ai.djl.modality.nlp

Class Decoder

java.lang.Object

ai.djl.nn.AbstractBlock

ai.djl.modality.nlp.Decoder

All Implemented Interfaces:

Block

public abstract class Decoder
extends AbstractBlock

Decoder is an abstract block that be can used as decoder in encoder-decoder architecture. This abstraction, along with Encoder, 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 and Description
protected Block	block

Fields inherited from class ai.djl.nn.AbstractBlock

children, inputNames, inputShapes, parameters, parameterShapeCallbacks, version

Constructor Summary

Constructors

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

Method Summary

Modifier and Type	Method and Description
NDList	forward(ParameterStore parameterStore, NDList inputs, boolean training, ai.djl.util.PairList<java.lang.String,java.lang.Object> params) Applies the operating function of the block once.
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.
abstract void	initState(NDList encoderStates) Sets the state of the encoder as the initial state of the decoder.
void	loadParameters(NDManager manager, java.io.DataInputStream is) Loads the parameters from the given input stream.
void	saveParameters(java.io.DataOutputStream os) Writes the parameters of the block to the given outputStream.

Methods inherited from class ai.djl.nn.AbstractBlock

addChildBlock, addParameter, addParameter, addParameter, beforeInitialize, cast, clear, describeInput, getChildren, getDirectParameters, getParameters, getParameterShape, initialize, isInitialized, loadMetadata, readInputShapes, saveInputShapes, saveMetadata, 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, forward, validateLayout

Field Detail

block

protected Block block

Constructor Detail

Decoder

public Decoder(byte version,
               Block block)

Constructs a new instance of Decoder with the given block. Use this constructor if you are planning to use pre-trained embeddings that don't need further training.

Parameters:

block - the block to be used to decode

version - the version to use for parameter and metadata serialization

Method Detail

initState

public abstract void initState(NDList encoderStates)

Sets the state of the encoder as the initial state of the decoder.

Parameters:

encoderStates - the states of the encoder

forward

public NDList forward(ParameterStore parameterStore,
                      NDList inputs,
                      boolean training,
                      ai.djl.util.PairList<java.lang.String,java.lang.Object> params)

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

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

initializeChildBlocks

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

Description copied from class: AbstractBlock

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

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

saveParameters

public void saveParameters(java.io.DataOutputStream os)
                    throws java.io.IOException

Writes the parameters of the block to the given outputStream.

Specified by:

saveParameters in interface Block

Overrides:

saveParameters in class AbstractBlock

Parameters:

os - the outputstream to save the parameters to

Throws:

java.io.IOException - if an I/O error occurs

loadParameters

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

Loads the parameters from the given input stream.

Specified by:

loadParameters in interface Block

Overrides:

loadParameters in class AbstractBlock

Parameters:

manager - an NDManager to create the parameter arrays

is - the inputstream that stream the parameter values

Throws:

java.io.IOException - if an I/O error occurs

MalformedModelException - if the model file is corrupted or unsupported