source: branches/HeuristicLab.Problems.GrammaticalOptimization-gkr/HeuristicLab.Algorithms.Bandits/ActionInfos/MeanAndVariancePolicyActionInfo.cs @ 13234

using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.Linq;
using System.Text;
using System.Threading.Tasks;

namespace HeuristicLab.Algorithms.Bandits.BanditPolicies {
  public class MeanAndVariancePolicyActionInfo : IBanditPolicyActionInfo {
    private OnlineMeanAndVarianceEstimator estimator = new OnlineMeanAndVarianceEstimator();
    public int Tries { get { return estimator.N; } }
    public double SumReward { get { return estimator.Sum; } }
    public double AvgReward { get { return estimator.Avg; } }
    public double MaxReward { get; private set; }
    public double RewardVariance { get { return estimator.Variance; } }
    public double Value {
      get {
        return AvgReward;
      }
    }

    public void UpdateReward(double reward) {
      MaxReward = Math.Max(MaxReward, reward);
      estimator.UpdateReward(reward);
    }

    public void Reset() {
      MaxReward = double.NegativeInfinity;
      estimator.Reset();
    }

    public override string ToString() {
      return string.Format("{0:N3} {1,3}", AvgReward, Tries);
    }
  }
}