Class GaussianAnomalyDataSource

java.lang.Object

org.tribuo.anomaly.example.GaussianAnomalyDataSource

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<DataSourceProvenance>, Iterable<Example<Event>>, ConfigurableDataSource<Event>, DataSource<Event>

public final class GaussianAnomalyDataSource
extends Object
implements ConfigurableDataSource<Event>

Generates an anomaly detection dataset sampling each feature uniformly from a univariate Gaussian.

Or equivalently sampling all the features from a spherical Gaussian. Can accept at most 26 features.

By default the expected means are (1.0, 2.0, 1.0, 2.0, 5.0), with variances (1.0, 0.5, 0.25, 1.0, 0.1). The anomalous means are (-2.0, 2.0, -2.0, 2.0, -10.0), with variances (1.0, 0.5, 0.25, 1.0, 0.1) which are the same as the default expected variances.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final class	GaussianAnomalyDataSource.GaussianAnomalyDataSourceProvenance	Provenance for GaussianAnomalyDataSource.

Constructor Summary

Constructors

Constructor	Description
GaussianAnomalyDataSource(int numSamples, double[] expectedMeans, double[] expectedVariances, double[] anomalousMeans, double[] anomalousVariances, float fractionAnomalous, long seed)	Generates anomaly detection examples sampling each feature uniformly from a univariate Gaussian.
GaussianAnomalyDataSource(int numSamples, float fractionAnomalous, long seed)	Generates anomaly detection examples sampling each feature uniformly from a univariate Gaussian.

Method Summary

Modifier and Type	Method	Description
static MutableDataset<Event>	generateDataset(int numSamples, double[] expectedMeans, double[] expectedVariances, double[] anomalousMeans, double[] anomalousVariances, float fractionAnomalous, long seed)	Generates an anomaly detection dataset sampling each feature uniformly from a univariate Gaussian.
OutputFactory<Event>	getOutputFactory()	Returns the OutputFactory associated with this Output subclass.
DataSourceProvenance	getProvenance()
Iterator<Example<Event>>	iterator()
void	postConfig()	Used by the OLCUT configuration system, and should not be called by external code.

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Constructor Details

GaussianAnomalyDataSource

public GaussianAnomalyDataSource(int numSamples,
 float fractionAnomalous,
 long seed)

Generates anomaly detection examples sampling each feature uniformly from a univariate Gaussian.

Or equivalently sampling all the features from a spherical Gaussian.

Can accept at most 26 features.

Parameters:

numSamples - The size of the output dataset.

fractionAnomalous - The fraction of anomalies in the generated data.

seed - The rng seed to use.

GaussianAnomalyDataSource

public GaussianAnomalyDataSource(int numSamples,
 double[] expectedMeans,
 double[] expectedVariances,
 double[] anomalousMeans,
 double[] anomalousVariances,
 float fractionAnomalous,
 long seed)

Generates anomaly detection examples sampling each feature uniformly from a univariate Gaussian.

Or equivalently sampling all the features from a spherical Gaussian.

Can accept at most 26 features.

Parameters:

numSamples - The size of the output dataset.

expectedMeans - The means of the expected event features.

expectedVariances - The variances of the expected event features.

anomalousMeans - The means of the anomalous event features.

anomalousVariances - The variances of the anomalous event features.

fractionAnomalous - The fraction of anomalies to generate.

seed - The rng seed to use.

Method Details

postConfig

public void postConfig()

Used by the OLCUT configuration system, and should not be called by external code.

Specified by:

postConfig in interface com.oracle.labs.mlrg.olcut.config.Configurable

getOutputFactory

public OutputFactory<Event> getOutputFactory()

Description copied from interface: DataSource

Returns the OutputFactory associated with this Output subclass.

Specified by:

getOutputFactory in interface DataSource<Event>

Returns:

The output factory.

getProvenance

public DataSourceProvenance getProvenance()

Specified by:

getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<DataSourceProvenance>

iterator

public Iterator<Example<Event>> iterator()

Specified by:

iterator in interface Iterable<Example<Event>>

generateDataset

public static MutableDataset<Event> generateDataset(int numSamples,
 double[] expectedMeans,
 double[] expectedVariances,
 double[] anomalousMeans,
 double[] anomalousVariances,
 float fractionAnomalous,
 long seed)

Generates an anomaly detection dataset sampling each feature uniformly from a univariate Gaussian.

Or equivalently sampling all the features from a spherical Gaussian.

Can accept at most 26 features.

Parameters:

numSamples - The size of the output dataset.

expectedMeans - The means of the expected event features.

expectedVariances - The variances of the expected event features.

anomalousMeans - The means of the anomalous event features.

anomalousVariances - The variances of the anomalous event features.

fractionAnomalous - The fraction of anomalies to generate.

seed - The rng seed to use.

Returns:

A dataset drawn from a gaussian.