Trim some public fields on the Agent (#3269)

* Triming some of the methods of the agent but left SetReward * Fixing bugs * modifying the environments * Reintroducing IsDone and IsMaxStepReached * Updating the Migrating doc * more details on the Migration
5 年前 · a1a1126d
--- a/UnitySDK/Assets/ML-Agents/Editor/AgentEditor.cs
+++ b/UnitySDK/Assets/ML-Agents/Editor/AgentEditor.cs
            serializedAgent.Update();

            var maxSteps = serializedAgent.FindProperty(
-                "agentParameters.maxStep");
+                "maxStep");

            EditorGUILayout.PropertyField(
                maxSteps,
--- a/UnitySDK/Assets/ML-Agents/Editor/Tests/EditModeTestInternalBrainTensorGenerator.cs
+++ b/UnitySDK/Assets/ML-Agents/Editor/Tests/EditModeTestInternalBrainTensorGenerator.cs
            }
        }

-        static List<Agent> GetFakeAgents()
+        static List<TestAgent> GetFakeAgents()
        {
            var goA = new GameObject("goA");
            var bpA = goA.AddComponent<BehaviorParameters>();
            bpB.brainParameters.numStackedVectorObservations = 1;
            var agentB = goB.AddComponent<TestAgent>();

-            var agents = new List<Agent> { agentA, agentB };
+            var agents = new List<TestAgent> { agentA, agentB };
            foreach (var agent in agents)
            {
                var agentEnableMethod = typeof(Agent).GetMethod("OnEnableHelper",
                actionMasks = new[] { true, false, false, false, false },
            };

-            agentA.Info = infoA;
-            agentB.Info = infoB;
+
+            agentA._Info = infoA;
+            agentB._Info = infoB;
            return agents;
        }

            var agent1 = agentInfos[1];
            var inputs = new List<AgentInfoSensorsPair>
            {
-                new AgentInfoSensorsPair{agentInfo = agent0.Info, sensors = agent0.sensors},
-                new AgentInfoSensorsPair{agentInfo = agent1.Info, sensors = agent1.sensors},
+                new AgentInfoSensorsPair{agentInfo = agent0._Info, sensors = agent0.sensors},
+                new AgentInfoSensorsPair{agentInfo = agent1._Info, sensors = agent1.sensors},
            };
            generator.Generate(inputTensor, batchSize, inputs);
            Assert.IsNotNull(inputTensor.data);
            var agent1 = agentInfos[1];
            var inputs = new List<AgentInfoSensorsPair>
            {
-                new AgentInfoSensorsPair{agentInfo = agent0.Info, sensors = agent0.sensors},
-                new AgentInfoSensorsPair{agentInfo = agent1.Info, sensors = agent1.sensors},
+                new AgentInfoSensorsPair{agentInfo = agent0._Info, sensors = agent0.sensors},
+                new AgentInfoSensorsPair{agentInfo = agent1._Info, sensors = agent1.sensors},
            };
            generator.Generate(inputTensor, batchSize, inputs);
            Assert.IsNotNull(inputTensor.data);
            var agent1 = agentInfos[1];
            var inputs = new List<AgentInfoSensorsPair>
            {
-                new AgentInfoSensorsPair{agentInfo = agent0.Info, sensors = agent0.sensors},
-                new AgentInfoSensorsPair{agentInfo = agent1.Info, sensors = agent1.sensors},
+                new AgentInfoSensorsPair{agentInfo = agent0._Info, sensors = agent0.sensors},
+                new AgentInfoSensorsPair{agentInfo = agent1._Info, sensors = agent1.sensors},
            };

