ai.djl.nn.transformer

Class TransformerEncoderBlock

java.lang.Object

ai.djl.nn.AbstractBlock

ai.djl.nn.transformer.TransformerEncoderBlock

All Implemented Interfaces:

Block

public class TransformerEncoderBlock
extends AbstractBlock

Self-Attention based transformer encoder block.

Field Summary

Fields inherited from class ai.djl.nn.AbstractBlock

children, inputNames, inputShapes, parameters, parameterShapeCallbacks, version

Constructor Summary

Constructors

Constructor and Description
TransformerEncoderBlock(int embeddingSize, int headCount, int hiddenSize, float dropoutProbability, java.util.function.Function<NDList,NDList> activationFunction) Creates a transformer encoder block.

Method Summary

Modifier and Type	Method and Description
NDList	forward(ParameterStore ps, NDList inputs, boolean training) Applies the operating function of the block once.
protected NDList	forwardInternal(ParameterStore ps, NDList inputs, boolean training, ai.djl.util.PairList<java.lang.String,java.lang.Object> params)
Shape[]	getOutputShapes(NDManager manager, Shape[] inputShapes) Returns the expected output shapes of the block for the specified input shapes.
void	initializeChildBlocks(NDManager manager, DataType dataType, Shape... inputShapes) Initializes the Child blocks of this block.

Methods inherited from class ai.djl.nn.AbstractBlock

addChildBlock, addParameter, addParameter, addParameter, beforeInitialize, cast, clear, describeInput, forward, getChildren, getDirectParameters, getParameters, getParameterShape, initialize, isInitialized, loadMetadata, loadParameters, readInputShapes, saveInputShapes, saveMetadata, saveParameters, setInitializer, setInitializer, toString

Methods inherited from class java.lang.Object

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

Methods inherited from interface ai.djl.nn.Block

forward, validateLayout

Constructor Detail

TransformerEncoderBlock

public TransformerEncoderBlock(int embeddingSize,
                               int headCount,
                               int hiddenSize,
                               float dropoutProbability,
                               java.util.function.Function<NDList,NDList> activationFunction)

Creates a transformer encoder block.

Parameters:

embeddingSize - the embedding size for tokens

headCount - number of attention blocks

hiddenSize - the hidden size for fully connected networks

dropoutProbability - dropout probability

activationFunction - activation function

Method Detail

getOutputShapes

public Shape[] getOutputShapes(NDManager manager,
                               Shape[] inputShapes)

Returns the expected output shapes of the block for the specified input shapes.

Parameters:

manager - an NDManager

inputShapes - the shapes of the inputs

Returns:

the expected output shapes of the block

initializeChildBlocks

public void initializeChildBlocks(NDManager manager,
                                  DataType dataType,
                                  Shape... inputShapes)

Initializes the Child blocks of this block. You need to override this method if your subclass has child blocks. Used to determine the correct input shapes for child blocks based on the requested input shape for this block.

Overrides:

initializeChildBlocks in class AbstractBlock

Parameters:

manager - the manager to use for initialization

dataType - the requested data type

inputShapes - the expected input shapes for this block

forwardInternal

protected NDList forwardInternal(ParameterStore ps,
                                 NDList inputs,
                                 boolean training,
                                 ai.djl.util.PairList<java.lang.String,java.lang.Object> params)

Specified by:

forwardInternal in class AbstractBlock

forward

public NDList forward(ParameterStore ps,
                      NDList inputs,
                      boolean training)

Applies the operating function of the block once. This method should be called only on blocks that are initialized.

Parameters:

ps - the parameter store

inputs - the input NDList

training - true for a training forward pass

Returns:

the output of the forward pass