source: branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/BanditPolicies/BernoulliPolicyActionInfo.cs @ 11742

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

namespace HeuristicLab.Algorithms.Bandits.BanditPolicies {
  public class BernoulliPolicyActionInfo : IBanditPolicyActionInfo {
    public bool Disabled { get { return NumSuccess == -1; } }
    public int NumSuccess { get; private set; }
    public int NumFailure { get; private set; }
    public int Tries { get { return NumSuccess + NumFailure; } }
    public double Value { get { return NumSuccess / (double)(Tries); } }
    public void UpdateReward(double reward) {
      Debug.Assert(!Disabled);
      //Debug.Assert(reward.IsAlmost(0.0) || reward.IsAlmost(1.0));

      //if (reward.IsAlmost(1.0)) NumSuccess++;
      if (reward > 0) NumSuccess++;
      else NumFailure++;
    }
    public void Disable() {
      this.NumSuccess = -1;
      this.NumFailure = -1;
    }
    public void Reset() {
      NumSuccess = 0;
      NumFailure = 0;
    }
    public void PrintStats() {
      Console.WriteLine("expected value {0,5:F2} disabled {1}", Value, Disabled);
    }
  }
}