Package ai.djl.nn.convolutional

Class Deconvolution

java.lang.Object

ai.djl.nn.AbstractBaseBlock

ai.djl.nn.AbstractBlock

ai.djl.nn.convolutional.Deconvolution

All Implemented Interfaces:

Block

Direct Known Subclasses:

Conv1dTranspose, Conv2dTranspose

public abstract class Deconvolution
extends AbstractBlock

Transposed convolution, also named fractionally-strided convolution Dumoulin & Visin or deconvolution Long et al., 2015, serves this purpose.

The need for transposed convolutions generally arises from the desire to use a transformation going in the opposite direction of a normal convolution, i.e., from something that has the shape of the output of some convolution to something that has the shape of its input while maintaining a connectivity pattern that is compatible with said convolution.

Current implementations of Deconvolution are Conv1dTranspose with input dimension of LayoutType.WIDTH and Conv2dTranspose with input dimension of LayoutType.WIDTH and LayoutType.HEIGHT. These implementations share the same core principal as a Deconvolution layer does, with the difference being the number of input dimension each operates on as denoted by ConvXdTranspose for X dimension(s).

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	Deconvolution.DeconvolutionBuilder<T extends Deconvolution.DeconvolutionBuilder>	A builder that can build any Deconvolution block.

Field Summary

Fields

Modifier and Type	Field	Description
protected Parameter	bias
protected Shape	dilation
protected int	filters
protected int	groups
protected boolean	includeBias
protected Shape	kernelShape
protected Shape	outPadding
protected Shape	padding
protected Shape	stride
protected Parameter	weight

Fields inherited from class ai.djl.nn.AbstractBlock

children, parameters

Fields inherited from class ai.djl.nn.AbstractBaseBlock

inputNames, inputShapes, version

Constructor Summary

Constructors

Constructor	Description
Deconvolution(Deconvolution.DeconvolutionBuilder<?> builder)	Creates a Deconvolution object.

Method Summary

Modifier and Type	Method	Description
protected void	beforeInitialize(Shape... inputShapes)	Performs any action necessary before initialization.
protected NDList	forwardInternal(ParameterStore parameterStore, NDList inputs, boolean training, ai.djl.util.PairList<java.lang.String,java.lang.Object> params)	A helper for Block.forward(ParameterStore, NDList, boolean, PairList) after initialization.
protected abstract LayoutType[]	getExpectedLayout()	Returns the expected layout of the input.
Shape[]	getOutputShapes(Shape[] inputs)	Returns the expected output shapes of the block for the specified input shapes.
protected abstract java.lang.String	getStringLayout()	Returns the string representing the layout of the input.
void	loadMetadata(byte loadVersion, java.io.DataInputStream is)	Overwrite this to load additional metadata with the parameter values.
protected abstract int	numDimensions()	Returns the number of dimensions of the input.
protected void	prepare(Shape[] inputs)	Sets the shape of Parameters.

Methods inherited from class ai.djl.nn.AbstractBlock

addChildBlock, addChildBlock, addChildBlockSingleton, addParameter, getChildren, getDirectParameters

Methods inherited from class ai.djl.nn.AbstractBaseBlock

cast, clear, describeInput, forward, forward, forwardInternal, getInputShapes, getParameters, initialize, initializeChildBlocks, isInitialized, loadParameters, readInputShapes, saveInputShapes, saveMetadata, saveParameters, 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

Field Detail

kernelShape

protected Shape kernelShape

stride

protected Shape stride

padding

protected Shape padding

outPadding

protected Shape outPadding

dilation

protected Shape dilation

filters

protected int filters

groups

protected int groups

includeBias

protected boolean includeBias

weight

protected Parameter weight

bias

protected Parameter bias

Constructor Detail

Deconvolution

public Deconvolution(Deconvolution.DeconvolutionBuilder<?> builder)

Creates a Deconvolution object.

Parameters:

builder - the Builder that has the necessary configurations

Method Detail

getExpectedLayout

protected abstract LayoutType[] getExpectedLayout()

Returns the expected layout of the input.

Returns:

the expected layout of the input

getStringLayout

protected abstract java.lang.String getStringLayout()

Returns the string representing the layout of the input.

Returns:

the string representing the layout of the input

numDimensions

protected abstract int numDimensions()

Returns the number of dimensions of the input.

Returns:

the number of dimensions of the input

forwardInternal

protected NDList forwardInternal(ParameterStore parameterStore,
                                 NDList inputs,
                                 boolean training,
                                 ai.djl.util.PairList<java.lang.String,java.lang.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

beforeInitialize

protected void beforeInitialize(Shape... inputShapes)

Performs any action necessary before initialization. For example, keep the input information or verify the layout.

Overrides:

beforeInitialize in class AbstractBaseBlock

Parameters:

inputShapes - the expected shapes of the input

prepare

protected void prepare(Shape[] inputs)

Sets the shape of Parameters.

Overrides:

prepare in class AbstractBaseBlock

Parameters:

inputs - the shapes of inputs

getOutputShapes

public Shape[] getOutputShapes(Shape[] inputs)

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

Parameters:

inputs - the shapes of the inputs

Returns:

the expected output shapes of the block

loadMetadata

public void loadMetadata(byte loadVersion,
                         java.io.DataInputStream is)
                  throws java.io.IOException,
                         MalformedModelException

Overwrite this to load additional metadata with the parameter values.

If you overwrite AbstractBaseBlock.saveMetadata(DataOutputStream) or need to provide backward compatibility to older binary formats, you prabably need to overwrite this. This default implementation checks if the version number fits, if not it throws an MalformedModelException. After that it restores the input shapes.

Overrides:

loadMetadata in class AbstractBaseBlock

Parameters:

loadVersion - the version used for loading this metadata.

is - the input stream we are loading from

Throws:

java.io.IOException - loading failed

MalformedModelException - data can be loaded but has wrong format