Package ai.djl.modality.nlp

Class Encoder

java.lang.Object

ai.djl.nn.AbstractBaseBlock

ai.djl.nn.AbstractBlock

ai.djl.modality.nlp.Encoder

All Implemented Interfaces:

Block

public abstract class Encoder
extends AbstractBlock

Encoder is an abstract block that be can used as encoder in encoder-decoder architecture. This abstraction, along with Decoder, comes into play in the EncoderDecoder class, and facilitate implementing encoder-decoder models for different tasks and inputs.

Field Summary

Fields

Modifier and Type	Field	Description
protected Block	block

Fields inherited from class ai.djl.nn.AbstractBlock

children, parameters

Fields inherited from class ai.djl.nn.AbstractBaseBlock

inputNames, inputShapes, outputDataTypes, version

Constructor Summary

Constructors

Constructor	Description
Encoder(byte version, Block block)	Constructs a new instance of Encoder with the given block.

Method Summary

Modifier and Type	Method	Description
protected NDList	forwardInternal(ParameterStore parameterStore, NDList inputs, boolean training, ai.djl.util.PairList<String,Object> params)	A helper for Block.forward(ParameterStore, NDList, boolean, PairList) after initialization.
protected NDList	forwardInternal(ParameterStore parameterStore, NDList data, NDList labels, ai.djl.util.PairList<String,Object> params)	A helper for Block.forward(ParameterStore, NDList, NDList, PairList) after initialization.
Shape[]	getOutputShapes(Shape[] inputShapes)	Returns the expected output shapes of the block for the specified input shapes.
abstract NDList	getStates(NDList encoderOutput)	Gets the state of the encoder from the given encoder output.
void	initializeChildBlocks(NDManager manager, DataType dataType, Shape... inputShapes)	Initializes the Child blocks of this block.
void	loadParameters(NDManager manager, DataInputStream is)	Loads the parameters from the given input stream.
void	saveParameters(DataOutputStream os)	Writes the parameters of the block to the given outputStream.

Methods inherited from class ai.djl.nn.AbstractBlock

addChildBlock, addChildBlock, addChildBlockSingleton, addParameter, getChildren, getDirectParameters

Methods inherited from class ai.djl.nn.AbstractBaseBlock

beforeInitialize, cast, clear, describeInput, forward, forward, getInputShapes, getOutputDataTypes, getParameters, initialize, isInitialized, loadMetadata, prepare, readInputShapes, saveInputShapes, saveMetadata, setInitializer, 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, freezeParameters, freezeParameters, getOutputShapes

Field Details

block

protected Block block

Constructor Details

Encoder

public Encoder(byte version,
 Block block)

Constructs a new instance of Encoder with the given block.

Parameters:

version - the version to use for parameter and metadata serialization

block - the encoder block

Method Details

getStates

public abstract NDList getStates(NDList encoderOutput)

Gets the state of the encoder from the given encoder output.

Parameters:

encoderOutput - an NDList that contains the encoder output

Returns:

the state of the encoder

forwardInternal

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

A helper for Block.forward(ParameterStore, NDList, boolean, PairList) after initialization.

Specified by:

forwardInternal in class AbstractBaseBlock

Parameters:

parameterStore - the parameter store

inputs - the input NDList

training - true for a training forward pass

params - optional parameters

Returns:

the output of the forward pass

forwardInternal

protected NDList forwardInternal(ParameterStore parameterStore,
 NDList data,
 NDList labels,
 ai.djl.util.PairList<String,Object> params)

Description copied from class: AbstractBaseBlock

A helper for Block.forward(ParameterStore, NDList, NDList, PairList) after initialization.

Overrides:

forwardInternal in class AbstractBaseBlock

Parameters:

parameterStore - the parameter store

data - the input data NDList

labels - the input labels NDList

params - optional parameters

Returns:

the output of the forward pass

See Also:

• AbstractBaseBlock.forward(ParameterStore, NDList, boolean, PairList)

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 AbstractBaseBlock

Parameters:

manager - the manager to use for initialization

dataType - the requested data type

inputShapes - the expected input shapes for this block

getOutputShapes

public Shape[] getOutputShapes(Shape[] inputShapes)

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

Parameters:

inputShapes - the shapes of the inputs

Returns:

the expected output shapes of the block

saveParameters

public void saveParameters(DataOutputStream os)
                    throws IOException

Writes the parameters of the block to the given outputStream.

Specified by:

saveParameters in interface Block

Overrides:

saveParameters in class AbstractBaseBlock

Parameters:

os - the outputstream to save the parameters to

Throws:

IOException - if an I/O error occurs

loadParameters

public void loadParameters(NDManager manager,
 DataInputStream is)
                    throws IOException,
MalformedModelException

Loads the parameters from the given input stream.

Specified by:

loadParameters in interface Block

Overrides:

loadParameters in class AbstractBaseBlock

Parameters:

manager - an NDManager to create the parameter arrays

is - the inputstream that stream the parameter values

Throws:

IOException - if an I/O error occurs

MalformedModelException - if the model file is corrupted or unsupported