magpie.cluster

Class BaseClusterer

java.lang.Object

magpie.cluster.BaseClusterer

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, Commandable, Options, Printable

Direct Known Subclasses:

WekaClusterer

public abstract class BaseClusterer
extends java.lang.Object
implements Options, Printable, java.io.Serializable, java.lang.Cloneable, Commandable

Abstract model for all clustering algorithms. Splits dataset into multiple sets based on its attributes.

Implemented Commands:

train $<dataset> - Train a clustering algorithm
dataset: Dataset to use for training

<output> = get <num> $<input> - Get a certain cluster from a dataset
num: Index of cluster to extract
input: Dataset containing entries to cluster
output: Dataset Single cluster from the input dataset.

selector $<selector> - Define an attribute selector to use before clustering
selector: BaseAttributeSelector to use for selection

split $<dataset> <filename> [<format>] - Split dataset into several clusters, save them
dataset: Dataset to be partitioned
filename: Base filename to use (actual file name will be <filename><cluster #>.<format>)
format: Optional: Format to use when saving clusters

Implemented Print Commands:

selector - Print out attributes used to generate cluster, if an BaseAttributeSelector was provided

stats [<command>] - Print out statistics about clustering
command: Command to be passed to and run by internal ClustererStatistics class.

Implementation guide:

In order to create a new instance of this class, one must implement:

• train_protected(magpie.data.Dataset) - Train the clusterer
• label_protected(magpie.data.Dataset) - Label entries based on their predicted cluster
• NClusters_protected() - Return the number of possible clusters
• Options.setOptions(java.util.List<java.lang.Object>) - Define options for this subset given options
• clone() - Make sure no clones overwrite data in original

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
BaseClusterer()

Method Summary

Modifier and Type	Method and Description
java.lang.String	about() Prints a simple status message about this object
BaseClusterer	clone()
Dataset	getCluster(Dataset From, int number) Retrieve a specific cluster from a dataset.
protected abstract java.util.List<java.lang.String>	getClustererDetails(boolean htmlFormat) Get any user-specified options of this clusterer.
boolean	isTrained()
protected abstract int[]	label_protected(Dataset Data) Actual implementation of label(magpie.data.Dataset).
int[]	label(Dataset Data) Mark the most likely cluster for each entry
protected abstract int	NClusters_protected() Actual implementation of NClusters().
int	NClusters() How many different clusters data may be partitioned into
Dataset[]	partitionDataset(Dataset Data) Separate the Dataset based on predicted classes.
java.lang.String	printCommand(java.util.List<java.lang.String> Command) Handles more complicated printing commands.
java.lang.String	printDescription(boolean htmlFormat) Print full name of object, and a simple description of the options.
java.lang.Object	runCommand(java.util.List<java.lang.Object> Command) Process some command described by a list of Objects.
void	setAttributeSelector(BaseAttributeSelector AttributeSelector) Define an attribute selector to be used before clustering
protected abstract void	train_protected(Dataset Data) Perform the actual training.
void	train(Dataset TrainingData) Train the clusterer

Methods inherited from class java.lang.Object

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

Methods inherited from interface magpie.utility.interfaces.Options

printUsage, setOptions

Constructor Detail

BaseClusterer

public BaseClusterer()

Method Detail

clone

public BaseClusterer clone()

Overrides:

clone in class java.lang.Object

setAttributeSelector

public void setAttributeSelector(BaseAttributeSelector AttributeSelector)

Define an attribute selector to be used before clustering

Parameters:

AttributeSelector - Instantiated Attribute Selector

isTrained

public boolean isTrained()

Returns:

Whether this clusterer has been trained

train

public void train(Dataset TrainingData)

Train the clusterer

Parameters:

TrainingData - Training data

train_protected

protected abstract void train_protected(Dataset Data)

Perform the actual training.

Parameters:

Data - Dataset to use for training

label

public int[] label(Dataset Data)

Mark the most likely cluster for each entry

Parameters:

Data - Dataset to label

Returns:

List of most likely cluster for each entry

label_protected

protected abstract int[] label_protected(Dataset Data)

Actual implementation of label(magpie.data.Dataset).

Parameters:

Data - Dataset to be labeled

Returns:

List of most likely cluster for each entry

NClusters

public int NClusters()

How many different clusters data may be partitioned into

Returns:

How many clusters are defined

NClusters_protected

protected abstract int NClusters_protected()

Actual implementation of NClusters().

Returns:

How many different clusters are defined

partitionDataset

public Dataset[] partitionDataset(Dataset Data)

Separate the Dataset based on predicted classes. Does not alter input

Parameters:

Data - Dataset to be partitioned (does not get altered)s

Returns:

Array of datasets containing only entries classified into certain clusters

getCluster

public Dataset getCluster(Dataset From,
                          int number)

Retrieve a specific cluster from a dataset. Input dataset is not changed

Parameters:

From - Dataset to be clustered

number - Cluster number to extract

Returns:

Dataset containing only entries

about

public java.lang.String about()

Description copied from interface: Printable

Prints a simple status message about this object

Specified by:

about in interface Printable

Returns:

Short status message

printDescription

public java.lang.String printDescription(boolean htmlFormat)

Description copied from interface: Printable

Print full name of object, and a simple description of the options. This description should print enough information to reproduce how this object was created, but not necessarily enough to run it (e.g., coefficients for the model).

Example: For a model training a separate WekaRegression for intermetallics

magpie.models.regression.SplitRegression

Splitter: AllMetalsSplitter
All Metals: magpie.models.regression.WekaRegression trees.REPTree
Contains Nonmetal: magpie.regression.LASSORegreession -maxterms 2

Specified by:

printDescription in interface Printable

Parameters:

htmlFormat - Whether format for output to an HTML page (e.g., <div> to create indentation) or for printing to screen.

Returns:

String describing the model

See Also:

#printModel()

getClustererDetails

protected abstract java.util.List<java.lang.String> getClustererDetails(boolean htmlFormat)

Get any user-specified options of this clusterer. Used with printDescription(boolean)

Returns:

List of clusterer options

printCommand

public java.lang.String printCommand(java.util.List<java.lang.String> Command)
                              throws java.lang.Exception

Description copied from interface: Printable

Handles more complicated printing commands.

Specified by:

printCommand in interface Printable

Parameters:

Command - Command specifying what to print

Returns:

String formatted as requested

Throws:

java.lang.Exception - If command not understood

runCommand

public java.lang.Object runCommand(java.util.List<java.lang.Object> Command)
                            throws java.lang.Exception

Description copied from interface: Commandable

Process some command described by a list of Objects.

Specified by:

runCommand in interface Commandable

Parameters:

Command - Command as a list of objects

Returns:

Any output (null if no output generated by command)

Throws:

java.lang.Exception - If something goes wrong