org.deeplearning4j.bagofwords.vectorizer

Class BaseTextVectorizer

java.lang.Object

org.deeplearning4j.bagofwords.vectorizer.BaseTextVectorizer

All Implemented Interfaces:

Serializable, TextVectorizer, Vectorizer

Direct Known Subclasses:

BagOfWordsVectorizer, TfidfVectorizer

public abstract class BaseTextVectorizer
extends Object
implements TextVectorizer

Base text vectorizer for handling creation of vocab

Author:

Adam Gibson

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected int	batchSize
protected VocabCache	cache
protected DocumentIterator	docIter
protected InvertedIndex	index
protected List<String>	labels
protected int	minWordFrequency
protected AtomicLong	numWordsEncountered
protected double	sample
protected SentenceIterator	sentenceIterator
protected List<String>	stopWords
protected TokenizerFactory	tokenizerFactory
protected static akka.actor.ActorSystem	trainingSystem

Constructor Summary

Constructors

Modifier	Constructor and Description
	BaseTextVectorizer()
protected	BaseTextVectorizer(VocabCache cache, TokenizerFactory tokenizerFactory, List<String> stopWords, int layerSize, int minWordFrequency, DocumentIterator docIter, SentenceIterator sentenceIterator, List<String> labels, InvertedIndex index, int batchSize, double sample)

Method Summary

Modifier and Type	Method and Description
int	batchSize() For word vectors, this is the batch size for how to partition documents in to workloads
void	fit() Train the model
VocabCache	getCache()
DocumentIterator	getDocIter()
int	getLayerSize()
int	getMinWordFrequency()
SentenceIterator	getSentenceIterator()
List<String>	getStopWords()
TokenizerFactory	getTokenizerFactory()
InvertedIndex	index() Inverted index
long	numWordsEncountered() Returns the number of words encountered so far
double	sample() Sampling for building mini batches
void	setCache(VocabCache cache)
void	setDocIter(DocumentIterator docIter)
void	setLayerSize(int layerSize)
void	setMinWordFrequency(int minWordFrequency)
void	setSentenceIterator(SentenceIterator sentenceIterator)
void	setStopWords(List<String> stopWords)
void	setTokenizerFactory(TokenizerFactory tokenizerFactory)
VocabCache	vocab() The vocab sorted in descending order

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.deeplearning4j.bagofwords.vectorizer.TextVectorizer

transform, vectorize, vectorize, vectorize

Methods inherited from interface org.deeplearning4j.datasets.vectorizer.Vectorizer

vectorize

Field Detail

cache

protected transient VocabCache cache

trainingSystem

protected static akka.actor.ActorSystem trainingSystem

tokenizerFactory

protected transient TokenizerFactory tokenizerFactory

stopWords

protected List<String> stopWords

minWordFrequency

protected int minWordFrequency

docIter

protected transient DocumentIterator docIter

labels

protected List<String> labels

sentenceIterator

protected transient SentenceIterator sentenceIterator

numWordsEncountered

protected AtomicLong numWordsEncountered

index

protected InvertedIndex index

batchSize

protected int batchSize

sample

protected double sample

Constructor Detail

BaseTextVectorizer

public BaseTextVectorizer()

BaseTextVectorizer

protected BaseTextVectorizer(VocabCache cache,
                             TokenizerFactory tokenizerFactory,
                             List<String> stopWords,
                             int layerSize,
                             int minWordFrequency,
                             DocumentIterator docIter,
                             SentenceIterator sentenceIterator,
                             List<String> labels,
                             InvertedIndex index,
                             int batchSize,
                             double sample)

Method Detail

batchSize

public int batchSize()

Description copied from interface: TextVectorizer

For word vectors, this is the batch size for how to partition documents in to workloads

Specified by:

batchSize in interface TextVectorizer

Returns:

the batchsize for partitioning documents in to workloads

sample

public double sample()

Description copied from interface: TextVectorizer

Sampling for building mini batches

Specified by:

sample in interface TextVectorizer

Returns:

the sampling

fit

public void fit()

Description copied from interface: TextVectorizer

Train the model

Specified by:

fit in interface TextVectorizer

vocab

public VocabCache vocab()

Description copied from interface: TextVectorizer

The vocab sorted in descending order

Specified by:

vocab in interface TextVectorizer

Returns:

the vocab sorted in descending order

getSentenceIterator

public SentenceIterator getSentenceIterator()

setSentenceIterator

public void setSentenceIterator(SentenceIterator sentenceIterator)

getDocIter

public DocumentIterator getDocIter()

setDocIter

public void setDocIter(DocumentIterator docIter)

getMinWordFrequency

public int getMinWordFrequency()

setMinWordFrequency

public void setMinWordFrequency(int minWordFrequency)

getLayerSize

public int getLayerSize()

setLayerSize

public void setLayerSize(int layerSize)

getStopWords

public List<String> getStopWords()

setStopWords

public void setStopWords(List<String> stopWords)

getTokenizerFactory

public TokenizerFactory getTokenizerFactory()

setTokenizerFactory

public void setTokenizerFactory(TokenizerFactory tokenizerFactory)

getCache

public VocabCache getCache()

setCache

public void setCache(VocabCache cache)

numWordsEncountered

public long numWordsEncountered()

Description copied from interface: TextVectorizer

Returns the number of words encountered so far

Specified by:

numWordsEncountered in interface TextVectorizer

Returns:

the number of words encountered so far

index

public InvertedIndex index()

Description copied from interface: TextVectorizer

Inverted index

Specified by:

index in interface TextVectorizer

Returns:

the inverted index for this vectorizer