org.lenskit.eval.crossfold

Class Crossfolder

java.lang.Object

org.lenskit.eval.crossfold.Crossfolder

public class Crossfolder
extends java.lang.Object

Partitions a data set for cross-validation.

The resulting data is placed in an output directory with the following files:

• datasets.yaml - a manifest file listing all the data sets
• partNN.train.csv - a CSV file containing the train data for part NN
• partNN.train.yaml - a YAML manifest for the training data for part NN
• partNN.test.csv - a CSV file containing the test data for part NN
• partNN.test.yaml - a YAML manifest for the test data for part NN

Field Summary

Fields

Modifier and Type	Field and Description
static java.lang.String	ITEM_FILE_NAME

Constructor Summary

Constructors

Constructor and Description
Crossfolder()
Crossfolder(java.lang.String n)

Method Summary

Modifier and Type	Method and Description
void	execute() Run the crossfold command.
java.util.List<DataSet>	getDataSets() Get the train-test splits as data sets.
EntityType	getEntityType() Get the entity type that this crossfolder will crossfold.
CrossfoldMethod	getMethod() Get the method to be used for crossfolding.
java.lang.String	getName() Get the visible name of this crossfold split.
java.nio.file.Path	getOutputDir() Get the output directory.
OutputFormat	getOutputFormat() Get the output format for the crossfolder.
int	getPartitionCount() Get the partition count.
StaticDataSource	getSource() Get the data source backing this crossfold manager.
boolean	getWriteTimestamps() Query whether timestamps will be written.
void	setEntityType(EntityType entityType) Set the entity type that this crossfolder will crossfold.
Crossfolder	setMethod(CrossfoldMethod meth) Set the method to be used by the crossfolder.
Crossfolder	setName(java.lang.String n) Set a name for this crossfolder.
Crossfolder	setOutputDir(java.io.File dir) Set the output directory for this crossfold operation.
Crossfolder	setOutputDir(java.nio.file.Path dir) Set the output directory for this crossfold operation.
Crossfolder	setOutputDir(java.lang.String dir) Set the output directory for this crossfold operation.
Crossfolder	setOutputFormat(OutputFormat format) Set the output format for the crossfolder.
Crossfolder	setPartitionCount(int partition) Set the number of partitions to generate.
Crossfolder	setSource(StaticDataSource src) Set the data source.
Crossfolder	setWriteTimestamps(boolean pack) Configure whether to include timestamps in the output file.
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Field Detail

ITEM_FILE_NAME

public static final java.lang.String ITEM_FILE_NAME

See Also:

Constant Field Values

Constructor Detail

Crossfolder

public Crossfolder()

Crossfolder

public Crossfolder(java.lang.String n)

Method Detail

getEntityType

public EntityType getEntityType()

Get the entity type that this crossfolder will crossfold.

Returns:

The entity type to crossfold.

setEntityType

public void setEntityType(EntityType entityType)

Set the entity type that this crossfolder will crossfold.

Parameters:

entityType - The entity type to crossfold.

setPartitionCount

public Crossfolder setPartitionCount(int partition)

Set the number of partitions to generate.

Parameters:

partition - The number of paritions

Returns:

The CrossfoldCommand object (for chaining)

getPartitionCount

public int getPartitionCount()

Get the partition count.

Returns:

The number of partitions that will be generated.

setOutputFormat

public Crossfolder setOutputFormat(OutputFormat format)

Set the output format for the crossfolder.

Parameters:

format - The output format.

Returns:

The crossfolder (for chaining).

getOutputFormat

public OutputFormat getOutputFormat()

Get the output format for the crossfolder.

Returns:

The format the crossfolder will use for writing its output.

setOutputDir

public Crossfolder setOutputDir(java.nio.file.Path dir)

Set the output directory for this crossfold operation.

Parameters:

dir - The output directory.

Returns:

The crossfolder (for chaining).

setOutputDir

public Crossfolder setOutputDir(java.io.File dir)

Set the output directory for this crossfold operation.

Parameters:

dir - The output directory.

Returns:

The crossfolder (for chaining).

setOutputDir

public Crossfolder setOutputDir(java.lang.String dir)

Set the output directory for this crossfold operation.

Parameters:

dir - The output directory.

Returns:

The crossfolder (for chaining).

getOutputDir

public java.nio.file.Path getOutputDir()

Get the output directory.

Returns:

The directory into which crossfolding output will be placed.

setSource

public Crossfolder setSource(StaticDataSource src)

Set the data source.

Parameters:

src -

Returns:

The crossfolder (for chaining)

setMethod

public Crossfolder setMethod(CrossfoldMethod meth)

Set the method to be used by the crossfolder.

Parameters:

meth - The method to use.

Returns:

The crossfolder (for chaining).

getMethod

public CrossfoldMethod getMethod()

Get the method to be used for crossfolding.

Returns:

The configured crossfold method.

setWriteTimestamps

public Crossfolder setWriteTimestamps(boolean pack)

Configure whether to include timestamps in the output file.

Parameters:

pack - true to include timestamps (the default), false otherwise.

Returns:

The task (for chaining).

getWriteTimestamps

public boolean getWriteTimestamps()

Query whether timestamps will be written.

Returns:

true if output will include timestamps.

getName

public java.lang.String getName()

Get the visible name of this crossfold split.

Returns:

The name of the crossfold split.

setName

public Crossfolder setName(java.lang.String n)

Set a name for this crossfolder. It will be used to generate the names of individual data sets, for example.

Parameters:

n - The crossfolder name.

Returns:

The crossfolder (for chaining).

getSource

public StaticDataSource getSource()

Get the data source backing this crossfold manager.

Returns:

The underlying data source.

execute

public void execute()
             throws java.io.IOException

Run the crossfold command. Write the partition files to the disk by reading in the source file.

Throws:

java.io.IOException

getDataSets

public java.util.List<DataSet> getDataSets()

Get the train-test splits as data sets.

Returns:

The data sets produced by this crossfolder.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object