org.tribuo.sequence

Class MinimumCardinalitySequenceDataset<T extends Output<T>>

java.lang.Object

org.tribuo.sequence.SequenceDataset<T>

org.tribuo.sequence.ImmutableSequenceDataset<T>

org.tribuo.sequence.MinimumCardinalitySequenceDataset<T>

Type Parameters:

T - The type of the outputs in this SequenceDataset.

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.provenance.Provenancable<DatasetProvenance>, Serializable, Iterable<SequenceExample<T>>

public class MinimumCardinalitySequenceDataset<T extends Output<T>>
extends ImmutableSequenceDataset<T>

This class creates a pruned dataset in which low frequency features that occur less than the provided minimum cardinality have been removed. This can be useful when the dataset is very large due to many low-frequency features. Here, a new dataset is created so that the feature counts are recalculated and so that the original, passed-in dataset is not modified. The returned dataset may have fewer sequence examples because if any of the sequence examples have examples with no features after the minimum cardinality has been applied, then those sequence examples will not be added to the constructed dataset.

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	MinimumCardinalitySequenceDataset.MinimumCardinalitySequenceDatasetProvenance Provenance for MinimumCardinalitySequenceDataset.

Field Summary

Fields inherited from class org.tribuo.sequence.ImmutableSequenceDataset

featureIDMap, outputIDInfo

Fields inherited from class org.tribuo.sequence.SequenceDataset

data, outputFactory, sourceProvenance

Constructor Summary

Constructors

Constructor and Description
MinimumCardinalitySequenceDataset(SequenceDataset<T> sequenceDataset, int minCardinality)

Method Summary

Modifier and Type	Method and Description
int	getMinCardinality() The minimum cardinality threshold for the features.
int	getNumExamplesRemoved() The number of examples removed due to a lack of features.
DatasetProvenance	getProvenance()
Set<String>	getRemoved() The feature names that were removed.

Methods inherited from class org.tribuo.sequence.ImmutableSequenceDataset

add, add, copyDataset, copyDataset, copyDataset, getFeatureIDMap, getFeatureMap, getOutputIDInfo, getOutputInfo, getOutputs, toString

Methods inherited from class org.tribuo.sequence.SequenceDataset

getData, getExample, getFlatDataset, getOutputFactory, getSourceDescription, getSourceProvenance, iterator, size

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Constructor Detail

MinimumCardinalitySequenceDataset

public MinimumCardinalitySequenceDataset(SequenceDataset<T> sequenceDataset,
                                         int minCardinality)

Parameters:

sequenceDataset - this dataset is left untouched and is used to populate the constructed dataset.

minCardinality - features with a frequency less than minCardinality will be removed.

Method Detail

getRemoved

public Set<String> getRemoved()

The feature names that were removed.

Returns:

The feature names.

getNumExamplesRemoved

public int getNumExamplesRemoved()

The number of examples removed due to a lack of features.

Returns:

The number of removed examples.

getMinCardinality

public int getMinCardinality()

The minimum cardinality threshold for the features.

Returns:

The cardinality threshold.

getProvenance

public DatasetProvenance getProvenance()

Specified by:

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

Overrides:

getProvenance in class ImmutableSequenceDataset<T extends Output<T>>