abstract String	getIdentifier()
long	getNumPoints() Get an integer denoting the number of points observed by this cluster
ClusterObservations	getObservations()
Collection<Parameter<?>>	getParameters()
Vector	getRadius() Get the "radius" of the Cluster as a Vector.
protected double	getS0()
protected Vector	getS1()
protected Vector	getS2()
boolean	isConverged()
void	observe(ClusterObservations observations)
void	observe(Model<VectorWritable> x) Observe the given model, retaining information about its observations
void	observe(Vector x)
void	observe(Vector x, double weight)
void	observe(VectorWritable x) Observe the given observation, retaining information about it
void	observe(VectorWritable x, double weight) Observe the given observation, retaining information about it
void	readFields(DataInput in)
protected void	setCenter(Vector center)
protected void	setId(int id)
protected void	setNumPoints(long l)
protected void	setRadius(Vector radius)
protected void	setS0(double s0)
protected void	setS1(Vector s1)
protected void	setS2(Vector s2)
void	write(DataOutput out)

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.mahout.clustering.Model

pdf, sampleFromPosterior

Constructor Detail

AbstractCluster

protected AbstractCluster()

AbstractCluster

protected AbstractCluster(Vector point,
                          int id2)

AbstractCluster

protected AbstractCluster(Vector center2,
                          Vector radius2,
                          int id2)

Method Detail

configure

public void configure(org.apache.hadoop.conf.Configuration job)

Specified by:

configure in interface Parametered

getParameters

public Collection<Parameter<?>> getParameters()

Specified by:

getParameters in interface Parametered

createParameters

public void createParameters(String prefix,
                             org.apache.hadoop.conf.Configuration jobConf)

Description copied from interface: Parametered

EXPERT: consumers should never have to call this method. It would be friendly visible to Parametered.ParameteredGeneralizations if java supported it. Calling this method should create a new list of parameters and is called

Specified by:

createParameters in interface Parametered

Parameters:

prefix - ends with a dot if not empty.

jobConf - configuration used for retrieving values

See Also:

invoking method, invoking method

setId

protected void setId(int id)

Parameters:

id - the id to set

setNumPoints

protected void setNumPoints(long l)

Parameters:

l - the numPoints to set

setCenter

protected void setCenter(Vector center)

Parameters:

center - the center to set

setRadius

protected void setRadius(Vector radius)

Parameters:

radius - the radius to set

getS0

protected double getS0()

Returns:

the s0

getS1

protected Vector getS1()

Returns:

the s1

getS2

protected Vector getS2()

Returns:

the s2

observe

public void observe(Model<VectorWritable> x)

Description copied from interface: Model

Observe the given model, retaining information about its observations

Specified by:

observe in interface Model<VectorWritable>

Parameters:

x - a Model<0>

observe

public void observe(ClusterObservations observations)

observe

public void observe(VectorWritable x)

Description copied from interface: Model

Observe the given observation, retaining information about it

Specified by:

observe in interface Model<VectorWritable>

Parameters:

x - an Observation from the posterior

observe

public void observe(VectorWritable x,
                    double weight)

Description copied from interface: Model

Observe the given observation, retaining information about it

Specified by:

observe in interface Model<VectorWritable>

Parameters:

x - an Observation from the posterior

weight - a double weighting factor

observe

public void observe(Vector x,
                    double weight)

observe

public void observe(Vector x)

getNumPoints

public long getNumPoints()

Description copied from interface: Cluster

Get an integer denoting the number of points observed by this cluster

Specified by:

getNumPoints in interface Cluster

Returns:

an integer

getObservations

public ClusterObservations getObservations()

computeParameters

public void computeParameters()

Description copied from interface: Model

Compute a new set of posterior parameters based upon the Observations that have been observed since my creation

Specified by:

computeParameters in interface Model<VectorWritable>

readFields

public void readFields(DataInput in)
                throws IOException

Specified by:

readFields in interface org.apache.hadoop.io.Writable

Throws:

IOException

write

public void write(DataOutput out)
           throws IOException

Specified by:

write in interface org.apache.hadoop.io.Writable

Throws:

IOException

asFormatString

public String asFormatString(String[] bindings)

Description copied from interface: Cluster

Produce a custom, human-friendly, printable representation of the Cluster.

Specified by:

asFormatString in interface Cluster

Parameters:

bindings - an optional String[] containing labels used to format the primary Vector/s of this implementation.

Returns:

a String

getIdentifier

public abstract String getIdentifier()

getCenter

public Vector getCenter()

Description copied from interface: Cluster

Get the "center" of the Cluster as a Vector

Specified by:

getCenter in interface Cluster

Returns:

a Vector

getId

public int getId()

Description copied from interface: Cluster

Get the id of the Cluster

Specified by:

getId in interface Cluster

Returns:

a unique integer

getRadius

public Vector getRadius()

Description copied from interface: Cluster

Get the "radius" of the Cluster as a Vector. Usually the radius is the standard deviation expressed as a Vector of size equal to the center. Some clusters may return zero values if not appropriate.

Specified by:

getRadius in interface Cluster

Returns:

aVector

computeCentroid

public Vector computeCentroid()

Compute the centroid by averaging the pointTotals

Returns:

the new centroid

formatVector

public static String formatVector(Vector v,
                                  String[] bindings)

Return a human-readable formatted string representation of the vector, not intended to be complete nor usable as an input/output representation

count

public long count()

Description copied from interface: Model

Return the number of observations that have been observed by this model

Specified by:

count in interface Model<VectorWritable>

Returns:

an int

isConverged

public boolean isConverged()

Specified by:

isConverged in interface Cluster

Returns:

if the receiver has converged, or false if that has no meaning for the implementation

setS0

protected void setS0(double s0)

setS1

protected void setS1(Vector s1)

setS2

protected void setS2(Vector s2)