#85: added new interface for c# tau-first & generic explorers. cleaned up sample code in both c++ and c#

1 files changed, 25 insertions, 112 deletions

diff --git a/cs_test/ExploreOnlySample.cs b/cs_test/ExploreOnlySample.cs
index 6b2a19eb..16e7cc64 100644
--- a/cs_test/ExploreOnlySample.cs
+++ b/cs_test/ExploreOnlySample.cs
@@ -8,54 +8,8 @@ namespace cs_test
 {

     class ExploreOnlySample

     {

-        private static UInt32 SampleStatefulPolicyFunc(int policyParams, BaseContext appContext)

-        {

-            return (uint)((policyParams + appContext.GetFeatures().Length) % 10 + 1);

-        }

-

-        private static UInt32 SampleStatefulPolicyFunc2(int policyParams, BaseContext appContext)

-        {

-            return (uint)((policyParams + appContext.GetFeatures().Length) % 10 + 2);

-        }

-

-        private static UInt32 SampleStatefulPolicyFunc(CustomParams policyParams, BaseContext appContext)

-        {

-            return (uint)((policyParams.Value1 + policyParams.Value2 + appContext.GetFeatures().Length) % 10 + 1);

-        }

-

-        private static UInt32 SampleStatelessPolicyFunc(BaseContext appContext)

-        {

-            return (UInt32)appContext.GetFeatures().Length;

-        }

-

-        private static UInt32 SampleStatelessPolicyFunc2(BaseContext appContext)

-        {

-            return (UInt32)appContext.GetFeatures().Length + 1;

-        }

-

-        private static void SampleStatefulScorerFunc(int policyParams, BaseContext appContext, float[] scores)

-        {

-            for (uint i = 0; i < scores.Length; i++)

-            {

-                scores[i] = (int)policyParams + i;

-            }

-        }

-

-        private static void SampleStatelessScorerFunc(BaseContext appContext, float[] scores)

-        {

-            for (uint i = 0; i < scores.Length; i++)

-            {

-                scores[i] = appContext.GetFeatures().Length + i;

-            }

-        }

-

-        class CustomParams

-        {

-            public int Value1;

-            public int Value2;

-        }

-

         class MyContext { }

+

         class MyRecorder : IRecorder<MyContext>

         {

             public void Record(MyContext context, UInt32 action, float probability, string uniqueKey)

@@ -64,6 +18,7 @@ namespace cs_test
             }

             private List<uint> actions = new List<uint>();

         }

+

         class MyPolicy : IPolicy<MyContext>

         {

             public MyPolicy() : this(-1) { }

@@ -96,36 +51,13 @@ namespace cs_test
 

         public static void Run()

         {

-            string exploration_type = "greedy";

-            bool stateful = true;

-

-            MwtExplorer mwt = new MwtExplorer("test");

-

-            uint numActions = 10;

-

-            float epsilon = 0.2f;

-            uint tau = 0;

-            uint numbags = 2;

-            float lambda = 0.5f;

-

-            int policyParams = 1003;

-            int policyParams2 = 1004;

-            CustomParams customParams = new CustomParams() { Value1 = policyParams, Value2 = policyParams2 };

-            StatefulPolicyDelegate<int>[] bags = 

-            {

-                  new StatefulPolicyDelegate<int>(SampleStatefulPolicyFunc), 

-                  new StatefulPolicyDelegate<int>(SampleStatefulPolicyFunc2) 

-            };

-            int[] parameters = { policyParams, policyParams };

-            StatelessPolicyDelegate[] statelessbags = 

-            {

-                new StatelessPolicyDelegate(SampleStatelessPolicyFunc), 

-                new StatelessPolicyDelegate(SampleStatelessPolicyFunc2) 

-            };

+            string exploration_type = "generic";

 

             if (exploration_type == "greedy")

             {

                 // Initialize Epsilon-Greedy explore algorithm using custom Recorder, Policy & Context types

+                uint numActions = 10;

