org.deeplearning4j.nn.api

Interface Layer

All Superinterfaces:

Cloneable, Model, Serializable

All Known Implementing Classes:

AutoEncoder, BaseLayer, BasePretrainNetwork, ConvolutionDownSampleLayer, ConvolutionLayer, LSTM, OutputLayer, RBM, RecursiveAutoEncoder, SubsamplingLayer

public interface Layer
extends Serializable, Cloneable, Model

Interface for a layer of a neural network. This has an activation function, an input and output size, weights, and a bias

Author:

Adam Gibson

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static class	Layer.Type

Method Summary

Modifier and Type	Method and Description
org.nd4j.linalg.api.ndarray.INDArray	activate() Trigger an activation with the last specified input
org.nd4j.linalg.api.ndarray.INDArray	activate(org.nd4j.linalg.api.ndarray.INDArray input) Initialize the layer with the given input and return the activation for this layer given this input
org.nd4j.linalg.api.ndarray.INDArray	activationMean() Calculate the mean representation for the activation for this layer
Pair<Gradient,Gradient>	backWard(Gradient errors, Gradient deltas, org.nd4j.linalg.api.ndarray.INDArray activation, String previousActivation) Propagate errors backwards for a particular layer.
Gradient	backwardGradient(org.nd4j.linalg.api.ndarray.INDArray activation, Gradient errorSignal) Calculate the gradient relative to the error in the next layer
Gradient	calcGradient(Gradient layerError, org.nd4j.linalg.api.ndarray.INDArray indArray) Calculate the gradient
Layer	clone() Clone the layer
org.nd4j.linalg.api.ndarray.INDArray	derivativeActivation(org.nd4j.linalg.api.ndarray.INDArray input) Take the derivative of the given input based on the activation
Gradient	error(org.nd4j.linalg.api.ndarray.INDArray input) Calculate error with respect to the current layer.
Gradient	errorSignal(Gradient error, org.nd4j.linalg.api.ndarray.INDArray input) Error signal for this layer Using the amount of error caused by this layer calculate the error signal used as input in to the next layer.
Collection<IterationListener>	getIterationListeners() Get the iteration listeners for this layer.
void	merge(Layer layer, int batchSize) Parameter averaging
org.nd4j.linalg.api.ndarray.INDArray	preOutput(org.nd4j.linalg.api.ndarray.INDArray x) Raw activations
void	setIterationListeners(Collection<IterationListener> listeners) Set the iteration listeners for this layer.
Layer	transpose() Return a transposed copy of the weights/bias (this means reverse the number of inputs and outputs on the weights)
Layer.Type	type() Returns the layer type

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

accumulateScore, batchSize, clear, conf, fit, fit, getOptimizer, getParam, gradient, gradientAndScore, initParams, input, iterate, numParams, params, paramTable, score, setConf, setParam, setParams, setParamTable, setScore, transform, update, validateInput

Method Detail

type

Layer.Type type()

Returns the layer type

Returns:

error

Gradient error(org.nd4j.linalg.api.ndarray.INDArray input)

Calculate error with respect to the current layer. This gradient will contain the error signal

Parameters:

input - the gradient for the forward layer If this is the final layer, it will start with the error from the output. This is on the user to initialize.

Returns:

the gradient wrt the parameters on the current layer

derivativeActivation

org.nd4j.linalg.api.ndarray.INDArray derivativeActivation(org.nd4j.linalg.api.ndarray.INDArray input)

Take the derivative of the given input based on the activation

Parameters:

input - the input to take the derivative of

Returns:

the derivative of the action

calcGradient

Gradient calcGradient(Gradient layerError,
                      org.nd4j.linalg.api.ndarray.INDArray indArray)

Calculate the gradient

Parameters:

layerError - the layer error

indArray -

Returns:

the gradient

errorSignal

Gradient errorSignal(Gradient error,
                     org.nd4j.linalg.api.ndarray.INDArray input)

Error signal for this layer Using the amount of error caused by this layer calculate the error signal used as input in to the next layer. This is used for actually calculating the gradient of the layer

Parameters:

error -

input -

Returns:

backwardGradient

Gradient backwardGradient(org.nd4j.linalg.api.ndarray.INDArray activation,
                          Gradient errorSignal)

Calculate the gradient relative to the error in the next layer

Parameters:

activation - the activation from the network

errorSignal - the error signal caused by this network.

Returns:

merge

void merge(Layer layer,
           int batchSize)

Parameter averaging

Parameters:

layer - the layer to merge

batchSize - the batch size to merge on

activationMean

org.nd4j.linalg.api.ndarray.INDArray activationMean()

Calculate the mean representation for the activation for this layer

Returns:

the activation mean for this layer

preOutput

org.nd4j.linalg.api.ndarray.INDArray preOutput(org.nd4j.linalg.api.ndarray.INDArray x)

Raw activations

Parameters:

x - the input to transform

Returns:

the raw activation for this layer

activate

org.nd4j.linalg.api.ndarray.INDArray activate()

Trigger an activation with the last specified input

Returns:

the activation of the last specified input

activate

org.nd4j.linalg.api.ndarray.INDArray activate(org.nd4j.linalg.api.ndarray.INDArray input)

Initialize the layer with the given input and return the activation for this layer given this input

Parameters:

input - the input to use

Returns:

transpose

Layer transpose()

Return a transposed copy of the weights/bias (this means reverse the number of inputs and outputs on the weights)

Returns:

the transposed layer

clone

Layer clone()

Clone the layer

Returns:

backWard

Pair<Gradient,Gradient> backWard(Gradient errors,
                                 Gradient deltas,
                                 org.nd4j.linalg.api.ndarray.INDArray activation,
                                 String previousActivation)

Propagate errors backwards for a particular layer. This calculates the node error for a given true output.

Parameters:

errors - the errors to propagate

deltas - the previous deltas

previousActivation - the previous layer's activation

activation - the activation from the previous layer

getIterationListeners

Collection<IterationListener> getIterationListeners()

Get the iteration listeners for this layer.

setIterationListeners

void setIterationListeners(Collection<IterationListener> listeners)

Set the iteration listeners for this layer.