            generator.Generate(inputTensor, batchSize, inputs);
--- a/UnitySDK/Assets/ML-Agents/Editor/Tests/MLAgentsEditModeTest.cs
+++ b/UnitySDK/Assets/ML-Agents/Editor/Tests/MLAgentsEditModeTest.cs
 {
    public class TestAgent : Agent
    {
+
+        public AgentInfo _Info
+        {
+            get
+            {
+                return (AgentInfo)typeof(Agent).GetField("m_Info", BindingFlags.Instance | BindingFlags.NonPublic).GetValue(this);
+            }
+            set
+            {
+                typeof(Agent).GetField("m_Info", BindingFlags.Instance | BindingFlags.NonPublic).SetValue(this, value);
+            }
+        }
+
+        public bool IsDone()
+        {
+            return (bool)typeof(Agent).GetField("m_Done", BindingFlags.Instance | BindingFlags.NonPublic).GetValue(this);
+        }
        public int initializeAgentCalls;
        public int collectObservationsCalls;
        public int agentActionCalls;
            var agentEnableMethod = typeof(Agent).GetMethod(
                "OnEnableHelper", BindingFlags.Instance | BindingFlags.NonPublic);

-            agent1.agentParameters = new AgentParameters();
-            agent2.agentParameters = new AgentParameters();
            var decisionRequester = agent1.gameObject.AddComponent<DecisionRequester>();
            decisionRequester.DecisionPeriod = 2;
            decisionRequester.Awake();
            var agentEnableMethod = typeof(Agent).GetMethod(
                "OnEnableHelper", BindingFlags.Instance | BindingFlags.NonPublic);

-            agent1.agentParameters = new AgentParameters();
-            agent2.agentParameters = new AgentParameters();
            var decisionRequester = agent1.gameObject.AddComponent<DecisionRequester>();
            decisionRequester.DecisionPeriod = 2;


            var agentEnableMethod = typeof(Agent).GetMethod(
                "OnEnableHelper", BindingFlags.Instance | BindingFlags.NonPublic);
-            agent1.agentParameters = new AgentParameters();
-            agent2.agentParameters = new AgentParameters();
-            agent1.agentParameters.maxStep = 20;
+            agent1.maxStep = 20;

