Examples of NeighbourhoodSizeFunction

org.apache.commons.math3.ml.neuralnet.sofm.NeighbourhoodSizeFunction
Provides the network neighbourhood's size as a function of the number of calls already performed during the learning task. The "neighbourhood" is the set of neurons that can be reached by traversing at most the number of links returned by this function. @version $Id: NeighbourhoodSizeFunction.java 1566092 2014-02-08 18:48:29Z tn $ @since 3.3

Examples of org.apache.commons.math3.ml.neuralnet.sofm.NeighbourhoodSizeFunction

        final LearningFactorFunction learning
            = LearningFactorFunctionFactory.exponentialDecay(1e-1,
                                                             5e-2,
                                                             numSamplesPerTask / 2);
        final double numNeurons = FastMath.sqrt(sofm.getNumberOfRows() * sofm.getNumberOfColumns());
        final NeighbourhoodSizeFunction neighbourhood
            = NeighbourhoodSizeFunctionFactory.exponentialDecay(0.5 * numNeurons,
                                                                0.2 * numNeurons,
                                                                numSamplesPerTask / 2);


        for (int i = 0; i < numTasks; i++) {

View Full Code Here

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.