org.deeplearning4j.nn.conf.layers

Class CnnLossLayer

java.lang.Object

org.deeplearning4j.nn.conf.layers.Layer

org.deeplearning4j.nn.conf.layers.BaseLayer

org.deeplearning4j.nn.conf.layers.FeedForwardLayer

org.deeplearning4j.nn.conf.layers.CnnLossLayer

All Implemented Interfaces:

Serializable, Cloneable

public class CnnLossLayer
extends FeedForwardLayer

Convolutional Neural Network Loss Layer.
Handles calculation of gradients etc for various objective functions.
NOTE: CnnLossLayer does not have any parameters. Consequently, the output activations size is equal to the input size.
Input and output activations are same as other CNN layers: 4 dimensions with shape [miniBatchSize,channels,height,width]
CnnLossLayer has support for a built-in activation function (tanh, softmax etc) - if this is not required, set activation function to Activation.IDENTITY. For activations such as softmax, note that this is applied depth-wise: that is, softmax is applied along dimension 1 (depth) for each minibatch, and x/y location separately.

Note that 3 types of masking are supported: (n=minibatchSize, c=channels, h=height, w=width)
- Per example masking: Where an example is present or not (and all outputs are masked by it). Mask shape [n,1]
- Per x/y location masking: where each spatial X/Y location is present or not (all channels at a given x/y are masked by it). Mask shape: [n,h,w].
- Per output masking: Where each output activation value is present or not - mask shape [n,c,h,w] (same as output)

Author:

Alex Black

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	CnnLossLayer.Builder

Field Summary

Fields

Modifier and Type	Field and Description
protected org.nd4j.linalg.lossfunctions.ILossFunction	lossFn

Fields inherited from class org.deeplearning4j.nn.conf.layers.FeedForwardLayer

nIn, nOut

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

activationFn, biasInit, biasUpdater, dist, gradientNormalization, gradientNormalizationThreshold, iUpdater, l1, l1Bias, l2, l2Bias, weightInit, weightNoise

Fields inherited from class org.deeplearning4j.nn.conf.layers.Layer

constraints, iDropout, layerName

Method Summary

Modifier and Type	Method and Description
LayerMemoryReport	getMemoryReport(InputType inputType) This is a report of the estimated memory consumption for the given layer
InputType	getOutputType(int layerIndex, InputType inputType) For a given type of input to this layer, what is the type of the output?
InputPreProcessor	getPreProcessorForInputType(InputType inputType) For the given type of input to this layer, what preprocessor (if any) is required? Returns null if no preprocessor is required, otherwise returns an appropriate InputPreProcessor for this layer, such as a CnnToFeedForwardPreProcessor
ParamInitializer	initializer()
Layer	instantiate(NeuralNetConfiguration conf, Collection<TrainingListener> trainingListeners, int layerIndex, org.nd4j.linalg.api.ndarray.INDArray layerParamsView, boolean initializeParams)
void	setNIn(InputType inputType, boolean override) Set the nIn value (number of inputs, or input channels for CNNs) based on the given input type

Methods inherited from class org.deeplearning4j.nn.conf.layers.FeedForwardLayer

getL1ByParam, getL2ByParam, isPretrainParam

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

clone, getUpdaterByParam, resetLayerDefaultConfig

Methods inherited from class org.deeplearning4j.nn.conf.layers.Layer

initializeConstraints

Methods inherited from class java.lang.Object

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

Field Detail

lossFn

protected org.nd4j.linalg.lossfunctions.ILossFunction lossFn

Method Detail

instantiate

public Layer instantiate(NeuralNetConfiguration conf,
                         Collection<TrainingListener> trainingListeners,
                         int layerIndex,
                         org.nd4j.linalg.api.ndarray.INDArray layerParamsView,
                         boolean initializeParams)

Specified by:

instantiate in class Layer

initializer

public ParamInitializer initializer()

Specified by:

initializer in class Layer

Returns:

The parameter initializer for this model

getOutputType

public InputType getOutputType(int layerIndex,
                               InputType inputType)

Description copied from class: Layer

For a given type of input to this layer, what is the type of the output?

Overrides:

getOutputType in class FeedForwardLayer

Parameters:

layerIndex - Index of the layer

inputType - Type of input for the layer

Returns:

Type of output from the layer

getPreProcessorForInputType

public InputPreProcessor getPreProcessorForInputType(InputType inputType)

Description copied from class: Layer

For the given type of input to this layer, what preprocessor (if any) is required?
Returns null if no preprocessor is required, otherwise returns an appropriate InputPreProcessor for this layer, such as a CnnToFeedForwardPreProcessor

Overrides:

getPreProcessorForInputType in class FeedForwardLayer

Parameters:

inputType - InputType to this layer

Returns:

Null if no preprocessor is required, otherwise the type of preprocessor necessary for this layer/input combination

getMemoryReport

public LayerMemoryReport getMemoryReport(InputType inputType)

Description copied from class: Layer

This is a report of the estimated memory consumption for the given layer

Specified by:

getMemoryReport in class Layer

Parameters:

inputType - Input type to the layer. Memory consumption is often a function of the input type

Returns:

Memory report for the layer

setNIn

public void setNIn(InputType inputType,
                   boolean override)

Description copied from class: Layer

Set the nIn value (number of inputs, or input channels for CNNs) based on the given input type

Overrides:

setNIn in class FeedForwardLayer

Parameters:

inputType - Input type for this layer

override - If false: only set the nIn value if it's not already set. If true: set it regardless of whether it's already set or not.