org.deeplearning4j.nn.layers.pooling

Class GlobalPoolingLayer

java.lang.Object

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

org.deeplearning4j.nn.layers.pooling.GlobalPoolingLayer

All Implemented Interfaces:

Serializable, Cloneable, Layer, Model, Trainable

public class GlobalPoolingLayer
extends AbstractLayer<GlobalPoolingLayer>

Global pooling layer - used to do pooling over time for RNNs, and 2d pooling for CNNs.
Supports the following PoolingTypes: SUM, AVG, MAX, PNORM
Global pooling layer can also handle mask arrays when dealing with variable length inputs.
mask arrays are assumed to be 2d, and are fed forward through the network during training or post-training forward pass:
- Time series (RNNs, 1d CNNs): mask arrays are shape [miniBatchSize, maxTimeSeriesLength] and contain values 0 or 1 only
- CNNs (2d): mask have shape [miniBatchSize, 1, height, 1] or [miniBatchSize, 1, 1, width] or [minibatch, 1, height, width]. When used activations of shape [minibatch, channels, height, width] the size 1 dimensions are broadcast along the input

Behaviour with default settings:
- 3d (time series) input with shape [miniBatchSize, vectorSize, timeSeriesLength] -> 2d output [miniBatchSize, vectorSize]
- 4d (CNN) input with shape [miniBatchSize, channels, height, width] -> 2d output [miniBatchSize, channels]
- 5d (CNN3D) input with shape [miniBatchSize, channels, depth, height, width] -> 2d output [miniBatchSize, channels]

Alternatively, by setting collapseDimensions = false in the configuration, it is possible to retain the reduced dimensions as 1s: this gives - [miniBatchSize, vectorSize, 1] for RNN output, - [miniBatchSize, channels, 1, 1] for CNN output, and - [miniBatchSize, channels, 1, 1, 1] for CNN3D output.

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.AbstractLayer

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

Constructor Summary

Constructors

Constructor and Description
GlobalPoolingLayer(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
void	clearNoiseWeightParams()
Layer	clone()
Pair<INDArray,MaskState>	feedForwardMaskArray(INDArray maskArray, MaskState currentMaskState, int minibatchSize) Feed forward the input mask array, setting in the layer as appropriate.
boolean	isPretrainLayer() Returns true if the layer can be trained in an unsupervised/pretrain manner (AE, VAE, etc)
Layer.Type	type() Returns the layer type

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

activate, addListeners, allowInputModification, applyConstraints, applyDropOutIfNecessary, applyMask, assertInputSet, backpropDropOutIfPresent, batchSize, calcRegularizationScore, clear, close, computeGradientAndScore, conf, fit, fit, getConfig, getEpochCount, getGradientsViewArray, getHelper, getIndex, getInput, getInputMiniBatchSize, getListeners, getMaskArray, getOptimizer, getParam, gradient, gradientAndScore, init, input, layerConf, layerId, numParams, numParams, params, paramTable, paramTable, score, setBackpropGradientsViewArray, setCacheMode, setConf, setEpochCount, setIndex, setInput, setInputMiniBatchSize, setListeners, setListeners, setMaskArray, setParam, setParams, setParams, setParamsViewArray, setParamTable, update, update, updaterDivideByMinibatch

Methods inherited from class java.lang.Object

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

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

getIterationCount, setIterationCount

Constructor Detail

GlobalPoolingLayer

public GlobalPoolingLayer(NeuralNetConfiguration conf,
                          DataType dataType)

Method Detail

isPretrainLayer

public boolean isPretrainLayer()

Description copied from interface: Layer

Returns true if the layer can be trained in an unsupervised/pretrain manner (AE, VAE, etc)

Returns:

true if the layer can be pretrained (using fit(INDArray), false otherwise

clearNoiseWeightParams

public void clearNoiseWeightParams()

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<GlobalPoolingLayer>

Returns:

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

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

clone

public Layer clone()

Overrides:

clone in class Object

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

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

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<GlobalPoolingLayer>

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.