smile.nlp.embedding

Class Word2Vec

java.lang.Object

smile.nlp.embedding.Word2Vec

public class Word2Vec
extends java.lang.Object

Word2vec is a group of related models that are used to produce word embeddings. These models are shallow, two-layer neural networks that are trained to reconstruct linguistic contexts of words. Word2vec takes as its input a large corpus of text and produces a vector space, typically of several hundred dimensions, with each unique word in the corpus being assigned a corresponding vector in the space. Word vectors are positioned in the vector space such that words that share common contexts in the corpus are located close to one another in the space.

Word2vec can utilize either of two model architectures to produce a distributed representation of words: continuous bag-of-words (CBOW) or continuous skip-gram. In the continuous bag-of-words architecture, the model predicts the current word from a window of surrounding context words. The order of context words does not influence prediction (bag-of-words assumption). In the continuous skip-gram architecture, the model uses the current word to predict the surrounding window of context words. The skip-gram architecture weighs nearby context words more heavily than more distant context words. According to the authors' note, CBOW is faster while skip-gram is slower but does a better job for infrequent words.

Field Summary

Fields

Modifier and Type	Field and Description
smile.data.DataFrame	vectors The vector space.
java.lang.String[]	words The vocabulary.

Constructor Summary

Constructors

Constructor and Description
Word2Vec(java.lang.String[] words, float[][] vectors) Constructor.

Method Summary

Modifier and Type	Method and Description
float[]	apply(java.lang.String word) Returns the vector embedding of a word.
int	dimension() Returns the dimension of vector space.
float[]	get(java.lang.String word) Returns the vector embedding of a word.
static Word2Vec	of(java.nio.file.Path file) Loads a pre-trained word2vec model from binary file of ByteOrder.LITTLE_ENDIAN.
static Word2Vec	of(java.nio.file.Path file, java.nio.ByteOrder order) Loads a pre-trained word2vec model from binary file.

Methods inherited from class java.lang.Object

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

Field Detail

words

public final java.lang.String[] words

The vocabulary.

vectors

public final smile.data.DataFrame vectors

The vector space.

Constructor Detail

Word2Vec

public Word2Vec(java.lang.String[] words,
                float[][] vectors)

Constructor.

Parameters:

words - the vocabulary.

vectors - the vectors of d x n, where d is the dimension and n is the size of vocabulary.

Method Detail

dimension

public int dimension()

Returns the dimension of vector space.

get

public float[] get(java.lang.String word)

Returns the vector embedding of a word.

apply

public float[] apply(java.lang.String word)

Returns the vector embedding of a word. For Scala convenience.

of

public static Word2Vec of(java.nio.file.Path file)
                   throws java.io.IOException

Loads a pre-trained word2vec model from binary file of ByteOrder.LITTLE_ENDIAN.

Throws:

java.io.IOException

of

public static Word2Vec of(java.nio.file.Path file,
                          java.nio.ByteOrder order)
                   throws java.io.IOException

Loads a pre-trained word2vec model from binary file.

Throws:

java.io.IOException