            agentEnableMethod?.Invoke(agent2, new object[] { });
            agentEnableMethod?.Invoke(agent1, new object[] { });
--- a/UnitySDK/Assets/ML-Agents/Examples/3DBall/Prefabs/3DBall.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/3DBall/Prefabs/3DBall.prefab
  m_Script: {fileID: 11500000, guid: aaba48bf82bee4751aa7b89569e57f73, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  ball: {fileID: 1036225416237908}
 --- !u!114 &1306725529891448089
 MonoBehaviour:
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1533320402322554
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/3DBall/Prefabs/3DBallHardNew.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/3DBall/Prefabs/3DBallHardNew.prefab
  m_Script: {fileID: 11500000, guid: edf26e11cf4ed42eaa3ffb7b91bb4676, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  ball: {fileID: 1142513601053358}
 --- !u!114 &8193279139064749781
 MonoBehaviour:
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1978072206102878
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Crawler/Prefabs/DynamicPlatform.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Crawler/Prefabs/DynamicPlatform.prefab
  m_Script: {fileID: 11500000, guid: 2f37c30a5e8d04117947188818902ef3, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  target: {fileID: 4490950947783742}
  ground: {fileID: 4684408634944056}
  detectTargets: 1
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 0
+  offsetStep: 0
 --- !u!1 &1520563409393552
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Crawler/Prefabs/FixedPlatform.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Crawler/Prefabs/FixedPlatform.prefab
  m_Script: {fileID: 11500000, guid: 2f37c30a5e8d04117947188818902ef3, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  target: {fileID: 4749909135913778}
  ground: {fileID: 4856650706546504}
  detectTargets: 0
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 0
+  offsetStep: 0
 --- !u!1 &1492926997393242
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Crawler/Scripts/CrawlerAgent.cs
+++ b/UnitySDK/Assets/ML-Agents/Examples/Crawler/Scripts/CrawlerAgent.cs
        {
            foreach (var bodyPart in m_JdController.bodyPartsDict.Values)
            {
-                if (bodyPart.targetContact && !IsDone() && bodyPart.targetContact.touchingTarget)
+                if (bodyPart.targetContact && bodyPart.targetContact.touchingTarget)
                {
                    TouchedTarget();
                }
--- a/UnitySDK/Assets/ML-Agents/Examples/FoodCollector/Prefabs/FoodCollectorArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/FoodCollector/Prefabs/FoodCollectorArea.prefab
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  area: {fileID: 1819751139121548}
  turnSpeed: 300
  moveSpeed: 2
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1482701732800114
 GameObject:
  m_ObjectHideFlags: 0
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1819751139121548}
  turnSpeed: 300
  moveSpeed: 2
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1819751139121548}
  turnSpeed: 300
  moveSpeed: 2
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1819751139121548}
  turnSpeed: 300
  moveSpeed: 2
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1819751139121548}
  turnSpeed: 300
  moveSpeed: 2
--- a/UnitySDK/Assets/ML-Agents/Examples/FoodCollector/Prefabs/VisualFoodCollectorArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/FoodCollector/Prefabs/VisualFoodCollectorArea.prefab
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1145096862361766}
  turnSpeed: 300
  moveSpeed: 2
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1145096862361766}
  turnSpeed: 300
  moveSpeed: 2
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  area: {fileID: 1145096862361766}
  turnSpeed: 300
  moveSpeed: 2
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1399553220224106
 GameObject:
  m_ObjectHideFlags: 0
  m_Script: {fileID: 11500000, guid: c66e6845309d241c78a6d77ee2567928, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 0
  area: {fileID: 1145096862361766}
  turnSpeed: 300
  moveSpeed: 2
--- a/UnitySDK/Assets/ML-Agents/Examples/GridWorld/Prefabs/Area.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/GridWorld/Prefabs/Area.prefab
--- a/UnitySDK/Assets/ML-Agents/Examples/Hallway/Prefabs/SymbolFinderArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Hallway/Prefabs/SymbolFinderArea.prefab
  m_Script: {fileID: 11500000, guid: b446afae240924105b36d07e8d17a608, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 3000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 6
+  maxStep: 3000
  ground: {fileID: 1510027348950282}
  area: {fileID: 1745841960385024}
  symbolOGoal: {fileID: 1410733827718496}
  m_EditorClassIdentifier: 
  DecisionPeriod: 6
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1510027348950282
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Hallway/Prefabs/VisualSymbolFinderArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Hallway/Prefabs/VisualSymbolFinderArea.prefab
  m_Script: {fileID: 11500000, guid: b446afae240924105b36d07e8d17a608, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 3000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 6
+  maxStep: 3000
  ground: {fileID: 1625056884785366}
  area: {fileID: 1689874756253538}
  symbolOGoal: {fileID: 1800868804754718}
  m_EditorClassIdentifier: 
  DecisionPeriod: 6
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1377584197416466
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Hallway/Scripts/HallwayAgent.cs
+++ b/UnitySDK/Assets/ML-Agents/Examples/Hallway/Scripts/HallwayAgent.cs
    {
        if (useVectorObs)
        {
-            AddVectorObs(GetStepCount() / (float)agentParameters.maxStep);
+            AddVectorObs(GetStepCount() / (float)maxStep);
        }
    }


    public override void AgentAction(float[] vectorAction)
    {
-        AddReward(-1f / agentParameters.maxStep);
+        AddReward(-1f / maxStep);
        MoveAgent(vectorAction);
    }

--- a/UnitySDK/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockArea.prefab
  m_Script: {fileID: 11500000, guid: dea8c4f2604b947e6b7b97750dde87ca, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  ground: {fileID: 1500989011945850}
  area: {fileID: 1125452240183160}
  areaBounds:
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1500989011945850
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockVisualArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/PushBlock/Prefabs/PushBlockVisualArea.prefab
  m_Script: {fileID: 11500000, guid: dea8c4f2604b947e6b7b97750dde87ca, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  ground: {fileID: 1913379827958244}
  area: {fileID: 1632733799967290}
  areaBounds:
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1626651094211584
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/PushBlock/Scripts/PushAgentBasic.cs
+++ b/UnitySDK/Assets/ML-Agents/Examples/PushBlock/Scripts/PushAgentBasic.cs
        MoveAgent(vectorAction);

