org.tribuo.clustering.example

Class ClusteringDataGenerator

java.lang.Object

org.tribuo.clustering.example.ClusteringDataGenerator

public abstract class ClusteringDataGenerator
extends Object

Generates three example train and test datasets, used for unit testing. They don't necessarily have sensible cluster boundaries, it's for testing the machinery rather than accuracy.

Also has a dataset generator which returns a dataset sampled from a mixture of 2 dimensional gaussians.

Constructor Summary

Constructors

Constructor and Description
ClusteringDataGenerator()

Method Summary

Modifier and Type	Method and Description
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>>	denseTrainTest()
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>>	denseTrainTest(double negate) Generates a train/test dataset pair which is dense in the features, each example has 4 features,{A,B,C,D}, and there are 4 clusters, {0,1,2,3}.
static Example<ClusterID>	emptyExample() Generates an example with no features.
static Dataset<ClusterID>	gaussianClusters(long size, long seed) Generates a dataset drawn from a mixture of 5 2d gaussians.
static Example<ClusterID>	invalidSparseExample() Generates an example with the feature ids 1,5,8, which does not intersect with the ids used elsewhere in this class.
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>>	sparseTrainTest()
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>>	sparseTrainTest(double negate) Generates a pair of datasets, where the features are sparse, and unknown features appear in the test data.

Methods inherited from class java.lang.Object

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

Constructor Detail

ClusteringDataGenerator

public ClusteringDataGenerator()

Method Detail

gaussianClusters

public static Dataset<ClusterID> gaussianClusters(long size,
                                                  long seed)

Generates a dataset drawn from a mixture of 5 2d gaussians.

Parameters:

size - The number of points to sample for the dataset.

seed - The RNG seed.

Returns:

A pair of datasets.

denseTrainTest

public static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>> denseTrainTest()

denseTrainTest

public static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>> denseTrainTest(double negate)

Generates a train/test dataset pair which is dense in the features, each example has 4 features,{A,B,C,D}, and there are 4 clusters, {0,1,2,3}.

Parameters:

negate - Supply -1.0 to negate some feature values.

Returns:

A pair of datasets.

sparseTrainTest

public static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>> sparseTrainTest()

sparseTrainTest

public static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<ClusterID>,Dataset<ClusterID>> sparseTrainTest(double negate)

Generates a pair of datasets, where the features are sparse, and unknown features appear in the test data. It has the same 4 clusters {0,1,2,3}.

Parameters:

negate - Supply -1.0 to negate some feature values.

Returns:

A pair of datasets.

invalidSparseExample

public static Example<ClusterID> invalidSparseExample()

Generates an example with the feature ids 1,5,8, which does not intersect with the ids used elsewhere in this class. This should make the example empty at prediction time.

Returns:

An example with features {1:1.0,5:5.0,8:8.0}.

emptyExample

public static Example<ClusterID> emptyExample()

Generates an example with no features.

Returns:

An example with no features.