org.deeplearning4j.nn.layers.recurrent

Class RnnOutputLayer

java.lang.Object

org.deeplearning4j.nn.layers.AbstractLayer<LayerConfT>

org.deeplearning4j.nn.layers.BaseLayer<LayerConfT>

org.deeplearning4j.nn.layers.BaseOutputLayer<RnnOutputLayer>

org.deeplearning4j.nn.layers.recurrent.RnnOutputLayer

All Implemented Interfaces:

Serializable, Cloneable, Classifier, Layer, IOutputLayer, Model, Trainable

public class RnnOutputLayer
extends BaseOutputLayer<RnnOutputLayer>

Recurrent Neural Network Output Layer.
Handles calculation of gradients etc for various objective functions.
Functionally the same as OutputLayer, but handles output and label reshaping automatically.
Input and output activations are same as other RNN layers: 3 dimensions with shape [miniBatchSize,nIn,timeSeriesLength] and [miniBatchSize,nOut,timeSeriesLength] respectively.

Author:

Alex Black

See Also:

BaseOutputLayer, OutputLayer, 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.BaseOutputLayer

inputMaskArray, inputMaskArrayState, labels

Fields inherited from class org.deeplearning4j.nn.layers.BaseLayer

gradient, gradientsFlattened, gradientViews, optimizer, params, paramsFlattened, score, weightNoiseParams

Fields inherited from class org.deeplearning4j.nn.layers.AbstractLayer

cacheMode, conf, dataType, dropoutApplied, epochCount, index, input, inputModificationAllowed, iterationCount, maskArray, maskState, preOutput, trainingListeners

Constructor Summary

Constructors

Constructor and Description
RnnOutputLayer(NeuralNetConfiguration conf, DataType dataType)

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
Pair<Gradient,INDArray>	backpropGradient(INDArray epsilon, LayerWorkspaceMgr workspaceMgr) Calculate the gradient relative to the error in the next layer
INDArray	computeScoreForExamples(double fullNetRegTerm, LayerWorkspaceMgr workspaceMgr) Compute the score for each example individually, after labels and input have been set.
double	f1Score(INDArray examples, INDArray labels) Returns the f1 score for the given examples.
Pair<INDArray,MaskState>	feedForwardMaskArray(INDArray maskArray, MaskState currentMaskState, int minibatchSize) Feed forward the input mask array, setting in the layer as appropriate.
INDArray	getInput()
protected INDArray	getLabels2d(LayerWorkspaceMgr workspaceMgr, ArrayType arrayType)
protected INDArray	preOutput2d(boolean training, LayerWorkspaceMgr workspaceMgr)
void	setMaskArray(INDArray maskArray) Set the mask array.
Layer.Type	type() Returns the layer type

Methods inherited from class org.deeplearning4j.nn.layers.BaseOutputLayer

activate, applyMask, clear, computeGradientAndScore, computeScore, f1Score, fit, fit, fit, fit, fit, getLabels, gradient, gradientAndScore, hasBias, isPretrainLayer, needsLabels, numLabels, predict, predict, setLabels, setScoreWithZ

Methods inherited from class org.deeplearning4j.nn.layers.BaseLayer

calcRegularizationScore, clearNoiseWeightParams, clone, fit, getGradientsViewArray, getOptimizer, getParam, getParamWithNoise, hasLayerNorm, layerConf, numParams, params, paramTable, paramTable, preOutput, preOutputWithPreNorm, score, setBackpropGradientsViewArray, setParam, setParams, setParams, setParamsViewArray, setParamTable, toString, update, update

Methods inherited from class org.deeplearning4j.nn.layers.AbstractLayer

addListeners, allowInputModification, applyConstraints, applyDropOutIfNecessary, assertInputSet, backpropDropOutIfPresent, batchSize, close, conf, getConfig, getEpochCount, getHelper, getIndex, getInputMiniBatchSize, getListeners, getMaskArray, init, input, layerId, numParams, setCacheMode, setConf, setEpochCount, setIndex, setInput, setInputMiniBatchSize, setListeners, setListeners, updaterDivideByMinibatch

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.deeplearning4j.nn.api.Layer

allowInputModification, calcRegularizationScore, clearNoiseWeightParams, getEpochCount, getHelper, getIndex, getInputMiniBatchSize, getIterationCount, getListeners, getMaskArray, setCacheMode, setEpochCount, setIndex, setInput, setInputMiniBatchSize, setIterationCount, setListeners, setListeners

Methods inherited from interface org.deeplearning4j.nn.api.Trainable

getConfig, getGradientsViewArray, numParams, params, paramTable, updaterDivideByMinibatch

Methods inherited from interface org.deeplearning4j.nn.api.Model

addListeners, applyConstraints, batchSize, close, conf, fit, getGradientsViewArray, getOptimizer, getParam, init, input, numParams, numParams, params, paramTable, paramTable, score, setBackpropGradientsViewArray, setConf, setParam, setParams, setParamsViewArray, setParamTable, update, update

Constructor Detail

RnnOutputLayer

public RnnOutputLayer(NeuralNetConfiguration conf,
                      DataType dataType)

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 BaseOutputLayer<RnnOutputLayer>

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

f1Score

public double f1Score(INDArray examples,
                      INDArray labels)

Returns the f1 score for the given examples.

Specified by:

f1Score in interface Classifier

Overrides:

f1Score in class BaseOutputLayer<RnnOutputLayer>

Parameters:

examples - te the examples to classify (one example in each row)

labels - the true labels

Returns:

the scores for each ndarray

getInput

public INDArray getInput()

Overrides:

getInput in class AbstractLayer<RnnOutputLayer>

type

public Layer.Type type()

Description copied from interface: Layer

Returns the layer type

Specified by:

type in interface Layer

Overrides:

type in class AbstractLayer<RnnOutputLayer>

Returns:

preOutput2d

protected INDArray preOutput2d(boolean training,
                               LayerWorkspaceMgr workspaceMgr)

Overrides:

preOutput2d in class BaseOutputLayer<RnnOutputLayer>

getLabels2d

protected INDArray getLabels2d(LayerWorkspaceMgr workspaceMgr,
                               ArrayType arrayType)

Specified by:

getLabels2d in class BaseOutputLayer<RnnOutputLayer>

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 BaseLayer<RnnOutputLayer>

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

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 AbstractLayer<RnnOutputLayer>

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 AbstractLayer<RnnOutputLayer>

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.

computeScoreForExamples

public INDArray computeScoreForExamples(double fullNetRegTerm,
                                        LayerWorkspaceMgr workspaceMgr)

Compute the score for each example individually, after labels and input have been set.

Specified by:

computeScoreForExamples in interface IOutputLayer

Overrides:

computeScoreForExamples in class BaseOutputLayer<RnnOutputLayer>

Parameters:

fullNetRegTerm - Regularization score term for the entire network (or, 0.0 to not include regularization)

Returns:

A column INDArray of shape [numExamples,1], where entry i is the score of the ith example