org.deeplearning4j.bagofwords.vectorizer

Class TfidfVectorizer.Builder

java.lang.Object

org.deeplearning4j.bagofwords.vectorizer.TfidfVectorizer.Builder

Enclosing class:

TfidfVectorizer

public static class TfidfVectorizer.Builder
extends java.lang.Object

Field Summary

Fields

Modifier and Type	Field and Description
protected boolean	isParallel
protected LabelAwareIterator	iterator
protected LabelsSource	labelsSource
protected int	minWordFrequency
protected java.util.Collection<java.lang.String>	stopWords
protected TokenizerFactory	tokenizerFactory
protected VocabCache<VocabWord>	vocabCache

Constructor Summary

Constructors

Constructor and Description
Builder()

Method Summary

Modifier and Type	Method and Description
TfidfVectorizer.Builder	allowParallelTokenization(boolean reallyAllow)
TfidfVectorizer	build()
TfidfVectorizer.Builder	setIterator(DocumentIterator iterator)
TfidfVectorizer.Builder	setIterator(LabelAwareIterator iterator)
TfidfVectorizer.Builder	setIterator(SentenceIterator iterator)
TfidfVectorizer.Builder	setMinWordFrequency(int minWordFrequency)
TfidfVectorizer.Builder	setStopWords(java.util.Collection<java.lang.String> stopWords)
TfidfVectorizer.Builder	setTokenizerFactory(TokenizerFactory tokenizerFactory)
TfidfVectorizer.Builder	setVocab(VocabCache<VocabWord> vocab)

Methods inherited from class java.lang.Object

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

Field Detail

tokenizerFactory

protected TokenizerFactory tokenizerFactory

iterator

protected LabelAwareIterator iterator

minWordFrequency

protected int minWordFrequency

vocabCache

protected VocabCache<VocabWord> vocabCache

labelsSource

protected LabelsSource labelsSource

stopWords

protected java.util.Collection<java.lang.String> stopWords

isParallel

protected boolean isParallel

Constructor Detail

Builder

public Builder()

Method Detail

allowParallelTokenization

public TfidfVectorizer.Builder allowParallelTokenization(boolean reallyAllow)

setTokenizerFactory

public TfidfVectorizer.Builder setTokenizerFactory(@NonNull
                                                   TokenizerFactory tokenizerFactory)

setIterator

public TfidfVectorizer.Builder setIterator(@NonNull
                                           LabelAwareIterator iterator)

setIterator

public TfidfVectorizer.Builder setIterator(@NonNull
                                           DocumentIterator iterator)

setIterator

public TfidfVectorizer.Builder setIterator(@NonNull
                                           SentenceIterator iterator)

setVocab

public TfidfVectorizer.Builder setVocab(@NonNull
                                        VocabCache<VocabWord> vocab)

setMinWordFrequency

public TfidfVectorizer.Builder setMinWordFrequency(int minWordFrequency)

setStopWords

public TfidfVectorizer.Builder setStopWords(java.util.Collection<java.lang.String> stopWords)

build

public TfidfVectorizer build()