org.neuroph.nnet.comp.layer

Class ConvolutionalLayer

java.lang.Object

org.neuroph.core.Layer

org.neuroph.nnet.comp.layer.FeatureMapsLayer

org.neuroph.nnet.comp.layer.ConvolutionalLayer

All Implemented Interfaces:

java.io.Serializable, java.lang.Iterable<Neuron>

public class ConvolutionalLayer
extends FeatureMapsLayer

Convolutional layer is a special type of layer, used in convolutional neural networks. It contains a collection of feature maps, default neuron settings for convolutional layers, and method for creating connections to feature maps. The role of the convolutional layer is extaction of high level features.

See Also:

FeatureMapsLayer, Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static NeuronProperties	DEFAULT_NEURON_PROP Default neuron properties for convolutional layer

Constructor Summary

Constructors

Constructor and Description
ConvolutionalLayer(FeatureMapsLayer fromLayer, Dimension2D kernelDimension, int numberOfMaps) Creates convolutional layer with specified kernel, appropriate map dimensions in regard to previous layer (fromLayer param) and specified number of feature maps with default neuron settings for convolutional layer.
ConvolutionalLayer(FeatureMapsLayer fromLayer, Dimension2D kernelDimension, int numberOfMaps, java.lang.Class<? extends TransferFunction> transferFunction) Creates convolutional layer with specified kernel, appropriate map dimensions in regard to previous layer (fromLayer param) and specified number of feature maps with default neuron settings for convolutional layer.
ConvolutionalLayer(FeatureMapsLayer fromLayer, Dimension2D kernelDimension, int numberOfMaps, NeuronProperties neuronProp) Creates convolutional layer with specified kernel, appropriate map dimensions in regard to previous layer (fromLayer param) and specified number of feature maps with given neuron properties.

Method Summary

Modifier and Type	Method and Description
void	connectMaps(FeatureMapLayer fromMap, FeatureMapLayer toMap) Creates connections with shared weights between two feature maps Assumes that toMap is from Convolutional layer.

Methods inherited from class org.neuroph.nnet.comp.layer.FeatureMapsLayer

addFeatureMap, getFeatureMap, getFeatureMaps, getMapDimensions, getNeuronAt, getNeuronsCount, getNumberOfMaps

Methods inherited from class org.neuroph.core.Layer

addNeuron, addNeuron, calculate, getLabel, getNeuronAt, getNeurons, getParentNetwork, indexOf, initializeWeights, isEmpty, iterator, removeAllNeurons, removeNeuron, removeNeuronAt, reset, setLabel, setNeuron, setParentNetwork

Methods inherited from class java.lang.Object

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

Methods inherited from interface java.lang.Iterable

forEach, spliterator

Field Detail

DEFAULT_NEURON_PROP

public static final NeuronProperties DEFAULT_NEURON_PROP

Default neuron properties for convolutional layer

Constructor Detail

ConvolutionalLayer

public ConvolutionalLayer(FeatureMapsLayer fromLayer,
                          Dimension2D kernelDimension,
                          int numberOfMaps)

Creates convolutional layer with specified kernel, appropriate map dimensions in regard to previous layer (fromLayer param) and specified number of feature maps with default neuron settings for convolutional layer.

Parameters:

fromLayer - previous layer, which will be connected to this layer

kernel - kernel for all feature maps

numberOfMaps - number of feature maps to create in this layer

ConvolutionalLayer

public ConvolutionalLayer(FeatureMapsLayer fromLayer,
                          Dimension2D kernelDimension,
                          int numberOfMaps,
                          java.lang.Class<? extends TransferFunction> transferFunction)

Creates convolutional layer with specified kernel, appropriate map dimensions in regard to previous layer (fromLayer param) and specified number of feature maps with default neuron settings for convolutional layer.

Parameters:

fromLayer - previous layer, which will be connected to this layer

kernel - kernel for all feature maps

numberOfMaps - number of feature maps to create in this layer

transferFunction - neuron's transfer function to use

ConvolutionalLayer

public ConvolutionalLayer(FeatureMapsLayer fromLayer,
                          Dimension2D kernelDimension,
                          int numberOfMaps,
                          NeuronProperties neuronProp)

Creates convolutional layer with specified kernel, appropriate map dimensions in regard to previous layer (fromLayer param) and specified number of feature maps with given neuron properties.

Parameters:

fromLayer - previous layer, which will be connected to this layer

kernel - kernel for all feature maps

numberOfMaps - number of feature maps to create in this layer

neuronProp - settings for neurons in feature maps

Method Detail

connectMaps

public void connectMaps(FeatureMapLayer fromMap,
                        FeatureMapLayer toMap)

Creates connections with shared weights between two feature maps Assumes that toMap is from Convolutional layer.

Kernel is used as a sliding window, and kernel positions overlap. Kernel is shifting right by one position at a time. Neurons at the same kernel position share the same weights

Specified by:

connectMaps in class FeatureMapsLayer

Parameters:

fromMap - source feature map

toMap - destination feature map