Class RegressionDataGenerator

java.lang.Object

org.tribuo.regression.example.RegressionDataGenerator

public abstract class RegressionDataGenerator
extends Object

Generates two example train and test datasets, used for unit testing. They don't necessarily have linear regressed values, it's for testing the machinery rather than accuracy.

Also can generate a variety of single dimensional gaussian datasets.

Field Summary

Fields

Modifier and Type	Field	Description
static final String	firstDimensionName
static final String	secondDimensionName
static final String	SINGLE_DIM_NAME

Constructor Summary

Constructors

Constructor	Description
RegressionDataGenerator()

Method Summary

Modifier and Type	Method	Description
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	denseTrainTest()
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	denseTrainTest(double negate)	Generates a train/test dataset pair which is dense in the features, each example has 4 features,{A,B,C,D}.
static Example<Regressor>	emptyExample()	Generates an example with no features.
static Example<Regressor>	emptyMultiDimExample()	Generates an example with no features.
static Example<Regressor>	invalidMultiDimSparseExample()	Generates an example with the feature ids 1,5,8, which does not intersect with the ids used elsewhere in this class.
static Example<Regressor>	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<Regressor>, Dataset<Regressor>>	multiDimDenseTrainTest()
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	multiDimDenseTrainTest(double negate)	Generates a train/test dataset pair which is dense in the features, each example has 4 features,{A,B,C,D}.
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	multiDimSparseTrainTest()
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	multiDimSparseTrainTest(double negate)	Generates a pair of datasets, where the features are sparse, and unknown features appear in the test data.
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	sparseTrainTest()
static com.oracle.labs.mlrg.olcut.util.Pair<Dataset<Regressor>, Dataset<Regressor>>	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

Field Details

firstDimensionName

public static final String firstDimensionName

See Also:

• Constant Field Values

secondDimensionName

public static final String secondDimensionName

See Also:

• Constant Field Values

SINGLE_DIM_NAME

public static final String SINGLE_DIM_NAME

See Also:

• Constant Field Values

Constructor Details

RegressionDataGenerator

public RegressionDataGenerator()

Method Details

multiDimDenseTrainTest

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

multiDimDenseTrainTest

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

Generates a train/test dataset pair which is dense in the features, each example has 4 features,{A,B,C,D}.

Parameters:

negate - Supply -1.0 to negate some features.

Returns:

A pair of datasets.

multiDimSparseTrainTest

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

multiDimSparseTrainTest

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

Generates a pair of datasets, where the features are sparse, and unknown features appear in the test data.

Parameters:

negate - Supply -1.0 to negate some features.

Returns:

A pair of datasets.

invalidMultiDimSparseExample

public static Example<Regressor> invalidMultiDimSparseExample()

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}.

emptyMultiDimExample

public static Example<Regressor> emptyMultiDimExample()

Generates an example with no features.

Returns:

An example with no features.

denseTrainTest

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

denseTrainTest

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

Generates a train/test dataset pair which is dense in the features, each example has 4 features,{A,B,C,D}.

Parameters:

negate - Supply -1.0 to negate some values in this dataset.

Returns:

A pair of datasets.

sparseTrainTest

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

sparseTrainTest

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

Generates a pair of datasets, where the features are sparse, and unknown features appear in the test data.

Parameters:

negate - Supply -1.0 to negate some values in this dataset.

Returns:

A pair of datasets.

invalidSparseExample

public static Example<Regressor> 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<Regressor> emptyExample()

Generates an example with no features.

Returns:

An example with no features.