org.deeplearning4j.models.glove

Class CoOccurrences

java.lang.Object

org.deeplearning4j.models.glove.CoOccurrences

All Implemented Interfaces:

Serializable

public class CoOccurrences
extends Object
implements Serializable

Co occurrence counts

Author:

Adam Gibson

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	CoOccurrences.Builder
class	CoOccurrences.CoOccurrenceBatchIterator
class	CoOccurrences.CoOccurrenceIterator

Field Summary

Fields

Modifier and Type	Field and Description
protected VocabCache	cache
protected InvertedIndex	index
protected boolean	symmetric
protected akka.actor.ActorSystem	trainingSystem

Constructor Summary

Constructors

Constructor and Description
CoOccurrences(TokenizerFactory tokenizerFactory, SentenceIterator sentenceIterator, int windowSize, VocabCache cache, CounterMap<String,String> coOCurreneCounts, boolean symmetric)

Method Summary

Modifier and Type	Method and Description
Iterator<Pair<String,String>>	coOccurrenceIterator() Get an iterator over all possible non zero co occurrences
Iterator<Pair<VocabWord,VocabWord>>	coOccurrenceIteratorVocab()
Iterator<List<Pair<VocabWord,VocabWord>>>	coOccurrenceIteratorVocabBatch(int batchSize)
List<Pair<String,String>>	coOccurrenceList() Return a list of all of the co occurrences
double	count(String w1, String w2)
void	fit()
CounterMap<String,String>	getCoOCurreneCounts()
Counter<Integer>	getSentenceOccurrences()
static CoOccurrences	load(InputStream from) Load from an input stream with the following format: w1 w2 score
int	numCoOccurrences() The number of co occurrences
List<Pair<String,String>>	randomizedList() Return a randomized list of the co occurrences
void	setCoOCurreneCounts(CounterMap<String,String> coOCurreneCounts)
void	setSentenceOccurrences(Counter<Integer> sentenceOccurrences)

Methods inherited from class java.lang.Object

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

Field Detail

cache

protected transient VocabCache cache

index

protected InvertedIndex index

trainingSystem

protected transient akka.actor.ActorSystem trainingSystem

symmetric

protected boolean symmetric

Constructor Detail

CoOccurrences

public CoOccurrences(TokenizerFactory tokenizerFactory,
                     SentenceIterator sentenceIterator,
                     int windowSize,
                     VocabCache cache,
                     CounterMap<String,String> coOCurreneCounts,
                     boolean symmetric)

Method Detail

fit

public void fit()

coOccurrenceIteratorVocabBatch

public Iterator<List<Pair<VocabWord,VocabWord>>> coOccurrenceIteratorVocabBatch(int batchSize)

coOccurrenceIteratorVocab

public Iterator<Pair<VocabWord,VocabWord>> coOccurrenceIteratorVocab()

load

public static CoOccurrences load(InputStream from)

Load from an input stream with the following format: w1 w2 score

Parameters:

from - the input stream to read from

Returns:

the co occurrences based on the input stream

getSentenceOccurrences

public Counter<Integer> getSentenceOccurrences()

setSentenceOccurrences

public void setSentenceOccurrences(Counter<Integer> sentenceOccurrences)

coOccurrenceList

public List<Pair<String,String>> coOccurrenceList()

Return a list of all of the co occurrences

Returns:

a list of all of the co occurrences

randomizedList

public List<Pair<String,String>> randomizedList()

Return a randomized list of the co occurrences

Returns:

numCoOccurrences

public int numCoOccurrences()

The number of co occurrences

Returns:

count

public double count(String w1,
                    String w2)

coOccurrenceIterator

public Iterator<Pair<String,String>> coOccurrenceIterator()

Get an iterator over all possible non zero co occurrences

Returns:

the iterator

getCoOCurreneCounts

public CounterMap<String,String> getCoOCurreneCounts()

setCoOCurreneCounts

public void setCoOCurreneCounts(CounterMap<String,String> coOCurreneCounts)