Package planner.utils.evaluators.regrmetrics

Class RAE

java.lang.Object
planner.utils.evaluators.regrmetrics.RegressionMetric
planner.utils.evaluators.regrmetrics.RAE
All Implemented Interfaces:
Evaluator
public class RAE extends RegressionMetric
RAE is a class implementing the Relative Absolute Error over the RegressionMetric abstract super class.
Since:
1.0
Version:
1.0
Author:
Alicia Vazquez Ramos

  • Field Summary

    Fields inherited from class planner.utils.evaluators.regrmetrics.RegressionMetric

    exe

  • Constructor Summary

    Constructors
    Constructor
    Description
    RAE(Execution exp)
    Constructor.

  • Method Summary

    Modifier and Type
    Method
    Description
    double
    compute(double[] real, double[] estimation)
    Computes Relative Absolute Error (RAE) given the real predicted value for an instance of a dataset and the estimation computed from an objective function.

    Methods inherited from class planner.utils.evaluators.regrmetrics.RegressionMetric

    evaluate

    Methods inherited from class java.lang.Object

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

  • Constructor Details

    • RAE

      public RAE(Execution exp)
      Constructor.
      Parameters:
      exp - configuration of the current execution.

  • Method Details

    • compute

      public double compute(double[] real, double[] estimation)
      Computes Relative Absolute Error (RAE) given the real predicted value for an instance of a dataset and the estimation computed from an objective function.
      Specified by:
      compute in class RegressionMetric
      Parameters:
      real - real value for a given instance of a dataset.
      estimation - prediction of the objective function used.
      Returns:
      error of the estimation.