public class KohonenUpdateAction extends Object implements UpdateAction
update
method modifies the
features w
of the "winning" neuron and its neighbours
according to the following rule:
wnew = wold + α e(-d / σ) * (sample - wold)
where
d
is the number of links to traverse in order to reach
the neuron from the winning neuron.constructor
are instances of thread-safe
classes.
update
method
will increment the internal counter used to compute the current
values for
Constructor and Description |
---|
KohonenUpdateAction(DistanceMeasure distance,
LearningFactorFunction learningFactor,
NeighbourhoodSizeFunction neighbourhoodSize) |
Modifier and Type | Method and Description |
---|---|
long |
getNumberOfCalls()
Retrieves the number of calls to the
update
method. |
void |
update(Network net,
double[] features)
Updates the network in response to the sample
features . |
public KohonenUpdateAction(DistanceMeasure distance, LearningFactorFunction learningFactor, NeighbourhoodSizeFunction neighbourhoodSize)
distance
- Distance function.learningFactor
- Learning factor update function.neighbourhoodSize
- Neighbourhood size update function.public void update(Network net, double[] features)
features
.update
in interface UpdateAction
net
- Network.features
- Training data.public long getNumberOfCalls()
update
method.Copyright © 2003–2016 The Apache Software Foundation. All rights reserved.