        // Penalty given each step to encourage agent to finish task quickly.
-        AddReward(-1f / agentParameters.maxStep);
+        AddReward(-1f / maxStep);
    }

    public override float[] Heuristic()
--- a/UnitySDK/Assets/ML-Agents/Examples/Pyramids/Prefabs/AreaPB.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Pyramids/Prefabs/AreaPB.prefab
  m_Script: {fileID: 11500000, guid: b8db44472779248d3be46895c4d562d5, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  area: {fileID: 1464170487903594}
  areaSwitch: {fileID: 1432086782037750}
  useVectorObs: 1
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1148882946833254
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Pyramids/Prefabs/VisualAreaPyramids.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Pyramids/Prefabs/VisualAreaPyramids.prefab
  m_Script: {fileID: 11500000, guid: b8db44472779248d3be46895c4d562d5, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  area: {fileID: 1055559745433172}
  areaSwitch: {fileID: 1212218760704844}
  useVectorObs: 0
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1747856067778386
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Pyramids/Scripts/PyramidAgent.cs
+++ b/UnitySDK/Assets/ML-Agents/Examples/Pyramids/Scripts/PyramidAgent.cs

    public override void AgentAction(float[] vectorAction)
    {
-        AddReward(-1f / agentParameters.maxStep);
+        AddReward(-1f / maxStep);
        MoveAgent(vectorAction);
    }

--- a/UnitySDK/Assets/ML-Agents/Examples/Reacher/Prefabs/Agent.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Reacher/Prefabs/Agent.prefab
  m_Script: {fileID: 11500000, guid: 220b156e3b142406c8b76d4db981d044, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 4000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 4
+  maxStep: 4000
  pendulumA: {fileID: 1644872085946016}
  pendulumB: {fileID: 1053261483945176}
  hand: {fileID: 1654288206095398}
  m_EditorClassIdentifier: 
  DecisionPeriod: 4
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1644872085946016
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Soccer/Prefabs/SoccerFieldTwos.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Soccer/Prefabs/SoccerFieldTwos.prefab
  m_Script: {fileID: 11500000, guid: 2a2688ef4a36349f9aa010020c32d198, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 3000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 3000
  team: 0
  agentRole: 0
  area: {fileID: 114559182131992928}
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1100217258374548
 GameObject:
  m_ObjectHideFlags: 0
  m_Script: {fileID: 11500000, guid: 2a2688ef4a36349f9aa010020c32d198, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 3000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 3000
  team: 1
  agentRole: 1
  area: {fileID: 114559182131992928}
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1131626411948014
 GameObject:
  m_ObjectHideFlags: 0
  m_Script: {fileID: 11500000, guid: 2a2688ef4a36349f9aa010020c32d198, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 3000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 3000
  team: 1
  agentRole: 0
  area: {fileID: 114559182131992928}
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1141134673700168
 GameObject:
  m_ObjectHideFlags: 0
  m_Script: {fileID: 11500000, guid: 2a2688ef4a36349f9aa010020c32d198, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 3000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 3000
  team: 0
  agentRole: 1
  area: {fileID: 114559182131992928}
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1931023723143276
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Tennis/Prefabs/TennisArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Tennis/Prefabs/TennisArea.prefab
  m_Script: {fileID: 11500000, guid: e51a3fb0b3186433ea84fc1e0549cc91, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  ball: {fileID: 1273406647218856}
  invertX: 0
  score: 0
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1194790474478638
 GameObject:
  m_ObjectHideFlags: 0
  m_Script: {fileID: 11500000, guid: e51a3fb0b3186433ea84fc1e0549cc91, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  ball: {fileID: 1273406647218856}
  invertX: 1
  score: 0
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1969551055586186
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/Walker/Prefabs/WalkerPair.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/Walker/Prefabs/WalkerPair.prefab
  m_Script: {fileID: 11500000, guid: ccb0f85f0009540d7ad997952e2aed7b, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 5000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 5000
  target: {fileID: 4085853164035250}
  hips: {fileID: 4333477265252406}
  chest: {fileID: 4027052323869384}
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 0
+  offsetStep: 0
 --- !u!1 &1907539933197724
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Examples/WallJump/Prefabs/WallJumpArea.prefab
+++ b/UnitySDK/Assets/ML-Agents/Examples/WallJump/Prefabs/WallJumpArea.prefab
  m_Script: {fileID: 11500000, guid: 676fca959b8ee45539773905ca71afa1, type: 3}
  m_Name: 
  m_EditorClassIdentifier: 
-  agentParameters:
-    maxStep: 2000
-    onDemandDecision: 0
-    numberOfActionsBetweenDecisions: 5
+  maxStep: 2000
  noWallBrain: {fileID: 11400000, guid: fb2ce36eb40b6480e94ea0b5d7573e47, type: 3}
  smallWallBrain: {fileID: 11400000, guid: fb2ce36eb40b6480e94ea0b5d7573e47, type: 3}
  bigWallBrain: {fileID: 11400000, guid: 0468bf44b1efd4992b6bf22cadb50d89, type: 3}
  m_EditorClassIdentifier: 
  DecisionPeriod: 5
  RepeatAction: 1
+  offsetStep: 0
 --- !u!1 &1264699583886832
 GameObject:
  m_ObjectHideFlags: 0
--- a/UnitySDK/Assets/ML-Agents/Scripts/Agent.cs
+++ b/UnitySDK/Assets/ML-Agents/Scripts/Agent.cs
    public struct AgentAction
    {
        public float[] vectorActions;
-        public float value;
-    /// <summary>
-    /// Struct that contains all the Agent-specific parameters provided in the
-    /// Editor. This excludes the Brain linked to the Agent since it can be
-    /// modified programmatically.
-    /// </summary>
-    [Serializable]
-    public class AgentParameters
-    {
-        /// <summary>
-        /// The maximum number of steps the agent takes before being done.
-        /// </summary>
-        /// <remarks>
-        /// If set to 0, the agent can only be set to done programmatically (or
-        /// when the Academy is done).
-        /// If set to any positive integer, the agent will be set to done after
-        /// that many steps. Note that setting the max step to a value greater
-        /// than the academy max step value renders it useless.
-        /// </remarks>
-        public int maxStep;
-
-    }


