Package org.joone.engine

Interface Summary
IErrorPatternListener  
InputPatternListener This interface represents an input synapse for a generic layer.
Learnable  
LearnableLayer  
LearnableSynapse  
Learner  
LearnerFactory Learner factories are used to provide the synapses and layers, through the monitor object with Leaners.
NeuralElement This interface represents a generic element of a neural network
NeuralLayer This is the interface for all the layer objects of the neural network
NeuralNetListener  
OutputPatternListener This interface represents an output synapse for a generic layer.
 

Class Summary
AbstractEventNotifier This class raises an event notification invoking the corrisponnding Monitor.fireXXX method.
AbstractLearner This class provides some basic simple functionality that can be used (extended) by other learners.
BasicLearner  
BatchLearner BatchLearner stores the weight/bias changes during the batch and updates them after the batch is done.
BiasedLinearLayer This layer consists of linear neurons, i.e.
BufferedSynapse This class implements a synapse that permits to have asynchronous methods to write output patterns.
CircularSpatialMap This class implements the SpatialMap interface providing a circular spatial map for use with the GaussianLayer and Kohonen Networks.
ContextLayer The context layer is similar to the linear layer except that it has an auto-recurrent connection between its output and input.
DelayLayer Delay unit to create temporal windows from time series
O---> Yk(t-N)
|
...
DelayLayerBeanInfo  
DelaySynapse This Synapse connects the N input neurons with the M output neurons using a matrix of FIRFilter elements of size NxM.
DirectSynapse This is forward-only synapse.
EKFFFNLearnerPlugin A plugin listener that implements the EKFFFN learner used to train feed forward neural networks.
EKFRNNLearnerPlugin A plugin listener that implements the EKF learner, based on "Some observations on the use of the extended Kalman filter as a recurrent network learning algorithm" by Williams (1992) in order to train a network.
ExtendableLearner Learners that extend this class are forced to implement certain functions, a so-called skeleton.
ExtendedKalmanFilterFFN Implements the extended Kalman filter (EKF) as described in "Using an extended Kalman filter learning algorithm for feed-forward neural networks to describe tracer correlations" by Lary and Mussa (2004) in order to train a feed-forward neural network.
ExtendedKalmanFilterRNN Implements the extended Kalman filter (EKF) as described in "Some observations on the use of the extended Kalman filter as a recurrent network learning algorithm" by Williams (1992) in order to train a recurrent neural network.
Fifo The Fifo class represents a first-in-first-out (FIFO) stack of objects.
FIRFilter Element of a connection representing a FIR filter (Finite Impulse Response).
FreudRuleFullSynapse Deprecated. possible bug in implementation
FullSynapse  
GaussianLayer This layer implements the Gaussian Neighborhood SOM strategy.
GaussianLayerBeanInfo  
GaussianSpatialMap This class implements the SpatialMap interface providing a circular spatial map for use with the GaussianLayer and Kohonen Networks.
GaussLayer The output of a Gauss(ian) layer neuron is the sum of the weighted input values, applied to a gaussian curve (exp(- x * x)).
KohonenSynapse This is an unsupervised Kohonen Synapse which is a Self Organising Map.
KohonenSynapseBeanInfo  
Layer The Layer object is the basic element forming the neural net.
LayerBeanInfo  
LinearLayer The output of a linear layer neuron is the sum of the weighted input values, scaled by the beta parameter.
LinearLayerBeanInfo  
LogarithmicLayer This layer implements a logarithmic transfer function.
Matrix The Matrix object represents the connection matrix of the weights of a synapse or the biases of a layer.
MatrixBeanInfo  
MemoryLayer  
MemoryLayerBeanInfo  
Monitor The Monitor object is the controller of the behavior of the neural net.
MonitorBeanInfo  
NetErrorManager This class should be used when ever a critical error occurs that would impact on the training or running of the network.
NetStoppedEventNotifier Raises the netStopped event from within a separate Thread
NeuralNetAdapter  
NeuralNetEvent Transport class used to notify the events raised from a neural network
OutputSwitchSynapse This class acts as a switch that can connect its input to one of its connected output synapses.
OutputSwitchSynapseBeanInfo  
Pattern The pattern object contains the data that must be processed from a neural net.
PatternBeanInfo  
RbfGaussianLayer This class implements the nonlinear layer in Radial Basis Function (RBF) networks using Gaussian functions.
RbfGaussianParameters This class defines the parameters, like center, sigma, etc.
RbfInputSynapse The synapse to the input of a radial basis function layer should't provide a single value to every neuron in the output (RBF) layer, as is usual the case.
RbfLayer This is the basis (helper) for radial basis function layers.
RpropLearner This class implements the RPROP learning algorithm.
RpropParameters This object holds the global parameters for the RPROP learning algorithm (RpropLearner).
RTRL A RTRL implementation.
RTRLLearnerFactory A RTRL implementation.
RTRLLearnerPlugin A plugin listener that applies the RTRL algorithm to a network.
SangerSynapse This is the synapse useful to extract the principal components from an input data set.
SigmoidLayer The output of a sigmoid layer neuron is the sum of the weighted input values, applied to a sigmoid function.
SimpleLayer This abstract class represents layers that are composed by neurons that implement some transfer function.
SimpleLayerBeanInfo  
SineLayer The output of a sine layer neuron is the sum of the weighted input values, applied to a sine (sin(x)).
SoftmaxLayer The outputs of the Softmax layer must be interpreted as probabilities.
SpatialMap SpatialMap is intended to be an abstract spatial map for use with a GaussianLayer.
Synapse The Synapse is the connection element between two Layer objects.
SynapseBeanInfo  
TanhLayer Layer that applies the tangent hyperbolic transfer function to its input patterns
TanhLayerBeanInfo  
WTALayer This layer implements the Winner Takes All SOM strategy.
WTALayerBeanInfo  
 



Submit Feedback to pmarrone@users.sourceforge.net