+                float epsilon = 0.2f;

                 MyRecorder mc = new MyRecorder();

                 MyPolicy mp = new MyPolicy();

                 MWT<MyContext> mwtt = new MWT<MyContext>("mwt", mc);

@@ -134,20 +66,20 @@ namespace cs_test
             }

             else if (exploration_type == "tau-first")

             {

-                if (stateful)

-                {

-                    /*** Initialize Tau-First explore algorithm using a default policy function that accepts parameters ***/

-                    mwt.InitializeTauFirst<CustomParams>(tau, new StatefulPolicyDelegate<CustomParams>(SampleStatefulPolicyFunc), customParams, numActions);

-                }

-                else

-                {

-                    /*** Initialize Tau-First explore algorithm using a stateless default policy function ***/

-                    mwt.InitializeTauFirst(tau, new StatelessPolicyDelegate(SampleStatelessPolicyFunc), numActions);

-                }

+                // Initialize Tau-First explore algorithm using custom Recorder, Policy & Context types

+                uint numActions = 10;

+                uint tau = 0;

+                MyRecorder mc = new MyRecorder();

+                MyPolicy mp = new MyPolicy();

+                MWT<MyContext> mwtt = new MWT<MyContext>("mwt", mc);

+                uint action = mwtt.Choose_Action(new TauFirstExplorer<MyContext>(mp, tau, numActions), "key", new MyContext());

+                return;

             }

             else if (exploration_type == "bagging")

             {

                 // Initialize Bagging explore algorithm using custom Recorder, Policy & Context types

+                uint numActions = 10;

+                uint numbags = 2;

                 MyRecorder mc = new MyRecorder();

                 MyPolicy[] mps = new MyPolicy[numbags];

                 for (int i = 0; i < numbags; i++)

@@ -162,6 +94,8 @@ namespace cs_test
             else if (exploration_type == "softmax")

             {

                 // Initialize Softmax explore algorithm using custom Recorder, Scorer & Context types

+                uint numActions = 10;

+                float lambda = 0.5f;

                 MyRecorder mc = new MyRecorder();

                 MyScorer ms = new MyScorer(numActions);

 

@@ -171,41 +105,20 @@ namespace cs_test
             }

             else if (exploration_type == "generic")

             {

-                if (stateful)

-                {

-                    /*** Initialize Generic explore algorithm using a default policy function that accepts parameters ***/

-                    mwt.InitializeGeneric<int>(new StatefulScorerDelegate<int>(SampleStatefulScorerFunc), policyParams, numActions);

-                }

-                else

-                {

-                    /*** Initialize Generic explore algorithm using a stateless default policy function ***/

-                    mwt.InitializeGeneric(new StatelessScorerDelegate(SampleStatelessScorerFunc), numActions);

-                }

+                // Initialize Generic explore algorithm using custom Recorder, Scorer & Context types

+                uint numActions = 10;

+                MyRecorder mc = new MyRecorder();

+                MyScorer ms = new MyScorer(numActions);

+

+                MWT<MyContext> mwtt = new MWT<MyContext>("mwt", mc);

+                uint action = mwtt.Choose_Action(new GenericExplorer<MyContext>(ms, numActions), "key", new MyContext());

+                return;

             }

             else

             {  //add error here

 

 

             }

-            Feature[] f = new Feature[2];

-            f[0].Value = 0.5f;

-            f[0].Id = 1;

-            f[1].Value = 0.9f;

-            f[1].Id = 2;

-

-            string otherContext = "Some other context data that might be helpful to log";

-            OldSimpleContext appContext = new OldSimpleContext(f, otherContext);

-

-            UInt32 chosenAction = mwt.ChooseAction("myId", appContext);

-

-            Interaction[] interactions = mwt.GetAllInteractions();

-            // string interactions = mwt.GetAllInteractionsAsString();

-

-            mwt.Uninitialize();

-

-            Console.WriteLine(chosenAction);

-            Console.WriteLine(interactions);

-

         }

     }

 }