org.lenskit.eval.traintest

Class DataSet

java.lang.Object

org.lenskit.eval.traintest.DataSet

public class DataSet
extends java.lang.Object

A train-test data set.

Constructor Summary

Constructors

Constructor and Description
DataSet(java.lang.String name, StaticDataSource train, StaticDataSource rt, StaticDataSource test, java.util.UUID grp, java.util.Map<java.lang.String,java.lang.Object> attrs, java.util.List<EntityType> entityTypes) Create a new data set.

Method Summary

Modifier and Type	Method and Description
DataSetBuilder	copyBuilder() Create a new builder initialized with this data set’s values.
static DataSetBuilder	copyBuilder(DataSet data) Create a new builder initialized with this data set’s values.
static java.util.List<DataSet>	fromJSON(com.fasterxml.jackson.databind.JsonNode json, java.net.URI base) Load one or more data sets from JSON data.
LongSet	getAllItems()
java.util.Map<java.lang.String,java.lang.Object>	getAttributes() Get the data set attributes (used for identification in output).
java.util.List<EntityType>	getEntityTypes() Get the entity types registered with this builder so far.
LenskitConfiguration	getExtraConfiguration() Get extra LensKit configuration required by this data set.
java.util.UUID	getIsolationGroup() Get the isolation group ID for this data set.
java.lang.String	getName() Get the data set name.
StaticDataSource	getRuntimeData() Get the runtime data set.
StaticDataSource	getTestData() Get the training data.
StaticDataSource	getTrainingData() Get the training data.
static java.util.List<DataSet>	load(java.nio.file.Path file) Load one or more data sets from a YAML manifest file.
static java.util.List<DataSet>	load(java.net.URL url) Load one or more data sets from a YAML manifest file.
static DataSetBuilder	newBuilder() Create a new generic train-test data set builder.
static DataSetBuilder	newBuilder(java.lang.String name) Create a new generic train-test data set builder.
java.lang.String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

DataSet

public DataSet(@Nonnull
               java.lang.String name,
               @Nonnull
               StaticDataSource train,
               @Nullable
               StaticDataSource rt,
               @Nonnull
               StaticDataSource test,
               @Nonnull
               java.util.UUID grp,
               java.util.Map<java.lang.String,java.lang.Object> attrs,
               @Nonnull
               java.util.List<EntityType> entityTypes)

Create a new data set.

Parameters:

name - The name.

train - The training source.

test - The test data source.

grp - The data set isolation group.

attrs - The data set attributes.

Method Detail

getName

@Nonnull
public java.lang.String getName()

Get the data set name.

Returns:

A name for the data set. Used in the output file.

getAttributes

public java.util.Map<java.lang.String,java.lang.Object> getAttributes()

Get the data set attributes (used for identification in output).

Returns:

A key → value map of the attributes used to identify this data set. For example, a crossfold data set may include the source name and fold number.

getIsolationGroup

public java.util.UUID getIsolationGroup()

Get the isolation group ID for this data set. Data sets in the same group will be allowed to run in parallel. This is used to implement data set isolation.

Returns:

The group ID for this data set.

getTestData

@Nonnull
public StaticDataSource getTestData()

Get the training data.

Returns:

A data source containing the test data.

getTrainingData

@Nonnull
public StaticDataSource getTrainingData()

Get the training data.

Returns:

A data source containing the training data.

getRuntimeData

@Nullable
public StaticDataSource getRuntimeData()

Get the runtime data set. This is the data that should be available when the recommender is run, but not when its model is trained.

Returns:

The runtime data set.

getAllItems

public LongSet getAllItems()

getEntityTypes

@Nonnull
public java.util.List<EntityType> getEntityTypes()

Get the entity types registered with this builder so far.

Returns:

The entity types registered so far.

getExtraConfiguration

public LenskitConfiguration getExtraConfiguration()

Get extra LensKit configuration required by this data set.

Returns:

A LensKit configuration with additional configuration data for this data set.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

newBuilder

public static DataSetBuilder newBuilder()

Create a new generic train-test data set builder.

Returns:

The new builder.

newBuilder

public static DataSetBuilder newBuilder(java.lang.String name)

Create a new generic train-test data set builder.

Parameters:

name - The data set name.

Returns:

The new builder.

copyBuilder

public DataSetBuilder copyBuilder()

Create a new builder initialized with this data set’s values.

Returns:

A new builder initialized to make a copy of this data set definition.

copyBuilder

public static DataSetBuilder copyBuilder(DataSet data)

Create a new builder initialized with this data set’s values.

Returns:

A new builder initialized to make a copy of this data set definition.

fromJSON

public static java.util.List<DataSet> fromJSON(com.fasterxml.jackson.databind.JsonNode json,
                                               java.net.URI base)
                                        throws java.io.IOException

Load one or more data sets from JSON data.

Throws:

java.io.IOException

load

public static java.util.List<DataSet> load(java.nio.file.Path file)
                                    throws java.io.IOException

Load one or more data sets from a YAML manifest file.

Parameters:

file - The path to the YAML manifest file.

Returns:

The list of data sets.

Throws:

java.io.IOException

load

public static java.util.List<DataSet> load(java.net.URL url)
                                    throws java.io.IOException

Load one or more data sets from a YAML manifest file.

Parameters:

url - The URL of a the YAML manifest file.

Returns:

The list of data sets.

Throws:

java.io.IOException