org.deeplearning4j.nn.layers.recurrent

Class TimeDistributedLayer

java.lang.Object

org.deeplearning4j.nn.layers.wrapper.BaseWrapperLayer

org.deeplearning4j.nn.layers.recurrent.TimeDistributedLayer

All Implemented Interfaces:

Serializable, Cloneable, Layer, Model, Trainable

public class TimeDistributedLayer
extends BaseWrapperLayer

TimeDistributed wrapper layer.
Note: only the "Feed forward layer time distributed in an RNN" is currently supported. For example, a time distributed dense layer.
Usage: .layer(new TimeDistributed(new DenseLayer.Builder()....build(), timeAxis))
Note that for DL4J RNNs, time axis is always 2 - i.e., RNN activations have shape [minibatch, size, sequenceLength]

Author:

Alex Black

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from interface org.deeplearning4j.nn.api.Layer

Layer.TrainingMode, Layer.Type

Field Summary

Fields inherited from class org.deeplearning4j.nn.layers.wrapper.BaseWrapperLayer

underlying

Constructor Summary

Constructors

Constructor and Description
TimeDistributedLayer(Layer underlying, RNNFormat rnnDataFormat)

Method Summary

Modifier and Type	Method and Description
INDArray	activate(boolean training, LayerWorkspaceMgr workspaceMgr) Perform forward pass and return the activations array with the last set input
INDArray	activate(INDArray input, boolean training, LayerWorkspaceMgr workspaceMgr) Perform forward pass and return the activations array with the specified input
Pair<Gradient,INDArray>	backpropGradient(INDArray epsilon, LayerWorkspaceMgr workspaceMgr) Calculate the gradient relative to the error in the next layer
Pair<INDArray,MaskState>	feedForwardMaskArray(INDArray maskArray, MaskState currentMaskState, int minibatchSize) Feed forward the input mask array, setting in the layer as appropriate.
protected int[]	permuteAxes(int rank, int timeAxis)
protected INDArray	reshape(INDArray array)
protected INDArray	revertReshape(INDArray toRevert, long minibatch)
void	setMaskArray(INDArray maskArray) Set the mask array.

Methods inherited from class org.deeplearning4j.nn.layers.wrapper.BaseWrapperLayer

addListeners, allowInputModification, applyConstraints, batchSize, calcRegularizationScore, clear, clearNoiseWeightParams, close, computeGradientAndScore, conf, fit, fit, getConfig, getEpochCount, getGradientsViewArray, getHelper, getIndex, getInputMiniBatchSize, getIterationCount, getListeners, getMaskArray, getOptimizer, getParam, gradient, gradientAndScore, init, input, isPretrainLayer, numParams, numParams, params, paramTable, paramTable, score, setBackpropGradientsViewArray, setCacheMode, setConf, setEpochCount, setIndex, setInput, setInputMiniBatchSize, setIterationCount, setListeners, setListeners, setParam, setParams, setParamsViewArray, setParamTable, type, update, update, updaterDivideByMinibatch

Methods inherited from class java.lang.Object

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

Constructor Detail

TimeDistributedLayer

public TimeDistributedLayer(Layer underlying,
                            RNNFormat rnnDataFormat)

Method Detail

backpropGradient

public Pair<Gradient,INDArray> backpropGradient(INDArray epsilon,
                                                LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Layer

Calculate the gradient relative to the error in the next layer

Specified by:

backpropGradient in interface Layer

Overrides:

backpropGradient in class BaseWrapperLayer

Parameters:

epsilon - w^(L+1)*delta^(L+1). Or, equiv: dC/da, i.e., (dC/dz)*(dz/da) = dC/da, where C is cost function a=sigma(z) is activation.

workspaceMgr - Workspace manager

Returns:

Pair where Gradient is gradient for this layer, INDArray is epsilon (activation gradient) needed by next layer, but before element-wise multiply by sigmaPrime(z). So for standard feed-forward layer, if this layer is L, then return.getSecond() == dL/dIn = (w^(L)*(delta^(L))^T)^T. Note that the returned array should be placed in the ArrayType.ACTIVATION_GRAD workspace via the workspace manager

activate

public INDArray activate(boolean training,
                         LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Layer

Perform forward pass and return the activations array with the last set input

Specified by:

activate in interface Layer

Overrides:

activate in class BaseWrapperLayer

Parameters:

training - training or test mode

workspaceMgr - Workspace manager

Returns:

the activation (layer output) of the last specified input. Note that the returned array should be placed in the ArrayType.ACTIVATIONS workspace via the workspace manager

activate

public INDArray activate(INDArray input,
                         boolean training,
                         LayerWorkspaceMgr workspaceMgr)

Description copied from interface: Layer

Perform forward pass and return the activations array with the specified input

Specified by:

activate in interface Layer

Overrides:

activate in class BaseWrapperLayer

Parameters:

input - the input to use

training - train or test mode

workspaceMgr - Workspace manager.

Returns:

Activations array. Note that the returned array should be placed in the ArrayType.ACTIVATIONS workspace via the workspace manager

reshape

protected INDArray reshape(INDArray array)

permuteAxes

protected int[] permuteAxes(int rank,
                            int timeAxis)

revertReshape

protected INDArray revertReshape(INDArray toRevert,
                                 long minibatch)

setMaskArray

public void setMaskArray(INDArray maskArray)

Description copied from interface: Layer

Set the mask array. Note: In general, Layer.feedForwardMaskArray(INDArray, MaskState, int) should be used in preference to this.

Specified by:

setMaskArray in interface Layer

Overrides:

setMaskArray in class BaseWrapperLayer

Parameters:

maskArray - Mask array to set

feedForwardMaskArray

public Pair<INDArray,MaskState> feedForwardMaskArray(INDArray maskArray,
                                                     MaskState currentMaskState,
                                                     int minibatchSize)

Description copied from interface: Layer

Feed forward the input mask array, setting in the layer as appropriate. This allows different layers to handle masks differently - for example, bidirectional RNNs and normal RNNs operate differently with masks (the former sets activations to 0 outside of the data present region (and keeps the mask active for future layers like dense layers), whereas normal RNNs don't zero out the activations/errors )instead relying on backpropagated error arrays to handle the variable length case.
This is also used for example for networks that contain global pooling layers, arbitrary preprocessors, etc.

Specified by:

feedForwardMaskArray in interface Layer

Overrides:

feedForwardMaskArray in class BaseWrapperLayer

Parameters:

maskArray - Mask array to set

currentMaskState - Current state of the mask - see MaskState

minibatchSize - Current minibatch size. Needs to be known as it cannot always be inferred from the activations array due to reshaping (such as a DenseLayer within a recurrent neural network)

Returns:

New mask array after this layer, along with the new mask state.