org.deeplearning4j.models.embeddings.wordvectors

Interface WordVectors

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

Glove, Node2Vec, ParagraphVectors, SequenceVectors, SparkParagraphVectors, SparkSequenceVectors, SparkWord2Vec, StaticWord2Vec, Word2Vec, Word2Vec, WordVectorsImpl

public interface WordVectors
extends java.io.Serializable

Word vectors. Handles operations based on the lookup table and vocab.

Method Summary

Modifier and Type	Method and Description
java.util.Map<java.lang.String,java.lang.Double>	accuracy(java.util.List<java.lang.String> questions) Accuracy based on questions which are a space separated list of strings where the first word is the query word, the next 2 words are negative, and the last word is the predicted word to be nearest
java.lang.String	getUNK()
double[]	getWordVector(java.lang.String word) Get the word vector for a given matrix
org.nd4j.linalg.api.ndarray.INDArray	getWordVectorMatrix(java.lang.String word) Get the word vector for a given matrix
org.nd4j.linalg.api.ndarray.INDArray	getWordVectorMatrixNormalized(java.lang.String word) Returns the word vector divided by the norm2 of the array
org.nd4j.linalg.api.ndarray.INDArray	getWordVectors(java.util.Collection<java.lang.String> labels) This method returns 2D array, where each row represents corresponding word/label
org.nd4j.linalg.api.ndarray.INDArray	getWordVectorsMean(java.util.Collection<java.lang.String> labels) This method returns mean vector, built from words/labels passed in
boolean	hasWord(java.lang.String word) Returns true if the model has this word in the vocab
int	indexOf(java.lang.String word)
WeightLookupTable	lookupTable() Lookup table for the vectors
void	setModelUtils(ModelUtils utils) Specifies ModelUtils to be used to access model
void	setUNK(java.lang.String newUNK)
double	similarity(java.lang.String word, java.lang.String word2) Returns the similarity of 2 words
java.util.List<java.lang.String>	similarWordsInVocabTo(java.lang.String word, double accuracy) Find all words with a similar characters in the vocab
VocabCache	vocab() Vocab for the vectors
java.util.Collection<java.lang.String>	wordsNearest(java.util.Collection<java.lang.String> positive, java.util.Collection<java.lang.String> negative, int top) Words nearest based on positive and negative words
java.util.Collection<java.lang.String>	wordsNearest(org.nd4j.linalg.api.ndarray.INDArray words, int top)
java.util.Collection<java.lang.String>	wordsNearest(java.lang.String word, int n) Get the top n words most similar to the given word
java.util.Collection<java.lang.String>	wordsNearestSum(java.util.Collection<java.lang.String> positive, java.util.Collection<java.lang.String> negative, int top) Words nearest based on positive and negative words
java.util.Collection<java.lang.String>	wordsNearestSum(org.nd4j.linalg.api.ndarray.INDArray words, int top)
java.util.Collection<java.lang.String>	wordsNearestSum(java.lang.String word, int n) Get the top n words most similar to the given word

Method Detail

getUNK

java.lang.String getUNK()

setUNK

void setUNK(java.lang.String newUNK)

hasWord

boolean hasWord(java.lang.String word)

Returns true if the model has this word in the vocab

Parameters:

word - the word to test for

Returns:

true if the model has the word in the vocab

wordsNearest

java.util.Collection<java.lang.String> wordsNearest(org.nd4j.linalg.api.ndarray.INDArray words,
                                                    int top)

wordsNearestSum

java.util.Collection<java.lang.String> wordsNearestSum(org.nd4j.linalg.api.ndarray.INDArray words,
                                                       int top)

wordsNearestSum

java.util.Collection<java.lang.String> wordsNearestSum(java.lang.String word,
                                                       int n)

Get the top n words most similar to the given word

Parameters:

word - the word to compare

n - the n to get

Returns:

the top n words

wordsNearestSum

java.util.Collection<java.lang.String> wordsNearestSum(java.util.Collection<java.lang.String> positive,
                                                       java.util.Collection<java.lang.String> negative,
                                                       int top)

Words nearest based on positive and negative words

Parameters:

positive - the positive words

negative - the negative words

top - the top n words

Returns:

the words nearest the mean of the words

accuracy

java.util.Map<java.lang.String,java.lang.Double> accuracy(java.util.List<java.lang.String> questions)

Accuracy based on questions which are a space separated list of strings where the first word is the query word, the next 2 words are negative, and the last word is the predicted word to be nearest

Parameters:

questions - the questions to ask

Returns:

the accuracy based on these questions

indexOf

int indexOf(java.lang.String word)

similarWordsInVocabTo

java.util.List<java.lang.String> similarWordsInVocabTo(java.lang.String word,
                                                       double accuracy)

Find all words with a similar characters in the vocab

Parameters:

word - the word to compare

accuracy - the accuracy: 0 to 1

Returns:

the list of words that are similar in the vocab

getWordVector

double[] getWordVector(java.lang.String word)

Get the word vector for a given matrix

Parameters:

word - the word to get the matrix for

Returns:

the ndarray for this word

getWordVectorMatrixNormalized

org.nd4j.linalg.api.ndarray.INDArray getWordVectorMatrixNormalized(java.lang.String word)

Returns the word vector divided by the norm2 of the array

Parameters:

word - the word to get the matrix for

Returns:

the looked up matrix

getWordVectorMatrix

org.nd4j.linalg.api.ndarray.INDArray getWordVectorMatrix(java.lang.String word)

Get the word vector for a given matrix

Parameters:

word - the word to get the matrix for

Returns:

the ndarray for this word

getWordVectors

org.nd4j.linalg.api.ndarray.INDArray getWordVectors(java.util.Collection<java.lang.String> labels)

This method returns 2D array, where each row represents corresponding word/label

Parameters:

labels -

Returns:

getWordVectorsMean

org.nd4j.linalg.api.ndarray.INDArray getWordVectorsMean(java.util.Collection<java.lang.String> labels)

This method returns mean vector, built from words/labels passed in

Parameters:

labels -

Returns:

wordsNearest

java.util.Collection<java.lang.String> wordsNearest(java.util.Collection<java.lang.String> positive,
                                                    java.util.Collection<java.lang.String> negative,
                                                    int top)

Words nearest based on positive and negative words

Parameters:

positive - the positive words

negative - the negative words

top - the top n words

Returns:

the words nearest the mean of the words

wordsNearest

java.util.Collection<java.lang.String> wordsNearest(java.lang.String word,
                                                    int n)

Get the top n words most similar to the given word

Parameters:

word - the word to compare

n - the n to get

Returns:

the top n words

similarity

double similarity(java.lang.String word,
                  java.lang.String word2)

Returns the similarity of 2 words

Parameters:

word - the first word

word2 - the second word

Returns:

a normalized similarity (cosine similarity)

vocab

VocabCache vocab()

Vocab for the vectors

Returns:

lookupTable

WeightLookupTable lookupTable()

Lookup table for the vectors

Returns:

setModelUtils

void setModelUtils(ModelUtils utils)

Specifies ModelUtils to be used to access model

Parameters:

utils -