    /// <summary>
        BehaviorParameters m_PolicyFactory;

        /// <summary>
-        /// Agent parameters specified within the Editor via AgentEditor.
+        /// The maximum number of steps the agent takes before being done.
-        [HideInInspector] public AgentParameters agentParameters;
+        /// <remarks>
+        /// If set to 0, the agent can only be set to done programmatically (or
+        /// when the Academy is done).
+        /// If set to any positive integer, the agent will be set to done after
+        /// that many steps. Note that setting the max step to a value greater
+        /// than the academy max step value renders it useless.
+        /// </remarks>
+        [HideInInspector] public int maxStep;
-        public AgentInfo Info
-        {
-            get { return m_Info; }
-            set { m_Info = value; }
-        }

        /// Current Agent action (message sent from Brain).
        AgentAction m_Action;
        }

        /// <summary>
-        /// Returns the current step counter (within the current epside).
+        /// Returns the current step counter (within the current episode).
        /// </summary>
        /// <returns>
        /// Current episode number.
            return m_StepCount;
-        }
-
-        /// <summary>
-        /// Resets the step reward and possibly the episode reward for the agent.
-        /// </summary>
-        public void ResetReward()
-        {
-            m_Reward = 0f;
-            if (m_Done)
-            {
-                m_CumulativeReward = 0f;
-            }
        }

        /// <summary>
 #endif
            m_Reward += increment;
            m_CumulativeReward += increment;
-        }
-
-        /// <summary>
-        /// Retrieves the step reward for the Agent.
-        /// </summary>
-        /// <returns>The step reward.</returns>
-        public float GetReward()
-        {
-            return m_Reward;
        }

