org.deeplearning4j.models.embeddings

Interface WeightLookupTable<T extends SequenceElement>

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

GloveWeightLookupTable, InMemoryLookupTable

public interface WeightLookupTable<T extends SequenceElement>
extends java.io.Serializable

General weight lookup table

Method Summary

Modifier and Type	Method and Description
double	getGradient(int column, double gradient) Returns gradient for specified word
java.lang.Long	getTableId() Returns unique ID of this table.
VocabCache<T>	getVocabCache() Returns corresponding vocabulary
org.nd4j.linalg.api.ndarray.INDArray	getWeights()
void	iterate(T w1, T w2) Deprecated.
void	iterateSample(T w1, T w2, java.util.concurrent.atomic.AtomicLong nextRandom, double alpha) Deprecated.
int	layerSize() The layer size for the lookup table
org.nd4j.linalg.api.ndarray.INDArray	loadCodes(int[] codes) Loads the co-occurrences for the given codes
void	plotVocab(BarnesHutTsne tsne, int numWords, java.io.File file) Render the words via TSNE
void	plotVocab(BarnesHutTsne tsne, int numWords, UiConnectionInfo connectionInfo) Render the words via TSNE
void	plotVocab(int numWords, java.io.File file) Render the words via tsne
void	plotVocab(int numWords, UiConnectionInfo connectionInfo) Render the words via tsne
void	putCode(int codeIndex, org.nd4j.linalg.api.ndarray.INDArray code)
void	putVector(java.lang.String word, org.nd4j.linalg.api.ndarray.INDArray vector) Inserts a word vector
void	resetWeights() Reset the weights of the cache
void	resetWeights(boolean reset) Clear out all weights regardless
void	setLearningRate(double lr) Sets the learning rate
void	setTableId(java.lang.Long tableId) Set's table Id.
org.nd4j.linalg.api.ndarray.INDArray	vector(java.lang.String word)
java.util.Iterator<org.nd4j.linalg.api.ndarray.INDArray>	vectors() Iterates through all of the vectors in the cache

Method Detail

getTableId

java.lang.Long getTableId()

Returns unique ID of this table. Used for JointStorage/DistributedLookupTable mechanics

Returns:

ID of this table

setTableId

void setTableId(java.lang.Long tableId)

Set's table Id. Please note, it should be unique withing Joint/Distributed LookupTable

Parameters:

tableId -

layerSize

int layerSize()

The layer size for the lookup table

Returns:

the layer size for the lookup table

getGradient

double getGradient(int column,
                   double gradient)

Returns gradient for specified word

Parameters:

column -

gradient -

Returns:

resetWeights

void resetWeights(boolean reset)

Clear out all weights regardless

Parameters:

reset -

plotVocab

void plotVocab(BarnesHutTsne tsne,
               int numWords,
               UiConnectionInfo connectionInfo)

Render the words via TSNE

Parameters:

tsne - the tsne to use

plotVocab

void plotVocab(BarnesHutTsne tsne,
               int numWords,
               java.io.File file)

Render the words via TSNE

Parameters:

tsne - the tsne to use

plotVocab

void plotVocab(int numWords,
               UiConnectionInfo connectionInfo)

Render the words via tsne

plotVocab

void plotVocab(int numWords,
               java.io.File file)

Render the words via tsne

putCode

void putCode(int codeIndex,
             org.nd4j.linalg.api.ndarray.INDArray code)

Parameters:

codeIndex -

code -

loadCodes

org.nd4j.linalg.api.ndarray.INDArray loadCodes(int[] codes)

Loads the co-occurrences for the given codes

Parameters:

codes - the codes to load

Returns:

an ndarray of code.length by layerSize

iterate

@Deprecated
void iterate(T w1,
                         T w2)

Deprecated.

Iterate on the given 2 vocab words

Parameters:

w1 - the first word to iterate on

w2 - the second word to iterate on

iterateSample

@Deprecated
void iterateSample(T w1,
                               T w2,
                               java.util.concurrent.atomic.AtomicLong nextRandom,
                               double alpha)

Deprecated.

Iterate on the given 2 vocab words

Parameters:

w1 - the first word to iterate on

w2 - the second word to iterate on

nextRandom - nextRandom for sampling

alpha - the alpha to use for learning

putVector

void putVector(java.lang.String word,
               org.nd4j.linalg.api.ndarray.INDArray vector)

Inserts a word vector

Parameters:

word - the word to insert

vector - the vector to insert

vector

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

Parameters:

word -

Returns:

resetWeights

void resetWeights()

Reset the weights of the cache

setLearningRate

void setLearningRate(double lr)

Sets the learning rate

Parameters:

lr -

vectors

java.util.Iterator<org.nd4j.linalg.api.ndarray.INDArray> vectors()

Iterates through all of the vectors in the cache

Returns:

an iterator for all vectors in the cache

getWeights

org.nd4j.linalg.api.ndarray.INDArray getWeights()

getVocabCache

VocabCache<T> getVocabCache()

Returns corresponding vocabulary