Changeset 11727 for branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/Policies/RandomPolicy.cs

Timestamp:

12/29/14 11:02:36 (9 years ago)

Author:

gkronber

Message:

#2283: worked on grammatical optimization problem solvers (simple MCTS done)

File:

• branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/Policies/RandomPolicy.cs (modified) (2 diffs)

branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/Policies/RandomPolicy.cs

@@ -1 +1 @@
 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 {
   public class RandomPolicy : BanditPolicy {
     private readonly Random random;
+
     public RandomPolicy(Random random, int numActions)
       : base(numActions) {
@@ -14 +17 @@
 
     public override int SelectAction() {
-      return random.Next(NumActions);
+      Debug.Assert(Actions.Any());
+      return Actions.SelectRandom(random);
     }
     public override void UpdateReward(int action, double reward) {
       // do nothing
     }
-    public override void Reset() {
-      // do nothing
-    }
+
   }
 }