Documentation of 'com.datumbox.framework.core.mathematics.regularization.ElasticNetRegularizer' Java class
ElasticNetRegularizer
com.datumbox.framework.core.mathematics.regularization

Class ElasticNetRegularizer



  • public class ElasticNetRegularizer
    extends java.lang.Object
    Utility class for ElasticNet regularization. https://web.stanford.edu/~hastie/Papers/B67.2%20(2005)%20301-320%20Zou%20&%20Hastie.pdf http://web.stanford.edu/~hastie/TALKS/enet_talk.pdf
    • Method Summary

      All Methods Static Methods Concrete Methods 
      Modifier and Type Method and Description
      static <K> double estimatePenalty(double l1, double l2, java.util.Map<K,java.lang.Double> weights)
      Estimates the penalty by adding the ElasticNet regularization.
      static <K> void updateWeights(double l1, double l2, double learningRate, java.util.Map<K,java.lang.Double> weights, java.util.Map<K,java.lang.Double> newWeights)
      Updates the weights by applying the ElasticNet regularization.
      • Methods inherited from class java.lang.Object

        equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
    • Constructor Detail

      • ElasticNetRegularizer

        public ElasticNetRegularizer()
    • Method Detail

      • updateWeights

        public static <K> void updateWeights(double l1,
                                             double l2,
                                             double learningRate,
                                             java.util.Map<K,java.lang.Double> weights,
                                             java.util.Map<K,java.lang.Double> newWeights)
        Updates the weights by applying the ElasticNet regularization.
        Type Parameters:
        K -
        Parameters:
        l1 -
        l2 -
        learningRate -
        weights -
        newWeights -
      • estimatePenalty

        public static <K> double estimatePenalty(double l1,
                                                 double l2,
                                                 java.util.Map<K,java.lang.Double> weights)
        Estimates the penalty by adding the ElasticNet regularization.
        Type Parameters:
        K -
        Parameters:
        l1 -
        l2 -
        weights -
        Returns:

DataMelt 3.0 © DataMelt by jWork.ORG

You see the box below because you did not login.