        /// <summary>
        }

        /// <summary>
-        /// Updates the value of the agent.
-        /// </summary>
-        public void UpdateValueAction(float value)
-        {
-            m_Action.value = value;
-        }
-
-        protected float GetValueEstimate()
-        {
-            return m_Action.value;
-        }
-
-        /// <summary>
        /// Scales continuous action from [-1, 1] to arbitrary range.
        /// </summary>
        /// <param name="rawAction"></param>
        /// Signals the agent that it must reset if its done flag is set to true.
        void ResetIfDone()
        {
-            if (IsDone())
+            if (m_Done)
            {
                _AgentReset();
            }
        void SendInfo()
        {
            // If the Agent is done, it has just reset and thus requires a new decision
-            if (m_RequestDecision || IsDone())
+            if (m_RequestDecision || m_Done)
-                ResetReward();
+                m_Reward = 0f;
+                if (m_Done)
+                {
+                    m_CumulativeReward = 0f;
+                }
                m_Done = false;
                m_MaxStepReached = false;
                m_RequestDecision = false;
                AgentAction(m_Action.vectorActions);
            }

-            if ((m_StepCount >= agentParameters.maxStep)
-                && (agentParameters.maxStep > 0))
+            if ((m_StepCount >= maxStep) && (maxStep > 0))
            {
                m_MaxStepReached = true;
                Done();
--- a/UnitySDK/Assets/ML-Agents/Scripts/Grpc/GrpcExtensions.cs
+++ b/UnitySDK/Assets/ML-Agents/Scripts/Grpc/GrpcExtensions.cs
        {
            return new AgentAction
            {
-                vectorActions = aap.VectorActions.ToArray(),
-                value = aap.Value,
+                vectorActions = aap.VectorActions.ToArray()
            };
        }

--- a/docs/Migrating.md
+++ b/docs/Migrating.md
 ## Migrating from 0.13 to latest

 ### Important changes
+* The `Decision Period` and `On Demand decision` checkbox have been removed from the Agent. On demand decision is now the default (calling `RequestDecision` on the Agent manually.)
 * The Academy class was changed to a singleton, and its virtual methods were removed.
 * Trainer steps are now counted per-Agent, not per-environment as in previous versions. For instance, if you have 10 Agents in the scene, 20 environment steps now corresponds to 200 steps as printed in the terminal and in Tensorboard.
 * Curriculum config files are now YAML formatted and all curricula for a training run are combined into a single file.
-* The `Decision Period` and `On Demand decision` checkbox have been removed from the Agent. On demand decision is now the default (calling `RequestDecision` on the Agent manually.)
+* The `agentParameters` field of the Agent has been removed. (Contained only `maxStep` information)
+* `maxStep` is now a public field on the Agent. (Was moved from `agentParameters`)
+* The `Info` field of the Agent has been made private. (Was only used internally and not meant to be modified outside of the Agent)
+* The `GetReward()` method on the Agent has been removed. (It was being confused with `GetCumulativeReward()`)
+* The `AgentAction` struct no longer contains a `value` field. (Value estimates were not set during inference)
+* The `GetValueEstimate()` method on the Agent has been removed.
+* The `UpdateValueAction()` method on the Agent has been removed.
+* If you were not using `On Demand Decision` for your Agent, you **must** add a `DecisionRequester` component to your Agent GameObject and set its `Decision Period` field to the old `Decision Period` of the Agent.
 * If you have a class that inherits from Academy:
  * If the class didn't override any of the virtual methods and didn't store any additional data, you can just remove the old script from the scene.
  * If the class had additional data, create a new MonoBehaviour and store the data on this instead.
 * Combine curriculum configs into a single file.  See [the WallJump curricula](../config/curricula/wall_jump.yaml) for an example of the new curriculum config format.
  A tool like https://www.json2yaml.com may be useful to help with the conversion.
 * If your Agent implements `AgentOnDone` and your Agent does not have the checkbox `Reset On Done` checked in the inspector, you must call the code that was in `AgentOnDone` manually.
-* If you were not using `On Demand Decision` for your Agent, you must add a `DecisionRequester` component to your Agent GameObject and set its `Decision Period` field to the old `Decision Period` of the Agent.
+

 ## Migrating from ML-Agents toolkit v0.12.0 to v0.13.0