Package de.bwaldvogel.liblinear

Class Parameter

java.lang.Object
de.bwaldvogel.liblinear.Parameter
All Implemented Interfaces:
java.lang.Cloneable
public final class Parameter
extends java.lang.Object
implements java.lang.Cloneable

  • Constructor Summary

    Constructors 
    Constructor Description
    Parameter​(SolverType solver, double C, double eps)  
    Parameter​(SolverType solverType, double C, double eps, double p)  
    Parameter​(SolverType solverType, double C, double eps, int max_iters, double p)  
    Parameter​(SolverType solver, double C, int max_iters, double eps)  

  • Method Summary

    Modifier and Type Method Description
    Parameter clone()  
    double getC()  
    double getEps()  
    double[] getInitSol()  
    int getMaxIters()  
    double getNu()  
    int getNumWeights()
    the number of weights
    double getP()  
    SolverType getSolverType()  
    int[] getWeightLabels()  
    double[] getWeights()  
    boolean isRegularizeBias()  
    void setC​(double C)
    C is the cost of constraints violation.
    void setEps​(double eps)
    eps is the stopping criterion.
    void setInitSol​(double[] init_sol)  
    void setMaxIters​(int iters)  
    void setNu​(double nu)  
    void setP​(double p)
    set the epsilon in loss function of epsilon-SVR (default 0.1)
    void setRegularizeBias​(boolean regularizeBias)  
    void setSolverType​(SolverType solverType)  
    void setWeights​(double[] weights, int[] weightLabels)
    nr_weight, weight_label, and weight are used to change the penalty for some classes (If the weight for a class is not changed, it is set to 1).

    Methods inherited from class java.lang.Object

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

  • Constructor Details

  • Method Details

    • setWeights

      public void setWeights​(double[] weights, int[] weightLabels)

      nr_weight, weight_label, and weight are used to change the penalty for some classes (If the weight for a class is not changed, it is set to 1). This is useful for training classifier using unbalanced input data or with asymmetric misclassification cost.

      Each weight[i] corresponds to weight_label[i], meaning that the penalty of class weight_label[i] is scaled by a factor of weight[i].

      If you do not want to change penalty for any of the classes, just set nr_weight to 0.

    • getWeights

      public double[] getWeights()
      See Also:
      setWeights(double[], int[])

    • getWeightLabels

      public int[] getWeightLabels()
      See Also:
      setWeights(double[], int[])

    • getNumWeights

      public int getNumWeights()
      the number of weights
      See Also:
      setWeights(double[], int[])

    • setC

      public void setC​(double C)
      C is the cost of constraints violation. (we usually use 1 to 1000)

    • getC

      public double getC()

    • setEps

      public void setEps​(double eps)
      eps is the stopping criterion. (we usually use 0.01).

    • getEps

      public double getEps()

    • setMaxIters

      public void setMaxIters​(int iters)

    • getMaxIters

      public int getMaxIters()

    • setSolverType

      public void setSolverType​(SolverType solverType)

    • getSolverType

      public SolverType getSolverType()

    • setP

      public void setP​(double p)
      set the epsilon in loss function of epsilon-SVR (default 0.1)

    • getP

      public double getP()

    • setInitSol

      public void setInitSol​(double[] init_sol)

    • getInitSol

      public double[] getInitSol()

    • setNu

      public void setNu​(double nu)

    • getNu

      public double getNu()

    • setRegularizeBias

      public void setRegularizeBias​(boolean regularizeBias)

    • isRegularizeBias

      public boolean isRegularizeBias()

    • clone

      public Parameter clone()
      Overrides:
      clone in class java.lang.Object