Modifier and Type | Method and Description |
---|---|
CnnSentenceDataSetIterator.Builder |
CnnSentenceDataSetIterator.Builder.wordVectors(WordVectors wordVectors)
Provide the WordVectors instance that should be used for training
|
Modifier and Type | Method and Description |
---|---|
static WordVectors |
WordVectorSerializer.fromTableAndVocab(WeightLookupTable table,
VocabCache vocab)
Load word vectors for the given vocab and table
|
static WordVectors |
WordVectorSerializer.loadStaticModel(@NonNull File file)
This method restores previously saved w2v model.
|
static WordVectors |
WordVectorSerializer.loadStaticModel(InputStream inputStream)
This method restores previously saved w2v model.
|
static WordVectors |
WordVectorSerializer.loadTxtVectors(File vectorsFile)
Deprecated.
|
static WordVectors |
WordVectorSerializer.loadTxtVectors(@NonNull InputStream stream,
boolean skipFirstLine)
Deprecated.
Use readWord2VecModel() or loadStaticModel() method instead
|
Modifier and Type | Class and Description |
---|---|
class |
WordVectorsImpl<T extends SequenceElement>
Common word vector operations
|
Modifier and Type | Class and Description |
---|---|
class |
FastText |
Modifier and Type | Class and Description |
---|---|
class |
Node2Vec<V extends SequenceElement,E extends Number>
Deprecated.
|
Modifier and Type | Method and Description |
---|---|
protected Node2Vec.Builder<V,E> |
Node2Vec.Builder.useExistingWordVectors(@NonNull WordVectors vec) |
Modifier and Type | Class and Description |
---|---|
class |
ParagraphVectors
Basic ParagraphVectors (aka Doc2Vec) implementation for DL4j, as wrapper over SequenceVectors
|
Modifier and Type | Method and Description |
---|---|
ParagraphVectors.Builder |
ParagraphVectors.Builder.useExistingWordVectors(@NonNull WordVectors vec)
This method allows you to use pre-built WordVectors model (e.g.
|
Modifier and Type | Class and Description |
---|---|
class |
SequenceVectors<T extends SequenceElement>
SequenceVectors implements abstract features extraction for Sequences and SequenceElements, using SkipGram, CBOW or DBOW (for Sequence features extraction).
|
Modifier and Type | Field and Description |
---|---|
protected WordVectors |
SequenceVectors.existingModel |
protected WordVectors |
SequenceVectors.Builder.existingVectors |
protected WordVectors |
SequenceVectors.intersectModel |
Modifier and Type | Method and Description |
---|---|
protected SequenceVectors.Builder<T> |
SequenceVectors.Builder.useExistingWordVectors(@NonNull WordVectors vec)
This method allows you to use pre-built WordVectors model (e.g.
|
Modifier and Type | Class and Description |
---|---|
class |
StaticWord2Vec
This is special limited Word2Vec implementation, suited for serving as lookup table in concurrent multi-gpu environment
This implementation DOES NOT load all vectors onto any of gpus, instead of that it holds vectors in, optionally, compressed state in host memory.
|
Modifier and Type | Method and Description |
---|---|
protected Word2Vec.Builder |
Word2Vec.Builder.useExistingWordVectors(@NonNull WordVectors vec)
This method has no effect for Word2Vec
|
Modifier and Type | Method and Description |
---|---|
VocabCache<T> |
VocabConstructor.buildMergedVocabulary(@NonNull WordVectors wordVectors,
boolean fetchLabels)
This method transfers existing WordVectors model into current one
|
Modifier and Type | Class and Description |
---|---|
class |
Word2Vec
Spark version of word2vec
|
Modifier and Type | Class and Description |
---|---|
class |
SparkParagraphVectors |
Modifier and Type | Class and Description |
---|---|
class |
SparkSequenceVectors<T extends SequenceElement>
Generic SequenceVectors implementation for dl4j-spark-nlp
|
Modifier and Type | Class and Description |
---|---|
class |
SparkWord2Vec |
Modifier and Type | Method and Description |
---|---|
static List<Window> |
Windows.windows(String words,
@NonNull TokenizerFactory tokenizerFactory,
int windowSize,
WordVectors vectors)
Constructs a list of window of size windowSize.
|
Copyright © 2020. All rights reserved.