org.tribuo.impl

Class IndexedArrayExample<T extends Output<T>>

java.lang.Object

org.tribuo.Example<T>

org.tribuo.impl.ArrayExample<T>

org.tribuo.impl.IndexedArrayExample<T>

All Implemented Interfaces:

Serializable, Iterable<Feature>

public class IndexedArrayExample<T extends Output<T>>
extends ArrayExample<T>

A version of ArrayExample which also has the id numbers.

Used in feature selection to provide log n lookups. May be used elsewhere in the future as a performance optimisation.

Note: output id caching is only valid with single dimensional Outputs like ClusterID, Event and Label. Other outputs may return -1 from getOutputID().

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	IndexedArrayExample.FeatureTuple A tuple of the feature name, id and value.

Field Summary

Fields

Modifier and Type	Field and Description
protected int[]	featureIDs
protected int	outputID

Fields inherited from class org.tribuo.impl.ArrayExample

DEFAULT_SIZE, featureNames, featureValues, size

Fields inherited from class org.tribuo.Example

DEFAULT_METADATA_SIZE, DEFAULT_WEIGHT, metadata, NAME, output, weight

Constructor Summary

Constructors

Constructor and Description
IndexedArrayExample(Example<T> other, ImmutableFeatureMap featureMap, ImmutableOutputInfo<T> outputMap) This constructor removes unknown features.
IndexedArrayExample(IndexedArrayExample<T> other) Copy constructor.

Method Summary

Modifier and Type	Method and Description
void	add(Feature feature) Adds a feature.
void	addAll(Collection<? extends Feature> features) Adds a collection of features.
boolean	contains(int i) Does this example contain a feature with id i.
IndexedArrayExample<T>	copy() Returns a deep copy of this Example.
void	densify(List<String> featureList) Converts all implicit zeros into explicit zeros based on the supplied feature names.
boolean	equals(Object o)
int	getIdx(int i) Gets the feature at internal index i.
int	getOutputID() Gets the output id dimension number.
protected void	growArray(int minCapacity) Grows the backing arrays storing the names and values.
int	hashCode()
Iterable<IndexedArrayExample.FeatureTuple>	idIterator() Iterator over the feature ids and values.
void	reduceByName(Merger merger) Merges features with the same name using the supplied Merger.
void	removeFeatures(List<Feature> featureList) Removes all features in this list from the Example.
protected void	sort() Sorts the feature list to maintain the lexicographic order invariant.

Methods inherited from class org.tribuo.impl.ArrayExample

add, canonicalize, copyValues, growArray, isDense, iterator, lookup, newCapacity, set, size, toString, transform, validateExample

Methods inherited from class org.tribuo.Example

containsMetadata, densify, getMetadata, getMetadataValue, getOutput, getWeight, setMetadataValue, setWeight

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Field Detail

featureIDs

protected int[] featureIDs

outputID

protected final int outputID

Constructor Detail

IndexedArrayExample

public IndexedArrayExample(IndexedArrayExample<T> other)

Copy constructor.

Parameters:

other - The example to copy.

IndexedArrayExample

public IndexedArrayExample(Example<T> other,
                           ImmutableFeatureMap featureMap,
                           ImmutableOutputInfo<T> outputMap)

This constructor removes unknown features.

Parameters:

other - The example to copy from.

featureMap - The feature map.

outputMap - The output info.

Method Detail

equals

public boolean equals(Object o)

Overrides:

equals in class ArrayExample<T extends Output<T>>

hashCode

public int hashCode()

Overrides:

hashCode in class ArrayExample<T extends Output<T>>

growArray

protected void growArray(int minCapacity)

Description copied from class: ArrayExample

Grows the backing arrays storing the names and values.

Overrides:

growArray in class ArrayExample<T extends Output<T>>

Parameters:

minCapacity - The new minimum capacity required.

add

public void add(Feature feature)

Description copied from class: Example

Adds a feature. This maintains the sorted invariant and has a lg(example.size()) cost per insertion.

Overrides:

add in class ArrayExample<T extends Output<T>>

Parameters:

feature - The feature to add.

addAll

public void addAll(Collection<? extends Feature> features)

Description copied from class: Example

Adds a collection of features. This maintains the sorted invariant but is more efficient than adding one at a time due to allocation.

Overrides:

addAll in class ArrayExample<T extends Output<T>>

Parameters:

features - The features to add.

sort

protected void sort()

Description copied from class: ArrayExample

Sorts the feature list to maintain the lexicographic order invariant.

Overrides:

sort in class ArrayExample<T extends Output<T>>

reduceByName

public void reduceByName(Merger merger)

Description copied from class: Example

Merges features with the same name using the supplied Merger.

Overrides:

reduceByName in class ArrayExample<T extends Output<T>>

Parameters:

merger - A function to merge two doubles.

removeFeatures

public void removeFeatures(List<Feature> featureList)

Description copied from class: Example

Removes all features in this list from the Example.

Overrides:

removeFeatures in class ArrayExample<T extends Output<T>>

Parameters:

featureList - Features to remove from this Example.

contains

public boolean contains(int i)

Does this example contain a feature with id i.

Parameters:

i - The index to check.

Returns:

True if the example contains the id.

copy

public IndexedArrayExample<T> copy()

Description copied from class: Example

Returns a deep copy of this Example.

Overrides:

copy in class ArrayExample<T extends Output<T>>

Returns:

A deep copy of this example.

densify

public void densify(List<String> featureList)

Description copied from class: Example

Converts all implicit zeros into explicit zeros based on the supplied feature names.

Overrides:

densify in class ArrayExample<T extends Output<T>>

Parameters:

featureList - A *sorted* list of feature names.

getIdx

public int getIdx(int i)

Gets the feature at internal index i.

Parameters:

i - The internal index.

Returns:

The feature index.

getOutputID

public int getOutputID()

Gets the output id dimension number.

Returns:

The output id.

idIterator

public Iterable<IndexedArrayExample.FeatureTuple> idIterator()

Iterator over the feature ids and values.

Returns:

The feature ids and values.